Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Laboratory evolution for forced glucose-xylose co-consumption enables identification of mutations that improve mixed-sugar fermentation by xylose-fermenting Saccharomyces cerevisiae

Laboratory evolution for forced glucose-xylose co-consumption enables identification of mutations... Abstract Simultaneous fermentation of glucose and xylose can contribute to improved productivity and robustness of yeast-based processes for bioethanol production from lignocellulosic hydrolysates. This study explores a novel laboratory evolution strategy for identifying mutations that contribute to simultaneous utilisation of these sugars in batch cultures of Saccharomyces cerevisiae. To force simultaneous utilisation of xylose and glucose, the genes encoding glucose-6-phosphate isomerase (PGI1) and ribulose-5-phosphate epimerase (RPE1) were deleted in a xylose-isomerase-based xylose-fermenting strain with a modified oxidative pentose-phosphate pathway. Laboratory evolution of this strain in serial batch cultures on glucose–xylose mixtures yielded mutants that rapidly co-consumed the two sugars. Whole-genome sequencing of evolved strains identified mutations in HXK2, RSP5 and GAL83, whose introduction into a non-evolved xylose-fermenting S. cerevisiae strain improved co-consumption of xylose and glucose under aerobic and anaerobic conditions. Combined deletion of HXK2 and introduction of a GAL83G673T allele yielded a strain with a 2.5-fold higher xylose and glucose co-consumption ratio than its xylose-fermenting parental strain. These two modifications decreased the time required for full sugar conversion in anaerobic bioreactor batch cultures, grown on 20 g L−1 glucose and 10 g L−1 xylose, by over 24 h. This study demonstrates that laboratory evolution and genome resequencing of microbial strains engineered for forced co-consumption is a powerful approach for studying and improving simultaneous conversion of mixed substrates. biofuels, yeast, industrial biotechnology, redox engineering, fermentation, pentoses INTRODUCTION Industrial biotechnology can contribute to reconciliating global demands for liquid transport fuels with the almost universally accepted need to limit anthropogenic CO2 emissions (Fulton et al.2015). Bioethanol, the biofuel with the largest annual global production volume, is still predominantly produced by fermentation of sucrose or glucose, derived from sugar cane or corn starch, respectively (Gombert and van Maris 2015). These ‘first-generation’ bioethanol processes exploit the natural high fermentation rates and ethanol yield of the yeast Saccharomyces cerevisiae. Optimisation of yeast strains and production processes enables many industrial processes to operate at >90% of the theoretical ethanol yield on sugar (Basso, Basso and Nitsche Rocha 2011; Gombert and van Maris 2015; Lopes et al.2016). However, the massive scaling up of production volumes that would be required to replace a substantial fraction of petroleum-based transport fuels cannot be sustainably achieved with corn starch and cane sugar as the only feedstocks. Instead, a large fraction of the feedstock will have to be derived from lignocellulosic plant biomass, such as agricultural residues and energy crops (Lynd et al.2017). In comparison with first-generation feedstocks, lignocellulosic hydrolysates pose additional challenges for yeasts and yeast researchers. In addition to containing mixtures of hexose and pentose (mainly D-xylose and L-arabinose) sugars, the deconstruction of lignocellulosic biomass that precedes yeast-based fermentation releases fermentation inhibitors (Palmqvist and Hahn-Hägerdal 2000; van Maris et al.2006; Jansen et al.2017). Intensive metabolic engineering studies, encompassing functional expression of heterologous pathways for xylose and arabinose catabolism, improvements in inhibitor tolerance and minimisation of by-product formation have yielded S. cerevisiae strains that are now applied in the first full-scale ‘second-generation’ industrial bioethanol plants (Moysés et al.2016; Jansen et al.2017). However, further improvements in ethanol titres, yields and productivities are important to increase the economic viability of this nascent technology. Current strain engineering strategies for enabling pentose fermentation by S. cerevisiae typically yield strains that, in anaerobic batch cultures grown on sugar mixtures, preferentially ferment glucose, while xylose and/or arabinose are predominantly converted in a second, slower fermentation phase (Jansen et al.2017). This strong preference for glucose over pentoses persists even after extensive laboratory evolution on sugar mixtures (Kuyper et al.2005b; Wisselink et al.2009; Garcia Sanchez et al.2010; Shen et al.2012; Zhou et al.2012). Achieving efficient co-fermentation of glucose and pentoses, while maintaining a high overall rate of sugar conversion, could increase volumetric productivity of industrial processes. Moreover, since several inhibitors of yeast performance are more harmful during the slower pentose fermentation phase (Bellissimi et al.2009; Kim et al.2012; Ask et al.2013; Jansen et al.2017), simultaneous fermentation of glucose and pentose sugars can also contribute to robustness under industrial process conditions. Random mutagenesis, laboratory evolution and protein engineering of xylose transporters have yielded transporter variants with improved xylose affinity and reduced glucose inhibition, which enabled the construction of yeast strains with improved xylose consumption in the presence of glucose (Farwick et al.2014; Nijland et al.2014; Reznicek et al.2015; Shin et al.2015). In an alternative approach, expression of a heterologous cellodextrin transporter and an intracellular β-glucosidase, along with a heterologous xylose reductase/xylitol dehydrogenase (XR/XDH) pathway, enabled simultaneous consumption of cellobiose and xylose in S. cerevisiae by reducing the impact of glucose repression (Wei et al.2015). However, despite progress in this area, engineering of yeast strains showing simultaneous, fast fermentation of glucose and xylose remains a key challenge. Deletion of RPE1, which encodes the pentose-phosphate-pathway (PPP) enzyme ribulose-5-phosphate epimerase, was recently shown to result in coupling of glucose and xylose catabolism, at a ratio of 10:1, in an engineered xylose-utilising S. cerevisiae strain (Shen et al.2015). Despite a low xylose-to-glucose consumption ratio, this strategy indicated the potential of forced stoichiometric coupling of glucose and pentose metabolism in S. cerevisiae. In S. cerevisiae, phosphoglucose isomerase (Pgi1) catalyses interconversion of glucose-6-phosphate to fructose-6-phosphate in upper glycolysis (Aguilera 1986). Since deletion of PGI1 blocks glycolysis, pgi1Δ strains cannot grow on glucose as the sole carbon source unless all glucose-6-phosphate is rerouted through the PPP. Deletion of eda, rpe and pgi in Escherichia coli was previously shown to enable co-consumption of xylose and glucose (Gawand et al.2013). As conversion of 1 mol glucose-6-phosphate to 1 mol ribulose-5-phosphate via the oxidative branch of the PPP results in a net generation of 2 mol NADPH, the absence of a redox imbalance relied on conversion of excess NADPH to NADH by the native E. coli transhydrogenases. Wild-type S. cerevisiae strains cannot reoxidise all NADPH generated by such a redirection of metabolism and, consequently, pgi1-null mutants cannot grow on glucose as sole carbon source (Boles, Lehnert and Zimmermann 1993; Dickinson, Sobanski and Hewlins 1995). Overexpression of GDH2, which encodes NAD+-dependent glutamate dehydrogenase, can restore growth of pgi1Δ strains by enabling a transhydrogenase-like cycle that couples the interconversion of 2-oxoglutarate and glutamate to the conversion of NADPH and NAD+ to NADP+ and NADH (Boles, Lehnert and Zimmermann 1993). Based on the impact of a pgi1Δ mutation on glucose metabolism, we reasoned that inactivation of PGI1 might be used to construct strains with a stringent requirement for co-utilisation of xylose and glucose, at much higher ratios than hitherto demonstrated. The goal of this study was to explore a new strategy for identifying mutations that stimulate glucose-xylose co-consumption by S. cerevisiae. The strategy was based on enforcing a strict stoichiometric coupling of glucose and xylose fermentation by the combined deletion of RPE1 and PGI1 in an engineered, xylose-isomerase-based S. cerevisiae strain (Kuyper et al.2005a; Verhoeven et al.2017). Furthermore, to reduce the impact of these modifications on NADP+/NADPH redox cofactor balancing, the native NADP+-dependent 6-phosphogluconate dehydrogenases (Gnd1 and Gnd2) were replaced by a heterologous NAD+-dependent enzyme (Papapetridis et al.2016). After laboratory evolution for improved growth on glucose-xylose mixtures, the physiology of evolved strains was analysed in aerobic shake-flask and bioreactor batch cultures. Potential causal mutations identified by whole-genome sequencing were introduced into a non-evolved (PGI1 RPE1 GND1 GND2) xylose-fermenting strain background. The resulting reverse engineered strains were then analysed in shake-flask and anaerobic bioreactor batch cultures, grown on mixtures of glucose and xylose. MATERIALS AND METHODS Maintenance of strains The CEN.PK lineage of S. cerevisiae laboratory strains (Entian and Kötter 2007) was used to construct and evolve all strains used in this study (Table 1, Additional File 1). Depending on strain auxotrophies, cultures were grown in YP (10 g L−1 yeast extract, 20 g L−1 peptone) (BD, Franklin Lakes, NJ) or synthetic medium (SM) (Verduyn et al.1992), supplemented with glucose (20 g L−1), xylose (20 g L−1), a glucose/xylose mixture (10 g L−1 of each sugar) or a xylose/fructose/glucose mixture (20, 10 and 1 g L−1 respectively). Propagation of E. coli XL-1 Blue cultures was performed in LB medium (5 g L−1 Bacto yeast extract, 10 g L−1 Bacto tryptone, 5 g L−1 NaCl, 100 μg mL−1 ampicillin). Frozen stock cultures were stored at –80°C, after addition of glycerol (30% v/v final concentration). Table 1. Strains used in this study. Strain name . Relevant Genotype . Origin . CEN.PK113–7D MATa MAL2–8c SUC2 CAN1 Entian and Kötter (2007) IMX581 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2 Mans et al. (2015) IMX705 MATa MAL2–8c SUC2 can1::cas9-natNT2 gnd2Δ gnd1::gndA Papapetridis et al. (2016) IMX1046 MATa MAL2–8c SUC2 can1::cas9-natNT2 gnd2Δ gnd1::gndA gre3::ZWF1, SOL3, TKL1, TAL1, NQM1, RKI1, TKL2 sga1::9*xylA, XKS1 rpe1Δ pgi1Δ This work IMX994 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2, gre3::RPE1, TKL1, TAL1, RKI1, XKS1 This work IMU079 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2, gre3::RPE1, TKL1, TAL1, RKI1, XKS1 pAKX002 This work IMX1485 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RRKI1, XKS1 hxk2Δ pAKX002 This work IMX1486 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83Δ pAKX002 This work IMX1487 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 rsp5Δ pAKX002 This work IMX1488 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83::GAL83G673T pAKX002 This work IMX1515 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 rsp5Δ hxk2Δ pAKX002 This work IMX1583 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83::GAL83G673T hxk2Δ pAKX002 This work Strain name . Relevant Genotype . Origin . CEN.PK113–7D MATa MAL2–8c SUC2 CAN1 Entian and Kötter (2007) IMX581 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2 Mans et al. (2015) IMX705 MATa MAL2–8c SUC2 can1::cas9-natNT2 gnd2Δ gnd1::gndA Papapetridis et al. (2016) IMX1046 MATa MAL2–8c SUC2 can1::cas9-natNT2 gnd2Δ gnd1::gndA gre3::ZWF1, SOL3, TKL1, TAL1, NQM1, RKI1, TKL2 sga1::9*xylA, XKS1 rpe1Δ pgi1Δ This work IMX994 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2, gre3::RPE1, TKL1, TAL1, RKI1, XKS1 This work IMU079 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2, gre3::RPE1, TKL1, TAL1, RKI1, XKS1 pAKX002 This work IMX1485 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RRKI1, XKS1 hxk2Δ pAKX002 This work IMX1486 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83Δ pAKX002 This work IMX1487 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 rsp5Δ pAKX002 This work IMX1488 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83::GAL83G673T pAKX002 This work IMX1515 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 rsp5Δ hxk2Δ pAKX002 This work IMX1583 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83::GAL83G673T hxk2Δ pAKX002 This work Open in new tab Table 1. Strains used in this study. Strain name . Relevant Genotype . Origin . CEN.PK113–7D MATa MAL2–8c SUC2 CAN1 Entian and Kötter (2007) IMX581 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2 Mans et al. (2015) IMX705 MATa MAL2–8c SUC2 can1::cas9-natNT2 gnd2Δ gnd1::gndA Papapetridis et al. (2016) IMX1046 MATa MAL2–8c SUC2 can1::cas9-natNT2 gnd2Δ gnd1::gndA gre3::ZWF1, SOL3, TKL1, TAL1, NQM1, RKI1, TKL2 sga1::9*xylA, XKS1 rpe1Δ pgi1Δ This work IMX994 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2, gre3::RPE1, TKL1, TAL1, RKI1, XKS1 This work IMU079 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2, gre3::RPE1, TKL1, TAL1, RKI1, XKS1 pAKX002 This work IMX1485 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RRKI1, XKS1 hxk2Δ pAKX002 This work IMX1486 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83Δ pAKX002 This work IMX1487 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 rsp5Δ pAKX002 This work IMX1488 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83::GAL83G673T pAKX002 This work IMX1515 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 rsp5Δ hxk2Δ pAKX002 This work IMX1583 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83::GAL83G673T hxk2Δ pAKX002 This work Strain name . Relevant Genotype . Origin . CEN.PK113–7D MATa MAL2–8c SUC2 CAN1 Entian and Kötter (2007) IMX581 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2 Mans et al. (2015) IMX705 MATa MAL2–8c SUC2 can1::cas9-natNT2 gnd2Δ gnd1::gndA Papapetridis et al. (2016) IMX1046 MATa MAL2–8c SUC2 can1::cas9-natNT2 gnd2Δ gnd1::gndA gre3::ZWF1, SOL3, TKL1, TAL1, NQM1, RKI1, TKL2 sga1::9*xylA, XKS1 rpe1Δ pgi1Δ This work IMX994 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2, gre3::RPE1, TKL1, TAL1, RKI1, XKS1 This work IMU079 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2, gre3::RPE1, TKL1, TAL1, RKI1, XKS1 pAKX002 This work IMX1485 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RRKI1, XKS1 hxk2Δ pAKX002 This work IMX1486 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83Δ pAKX002 This work IMX1487 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 rsp5Δ pAKX002 This work IMX1488 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83::GAL83G673T pAKX002 This work IMX1515 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 rsp5Δ hxk2Δ pAKX002 This work IMX1583 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83::GAL83G673T hxk2Δ pAKX002 This work Open in new tab Construction of plasmids and cassettes PCR amplification for construction of plasmid fragments and yeast integration cassettes was performed with Phusion High Fidelity DNA Polymerase (Thermo-Scientific, Waltham, MA), according to the manufacturer's guidelines. Plasmid assembly was performed in vitro with a Gibson Assembly Cloning kit (New England Biolabs, Ipswich, MA), following the supplier's guidelines, or in vivo by transformation of plasmid fragments into yeast cells (Kuijpers et al.2013). For all constructs, correct assembly was confirmed by diagnostic PCR with DreamTaq polymerase (Thermo-Scientific), following the manufacturer's protocol. Plasmids used and constructed in this work are described in Additional File 2. All yeast genetic modifications were performed using CRISPR/Cas9-based genome editing (Mans et al.2015). Unique guide-RNA (gRNA) sequences targeting GRE3, GAL83 and RSP5 were selected from a publicly available list (DiCarlo et al.2013) and synthesised (Baseclear, Leiden, The Netherlands). Primers and oligonucleotides used in this work are listed in Additional File 3. To construct the GRE3-targeting CRISPR-plasmid pUDR204, the plasmid backbone of pMEL11 was PCR amplified using primer combination 5980/5792. The insert fragment, expressing the GRE3-targeting gRNA, was amplified using primer combination 5979/5978 and pMEL11 as template. To construct the RPE1/PGI1 double-targeting CRISPR-plasmid pUDR202, the plasmid backbone and the insert fragment were PCR amplified using primer combinations 5941/6005 and 9269/9401, respectively, using pROS11 as template. Both plasmids were assembled in vitro and cloned in E. coli. To construct CRISPR-plasmids for single deletion of GAL83 and RSP5, the plasmid backbone, the GAL83-gRNA insert and the RSP5-gRNA insert were amplified using primer combination 5792/5980, 5979/11 270 and 5979/11 373, respectively, using pMEL10 as template and assembled in vivo. To generate ZWF1 and SOL3 overexpression cassettes, promoter regions of ADH1 and ENO1 and the coding regions of ZWF1 and SOL3 (including their terminator regions) were PCR amplified using primer combinations 8956/8960, 8958/8961, 8953/8964 and 8984/8986, respectively, using CEN.PK113–7D genomic DNA as a template. The resulting products were used as templates for fusion-PCR assembly of the pADH1-ZWF1-tZWF1 and pENO1-SOL3-tSOL3 overexpression cassettes with primer combinations 8956/8964 and 8958/8986, respectively, which yielded plasmids pUD426 and pUD427 after ligation to pJET-blunt vectors (Thermo-Scientific) and cloning in E. coli. To generate yeast-integration cassettes for overexpression of the major genes of the complete PPP, pADH1-ZWF1-tZWF1, pENO1-SOL3-tSOL3, pPGK1-TKL1-tTKL1, pTEF1-TAL1-tTAL1, pPGI1-NQM1-tNQM1, pTPI1-RKI1-tRKI1 and pPYK1-TKL2-tTKL2 cassettes were PCR amplified using primer combinations 4870/7369, 8958/3290, 3291/4068, 3274/3275, 3847/3276, 4691/3277 and 3283/3288, respectively, using plasmids pUD426, pUD427, pUD348, pUD349, pUD344, pUD345 and pUD346, respectively, as templates. To generate yeast-integration cassettes of the genes of the non-oxidative PPP, the pTDH3-RPE1-tRPE1, pPGK1-TKL1-tTKL1, pTEF1-TAL1-tTAL1, pTPI1-RKI1-tRKI1 overexpression cassettes were PCR-amplified using primer pairs 7133/3290, 3291/4068, 3724/3725 and 10 460/10 461, respectively and plasmids pUD347, pUD348, pUD34 and pUD345 as templates. Yeast-integration cassettes for overexpression of Piromyces sp. xylose isomerase (pTPI1-xylA-tCYC1) were PCR-amplified using primer combinations 6285/7548, 6280/6273, 6281/6270, 6282/6271, 6284/6272, 6283/6275, 6287/6276, 6288/6277 or 6289/6274, using pUD350 as template. Yeast xylulokinase overexpression cassettes (pTEF1-XKS1-tXKS1) were PCR-amplified from plasmid pUD353, using primer combination 5920/9029 or 7135/7222. A yeast-integration cassette of pGAL83-gal83::GAL83G673T-tGAL83 was PCR-amplified from genomic DNA of IMS0629, using primer combination 11 273/11 274. Strain construction Yeast transformation was performed as previously described (Gietz and Woods 2002). Transformation mixtures were plated on SM or YP agar plates (2% Bacto Agar, BD), supplemented with the appropriate carbon sources. For transformations with the amdS marker cassette, agar plates were prepared and counter selection was performed as previously described (Solis-Escalante et al.2013). For transformations with the URA3 selection marker counterselection was performed using 5-fluoro-orotic acid (Zymo Research, Irvine, CA), following the supplier's protocol. For transformations with the hphNT marker, agar plates were additionally supplemented with 200 mg L−1 hygromycin B (Invivogen, San Diego, CA) and plasmid loss was induced by cultivation in non-selective medium. After each transformation, correct genotypes were confirmed by diagnostic PCR using DreamTaq polymerase (Thermo-Scientific, see Additional File 3 for primer sequences). Co-transformation of pUDR204 along with the pADH1-ZWF1-tZWF1, pENO1-SOL3-tSOL3, pPGK1-TKL1-tTKL1, pTEF1-TAL1-tTAL1, pPGI1-NQM1-tNQM1, pTPI1-RKI1-tRKI1 and pPYK1-TKL2-tTKL2 integration cassettes to IMX705 (Papapetridis et al.2016) and subsequent plasmid counterselection yielded strain IMX963, which overexpresses the major enzymes of the PPP. Co-transformation of pUDR119, 9 copies of the pTPI1-xylA-tCYC1 integration cassette, along with a single copy of the pTEF1-XKS1-tXKS1cassette, to IMX963 followed by plasmid counterselection yielded the xylose-fermenting strain IMX990. In IMX990, the pTPI1-xylA-tCYC1 cassettes recombined in vivo to form a multicopy construct of xylose isomerase overexpression (Verhoeven et al.2017). To construct IMX1046, in which RPE1 and PGI1 were deleted, plasmid pUDR202 and the repair oligonucleotides 9279/9280/9281/9282 were co-transformed to IMX990. Transformation mixes of IMX1046 were plated on SM agar supplemented with a xylose/fructose/glucose mixture (20, 10 and 1 g L−1 final concentrations respectively), to avoid potential glucose toxicity (Boles, Lehnert and Zimmermann 1993). To construct strain IMX994, plasmid pUDE335 was co-transformed to IMX581, along with the pTDH3-RPE1-tRPE1, pPGK1-TKL1-tTKL1, pTEF1-TAL1-tTAL1, pTPI1-RKI1-tRKI1 and pTEF1-XKS1-tXKS1 integration cassettes, after which the CRISPR plasmid was recycled. Transformation of pAKX002 to IMX994 yielded the xylose-fermenting strain IMU079. Co-transformation of pUDE327 along with the repair oligonucleotides 5888/5889 to IMX994 yielded strain IMX1384, in which HXK2 was deleted. Co-transformation of the pMEL10 backbone fragment, along with the GAL83-gRNA insert or the RSP5-gRNA insert and repair oligonucleotides 11 271/11 272 or 11 374/11 375, respectively, yielded strains IMX1385 (GAL83 deletion) and IMX1442 (RSP5 deletion). Counterselection of the CRISPR plasmids from IMX1384, IMX1385 and IMX1442 yielded, respectively, strains IMX1408, IMX1409 and IMX1451. Transformation of pAKX002 to IMX1408, IMX1409 and IMX1451 yielded the xylose-fermenting strains IMX1485, IMX1486 and IMX1487, respectively. To construct strain IMX1453, in which the mutated GAL83G673T gene replaced the wild-type GAL83 allele, plasmid pUDR105 was co-transformed to IMX1409 with the pGAL83-gal83::GAL83G673T-tGAL83 cassette. Transformation of pAKX002 to IMX1453 yielded the xylose-fermenting strain IMX1488. To construct the hxk2Δ rsp5Δ strain IMX1484, plasmid pUDE327 was co-transformed to IMX1451, along with the repair oligonucleotides 5888/5889. Counterselection of pUDE327 from IMX1484 yielded strain IMX1510. Transformation of pAKX002 to IMX1510 yielded the xylose-fermenting strain IMX1515. To construct the hxk2Δ gal83::GAL83G673T strain IMX1563, plasmid pUDE327 along with the repair oligonucleotides 5888/5889 was co-transformed to IMX1453. Counterselection of pUDE327 from IMX1563 yielded IMX1571. The xylose-fermenting strain IMX1583 was obtained by transformation of pAKX002 to IMX1571. Cultivation and media Shake-flask growth experiments were performed in 500-mL conical shake flasks containing 100 mL of SM with urea as nitrogen source (2.3 g L−1 urea, 6.6 g L−1 K2SO4, 3 g L−1 KH2PO4, 1 mL L−1 trace elements solution (Verduyn et al.1992) and 1 mL L−1 vitamin solution (Verduyn et al.1992)) to prevent medium acidification. The initial pH of the medium was set to 6.0 by titration with 2 mol L−1 KOH. Depending on the strains grown, different mixtures of carbon sources (glucose/xylose/fructose) were added and media were filter-sterilised (0.2 μm, Merck, Darmstadt, Germany). The temperature was set to 30°C and the shaking speed to 200 rpm in an Innova incubator (New Brunswick Scientific, Edison, NJ). In each case, pre-culture shake flasks were inoculated from frozen stocks. After 8–12 h of growth, exponentially growing cells from the initial shake flasks were used to inoculate fresh cultures that, after 12–18 h of growth, were used as inoculum for the growth experiments to a starting OD660 of 0.4–0.5 in the case of shake-flask growth experiments and of 0.2–0.3 in the case of bioreactor cultivation. Bioreactor cultures were grown on SM (Verduyn et al.1992), supplemented with a glucose/xylose mixture (10 g L−1/20 g L−1 for aerobic cultivation or 20 g L−1/10 g L−1 for anaerobic cultivation). Sterilisation of the salt solution was performed by autoclaving at 121°C for 20 min. Sugar solutions were sterilised separately by autoclaving at 110°C for 20 min and added to the sterile salt media along with filter-sterilised vitamin solutions. In the case of anaerobic cultivation, media were additionally supplemented with ergosterol (10 mg L−1) and Tween 80 (420 mg L−1). Sterile antifoam C (0.2 g L−1; Sigma-Aldrich, St. Louis, MO) was added to all media used for bioreactor cultivation. Batch cultures were grown in 2-L bioreactors (Applikon, Delft, The Netherlands) with a 1-L working volume, stirred at 800 rpm. Culture pH was maintained at 5.0 by automatic titration with 2 mol L−1 KOH. Temperature was maintained at 30°C. Bioreactors were sparged at 0.5 L min−1 with either pressurised air (aerobic cultivation) or nitrogen gas (<10 ppm oxygen, anaerobic cultivation). All reactors were equipped with Viton O-rings and Norprene tubing to minimise oxygen diffusion. Evaporation in bioreactor cultures was minimised by cooling the offgas outlet to 4°C. Laboratory evolution Laboratory evolution of strain IMX1046 was performed via serial shake-flask cultivation on SM (Verduyn et al.1992). Cultures were grown in 500-mL shake flasks with 100 mL working volume. Growth conditions were the same as described above. Initially, the cultures were grown on a glucose/xylose concentration ratio of 1.0 g L−1/20 g L−1. After growth was observed, exponentially growing cells (0.05 mL of culture) were transferred to SM with a glucose/xylose concentration ratio of 2.0 g L−1/20 g L−1. During subsequent serial transfers, the glucose content was progressively increased as high growth rates were established at each sugar composition, reaching a final glucose/xylose ratio of 20 g L−1/20 g L−1, to increase the selective pressure for alleviation of glucose repression and/or inhibition of xylose conversion. At that point three single colonies were isolated from two replicate evolution experiments (IMS0628–630 and IMS0634–636, respectively) by plating on SM with 10 g L−1 glucose and 20 g L−1 xylose. This medium composition supported fast growth on plates during three consecutive restreaks. Analytical methods Offgas analysis, biomass dry weight measurements, HPLC analysis of culture supernatants and correction for ethanol evaporation in bioreactor experiments were performed as previously described (Papapetridis et al.2016). Determination of optical density was performed at 660 nm using a Jenway 7200 spectrophotometer (Cole-Palmer, Staffordshire, UK). Yields of products and biomass-specific sugar uptake rates in bioreactor batch cultures were determined as previously described (Wisselink et al.2009; Papapetridis et al.2016). All values are represented as averages ± mean deviation of independent biological duplicate cultures. In silico determination of sugar uptake The Yeast v7.6 consensus metabolic model (Aung, Henry and Walker 2013) was used for in silico prediction of relative xylose and glucose consumption rates in aerobic bioreactor batch cultures of strain IMS0629. The COBRA v2 toolbox (Becker et al.2007) was used to read the model in MATLAB vR2017b (Mathworks, Natick, MA), supported by the SBML Toolbox v4.1 and the libSBML v5.12 (Hucka et al.2003). The Gurobi v6.5 linear programming solver (Gurobi Optimization Inc, Houston, TX) was installed and used according to the manual provided. The MATLAB script is provided in Additional File 4. Genome sequencing Genomic DNA of strains IMS0629 and IMS0634 was isolated from exponentially growing shake-flask cultures on SM (10 g L−1 glucose/20 g L−1 xylose) with a Qiagen Blood & Cell culture DNA kit (Qiagen, Germantown, MD), according to the manufacturer's specifications. Whole-genome sequencing was performed on an Illumina HiSeq PE150 sequencer (Novogene Company Limited, Hong Kong), as previously described (Papapetridis et al.2018). Sequence data were mapped to the reference CEN.PK113–7D genome (Nijkamp et al.2012), to which the sequences of the pTPI1-gndA-tCYC1 and pTPI1-xylA-tCYC1 cassettes were manually added. Data processing and chromosome copy number analysis were carried as previously described (Papapetridis et al.2018). RESULTS Design of an S. cerevisiae strain with a forced, high stoichiometry of xylose and glucose co-consumption Design of an S. cerevisiae strain whose growth depended on extensive co-consumption of xylose and glucose was based on the observation that inactivation of PGI1 blocks entry of glucose-6-phosphate into glycolysis, while inactivation of RPE1 prevents entry of ribulose-5-phosphate into the non-oxidative PPP (Fig. 1). As a consequence, a pgi1Δ rpe1Δ strain is unable to grow on glucose. If conversion of xylose into xylulose-5-phosphate in such a strain is enabled by expression of a heterologous xylose isomerase and overexpression of the native xylulose kinase Xks1 (Kuyper et al.2005a), co-consumption of xylose and glucose should enable growth (Fig. 1). Overexpression of native genes encoding the enzymes of the non-oxidative PPP has previously been shown to stimulate the required conversion of xylulose-5-phosphate into the glycolytic intermediates fructose-6-phosphate and glyceraldehyde-3-phosphate (Fig. 1) (Johansson and Hahn-Hägerdal 2002; Kuyper et al.2005a). The predicted stoichiometry for conversion of glucose and xylose into pyruvate in a yeast strain that combines these genetic modifications is summarised in Equation 1: \begin{eqnarray} {{\rm{C}}_6}{{\rm{H}}_{12}}{{\rm{O}}_6} &+& 2{{\rm{C}}_5}{{\rm{H}}_{10}}{{\rm{O}}_5} + 5{\rm{ADP}} + 5{{\rm{P}}_i} + 2{\rm{NAD}}{{\rm{P}}^ + } + 5{\rm{NA}}{{\rm{D}}^ + } \nonumber\\ &\to & 5{{\rm{C}}_3}{{\rm{H}}_4}{{\rm{O}}_3} + 5{\rm{ATP}} + 4{{\rm{H}}_2}{\rm{O}} + 2{\rm{NADPH}} + 5{\rm{NADH}}\nonumber\\ &+& 7{{\rm{H}}^ + } + {\rm{C}}{{\rm{O}}_2} \end{eqnarray} (1) Figure 1. Open in new tabDownload slide Schematic representation of central carbon metabolism in a yeast strain engineered for forced co-consumption of glucose and xylose. In pgi1Δ rpe1Δ S. cerevisiae expressing a heterologous xylose isomerase (XI, Kuyper et al.2003), the native 6-phosphogluconate dehydrogenases (Gnd1 and Gnd2) were replaced by a bacterial NAD+-dependent enzyme (GndA, Papapetridis et al.2016). Additionally, xylulokinase (Xks1) and enzymes of the pentose phosphate pathway (PPP) were overexpressed. F6P fructose-6-phosphate; G3P glyceraldehyde-3-phosphate; DHAP dihydroxyacetone phosphate; 6PGL 6-phosphogluconolactone; 6PG 6-phosphogluconate; RU5P ribulose-5-phosphate; XU5P xylulose-5-phosphate; R5P ribose-5-phosphate; S7P sedoheptulose-7-phosphate; E4P erythrose-4-phosphate. To prevent a potential excessive formation of NADPH (Boles, Lehnert and Zimmermann 1993; Dickinson, Sobanski and Hewlins 1995), the strain design further included replacement of the native S. cerevisiae NADP+-dependent 6-phosphogluconate dehydrogenases (Gnd1 and Gnd2) by the NAD+-dependent bacterial enzyme GndA (Papapetridis et al.2016), leading to the stoichiometry shown in Equation 2: \begin{eqnarray} {{\rm{C}}_6}{{\rm{H}}_{12}}{{\rm{O}}_6} &+& 2\,{{\rm{C}}_5}{{\rm{H}}_{10}}{{\rm{O}}_5} + 5{\rm{ADP}} + 5{{\rm{P}}_i} + 1{\rm{NAD}}{{\rm{P}}^ + } + 6{\rm{NA}}{{\rm{D}}^ + }\nonumber\\ &\to & 5\,{{\rm{C}}_3}{{\rm{H}}_4}{{\rm{O}}_3} + 5{\rm{ATP}} + 4{{\rm{H}}_2}{\rm{O}} + {\rm{NADPH}} + 6{\rm{NADH}}\nonumber\\ &+& 7{{\rm{H}}^ + } + {\rm{C}}{{\rm{O}}_2} \end{eqnarray} (2) As indicated by Equation 2, this strain design forces co-consumption of 2 mol xylose and 1 mol glucose for the production of 5 mol pyruvate, with a concomitant formation of 1 mol NADPH, 6 mol NADH and 5 mol ATP. NADPH generated in this process can be reoxidised in biosynthetic reactions (Bruinenberg, Van Dijken and Scheffers 1983) or via an L-glutamate-2-oxoglutarate transhydrogenase cycle catalysed by Gdh1 and Gdh2 (Boles, Lehnert and Zimmermann 1993). Actual in vivo stoichiometries of mixed-sugar consumption will depend on the relative contribution of precursors derived from glucose and xylose to biomass synthesis and on the biomass yield (Verduyn et al.1990). In aerobic cultures, the latter strongly depends on the mode of NADH reoxidation (mitochondrial respiration, alcoholic fermentation and/or glycerol production; Bakker et al.2001). While quantitation of precise co-consumption stoichiometries will therefore require experimental analysis, this strain design clearly has the potential to force xylose and glucose co-consumption at much higher stoichiometries than previously reported (Katahira et al.2008; Shen et al.2012, 2015; Gonçalves et al.2014; Nijland et al.2014; Westman et al.2014; Shin et al.2015). Construction, laboratory evolution and growth stoichiometry of glucose-xylose co-consuming S. cerevisiae strains To implement the proposed strain design for forced co-consumption of xylose and glucose, multiple copies of a codon-optimised expression cassette for Piromyces xylA (Verhoeven et al.2017) were integrated into the genome of S. cerevisiae IMX705 (gnd1Δ gnd2Δ gndA; Papapetridis et al.2016), along with overexpression cassettes for S. cerevisiae XKS1 and for structural genes encoding PPP enzymes. Deletion of RPE1 and PGI1 in the resulting xylose-consuming strain IMX990 yielded strain IMX1046, which grew in aerobic shake-flask cultures on SM with 1 g L−1 glucose and 20 g L−1 xylose as sole carbon sources without a requirement for laboratory evolution. However, this strain did not grow at the same xylose concentration when the glucose concentration was increased to 10 g L−1. Kinetic and/or regulatory constraints in glucose-xylose co-consumption at higher glucose concentrations could, for example, reflect interference of glucose with expression or activity of Hxt transporters involved in xylose uptake. To select for co-consumption of xylose at higher glucose concentrations, duplicate serial-transfer experiments were performed in aerobic shake-flask cultures on SM with 20 g L−1 xylose. During serial transfer, the glucose concentration in the medium was gradually increased from 1 to 20 g L−1 (Fig. 2). Samples of the evolving cultures were regularly inoculated in SM containing either 20 g L−1 glucose or 20 g L−1 xylose as sole carbon source. Absence of growth on these single sugars showed that laboratory evolution did not result in an escape from their forced co-consumption. When, after 13 transfers, vigorous growth was observed on a mixture of 20 g L−1 glucose and 20 g L−1 xylose, three single-colony isolates were obtained from each laboratory evolution experiment by streaking on SM agar (10 g L−1 glucose/20 g L−1 xylose). Figure 2. Open in new tabDownload slide Laboratory evolution of S. cerevisiae IMX1046 (pgi1Δ rpe1Δ gnd1Δ gnd2Δ gndA xylA XKS1↑ PPP↑) for improved co-consumption of xylose at high glucose concentrations. Cultures were grown in shake flasks containing 100 mL SM (pH 6) supplemented with 20 g L−1 xylose and progressively increasing glucose concentrations. In every transfer, 0.05 mL of an exponentially growing culture was used to inoculate the next shake flask. Growth studies with the six evolved isolates in shake-flask cultures on SM with 10 g L−1 glucose and 20 g L−1 xylose (Additional File 5) identified isolate IMS0629 (Evolution Line 1) as the fastest growing isolate (μ = 0.21 h−1). The physiology of this strain was further characterised in aerobic bioreactor batch cultures on SM containing 10 g L−1 glucose and 20 g L−1 xylose. After a ca. 10-h lag phase (Fig. 3, Additional Files 6 and 7), exponential growth was observed at a specific growth rate of 0.18 h−1. Biomass, ethanol and CO2 were the main products, with additional minor formation of glycerol and acetate (Table 2, Fig. 3, Additional File 6). During the exponential growth phase, xylose and glucose were co-consumed at a fixed molar ratio of 1.64 mol mol−1 (Table 2, Fig. 3). Growth ceased after glucose depletion, at which point xylose consumption rates drastically decreased and corresponded to a simultaneous low rate of xylitol formation (Fig. 3, Additional File 6). As previously reported for XylA-expressing, xylose-fermenting S. cerevisiae strains (Kuyper et al.2005a; Verhoeven et al.2017), no production of xylitol was observed during the exponential growth phase. The biomass and ethanol yields on total sugars consumed were 0.28 g biomass (g sugar)−1 and 0.18 g ethanol (g sugar)−1, respectively. Together with a respiratory quotient of 1.5, these observations indicated a respiro-fermentative sugar dissimilation. In line with the inability of pgi1Δ S. cerevisiae to generate glucose-6-phosphate from ethanol and acetate, reconsumption of these fermentation products after glucose depletion was not coupled to growth (Fig. 3, Additional File 6). However, their oxidation may have provided redox equivalents for the observed slow production of xylitol from xylose (Fig. 3). Figure 3. Open in new tabDownload slide Sugar consumption, biomass and metabolite production profiles of the evolved S. cerevisiae strain IMS0629 (pgi1Δ rpe1Δ gnd1Δ gnd2Δ gndA xylA XKS1↑ PPP↑), grown on SM with 10 g L−1 glucose and 20 g L−1 xylose in aerobic bioreactor batch cultures (pH 5, 30°C). Cultures were grown in duplicate; the data shown are from a single representative culture. a: filled circles glucose, open circles xylose; b: filled squares biomass, open squares ethanol, filled triangles acetate, open triangles glycerol, open diamonds xylitol; c: ratio of xylose and glucose consumption during exponential growth phase. Table 2. Product yields, biomass-specific sugar uptake and production rates in aerobic bioreactor batch cultures of evolved strain S. cerevisiae IMS0629 (pgi1Δ rpe1Δ gnd1Δ gnd2Δ gndA xylA XKS1↑ PPP↑) on SM supplemented with 10 g L−1 glucose and 20 g L−1 xylose (pH 5, 30°C). Growth rate (h−1) 0.18 ± 0.00 Glucose-xylose consumption ratio mol mol−1 1.64 ± 0.00 Spec. xylose uptake rate mmol (g biomass)−1 h−1 2.52 ± 0.00 Spec. glucose uptake rate mmol (g biomass)−1 h−1 1.54 ± 0.07 Spec. glycerol production rate mmol (g biomass)−1 h−1 0.23 ± 0.01 Spec. ethanol production rate mmol (g biomass)−1 h−1 2.25 ± 0.37 Spec. CO2 production rate mmol (g biomass)−1 h−1 10.43 ± 0.98 Spec. O2 uptake rate mmol (g biomass)−1 h−1 6.87 ± 0.56 Respiratory quotient 1.52 ± 0.03 Biomass yield g biomass (g sugars)−1 0.28 ± 0.00 Ethanol yield g (g sugars)−1 0.18 ± 0.00 Growth rate (h−1) 0.18 ± 0.00 Glucose-xylose consumption ratio mol mol−1 1.64 ± 0.00 Spec. xylose uptake rate mmol (g biomass)−1 h−1 2.52 ± 0.00 Spec. glucose uptake rate mmol (g biomass)−1 h−1 1.54 ± 0.07 Spec. glycerol production rate mmol (g biomass)−1 h−1 0.23 ± 0.01 Spec. ethanol production rate mmol (g biomass)−1 h−1 2.25 ± 0.37 Spec. CO2 production rate mmol (g biomass)−1 h−1 10.43 ± 0.98 Spec. O2 uptake rate mmol (g biomass)−1 h−1 6.87 ± 0.56 Respiratory quotient 1.52 ± 0.03 Biomass yield g biomass (g sugars)−1 0.28 ± 0.00 Ethanol yield g (g sugars)−1 0.18 ± 0.00 Biomass-specific rates, yields and ratios were calculated from samples taken during the mid-exponential growth phase and represent averages ± mean deviation of independent duplicate cultures. Ethanol yield was corrected for evaporation. Open in new tab Table 2. Product yields, biomass-specific sugar uptake and production rates in aerobic bioreactor batch cultures of evolved strain S. cerevisiae IMS0629 (pgi1Δ rpe1Δ gnd1Δ gnd2Δ gndA xylA XKS1↑ PPP↑) on SM supplemented with 10 g L−1 glucose and 20 g L−1 xylose (pH 5, 30°C). Growth rate (h−1) 0.18 ± 0.00 Glucose-xylose consumption ratio mol mol−1 1.64 ± 0.00 Spec. xylose uptake rate mmol (g biomass)−1 h−1 2.52 ± 0.00 Spec. glucose uptake rate mmol (g biomass)−1 h−1 1.54 ± 0.07 Spec. glycerol production rate mmol (g biomass)−1 h−1 0.23 ± 0.01 Spec. ethanol production rate mmol (g biomass)−1 h−1 2.25 ± 0.37 Spec. CO2 production rate mmol (g biomass)−1 h−1 10.43 ± 0.98 Spec. O2 uptake rate mmol (g biomass)−1 h−1 6.87 ± 0.56 Respiratory quotient 1.52 ± 0.03 Biomass yield g biomass (g sugars)−1 0.28 ± 0.00 Ethanol yield g (g sugars)−1 0.18 ± 0.00 Growth rate (h−1) 0.18 ± 0.00 Glucose-xylose consumption ratio mol mol−1 1.64 ± 0.00 Spec. xylose uptake rate mmol (g biomass)−1 h−1 2.52 ± 0.00 Spec. glucose uptake rate mmol (g biomass)−1 h−1 1.54 ± 0.07 Spec. glycerol production rate mmol (g biomass)−1 h−1 0.23 ± 0.01 Spec. ethanol production rate mmol (g biomass)−1 h−1 2.25 ± 0.37 Spec. CO2 production rate mmol (g biomass)−1 h−1 10.43 ± 0.98 Spec. O2 uptake rate mmol (g biomass)−1 h−1 6.87 ± 0.56 Respiratory quotient 1.52 ± 0.03 Biomass yield g biomass (g sugars)−1 0.28 ± 0.00 Ethanol yield g (g sugars)−1 0.18 ± 0.00 Biomass-specific rates, yields and ratios were calculated from samples taken during the mid-exponential growth phase and represent averages ± mean deviation of independent duplicate cultures. Ethanol yield was corrected for evaporation. Open in new tab The quantitative data on biomass and product formation obtained from the bioreactor batch cultures enabled a comparison of the observed molar ratio of xylose and glucose consumption with a model-based prediction. To this end, the engineered metabolic network of strain IMX1046 was re-created in silico, using the Yeast v7.6 consensus metabolic model as a basis (Aung, Henry and Walker 2013; Additional File 4). Consistent with the experimental observations on forced co-consumption, inactivation of either xylose or glucose uptake in the model network did not result in any feasible growth solutions. Using the experimentally determined average specific growth rates and oxygen consumption rates from the aerobic bioreactor batch cultures of strain IMS0629 as constraints on the model resulted in predicted xylose and glucose uptake rates of 2.68 and 1.93 mmol (g biomass)−1 h−1, respectively, corresponding to a molar ratio of the xylose and glucose consumption rates of 1.4. In view of the complexity of the model and the potential impact of differences in biomass composition, this number corresponded well with the experimentally measured value of 1.64 (Table 2). Whole-genome sequencing of evolved glucose-xylose co-consuming S. cerevisiae To identify causal mutations for the improved growth of the evolved glucose-xylose co-consuming S. cerevisiae strains at high glucose concentrations, the genomes of strains IMS0629 and IMS0634 (fastest growing isolates from evolution line 1 and 2, respectively, Additional File 5) were sequenced and compared to that of their common parental strain. Despite the well-documented role of S. cerevisiae hexose transporters in xylose uptake (Reifenberger, Boles and Ciriacy 1997; Hamacher et al.2002; Lee et al.2002; Farwick et al.2014), no mutations were found in the coding region of any of the 18 genes encoding these transporters (HXT1–17 and GAL2), or in other known transporter genes. Both evolved strains harboured mutations in HXK2 (Table 3). This gene encodes the major S. cerevisiae hexokinase which, in addition to its catalytic role, is involved in glucose repression (De Winde et al.1996; Raamsdonk et al.2001). The mutation in IMS0629 caused a premature stop codon at position 309 of Hxk2. Both strains also harboured mutations in RSP5, which encodes an E3-ubiquitin ligase linked to ubiquitination and endocytosis of membrane proteins (Belgareh-Touzé et al.2008). In strain IMS0629, a substitution at position 686 caused a glycine to aspartic acid change at position 229 of Rsp5 (Table 3). Strain IMS0634 carried a 41 bp internal deletion in RSP5, which included the location of the mutation in strain IMS0629 and probably caused loss of function. Table 3. Mutations identified by whole-genome sequencing of glucose-xylose co-consuming S. cerevisiae strains evolved for fast growth at high glucose concentrations. Strain and gene . Nucleotide change . Amino acid change . Description . IMS0629 CYC8 G896A W299 →Stop General transcriptional co-repressor; acts together with Tup1; also acts as part of a transcriptional co-activator complex that recruits the SWI/SNF and SAGA complexes to promoters; can form the prion [OCT+] GAL83 G673T D225Y One of three possible beta-subunits of the Snf1 kinase complex; allows nuclear localisation of the Snf1 kinase complex in the presence of a non-fermentable carbon source RSP5 G686A G229D NEDD4 family E3 ubiquitin ligase; regulates processes including MVB sorting, the heat shock response, transcription, endocytosis and ribosome stability; ubiquitinates Sec23, Sna3, Ste4, Nfi1, Rpo21 and Sem1; autoubiquitinates; deubiquitinated by Ubp2; regulated by SUMO ligase Siz1, in turn regulates Siz1p SUMO ligase activity; required for efficient Golgi-to-ER trafficking in COPI mutants HXK2 C927G Y309 →Stop Hexokinase isoenzyme 2; phosphorylates glucose in cytosol; predominant hexokinase during growth on glucose; represses expression of HXK1, GLK1 RBH1 C190A Q64K Putative protein of unknown function; expression is cell-cycle regulated as shown by microarray analysis; potential regulatory target of Mbp1, which binds to the YJL181W promoter region; contains a PH-like domain DCS1 C636G Y212→Stop Non-essential hydrolase involved in mRNA decapping; activates Xrn1; may function in a feedback mechanism to regulate deadenylation, contains pyrophosphatase activity and a HIT (histidine triad) motif; acts as inhibitor of neutral trehalase Nth1; required for growth on glycerol medium IMS0634 RSP5 Internal deletion (41 nucleotides) Frameshift NEDD4 family E3 ubiquitin ligase; regulates processes including: MVB sorting, the heat shock response, transcription, endocytosis and ribosome stability; ubiquitinates Sec23, Sna3, Ste4, Nfi1, Rpo21 and Sem1; autoubiquitinates; deubiquitinated by Ubp2; regulated by SUMO ligase Siz1, in turn regulates Siz1p SUMO ligase activity; required for efficient Golgi-to-ER trafficking in COPI mutants HXK2 G1027C D343H Hexokinase isoenzyme 2; phosphorylates glucose in cytosol; predominant hexokinase during growth on glucose; represses expression of HXK1, GLK1 Strain and gene . Nucleotide change . Amino acid change . Description . IMS0629 CYC8 G896A W299 →Stop General transcriptional co-repressor; acts together with Tup1; also acts as part of a transcriptional co-activator complex that recruits the SWI/SNF and SAGA complexes to promoters; can form the prion [OCT+] GAL83 G673T D225Y One of three possible beta-subunits of the Snf1 kinase complex; allows nuclear localisation of the Snf1 kinase complex in the presence of a non-fermentable carbon source RSP5 G686A G229D NEDD4 family E3 ubiquitin ligase; regulates processes including MVB sorting, the heat shock response, transcription, endocytosis and ribosome stability; ubiquitinates Sec23, Sna3, Ste4, Nfi1, Rpo21 and Sem1; autoubiquitinates; deubiquitinated by Ubp2; regulated by SUMO ligase Siz1, in turn regulates Siz1p SUMO ligase activity; required for efficient Golgi-to-ER trafficking in COPI mutants HXK2 C927G Y309 →Stop Hexokinase isoenzyme 2; phosphorylates glucose in cytosol; predominant hexokinase during growth on glucose; represses expression of HXK1, GLK1 RBH1 C190A Q64K Putative protein of unknown function; expression is cell-cycle regulated as shown by microarray analysis; potential regulatory target of Mbp1, which binds to the YJL181W promoter region; contains a PH-like domain DCS1 C636G Y212→Stop Non-essential hydrolase involved in mRNA decapping; activates Xrn1; may function in a feedback mechanism to regulate deadenylation, contains pyrophosphatase activity and a HIT (histidine triad) motif; acts as inhibitor of neutral trehalase Nth1; required for growth on glycerol medium IMS0634 RSP5 Internal deletion (41 nucleotides) Frameshift NEDD4 family E3 ubiquitin ligase; regulates processes including: MVB sorting, the heat shock response, transcription, endocytosis and ribosome stability; ubiquitinates Sec23, Sna3, Ste4, Nfi1, Rpo21 and Sem1; autoubiquitinates; deubiquitinated by Ubp2; regulated by SUMO ligase Siz1, in turn regulates Siz1p SUMO ligase activity; required for efficient Golgi-to-ER trafficking in COPI mutants HXK2 G1027C D343H Hexokinase isoenzyme 2; phosphorylates glucose in cytosol; predominant hexokinase during growth on glucose; represses expression of HXK1, GLK1 Gene descriptions were taken from the Saccharomyces Genome Database (https://www.yeastgenome.org/, Accessed 14 December 2017). Open in new tab Table 3. Mutations identified by whole-genome sequencing of glucose-xylose co-consuming S. cerevisiae strains evolved for fast growth at high glucose concentrations. Strain and gene . Nucleotide change . Amino acid change . Description . IMS0629 CYC8 G896A W299 →Stop General transcriptional co-repressor; acts together with Tup1; also acts as part of a transcriptional co-activator complex that recruits the SWI/SNF and SAGA complexes to promoters; can form the prion [OCT+] GAL83 G673T D225Y One of three possible beta-subunits of the Snf1 kinase complex; allows nuclear localisation of the Snf1 kinase complex in the presence of a non-fermentable carbon source RSP5 G686A G229D NEDD4 family E3 ubiquitin ligase; regulates processes including MVB sorting, the heat shock response, transcription, endocytosis and ribosome stability; ubiquitinates Sec23, Sna3, Ste4, Nfi1, Rpo21 and Sem1; autoubiquitinates; deubiquitinated by Ubp2; regulated by SUMO ligase Siz1, in turn regulates Siz1p SUMO ligase activity; required for efficient Golgi-to-ER trafficking in COPI mutants HXK2 C927G Y309 →Stop Hexokinase isoenzyme 2; phosphorylates glucose in cytosol; predominant hexokinase during growth on glucose; represses expression of HXK1, GLK1 RBH1 C190A Q64K Putative protein of unknown function; expression is cell-cycle regulated as shown by microarray analysis; potential regulatory target of Mbp1, which binds to the YJL181W promoter region; contains a PH-like domain DCS1 C636G Y212→Stop Non-essential hydrolase involved in mRNA decapping; activates Xrn1; may function in a feedback mechanism to regulate deadenylation, contains pyrophosphatase activity and a HIT (histidine triad) motif; acts as inhibitor of neutral trehalase Nth1; required for growth on glycerol medium IMS0634 RSP5 Internal deletion (41 nucleotides) Frameshift NEDD4 family E3 ubiquitin ligase; regulates processes including: MVB sorting, the heat shock response, transcription, endocytosis and ribosome stability; ubiquitinates Sec23, Sna3, Ste4, Nfi1, Rpo21 and Sem1; autoubiquitinates; deubiquitinated by Ubp2; regulated by SUMO ligase Siz1, in turn regulates Siz1p SUMO ligase activity; required for efficient Golgi-to-ER trafficking in COPI mutants HXK2 G1027C D343H Hexokinase isoenzyme 2; phosphorylates glucose in cytosol; predominant hexokinase during growth on glucose; represses expression of HXK1, GLK1 Strain and gene . Nucleotide change . Amino acid change . Description . IMS0629 CYC8 G896A W299 →Stop General transcriptional co-repressor; acts together with Tup1; also acts as part of a transcriptional co-activator complex that recruits the SWI/SNF and SAGA complexes to promoters; can form the prion [OCT+] GAL83 G673T D225Y One of three possible beta-subunits of the Snf1 kinase complex; allows nuclear localisation of the Snf1 kinase complex in the presence of a non-fermentable carbon source RSP5 G686A G229D NEDD4 family E3 ubiquitin ligase; regulates processes including MVB sorting, the heat shock response, transcription, endocytosis and ribosome stability; ubiquitinates Sec23, Sna3, Ste4, Nfi1, Rpo21 and Sem1; autoubiquitinates; deubiquitinated by Ubp2; regulated by SUMO ligase Siz1, in turn regulates Siz1p SUMO ligase activity; required for efficient Golgi-to-ER trafficking in COPI mutants HXK2 C927G Y309 →Stop Hexokinase isoenzyme 2; phosphorylates glucose in cytosol; predominant hexokinase during growth on glucose; represses expression of HXK1, GLK1 RBH1 C190A Q64K Putative protein of unknown function; expression is cell-cycle regulated as shown by microarray analysis; potential regulatory target of Mbp1, which binds to the YJL181W promoter region; contains a PH-like domain DCS1 C636G Y212→Stop Non-essential hydrolase involved in mRNA decapping; activates Xrn1; may function in a feedback mechanism to regulate deadenylation, contains pyrophosphatase activity and a HIT (histidine triad) motif; acts as inhibitor of neutral trehalase Nth1; required for growth on glycerol medium IMS0634 RSP5 Internal deletion (41 nucleotides) Frameshift NEDD4 family E3 ubiquitin ligase; regulates processes including: MVB sorting, the heat shock response, transcription, endocytosis and ribosome stability; ubiquitinates Sec23, Sna3, Ste4, Nfi1, Rpo21 and Sem1; autoubiquitinates; deubiquitinated by Ubp2; regulated by SUMO ligase Siz1, in turn regulates Siz1p SUMO ligase activity; required for efficient Golgi-to-ER trafficking in COPI mutants HXK2 G1027C D343H Hexokinase isoenzyme 2; phosphorylates glucose in cytosol; predominant hexokinase during growth on glucose; represses expression of HXK1, GLK1 Gene descriptions were taken from the Saccharomyces Genome Database (https://www.yeastgenome.org/, Accessed 14 December 2017). Open in new tab Compared to strain IMS0634, strain IMS0629 harboured four additional nucleotide changes in protein-coding regions (Table 3). A G-A change at position 896 of the transcriptional regulator gene CYC8 introduced a stop codon at position 299 of the protein. Deletion of CYC8 was previously shown to enhance xylose uptake in the presence of glucose, albeit at the expense of growth rate (Nijland et al.2017). A G-T change at position 673 of the transcriptional regulator gene GAL83 caused an amino acid change from aspartic acid to tyrosine at position 225 of the protein. Gal83 plays a vital role in the function of the Snf1-kinase complex of S. cerevisiae, which is involved in activation of glucose-repressed genes in the absence of the sugar (Rahner et al.1996; Vincent and Carlson 1998, 1999; Schmidt and McCartney 2000). Analysis of chromosomal copy number variations showed no chromosomal rearrangements in strain IMS0629 (Additional File 8). In contrast, strain IMS0634 carried a duplication of the right arm of chromosome 3, a duplication of the middle part of chromosome 8 and a duplication of chromosome 9 (Additional File 8). The duplications in chromosomes 8 and 9 in IMS0634 spanned the GND1, GRE3 and SGA1 loci at which the expressing cassettes for heterologous genes were integrated (Table 1). In the evolved strains IMS0629 and IMS0634, xylA copy numbers had increased to ca. 27 and 20, respectively. This observation is consistent with previous research that showed a requirement for high copy numbers of xylA expression cassettes to support fast xylose consumption (Alper and Stephanopoulos 2009; Verhoeven et al.2017). The duplication of a segment of chromosome 8 in strain IMS0634 also spanned the locations of the low-to-moderate affinity hexose transporter genes HXT1 and HXT5 and the high-affinity hexose transporter gene HXT4. Mutations in HXK2, RSP5 and GAL83 stimulate co-consumption of xylose and glucose in aerobic cultures of xylose-consuming S. cerevisiae To investigate whether mutations acquired during evolution for forced co-consumption of glucose and xylose were relevant for mixed-sugar utilisation in a strain without forced glucose-xylose co-consumption, we focused on mutations in HXK2 and RSP5 (which were present in both IMS0629 and IMS0634) and/or GAL83. Mutations in these genes were introduced into an engineered, non-evolved xylose-consuming S. cerevisiae strain background (IMX994, Table 1). Overexpression of xylA was accomplished by transforming strains with the multicopy xylA expression vector pAKX002 (Kuyper et al.2003). In aerobic shake-flask cultures grown on 10 g L−1 glucose and 10 g L−1 xylose, the reference strain IMU079 (XKS1↑ PPP↑ pAKX002) displayed a pronounced biphasic growth profile and only a minor co-consumption of the two sugars (0.13 mol xylose (mol glucose)−1; Table 4, Fig. 4, Additional File 9). Co-consumption was strongly enhanced in the congenic hxk2Δ strain IMX1485, which showed a 3-fold higher molar ratio of xylose and glucose consumption (0.41 mol mol−1). However, its specific growth rate before glucose depletion (0.28 h−1) was 13% lower than that of the reference strain (Table 4). Strain IMX1487 (rsp5Δ), which showed a 20% lower specific growth rate than the reference strain, showed a slight improvement in co-consumption (Table 4). Deletion of GAL83 (strain IMX1486) affected neither sugar co-consumption nor growth rate. In contrast, replacement of GAL83 by GAL83G673T (strain IMX1488) resulted in a 40% higher co-consumption of glucose and xylose than observed in the reference strain IMU079, without affecting growth rate (Table 4). Figure 4. Open in new tabDownload slide Consumption of glucose and xylose and growth of strains IMU079 (XKS1↑ PPP↑ pAKX002; a, d), IMX1515 (hxk2Δ rsp5Δ XKS1↑ PPP↑ pAKX002; b, e) and IMX1583 (hxk2Δ gal83::GAL83G673T XKS1↑ PPP↑ pAKX002; c, f) in batch cultures. The three strains were grown on SM (urea as nitrogen source) with 10 g L−1 glucose and 10 g L−1 xylose in aerobic shake-flask cultures (pH 6, 30°C). a, b, c: filled circles glucose, open circles xylose, open squares OD660; d, e, f: ratio of xylose and glucose consumption during exponential growth phase. Data for independent duplicate cultures are provided in Additional File 9. Table 4. Specific growth rates (μ) and ratio of xylose and glucose consumption in aerobic shake-flask cultures of strains IMU079 (XKS1↑ PPP↑ pAKX002), IMX1485 (hkx2Δ XKS1↑ PPP↑ pAKX002), IMX1486 (gal83Δ XKS1↑ PPP↑ pAKX002), IMX1487 (rsp5Δ XKS1↑ PPP↑ pAKX002), IMX1488 (gal83::GAL83G673T XKS1↑ PPP↑ pAKX002), IMX1515 (hxk2Δ rsp5Δ XKS1↑ PPP↑ pAKX002) and IMX1583 (hxk2Δ gal83::GAL83G673T XKS1↑ PPP↑ pAKX002) grown on SM (urea as nitrogen source) with 10 g L−1 glucose and 10 g L−1 xylose (pH 6, 30°C). Strain . Relevant genotype . μ (h−1) . Xylose-glucose consumption ratio (mol mol−1) . IMU079 HXK2 RSP5 GAL83 0.32 ± 0.01 0.13 ± 0.00 IMX1485 hxk2Δ 0.28 ± 0.00 0.41 ± 0.01 IMX1486 gal83Δ 0.31 ± 0.00 0.14 ± 0.01 IMX1487 rsp5Δ 0.26 ± 0.00 0.15 ± 0.00 IMX1488 gal83::GAL83G673T 0.31 ± 0.00 0.18 ± 0.01 IMX1515 hxk2Δ rsp5Δ 0.19 ± 0.00 0.60 ± 0.00 IMX1583 hxk2Δ gal83::GAL83G673T 0.31 ± 0.00 0.49 ± 0.00 Strain . Relevant genotype . μ (h−1) . Xylose-glucose consumption ratio (mol mol−1) . IMU079 HXK2 RSP5 GAL83 0.32 ± 0.01 0.13 ± 0.00 IMX1485 hxk2Δ 0.28 ± 0.00 0.41 ± 0.01 IMX1486 gal83Δ 0.31 ± 0.00 0.14 ± 0.01 IMX1487 rsp5Δ 0.26 ± 0.00 0.15 ± 0.00 IMX1488 gal83::GAL83G673T 0.31 ± 0.00 0.18 ± 0.01 IMX1515 hxk2Δ rsp5Δ 0.19 ± 0.00 0.60 ± 0.00 IMX1583 hxk2Δ gal83::GAL83G673T 0.31 ± 0.00 0.49 ± 0.00 Growth rates and ratios were calculated from samples taken during the mid-exponential growth phase and represent averages ± mean deviation of independent duplicate cultures. Open in new tab Table 4. Specific growth rates (μ) and ratio of xylose and glucose consumption in aerobic shake-flask cultures of strains IMU079 (XKS1↑ PPP↑ pAKX002), IMX1485 (hkx2Δ XKS1↑ PPP↑ pAKX002), IMX1486 (gal83Δ XKS1↑ PPP↑ pAKX002), IMX1487 (rsp5Δ XKS1↑ PPP↑ pAKX002), IMX1488 (gal83::GAL83G673T XKS1↑ PPP↑ pAKX002), IMX1515 (hxk2Δ rsp5Δ XKS1↑ PPP↑ pAKX002) and IMX1583 (hxk2Δ gal83::GAL83G673T XKS1↑ PPP↑ pAKX002) grown on SM (urea as nitrogen source) with 10 g L−1 glucose and 10 g L−1 xylose (pH 6, 30°C). Strain . Relevant genotype . μ (h−1) . Xylose-glucose consumption ratio (mol mol−1) . IMU079 HXK2 RSP5 GAL83 0.32 ± 0.01 0.13 ± 0.00 IMX1485 hxk2Δ 0.28 ± 0.00 0.41 ± 0.01 IMX1486 gal83Δ 0.31 ± 0.00 0.14 ± 0.01 IMX1487 rsp5Δ 0.26 ± 0.00 0.15 ± 0.00 IMX1488 gal83::GAL83G673T 0.31 ± 0.00 0.18 ± 0.01 IMX1515 hxk2Δ rsp5Δ 0.19 ± 0.00 0.60 ± 0.00 IMX1583 hxk2Δ gal83::GAL83G673T 0.31 ± 0.00 0.49 ± 0.00 Strain . Relevant genotype . μ (h−1) . Xylose-glucose consumption ratio (mol mol−1) . IMU079 HXK2 RSP5 GAL83 0.32 ± 0.01 0.13 ± 0.00 IMX1485 hxk2Δ 0.28 ± 0.00 0.41 ± 0.01 IMX1486 gal83Δ 0.31 ± 0.00 0.14 ± 0.01 IMX1487 rsp5Δ 0.26 ± 0.00 0.15 ± 0.00 IMX1488 gal83::GAL83G673T 0.31 ± 0.00 0.18 ± 0.01 IMX1515 hxk2Δ rsp5Δ 0.19 ± 0.00 0.60 ± 0.00 IMX1583 hxk2Δ gal83::GAL83G673T 0.31 ± 0.00 0.49 ± 0.00 Growth rates and ratios were calculated from samples taken during the mid-exponential growth phase and represent averages ± mean deviation of independent duplicate cultures. Open in new tab Since independently evolved glucose-xylose co-consuming strains contained putative loss-of-function mutations in HXK2 and RSP5, both genes were deleted in strain IMX1515 (hxk2Δ rsp5Δ XKS1↑ PPP↑ pAKX002). Similarly, deletion of HXK2 and introduction of GAL83G673Twere combined in strain IMX1583 (hxk2Δ gal83::GAL83G673T XKS1↑ PPP↑ pAKX002). Co-consumption ratios in the two strains (0.60 and 0.49 mol xylose (mol glucose)−1, respectively) were 4- to 5-fold higher than in the reference strain IMU079 (Table 4, Fig. 4, Additional File 9). However, strain IMX1515 exhibited a 40% lower specific growth rate (0.19 h−1) than the reference strain, resulting in a 9 h extension of the fermentation experiments (Fig. 4, Additional File 9). In contrast, strain IMX1583 combined a high co-consumption ratio with the same specific growth rate as that of the reference strain. Combined mutations in HXK2 and GAL83 significantly accelerate conversion of glucose-xylose mixtures by anaerobic cultures of xylose-consuming S. cerevisiae To investigate the impact of the identified mutations under more industrially relevant conditions, anaerobic growth of the reference xylose-fermenting strain IMU079 (XKS1↑ PPP↑ pAKX002) in bioreactor batch experiments was compared with that of the two congenic double mutants IMX1515 (hxk2Δ rsp5Δ) and IMX1583 (hxk2Δ gal83::GAL83G673T) that showed the highest glucose-xylose co-consumption in the aerobic shake-flask experiments. The anaerobic cultures were grown on 20 g L−1 glucose and 10 g L−1 xylose to simulate the relative concentrations of these sugars typically found in corn stover and wheat straw hydrolysates (van Maris et al.2006). In the anaerobic batch cultures, strains IMU079, IMX1515 and IMX1583 all produced CO2, biomass, ethanol and glycerol as main products, with a minor production of acetate (Table 5, Fig. 5, Additional File 10). The strains did not produce xylitol during exponential growth and low concentrations of xylitol in cultures of strain IMU079 (2.2 ± 0.1 mmol L−1) were only observed at the end of fermentation. As observed in aerobic cultures (Fig. 4), strain IMU079 showed a clear biphasic growth profile in the anaerobic bioreactors (Fig. 5, Additional File 10), during which a fast glucose phase (ca. 16 h) was followed by a much slower and decelerating xylose consumption phase. During the glucose phase, this reference strain maintained a specific growth rate of 0.29 h−1 and a glucose-xylose co-consumption ratio of 0.14 mol mol−1 (Table 5). After a ca. 30-h lag phase (Fig. 5, Additional File 10), strain IMX1515 exhibited an exponential growth rate of 0.07 ± 0.00 h−1, with a high glucose-xylose co-consumption ratio (0.45 ± 0.03 mol mol−1). Mainly as a result of its lag phase, strain IMX1515 took longer to consume all sugars than the reference strain IMU079, but its xylose-consumption phase was ca. 65% shorter (ca. 14 and 43 h, respectively; Fig. 5, Additional File 10). Figure 5. Open in new tabDownload slide Sugar consumption, biomass and metabolite production profiles of S. cerevisiae strains IMU079 (XKS1↑ PPP↑ pAKX002; a, b), IMX1515 (hxk2Δ rsp5Δ XKS1↑ PPP↑ pAKX002; c, d) and IMX1583 (hxk2Δ gal83::GAL83G673T XKS1↑ PPP↑ pAKX002; e, f), grown on SM with 20 g L−1 glucose and 10 g L−1 xylose in anaerobic bioreactor batch cultures (pH 5, 30°C). Cultures were grown in duplicate, the data shown are from a single representative culture. a, c, e: filled circles glucose, open circles xylose; b, d, f: filled squares biomass, open squares ethanol, filled triangles acetate, open triangles glycerol. Data on ethanol corrected for evaporation. For duplicate cultures, refer to Additional File 10. Table 5. Product yields, biomass-specific rates and sugar uptake ratios in anaerobic bioreactor batch cultures of strains IMU079 (XKS1↑ PPP↑ pAKX002), IMX1515 (hxk2Δ rsp5Δ XKS1↑ PPP↑ pAKX002) and IMX1583 (hxk2Δ gal83::GAL83G673T XKS1↑ PPP↑ pAKX002) grown on SM supplemented with 20 g L−1 glucose and 10 g L−1 xylose (pH 5, 30°C). Strain . IMU079 . IMX1515 . IMX1583 . Relevant genotype HXK2 RSP5 GAL83 hxk2Δ rsp5Δ hxk2Δ gal83::GAL83G673T μ (h−1) 0.29 ± 0.01 0.07 ± 0.00 0.21 ± 0.00 Spec. xylose uptake rate mmol (g biomass)−1 h−1 2.22 ± 0.14 2.50 ± 0.12 3.19 ± 0.02 Spec. glucose uptake rate mmol (g biomass)−1 h−1 15.65 ± 0.52 5.58 ± 0.08 10.09 ± 0.08 Glucose-xylose consumption ratio (mol mol−1) 0.14 ± 0.00 0.45 ± 0.03 0.32 ± 0.01 Biomass yield on sugars (g biomass g−1) 0.09 ± 0.01 0.05 ± 0.00 0.09 ± 0.00 Ethanol yield on sugars (g g−1) 0.37 ± 0.00 0.43 ± 0.00 0.38 ± 0.01 Glycerol yield on sugars (g g−1) 0.10 ± 0.00 0.06 ± 0.00 0.08 ± 0.00 Ratio glycerol production on biomass production (mmol (g biomass)−1) 11.5 ± 0.60 12.0 ± 0.50 9.9 ± 0.10 Xylitol production (mmol L−1) 2.22 ± 0.06 0.90 ± 0.04 0.35 ± 0.04 Strain . IMU079 . IMX1515 . IMX1583 . Relevant genotype HXK2 RSP5 GAL83 hxk2Δ rsp5Δ hxk2Δ gal83::GAL83G673T μ (h−1) 0.29 ± 0.01 0.07 ± 0.00 0.21 ± 0.00 Spec. xylose uptake rate mmol (g biomass)−1 h−1 2.22 ± 0.14 2.50 ± 0.12 3.19 ± 0.02 Spec. glucose uptake rate mmol (g biomass)−1 h−1 15.65 ± 0.52 5.58 ± 0.08 10.09 ± 0.08 Glucose-xylose consumption ratio (mol mol−1) 0.14 ± 0.00 0.45 ± 0.03 0.32 ± 0.01 Biomass yield on sugars (g biomass g−1) 0.09 ± 0.01 0.05 ± 0.00 0.09 ± 0.00 Ethanol yield on sugars (g g−1) 0.37 ± 0.00 0.43 ± 0.00 0.38 ± 0.01 Glycerol yield on sugars (g g−1) 0.10 ± 0.00 0.06 ± 0.00 0.08 ± 0.00 Ratio glycerol production on biomass production (mmol (g biomass)−1) 11.5 ± 0.60 12.0 ± 0.50 9.9 ± 0.10 Xylitol production (mmol L−1) 2.22 ± 0.06 0.90 ± 0.04 0.35 ± 0.04 Rates, yields and ratios were calculated from samples taken during the mid-exponential growth phase and represent averages ± mean deviation of independent duplicate cultures. Ethanol yields were corrected for evaporation. Open in new tab Table 5. Product yields, biomass-specific rates and sugar uptake ratios in anaerobic bioreactor batch cultures of strains IMU079 (XKS1↑ PPP↑ pAKX002), IMX1515 (hxk2Δ rsp5Δ XKS1↑ PPP↑ pAKX002) and IMX1583 (hxk2Δ gal83::GAL83G673T XKS1↑ PPP↑ pAKX002) grown on SM supplemented with 20 g L−1 glucose and 10 g L−1 xylose (pH 5, 30°C). Strain . IMU079 . IMX1515 . IMX1583 . Relevant genotype HXK2 RSP5 GAL83 hxk2Δ rsp5Δ hxk2Δ gal83::GAL83G673T μ (h−1) 0.29 ± 0.01 0.07 ± 0.00 0.21 ± 0.00 Spec. xylose uptake rate mmol (g biomass)−1 h−1 2.22 ± 0.14 2.50 ± 0.12 3.19 ± 0.02 Spec. glucose uptake rate mmol (g biomass)−1 h−1 15.65 ± 0.52 5.58 ± 0.08 10.09 ± 0.08 Glucose-xylose consumption ratio (mol mol−1) 0.14 ± 0.00 0.45 ± 0.03 0.32 ± 0.01 Biomass yield on sugars (g biomass g−1) 0.09 ± 0.01 0.05 ± 0.00 0.09 ± 0.00 Ethanol yield on sugars (g g−1) 0.37 ± 0.00 0.43 ± 0.00 0.38 ± 0.01 Glycerol yield on sugars (g g−1) 0.10 ± 0.00 0.06 ± 0.00 0.08 ± 0.00 Ratio glycerol production on biomass production (mmol (g biomass)−1) 11.5 ± 0.60 12.0 ± 0.50 9.9 ± 0.10 Xylitol production (mmol L−1) 2.22 ± 0.06 0.90 ± 0.04 0.35 ± 0.04 Strain . IMU079 . IMX1515 . IMX1583 . Relevant genotype HXK2 RSP5 GAL83 hxk2Δ rsp5Δ hxk2Δ gal83::GAL83G673T μ (h−1) 0.29 ± 0.01 0.07 ± 0.00 0.21 ± 0.00 Spec. xylose uptake rate mmol (g biomass)−1 h−1 2.22 ± 0.14 2.50 ± 0.12 3.19 ± 0.02 Spec. glucose uptake rate mmol (g biomass)−1 h−1 15.65 ± 0.52 5.58 ± 0.08 10.09 ± 0.08 Glucose-xylose consumption ratio (mol mol−1) 0.14 ± 0.00 0.45 ± 0.03 0.32 ± 0.01 Biomass yield on sugars (g biomass g−1) 0.09 ± 0.01 0.05 ± 0.00 0.09 ± 0.00 Ethanol yield on sugars (g g−1) 0.37 ± 0.00 0.43 ± 0.00 0.38 ± 0.01 Glycerol yield on sugars (g g−1) 0.10 ± 0.00 0.06 ± 0.00 0.08 ± 0.00 Ratio glycerol production on biomass production (mmol (g biomass)−1) 11.5 ± 0.60 12.0 ± 0.50 9.9 ± 0.10 Xylitol production (mmol L−1) 2.22 ± 0.06 0.90 ± 0.04 0.35 ± 0.04 Rates, yields and ratios were calculated from samples taken during the mid-exponential growth phase and represent averages ± mean deviation of independent duplicate cultures. Ethanol yields were corrected for evaporation. Open in new tab In contrast to strain IMX1515, strain IMX1583 (hxk2Δ gal83::GAL83G673T XKS1↑ PPP↑ pAKX002) did not exhibit a lag phase but immediately started exponential growth at 0.21 h−1 (Fig. 5). A comparison of biomass-specific uptake rates of xylose and glucose in the anaerobic batch experiments showed that strain IMX1583 maintained a 44% higher xylose uptake rate than strain IMU079 before glucose exhaustion (Table 5). Moreover, both strains IMX1515 and IMX1583 did not show the pronounced decline of xylose consumption after glucose exhaustion that was observed in the reference strain (Fig. 5, Additional File 10). As a result, the xylose consumption phase in anaerobic cultures of strain IMX1583 was 80% shorter than that strain IMU079 (ca. 9 h compared to 43 h), thereby reducing the time required for complete sugar conversion by over 24 h (Fig. 5). DISCUSSION Engineering S. cerevisiae for forced co-consumption of xylose and glucose The key objective of this study was to develop and test a strain platform that, via laboratory evolution and subsequent genome resequencing, can be used to identify mutations that support co-metabolism of xylose and glucose. In previous studies, laboratory evolution of glucose-phosphorylation-negative, pentose-fermenting strains in the presence of glucose yielded valuable leads for improving utilisation of glucose-xylose mixtures, including mutations in HXT genes that improved pentose uptake in the presence of glucose (Farwick et al.2014; Nijland et al.2014, 2017; Wisselink, Van Maris and Pronk 2015). The strategy described in this study differs from these previous studies as it not only enabled selection for xylose utilisation in the presence of glucose, but also for simultaneous metabolism of the two sugars. The molar ratio of xylose and glucose co-consumption (1.64 mol xylose (mol glucose)−1) by the evolved strain IMS0629 is the highest reported to date for batch cultures of S. cerevisiae (Katahira et al.2008; Shen et al.2012, 2015; Gonçalves et al.2014; Nijland et al.2014; Westman et al.2014; Shin et al.2015). Inactivation of PGI1 played a key role in forcing simultaneous utilisation of xylose and glucose (Fig. 1). To pre-empt the well-documented inability of S. cerevisiae pgi1Δ mutants to reoxidise the NADPH that is generated when metabolism of glucose-6-phosphate is rerouted through the oxidative PPP (Boles, Lehnert and Zimmermann 1993; Heux, Cadiere and Dequin 2008), we replaced Gnd1 and Gnd2 with the NAD+-linked bacterial 6-phosphogluconate dehydrogenase GndA (Papapetridis et al.2016). Together with the co-consumption of xylose, via an engineered pathway that did not involve NAD(P)H generation (Fig. 1), these modifications enabled the engineered strain IMX1046 to grow on mixtures of xylose and 1 g L−1 glucose, without the fructose supplementation that is normally required for growth of pgi1-null mutants on glucose (Aguilera 1986; Boles, Lehnert and Zimmermann 1993). The evolved strain IMS0629 consumed 8.6 mmol glucose (g biomass)−1 in aerobic batch cultures (Table 2), which is close to the requirement of 9.3 mmol NADPH (g biomass)−1 for aerobic growth on glucose of wild-type S. cerevisiae (Bruinenberg, Van Dijken and Scheffers 1983). However, since glucose-6-phosphate is a key biosynthetic precursor, not all glucose consumed by the cultures can be converted through the oxidative PPP. Additional enzymes such as NADP+-dependent acetaldehyde dehydrogenase Ald6 (Grabowska and Chelstowska 2003; Celton et al.2012; Papapetridis et al.2016) are therefore likely to have supplemented NADPH generation via the oxidative PPP in the ‘forced co-consumption’ strains. Improvement of mixed-sugar fermentation in aerobic and anaerobic cultures of xylose-fermenting S. cerevisiae Mutations in HXK2, RSP5 and GAL83 were reverse engineered into a non-evolved strain (XKS1↑ PPP↑) to investigate whether mutations acquired during evolution of strain IMX1046 (pgi1Δ rpe1Δ gnd1Δ gnd2Δ gndA xylA XKS1↑ PPP↑) on glucose-xylose mixtures would also stimulate mixed-substrate utilisation in a genetic background that does not impose forced co-utilisation. During growth on glucose-xylose mixtures, the reference strain IMU079 (XKS1↑ PPP↑ pAKX002) displayed the typical biphasic growth profile seen in non-evolved, xylose-consuming strains that express a basic, functional xylose-isomerase (XI)-based xylose fermentation pathway (Kuyper et al.2003, 2005a). Biphasic growth was especially pronounced in anaerobic cultures, in which the xylose consumption rate collapsed upon glucose depletion (Fig. 5). Deletion of HXK2, either combined with the deletion of RSP5 or with the introduction of a GAL83G673T mutation, strongly improved mixed-sugar fermentation kinetics, both by increased co-utilisation and by faster conversion of xylose after glucose had been depleted (Figs 4 and 5). Hxk2, the major hexokinase in S. cerevisiae, plays an additional key role in transcriptional repression of a large set of yeast genes (Entian 1980; Entian and Fröhlich 1984; Peláez, Herrero and Moreno 2010) by glucose. Deletion of HXK2 has been shown to enhance co-consumption of combinations of natural substrates (glucose-galactose, glucose-sucrose and glucose-ethanol) in batch cultures (Raamsdonk et al.2001). During exponential growth on glucose in batch cultures, hxk2Δ mutants show increased transcription of the high-affinity hexose transporter genes HXT2 and HXT7 and decreased transcription of the low-affinity hexose transporter genes HXT1 and HXT3 (Petit et al.2000). The high-affinity Hxt transporters, which in wild-type strains are only expressed at low glucose concentrations (Diderich et al.1999), have a much lower Km for xylose than their low-affinity counterparts (Hamacher et al.2002; Lee et al.2002). Several studies have demonstrated that overexpression of high-affinity hexose transporters stimulates xylose uptake (Subtil and Boles 2012; Farwick et al.2014; Gonçalves et al.2014; Nijland et al.2014, 2016; Reznicek et al.2015; Shin et al.2015; Jansen et al.2017). The observed improved co-utilisation of glucose and xylose upon inactivation of HXK2 may therefore reflect an increased abundance of high-affinity hexose transporters in the yeast plasma membrane during growth on glucose-xylose mixtures. A recent in silico study identified HXK2 as potential target for improving xylose uptake rates in S. cerevisiae (Miskovic et al.2017). However, subsequent deletion of HXK2 in a strain expressing a XR/XDH-based pathway resulted in faster xylose uptake but also increased by-product formation and reduced ethanol productivity (Miskovic et al.2017). The absence of such negative effects in the present study is in line with reports that XI-based strains are less prone to by-product formation than XR/XDH-based strains (Moysés et al.2016; Jansen et al.2017). Rsp5, the only representative of the NEDD4 family of E3-ubiquitin ligases in S. cerevisiae, is involved in regulation of a multitude of cellular processes, including intracellular protein trafficking, regulation of the large subunit of RNA polymerase II, ribosome stability, regulation of fatty acid synthesis and stress response (Huibregtse, Yang and Beaudenon 1997; Belgareh-Touzé et al.2008; Kaliszewski and Zoładek 2008; Shcherbik and Pestov 2011). This multitude of roles may explain the reduced growth rate of the rsp5Δ strains (Tables 4 and 5) in this study. Involvement of Rsp5 in ubiquitination and subsequent endocytosis of the high-affinity hexose transporters Hxt6 and Hxt7 (Krampe et al.1998; Haguenauer-Tsapis and André 2004) could explain the strong synergistic effect of the hxk2Δ and rsp5Δ deletions: while deletion of HXK2 prevents glucose repression of the synthesis of these transporters, deletion of RSP5 could prevent their ubiquitination and removal from the membrane. Removal of ubiquitination sites in the hexose transporters Hxt1 and Hxt36 was previously shown to enhance xylose uptake by S. cerevisiae (Nijland et al.2016). Our results suggest that a similar modification of Hxt6 and Hxt7 could also be beneficial. Gal83 is one of three possible β-subunits of the Snf1-kinase complex, which enables transcription of glucose-repressed genes at low glucose concentrations (for reviews, see Gancedo 1998; Schüller 2003). At non-repressing glucose concentrations, Gal83 directs the Snf1-Gal83 complex to the cell nucleus (Vincent et al.2001), where it mediates transcriptional upregulation of genes involved in utilisation of alternative carbon sources (Vincent and Carlson 1999). Targets of the Snf1-Gal83 complex include the GAL regulon (Matsumoto, Toh-e and Oshima 1981; Erickson and Johnston 1993) and the high-affinity hexose transporter genes HXT2 and HXT4 (Özcan and Johnston 1999). The D225Y substitution, which stimulated glucose-xylose co-consumption in the present study, is located in the glycogen-binding domain of Gal83 (residues 161–243; Momcilovic et al.2008). Other mutations in this domain have been shown to cause transcription of Snf1-Gal83 targets in the presence of glucose (Wiatrowski et al.2004; Momcilovic et al.2008; Mangat et al.2010). In contrast to deletion of the transcriptional regulator CYC8 (Nijland et al.2017), which also stimulated co-utilisation but caused severe reductions of the specific growth rate of engineered strains, the GAL83G673T mutation did not have a strong impact on growth rate (Tables 4 and 5). The synergistic effect of the hxk2Δ and GAL83G673T mutations may be related to the involvement of Hxk2 in deactivation of Snf1 in the presence of glucose, causing constitutive activity of the Snf1-Gal83 complex in hxk2-null mutants (Gancedo 1998; Schüller 2003). The reverse engineered mutations in HXK2 and GAL83 or RSP5 not only stimulated simultaneous utilisation of xylose and glucose when both sugars were present, but also prevented the sharp decline in xylose uptake rates that occurred in the reference strain IMU079 (XKS1↑ PPP↑ pAKX002). In the reference strain, the biomass-specific rate of xylose consumption, probably mediated by low- or moderate-affinity Hxt transporters, declined to values below 0.5 mmol (g biomass)−1 h−1 after glucose depletion (Fig. 5, Additional File 10). Under anaerobic conditions, this low rate of xylose fermentation would correspond to a biomass-specific rate of ATP production of 0.8 mmol (g biomass)−1 h−1. This value is lower than the estimated ATP requirement for cellular maintenance of S. cerevisiae (ca. 1 mmol ATP (g biomass)−1 h−1; Boender et al.2009). Since protein synthesis is a highly ATP-intensive process (Canelas et al.2010), an inability of the reference strain to functionally express high-affinity hexose transporters upon glucose depletion may therefore reflect an energy shortage. A similar effect was observed during transitions between glucose and galactose growth in anaerobic S. cerevisiae cultures (van den Brink et al.2009). By already expressing functional high-affinity transporters before glucose was depleted, the hxk2Δ rsp5Δ and hxk2Δ GAL83G673Tgenotypes may have enabled cells to avoid such a bioenergetic ‘valley of death’ upon the transition to xylose fermentation. In industrial processes using lignocellulosic feedstocks, this energetic challenge is likely to be even more stringent due to the presence of compounds such as acetic acid that increase maintenance energy requirements (Verduyn et al.1992; Abbott et al.2007; Bellissimi et al.2009). While reverse engineering of HXK2, RSP5 and GAL83 mutations demonstrated the relevance of the forced co-utilisation strategy demonstrated in this study, they do not exhaust its possibilities. Prolongation of the evolution experiments, combined with anaerobic conditions and/or a further increase of the glucose to xylose ratio in the medium, may allow for selection of additional relevant mutations. The IMS0629 strain may also be used to select mutations that enable efficient co-utilisation of glucose and xylose at different concentrations or in lignocellulosic hydrolysates that, in addition to fermentable sugars, contain inhibitors of yeast performance (Palmqvist and Hahn-Hägerdal 2000; Klinke et al.2002; Almeida et al.2007). Alternatively, the strain design can be adapted to enable selection for co-metabolism of other sugars, for example by replacing the xylose pathway by a bacterial pathway for conversion of L-arabinose into xylulose-5-phosphate (Becker and Boles 2003; Wisselink et al.2007). This strategy can be extended to optimising the yeast metabolic network for aerobic production of economically relevant compounds other than ethanol from lignocellulosic hydrolysates. In particular, imposing fixed stoichiometries of glycolytic and (non-oxidative) PPP reactions may offer interesting options for high-yield production of compounds whose synthesis requires a large net input of PPP intermediates and/or NADPH, such as aromatic compounds and lipids (Tehlivets, Scheuringer and Kohlwein 2007; Klug and Daum 2014; Gottardi et al.2017). CONCLUSIONS Engineering of carbon and redox metabolism yielded an S. cerevisiae strain whose growth was strictly dependent on the simultaneous uptake and metabolism of xylose and glucose. Laboratory evolution improved growth of the resulting strains on mixtures of xylose and glucose at elevated glucose concentrations. Mutations in HXK2, RSP5 and GAL83 were identified by genome sequencing of the evolved strains. Upon their combined introduction into an engineered xylose-fermenting yeast strain, these mutations strongly stimulated simultaneous utilisation of xylose and glucose and, after depletion of glucose, fast conversion of the remaining xylose. The developed strain platform and modified versions thereof can be used for identification of further metabolic engineering targets for improving the performance of yeast strains in industrial processes based on lignocellulosic feedstocks. SUPPLEMENTARY DATA Supplementary data are available at FEMSYR online. Acknowledgements We thank Jasmine Bracher and Oscar Martinez for construction of strain IMU079. We thank Jean-Marc Daran and Pilar de la Torre for advice on molecular biology, Marcel van den Broek for assistance with bioinformatics analyses and Erik de Hulster for advice on fermentations. We thank Laura Valk for stimulating discussions. Availability of supporting data The genomic sequence of strains IMS0629 and IMS0634 have been deposited in Genbank (http://www.ncbi.nlm.nih.gov/), BioProject number PRJNA433919. Author's contributions IP, MDV, AJAvM and JTP designed experiments. IP, MDV and JTP wrote the first draft of the manuscript. IP and MDV constructed yeast strains and performed fermentations. SJW constructed yeast strains and performed in silico analyses. MG designed and carried out fermentations. All authors read and commented upon the draft manuscript and approved the final version. Funding The PhD project of IP is funded by DSM. This work was performed within the BE-Basic R&D Program (http://www.be-basic.org/), which is financially supported by an EOS Long Term grant from the Dutch Ministry of Economic Affairs, Agriculture and Innovation (EL&I). Competing interests IP, MDV, AJAvM and JTP are co-inventors on a patent application related to the content of this work. REFERENCES Abbott DA , Knijnenburg TA, De Poorter LMI et al. Generic and specific transcriptional responses to different weak organic acids in anaerobic chemostat cultures of Saccharomyces cerevisiae . FEMS Yeast Res 2007 ; 7 : 819 – 33 . Google Scholar Crossref Search ADS PubMed WorldCat Aguilera A . Deletion of the phosphoglucose isomerase structural gene makes growth and sporulation glucose dependent in Saccharomyces cerevisiae . Mol Gen Genet 1986 ; 204 : 310 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat Almeida JRM , Modig T, Petersson A et al. Increased tolerance and conversion of inhibitors in lignocellulosic hydrolysates by Saccharomyces cerevisiae . J Chem Technol Biot 2007 ; 82 : 340 – 9 . Google Scholar Crossref Search ADS WorldCat Alper H , Stephanopoulos G. Engineering for biofuels: exploiting innate microbial capacity or importing biosynthetic potential ? Nat Rev Microbiol 2009 ; 7 : 715 – 23 . Google Scholar Crossref Search ADS PubMed WorldCat Ask M , Bettiga M, Duraiswamy VR et al. Pulsed addition of HMF and furfural to batch-grown xylose-utilizing Saccharomyces cerevisiae results in different physiological responses in glucose and xylose consumption phase . Biotechnol Biofuels 2013 ; 6 : 181 . Google Scholar Crossref Search ADS PubMed WorldCat Aung HW , Henry SA, Walker LP. Revising the representation of fatty acid, glycerolipid, and glycerophospholipid metabolism in the consensus model of yeast metabolism . Ind Biotechnol 2013 ; 9 : 215 – 28 . Google Scholar Crossref Search ADS WorldCat Bakker BM , Overkamp KM, van Maris AJA et al. Stoichiometry and compartmentation of NADH metabolism in Saccharomyces cerevisiae . FEMS Microbiol Rev 2001 ; 25 : 15 – 37 . Google Scholar Crossref Search ADS PubMed WorldCat Basso LC , Basso T, Nitsche Rocha S. Ethanol production in Brazil: the industrial process and its impact on yeast fermentation . Biofuel Production - Recent Developments and Prospects . Rijeka : IntechOpen , 2011 , DOI: 10.5772/17047 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Becker J , Boles E. A modified Saccharomyces cerevisiae strain that consumes l-arabinose and produces ethanol . Appl Environ Microb 2003 ; 69 : 4144 – 50 . Google Scholar Crossref Search ADS WorldCat Becker SA , Feist AM, Mo ML et al. Quantitative prediction of cellular metabolism with constraint-based models: the COBRA Toolbox . Nat Protoc 2007 ; 2 : 727 – 38 . Google Scholar Crossref Search ADS PubMed WorldCat Belgareh-Touzé N , Léon S, Erpapazoglou Z et al. Versatile role of the yeast ubiquitin ligase Rsp5p in intracellular trafficking: Figure 1 . Biochem Soc Trans 2008 ; 36 : 791 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat Bellissimi E , Van Dijken JP, Pronk JT et al. Effects of acetic acid on the kinetics of xylose fermentation by an engineered, xylose-isomerase-based Saccharomyces cerevisiae strain . FEMS Yeast Res 2009 ; 9 : 358 – 64 . Google Scholar Crossref Search ADS PubMed WorldCat Boender LGM , de Hulster EAF, van Maris AJA et al. Quantitative physiology of Saccharomyces cerevisiae at near-zero specific growth rates . Appl Environ Microb 2009 ; 75 : 5607 – 14 . Google Scholar Crossref Search ADS WorldCat Boles E , Lehnert W, Zimmermann FK. The role of the NAD-dependent glutamate dehydrogenase in restoring growth on glucose of a Saccharomyces cerevisiae phosphoglucose isomerase mutant . Eur J Biochem 1993 ; 217 : 469 – 77 . Google Scholar Crossref Search ADS PubMed WorldCat Bruinenberg PM , Van Dijken JP, Scheffers WA. A theoretical analysis of NADPH production and consumption in yeasts . Microbiology 1983 ; 129 : 953 – 64 . Google Scholar Crossref Search ADS WorldCat Canelas AB , Harrison N, Fazio A et al. Integrated multilaboratory systems biology reveals differences in protein metabolism between two reference yeast strains . Nat Commun 2010 ; 1 : 145 . Google Scholar Crossref Search ADS PubMed WorldCat Celton M , Sanchez I, Goelzer A et al. A comparative transcriptomic, fluxomic and metabolomic analysis of the response of Saccharomyces cerevisiae to increases in NADPH oxidation . BMC Genomics 2012 ; 13 : 317 . Google Scholar Crossref Search ADS PubMed WorldCat De Winde JH , Crauwels M, Hohmann S et al. Differential requirement of the yeast sugar kinases for sugar sensing in establishing the catabolite-repressed state . Eur J Biochem 1996 ; 241 : 633 – 43 . Google Scholar Crossref Search ADS PubMed WorldCat DiCarlo JE , Norville JE, Mali P et al. Genome engineering in Saccharomyces cerevisiae using CRISPR-Cas systems . Nucleic Acids Res 2013 ; 41 : 4336 – 43 . Google Scholar Crossref Search ADS PubMed WorldCat Dickinson JR , Sobanski MA, Hewlins MJE. In Saccharomyces cerevisiae deletion of phosphoglucose isomerase can be suppressed by increased activities of enzymes of the hexose monophosphate pathway . Microbiology 1995 ; 141 : 385 – 91 . Google Scholar Crossref Search ADS PubMed WorldCat Diderich JA , Schepper M, van Hoek P et al. Glucose uptake kinetics and transcription of HXTGenes in chemostat cultures of Saccharomyces cerevisiae . J Biol Chem 1999 ; 274 : 15350 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Entian K-D . Genetic and biochemical evidence for hexokinase PII as a key enzyme involved in carbon catabolite repression in yeast . Mol Gen Genet 1980 ; 178 : 633 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat Entian K-D , Fröhlich K-U. Saccharomyces cerevisiae mutants provide evidence of hexokinase PII as a bifunctional enzyme with catalytic and regulatory domains for triggering catabolite repression . J Bacteriol 1984 ; 158 : 29 – 35 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Entian K-D , Kötter P. 25 yeast genetic strain and plasmid collections . In: Stansfield I, Stark MJR (eds.) Methods in Microbiology , vol. 36 . London : Academic Press , 2007 , 629 – 66 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Erickson JR , Johnston M. Genetic and molecular characterization of Gal83: Its interaction and similarities with other genes involved in glucose repression in Saccharomyces cerevisiae . Genetics 1993 ; 135 : 655 – 64 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Farwick A , Bruder S, Schadeweg V et al. Engineering of yeast hexose transporters to transport D-xylose without inhibition by D-glucose . Proc Natl Acad Sci USA 2014 ; 111 : 5159 – 64 . Google Scholar Crossref Search ADS PubMed WorldCat Fulton LM , Lynd LR, Körner A et al. The need for biofuels as part of a low carbon energy future . Biofuels Bioprod Bioref 2015 ; 9 : 476 – 83 . Google Scholar Crossref Search ADS WorldCat Gancedo JM . Yeast carbon catabolite repression . Microbiol Mol Biol R 1998 ; 62 : 334 – 61 . Google Scholar OpenURL Placeholder Text WorldCat Garcia Sanchez R , Karhumaa K, Fonseca C et al. Improved xylose and arabinose utilization by an industrial recombinant Saccharomyces cerevisiae strain using evolutionary engineering . Biotechnol Biofuels 2010 ; 3 : 13 . Google Scholar Crossref Search ADS PubMed WorldCat Gawand P , Hyland P, Ekins A et al. Novel approach to engineer strains for simultaneous sugar utilization . Metab Eng 2013 ; 20 : 63 – 72 . Google Scholar Crossref Search ADS PubMed WorldCat Gietz RD , Woods RA. Transformation of yeast by lithium acetate/single-stranded carrier DNA/polyethylene glycol method . In: Guthrie C, Fink GR (eds.) Methods in Enzymology , vol. 350 : Academic Press , 2002 , 87 – 96 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gombert AK , van Maris AJA. Improving conversion yield of fermentable sugars into fuel ethanol in 1st generation yeast-based production processes . Curr Opin Biotechnol 2015 ; 33 : 81 – 86 . Google Scholar Crossref Search ADS PubMed WorldCat Gonçalves DL , Matsushika A, de Sales BB et al. Xylose and xylose/glucose co-fermentation by recombinant Saccharomyces cerevisiae strains expressing individual hexose transporters . Enzyme Microb Technol 2014 ; 63 : 13 – 20 . Google Scholar Crossref Search ADS PubMed WorldCat Gottardi M , Reifenrath M, Boles E et al. Pathway engineering for the production of heterologous aromatic chemicals and their derivatives in Saccharomyces cerevisiae: bioconversion from glucose . FEMS Yeast Res 2017 ; 17 : fox035 . Google Scholar OpenURL Placeholder Text WorldCat Grabowska D , Chelstowska A. The ALD6 gene product is indispensable for providing NADPH in yeast cells lacking glucose-6-phosphate dehydrogenase activity . J Biol Chem 2003 ; 278 : 13984 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat Haguenauer-Tsapis R , André B. Membrane trafficking of yeast transporters: mechanisms and physiological control of downregulation . In: Boles E, Krämer R (eds). Molecular Mechanisms Controlling Transmembrane Transport . Berlin, Heidelberg : Springer , 2004 , 273 – 323 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hamacher T , Becker J, Gárdonyi M et al. Characterization of the xylose-transporting properties of yeast hexose transporters and their influence on xylose utilization . Microbiology 2002 ; 148 : 2783 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat Heux S , Cadiere A, Dequin S. Glucose utilization glucose utilization of strains lacking PGI1 and expressing a transhydrogenase suggests differences in the pentose phosphate capacity among Saccharomyces cerevisiae strains . FEMS Yeast Res 2008 ; 8 : 217 – 24 . Google Scholar Crossref Search ADS PubMed WorldCat Hucka M , Finney A, Sauro HM et al. The systems biology markup language (SBML): a medium for representation and exchange of biochemical network models . Bioinformatics 2003 ; 19 : 524 – 31 . Google Scholar Crossref Search ADS PubMed WorldCat Huibregtse JM , Yang JC, Beaudenon SL. The large subunit of RNA polymerase II is a substrate of the Rsp5 ubiquitin-protein ligase . Proc Natl Acad Sci USA 1997 ; 94 : 3656 – 61 . Google Scholar Crossref Search ADS PubMed WorldCat Jansen MLA , Bracher JM, Papapetridis I et al. Saccharomyces cerevisiae strains for second-generation ethanol production: from academic exploration to industrial implementation . FEMS Yeast Res 2017 ; 17 : fox044 . Google Scholar OpenURL Placeholder Text WorldCat Johansson B , Hahn-Hägerdal B. The non-oxidative pentose phosphate pathway controls the fermentation rate of xylulose but not of xylose in Saccharomyces cerevisiae TMB3001 . FEMS Yeast Res 2002 ; 2 : 277 – 82 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Kaliszewski P , Zoładek T. The role of Rsp5 ubiquitin ligase in regulation of diverse processes in yeast cells . Acta Biochim Pol 2008 ; 55 : 649 – 62 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Katahira S , Ito M, Takema H et al. Improvement of ethanol productivity during xylose and glucose co-fermentation by xylose-assimilating S. cerevisiae via expression of glucose transporter Sut1 . Enzyme Microb Technol 2008 ; 43 : 115 – 9 . Google Scholar Crossref Search ADS WorldCat Kim SR , Ha S-J, Wei N et al. Simultaneous co-fermentation of mixed sugars: a promising strategy for producing cellulosic ethanol . Trends Biotechnol 2012 ; 30 : 274 – 82 . Google Scholar Crossref Search ADS PubMed WorldCat Klinke HB , Ahring BK, Schmidt AS et al. Characterization of degradation products from alkaline wet oxidation of wheat straw . Bioresource Technol 2002 ; 82 : 15 – 26 . Google Scholar Crossref Search ADS WorldCat Klug L , Daum G. Yeast lipid metabolism at a glance . FEMS Yeast Res 2014 ; 14 : 369 – 88 . Google Scholar Crossref Search ADS PubMed WorldCat Krampe S , Stamm O, Hollenberg CP et al. Catabolite inactivation of the high-affinity hexose transporters Hxt6 and Hxt7 of Saccharomyces cerevisiae occurs in the vacuole after internalization by endocytosis . FEBS Lett 1998 ; 441 : 343 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat Kuijpers NG , Solis-Escalante D, Bosman L et al. A versatile, efficient strategy for assembly of multi-fragment expression vectors in Saccharomyces cerevisiae using 60 bp synthetic recombination sequences . Microb Cell Fact 2013 ; 12 : 47 . Google Scholar Crossref Search ADS PubMed WorldCat Kuyper M , Harhangi HR, Stave AK et al. High-level functional expression of a fungal xylose isomerase: the key to efficient ethanolic fermentation of xylose by Saccharomyces cerevisiae ? FEMS Yeast Res 2003 ; 4 : 69 – 78 . Google Scholar Crossref Search ADS PubMed WorldCat Kuyper M , Hartog MMP, Toirkens MJ et al. Metabolic engineering of a xylose-isomerase-expressing strain for rapid anaerobic xylose fermentation . FEMS Yeast Res 2005a ; 5 : 399 – 409 . Google Scholar Crossref Search ADS WorldCat Kuyper M , Toirkens MJ, Diderich JA et al. Evolutionary engineering of mixed-sugar utilization by a xylose-fermenting strain . FEMS Yeast Res 2005b ; 5 : 925 – 34 . Google Scholar Crossref Search ADS WorldCat Lee W-J , Kim M-D, Ryu Y-W et al. Kinetic studies on glucose and xylose transport in Saccharomyces cerevisiae . Appl Microbiol Biot 2002 ; 60 : 186 – 91 . Google Scholar Crossref Search ADS WorldCat Lopes ML , Paulillo SCdL, Godoy A et al. Ethanol production in Brazil: a bridge between science and industry . Braz J Microbiol 2016 ; 47 : 64 – 76 . Google Scholar Crossref Search ADS PubMed WorldCat Lynd LR , Liang X, Biddy MJ et al. Cellulosic ethanol: status and innovation . Curr Opin Biotechnol 2017 ; 45 : 202 – 11 . Google Scholar Crossref Search ADS PubMed WorldCat Mangat S , Chandrashekarappa D, McCartney RR et al. Differential roles of the glycogen-binding domains of beta subunits in regulation of the Snf1 kinase complex . Eukaryot Cell 2010 ; 9 : 173 – 83 . Google Scholar Crossref Search ADS PubMed WorldCat Mans R , van Rossum HM, Wijsman M et al. CRISPR/Cas9: a molecular Swiss army knife for simultaneous introduction of multiple genetic modifications in Saccharomyces cerevisiae . FEMS Yeast Res 2015 ; 15 : fov004 . Google Scholar Crossref Search ADS PubMed WorldCat Matsumoto K , Toh-e A, Oshima Y. Isolation and characterization of dominant mutations resistant to carbon catabolite repression of galactokinase synthesis in Saccharomyces cerevisiae. Mol Cell Biol 1981 ; 1 : 83 – 93 . Google Scholar Crossref Search ADS PubMed WorldCat Miskovic L , Alff-Tuomala S, Soh KC et al. A design–build–test cycle using modeling and experiments reveals interdependencies between upper glycolysis and xylose uptake in recombinant S. cerevisiae and improves predictive capabilities of large-scale kinetic models . Biotechnol Biofuels 2017 ; 10 : 166 . Google Scholar Crossref Search ADS PubMed WorldCat Momcilovic M , Iram SH, Liu Y et al. Roles of the glycogen-binding domain and Snf4 in glucose inhibition of SNF1 protein kinase . J Biol Chem 2008 ; 283 : 19521 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Moysés D , Reis V, Almeida J et al. Xylose fermentation by Saccharomyces cerevisiae: challenges and prospects . Int J Mol Sci 2016 ; 17 : 207 . Google Scholar Crossref Search ADS PubMed WorldCat Nijkamp JF , van den Broek M, Datema E et al. De novo sequencing, assembly and analysis of the genome of the laboratory strain Saccharomyces cerevisiae CEN.PK113-7D, a model for modern industrial biotechnology . Microb Cell Fact 2012 ; 11 : 36 . Google Scholar Crossref Search ADS PubMed WorldCat Nijland JG , Shin HY, Boender LGM et al. Improved xylose metabolism by a CYC8 mutant of Saccharomyces cerevisiae . Appl Environ Microb 2017 ; 83 : 95 – 117 . Google Scholar Crossref Search ADS WorldCat Nijland JG , Shin HY, de Jong RM et al. Engineering of an endogenous hexose transporter into a specific D-xylose transporter facilitates glucose-xylose co-consumption in Saccharomyces cerevisiae . Biotechnol Biofuels 2014 ; 7 : 168 . Google Scholar Crossref Search ADS PubMed WorldCat Nijland JG , Vos E, Shin HY et al. Improving pentose fermentation by preventing ubiquitination of hexose transporters in Saccharomyces cerevisiae . Biotechnol Biofuels 2016 ; 9 : 158 . Google Scholar Crossref Search ADS PubMed WorldCat Özcan S , Johnston M. Function and regulation of yeast hexose transporters . Microbiol Mol Biol R 1999 ; 63 : 554 – 69 . Google Scholar OpenURL Placeholder Text WorldCat Palmqvist E , Hahn-Hägerdal B. Fermentation of lignocellulosic hydrolysates. II: inhibitors and mechanisms of inhibition . Bioresource Technol 2000 ; 74 : 25 – 33 . Google Scholar Crossref Search ADS WorldCat Papapetridis I , Goudriaan M, Vázquez Vitali M et al. Optimizing anaerobic growth rate and fermentation kinetics in Saccharomyces cerevisiae strains expressing Calvin-cycle enzymes for improved ethanol yield . Biotechnol Biofuels 2018 ; 11 : 17 . Google Scholar Crossref Search ADS PubMed WorldCat Papapetridis I , van Dijk M, Dobbe APA et al. Improving ethanol yield in acetate-reducing Saccharomyces cerevisiae by cofactor engineering of 6-phosphogluconate dehydrogenase and deletion of ALD6 . Microb Cell Fact 2016 ; 15 : 67 . Google Scholar Crossref Search ADS PubMed WorldCat Peláez R , Herrero P, Moreno F. Functional domains of yeast hexokinase 2 . Biochem J 2010 ; 432 : 181 – 90 . Google Scholar Crossref Search ADS PubMed WorldCat Petit T , Diderich JA, Kruckeberg AL et al. Hexokinase regulates kinetics of glucose transport and expression of genes encoding hexose transporters in Saccharomyces cerevisiae . J Bacteriol 2000 ; 182 : 6815 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat Raamsdonk LM , Diderich JA, Kuiper A et al. Co-consumption of sugars or ethanol and glucose in a Saccharomyces cerevisiae strain deleted in theHXK2 gene . Yeast 2001 ; 18 : 1023 – 33 . Google Scholar Crossref Search ADS PubMed WorldCat Rahner A , Schöler A, Martens E et al. Dual influence of the yeast Catlp (Snflp) protein kinase on carbon source-dependent transcriptional activation of gluconeogenic genes by the regulatory gene CAT8 . Nucleic Acids Res 1996 ; 24 : 2331 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat Reifenberger E , Boles E, Ciriacy M. Kinetic characterization of individual hexose transporters of Saccharomyces cerevisiae and their relation to the triggering mechanisms of glucose repression . Eur J Biochem 1997 ; 245 : 324 – 33 . Google Scholar Crossref Search ADS PubMed WorldCat Reznicek O , Facey SJ, de Waal PP et al. Improved xylose uptake in Saccharomyces cerevisiae due to directed evolution of galactose permease Gal2 for sugar co-consumption . J Appl Microbiol 2015 ; 119 : 99 – 111 . Google Scholar Crossref Search ADS PubMed WorldCat Schmidt MC , McCartney RR. beta-subunits of Snf1 kinase are required for kinase function and substrate definition . EMBO J 2000 ; 19 : 4936 – 43 . Google Scholar Crossref Search ADS PubMed WorldCat Schüller H-J . Transcriptional control of nonfermentative metabolism in the yeast Saccharomyces cerevisiae . Curr Genet 2003 ; 43 : 139 – 60 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Shcherbik N , Pestov DG. The ubiquitin ligase Rsp5 is required for ribosome stability in Saccharomyces cerevisiae . RNA 2011 ; 17 : 1422 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat Shen M-H , Song H, Li B-Z et al. Deletion of D-ribulose-5-phosphate 3-epimerase (RPE1) induces simultaneous utilization of xylose and glucose in xylose-utilizing Saccharomyces cerevisiae . Biotechnol Lett 2015 ; 37 : 1031 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat Shen Y , Chen X, Peng B et al. An efficient xylose-fermenting recombinant Saccharomyces cerevisiae strain obtained through adaptive evolution and its global transcription profile . Appl Microbiol Biot 2012 ; 96 : 1079 – 91 . Google Scholar Crossref Search ADS WorldCat Shin HY , Nijland JG, de Waal PP et al. An engineered cryptic Hxt11 sugar transporter facilitates glucose–xylose co-consumption in Saccharomyces cerevisiae . Biotechnol Biofuels 2015 ; 8 : 176 . Google Scholar Crossref Search ADS PubMed WorldCat Solis-Escalante D , Kuijpers NGA, Bongaerts N et al. amdSYM, a new dominant recyclable marker cassette for Saccharomyces cerevisiae . FEMS Yeast Res 2013 ; 13 : 126 – 39 . Google Scholar Crossref Search ADS PubMed WorldCat Subtil T , Boles E. Competition between pentoses and glucose during uptake and catabolism in recombinant Saccharomyces cerevisiae . Biotechnol Biofuels 2012 ; 5 : 14 . Google Scholar Crossref Search ADS PubMed WorldCat Tehlivets O , Scheuringer K, Kohlwein SD. Fatty acid synthesis and elongation in yeast . BBA-Mol Cell Biol L 2007 ; 1771 : 255 – 70 . Google Scholar Crossref Search ADS WorldCat van den Brink J , Akeroyd M, van der Hoeven R et al. Energetic limits to metabolic flexibility: responses of Saccharomyces cerevisiae to glucose-galactose transitions . Microbiology 2009 ; 155 : 1340 – 50 . Google Scholar Crossref Search ADS PubMed WorldCat van Maris AJA , Abbott DA, Bellissimi E et al. Alcoholic fermentation of carbon sources in biomass hydrolysates by Saccharomyces cerevisiae: current status . Anton Leeuw 2006 ; 90 : 391 – 418 . Google Scholar Crossref Search ADS WorldCat Verduyn C , Postma E, Scheffers WA et al. Physiology of Saccharomyces cerevisiae in anaerobic glucose-limited chemostat cultures . Microbiology 1990 ; 136 : 395 – 403 . Google Scholar OpenURL Placeholder Text WorldCat Verduyn C , Postma E, Scheffers WA et al. Effect of benzoic acid on metabolic fluxes in yeasts: a continuous-culture study on the regulation of respiration and alcoholic fermentation . Yeast 1992 ; 8 : 501 – 17 . Google Scholar Crossref Search ADS PubMed WorldCat Verhoeven MD , Lee M, Kamoen L et al. Mutations in PMR1 stimulate xylose isomerase activity and anaerobic growth on xylose of engineered Saccharomyces cerevisiae by influencing manganese homeostasis . Sci Rep 2017 ; 7 : 46155 . Google Scholar Crossref Search ADS PubMed WorldCat Vincent O , Carlson M. Sip4, a Snf1 kinase-dependent transcriptional activator, binds to the carbon source-responsive element of gluconeogenic genes . EMBO J 1998 ; 17 : 7002 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat Vincent O , Carlson M. Gal83 mediates the interaction of the Snf1 kinase complex with the transcription activator Sip4 . EMBO J 1999 ; 18 : 6672 – 81 . Google Scholar Crossref Search ADS PubMed WorldCat Vincent O , Townley R, Kuchin S et al. Subcellular localization of the Snf1 kinase is regulated by specific beta subunits and a novel glucose signaling mechanism . Gene Dev 2001 ; 15 : 1104 – 14 . Google Scholar Crossref Search ADS PubMed WorldCat Wei N , Oh EJ, Million G et al. Simultaneous utilization of cellobiose, xylose, and acetic acid from lignocellulosic biomass for biofuel production by an engineered yeast platform . ACS Synth Biol 2015 ; 4 : 707 – 13 . Google Scholar Crossref Search ADS PubMed WorldCat Westman JO , Bonander N, Taherzadeh MJ et al. Improved sugar co-utilisation by encapsulation of a recombinant Saccharomyces cerevisiae strain in alginate-chitosan capsules . Biotechnol Biofuels 2014 ; 7 : 102 . Google Scholar Crossref Search ADS PubMed WorldCat Wiatrowski HA , van Denderen BJW, Berkey CD et al. Mutations in the Gal83 glycogen-binding domain activate the Snf1/Gal83 kinase pathway by a glycogen-independent mechanism . Mol Cell Biol 2004 ; 24 : 352 – 61 . Google Scholar Crossref Search ADS PubMed WorldCat Wisselink HW , Toirkens MJ, del Rosario Franco Berriel M et al. Engineering of Saccharomyces cerevisiae for efficient anaerobic alcoholic fermentation of l-arabinose . Appl Environ Microb 2007 ; 73 : 4881 – 91 . Google Scholar Crossref Search ADS WorldCat Wisselink HW , Toirkens MJ, Wu Q et al. Novel evolutionary engineering approach for accelerated utilization of glucose, xylose, and arabinose mixtures by engineered Saccharomyces cerevisiae strains . Appl Environ Microb 2009 ; 75 : 907 – 14 . Google Scholar Crossref Search ADS WorldCat Wisselink HW , van Maris AJA, Pronk JT. WO/2014/195522.Polypeptides with permease activity . 2015 . Zhou H , Cheng J-s, Wang BL et al. Xylose isomerase overexpression along with engineering of the pentose phosphate pathway and evolutionary engineering enable rapid xylose utilization and ethanol production by Saccharomyces cerevisiae . Metab Eng 2012 ; 14 : 611 – 22 . Google Scholar Crossref Search ADS PubMed WorldCat Author notes The authors have contributed equally and should be considered as joint first authors. Current address: School of Engineering Sciences in Chemistry, Biotechnology and Health, Department of Industrial Biotechnology, KTH Royal Institute of Technology, AlbaNova University Centre, SE 10691 Stockholm, Sweden © FEMS 2018. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs licence (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial reproduction and distribution of the work, in any medium, provided the original work is not altered or transformed in any way, and that the work is properly cited. For commercial re-use, please contact [email protected] © FEMS 2018. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png FEMS Yeast Research Oxford University Press

Laboratory evolution for forced glucose-xylose co-consumption enables identification of mutations that improve mixed-sugar fermentation by xylose-fermenting Saccharomyces cerevisiae

Download PDF
17 pages

Loading next page...
 
/lp/ou_press/laboratory-evolution-for-forced-glucose-xylose-co-consumption-enables-xLoflJukSu

References (111)

Publisher
Oxford University Press
Copyright
Copyright © 2022 Federation of European Microbiological Societies
ISSN
1567-1356
eISSN
1567-1364
DOI
10.1093/femsyr/foy056
Publisher site
See Article on Publisher Site

Abstract

Abstract Simultaneous fermentation of glucose and xylose can contribute to improved productivity and robustness of yeast-based processes for bioethanol production from lignocellulosic hydrolysates. This study explores a novel laboratory evolution strategy for identifying mutations that contribute to simultaneous utilisation of these sugars in batch cultures of Saccharomyces cerevisiae. To force simultaneous utilisation of xylose and glucose, the genes encoding glucose-6-phosphate isomerase (PGI1) and ribulose-5-phosphate epimerase (RPE1) were deleted in a xylose-isomerase-based xylose-fermenting strain with a modified oxidative pentose-phosphate pathway. Laboratory evolution of this strain in serial batch cultures on glucose–xylose mixtures yielded mutants that rapidly co-consumed the two sugars. Whole-genome sequencing of evolved strains identified mutations in HXK2, RSP5 and GAL83, whose introduction into a non-evolved xylose-fermenting S. cerevisiae strain improved co-consumption of xylose and glucose under aerobic and anaerobic conditions. Combined deletion of HXK2 and introduction of a GAL83G673T allele yielded a strain with a 2.5-fold higher xylose and glucose co-consumption ratio than its xylose-fermenting parental strain. These two modifications decreased the time required for full sugar conversion in anaerobic bioreactor batch cultures, grown on 20 g L−1 glucose and 10 g L−1 xylose, by over 24 h. This study demonstrates that laboratory evolution and genome resequencing of microbial strains engineered for forced co-consumption is a powerful approach for studying and improving simultaneous conversion of mixed substrates. biofuels, yeast, industrial biotechnology, redox engineering, fermentation, pentoses INTRODUCTION Industrial biotechnology can contribute to reconciliating global demands for liquid transport fuels with the almost universally accepted need to limit anthropogenic CO2 emissions (Fulton et al.2015). Bioethanol, the biofuel with the largest annual global production volume, is still predominantly produced by fermentation of sucrose or glucose, derived from sugar cane or corn starch, respectively (Gombert and van Maris 2015). These ‘first-generation’ bioethanol processes exploit the natural high fermentation rates and ethanol yield of the yeast Saccharomyces cerevisiae. Optimisation of yeast strains and production processes enables many industrial processes to operate at >90% of the theoretical ethanol yield on sugar (Basso, Basso and Nitsche Rocha 2011; Gombert and van Maris 2015; Lopes et al.2016). However, the massive scaling up of production volumes that would be required to replace a substantial fraction of petroleum-based transport fuels cannot be sustainably achieved with corn starch and cane sugar as the only feedstocks. Instead, a large fraction of the feedstock will have to be derived from lignocellulosic plant biomass, such as agricultural residues and energy crops (Lynd et al.2017). In comparison with first-generation feedstocks, lignocellulosic hydrolysates pose additional challenges for yeasts and yeast researchers. In addition to containing mixtures of hexose and pentose (mainly D-xylose and L-arabinose) sugars, the deconstruction of lignocellulosic biomass that precedes yeast-based fermentation releases fermentation inhibitors (Palmqvist and Hahn-Hägerdal 2000; van Maris et al.2006; Jansen et al.2017). Intensive metabolic engineering studies, encompassing functional expression of heterologous pathways for xylose and arabinose catabolism, improvements in inhibitor tolerance and minimisation of by-product formation have yielded S. cerevisiae strains that are now applied in the first full-scale ‘second-generation’ industrial bioethanol plants (Moysés et al.2016; Jansen et al.2017). However, further improvements in ethanol titres, yields and productivities are important to increase the economic viability of this nascent technology. Current strain engineering strategies for enabling pentose fermentation by S. cerevisiae typically yield strains that, in anaerobic batch cultures grown on sugar mixtures, preferentially ferment glucose, while xylose and/or arabinose are predominantly converted in a second, slower fermentation phase (Jansen et al.2017). This strong preference for glucose over pentoses persists even after extensive laboratory evolution on sugar mixtures (Kuyper et al.2005b; Wisselink et al.2009; Garcia Sanchez et al.2010; Shen et al.2012; Zhou et al.2012). Achieving efficient co-fermentation of glucose and pentoses, while maintaining a high overall rate of sugar conversion, could increase volumetric productivity of industrial processes. Moreover, since several inhibitors of yeast performance are more harmful during the slower pentose fermentation phase (Bellissimi et al.2009; Kim et al.2012; Ask et al.2013; Jansen et al.2017), simultaneous fermentation of glucose and pentose sugars can also contribute to robustness under industrial process conditions. Random mutagenesis, laboratory evolution and protein engineering of xylose transporters have yielded transporter variants with improved xylose affinity and reduced glucose inhibition, which enabled the construction of yeast strains with improved xylose consumption in the presence of glucose (Farwick et al.2014; Nijland et al.2014; Reznicek et al.2015; Shin et al.2015). In an alternative approach, expression of a heterologous cellodextrin transporter and an intracellular β-glucosidase, along with a heterologous xylose reductase/xylitol dehydrogenase (XR/XDH) pathway, enabled simultaneous consumption of cellobiose and xylose in S. cerevisiae by reducing the impact of glucose repression (Wei et al.2015). However, despite progress in this area, engineering of yeast strains showing simultaneous, fast fermentation of glucose and xylose remains a key challenge. Deletion of RPE1, which encodes the pentose-phosphate-pathway (PPP) enzyme ribulose-5-phosphate epimerase, was recently shown to result in coupling of glucose and xylose catabolism, at a ratio of 10:1, in an engineered xylose-utilising S. cerevisiae strain (Shen et al.2015). Despite a low xylose-to-glucose consumption ratio, this strategy indicated the potential of forced stoichiometric coupling of glucose and pentose metabolism in S. cerevisiae. In S. cerevisiae, phosphoglucose isomerase (Pgi1) catalyses interconversion of glucose-6-phosphate to fructose-6-phosphate in upper glycolysis (Aguilera 1986). Since deletion of PGI1 blocks glycolysis, pgi1Δ strains cannot grow on glucose as the sole carbon source unless all glucose-6-phosphate is rerouted through the PPP. Deletion of eda, rpe and pgi in Escherichia coli was previously shown to enable co-consumption of xylose and glucose (Gawand et al.2013). As conversion of 1 mol glucose-6-phosphate to 1 mol ribulose-5-phosphate via the oxidative branch of the PPP results in a net generation of 2 mol NADPH, the absence of a redox imbalance relied on conversion of excess NADPH to NADH by the native E. coli transhydrogenases. Wild-type S. cerevisiae strains cannot reoxidise all NADPH generated by such a redirection of metabolism and, consequently, pgi1-null mutants cannot grow on glucose as sole carbon source (Boles, Lehnert and Zimmermann 1993; Dickinson, Sobanski and Hewlins 1995). Overexpression of GDH2, which encodes NAD+-dependent glutamate dehydrogenase, can restore growth of pgi1Δ strains by enabling a transhydrogenase-like cycle that couples the interconversion of 2-oxoglutarate and glutamate to the conversion of NADPH and NAD+ to NADP+ and NADH (Boles, Lehnert and Zimmermann 1993). Based on the impact of a pgi1Δ mutation on glucose metabolism, we reasoned that inactivation of PGI1 might be used to construct strains with a stringent requirement for co-utilisation of xylose and glucose, at much higher ratios than hitherto demonstrated. The goal of this study was to explore a new strategy for identifying mutations that stimulate glucose-xylose co-consumption by S. cerevisiae. The strategy was based on enforcing a strict stoichiometric coupling of glucose and xylose fermentation by the combined deletion of RPE1 and PGI1 in an engineered, xylose-isomerase-based S. cerevisiae strain (Kuyper et al.2005a; Verhoeven et al.2017). Furthermore, to reduce the impact of these modifications on NADP+/NADPH redox cofactor balancing, the native NADP+-dependent 6-phosphogluconate dehydrogenases (Gnd1 and Gnd2) were replaced by a heterologous NAD+-dependent enzyme (Papapetridis et al.2016). After laboratory evolution for improved growth on glucose-xylose mixtures, the physiology of evolved strains was analysed in aerobic shake-flask and bioreactor batch cultures. Potential causal mutations identified by whole-genome sequencing were introduced into a non-evolved (PGI1 RPE1 GND1 GND2) xylose-fermenting strain background. The resulting reverse engineered strains were then analysed in shake-flask and anaerobic bioreactor batch cultures, grown on mixtures of glucose and xylose. MATERIALS AND METHODS Maintenance of strains The CEN.PK lineage of S. cerevisiae laboratory strains (Entian and Kötter 2007) was used to construct and evolve all strains used in this study (Table 1, Additional File 1). Depending on strain auxotrophies, cultures were grown in YP (10 g L−1 yeast extract, 20 g L−1 peptone) (BD, Franklin Lakes, NJ) or synthetic medium (SM) (Verduyn et al.1992), supplemented with glucose (20 g L−1), xylose (20 g L−1), a glucose/xylose mixture (10 g L−1 of each sugar) or a xylose/fructose/glucose mixture (20, 10 and 1 g L−1 respectively). Propagation of E. coli XL-1 Blue cultures was performed in LB medium (5 g L−1 Bacto yeast extract, 10 g L−1 Bacto tryptone, 5 g L−1 NaCl, 100 μg mL−1 ampicillin). Frozen stock cultures were stored at –80°C, after addition of glycerol (30% v/v final concentration). Table 1. Strains used in this study. Strain name . Relevant Genotype . Origin . CEN.PK113–7D MATa MAL2–8c SUC2 CAN1 Entian and Kötter (2007) IMX581 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2 Mans et al. (2015) IMX705 MATa MAL2–8c SUC2 can1::cas9-natNT2 gnd2Δ gnd1::gndA Papapetridis et al. (2016) IMX1046 MATa MAL2–8c SUC2 can1::cas9-natNT2 gnd2Δ gnd1::gndA gre3::ZWF1, SOL3, TKL1, TAL1, NQM1, RKI1, TKL2 sga1::9*xylA, XKS1 rpe1Δ pgi1Δ This work IMX994 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2, gre3::RPE1, TKL1, TAL1, RKI1, XKS1 This work IMU079 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2, gre3::RPE1, TKL1, TAL1, RKI1, XKS1 pAKX002 This work IMX1485 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RRKI1, XKS1 hxk2Δ pAKX002 This work IMX1486 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83Δ pAKX002 This work IMX1487 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 rsp5Δ pAKX002 This work IMX1488 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83::GAL83G673T pAKX002 This work IMX1515 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 rsp5Δ hxk2Δ pAKX002 This work IMX1583 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83::GAL83G673T hxk2Δ pAKX002 This work Strain name . Relevant Genotype . Origin . CEN.PK113–7D MATa MAL2–8c SUC2 CAN1 Entian and Kötter (2007) IMX581 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2 Mans et al. (2015) IMX705 MATa MAL2–8c SUC2 can1::cas9-natNT2 gnd2Δ gnd1::gndA Papapetridis et al. (2016) IMX1046 MATa MAL2–8c SUC2 can1::cas9-natNT2 gnd2Δ gnd1::gndA gre3::ZWF1, SOL3, TKL1, TAL1, NQM1, RKI1, TKL2 sga1::9*xylA, XKS1 rpe1Δ pgi1Δ This work IMX994 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2, gre3::RPE1, TKL1, TAL1, RKI1, XKS1 This work IMU079 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2, gre3::RPE1, TKL1, TAL1, RKI1, XKS1 pAKX002 This work IMX1485 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RRKI1, XKS1 hxk2Δ pAKX002 This work IMX1486 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83Δ pAKX002 This work IMX1487 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 rsp5Δ pAKX002 This work IMX1488 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83::GAL83G673T pAKX002 This work IMX1515 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 rsp5Δ hxk2Δ pAKX002 This work IMX1583 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83::GAL83G673T hxk2Δ pAKX002 This work Open in new tab Table 1. Strains used in this study. Strain name . Relevant Genotype . Origin . CEN.PK113–7D MATa MAL2–8c SUC2 CAN1 Entian and Kötter (2007) IMX581 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2 Mans et al. (2015) IMX705 MATa MAL2–8c SUC2 can1::cas9-natNT2 gnd2Δ gnd1::gndA Papapetridis et al. (2016) IMX1046 MATa MAL2–8c SUC2 can1::cas9-natNT2 gnd2Δ gnd1::gndA gre3::ZWF1, SOL3, TKL1, TAL1, NQM1, RKI1, TKL2 sga1::9*xylA, XKS1 rpe1Δ pgi1Δ This work IMX994 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2, gre3::RPE1, TKL1, TAL1, RKI1, XKS1 This work IMU079 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2, gre3::RPE1, TKL1, TAL1, RKI1, XKS1 pAKX002 This work IMX1485 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RRKI1, XKS1 hxk2Δ pAKX002 This work IMX1486 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83Δ pAKX002 This work IMX1487 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 rsp5Δ pAKX002 This work IMX1488 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83::GAL83G673T pAKX002 This work IMX1515 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 rsp5Δ hxk2Δ pAKX002 This work IMX1583 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83::GAL83G673T hxk2Δ pAKX002 This work Strain name . Relevant Genotype . Origin . CEN.PK113–7D MATa MAL2–8c SUC2 CAN1 Entian and Kötter (2007) IMX581 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2 Mans et al. (2015) IMX705 MATa MAL2–8c SUC2 can1::cas9-natNT2 gnd2Δ gnd1::gndA Papapetridis et al. (2016) IMX1046 MATa MAL2–8c SUC2 can1::cas9-natNT2 gnd2Δ gnd1::gndA gre3::ZWF1, SOL3, TKL1, TAL1, NQM1, RKI1, TKL2 sga1::9*xylA, XKS1 rpe1Δ pgi1Δ This work IMX994 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2, gre3::RPE1, TKL1, TAL1, RKI1, XKS1 This work IMU079 MATa ura3–52 MAL2–8c SUC2 can1::cas9-natN2, gre3::RPE1, TKL1, TAL1, RKI1, XKS1 pAKX002 This work IMX1485 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RRKI1, XKS1 hxk2Δ pAKX002 This work IMX1486 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83Δ pAKX002 This work IMX1487 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 rsp5Δ pAKX002 This work IMX1488 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83::GAL83G673T pAKX002 This work IMX1515 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 rsp5Δ hxk2Δ pAKX002 This work IMX1583 MATa ura3–52 MAL2–8c SUC2 gre3::RPE1, TKL1, TAL1, RKI1, XKS1 gal83::GAL83G673T hxk2Δ pAKX002 This work Open in new tab Construction of plasmids and cassettes PCR amplification for construction of plasmid fragments and yeast integration cassettes was performed with Phusion High Fidelity DNA Polymerase (Thermo-Scientific, Waltham, MA), according to the manufacturer's guidelines. Plasmid assembly was performed in vitro with a Gibson Assembly Cloning kit (New England Biolabs, Ipswich, MA), following the supplier's guidelines, or in vivo by transformation of plasmid fragments into yeast cells (Kuijpers et al.2013). For all constructs, correct assembly was confirmed by diagnostic PCR with DreamTaq polymerase (Thermo-Scientific), following the manufacturer's protocol. Plasmids used and constructed in this work are described in Additional File 2. All yeast genetic modifications were performed using CRISPR/Cas9-based genome editing (Mans et al.2015). Unique guide-RNA (gRNA) sequences targeting GRE3, GAL83 and RSP5 were selected from a publicly available list (DiCarlo et al.2013) and synthesised (Baseclear, Leiden, The Netherlands). Primers and oligonucleotides used in this work are listed in Additional File 3. To construct the GRE3-targeting CRISPR-plasmid pUDR204, the plasmid backbone of pMEL11 was PCR amplified using primer combination 5980/5792. The insert fragment, expressing the GRE3-targeting gRNA, was amplified using primer combination 5979/5978 and pMEL11 as template. To construct the RPE1/PGI1 double-targeting CRISPR-plasmid pUDR202, the plasmid backbone and the insert fragment were PCR amplified using primer combinations 5941/6005 and 9269/9401, respectively, using pROS11 as template. Both plasmids were assembled in vitro and cloned in E. coli. To construct CRISPR-plasmids for single deletion of GAL83 and RSP5, the plasmid backbone, the GAL83-gRNA insert and the RSP5-gRNA insert were amplified using primer combination 5792/5980, 5979/11 270 and 5979/11 373, respectively, using pMEL10 as template and assembled in vivo. To generate ZWF1 and SOL3 overexpression cassettes, promoter regions of ADH1 and ENO1 and the coding regions of ZWF1 and SOL3 (including their terminator regions) were PCR amplified using primer combinations 8956/8960, 8958/8961, 8953/8964 and 8984/8986, respectively, using CEN.PK113–7D genomic DNA as a template. The resulting products were used as templates for fusion-PCR assembly of the pADH1-ZWF1-tZWF1 and pENO1-SOL3-tSOL3 overexpression cassettes with primer combinations 8956/8964 and 8958/8986, respectively, which yielded plasmids pUD426 and pUD427 after ligation to pJET-blunt vectors (Thermo-Scientific) and cloning in E. coli. To generate yeast-integration cassettes for overexpression of the major genes of the complete PPP, pADH1-ZWF1-tZWF1, pENO1-SOL3-tSOL3, pPGK1-TKL1-tTKL1, pTEF1-TAL1-tTAL1, pPGI1-NQM1-tNQM1, pTPI1-RKI1-tRKI1 and pPYK1-TKL2-tTKL2 cassettes were PCR amplified using primer combinations 4870/7369, 8958/3290, 3291/4068, 3274/3275, 3847/3276, 4691/3277 and 3283/3288, respectively, using plasmids pUD426, pUD427, pUD348, pUD349, pUD344, pUD345 and pUD346, respectively, as templates. To generate yeast-integration cassettes of the genes of the non-oxidative PPP, the pTDH3-RPE1-tRPE1, pPGK1-TKL1-tTKL1, pTEF1-TAL1-tTAL1, pTPI1-RKI1-tRKI1 overexpression cassettes were PCR-amplified using primer pairs 7133/3290, 3291/4068, 3724/3725 and 10 460/10 461, respectively and plasmids pUD347, pUD348, pUD34 and pUD345 as templates. Yeast-integration cassettes for overexpression of Piromyces sp. xylose isomerase (pTPI1-xylA-tCYC1) were PCR-amplified using primer combinations 6285/7548, 6280/6273, 6281/6270, 6282/6271, 6284/6272, 6283/6275, 6287/6276, 6288/6277 or 6289/6274, using pUD350 as template. Yeast xylulokinase overexpression cassettes (pTEF1-XKS1-tXKS1) were PCR-amplified from plasmid pUD353, using primer combination 5920/9029 or 7135/7222. A yeast-integration cassette of pGAL83-gal83::GAL83G673T-tGAL83 was PCR-amplified from genomic DNA of IMS0629, using primer combination 11 273/11 274. Strain construction Yeast transformation was performed as previously described (Gietz and Woods 2002). Transformation mixtures were plated on SM or YP agar plates (2% Bacto Agar, BD), supplemented with the appropriate carbon sources. For transformations with the amdS marker cassette, agar plates were prepared and counter selection was performed as previously described (Solis-Escalante et al.2013). For transformations with the URA3 selection marker counterselection was performed using 5-fluoro-orotic acid (Zymo Research, Irvine, CA), following the supplier's protocol. For transformations with the hphNT marker, agar plates were additionally supplemented with 200 mg L−1 hygromycin B (Invivogen, San Diego, CA) and plasmid loss was induced by cultivation in non-selective medium. After each transformation, correct genotypes were confirmed by diagnostic PCR using DreamTaq polymerase (Thermo-Scientific, see Additional File 3 for primer sequences). Co-transformation of pUDR204 along with the pADH1-ZWF1-tZWF1, pENO1-SOL3-tSOL3, pPGK1-TKL1-tTKL1, pTEF1-TAL1-tTAL1, pPGI1-NQM1-tNQM1, pTPI1-RKI1-tRKI1 and pPYK1-TKL2-tTKL2 integration cassettes to IMX705 (Papapetridis et al.2016) and subsequent plasmid counterselection yielded strain IMX963, which overexpresses the major enzymes of the PPP. Co-transformation of pUDR119, 9 copies of the pTPI1-xylA-tCYC1 integration cassette, along with a single copy of the pTEF1-XKS1-tXKS1cassette, to IMX963 followed by plasmid counterselection yielded the xylose-fermenting strain IMX990. In IMX990, the pTPI1-xylA-tCYC1 cassettes recombined in vivo to form a multicopy construct of xylose isomerase overexpression (Verhoeven et al.2017). To construct IMX1046, in which RPE1 and PGI1 were deleted, plasmid pUDR202 and the repair oligonucleotides 9279/9280/9281/9282 were co-transformed to IMX990. Transformation mixes of IMX1046 were plated on SM agar supplemented with a xylose/fructose/glucose mixture (20, 10 and 1 g L−1 final concentrations respectively), to avoid potential glucose toxicity (Boles, Lehnert and Zimmermann 1993). To construct strain IMX994, plasmid pUDE335 was co-transformed to IMX581, along with the pTDH3-RPE1-tRPE1, pPGK1-TKL1-tTKL1, pTEF1-TAL1-tTAL1, pTPI1-RKI1-tRKI1 and pTEF1-XKS1-tXKS1 integration cassettes, after which the CRISPR plasmid was recycled. Transformation of pAKX002 to IMX994 yielded the xylose-fermenting strain IMU079. Co-transformation of pUDE327 along with the repair oligonucleotides 5888/5889 to IMX994 yielded strain IMX1384, in which HXK2 was deleted. Co-transformation of the pMEL10 backbone fragment, along with the GAL83-gRNA insert or the RSP5-gRNA insert and repair oligonucleotides 11 271/11 272 or 11 374/11 375, respectively, yielded strains IMX1385 (GAL83 deletion) and IMX1442 (RSP5 deletion). Counterselection of the CRISPR plasmids from IMX1384, IMX1385 and IMX1442 yielded, respectively, strains IMX1408, IMX1409 and IMX1451. Transformation of pAKX002 to IMX1408, IMX1409 and IMX1451 yielded the xylose-fermenting strains IMX1485, IMX1486 and IMX1487, respectively. To construct strain IMX1453, in which the mutated GAL83G673T gene replaced the wild-type GAL83 allele, plasmid pUDR105 was co-transformed to IMX1409 with the pGAL83-gal83::GAL83G673T-tGAL83 cassette. Transformation of pAKX002 to IMX1453 yielded the xylose-fermenting strain IMX1488. To construct the hxk2Δ rsp5Δ strain IMX1484, plasmid pUDE327 was co-transformed to IMX1451, along with the repair oligonucleotides 5888/5889. Counterselection of pUDE327 from IMX1484 yielded strain IMX1510. Transformation of pAKX002 to IMX1510 yielded the xylose-fermenting strain IMX1515. To construct the hxk2Δ gal83::GAL83G673T strain IMX1563, plasmid pUDE327 along with the repair oligonucleotides 5888/5889 was co-transformed to IMX1453. Counterselection of pUDE327 from IMX1563 yielded IMX1571. The xylose-fermenting strain IMX1583 was obtained by transformation of pAKX002 to IMX1571. Cultivation and media Shake-flask growth experiments were performed in 500-mL conical shake flasks containing 100 mL of SM with urea as nitrogen source (2.3 g L−1 urea, 6.6 g L−1 K2SO4, 3 g L−1 KH2PO4, 1 mL L−1 trace elements solution (Verduyn et al.1992) and 1 mL L−1 vitamin solution (Verduyn et al.1992)) to prevent medium acidification. The initial pH of the medium was set to 6.0 by titration with 2 mol L−1 KOH. Depending on the strains grown, different mixtures of carbon sources (glucose/xylose/fructose) were added and media were filter-sterilised (0.2 μm, Merck, Darmstadt, Germany). The temperature was set to 30°C and the shaking speed to 200 rpm in an Innova incubator (New Brunswick Scientific, Edison, NJ). In each case, pre-culture shake flasks were inoculated from frozen stocks. After 8–12 h of growth, exponentially growing cells from the initial shake flasks were used to inoculate fresh cultures that, after 12–18 h of growth, were used as inoculum for the growth experiments to a starting OD660 of 0.4–0.5 in the case of shake-flask growth experiments and of 0.2–0.3 in the case of bioreactor cultivation. Bioreactor cultures were grown on SM (Verduyn et al.1992), supplemented with a glucose/xylose mixture (10 g L−1/20 g L−1 for aerobic cultivation or 20 g L−1/10 g L−1 for anaerobic cultivation). Sterilisation of the salt solution was performed by autoclaving at 121°C for 20 min. Sugar solutions were sterilised separately by autoclaving at 110°C for 20 min and added to the sterile salt media along with filter-sterilised vitamin solutions. In the case of anaerobic cultivation, media were additionally supplemented with ergosterol (10 mg L−1) and Tween 80 (420 mg L−1). Sterile antifoam C (0.2 g L−1; Sigma-Aldrich, St. Louis, MO) was added to all media used for bioreactor cultivation. Batch cultures were grown in 2-L bioreactors (Applikon, Delft, The Netherlands) with a 1-L working volume, stirred at 800 rpm. Culture pH was maintained at 5.0 by automatic titration with 2 mol L−1 KOH. Temperature was maintained at 30°C. Bioreactors were sparged at 0.5 L min−1 with either pressurised air (aerobic cultivation) or nitrogen gas (<10 ppm oxygen, anaerobic cultivation). All reactors were equipped with Viton O-rings and Norprene tubing to minimise oxygen diffusion. Evaporation in bioreactor cultures was minimised by cooling the offgas outlet to 4°C. Laboratory evolution Laboratory evolution of strain IMX1046 was performed via serial shake-flask cultivation on SM (Verduyn et al.1992). Cultures were grown in 500-mL shake flasks with 100 mL working volume. Growth conditions were the same as described above. Initially, the cultures were grown on a glucose/xylose concentration ratio of 1.0 g L−1/20 g L−1. After growth was observed, exponentially growing cells (0.05 mL of culture) were transferred to SM with a glucose/xylose concentration ratio of 2.0 g L−1/20 g L−1. During subsequent serial transfers, the glucose content was progressively increased as high growth rates were established at each sugar composition, reaching a final glucose/xylose ratio of 20 g L−1/20 g L−1, to increase the selective pressure for alleviation of glucose repression and/or inhibition of xylose conversion. At that point three single colonies were isolated from two replicate evolution experiments (IMS0628–630 and IMS0634–636, respectively) by plating on SM with 10 g L−1 glucose and 20 g L−1 xylose. This medium composition supported fast growth on plates during three consecutive restreaks. Analytical methods Offgas analysis, biomass dry weight measurements, HPLC analysis of culture supernatants and correction for ethanol evaporation in bioreactor experiments were performed as previously described (Papapetridis et al.2016). Determination of optical density was performed at 660 nm using a Jenway 7200 spectrophotometer (Cole-Palmer, Staffordshire, UK). Yields of products and biomass-specific sugar uptake rates in bioreactor batch cultures were determined as previously described (Wisselink et al.2009; Papapetridis et al.2016). All values are represented as averages ± mean deviation of independent biological duplicate cultures. In silico determination of sugar uptake The Yeast v7.6 consensus metabolic model (Aung, Henry and Walker 2013) was used for in silico prediction of relative xylose and glucose consumption rates in aerobic bioreactor batch cultures of strain IMS0629. The COBRA v2 toolbox (Becker et al.2007) was used to read the model in MATLAB vR2017b (Mathworks, Natick, MA), supported by the SBML Toolbox v4.1 and the libSBML v5.12 (Hucka et al.2003). The Gurobi v6.5 linear programming solver (Gurobi Optimization Inc, Houston, TX) was installed and used according to the manual provided. The MATLAB script is provided in Additional File 4. Genome sequencing Genomic DNA of strains IMS0629 and IMS0634 was isolated from exponentially growing shake-flask cultures on SM (10 g L−1 glucose/20 g L−1 xylose) with a Qiagen Blood & Cell culture DNA kit (Qiagen, Germantown, MD), according to the manufacturer's specifications. Whole-genome sequencing was performed on an Illumina HiSeq PE150 sequencer (Novogene Company Limited, Hong Kong), as previously described (Papapetridis et al.2018). Sequence data were mapped to the reference CEN.PK113–7D genome (Nijkamp et al.2012), to which the sequences of the pTPI1-gndA-tCYC1 and pTPI1-xylA-tCYC1 cassettes were manually added. Data processing and chromosome copy number analysis were carried as previously described (Papapetridis et al.2018). RESULTS Design of an S. cerevisiae strain with a forced, high stoichiometry of xylose and glucose co-consumption Design of an S. cerevisiae strain whose growth depended on extensive co-consumption of xylose and glucose was based on the observation that inactivation of PGI1 blocks entry of glucose-6-phosphate into glycolysis, while inactivation of RPE1 prevents entry of ribulose-5-phosphate into the non-oxidative PPP (Fig. 1). As a consequence, a pgi1Δ rpe1Δ strain is unable to grow on glucose. If conversion of xylose into xylulose-5-phosphate in such a strain is enabled by expression of a heterologous xylose isomerase and overexpression of the native xylulose kinase Xks1 (Kuyper et al.2005a), co-consumption of xylose and glucose should enable growth (Fig. 1). Overexpression of native genes encoding the enzymes of the non-oxidative PPP has previously been shown to stimulate the required conversion of xylulose-5-phosphate into the glycolytic intermediates fructose-6-phosphate and glyceraldehyde-3-phosphate (Fig. 1) (Johansson and Hahn-Hägerdal 2002; Kuyper et al.2005a). The predicted stoichiometry for conversion of glucose and xylose into pyruvate in a yeast strain that combines these genetic modifications is summarised in Equation 1: \begin{eqnarray} {{\rm{C}}_6}{{\rm{H}}_{12}}{{\rm{O}}_6} &+& 2{{\rm{C}}_5}{{\rm{H}}_{10}}{{\rm{O}}_5} + 5{\rm{ADP}} + 5{{\rm{P}}_i} + 2{\rm{NAD}}{{\rm{P}}^ + } + 5{\rm{NA}}{{\rm{D}}^ + } \nonumber\\ &\to & 5{{\rm{C}}_3}{{\rm{H}}_4}{{\rm{O}}_3} + 5{\rm{ATP}} + 4{{\rm{H}}_2}{\rm{O}} + 2{\rm{NADPH}} + 5{\rm{NADH}}\nonumber\\ &+& 7{{\rm{H}}^ + } + {\rm{C}}{{\rm{O}}_2} \end{eqnarray} (1) Figure 1. Open in new tabDownload slide Schematic representation of central carbon metabolism in a yeast strain engineered for forced co-consumption of glucose and xylose. In pgi1Δ rpe1Δ S. cerevisiae expressing a heterologous xylose isomerase (XI, Kuyper et al.2003), the native 6-phosphogluconate dehydrogenases (Gnd1 and Gnd2) were replaced by a bacterial NAD+-dependent enzyme (GndA, Papapetridis et al.2016). Additionally, xylulokinase (Xks1) and enzymes of the pentose phosphate pathway (PPP) were overexpressed. F6P fructose-6-phosphate; G3P glyceraldehyde-3-phosphate; DHAP dihydroxyacetone phosphate; 6PGL 6-phosphogluconolactone; 6PG 6-phosphogluconate; RU5P ribulose-5-phosphate; XU5P xylulose-5-phosphate; R5P ribose-5-phosphate; S7P sedoheptulose-7-phosphate; E4P erythrose-4-phosphate. To prevent a potential excessive formation of NADPH (Boles, Lehnert and Zimmermann 1993; Dickinson, Sobanski and Hewlins 1995), the strain design further included replacement of the native S. cerevisiae NADP+-dependent 6-phosphogluconate dehydrogenases (Gnd1 and Gnd2) by the NAD+-dependent bacterial enzyme GndA (Papapetridis et al.2016), leading to the stoichiometry shown in Equation 2: \begin{eqnarray} {{\rm{C}}_6}{{\rm{H}}_{12}}{{\rm{O}}_6} &+& 2\,{{\rm{C}}_5}{{\rm{H}}_{10}}{{\rm{O}}_5} + 5{\rm{ADP}} + 5{{\rm{P}}_i} + 1{\rm{NAD}}{{\rm{P}}^ + } + 6{\rm{NA}}{{\rm{D}}^ + }\nonumber\\ &\to & 5\,{{\rm{C}}_3}{{\rm{H}}_4}{{\rm{O}}_3} + 5{\rm{ATP}} + 4{{\rm{H}}_2}{\rm{O}} + {\rm{NADPH}} + 6{\rm{NADH}}\nonumber\\ &+& 7{{\rm{H}}^ + } + {\rm{C}}{{\rm{O}}_2} \end{eqnarray} (2) As indicated by Equation 2, this strain design forces co-consumption of 2 mol xylose and 1 mol glucose for the production of 5 mol pyruvate, with a concomitant formation of 1 mol NADPH, 6 mol NADH and 5 mol ATP. NADPH generated in this process can be reoxidised in biosynthetic reactions (Bruinenberg, Van Dijken and Scheffers 1983) or via an L-glutamate-2-oxoglutarate transhydrogenase cycle catalysed by Gdh1 and Gdh2 (Boles, Lehnert and Zimmermann 1993). Actual in vivo stoichiometries of mixed-sugar consumption will depend on the relative contribution of precursors derived from glucose and xylose to biomass synthesis and on the biomass yield (Verduyn et al.1990). In aerobic cultures, the latter strongly depends on the mode of NADH reoxidation (mitochondrial respiration, alcoholic fermentation and/or glycerol production; Bakker et al.2001). While quantitation of precise co-consumption stoichiometries will therefore require experimental analysis, this strain design clearly has the potential to force xylose and glucose co-consumption at much higher stoichiometries than previously reported (Katahira et al.2008; Shen et al.2012, 2015; Gonçalves et al.2014; Nijland et al.2014; Westman et al.2014; Shin et al.2015). Construction, laboratory evolution and growth stoichiometry of glucose-xylose co-consuming S. cerevisiae strains To implement the proposed strain design for forced co-consumption of xylose and glucose, multiple copies of a codon-optimised expression cassette for Piromyces xylA (Verhoeven et al.2017) were integrated into the genome of S. cerevisiae IMX705 (gnd1Δ gnd2Δ gndA; Papapetridis et al.2016), along with overexpression cassettes for S. cerevisiae XKS1 and for structural genes encoding PPP enzymes. Deletion of RPE1 and PGI1 in the resulting xylose-consuming strain IMX990 yielded strain IMX1046, which grew in aerobic shake-flask cultures on SM with 1 g L−1 glucose and 20 g L−1 xylose as sole carbon sources without a requirement for laboratory evolution. However, this strain did not grow at the same xylose concentration when the glucose concentration was increased to 10 g L−1. Kinetic and/or regulatory constraints in glucose-xylose co-consumption at higher glucose concentrations could, for example, reflect interference of glucose with expression or activity of Hxt transporters involved in xylose uptake. To select for co-consumption of xylose at higher glucose concentrations, duplicate serial-transfer experiments were performed in aerobic shake-flask cultures on SM with 20 g L−1 xylose. During serial transfer, the glucose concentration in the medium was gradually increased from 1 to 20 g L−1 (Fig. 2). Samples of the evolving cultures were regularly inoculated in SM containing either 20 g L−1 glucose or 20 g L−1 xylose as sole carbon source. Absence of growth on these single sugars showed that laboratory evolution did not result in an escape from their forced co-consumption. When, after 13 transfers, vigorous growth was observed on a mixture of 20 g L−1 glucose and 20 g L−1 xylose, three single-colony isolates were obtained from each laboratory evolution experiment by streaking on SM agar (10 g L−1 glucose/20 g L−1 xylose). Figure 2. Open in new tabDownload slide Laboratory evolution of S. cerevisiae IMX1046 (pgi1Δ rpe1Δ gnd1Δ gnd2Δ gndA xylA XKS1↑ PPP↑) for improved co-consumption of xylose at high glucose concentrations. Cultures were grown in shake flasks containing 100 mL SM (pH 6) supplemented with 20 g L−1 xylose and progressively increasing glucose concentrations. In every transfer, 0.05 mL of an exponentially growing culture was used to inoculate the next shake flask. Growth studies with the six evolved isolates in shake-flask cultures on SM with 10 g L−1 glucose and 20 g L−1 xylose (Additional File 5) identified isolate IMS0629 (Evolution Line 1) as the fastest growing isolate (μ = 0.21 h−1). The physiology of this strain was further characterised in aerobic bioreactor batch cultures on SM containing 10 g L−1 glucose and 20 g L−1 xylose. After a ca. 10-h lag phase (Fig. 3, Additional Files 6 and 7), exponential growth was observed at a specific growth rate of 0.18 h−1. Biomass, ethanol and CO2 were the main products, with additional minor formation of glycerol and acetate (Table 2, Fig. 3, Additional File 6). During the exponential growth phase, xylose and glucose were co-consumed at a fixed molar ratio of 1.64 mol mol−1 (Table 2, Fig. 3). Growth ceased after glucose depletion, at which point xylose consumption rates drastically decreased and corresponded to a simultaneous low rate of xylitol formation (Fig. 3, Additional File 6). As previously reported for XylA-expressing, xylose-fermenting S. cerevisiae strains (Kuyper et al.2005a; Verhoeven et al.2017), no production of xylitol was observed during the exponential growth phase. The biomass and ethanol yields on total sugars consumed were 0.28 g biomass (g sugar)−1 and 0.18 g ethanol (g sugar)−1, respectively. Together with a respiratory quotient of 1.5, these observations indicated a respiro-fermentative sugar dissimilation. In line with the inability of pgi1Δ S. cerevisiae to generate glucose-6-phosphate from ethanol and acetate, reconsumption of these fermentation products after glucose depletion was not coupled to growth (Fig. 3, Additional File 6). However, their oxidation may have provided redox equivalents for the observed slow production of xylitol from xylose (Fig. 3). Figure 3. Open in new tabDownload slide Sugar consumption, biomass and metabolite production profiles of the evolved S. cerevisiae strain IMS0629 (pgi1Δ rpe1Δ gnd1Δ gnd2Δ gndA xylA XKS1↑ PPP↑), grown on SM with 10 g L−1 glucose and 20 g L−1 xylose in aerobic bioreactor batch cultures (pH 5, 30°C). Cultures were grown in duplicate; the data shown are from a single representative culture. a: filled circles glucose, open circles xylose; b: filled squares biomass, open squares ethanol, filled triangles acetate, open triangles glycerol, open diamonds xylitol; c: ratio of xylose and glucose consumption during exponential growth phase. Table 2. Product yields, biomass-specific sugar uptake and production rates in aerobic bioreactor batch cultures of evolved strain S. cerevisiae IMS0629 (pgi1Δ rpe1Δ gnd1Δ gnd2Δ gndA xylA XKS1↑ PPP↑) on SM supplemented with 10 g L−1 glucose and 20 g L−1 xylose (pH 5, 30°C). Growth rate (h−1) 0.18 ± 0.00 Glucose-xylose consumption ratio mol mol−1 1.64 ± 0.00 Spec. xylose uptake rate mmol (g biomass)−1 h−1 2.52 ± 0.00 Spec. glucose uptake rate mmol (g biomass)−1 h−1 1.54 ± 0.07 Spec. glycerol production rate mmol (g biomass)−1 h−1 0.23 ± 0.01 Spec. ethanol production rate mmol (g biomass)−1 h−1 2.25 ± 0.37 Spec. CO2 production rate mmol (g biomass)−1 h−1 10.43 ± 0.98 Spec. O2 uptake rate mmol (g biomass)−1 h−1 6.87 ± 0.56 Respiratory quotient 1.52 ± 0.03 Biomass yield g biomass (g sugars)−1 0.28 ± 0.00 Ethanol yield g (g sugars)−1 0.18 ± 0.00 Growth rate (h−1) 0.18 ± 0.00 Glucose-xylose consumption ratio mol mol−1 1.64 ± 0.00 Spec. xylose uptake rate mmol (g biomass)−1 h−1 2.52 ± 0.00 Spec. glucose uptake rate mmol (g biomass)−1 h−1 1.54 ± 0.07 Spec. glycerol production rate mmol (g biomass)−1 h−1 0.23 ± 0.01 Spec. ethanol production rate mmol (g biomass)−1 h−1 2.25 ± 0.37 Spec. CO2 production rate mmol (g biomass)−1 h−1 10.43 ± 0.98 Spec. O2 uptake rate mmol (g biomass)−1 h−1 6.87 ± 0.56 Respiratory quotient 1.52 ± 0.03 Biomass yield g biomass (g sugars)−1 0.28 ± 0.00 Ethanol yield g (g sugars)−1 0.18 ± 0.00 Biomass-specific rates, yields and ratios were calculated from samples taken during the mid-exponential growth phase and represent averages ± mean deviation of independent duplicate cultures. Ethanol yield was corrected for evaporation. Open in new tab Table 2. Product yields, biomass-specific sugar uptake and production rates in aerobic bioreactor batch cultures of evolved strain S. cerevisiae IMS0629 (pgi1Δ rpe1Δ gnd1Δ gnd2Δ gndA xylA XKS1↑ PPP↑) on SM supplemented with 10 g L−1 glucose and 20 g L−1 xylose (pH 5, 30°C). Growth rate (h−1) 0.18 ± 0.00 Glucose-xylose consumption ratio mol mol−1 1.64 ± 0.00 Spec. xylose uptake rate mmol (g biomass)−1 h−1 2.52 ± 0.00 Spec. glucose uptake rate mmol (g biomass)−1 h−1 1.54 ± 0.07 Spec. glycerol production rate mmol (g biomass)−1 h−1 0.23 ± 0.01 Spec. ethanol production rate mmol (g biomass)−1 h−1 2.25 ± 0.37 Spec. CO2 production rate mmol (g biomass)−1 h−1 10.43 ± 0.98 Spec. O2 uptake rate mmol (g biomass)−1 h−1 6.87 ± 0.56 Respiratory quotient 1.52 ± 0.03 Biomass yield g biomass (g sugars)−1 0.28 ± 0.00 Ethanol yield g (g sugars)−1 0.18 ± 0.00 Growth rate (h−1) 0.18 ± 0.00 Glucose-xylose consumption ratio mol mol−1 1.64 ± 0.00 Spec. xylose uptake rate mmol (g biomass)−1 h−1 2.52 ± 0.00 Spec. glucose uptake rate mmol (g biomass)−1 h−1 1.54 ± 0.07 Spec. glycerol production rate mmol (g biomass)−1 h−1 0.23 ± 0.01 Spec. ethanol production rate mmol (g biomass)−1 h−1 2.25 ± 0.37 Spec. CO2 production rate mmol (g biomass)−1 h−1 10.43 ± 0.98 Spec. O2 uptake rate mmol (g biomass)−1 h−1 6.87 ± 0.56 Respiratory quotient 1.52 ± 0.03 Biomass yield g biomass (g sugars)−1 0.28 ± 0.00 Ethanol yield g (g sugars)−1 0.18 ± 0.00 Biomass-specific rates, yields and ratios were calculated from samples taken during the mid-exponential growth phase and represent averages ± mean deviation of independent duplicate cultures. Ethanol yield was corrected for evaporation. Open in new tab The quantitative data on biomass and product formation obtained from the bioreactor batch cultures enabled a comparison of the observed molar ratio of xylose and glucose consumption with a model-based prediction. To this end, the engineered metabolic network of strain IMX1046 was re-created in silico, using the Yeast v7.6 consensus metabolic model as a basis (Aung, Henry and Walker 2013; Additional File 4). Consistent with the experimental observations on forced co-consumption, inactivation of either xylose or glucose uptake in the model network did not result in any feasible growth solutions. Using the experimentally determined average specific growth rates and oxygen consumption rates from the aerobic bioreactor batch cultures of strain IMS0629 as constraints on the model resulted in predicted xylose and glucose uptake rates of 2.68 and 1.93 mmol (g biomass)−1 h−1, respectively, corresponding to a molar ratio of the xylose and glucose consumption rates of 1.4. In view of the complexity of the model and the potential impact of differences in biomass composition, this number corresponded well with the experimentally measured value of 1.64 (Table 2). Whole-genome sequencing of evolved glucose-xylose co-consuming S. cerevisiae To identify causal mutations for the improved growth of the evolved glucose-xylose co-consuming S. cerevisiae strains at high glucose concentrations, the genomes of strains IMS0629 and IMS0634 (fastest growing isolates from evolution line 1 and 2, respectively, Additional File 5) were sequenced and compared to that of their common parental strain. Despite the well-documented role of S. cerevisiae hexose transporters in xylose uptake (Reifenberger, Boles and Ciriacy 1997; Hamacher et al.2002; Lee et al.2002; Farwick et al.2014), no mutations were found in the coding region of any of the 18 genes encoding these transporters (HXT1–17 and GAL2), or in other known transporter genes. Both evolved strains harboured mutations in HXK2 (Table 3). This gene encodes the major S. cerevisiae hexokinase which, in addition to its catalytic role, is involved in glucose repression (De Winde et al.1996; Raamsdonk et al.2001). The mutation in IMS0629 caused a premature stop codon at position 309 of Hxk2. Both strains also harboured mutations in RSP5, which encodes an E3-ubiquitin ligase linked to ubiquitination and endocytosis of membrane proteins (Belgareh-Touzé et al.2008). In strain IMS0629, a substitution at position 686 caused a glycine to aspartic acid change at position 229 of Rsp5 (Table 3). Strain IMS0634 carried a 41 bp internal deletion in RSP5, which included the location of the mutation in strain IMS0629 and probably caused loss of function. Table 3. Mutations identified by whole-genome sequencing of glucose-xylose co-consuming S. cerevisiae strains evolved for fast growth at high glucose concentrations. Strain and gene . Nucleotide change . Amino acid change . Description . IMS0629 CYC8 G896A W299 →Stop General transcriptional co-repressor; acts together with Tup1; also acts as part of a transcriptional co-activator complex that recruits the SWI/SNF and SAGA complexes to promoters; can form the prion [OCT+] GAL83 G673T D225Y One of three possible beta-subunits of the Snf1 kinase complex; allows nuclear localisation of the Snf1 kinase complex in the presence of a non-fermentable carbon source RSP5 G686A G229D NEDD4 family E3 ubiquitin ligase; regulates processes including MVB sorting, the heat shock response, transcription, endocytosis and ribosome stability; ubiquitinates Sec23, Sna3, Ste4, Nfi1, Rpo21 and Sem1; autoubiquitinates; deubiquitinated by Ubp2; regulated by SUMO ligase Siz1, in turn regulates Siz1p SUMO ligase activity; required for efficient Golgi-to-ER trafficking in COPI mutants HXK2 C927G Y309 →Stop Hexokinase isoenzyme 2; phosphorylates glucose in cytosol; predominant hexokinase during growth on glucose; represses expression of HXK1, GLK1 RBH1 C190A Q64K Putative protein of unknown function; expression is cell-cycle regulated as shown by microarray analysis; potential regulatory target of Mbp1, which binds to the YJL181W promoter region; contains a PH-like domain DCS1 C636G Y212→Stop Non-essential hydrolase involved in mRNA decapping; activates Xrn1; may function in a feedback mechanism to regulate deadenylation, contains pyrophosphatase activity and a HIT (histidine triad) motif; acts as inhibitor of neutral trehalase Nth1; required for growth on glycerol medium IMS0634 RSP5 Internal deletion (41 nucleotides) Frameshift NEDD4 family E3 ubiquitin ligase; regulates processes including: MVB sorting, the heat shock response, transcription, endocytosis and ribosome stability; ubiquitinates Sec23, Sna3, Ste4, Nfi1, Rpo21 and Sem1; autoubiquitinates; deubiquitinated by Ubp2; regulated by SUMO ligase Siz1, in turn regulates Siz1p SUMO ligase activity; required for efficient Golgi-to-ER trafficking in COPI mutants HXK2 G1027C D343H Hexokinase isoenzyme 2; phosphorylates glucose in cytosol; predominant hexokinase during growth on glucose; represses expression of HXK1, GLK1 Strain and gene . Nucleotide change . Amino acid change . Description . IMS0629 CYC8 G896A W299 →Stop General transcriptional co-repressor; acts together with Tup1; also acts as part of a transcriptional co-activator complex that recruits the SWI/SNF and SAGA complexes to promoters; can form the prion [OCT+] GAL83 G673T D225Y One of three possible beta-subunits of the Snf1 kinase complex; allows nuclear localisation of the Snf1 kinase complex in the presence of a non-fermentable carbon source RSP5 G686A G229D NEDD4 family E3 ubiquitin ligase; regulates processes including MVB sorting, the heat shock response, transcription, endocytosis and ribosome stability; ubiquitinates Sec23, Sna3, Ste4, Nfi1, Rpo21 and Sem1; autoubiquitinates; deubiquitinated by Ubp2; regulated by SUMO ligase Siz1, in turn regulates Siz1p SUMO ligase activity; required for efficient Golgi-to-ER trafficking in COPI mutants HXK2 C927G Y309 →Stop Hexokinase isoenzyme 2; phosphorylates glucose in cytosol; predominant hexokinase during growth on glucose; represses expression of HXK1, GLK1 RBH1 C190A Q64K Putative protein of unknown function; expression is cell-cycle regulated as shown by microarray analysis; potential regulatory target of Mbp1, which binds to the YJL181W promoter region; contains a PH-like domain DCS1 C636G Y212→Stop Non-essential hydrolase involved in mRNA decapping; activates Xrn1; may function in a feedback mechanism to regulate deadenylation, contains pyrophosphatase activity and a HIT (histidine triad) motif; acts as inhibitor of neutral trehalase Nth1; required for growth on glycerol medium IMS0634 RSP5 Internal deletion (41 nucleotides) Frameshift NEDD4 family E3 ubiquitin ligase; regulates processes including: MVB sorting, the heat shock response, transcription, endocytosis and ribosome stability; ubiquitinates Sec23, Sna3, Ste4, Nfi1, Rpo21 and Sem1; autoubiquitinates; deubiquitinated by Ubp2; regulated by SUMO ligase Siz1, in turn regulates Siz1p SUMO ligase activity; required for efficient Golgi-to-ER trafficking in COPI mutants HXK2 G1027C D343H Hexokinase isoenzyme 2; phosphorylates glucose in cytosol; predominant hexokinase during growth on glucose; represses expression of HXK1, GLK1 Gene descriptions were taken from the Saccharomyces Genome Database (https://www.yeastgenome.org/, Accessed 14 December 2017). Open in new tab Table 3. Mutations identified by whole-genome sequencing of glucose-xylose co-consuming S. cerevisiae strains evolved for fast growth at high glucose concentrations. Strain and gene . Nucleotide change . Amino acid change . Description . IMS0629 CYC8 G896A W299 →Stop General transcriptional co-repressor; acts together with Tup1; also acts as part of a transcriptional co-activator complex that recruits the SWI/SNF and SAGA complexes to promoters; can form the prion [OCT+] GAL83 G673T D225Y One of three possible beta-subunits of the Snf1 kinase complex; allows nuclear localisation of the Snf1 kinase complex in the presence of a non-fermentable carbon source RSP5 G686A G229D NEDD4 family E3 ubiquitin ligase; regulates processes including MVB sorting, the heat shock response, transcription, endocytosis and ribosome stability; ubiquitinates Sec23, Sna3, Ste4, Nfi1, Rpo21 and Sem1; autoubiquitinates; deubiquitinated by Ubp2; regulated by SUMO ligase Siz1, in turn regulates Siz1p SUMO ligase activity; required for efficient Golgi-to-ER trafficking in COPI mutants HXK2 C927G Y309 →Stop Hexokinase isoenzyme 2; phosphorylates glucose in cytosol; predominant hexokinase during growth on glucose; represses expression of HXK1, GLK1 RBH1 C190A Q64K Putative protein of unknown function; expression is cell-cycle regulated as shown by microarray analysis; potential regulatory target of Mbp1, which binds to the YJL181W promoter region; contains a PH-like domain DCS1 C636G Y212→Stop Non-essential hydrolase involved in mRNA decapping; activates Xrn1; may function in a feedback mechanism to regulate deadenylation, contains pyrophosphatase activity and a HIT (histidine triad) motif; acts as inhibitor of neutral trehalase Nth1; required for growth on glycerol medium IMS0634 RSP5 Internal deletion (41 nucleotides) Frameshift NEDD4 family E3 ubiquitin ligase; regulates processes including: MVB sorting, the heat shock response, transcription, endocytosis and ribosome stability; ubiquitinates Sec23, Sna3, Ste4, Nfi1, Rpo21 and Sem1; autoubiquitinates; deubiquitinated by Ubp2; regulated by SUMO ligase Siz1, in turn regulates Siz1p SUMO ligase activity; required for efficient Golgi-to-ER trafficking in COPI mutants HXK2 G1027C D343H Hexokinase isoenzyme 2; phosphorylates glucose in cytosol; predominant hexokinase during growth on glucose; represses expression of HXK1, GLK1 Strain and gene . Nucleotide change . Amino acid change . Description . IMS0629 CYC8 G896A W299 →Stop General transcriptional co-repressor; acts together with Tup1; also acts as part of a transcriptional co-activator complex that recruits the SWI/SNF and SAGA complexes to promoters; can form the prion [OCT+] GAL83 G673T D225Y One of three possible beta-subunits of the Snf1 kinase complex; allows nuclear localisation of the Snf1 kinase complex in the presence of a non-fermentable carbon source RSP5 G686A G229D NEDD4 family E3 ubiquitin ligase; regulates processes including MVB sorting, the heat shock response, transcription, endocytosis and ribosome stability; ubiquitinates Sec23, Sna3, Ste4, Nfi1, Rpo21 and Sem1; autoubiquitinates; deubiquitinated by Ubp2; regulated by SUMO ligase Siz1, in turn regulates Siz1p SUMO ligase activity; required for efficient Golgi-to-ER trafficking in COPI mutants HXK2 C927G Y309 →Stop Hexokinase isoenzyme 2; phosphorylates glucose in cytosol; predominant hexokinase during growth on glucose; represses expression of HXK1, GLK1 RBH1 C190A Q64K Putative protein of unknown function; expression is cell-cycle regulated as shown by microarray analysis; potential regulatory target of Mbp1, which binds to the YJL181W promoter region; contains a PH-like domain DCS1 C636G Y212→Stop Non-essential hydrolase involved in mRNA decapping; activates Xrn1; may function in a feedback mechanism to regulate deadenylation, contains pyrophosphatase activity and a HIT (histidine triad) motif; acts as inhibitor of neutral trehalase Nth1; required for growth on glycerol medium IMS0634 RSP5 Internal deletion (41 nucleotides) Frameshift NEDD4 family E3 ubiquitin ligase; regulates processes including: MVB sorting, the heat shock response, transcription, endocytosis and ribosome stability; ubiquitinates Sec23, Sna3, Ste4, Nfi1, Rpo21 and Sem1; autoubiquitinates; deubiquitinated by Ubp2; regulated by SUMO ligase Siz1, in turn regulates Siz1p SUMO ligase activity; required for efficient Golgi-to-ER trafficking in COPI mutants HXK2 G1027C D343H Hexokinase isoenzyme 2; phosphorylates glucose in cytosol; predominant hexokinase during growth on glucose; represses expression of HXK1, GLK1 Gene descriptions were taken from the Saccharomyces Genome Database (https://www.yeastgenome.org/, Accessed 14 December 2017). Open in new tab Compared to strain IMS0634, strain IMS0629 harboured four additional nucleotide changes in protein-coding regions (Table 3). A G-A change at position 896 of the transcriptional regulator gene CYC8 introduced a stop codon at position 299 of the protein. Deletion of CYC8 was previously shown to enhance xylose uptake in the presence of glucose, albeit at the expense of growth rate (Nijland et al.2017). A G-T change at position 673 of the transcriptional regulator gene GAL83 caused an amino acid change from aspartic acid to tyrosine at position 225 of the protein. Gal83 plays a vital role in the function of the Snf1-kinase complex of S. cerevisiae, which is involved in activation of glucose-repressed genes in the absence of the sugar (Rahner et al.1996; Vincent and Carlson 1998, 1999; Schmidt and McCartney 2000). Analysis of chromosomal copy number variations showed no chromosomal rearrangements in strain IMS0629 (Additional File 8). In contrast, strain IMS0634 carried a duplication of the right arm of chromosome 3, a duplication of the middle part of chromosome 8 and a duplication of chromosome 9 (Additional File 8). The duplications in chromosomes 8 and 9 in IMS0634 spanned the GND1, GRE3 and SGA1 loci at which the expressing cassettes for heterologous genes were integrated (Table 1). In the evolved strains IMS0629 and IMS0634, xylA copy numbers had increased to ca. 27 and 20, respectively. This observation is consistent with previous research that showed a requirement for high copy numbers of xylA expression cassettes to support fast xylose consumption (Alper and Stephanopoulos 2009; Verhoeven et al.2017). The duplication of a segment of chromosome 8 in strain IMS0634 also spanned the locations of the low-to-moderate affinity hexose transporter genes HXT1 and HXT5 and the high-affinity hexose transporter gene HXT4. Mutations in HXK2, RSP5 and GAL83 stimulate co-consumption of xylose and glucose in aerobic cultures of xylose-consuming S. cerevisiae To investigate whether mutations acquired during evolution for forced co-consumption of glucose and xylose were relevant for mixed-sugar utilisation in a strain without forced glucose-xylose co-consumption, we focused on mutations in HXK2 and RSP5 (which were present in both IMS0629 and IMS0634) and/or GAL83. Mutations in these genes were introduced into an engineered, non-evolved xylose-consuming S. cerevisiae strain background (IMX994, Table 1). Overexpression of xylA was accomplished by transforming strains with the multicopy xylA expression vector pAKX002 (Kuyper et al.2003). In aerobic shake-flask cultures grown on 10 g L−1 glucose and 10 g L−1 xylose, the reference strain IMU079 (XKS1↑ PPP↑ pAKX002) displayed a pronounced biphasic growth profile and only a minor co-consumption of the two sugars (0.13 mol xylose (mol glucose)−1; Table 4, Fig. 4, Additional File 9). Co-consumption was strongly enhanced in the congenic hxk2Δ strain IMX1485, which showed a 3-fold higher molar ratio of xylose and glucose consumption (0.41 mol mol−1). However, its specific growth rate before glucose depletion (0.28 h−1) was 13% lower than that of the reference strain (Table 4). Strain IMX1487 (rsp5Δ), which showed a 20% lower specific growth rate than the reference strain, showed a slight improvement in co-consumption (Table 4). Deletion of GAL83 (strain IMX1486) affected neither sugar co-consumption nor growth rate. In contrast, replacement of GAL83 by GAL83G673T (strain IMX1488) resulted in a 40% higher co-consumption of glucose and xylose than observed in the reference strain IMU079, without affecting growth rate (Table 4). Figure 4. Open in new tabDownload slide Consumption of glucose and xylose and growth of strains IMU079 (XKS1↑ PPP↑ pAKX002; a, d), IMX1515 (hxk2Δ rsp5Δ XKS1↑ PPP↑ pAKX002; b, e) and IMX1583 (hxk2Δ gal83::GAL83G673T XKS1↑ PPP↑ pAKX002; c, f) in batch cultures. The three strains were grown on SM (urea as nitrogen source) with 10 g L−1 glucose and 10 g L−1 xylose in aerobic shake-flask cultures (pH 6, 30°C). a, b, c: filled circles glucose, open circles xylose, open squares OD660; d, e, f: ratio of xylose and glucose consumption during exponential growth phase. Data for independent duplicate cultures are provided in Additional File 9. Table 4. Specific growth rates (μ) and ratio of xylose and glucose consumption in aerobic shake-flask cultures of strains IMU079 (XKS1↑ PPP↑ pAKX002), IMX1485 (hkx2Δ XKS1↑ PPP↑ pAKX002), IMX1486 (gal83Δ XKS1↑ PPP↑ pAKX002), IMX1487 (rsp5Δ XKS1↑ PPP↑ pAKX002), IMX1488 (gal83::GAL83G673T XKS1↑ PPP↑ pAKX002), IMX1515 (hxk2Δ rsp5Δ XKS1↑ PPP↑ pAKX002) and IMX1583 (hxk2Δ gal83::GAL83G673T XKS1↑ PPP↑ pAKX002) grown on SM (urea as nitrogen source) with 10 g L−1 glucose and 10 g L−1 xylose (pH 6, 30°C). Strain . Relevant genotype . μ (h−1) . Xylose-glucose consumption ratio (mol mol−1) . IMU079 HXK2 RSP5 GAL83 0.32 ± 0.01 0.13 ± 0.00 IMX1485 hxk2Δ 0.28 ± 0.00 0.41 ± 0.01 IMX1486 gal83Δ 0.31 ± 0.00 0.14 ± 0.01 IMX1487 rsp5Δ 0.26 ± 0.00 0.15 ± 0.00 IMX1488 gal83::GAL83G673T 0.31 ± 0.00 0.18 ± 0.01 IMX1515 hxk2Δ rsp5Δ 0.19 ± 0.00 0.60 ± 0.00 IMX1583 hxk2Δ gal83::GAL83G673T 0.31 ± 0.00 0.49 ± 0.00 Strain . Relevant genotype . μ (h−1) . Xylose-glucose consumption ratio (mol mol−1) . IMU079 HXK2 RSP5 GAL83 0.32 ± 0.01 0.13 ± 0.00 IMX1485 hxk2Δ 0.28 ± 0.00 0.41 ± 0.01 IMX1486 gal83Δ 0.31 ± 0.00 0.14 ± 0.01 IMX1487 rsp5Δ 0.26 ± 0.00 0.15 ± 0.00 IMX1488 gal83::GAL83G673T 0.31 ± 0.00 0.18 ± 0.01 IMX1515 hxk2Δ rsp5Δ 0.19 ± 0.00 0.60 ± 0.00 IMX1583 hxk2Δ gal83::GAL83G673T 0.31 ± 0.00 0.49 ± 0.00 Growth rates and ratios were calculated from samples taken during the mid-exponential growth phase and represent averages ± mean deviation of independent duplicate cultures. Open in new tab Table 4. Specific growth rates (μ) and ratio of xylose and glucose consumption in aerobic shake-flask cultures of strains IMU079 (XKS1↑ PPP↑ pAKX002), IMX1485 (hkx2Δ XKS1↑ PPP↑ pAKX002), IMX1486 (gal83Δ XKS1↑ PPP↑ pAKX002), IMX1487 (rsp5Δ XKS1↑ PPP↑ pAKX002), IMX1488 (gal83::GAL83G673T XKS1↑ PPP↑ pAKX002), IMX1515 (hxk2Δ rsp5Δ XKS1↑ PPP↑ pAKX002) and IMX1583 (hxk2Δ gal83::GAL83G673T XKS1↑ PPP↑ pAKX002) grown on SM (urea as nitrogen source) with 10 g L−1 glucose and 10 g L−1 xylose (pH 6, 30°C). Strain . Relevant genotype . μ (h−1) . Xylose-glucose consumption ratio (mol mol−1) . IMU079 HXK2 RSP5 GAL83 0.32 ± 0.01 0.13 ± 0.00 IMX1485 hxk2Δ 0.28 ± 0.00 0.41 ± 0.01 IMX1486 gal83Δ 0.31 ± 0.00 0.14 ± 0.01 IMX1487 rsp5Δ 0.26 ± 0.00 0.15 ± 0.00 IMX1488 gal83::GAL83G673T 0.31 ± 0.00 0.18 ± 0.01 IMX1515 hxk2Δ rsp5Δ 0.19 ± 0.00 0.60 ± 0.00 IMX1583 hxk2Δ gal83::GAL83G673T 0.31 ± 0.00 0.49 ± 0.00 Strain . Relevant genotype . μ (h−1) . Xylose-glucose consumption ratio (mol mol−1) . IMU079 HXK2 RSP5 GAL83 0.32 ± 0.01 0.13 ± 0.00 IMX1485 hxk2Δ 0.28 ± 0.00 0.41 ± 0.01 IMX1486 gal83Δ 0.31 ± 0.00 0.14 ± 0.01 IMX1487 rsp5Δ 0.26 ± 0.00 0.15 ± 0.00 IMX1488 gal83::GAL83G673T 0.31 ± 0.00 0.18 ± 0.01 IMX1515 hxk2Δ rsp5Δ 0.19 ± 0.00 0.60 ± 0.00 IMX1583 hxk2Δ gal83::GAL83G673T 0.31 ± 0.00 0.49 ± 0.00 Growth rates and ratios were calculated from samples taken during the mid-exponential growth phase and represent averages ± mean deviation of independent duplicate cultures. Open in new tab Since independently evolved glucose-xylose co-consuming strains contained putative loss-of-function mutations in HXK2 and RSP5, both genes were deleted in strain IMX1515 (hxk2Δ rsp5Δ XKS1↑ PPP↑ pAKX002). Similarly, deletion of HXK2 and introduction of GAL83G673Twere combined in strain IMX1583 (hxk2Δ gal83::GAL83G673T XKS1↑ PPP↑ pAKX002). Co-consumption ratios in the two strains (0.60 and 0.49 mol xylose (mol glucose)−1, respectively) were 4- to 5-fold higher than in the reference strain IMU079 (Table 4, Fig. 4, Additional File 9). However, strain IMX1515 exhibited a 40% lower specific growth rate (0.19 h−1) than the reference strain, resulting in a 9 h extension of the fermentation experiments (Fig. 4, Additional File 9). In contrast, strain IMX1583 combined a high co-consumption ratio with the same specific growth rate as that of the reference strain. Combined mutations in HXK2 and GAL83 significantly accelerate conversion of glucose-xylose mixtures by anaerobic cultures of xylose-consuming S. cerevisiae To investigate the impact of the identified mutations under more industrially relevant conditions, anaerobic growth of the reference xylose-fermenting strain IMU079 (XKS1↑ PPP↑ pAKX002) in bioreactor batch experiments was compared with that of the two congenic double mutants IMX1515 (hxk2Δ rsp5Δ) and IMX1583 (hxk2Δ gal83::GAL83G673T) that showed the highest glucose-xylose co-consumption in the aerobic shake-flask experiments. The anaerobic cultures were grown on 20 g L−1 glucose and 10 g L−1 xylose to simulate the relative concentrations of these sugars typically found in corn stover and wheat straw hydrolysates (van Maris et al.2006). In the anaerobic batch cultures, strains IMU079, IMX1515 and IMX1583 all produced CO2, biomass, ethanol and glycerol as main products, with a minor production of acetate (Table 5, Fig. 5, Additional File 10). The strains did not produce xylitol during exponential growth and low concentrations of xylitol in cultures of strain IMU079 (2.2 ± 0.1 mmol L−1) were only observed at the end of fermentation. As observed in aerobic cultures (Fig. 4), strain IMU079 showed a clear biphasic growth profile in the anaerobic bioreactors (Fig. 5, Additional File 10), during which a fast glucose phase (ca. 16 h) was followed by a much slower and decelerating xylose consumption phase. During the glucose phase, this reference strain maintained a specific growth rate of 0.29 h−1 and a glucose-xylose co-consumption ratio of 0.14 mol mol−1 (Table 5). After a ca. 30-h lag phase (Fig. 5, Additional File 10), strain IMX1515 exhibited an exponential growth rate of 0.07 ± 0.00 h−1, with a high glucose-xylose co-consumption ratio (0.45 ± 0.03 mol mol−1). Mainly as a result of its lag phase, strain IMX1515 took longer to consume all sugars than the reference strain IMU079, but its xylose-consumption phase was ca. 65% shorter (ca. 14 and 43 h, respectively; Fig. 5, Additional File 10). Figure 5. Open in new tabDownload slide Sugar consumption, biomass and metabolite production profiles of S. cerevisiae strains IMU079 (XKS1↑ PPP↑ pAKX002; a, b), IMX1515 (hxk2Δ rsp5Δ XKS1↑ PPP↑ pAKX002; c, d) and IMX1583 (hxk2Δ gal83::GAL83G673T XKS1↑ PPP↑ pAKX002; e, f), grown on SM with 20 g L−1 glucose and 10 g L−1 xylose in anaerobic bioreactor batch cultures (pH 5, 30°C). Cultures were grown in duplicate, the data shown are from a single representative culture. a, c, e: filled circles glucose, open circles xylose; b, d, f: filled squares biomass, open squares ethanol, filled triangles acetate, open triangles glycerol. Data on ethanol corrected for evaporation. For duplicate cultures, refer to Additional File 10. Table 5. Product yields, biomass-specific rates and sugar uptake ratios in anaerobic bioreactor batch cultures of strains IMU079 (XKS1↑ PPP↑ pAKX002), IMX1515 (hxk2Δ rsp5Δ XKS1↑ PPP↑ pAKX002) and IMX1583 (hxk2Δ gal83::GAL83G673T XKS1↑ PPP↑ pAKX002) grown on SM supplemented with 20 g L−1 glucose and 10 g L−1 xylose (pH 5, 30°C). Strain . IMU079 . IMX1515 . IMX1583 . Relevant genotype HXK2 RSP5 GAL83 hxk2Δ rsp5Δ hxk2Δ gal83::GAL83G673T μ (h−1) 0.29 ± 0.01 0.07 ± 0.00 0.21 ± 0.00 Spec. xylose uptake rate mmol (g biomass)−1 h−1 2.22 ± 0.14 2.50 ± 0.12 3.19 ± 0.02 Spec. glucose uptake rate mmol (g biomass)−1 h−1 15.65 ± 0.52 5.58 ± 0.08 10.09 ± 0.08 Glucose-xylose consumption ratio (mol mol−1) 0.14 ± 0.00 0.45 ± 0.03 0.32 ± 0.01 Biomass yield on sugars (g biomass g−1) 0.09 ± 0.01 0.05 ± 0.00 0.09 ± 0.00 Ethanol yield on sugars (g g−1) 0.37 ± 0.00 0.43 ± 0.00 0.38 ± 0.01 Glycerol yield on sugars (g g−1) 0.10 ± 0.00 0.06 ± 0.00 0.08 ± 0.00 Ratio glycerol production on biomass production (mmol (g biomass)−1) 11.5 ± 0.60 12.0 ± 0.50 9.9 ± 0.10 Xylitol production (mmol L−1) 2.22 ± 0.06 0.90 ± 0.04 0.35 ± 0.04 Strain . IMU079 . IMX1515 . IMX1583 . Relevant genotype HXK2 RSP5 GAL83 hxk2Δ rsp5Δ hxk2Δ gal83::GAL83G673T μ (h−1) 0.29 ± 0.01 0.07 ± 0.00 0.21 ± 0.00 Spec. xylose uptake rate mmol (g biomass)−1 h−1 2.22 ± 0.14 2.50 ± 0.12 3.19 ± 0.02 Spec. glucose uptake rate mmol (g biomass)−1 h−1 15.65 ± 0.52 5.58 ± 0.08 10.09 ± 0.08 Glucose-xylose consumption ratio (mol mol−1) 0.14 ± 0.00 0.45 ± 0.03 0.32 ± 0.01 Biomass yield on sugars (g biomass g−1) 0.09 ± 0.01 0.05 ± 0.00 0.09 ± 0.00 Ethanol yield on sugars (g g−1) 0.37 ± 0.00 0.43 ± 0.00 0.38 ± 0.01 Glycerol yield on sugars (g g−1) 0.10 ± 0.00 0.06 ± 0.00 0.08 ± 0.00 Ratio glycerol production on biomass production (mmol (g biomass)−1) 11.5 ± 0.60 12.0 ± 0.50 9.9 ± 0.10 Xylitol production (mmol L−1) 2.22 ± 0.06 0.90 ± 0.04 0.35 ± 0.04 Rates, yields and ratios were calculated from samples taken during the mid-exponential growth phase and represent averages ± mean deviation of independent duplicate cultures. Ethanol yields were corrected for evaporation. Open in new tab Table 5. Product yields, biomass-specific rates and sugar uptake ratios in anaerobic bioreactor batch cultures of strains IMU079 (XKS1↑ PPP↑ pAKX002), IMX1515 (hxk2Δ rsp5Δ XKS1↑ PPP↑ pAKX002) and IMX1583 (hxk2Δ gal83::GAL83G673T XKS1↑ PPP↑ pAKX002) grown on SM supplemented with 20 g L−1 glucose and 10 g L−1 xylose (pH 5, 30°C). Strain . IMU079 . IMX1515 . IMX1583 . Relevant genotype HXK2 RSP5 GAL83 hxk2Δ rsp5Δ hxk2Δ gal83::GAL83G673T μ (h−1) 0.29 ± 0.01 0.07 ± 0.00 0.21 ± 0.00 Spec. xylose uptake rate mmol (g biomass)−1 h−1 2.22 ± 0.14 2.50 ± 0.12 3.19 ± 0.02 Spec. glucose uptake rate mmol (g biomass)−1 h−1 15.65 ± 0.52 5.58 ± 0.08 10.09 ± 0.08 Glucose-xylose consumption ratio (mol mol−1) 0.14 ± 0.00 0.45 ± 0.03 0.32 ± 0.01 Biomass yield on sugars (g biomass g−1) 0.09 ± 0.01 0.05 ± 0.00 0.09 ± 0.00 Ethanol yield on sugars (g g−1) 0.37 ± 0.00 0.43 ± 0.00 0.38 ± 0.01 Glycerol yield on sugars (g g−1) 0.10 ± 0.00 0.06 ± 0.00 0.08 ± 0.00 Ratio glycerol production on biomass production (mmol (g biomass)−1) 11.5 ± 0.60 12.0 ± 0.50 9.9 ± 0.10 Xylitol production (mmol L−1) 2.22 ± 0.06 0.90 ± 0.04 0.35 ± 0.04 Strain . IMU079 . IMX1515 . IMX1583 . Relevant genotype HXK2 RSP5 GAL83 hxk2Δ rsp5Δ hxk2Δ gal83::GAL83G673T μ (h−1) 0.29 ± 0.01 0.07 ± 0.00 0.21 ± 0.00 Spec. xylose uptake rate mmol (g biomass)−1 h−1 2.22 ± 0.14 2.50 ± 0.12 3.19 ± 0.02 Spec. glucose uptake rate mmol (g biomass)−1 h−1 15.65 ± 0.52 5.58 ± 0.08 10.09 ± 0.08 Glucose-xylose consumption ratio (mol mol−1) 0.14 ± 0.00 0.45 ± 0.03 0.32 ± 0.01 Biomass yield on sugars (g biomass g−1) 0.09 ± 0.01 0.05 ± 0.00 0.09 ± 0.00 Ethanol yield on sugars (g g−1) 0.37 ± 0.00 0.43 ± 0.00 0.38 ± 0.01 Glycerol yield on sugars (g g−1) 0.10 ± 0.00 0.06 ± 0.00 0.08 ± 0.00 Ratio glycerol production on biomass production (mmol (g biomass)−1) 11.5 ± 0.60 12.0 ± 0.50 9.9 ± 0.10 Xylitol production (mmol L−1) 2.22 ± 0.06 0.90 ± 0.04 0.35 ± 0.04 Rates, yields and ratios were calculated from samples taken during the mid-exponential growth phase and represent averages ± mean deviation of independent duplicate cultures. Ethanol yields were corrected for evaporation. Open in new tab In contrast to strain IMX1515, strain IMX1583 (hxk2Δ gal83::GAL83G673T XKS1↑ PPP↑ pAKX002) did not exhibit a lag phase but immediately started exponential growth at 0.21 h−1 (Fig. 5). A comparison of biomass-specific uptake rates of xylose and glucose in the anaerobic batch experiments showed that strain IMX1583 maintained a 44% higher xylose uptake rate than strain IMU079 before glucose exhaustion (Table 5). Moreover, both strains IMX1515 and IMX1583 did not show the pronounced decline of xylose consumption after glucose exhaustion that was observed in the reference strain (Fig. 5, Additional File 10). As a result, the xylose consumption phase in anaerobic cultures of strain IMX1583 was 80% shorter than that strain IMU079 (ca. 9 h compared to 43 h), thereby reducing the time required for complete sugar conversion by over 24 h (Fig. 5). DISCUSSION Engineering S. cerevisiae for forced co-consumption of xylose and glucose The key objective of this study was to develop and test a strain platform that, via laboratory evolution and subsequent genome resequencing, can be used to identify mutations that support co-metabolism of xylose and glucose. In previous studies, laboratory evolution of glucose-phosphorylation-negative, pentose-fermenting strains in the presence of glucose yielded valuable leads for improving utilisation of glucose-xylose mixtures, including mutations in HXT genes that improved pentose uptake in the presence of glucose (Farwick et al.2014; Nijland et al.2014, 2017; Wisselink, Van Maris and Pronk 2015). The strategy described in this study differs from these previous studies as it not only enabled selection for xylose utilisation in the presence of glucose, but also for simultaneous metabolism of the two sugars. The molar ratio of xylose and glucose co-consumption (1.64 mol xylose (mol glucose)−1) by the evolved strain IMS0629 is the highest reported to date for batch cultures of S. cerevisiae (Katahira et al.2008; Shen et al.2012, 2015; Gonçalves et al.2014; Nijland et al.2014; Westman et al.2014; Shin et al.2015). Inactivation of PGI1 played a key role in forcing simultaneous utilisation of xylose and glucose (Fig. 1). To pre-empt the well-documented inability of S. cerevisiae pgi1Δ mutants to reoxidise the NADPH that is generated when metabolism of glucose-6-phosphate is rerouted through the oxidative PPP (Boles, Lehnert and Zimmermann 1993; Heux, Cadiere and Dequin 2008), we replaced Gnd1 and Gnd2 with the NAD+-linked bacterial 6-phosphogluconate dehydrogenase GndA (Papapetridis et al.2016). Together with the co-consumption of xylose, via an engineered pathway that did not involve NAD(P)H generation (Fig. 1), these modifications enabled the engineered strain IMX1046 to grow on mixtures of xylose and 1 g L−1 glucose, without the fructose supplementation that is normally required for growth of pgi1-null mutants on glucose (Aguilera 1986; Boles, Lehnert and Zimmermann 1993). The evolved strain IMS0629 consumed 8.6 mmol glucose (g biomass)−1 in aerobic batch cultures (Table 2), which is close to the requirement of 9.3 mmol NADPH (g biomass)−1 for aerobic growth on glucose of wild-type S. cerevisiae (Bruinenberg, Van Dijken and Scheffers 1983). However, since glucose-6-phosphate is a key biosynthetic precursor, not all glucose consumed by the cultures can be converted through the oxidative PPP. Additional enzymes such as NADP+-dependent acetaldehyde dehydrogenase Ald6 (Grabowska and Chelstowska 2003; Celton et al.2012; Papapetridis et al.2016) are therefore likely to have supplemented NADPH generation via the oxidative PPP in the ‘forced co-consumption’ strains. Improvement of mixed-sugar fermentation in aerobic and anaerobic cultures of xylose-fermenting S. cerevisiae Mutations in HXK2, RSP5 and GAL83 were reverse engineered into a non-evolved strain (XKS1↑ PPP↑) to investigate whether mutations acquired during evolution of strain IMX1046 (pgi1Δ rpe1Δ gnd1Δ gnd2Δ gndA xylA XKS1↑ PPP↑) on glucose-xylose mixtures would also stimulate mixed-substrate utilisation in a genetic background that does not impose forced co-utilisation. During growth on glucose-xylose mixtures, the reference strain IMU079 (XKS1↑ PPP↑ pAKX002) displayed the typical biphasic growth profile seen in non-evolved, xylose-consuming strains that express a basic, functional xylose-isomerase (XI)-based xylose fermentation pathway (Kuyper et al.2003, 2005a). Biphasic growth was especially pronounced in anaerobic cultures, in which the xylose consumption rate collapsed upon glucose depletion (Fig. 5). Deletion of HXK2, either combined with the deletion of RSP5 or with the introduction of a GAL83G673T mutation, strongly improved mixed-sugar fermentation kinetics, both by increased co-utilisation and by faster conversion of xylose after glucose had been depleted (Figs 4 and 5). Hxk2, the major hexokinase in S. cerevisiae, plays an additional key role in transcriptional repression of a large set of yeast genes (Entian 1980; Entian and Fröhlich 1984; Peláez, Herrero and Moreno 2010) by glucose. Deletion of HXK2 has been shown to enhance co-consumption of combinations of natural substrates (glucose-galactose, glucose-sucrose and glucose-ethanol) in batch cultures (Raamsdonk et al.2001). During exponential growth on glucose in batch cultures, hxk2Δ mutants show increased transcription of the high-affinity hexose transporter genes HXT2 and HXT7 and decreased transcription of the low-affinity hexose transporter genes HXT1 and HXT3 (Petit et al.2000). The high-affinity Hxt transporters, which in wild-type strains are only expressed at low glucose concentrations (Diderich et al.1999), have a much lower Km for xylose than their low-affinity counterparts (Hamacher et al.2002; Lee et al.2002). Several studies have demonstrated that overexpression of high-affinity hexose transporters stimulates xylose uptake (Subtil and Boles 2012; Farwick et al.2014; Gonçalves et al.2014; Nijland et al.2014, 2016; Reznicek et al.2015; Shin et al.2015; Jansen et al.2017). The observed improved co-utilisation of glucose and xylose upon inactivation of HXK2 may therefore reflect an increased abundance of high-affinity hexose transporters in the yeast plasma membrane during growth on glucose-xylose mixtures. A recent in silico study identified HXK2 as potential target for improving xylose uptake rates in S. cerevisiae (Miskovic et al.2017). However, subsequent deletion of HXK2 in a strain expressing a XR/XDH-based pathway resulted in faster xylose uptake but also increased by-product formation and reduced ethanol productivity (Miskovic et al.2017). The absence of such negative effects in the present study is in line with reports that XI-based strains are less prone to by-product formation than XR/XDH-based strains (Moysés et al.2016; Jansen et al.2017). Rsp5, the only representative of the NEDD4 family of E3-ubiquitin ligases in S. cerevisiae, is involved in regulation of a multitude of cellular processes, including intracellular protein trafficking, regulation of the large subunit of RNA polymerase II, ribosome stability, regulation of fatty acid synthesis and stress response (Huibregtse, Yang and Beaudenon 1997; Belgareh-Touzé et al.2008; Kaliszewski and Zoładek 2008; Shcherbik and Pestov 2011). This multitude of roles may explain the reduced growth rate of the rsp5Δ strains (Tables 4 and 5) in this study. Involvement of Rsp5 in ubiquitination and subsequent endocytosis of the high-affinity hexose transporters Hxt6 and Hxt7 (Krampe et al.1998; Haguenauer-Tsapis and André 2004) could explain the strong synergistic effect of the hxk2Δ and rsp5Δ deletions: while deletion of HXK2 prevents glucose repression of the synthesis of these transporters, deletion of RSP5 could prevent their ubiquitination and removal from the membrane. Removal of ubiquitination sites in the hexose transporters Hxt1 and Hxt36 was previously shown to enhance xylose uptake by S. cerevisiae (Nijland et al.2016). Our results suggest that a similar modification of Hxt6 and Hxt7 could also be beneficial. Gal83 is one of three possible β-subunits of the Snf1-kinase complex, which enables transcription of glucose-repressed genes at low glucose concentrations (for reviews, see Gancedo 1998; Schüller 2003). At non-repressing glucose concentrations, Gal83 directs the Snf1-Gal83 complex to the cell nucleus (Vincent et al.2001), where it mediates transcriptional upregulation of genes involved in utilisation of alternative carbon sources (Vincent and Carlson 1999). Targets of the Snf1-Gal83 complex include the GAL regulon (Matsumoto, Toh-e and Oshima 1981; Erickson and Johnston 1993) and the high-affinity hexose transporter genes HXT2 and HXT4 (Özcan and Johnston 1999). The D225Y substitution, which stimulated glucose-xylose co-consumption in the present study, is located in the glycogen-binding domain of Gal83 (residues 161–243; Momcilovic et al.2008). Other mutations in this domain have been shown to cause transcription of Snf1-Gal83 targets in the presence of glucose (Wiatrowski et al.2004; Momcilovic et al.2008; Mangat et al.2010). In contrast to deletion of the transcriptional regulator CYC8 (Nijland et al.2017), which also stimulated co-utilisation but caused severe reductions of the specific growth rate of engineered strains, the GAL83G673T mutation did not have a strong impact on growth rate (Tables 4 and 5). The synergistic effect of the hxk2Δ and GAL83G673T mutations may be related to the involvement of Hxk2 in deactivation of Snf1 in the presence of glucose, causing constitutive activity of the Snf1-Gal83 complex in hxk2-null mutants (Gancedo 1998; Schüller 2003). The reverse engineered mutations in HXK2 and GAL83 or RSP5 not only stimulated simultaneous utilisation of xylose and glucose when both sugars were present, but also prevented the sharp decline in xylose uptake rates that occurred in the reference strain IMU079 (XKS1↑ PPP↑ pAKX002). In the reference strain, the biomass-specific rate of xylose consumption, probably mediated by low- or moderate-affinity Hxt transporters, declined to values below 0.5 mmol (g biomass)−1 h−1 after glucose depletion (Fig. 5, Additional File 10). Under anaerobic conditions, this low rate of xylose fermentation would correspond to a biomass-specific rate of ATP production of 0.8 mmol (g biomass)−1 h−1. This value is lower than the estimated ATP requirement for cellular maintenance of S. cerevisiae (ca. 1 mmol ATP (g biomass)−1 h−1; Boender et al.2009). Since protein synthesis is a highly ATP-intensive process (Canelas et al.2010), an inability of the reference strain to functionally express high-affinity hexose transporters upon glucose depletion may therefore reflect an energy shortage. A similar effect was observed during transitions between glucose and galactose growth in anaerobic S. cerevisiae cultures (van den Brink et al.2009). By already expressing functional high-affinity transporters before glucose was depleted, the hxk2Δ rsp5Δ and hxk2Δ GAL83G673Tgenotypes may have enabled cells to avoid such a bioenergetic ‘valley of death’ upon the transition to xylose fermentation. In industrial processes using lignocellulosic feedstocks, this energetic challenge is likely to be even more stringent due to the presence of compounds such as acetic acid that increase maintenance energy requirements (Verduyn et al.1992; Abbott et al.2007; Bellissimi et al.2009). While reverse engineering of HXK2, RSP5 and GAL83 mutations demonstrated the relevance of the forced co-utilisation strategy demonstrated in this study, they do not exhaust its possibilities. Prolongation of the evolution experiments, combined with anaerobic conditions and/or a further increase of the glucose to xylose ratio in the medium, may allow for selection of additional relevant mutations. The IMS0629 strain may also be used to select mutations that enable efficient co-utilisation of glucose and xylose at different concentrations or in lignocellulosic hydrolysates that, in addition to fermentable sugars, contain inhibitors of yeast performance (Palmqvist and Hahn-Hägerdal 2000; Klinke et al.2002; Almeida et al.2007). Alternatively, the strain design can be adapted to enable selection for co-metabolism of other sugars, for example by replacing the xylose pathway by a bacterial pathway for conversion of L-arabinose into xylulose-5-phosphate (Becker and Boles 2003; Wisselink et al.2007). This strategy can be extended to optimising the yeast metabolic network for aerobic production of economically relevant compounds other than ethanol from lignocellulosic hydrolysates. In particular, imposing fixed stoichiometries of glycolytic and (non-oxidative) PPP reactions may offer interesting options for high-yield production of compounds whose synthesis requires a large net input of PPP intermediates and/or NADPH, such as aromatic compounds and lipids (Tehlivets, Scheuringer and Kohlwein 2007; Klug and Daum 2014; Gottardi et al.2017). CONCLUSIONS Engineering of carbon and redox metabolism yielded an S. cerevisiae strain whose growth was strictly dependent on the simultaneous uptake and metabolism of xylose and glucose. Laboratory evolution improved growth of the resulting strains on mixtures of xylose and glucose at elevated glucose concentrations. Mutations in HXK2, RSP5 and GAL83 were identified by genome sequencing of the evolved strains. Upon their combined introduction into an engineered xylose-fermenting yeast strain, these mutations strongly stimulated simultaneous utilisation of xylose and glucose and, after depletion of glucose, fast conversion of the remaining xylose. The developed strain platform and modified versions thereof can be used for identification of further metabolic engineering targets for improving the performance of yeast strains in industrial processes based on lignocellulosic feedstocks. SUPPLEMENTARY DATA Supplementary data are available at FEMSYR online. Acknowledgements We thank Jasmine Bracher and Oscar Martinez for construction of strain IMU079. We thank Jean-Marc Daran and Pilar de la Torre for advice on molecular biology, Marcel van den Broek for assistance with bioinformatics analyses and Erik de Hulster for advice on fermentations. We thank Laura Valk for stimulating discussions. Availability of supporting data The genomic sequence of strains IMS0629 and IMS0634 have been deposited in Genbank (http://www.ncbi.nlm.nih.gov/), BioProject number PRJNA433919. Author's contributions IP, MDV, AJAvM and JTP designed experiments. IP, MDV and JTP wrote the first draft of the manuscript. IP and MDV constructed yeast strains and performed fermentations. SJW constructed yeast strains and performed in silico analyses. MG designed and carried out fermentations. All authors read and commented upon the draft manuscript and approved the final version. Funding The PhD project of IP is funded by DSM. This work was performed within the BE-Basic R&D Program (http://www.be-basic.org/), which is financially supported by an EOS Long Term grant from the Dutch Ministry of Economic Affairs, Agriculture and Innovation (EL&I). Competing interests IP, MDV, AJAvM and JTP are co-inventors on a patent application related to the content of this work. REFERENCES Abbott DA , Knijnenburg TA, De Poorter LMI et al. Generic and specific transcriptional responses to different weak organic acids in anaerobic chemostat cultures of Saccharomyces cerevisiae . FEMS Yeast Res 2007 ; 7 : 819 – 33 . Google Scholar Crossref Search ADS PubMed WorldCat Aguilera A . Deletion of the phosphoglucose isomerase structural gene makes growth and sporulation glucose dependent in Saccharomyces cerevisiae . Mol Gen Genet 1986 ; 204 : 310 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat Almeida JRM , Modig T, Petersson A et al. Increased tolerance and conversion of inhibitors in lignocellulosic hydrolysates by Saccharomyces cerevisiae . J Chem Technol Biot 2007 ; 82 : 340 – 9 . Google Scholar Crossref Search ADS WorldCat Alper H , Stephanopoulos G. Engineering for biofuels: exploiting innate microbial capacity or importing biosynthetic potential ? Nat Rev Microbiol 2009 ; 7 : 715 – 23 . Google Scholar Crossref Search ADS PubMed WorldCat Ask M , Bettiga M, Duraiswamy VR et al. Pulsed addition of HMF and furfural to batch-grown xylose-utilizing Saccharomyces cerevisiae results in different physiological responses in glucose and xylose consumption phase . Biotechnol Biofuels 2013 ; 6 : 181 . Google Scholar Crossref Search ADS PubMed WorldCat Aung HW , Henry SA, Walker LP. Revising the representation of fatty acid, glycerolipid, and glycerophospholipid metabolism in the consensus model of yeast metabolism . Ind Biotechnol 2013 ; 9 : 215 – 28 . Google Scholar Crossref Search ADS WorldCat Bakker BM , Overkamp KM, van Maris AJA et al. Stoichiometry and compartmentation of NADH metabolism in Saccharomyces cerevisiae . FEMS Microbiol Rev 2001 ; 25 : 15 – 37 . Google Scholar Crossref Search ADS PubMed WorldCat Basso LC , Basso T, Nitsche Rocha S. Ethanol production in Brazil: the industrial process and its impact on yeast fermentation . Biofuel Production - Recent Developments and Prospects . Rijeka : IntechOpen , 2011 , DOI: 10.5772/17047 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Becker J , Boles E. A modified Saccharomyces cerevisiae strain that consumes l-arabinose and produces ethanol . Appl Environ Microb 2003 ; 69 : 4144 – 50 . Google Scholar Crossref Search ADS WorldCat Becker SA , Feist AM, Mo ML et al. Quantitative prediction of cellular metabolism with constraint-based models: the COBRA Toolbox . Nat Protoc 2007 ; 2 : 727 – 38 . Google Scholar Crossref Search ADS PubMed WorldCat Belgareh-Touzé N , Léon S, Erpapazoglou Z et al. Versatile role of the yeast ubiquitin ligase Rsp5p in intracellular trafficking: Figure 1 . Biochem Soc Trans 2008 ; 36 : 791 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat Bellissimi E , Van Dijken JP, Pronk JT et al. Effects of acetic acid on the kinetics of xylose fermentation by an engineered, xylose-isomerase-based Saccharomyces cerevisiae strain . FEMS Yeast Res 2009 ; 9 : 358 – 64 . Google Scholar Crossref Search ADS PubMed WorldCat Boender LGM , de Hulster EAF, van Maris AJA et al. Quantitative physiology of Saccharomyces cerevisiae at near-zero specific growth rates . Appl Environ Microb 2009 ; 75 : 5607 – 14 . Google Scholar Crossref Search ADS WorldCat Boles E , Lehnert W, Zimmermann FK. The role of the NAD-dependent glutamate dehydrogenase in restoring growth on glucose of a Saccharomyces cerevisiae phosphoglucose isomerase mutant . Eur J Biochem 1993 ; 217 : 469 – 77 . Google Scholar Crossref Search ADS PubMed WorldCat Bruinenberg PM , Van Dijken JP, Scheffers WA. A theoretical analysis of NADPH production and consumption in yeasts . Microbiology 1983 ; 129 : 953 – 64 . Google Scholar Crossref Search ADS WorldCat Canelas AB , Harrison N, Fazio A et al. Integrated multilaboratory systems biology reveals differences in protein metabolism between two reference yeast strains . Nat Commun 2010 ; 1 : 145 . Google Scholar Crossref Search ADS PubMed WorldCat Celton M , Sanchez I, Goelzer A et al. A comparative transcriptomic, fluxomic and metabolomic analysis of the response of Saccharomyces cerevisiae to increases in NADPH oxidation . BMC Genomics 2012 ; 13 : 317 . Google Scholar Crossref Search ADS PubMed WorldCat De Winde JH , Crauwels M, Hohmann S et al. Differential requirement of the yeast sugar kinases for sugar sensing in establishing the catabolite-repressed state . Eur J Biochem 1996 ; 241 : 633 – 43 . Google Scholar Crossref Search ADS PubMed WorldCat DiCarlo JE , Norville JE, Mali P et al. Genome engineering in Saccharomyces cerevisiae using CRISPR-Cas systems . Nucleic Acids Res 2013 ; 41 : 4336 – 43 . Google Scholar Crossref Search ADS PubMed WorldCat Dickinson JR , Sobanski MA, Hewlins MJE. In Saccharomyces cerevisiae deletion of phosphoglucose isomerase can be suppressed by increased activities of enzymes of the hexose monophosphate pathway . Microbiology 1995 ; 141 : 385 – 91 . Google Scholar Crossref Search ADS PubMed WorldCat Diderich JA , Schepper M, van Hoek P et al. Glucose uptake kinetics and transcription of HXTGenes in chemostat cultures of Saccharomyces cerevisiae . J Biol Chem 1999 ; 274 : 15350 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Entian K-D . Genetic and biochemical evidence for hexokinase PII as a key enzyme involved in carbon catabolite repression in yeast . Mol Gen Genet 1980 ; 178 : 633 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat Entian K-D , Fröhlich K-U. Saccharomyces cerevisiae mutants provide evidence of hexokinase PII as a bifunctional enzyme with catalytic and regulatory domains for triggering catabolite repression . J Bacteriol 1984 ; 158 : 29 – 35 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Entian K-D , Kötter P. 25 yeast genetic strain and plasmid collections . In: Stansfield I, Stark MJR (eds.) Methods in Microbiology , vol. 36 . London : Academic Press , 2007 , 629 – 66 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Erickson JR , Johnston M. Genetic and molecular characterization of Gal83: Its interaction and similarities with other genes involved in glucose repression in Saccharomyces cerevisiae . Genetics 1993 ; 135 : 655 – 64 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Farwick A , Bruder S, Schadeweg V et al. Engineering of yeast hexose transporters to transport D-xylose without inhibition by D-glucose . Proc Natl Acad Sci USA 2014 ; 111 : 5159 – 64 . Google Scholar Crossref Search ADS PubMed WorldCat Fulton LM , Lynd LR, Körner A et al. The need for biofuels as part of a low carbon energy future . Biofuels Bioprod Bioref 2015 ; 9 : 476 – 83 . Google Scholar Crossref Search ADS WorldCat Gancedo JM . Yeast carbon catabolite repression . Microbiol Mol Biol R 1998 ; 62 : 334 – 61 . Google Scholar OpenURL Placeholder Text WorldCat Garcia Sanchez R , Karhumaa K, Fonseca C et al. Improved xylose and arabinose utilization by an industrial recombinant Saccharomyces cerevisiae strain using evolutionary engineering . Biotechnol Biofuels 2010 ; 3 : 13 . Google Scholar Crossref Search ADS PubMed WorldCat Gawand P , Hyland P, Ekins A et al. Novel approach to engineer strains for simultaneous sugar utilization . Metab Eng 2013 ; 20 : 63 – 72 . Google Scholar Crossref Search ADS PubMed WorldCat Gietz RD , Woods RA. Transformation of yeast by lithium acetate/single-stranded carrier DNA/polyethylene glycol method . In: Guthrie C, Fink GR (eds.) Methods in Enzymology , vol. 350 : Academic Press , 2002 , 87 – 96 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gombert AK , van Maris AJA. Improving conversion yield of fermentable sugars into fuel ethanol in 1st generation yeast-based production processes . Curr Opin Biotechnol 2015 ; 33 : 81 – 86 . Google Scholar Crossref Search ADS PubMed WorldCat Gonçalves DL , Matsushika A, de Sales BB et al. Xylose and xylose/glucose co-fermentation by recombinant Saccharomyces cerevisiae strains expressing individual hexose transporters . Enzyme Microb Technol 2014 ; 63 : 13 – 20 . Google Scholar Crossref Search AD