Butanol production in S. cerevisiae via a synthetic ABE pathway is enhanced by specific metabolic engineering and butanol resistance
- R. Swidah†1,
- H. Wang†1,
- P.J. Reid1,
- H.Z. Ahmed1,
- A.M. Pisanelli2,
- K.C. Persaud2,
- C.M. Grant1 and
- M.P. Ashe1Email author
© Swidah et al. 2015
Received: 31 March 2015
Accepted: 29 June 2015
Published: 8 July 2015
The fermentation of sugars to alcohols by microbial systems underpins many biofuel initiatives. Short chain alcohols, like n-butanol, isobutanol and isopropanol, offer significant advantages over ethanol in terms of fuel attributes. However, production of ethanol from resistant Saccharomyces cerevisiae strains is significantly less complicated than for these alternative alcohols.
In this study, we have transplanted an n-butanol synthesis pathway largely from Clostridial sp. to the genome of an S. cerevisiae strain. Production of n-butanol is only observed when additional genetic manipulations are made to restore any redox imbalance and to drive acetyl-CoA production. We have used this butanol production strain to address a key question regarding the sensitivity of cells to short chain alcohols. In the past, we have defined specific point mutations in the translation initiation factor eIF2B based upon phenotypic resistance/sensitivity to high concentrations of exogenously added n-butanol. Here, we show that even during endogenous butanol production, a butanol resistant strain generates more butanol than a butanol sensitive strain.
These studies demonstrate that appreciable levels of n-butanol can be achieved in S. cerevisiae but that significant metabolic manipulation is required outside of the pathway converting acetyl-CoA to butanol. Furthermore, this work shows that the regulation of protein synthesis by short chain alcohols in yeast is a critical consideration if higher yields of these alcohols are to be attained.
Since fossil fuels represent a finite resource and their continued use contributes to climate change, alternative sources of energy have been widely sought . Biofuels produced from fermentation of renewable resources are expected to represent an important replacement for gasoline . Commercial bioethanol production from high yielding fermentations of the yeast Saccharomyces cerevisiae relies upon the inherent resistance of yeast cells to the damaging properties of ethanol . However, ethanol’s low energy content and high hygroscopicity are viewed as disadvantages in terms of its quality as a fuel [4, 5]. n-Butanol (1-butanol) and other short chain alcohols have a range of physical properties, which make them superior fuels to ethanol . For instance, in comparison to ethanol, n-butanol is less hygroscopic making it less corrosive, and it has a higher energy density and octane value. These characteristics mean that n-butanol can be mixed with gasoline in almost any proportion .
Engineered E. coli bearing the ABE pathway have been generated in a number of different ways and have been shown to produce high levels of butanol [11, 12]. However, as for Clostridia, some problems still exist in the use of engineered E. coli for butanol production, including the potential for phage infection/fermentation spoilage and product/degradation product toxicity . As S. cerevisiae is currently widely used for the production of bioethanol, it holds significant advantages in terms of scalable industrial fermentation for the production of butanol . However, initial attempts at introducing the ABE pathway into S. cerevisiae produced very low yields of 2.5 mg/L . Subsequent studies have generated improved yields by targeting specific metabolic pathways or utilising specific starting substrates [16, 17]. In addition, alternative pathways for butanol production have been sought with varying degrees of success [18, 19]. Recurrent issues associated with these butanol fermentations are relatively low yields and the potential for end-product toxicity.
Previously, we have studied, at the molecular level, mechanisms underlying the toxic effects of n-butanol and other alcohols in yeast [20–23]. We have found that these alcohols specifically inhibit protein synthesis at the translation initiation step by perturbing the guanine nucleotide exchange factor, eukaryotic initiation factor (eIF)2B [20, 21]. This factor recycles, eIF2, a key g-protein involved in translation initiation. eIF2 in the GTP bound form recruits the initiator methionyl tRNA to the ribosome . As a consequence of translation initiation, GTP is hydrolysed on eIF2 generating eIF2-GDP, which requires eIF2B-dependent recycling before further rounds of translation initiation are possible.
In this study, we explore the hypothesis that yeast strains, which are more resistant to the toxic effects of n-butanol and other alcohols, are capable of producing more alcohol. In order to assess this question, we generated strains bearing the entire ABE pathway, as well as specific metabolic mutations designed to increase carbon flux towards the ABE pathway. As a result, we obtained a strain of yeast that is capable of producing up to 300 mg/L n-butanol. Overall, even though this level of n-butanol does not begin to approach the level required to inhibit eIF2B and generate toxicity, we observe a significant difference in the level of n-butanol produced in strains that only vary in their sensitivity/resistance to alcohols. Therefore, the toxicity of alcohols on cells is a significant factor when considering biofuel production and strategies aimed at overcoming this toxicity hold significant promise in the quest towards commercially economic biofuel yields.
Results and discussion
Addition of the ABE pathway to S. cerevisiae results in very low levels of n-butanol
The goal of this project at the outset was to determine whether the toxic effects of alcohols such as n-butanol are important in determining the yield from producing strains. We started with two parent strains that are isogenic apart from a point mutation in a gene encoding a translation initiation factor; GCD1. GCD1-P180 (denoted BR throughout) is resistant to 1 % butanol, whereas GCD1-S180 (denoted BS throughout) is sensitive to this level of exogenously added butanol. In order to evaluate this question, we generated BS and BR strains of yeast expressing four Clostridia beijerinckii genes and one yeast gene that together encode the enzymes of an ABE pathway. Previous studies had shown that yeast strains harbouring the genes for these enzymes on extremely high copy plasmids produced n-butanol at quite low levels of ~2.5 mg/L . Therefore, we decided to integrate codon-optimised genes directly into specific sites associated with high expression on the genome  under the control of a highly efficient ubiquitous yeast TDH3 gene promoter with a CYC1 3′ end formation sequence downstream. Each open reading frame (ORF) was also tagged with Flag epitopes to aid protein detection (Fig. 1b).
Individual genes were integrated into opposing mating type haploid yeast strains, such that via a combination of genetic crosses (see Methods), strains were constructed bearing all five genes (Fig. 1c). Western blotting using an anti-Flag monoclonal antibody confirmed that proteins of an appropriate size were expressed (Fig. 1d). However, when butanol was quantified from the strains under a variety of conditions, including anaerobic fermentation, very little butanol was recovered in the media (<10 mg/L) and levels of ethanol production were equivalent to the parent strains (Fig. 1e, f). The low butanol production observed in this strain was entirely consistent with previous attempts to produce n-butanol in S. cerevisiae .
Deletion of the ADH1 gene improves the n-butanol yield significantly
Therefore, a strategy was designed whereby the ADH1 gene was deleted (Fig. 2b) to give strains with the previously described  actinomycin A sensitive phenotype (Fig. 2c). The ADH1 deletion was subsequently confirmed by PCR on genomic DNA from the selected transformants (Fig. 2d).
Consistent with the deletion of a major alcohol dehydrogenase, growth and the levels of ethanol produced by the adh1Δ strain were very low compared to the wild type strain under anaerobic conditions (Fig. 2d and Additional file 1: Figure S1). In addition, glucose present at the outset was not entirely consumed during the fermentation (Additional file 1: Figure S1). Interestingly, for the strain bearing the butanol production pathway (BR +5 g), adh1Δ still reduced ethanol levels dramatically but not to the same extent as an adh1Δ strain lacking the butanol pathway (Fig. 2d). Furthermore, the impact of deleting the ADH1 gene in this strain was less pronounced in terms of growth and glucose consumption (Additional file 1: Figure S1). It is possible these minor fermentation improvements stem from the fact that the clostridial Adhe2 alcohol dehydrogenase is expressed as part of the butanol pathway, and this enzyme might to a small extent rescue production of ethanol from acetaldehyde.
Intriguingly, deletion of ADH1 also leads to the production of n-butanol. Recent studies suggest that in the absence of ADH1, an endogenous pathway of n-butanol production can be activated . This pathway likely stems from threonine catabolism  and appears to be responsible for the production of roughly 40 mg/L n-butanol from our strain (Fig. 2e). However, when an adh1Δ mutant is generated in the context of the strain harbouring the butanol production pathway (BR adh1Δ +5 g), approximately 150 mg/L n-butanol is generated (Fig. 2e).
Replacement of the Bcd gene with Ter does not significantly improve butanol yields
A codon-optimised ORF for the Ter gene was used to precisely supplant the Bcd ORF in the integration cassette, and therefore, a directly comparable Ter containing strain was obtained (Fig. 4). In contrast to what has been observed in E. coli [11, 12] and even though the levels of Ter were as high as the other integrated genes of the butanol pathway (Fig. 4b), the presence of the Ter gene did not alter the level of ethanol (Fig. 4c) or lead to significant improvements in the butanol titre (Fig. 4d).
Improved flux of carbon to acetyl CoA generates higher butanol levels
Butanol resistant strains generate higher levels of butanol
Having generated a strain that yields a reasonable level of butanol, we assessed the impact of butanol resistance/sensitivity at the level of translation initiation. Previous work from the lab has defined specific butanol resistance and sensitive mutations in the genes for eIF2B. In this case, we generated strains, which harboured allelic variation at the GCD1 locus, which encodes the γ subunit of eIF2B. A proline at residue 180 gives a resistant phenotype, whereas a Serine at this position increases sensitivity to butanol.
The resulting strains were tested for alcohol production using our standard assay system, and the butanol resistant strain reproducibly generated up to 1.5–2-fold higher peak levels of butanol (Fig. 5d). These results were unforeseen, as the level of butanol generated by these strains is significantly lower than the level added exogenously during the tolerance studies [20, 21]. In addition, the level of ethanol production was slightly reduced at early time points in the butanol resistant strain (Fig. 5c). This is suggestive that in the butanol resistant strain, a higher flux is attained towards butanol and away from ethanol than in the butanol sensitive strain. These results provide proof of principle that strains that are more resistant to the effects of butanol (and other fusel alcohols) have improved yields of these alcohols from production pathways.
In this study, we show that an exogenous ABE pathway only generates substantial levels of butanol in yeast when a number of metabolic alterations are made. Deletion of the major alcohol dehydrogenase ADH1 not only leads to butanol production via a previously described endogenous pathway but also promotes much higher levels of butanol where an exogenous butanol production pathway has been added. These data support a view that both the endogenous and exogenous pathways are active in the cells.
Our GC-MS studies highlight a number of potential bottlenecks particularly with regard to the exogenous pathway. Accumulation of crotonal led us to take an approach previously validated in E. coli: the replacement of the Bcd enzyme with Ter [11, 12]. However in our studies in yeast, Ter gives little improvement in butanol levels. It seems possible that neither of these enzymes is particularly efficient in the context of the yeast cytoplasm, and this could represent an area where substantial further improvements in yield are possible. The GC-MS data also show that acetaldehyde, acetate and 2,3-butanediol accumulate in an adh1Δ mutant bearing the ABE pathway. The accumulation of these compounds suggests that production of cytosolic acetyl-CoA from acetaldehyde occurs inefficiently. Therefore, a high expression strategy was applied to the ALD6 and ACS2 genes involved in this process. In strains, this metabolic alteration generated a moderate improvement in the levels of butanol from the strain; peak levels increase from ~175 to ~300 mg/L. Therefore, while stimulating cytosolic acetyl-CoA production does lead to an improvement in butanol production, a deficiency in this area is not a major limitation. This begs the question what is the major limitation that prevents greater butanol production. Possible answers lie in an imbalance in redox potential or in sensitivity of the cells to butanol itself or intermediates in the pathway.
The initial goal of this project was to assess whether differences in the sensitivity of strains to butanol prompted equivalent changes in the yield of butanol. Here, we use previously characterised strains bearing butanol sensitive and butanol resistant alleles of the GCD1 gene to provide proof of principle that the inherent sensitivity of yeast strains to butanol impacts upon butanol production. Given that the concentrations of butanol that are required to inhibit protein synthesis and growth (1–2 %, 10–20 g/L) are very different to the levels that are produced in our yeast strains (0.3 g/L), it is inherently quite startling that greater levels of butanol are produced in a butanol resistant strain. Our current working hypothesis to explain this discrepancy is that butanol transport across the yeast cell membrane is inefficient. Thus, if extracellular butanol does not pass into a cell and intracellular butanol does not pass out of a cell particularly well, it is possible that the level of extracellular butanol required to inhibit growth and translation would be high, whereas the level of endogenous butanol required to elicit the same effect could be much lower. Indeed, a role for specific efflux pumps in increasing the tolerance of E. coli to exogenously added short chain alcohols has been described . This opens up the possibility of an integrated approach towards improved tolerance to, and hence, improved production of, short chain alcohols in S. cerevisiae, where both intracellular resistance at the level of proteins synthesis and the cells capacity to export alcohols are enhanced.
Yeast growth and strain construction/validation
Strains used in this study were grown at 30 °C on either standard yeast extract/peptone/dextrose media (YPD) or synthetic complete dextrose media (SCD) both supplemented with 2 % glucose . Individual genomic integration and deletion cassettes were generated and transformed into yeast using standard PCR-based integration methods to target the integration cassettes to specific high expression sites in the yeast genome  and validated using PCR, western blotting and phenotypic analysis. The individual cassettes carried yeast codon-optimised sequences Ter (from T. denticola), Cct, Adhe2, Bcd and Hbd (from Clostridium Beijerinckii) with a C-terminal Flag tag (two Flag peptide epitopes) and the CYC1 terminator sequences downstream. Each gene was first inserted into a specific pRS vector with a TDH3 promoter inserted upstream and the auxotrophic marker gene immediately downstream of the cassette. Integration primers were then designed to isolate the cassette upstream of the TDH3 promoter to downstream of the auxotrophic marker (Fig. 1b). The sites of integration were selected based on previous studies analysing the efficiency of gene expression from various sites across the yeast genome. The ADH1 gene was deleted using the ADE2 marker using standard yeast PCR-based gene disruption methods. The codon-optimised ERG10 yeast gene was synthesised downstream of the TDH3 gene promoter and upstream of the CYC1 terminator sequence, and flag epitope tags were placed at the C-terminus. The cassette was sub-cloned into the pFa6-KanMX4 plasmid upstream of the KanMX4 gene. Integration primers were designed to amplify the entire fragment prior to transformation into yeast. Codon-optimised versions of the yeast ALD6 and ACS2 genes were synthesised downstream of the TDH3 and TEF1 gene promoters and upstream of the CYC1 and ADH2 terminator sequences, respectively. Flag epitope tags were placed at the C-terminus of each cassette and a hygromycin marker gene was added (Fig. 5a). The whole cassette was bounded by 200 n sequences directing it to the TRP1 locus in the yeast genome. Finally, the cassette was flanked by sites for the type IIS restriction enzyme, BspQI, such that the whole fragment could be released and transformed into yeast. All commercial DNA synthesis was carried out by either Mr Gene GmbH (Regensburg, Germany) or GenScript (Piscataway, NJ).
Measurements of butanol and ethanol
Strains were grown in liquid YPD media from a starting OD600 of 0.1 using semi-anaerobic 50 ml vials over a 21-day period. On specific days, 2 ml samples were taken, passed through a 0.22 μ filter into gas chromatography (GC) vials and analysed by GC-FID using an Agilent 6850A GC system with an Agilent 4513A automatic injector, sampler and controller (Agilent technologies Ltd., Stockport, UK). A J&W DB-WAX capillary column (30 m × 0.25 mm, 0.25 μM) (Agilent technologies Ltd.) was used for separation. Samples were quantified relative to standards of ethanol and butanol.
GC-MS was carried out using media from anaerobic cultures grown in YPD for 5 days. Using a 6890 N GC system coupled to a 5973 Mass Selective Detector (MSD) (Agilent technologies Ltd.), 2 ml samples were collected and analysed. Data was analysed and processed using the MSD ChemStation software (Agilent technologies Ltd.).
Western blot analysis of Flag-tagged proteins
Yeast culture (5 ml) were grown to an OD600 of 0.7 in YPD, pelleted; then protein samples were prepared and processed for electrophoresis and immunoblot analysis as described previously . A monoclonal anti-Flag antibody (Sigma-Aldrich, Dorset, UK) was used as the primary antibody for the detection.
This work was supported by a University of Manchester Intellectual Property-Proof of Principle (UMIP-PoP) award and a Biotechnology and Biological Sciences Research Council (BBSRC) grant (BB/K002767/1). RS was initially supported by a Syria Ministry of Higher Education Capacity Building Project studentship, then subsequently by funds from the British Council and HZA was supported by a BBSRC DTP studentship.
- Kerr RA. Climate change. Global warming is changing the world. Science. 2007;316:188–90.View ArticleGoogle Scholar
- Peralta-Yahya PP, Zhang F, del Cardayre SB, Keasling JD. Microbial engineering for the production of advanced biofuels. Nature. 2012;488:320–8.View ArticleGoogle Scholar
- Laluce C, Schenberg AC, Gallardo JC, Coradello LF, Pombeiro-Sponchiado SR. Advances and developments in strategies to improve strains of Saccharomyces cerevisiae and processes to obtain the lignocellulosic ethanol—a review. Appl Biochem Biotechnol. 2012;166:1908–26.View ArticleGoogle Scholar
- Cascone R. Biobutanol—a replacement for bioethanol? Chem Eng Prog. 2008;104:S4–9.Google Scholar
- Zheng YN, Li LZ, Xian M, Ma YJ, Yang JM, Xu X, et al. Problems with the microbial production of butanol. J Ind Microbiol Biotechnol. 2009;36:1127–38.View ArticleGoogle Scholar
- Bud R. The uses of life: a history of biotechnology. Cambridge: Cambridge University Press; 1993.Google Scholar
- Jang YS, Lee J, Malaviya A, Seung do Y, Cho JH, Lee SY. Butanol production from renewable biomass: rediscovery of metabolic pathways and metabolic engineering. Biotechnol J. 2012;7:186–98.View ArticleGoogle Scholar
- Xue C, Zhao XQ, Liu CG, Chen LJ, Bai FW. Prospective and development of butanol as an advanced biofuel. Biotechnol Adv. 2013;31:1575–84.View ArticleGoogle Scholar
- Hong KK, Nielsen J. Metabolic engineering of Saccharomyces cerevisiae: a key cell factory platform for future biorefineries. Cell Mol Life Sci. 2012;69:2671–90.View ArticleGoogle Scholar
- Lan EI, Liao JC. Microbial synthesis of n-butanol, isobutanol, and other higher alcohols from diverse resources. Bioresour Technol. 2013;135:339–49.View ArticleGoogle Scholar
- Bond-Watts BB, Bellerose RJ, Chang MC. Enzyme mechanism as a kinetic control element for designing synthetic biofuel pathways. Nat Chem Biol. 2011;7:222–7.View ArticleGoogle Scholar
- Shen CR, Lan EI, Dekishima Y, Baez A, Cho KM, Liao JC. Driving forces enable high-titer anaerobic 1-butanol synthesis in Escherichia coli. Appl Environ Microbiol. 2011;77:2905–15.View ArticleGoogle Scholar
- Huffer S, Roche CM, Blanch HW, Clark DS. Escherichia coli for biofuel production: bridging the gap from promise to practice. Trends Biotechnol. 2012;30:538–45.View ArticleGoogle Scholar
- Generoso WC, Schadeweg V, Oreb M, Boles E. Metabolic engineering of Saccharomyces cerevisiae for production of butanol isomers. Curr Opin Biotechnol. 2014;33C:1–7.Google Scholar
- Steen EJ, Chan R, Prasad N, Myers S, Petzold CJ, Redding A, et al. Metabolic engineering of Saccharomyces cerevisiae for the production of n-butanol. Microb Cell Fact. 2008;7:36.View ArticleGoogle Scholar
- Sakuragi H, Morisaka H, Kuroda K, Ueda M. Enhanced butanol production by eukaryotic Saccharomyces cerevisiae engineered to contain an improved pathway. Biosci Biotechnol Biochem. 2015;79:314–20.View ArticleGoogle Scholar
- Krivoruchko A, Serrano-Amatriain C, Chen Y, Siewers V, Nielsen J. Improving biobutanol production in engineered Saccharomyces cerevisiae by manipulation of acetyl-CoA metabolism. J Ind Microbiol Biotechnol. 2013;40:1051–6.View ArticleGoogle Scholar
- Si T, Luo Y, Xiao H, Zhao H. Utilizing an endogenous pathway for 1-butanol production in Saccharomyces cerevisiae. Metab Eng. 2014;22:60–8.View ArticleGoogle Scholar
- Branduardi P, Longo V, Berterame NM, Rossi G, Porro D. A novel pathway to produce butanol and isobutanol in Saccharomyces cerevisiae. Biotechnol Biofuels. 2013;6:68.View ArticleGoogle Scholar
- Ashe MP, Slaven JW, De Long SK, Ibrahimo S, Sachs AB. A novel eIF2B-dependent mechanism of translational control in yeast as a response to fusel alcohols. Embo J. 2001;20:6464–74.View ArticleGoogle Scholar
- Taylor EJ, Campbell SG, Griffiths CD, Reid PJ, Slaven JW, Harrison RJ, et al. Fusel alcohols regulate translation initiation by inhibiting eIF2B to reduce ternary complex in a mechanism that may involve altering the integrity and dynamics of the eIF2B body. Mol Biol Cell. 2010;21:2202–16.View ArticleGoogle Scholar
- Smirnova JB, Selley JN, Sanchez-Cabo F, Carroll K, Eddy AA, McCarthy JE, et al. Global gene expression profiling reveals widespread yet distinctive translational responses to different eukaryotic translation initiation factor 2B-targeting stress pathways. Mol Cell Biol. 2005;25:9340–9.View ArticleGoogle Scholar
- Egbe NE, Paget CM, Wang H, Ashe MP. Alcohols inhibit translation to regulate morphogenesis in C. albicans. Fungal Genet Biol. 2015;77:50–60.View ArticleGoogle Scholar
- Pavitt GD. eIF2B, a mediator of general and gene-specific translational control. Biochem Soc Trans. 2005;33:1487–92.View ArticleGoogle Scholar
- Flagfeldt DB, Siewers V, Huang L, Nielsen J. Characterization of chromosomal integration sites for heterologous gene expression in Saccharomyces cerevisiae. Yeast. 2009;26:545–51.View ArticleGoogle Scholar
- Paquin CE, Williamson VM. Ty insertions at two loci account for most of the spontaneous antimycin A resistance mutations during growth at 15 degrees C of Saccharomyces cerevisiae strains lacking ADH1. Mol Cell Biol. 1986;6:70–9.Google Scholar
- Slavov N, Botstein D. Coupling among growth rate response, metabolic cycle, and cell division cycle in yeast. Mol Biol Cell. 2011;22:1997–2009.View ArticleGoogle Scholar
- Dunlop MJ, Dossani ZY, Szmidt HL, Chu HC, Lee TS, Keasling JD, et al. Engineering microbial biofuel tolerance and export using efflux pumps. Mol Syst Biol. 2011;7:487.View ArticleGoogle Scholar
- Guthrie C, Fink GR. Guide to yeast genetics and molecular biology. San Diego, California: Academic; 1991.Google Scholar
- Ashe MP, De Long SK, Sachs AB. Glucose depletion rapidly inhibits translation initiation in yeast. Mol Biol Cell. 2000;11:833–48.View ArticleGoogle Scholar
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.