Increasing free-energy (ATP) conservation in maltose-grown Saccharomyces cerevisiae by expression of a heterologous maltose phosphorylase


of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Increasing free-energy (ATP) conservation in maltose-grown Saccharomyces cerevisiae by expression of a heterologous maltose phosphorylase
  Increasing free-energy (ATP) conservation in maltose-grown  Saccharomycescerevisiae  by expression of a heterologous maltose phosphorylase Stefan de Kok, Duygu Yilmaz, Erwin Suir, Jack T. Pronk, Jean-Marc Daran, Antonius J.A. van Maris n Department of Biotechnology, Delft University of Technology and Kluyver Centre for Genomics of Industrial Fermentation, Julianalaan 67, 2628 BC Delft, The Netherlands a r t i c l e i n f o  Article history: Received 11 March 2011Received in revised form16 May 2011Accepted 1 June 2011Available online 14 June 2011 Keywords: Metabolic engineeringSynthetic biologyYeastEnergetics b -PhosphoglucomutaseMaltase a b s t r a c t Increasing free-energy conservation from the conversion of substrate into product is crucial for furtherdevelopment of many biotechnological processes. In theory, replacing the hydrolysis of disaccharidesby a phosphorolytic cleavage reaction provides an opportunity to increase the ATP yield on thedisaccharide. To test this concept, we first deleted the native maltose metabolism genes in Saccharomyces cerevisiae . The knockout strain showed no maltose-transport activity and a very lowresidual maltase activity (0.03 m mol mg protein  1 min  1 ). Expression of a maltose phosphorylase genefrom  Lactobacillus sanfranciscensis  and the  MAL11  maltose-transporter gene resulted in relatively slowgrowth ( m aerobic   0.09 7 0.03 h  1 ). Co-expression of   Lactococcus lactis  b -phosphoglucomutase acceleratedmaltose utilization via this route ( m aerobic   0.21 7 0.01 h  1 ,  m anaerobic   0.10 7 0.00 h  1 ). Replacing maltosehydrolysis with phosphorolysis increased the anaerobic biomass yield on maltose in anaerobic maltose-limited chemostat cultures by 26%, thus demonstrating the potential of phosphorolysis to improve thefree-energy conservation of disaccharide metabolism in industrial microorganisms. &  2011 Elsevier Inc. All rights reserved. 1. Introduction Showcases such as the biotechnological production of 1,3-propa-nediol with  Escherichia coli  (Nakamura and Whited, 2003) and the anti-malarial precursor artemisinic acid with  Saccharomyces cerevisiae (Ro et al., 2006) demonstrate the maturation of metabolic engineer- ing. Introduction and optimization of heterologous enzymes andpathways through metabolic modeling, synthetic biology and highthroughput screening allow production of a wide range of biologicalmolecules (Dietrich et al., 2010; Na et al., 2010). The development of  efficient microorganisms that closely approximate maximum theore-tical product yields requires cellular homeostasis of the redoxcofactors (e.g. NAD(P)(H)) and free energy (e.g. in the form of ATP)for growth, cellular maintenance and/or product formation (Boenderet al., 2009; Nasution et al., 2008; Sharma et al., 2007). Aerobic respiration enables redox cofactor regeneration forproduct pathways that would otherwise result in a surplusof NAD(P)H (Grewal and Kalra, 1995; Kimura, 2003) and can also provide the cells with ATP via oxidative phosphorylation.However, aeration of industrial scale fermentations is expensivedue to the cost of stirring, air compression and cooling. Inaddition, aeration results in dissimilation of part of the substrateto CO 2 , thereby decreasing the product yield. Therefore, wherepossible, it would be beneficial to produce commodity chemicalsthrough redox-neutral pathways, as is the case for alcohol, lacticacid and many metabolic engineering targets, which allowsindustrial production under anaerobic conditions.In the conversion of glucose via a classical Embden-Meyerhof glycolytic pathway, substrate-level phosphorylation results in thenet formation of 2 ATP for each molecule of glucose converted.A challenging situation arises for products of interest whose forma-tion from glucose does not result in a net formation of ATP whenproduced through redox-neutral routes under anaerobic conditions.In many such cases, the ATP formed in glycolysis by substrate-levelphosphorylation may subsequently be used for carboxylation reac-tions (Zelle et al., 2010; Zhang et al., 2009), product export (van Maris et al., 2004b) or the formation of acyl-CoA esters (Singh et al.,2010; van Maris et al., 2004a). Production of lactic acid by metabo- lically engineered  S. cerevisiae  is an illustrative example. Theconversion of glucose to 2 molecules of lactic acid yields 2 ATP.However, in  S. cerevisiae  export of lactic acid is hypothesized torequire 1 ATP per lactic acid. This results in a ‘zero-ATP pathway’from glucose to extracellular lactic acid. This phenomenon presentsan intrinsic limitation for efficient production of lactic acid underanaerobic conditions and at low pH (van Maris et al., 2004b) since,without a net formation of ATP, cells cannot grow or fulfill the free-energy requirements for cellular maintenance.Increasing free-energy (ATP) conservation from the conversionof substrate into product is of major importance for such ‘zeroATP pathways’. This study will explore the possibilities to increase Contents lists available at ScienceDirectjournal homepage: Metabolic Engineering 1096-7176/$-see front matter  &  2011 Elsevier Inc. All rights reserved.doi:10.1016/j.ymben.2011.06.001 n Corresponding author. Fax:  þ 31 15 278 2355. E-mail address: (A.J.A. van Maris).Metabolic Engineering 13 (2011) 518–526  free-energy conservation during growth on disaccharides. Inmany industrial microorganisms, the disaccharides are hydro-lyzed (reaction (1)), which dissipates the free energy availablefrom this cleavage. Alternatively, disaccharides can be cleavedwith inorganic phosphate (reaction (2)), thereby increasing free-energy conservation of this reaction through the direct formationof phosphorylated intermediates (Zhang and Lynd, 2005)disaccharide þ H 2 O - monosaccharide þ glucose  ð 1 Þ disaccharide þ P i - monosaccharide þ glucose-1-phosphate  ð 2 Þ Disaccharide phosphorylases are described for sucrose, maltose,cellobiose, trehalose and lactose (Alexander, 1968; Belocopitow and Mare´chal,1970;DeGroeveetal.,2009;Doudoroff,1955).Functional expression of heterologous cellobiose phosphorylase in  S. cerevisiae has recently been demonstrated (Sadie et al., 2011). The formed glucose-1-phosphate can be converted by a phosphoglucomutaseinto glucose-6-phosphate, which is further metabolized via glyco-lysis. As a result no ATP is hydrolyzed in the hexokinase reaction toconvert glucose to glucose-6-phosphate. This modification increasesthe net ATP yield by 1 ATP per disaccharide molecule.In this study, growth of   S. cerevisiae  on maltose was used as amodel. The genetic structure of maltose metabolism in  S. cerevisiae is well described (Needleman, 1991). It consists of several  MAL  loci,which contain a maltose permease ( MALx1 ), a maltase ( MALx2)  anda regulator ( MALx3) . These  MAL  loci are located subtelomericallyand the number of   MAL  loci is strain dependent (Naumov et al.,1994).Additionally, Mph2pand Mph3pcantransport maltose(Day et al., 2002). In  S. cerevisiae , maltose is imported via a protonsymport mechanism. The imported proton has to be exported tomaintain intracellular pH homeostasis. In  S. cerevisiae , the plasmamembrane ATPase expels 1 proton per ATP hydrolyzed. Maltose issubsequently hydrolyzed by the intracellular maltases into twomolecules of glucose ( D G 0 0  15.5kJmol  1 ), which can be con-verted through normal sugar metabolism. Anaerobic conversionof maltose to ethanol therefore yields 3 ATP (Van Leeuwen et al.,1992; Weusthuis et al., 1993). An anaerobic catabolic route that uses maltose phosphorolysis ( D G 0 0 þ 5.5kJmol  1 ) instead of mal-tose hydrolysis would generate 4 instead of 3 ATP per maltose.Theoretically, this ATP yield can be further increased to 5 ATP permaltose (2.5 ATP per hexose unit) when the proton coupling of maltose transport is abolished.The goal of the present study is to investigate whether maltosehydrolysis in  S. cerevisiae  can be replaced by phosphorolyticcleavage and to quantitatively analyze the impact of such areplacement on free-energy (ATP) conservation in anaerobicmaltose-grown cultures. To this end, the native maltose metabo-lism of   S. cerevisiae  was deleted and a bacterial maltose phos-phorylase was introduced. Subsequently, enzyme activities,growth rates and biomass yields on maltose were analyzed inbatch and chemostat cultures. 2. Materials and methods  2.1. Strains and maintenance The  S. cerevisiae  strains used and constructed in this study(Table 1) are congenic members of the CEN.PK family (Entian and Kotter, 1998; Van Dijken et al., 2000) and contain a constitutive MAL  activator gene  MAL2-8 C  (Gibson et al., 1997). Stock cultures were grown at 30  1 C in shake flasks containing 100 ml syntheticmedium (according to (Verduyn et al., 1990)) with 20 g l  1 glucose as carbon source. After overnight growth, 20% (v/v)glycerol was added and 1 ml aliquots were stored at   80  1 C.  2.2. Plasmid construction The plasmids and primers used in this study are listed inTables 2 and 3, respectively. The  hphNT1  marker was cloned frompFA6a- hphNT1  ( Janke et al., 2004) to pUG6 (Gueldener et al., 2002) via the  Sac  I and  Bgl II restriction sites, thereby replacing  KanMX4 marker, resulting in pUG- hphNT1 . The  ADH1  terminator was ampli-fied from plasmid pRW231 (Wisselink et al., 2007) using primersADH1t Fw and ADH1t Rv and cloned into p426GPD (Mumberg et al.,1995) via the  Kpn I and  Spe I restriction sites, thereby replacing the CYC1  terminator, resulting in pUD63.  MAL12  was amplified fromCEN.PK113-7D genomic DNA with primers MAL12 Fw and MAL12Rv and cloned into pUD63 via the  Hind III and  Spe I sites, resulting inpUDE44.  mapA  was amplified from  Lactobacillus sanfranciscensis (DSM 20451) genomic DNA using primers mapA Fw and mapA Rvand cloned into pENTR/D-TOPO using Gateway Technology (Invitro-gen, Carlsbad, USA), resulting in pUD37. The  mapA  gene wasrecombined into pvv214 (Van Mullem et al., 2003) via the LR reaction using Gateway Technology (Invitrogen, Carlsbad, USA),resulting in pUDE31. The  mapA  gene of plasmid pUD37 was clonedinto pUD63 via the  Hind III and  Spe I sites, resulting in pUDE45.  pgmB was ordered as a synthetic gene at Baseclear B.V. (Leiden, TheNetherlands). The gene was codon-optimized for  S. cerevisiae according to Grote et al. (2005), and was flanked by attB-sites and Pst  I and  BamH  I restriction sites 5 0 of the gene and  Hind III and  Sal Isites 3 0 of the gene (GenBank accession no. JF514919). The  pgmB gene was first cloned to pDONR221 via the BP reaction usingGateway Technology (Invitrogen, Carlsbad, USA), resulting inpUD88 and then cloned to pAG426GPD (Alberti et al., 2007) viathe LR reaction using Gateway Technology (Invitrogen, Carlsbad,USA), resulting in pUDE60. The gene was cloned into pUD63 via the Spe I and  Xho I sites, resulting in pUDE63. The  mapA  gene and  PGK1 promoter and  CYC1  terminator were amplified from pUDE31 usingprimers PGK1p Fw and CYC1t Rv and cloned into pUDE63 using the  Age I and  Sgf  I restriction sites, resulting in pUDE82. A multiplecloning site was constructed using primers Mcs Fw and Mcs Rvand cloned into pENTR/D-TOPO, resulting in pUD65. The  MAL11 gene was amplified from CEN.PK113-7D genomic DNA using  Table 1 Saccharomyces cerevisiae  strains used in this study. Strain Relevant genotype Source CEN.PK113-7D MATa  MAL1x MAL2x MAL3x MAL4x LEU2 URA3 MAL2-8 C  P. K¨otter, GermanyCEN.PK102-3A MATa  MAL1x MAL2x MAL3x leu2-112 ura3-52 MAL2-8 C  P. K¨otter, GermanyIMK289 ( mal D  mph D ) CEN.PK102-3A  mal11-mal12::loxP mal21-mal22::loxP mal31-mal32::loxP mph2::loxP mph3::loxP   This studyIMX020 IMK289 pRS405 ( LEU2 ) This studyIMX032 IMK289 pUDI35 ( LEU2 MAL11 ) This studyIMZ133 IMK289 pRS405 ( LEU2 ) pUD63 ( URA3)  This studyIMZ135 IMK289 pRS405 ( LEU2 ) pUDE44 ( URA3 MAL12 ) This studyIMZ229 IMK289 pUDI35 ( LEU2 MAL11 ) pUD63 ( URA3)  This studyIMZ199 IMK289 pUDI35 ( LEU2 MAL11 ) pUDE44 ( URA3 MAL12 ) This studyIMZ203 IMK289 pUDI35 ( LEU2 MAL11 ) pUDE45 ( URA3 mapA ) This studyIMZ226 IMK289 pUDI35 ( LEU2 MAL11 ) pUDE82 ( URA3 mapA pgmB ) This study S. de Kok et al. / Metabolic Engineering 13 (2011) 518–526   519  primers MAL11 Fw and MAL11 Rv, cloned into pUD65 using the Nde I and  Fse I sites and recombined into pAG305GPD (Alberti et al.,2007), resulting in pUDI35.  2.3. Strain construction Transformations of   S. cerevisiae  were carried out using theprotocols of  (Gietz and Woods, 2002). Gene deletions and subse- quent marker removal were performed using the loxP-marker-loxP/Cre recombinase system (Gueldener et al., 2002), using pUG6 and pUG- hphNT1  as templates for PCR amplification of the knockoutcassettes. Selection of mutants was performed on either 200mgl  1 G418 or 200 mgl  1 hygromycin B. Primers used for amplificationsof knockout cassettes are listed in Table 3. During the constructionof the  S. cerevisiae  strain deleted for the native maltose metabolism,the maltose transporter ( MALx1 ) and maltase ( MALx2 ) within thesame locus were deleted with one knockout cassette, starting in thedouble auxotrophic CEN.PK102-3A strain. The order of knockoutwas  MAL11  and  MAL12 ,  MAL31  and  MAL32 ,  MAL21  and  MAL22 ,followed by  MPH2  and  MPH3 . Knockouts were confirmed by south-ern blotting on separated chromosomes using a  MAL32  probe,amplified from CEN.PK113-7D genomic DNA with primers MAL32  Table 2 Plasmids used in this study. Plasmid Characteristic Reference pENTR/D-TOPO Gateway entry clone Invitrogen, USApUG6 PCR template for loxP- KanMX4 -loxP cassette Gueldener et al. (2002)pFA6a- hphNT1  Plasmid with  hphNT1  marker Janke et al. (2004)pSH47 Centromeric plasmid,  URA3 , P GAL1 -Cre- T CYC1  Gueldener et al. (2002)pRW231 2 m m ori,  URA3 , P TPI1 -  xylA -T CYC1  P TDH3 - araA -T  ADH1  P HXT7  - araD -T PGI1  Wisselink et al. (2007)p426GPD 2 m m ori,  URA3 , P TDH3 -T CYC1  Mumberg et al. (1995)pvv214 2 m m ori,  URA3 , P PGK1 - ccdB -T CYC1  Van Mullem et al. (2003)pAG426GPD 2 m m ori,  URA3 , P TDH3 - ccdB -T CYC1  Alberti et al. (2007)pAG305GPD Integration plasmid,  LEU2 , P TDH3 - ccdb -T CYC1  Alberti et al. (2007)pRS405 Integration plasmid,  LEU2  Sikorski and Hieter (1989)pUG- hphNT1  PCR template for loxP- hphNT1 -loxP cassette This studypUD37 Gateway entry clone,  mapA  This studypUD63 2 m m ori,  URA3 , P TDH3 -T  ADH1  This studypUD65 Gateway entry clone, multiple cloning site This studypUD88 Gateway entry clone,  pgmB  This studypUDE31 2 m m ori,  URA3 , P PGK1 - mapA -T CYC1  This studypUDE44 2 m m ori,  URA3 , P TDH3 - MAL12 -T  ADH1  This studypUDE45 2 m m ori,  URA3 , P TDH3 - mapA -T  ADH1  This studypUDE60 2 m m ori,  URA3 , P TDH3 -  pgmB -T CYC1  This studypUDE63 2 m m ori,  URA3 , P TDH3 -  pgmB -T  ADH11  This studypUDE82 2 m m ori,  URA3,  P TDH3 -  pgmB -T  ADH1  P PGK1 - mapA -T CYC1  This studypUDI35 Integration plasmid,  LEU2 , P TDH3 - MAL11 -T CYC1  This study  Table 3 Primers used in this study. Name  Sequence (5 0 - 3 0 ) Primers for knockouts MAL1 KO Fw CTTAACATTTATCAGCTGCATTTAATTCTCGCTGTTTTATGCTTGAGGACTGACTCCAATGGCAGGTAATGTACGCAGCTGAAGCTTCGTACGCMAL1 KO Rv GCACTAATTTTATTTGACGAGGTAGATTCTACCTTCCCATGGTTTCAAAACTCTGGCATAGGCCACTAGTGGATCTGMAL3 KO Fw A CATTTGTTCACAACAGATGAGGTGTTTCGCCCTTCATCTACCACAGAAGTTTCCACAATTAGACCTCCCTACAGTGCAGCTGAAGCTTCGTACGCMAL3 KO Rv A CTCTGGAGGAAACGTCAGTATCATCATAATTTCCAAGAATAAAAGATAAAGAAGCGCATAGGCCACTAGTGGATCTGMAL3 KO Fw B CCAAATCTTCCTTCGGATCTTTAACATTAATTTCTGCAGCTGCTGCTTTGAACAGAAGCCCATGTGGTGACAGCTGAAGCTTCGTACGCMAL3 KO Rv B ATAAAAGATAAAGAAGCACCTTCTCTTGGGAGGCTGAATTCAATTTCTTCGACCCTTCTGAGACGGTAGTGGCATAGGCCACTAGTGGATCTGMPH2/3 KO Fw A GTTTGTAATTCTTAGAGGCCTGTTCTTGGAATTATTATGCAAAGATTACCATCACCAGCTGAAGCTTCGTACGCMPH2/3 KO Rv A ACAAATCTACGTGTATTATACTCCGTAACATGTAGAGTAAATACCATAGTTACCTGCATAGGCCACTAGTGGATCTGMPH2/3 KO Fw B AAATCTAACACAAATAGAGATACAGGTCTTGTAAAGCCATTACCTAGCTATGAAACAGCTGAAGCTTCGTACGCMPH2/3 KO Rv B TACCTGTGTCGATAAATGTTCATTAGCTCATAAGTGATGGGATACATTGCTATTCGCATAGGCCACTAGTGGATCTG Primers for DNA probes MAL32 probe Fw GCAGAAGGGCAATCTTTGAAAGTGMAL32 probe Rv AGCAGCAAACAGCGTCTTGTCMPH2/3 probe Fw GGATACAAGAGCGCCAAGCGATAGMPH2/3 probe Rv GTCTTTCCGGCAGTTTCTGGTAGG Primers for cloning  mapA Fw CACCGTTTAAACACTAGTGGAGAATATCATGAAGCGAATTTTTGAAGmapA Rv GGCCGTTTAAACAAGCTTCATCTTAGGCCTCCAAAGTTAGCADH1t Fw CACTCGAGGGTTGGCATGACTAAGGCATTCADH1t Rv CTGGTACCCGATACCGGTGATGGCGATCGCGTGTGGAAGAACGATTACAACAGGMAL12 Fw CACCACTAGTCATAAATGACTATTTCTGATCATCCAGAAACAGMAL12 Rv GCAAGCTTGCCGGCACTAATTTTATTTGACGAGMcs Fw CACCCATATGATATGTTTAAACGCATGCTGAGCATCGCCATGGMcs Rv GCGGCCGGCCGCATCACCGGTGCGATCCATGGCGATGCTCAGCMAL11 Fw CACCAGTCATATGGTATAATATGAAAAATATCATTTCATTGGTAAGCAAGAAGMAL11 Rv GCATTAGGCCGGCCCTTAACATTTATCAGCTGCATTTAATTCTCGPGK1p Fw GCACCGGTACGTCGTACGCGACTCTTTTCTTCTAACCAAGGGCYC1t Rv CAGCGATCGCCTTCAAAGCTTGCAAATTAAAGCC S. de Kok et al. / Metabolic Engineering 13 (2011) 518–526  520  probe Fw and MAL32 probe Rv, or a  MPH2/3  probe, amplified fromCEN.PK113-7D genomic DNA with primers MPH2/3 probe Fw andMPH2/3 probe Rv. The resulting maltose transporter and maltaseknockoutstrainweredesignatedIMK289.PlasmidspRS405(Sikorskiand Hieter, 1989) and pUDI35 were linearized with  BstE  II andtransformed into IMK289, resulting in IMX020 and IMX032, respec-tively. IMX020 was transformed with pUD63 and pUDE44, resultingin IMZ133 and IMZ135, respectively. IMX032 was transformed withpUDE44, pUDE45, pUDE82 and pUD63, resulting in IMZ199,IMZ203, IMZ226 and IMZ229, respectively (Table 1).  2.4. Molecular biology techniques PCR amplification was performed using Phusion Hot Start HighFidelity Polymerase (Finnzymes, Espoo, Finland) according tomanufacturer’s instructions in a Biometra TGradient Thermocy-cler (Biometra, Gottingen, Germany). DNA fragments were sepa-rated on a 1% (w/v) agarose (Sigma, St. Louis USA) gel in 1  TAE(40 mM Tris-acetate pH 8.0 and 1 mM EDTA). Isolation of frag-ments from gel was performed with the Zymoclean Gel DNARecovery kit (Zymo Research, Orange, USA). Restriction endonu-cleases (New England Biolabs, Beverly, USA and Promega,Madison, USA) and DNA ligases (Roche, Basel, Switzerland) wereused according to manufacturer’s instructions. Transformationand amplification of plasmids was performed in  E. coli  One ShotTOP10 Competent cells (Invitrogen, Carlsbad, USA) according tomanufacturer’s instructions. Plasmids were isolated from  E. coli with the Sigma GenElute Plasmid Miniprep Kit (Sigma, St. Louis,USA). DNA constructs were routinely sequenced by Baseclear BV(Baseclear, Leiden, The Netherlands). Chromosomes were sepa-rated using the CHEF yeast genomic DNA Plug Kit (170–3593,Bio-Rad, Richmond, USA) according to manufacturer’s instruc-tions and subsequently transferred onto Hybond-N þ nylon mem-branes (RPN303, Amersham Biosciences, Piscataway, USA).Southern blotting, signal generation and signal detection wereperformed using the Gene Images AlkPhos Kit, CPD Star detectionreagent and Hyperfilm ECL (RPN 3680, RPN3682 and 28-9068,Amersham Biosciences, Piscataway, USA).  2.5. Cultivation Characterization in shake flasks was performed in a mediumcontaining per litre 20 g maltose (Molekula, Dorset, UK), 6.6 gK 2 SO 4 , 3 g KH 2 PO 4 , 2.3 g urea, 0.5 g MgSO 4 d 7H 2 O, trace elementsand vitamins as described previously (Verduyn et al., 1990). pH of  the medium was set to 5.0 with 2 M KOH and 2 M H 2 SO 4  prior tofilter sterilization. Cultures were prepared by inoculating 100 mlmedium in a 500 ml shake flask with 1 ml frozen stock cultureand incubated in an Innova incubator shaker (New BrunswickScientific, Edison, USA) at 200 rpm and 30  1 C. Optical density at660 nm was measured in regular time intervals with a Libra S11spectrophotometer (Biochrom, Cambridge, UK). Anaerobic batchand chemostat fermentations were carried out at 30  1 C in 2 llaboratory bioreactors (Applikon, Schiedam, The Netherlands)with a working volume of 1 l. A synthetic medium with 25 g l  1 maltose as the sole carbon and energy source was used for allchemostats (Verduyn et al., 1990). The medium was supplemen- ted with the anaerobic growth factors ergosterol (10 mg l  1 ) andTween 80 (420 mg l  1 ) dissolved in ethanol. Antifoam Emulsion C(Sigma, St. Louis, USA) was autoclaved separately (120  1 C) as a20% (w/v) solution and added to a final concentration of 0.2 g l  1 .The culture pH was maintained at 5.0 by the automatic additionof 2 M KOH. Cultures were stirred at 800 rpm and sparged with500 ml min  1 nitrogen ( o 10 ppm oxygen). To minimize diffusionof oxygen, the bioreactors were equipped with Norprene tubingand Viton O-rings and the medium vessels were flushed withnitrogen gas. The anaerobic growth rate was obtained by plottingthe CO 2  production rate, which is stoichiometrically coupled togrowth, on a logarithmic scale and calculating the slope of thelinear part of the graph.  2.6. Analytical methods Chemostat cultures were assumed to be in steady state when,after at least five volume changes, the culture dry weight andcarbon dioxide production rates changed by less than 4% over2 volume changes. Steady state samples were taken between 10and 15 volume changes after inoculation to avoid possibleevolutionary adaptation during long-term cultivation. Culturedry weights were determined via filtration of 20 ml samples overdry preweighed nitrocellulose filters (Gelman Laboratory, AnnArbor, USA) with a pore size of 0.45 m m. After removal of themedium, the filters were washed twice with demineralized water,dried in a microwave oven for 20 min at 350 W and weighed.Culture supernatants were obtained after centrifugation of chemostat broth. Supernatants and media were analyzed viaHPLC using an Aminex HPX-87H ion exchange column operatedat 60  1 C with 5 mM H 2 SO 4  as mobile phase at a flow rate of 0.6 ml min  1 . Samples for residual maltose were taken asdescribed previously (Canelas et al., 2011). Ethanol concentra- tions were corrected for evaporation as described previously(Guadalupe Medina et al., 2010). Off-gas was first cooled in acondenser (2  1 C) and dried with a Perma Pure Dryer (Permapure,Toms River, USA). CO 2  concentrations in the off-gas were mea-sured with a NGA 20000 Rosemount gas analyzer (RosemountAnalytical, Orrville, USA). Culture viability was assayed usingthe  Funga light CFDA, AM/propidium iodide yeast vitality kit(Invitrogen, Carlsbad, USA) on a Cell Lab Quanta TM SC MPL flowcytometer (Beckman Coulter, Woerden, The Netherlands). Viabi-lity (metabolic activity) was calculated as the number of CFDA þ and PI  cells divided by the total number of cells.  2.7. Enzyme activity measurements For preparation of cell extracts, culture samples (around62.5 mg cell dry weight) were harvested from exponentiallygrowing shake flask cultures on 20 g l  1 maltose. The sampleswere centrifuged (4  1 C, 10 min at 5000 g), washed twice withfreeze buffer (10 mM potassium phosphate, pH 7.5 and 2 mMEDTA), washed once with disruption buffer (100 mM potassiumphosphate, pH 7.5, 2 mM MgCl 2  and 1 mM DTT) and resuspendedin 2 ml disruption buffer. Cells were disrupted in a FastPrep 120(Qbiogene, Carlsbad, USA) by shaking 1 ml aliquots with 0.75 gglass beads (425–600 m m, Sigma-Aldrich, Zwijndrecht, The Neth-erlands) four times 20 s at power level 6, with 1 min cooling onice in between. Unbroken cells and cell debris were removed bycentrifugation (4  1 C, 20 min at 36 000 g), and the supernatant wasused as the cell extract for enzyme assays. Protein levels in cellextracts were determined using a Lowry assay (Lowry et al.,1951). For IMZ229, cells were harvested from shake flask cultureswith 20 g l  1 ethanol and 20 g l  1 maltose. Cell extracts for b -phosphoglucomutase activity measurements were preparedfrom fresh, non-frozen cultures to avoid the observed instabilityof the enzyme upon freezing. Maltase activity was measured at30  1 C by monitoring the reduction of NADP þ at 340 nm in a 1 mlreaction mixture containing 100 mM maltose, 50 mM imidazole-HCl buffer (pH 6.6), 1 mM NADP þ , 12.5 mM MgCl 2 , 1 mM ATP,3.5 units hexokinase and 3.5 units glucose-6-phosphate dehydro-genase. The reaction was started by the addition of 1–100 m l cellextract. Maltose phosphorylases and  b -phosphoglucomutaseactivities were measured at 30 1 C by monitoring the reduction of NADP þ at 340nm in a 1ml reaction mixture containing 100 mM S. de Kok et al. / Metabolic Engineering 13 (2011) 518–526   521  MES buffer (pH 7.0), 25mM potassium phosphate buffer (pH 7.0),5mM MgCl 2 , 1mM NADP þ , 0.01mM glucose-1,6-biphosphate,3.5 units glucose-6-phosphate dehydrogenase and 1–100 m l cellextract. The reaction was started by the addition of 100 mMmaltose. For determination of the individual maltose phosphor-ylases or  b -phosphoglucomutase activity, 3.5 units of commercial b -phosphoglucomutase (Sigma, St. Louis, USA) or maltose phos-phorylases (Sigma, St. Louis, USA) were added to the reactionmixture, respectively. An extinction coefficient of 6.3 mM  1 wasassumed for NADPH. Reaction rates were proportional (within25%) to the amounts of cell extract added. 3. Results  3.1. Deletion of native maltose transporters and maltases To enable a systematic analysis of the bioenergetic impact of genetic modifications in maltose metabolism, we first constructeda strain platform in which the native maltose metabolism genesof   S. cerevisiae  (consisting of the  MAL  loci,  MPH2  and  MPH3 ) weredeleted. The number of   MAL  loci is strain dependent (Naumovet al., 1994). Southern blot analysis on separated chromosomesusing a  MAL32  probe showed that CEN.PK102-3A contained three MAL  loci ( MAL1x ,  MAL2x  and  MAL3x , respectively, on chromo-somes VII, III and II; data not shown). The maltose transporters( MALx1 ) and maltases ( MALx2 ) are adjacent genes, which enableddeletion of both genes in one transformation using a singleknockout cassette. Using the Cre/loxP system (Gueldener et al.,2002) for removal of dominant marker genes, the maltosetransporter ( MALx1 ) and maltase ( MALx2 ) genes of the three MAL  loci were deleted, followed by  MPH2  and  MPH3 . Southernblot analysis with  MAL32  and  MPH2/3  probes confirmed that allknown maltases and maltose transporters were successfullydeleted (data not shown). The resulting strain ( mal11-mal12 D mal21-mal22 D  mal31-32 D  mph2 D  mph3 D ) strain was namedIMK289.  3.2. Characterization of the maltose knockout (mal D  mph D ) strain To determine whether the  S. cerevisiae  genome encodes addi-tional maltose transporters and maltases, growth tests wereemployed with strains derived from IMK289 ( mal D  mph D ). Toanalyze the residual activity of maltose transport and hydrolysis,the maltase-encoding  MAL12  gene and the maltose transportergene  MAL11  (also known as  AGT1 ) were re-introduced in IMK289,either alone or in combination. The resulting prototrophic strainswere tested in shake flasks with maltose as the sole carbon source.Under these conditions, CEN.PK113-7D (Mal þ , m ¼ 0.34 7 0.00h  1 )and IMZ199 ( mal D  mph D  MAL11 MAL12 ,  m ¼ 0.24 7 0.00h  1 ) grewfast. IMZ133 ( mal D  mph D ) and IMZ135 ( mal D  mph D  MAL12 ) didnot show any growth during 7 days ( m o 0.001 h  1 ), but IMZ229( mal D  mph D  MAL11 ) grew slowly ( m ¼ 0.007 7 0.001 h  1 ) (Table 4).These results showed that all maltose transport activity wasremoved and that a low residual maltose-hydrolyzing activitywas still present in the maltose knockout strain.To quantify the remaining maltase activity in the maltoseknockout strain background, cell extracts were prepared fromIMZ229 ( mal D  mph D  MAL11 ). Because IMZ229 only grew veryslowly on maltose ( m ¼ 0.007 7 0.001 h  1 ), a different carbonsource was chosen that supports faster growth. Considerationsin choosing another carbon source were that maltose metabolismis repressed by glucose and induced by maltose (Carlson, 1987), even though the CEN.PK strains contain a constitutive  MAL activator gene  MAL2-8 C  (Gibson et al., 1997). Therefore, cellsextracts were prepared from an IMZ229 culture grown on thenon-repressing carbon source ethanol (2% w/v) in the presence of the inducer maltose (2% w/v). This revealed a maltase activity of 0.03 7 0.00 m mol min  1 mg protein  1 for IMZ229 ( mal D  mph D MAL11 ), while CEN.PK113-7D (Mal þ ) and IMZ199 ( mal D  mph D MAL11 MAL12 ) grown on maltose showed a maltase activity of 3.6 7 0.1 and 5.9 7 1.2 m mol min  1 mg protein  1 , respectively.Thus, the residual maltose-hydrolyzing activity of the maltose-knockout strain was less than 1% of that of the reference strain.  3.3. Successful expression of Lactobacillus sanfransiscensis maltose phosphorylase After removing virtually all maltase activity, the next goal was toexpress a functional maltose phosphorylase in a strain expressing themaltose transporter  MAL11 . This modification should increase free-energy conservation with 1 ATP per maltose compared to normalmaltose hydrolysis in  S. cerevisiae . In literature, only two maltosephosphorylase genes have been cloned and characterized (Ehrmannand Vogel, 1998; Inoue et al., 2002). The  mapA  gene from L. sanfranciscensis  was chosen, because of a better  S. cerevisiae  codonadaptation index (calculated according to (Grote et al., 2005)). The L. sanfranciscensis mapA  gene was cloned behind the strong constitu-tive  TDH3  promoter on the multicopy plasmid pUDE45. The resulting mapA  expressing strain IMZ203 displayed a high maltose phosphor-ylase activity (11.3 7 1.1 m molmin  1 mgprotein  1 ), while the mal-tosephosphorylaseactivityofthereferencestrainCEN.PK113-7Dwasbelow the detection limit ( o 0.005 m molmin  1 mgprotein  1 ).  Table 4 Aerobic and anaerobic growth rates on maltose and maltase, maltose phosphorylase and  b -phosphoglucomutase activities of   Saccharomyces cerevisiae  strains carryingdifferent combinations of maltase, maltose phosphorylase and  b -phosphoglucomutase. Averages and mean deviations were obtained from duplicate experiments.Strain Relevant genotype Growth rate (h  1 ) Enzyme activity ( m mol mg protein  1 min  1 )Aerobic a Anaerobic b Maltase Maltose phosphorylase  b -phosphoglucomutaseCEN.PK113-7D  MAL1 MAL2 MAL3 MAL4  0.34 7 0.00 0.28 7 0.01 3.6 7 0.1  o 0.005  o 0.005IMZ133  mal D  mph D  o 0.001 N.D. N.D. N.D. N.D.IMZ135  mal D  mph D þ MAL12  o 0.001 N.D. N.D. N.D. N.D.IMZ229  mal D  mph D þ MAL11  0.007 7 0.001 N.D. 0.03 7 0.00 N.D. N.D.IMZ199  mal D  mph D þ MAL11 þ MAL12  0.24 7 0.00 0.12 7 0.00 5.9 7 1.2 N.D. N.D.IMZ203  mal D  mph D þ MAL11 þ mapA  0.09 7 0.03 0.03 7 0.01 c N.D. 11.3 7 1.1  o 0.005IMZ226  mal D  mph D þ MAL11 þ mapA þ  pgmB  0.21 7 0.01 0.10 7 0.00 N.D. 0.83 7 0.03 0.58 7 0.04N.D. ¼ not determined. a Aerobic growth rates were based on optical density measurements in shake flasks at pH 5.0 and with 20 g l  1 maltose. A growth rate  o 0.001 h  1 means that nogrowth was detected during 7 days. b Anaerobic growth rates were based on CO 2  production in batch fermentations at pH 5.0 and 25 g l  1 maltose. c No consistent exponential growth was observed for this culture. Average of 40 h of incubation. S. de Kok et al. / Metabolic Engineering 13 (2011) 518–526  522
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks