WO2004085627A1 - New saccharomyces cerevisiae strains utilizing xylose - Google Patents

Abstract

A mutant strain of Saccharomyces cerevisiae which utilises xylose. The strain overexpresses xylose reductase, xylitol dehydrogenase and xylokinase. In addition, one or more genes encoding NADPH-producing enzymes and one or more genes encoding NADPH-consuming enzymes are up-regulated. A method for selecting an improved S. cerevisiae strain, which is cultivated with xylose as the only carbon source under aerobic conditions followed by microaerobic and anaerobic conditions. Each cultivation condition stretches over many generations and is followed by a selection for growth on xylose. Random mutagenesis is carried out during or after the aerobic phase.

Description

TITLE

NEW SACCHAROMYCES cerevisiae STRAINS UTILIZING XYLOSE

DESCRIPTION Technical field

The present invention relates to novel Saccharomyces cerevisiae strains utilizing xylose for production of ethanol, as well as method for selecting a xylose-utilizing Saccharomyces cerevisiae strain.

Background of the invention

Over the last decade, metabolic engineering became a standard for strain improvement and has been very successful when simple cellular traits were targeted (Ostergaard et al . 2000; Stafford and Stephanopoulos 2001). Although the genomics age with the associated genome-wide analytical technologies gives further impetus to rational approaches, metabolic engineering of more complex or not fully understood cellular systems remains a challenge. Akin to random, combinatorial approaches such as directed evolution in contemporary protein engineering evolutionary approaches are becoming increasingly important to augment metabolic engineering of complex phenotypes. In certain cases, however, even seemingly simple metabolic systems resist straightforward rational engineering.

One example of a seemingly simple trait is expanding the substrate range of Saccharomyces cerevisiae for the utilization of pentoses for ethanol formation. The commercial interest in pentose utilization, in particular xylose, is related to the prevalence of pentoses in abundant plant material, for instance as the major structural unit in hemicelluloses. While metabolic engineering has successfully endowed S. cerevisiae with the ability to utilize the pentoses xylose (Kόtter and Ciriacy 1993; Walfridsson et al. 1995; Ho et al. 1998; Eliasson et al. 2000) and recently also arabinose, it has not yet succeeded in developing strains that convert pentose at a high yield and a high specific rate to ethanol (Ho et al. 1999; Hahn-Hagerdal et al. 2001).

An additional puzzling fact is the inability of xylose to support anaerobic growth in both natural and recombinant xylose-utilizing yeasts. Since many bacteria can grow anaerobically on xylose, the reason for this inability is not really understood at present, but has been ascribed to a general restriction of eucaryotic xylose metabolism to respirative conditions. This argument was based on the fundamental difference between eucaryotic and prokaryotic xylose catabolism because bacteria convert xylose directly to xylulose using xylose isomerase, whereas eukaryotes rely on two consecutive redox reactions that are catalyzed by the NADPH-dependent xylose reductase (X ) and the NADH-dependent xylitol dehydrogenase (XDH) with xylitol as the pathway intermediate. By providing NADPH through the oxidative pentose phosphate pathway, which operates actively in S. cerevisiae and Pichia stipitis (Gombert et al. 2001; Fiaux et al. 2003), and by respiring NADH, eukaryotes can efficiently drive these coupled redox reactions under aerobic conditions, but possibly not under anaerobic conditions. While this could potentially explain the inability of many yeasts to grow anaerobically on xylose, it does not suffice as an explanation for those xylose-utilizing S. cerevisiae strains that functionally overexpress xylose isomerase. Hence, it appears that at least one additional component in our understanding of xylose metabolism is missing.

Such understanding cannot be obtained from the available databases and the published knowledge, hence it is here attempted to evolve strains that are capable of anaerobic growth on xylose and to investigate the correlated molecular and functional mechanisms. Since anaerobic xylose-utilizing eukaryotes did apparently not evolve naturally, selection should be initiated with the best xylose-utilizing S. cerevisiae strains available. The presently best strains overexpress the XR and XDH genes from P. stipitis (Kόtter and Ciriacy 1993) in combination with the endogenous xylulokinase (XK) (Ho et al. 1998; Eliasson et al. 2000; Toivari et al. 2001). The S. cerevisiae strain TMB3001 that overexpresses the three genes of the xylose-utilization pathway from a chromosomal integration to initiate various long-term evolution experiments (Eliasson et al. 2000) was used.

Summary of the present invention

The present invention relates to novel xylose-utilizing Saccharomyces cerevisiae strains, in particular to those deposited at DSMZ under the accession numbers DSM 15519 and DSM 15520 on the 18th of March, 2003. DSM 15519 relates to the strain coded TMB3001C1 below, and DSM 15510 relates to the strain coded TMB3001C5 below.

In general the invention relates to new xylose-utilizing Saccharomyces cerevisiae mutant strain overexpressing the xylose-utilizing pathway by overexpressing xylose reductase (XR), xylitol dehydrogenase (XDH), and xylokinase (XK), wherein one or more of genes encoding NADPH producing enzymes are up-regulated and one or more of the genes encoding NADH consuming enzymes are up-regulated.

In one preferred embodiment, the genes XKS1, TALI, TKL1, ZWF1, SOLI, SOL3, GNDl, PGI1, PFK1, PFK2, FBA1, TPI1, GPD1, GPD2, TDH1, TDH2, GPM1, GPM2, EN02, PYK1, PYK2, PDC1, ADH 1, ADH3, ADH4, ADH5, PYC2, GAL2 and GPP1, SOL2, TDH3, PGK1, ENOl, ADH6, MAE1, GPP2, HXT6, GAL2, YEL041W, and GCY1 are up-regulated.

In another preferred embodiment, the genes GAL2, PYK2, PUT4, GPD2, and ADH4 are over- expressed.

In another preferred embodiment, regulators of glycolysis and central carbon metabolism : GCR1, RAP1, REB1, PDC2, HAP1, HAP5, and SNF1 are over-expressed.

In another preferred embodiment, a regulator of galactose metabolism: GAL3 is over- expressed.

In another preferred embodiment, regulators of osmotic stress response (HOG-pathway) : SSK22, RIM15, ASK10, SKN7, PBS2, MSN4, MSN1, and PTP2 are over-expressed.

In another preferred embodiment, regulators with other functions: TEA1, YAP5, ACAl, CLN3, RIOl, and SNF2 are over-expressed.

In another preferred embodiment, the genes YATl, YCR060W, YEL057C, YIL176C, MCH5, and ATM1 are expressed to improve the capability of anaerobic growth on xylose.

In further preferred embodiment, the genes encoding for xylose reductase (XR), and xylitol dehydrogenase (XDH) are over expressed to a higher degree than in the corresponding non-mutated strain TMB3001.

Another aspect of the invention relates to a method for selecting an improved xylose- utilizing Saccharomyces cerevisiae strain, which strain overexpresses the xylose-utilizring pathway by overexpressing xylose reductase (XR), xylitol dehydrogenase (XDH), and xylokinase (XK), which method is characterized in that a starting Saccharomyces cerevisiae strain is cultured under aerobic conditions in minimal medium containing xylose as sole carbon source, is selected with regard to growth on xylose, is continued cultured under microaerobic conditions in minimal medium containing xylose as sole carbon source, is selected with regard to growth on xylose, and is finally cultured under anaerobic conditions in minimal medium containing xylose as sole carbon source to produce a xylose-utilising Saccharomyces cerevisiae strain.

In one preferred embodiment, random mutagenesis using ethyl methane sulfonate is carried out on the starting Saccharomyces cerevisiae strain during the aerobic culture phase.

In one preferred embodiment, random mutagenesis using ethyl methane sulfonate is carried out on the Saccharomyces cerevisiae strain selected after the aerobic culture phase.

In one preferred embodiment, selection is done using continuous culture.

In one preferred embodiment, selection is done using chemostat selection.

In one preferred embodiment, culturing in a continuous culture takes place at constant dilution rate.

In one preferred embodiment, the dilution rate is 0.05 h^"1.

In one preferred embodiment, the first selection is made after at least 50 generations culturing in minimal medium using xylose as only carbon source, the second selection is made after further at least 100 generations, and the third selection is made after further at least 120 generations.

In one preferred embodiment, the culturing under aerobic conditions takes place under at least 90 generations, preferably 130 generations.

In one preferred embodiment, culturing under microaerobic conditions takes place under further at least 110 generations, preferably at least 140 generations. In one preferred embodiment, culturing under anaerobic conditions takes place under further at least 150 generations, preferably 190 generations.

In one preferred embodiment, single clones from anaerobic growth on xylose plates are selected, and plated on anaerobic minimal medium containing xylose as the sole carbon source, and plates are incubated in sealed environment, to provide for an anaerobic atmosphere.

In one preferred embodiment, a pre-selection step is carried out prior to the aerobic culture on sole xylose medium comprising an aerobic growth on a minimal medium containing glucose and xylose as carbon sources in a molar ratio of about 50: 50.

Detailed description of the present invention

The invention will be described more in detail in the following with reference to the below given technical disclosure.

Materials and methods

Strains and media

All evolution experiments were inoculated with the recombinant S. cerevisiae strain TMB3001 (CEN.PK 113-7A {MATa, his3-Δl, MAL2-8C, SUC2) /?/s3;.ΥIpXR/XDH/XK), which contains the entire xylose-utilization pathway (Eliasson et al. 2000). Overexpression of XR is controlled by the alcohol dehydrogenase promoter and terminator, whereas XDH and XK are both under the control of phosphoglycerate kinase promoters and terminators. The extensions Cl to C5 refer to clones that were isolated after 460 generations of selection. TMB3001C1 and TMB3001C5 are representatives of the phenotypic classes II and I, respectively.

For physiological analysis and evolution experiments, yeast cultures were grown at 30°C in minimal medium containing per litre: 5 g (NH₄)₂S0₄, 3 g KH₂P0₄, 0.5 g MgS0₄»7H₂0, 15 mg EDTA, 4.5 mg ZnS0₄»7H₂0, 0.3 mg CoCI₂»6H₂0, 1 mg MnCI₂«4H₂0, 0.3 mg

CuS0₄«4H₂0, 4.5 mg CaCI₂.2H₂0, 3 mg FeS0₄-7H₂0, 0.4 mg Na₂Mo0₄»2H₂0, 1 mg H₃B0₃, 0.1 mg KI, 0.05 mg biotin, 1 mg Ca pantothenate, 1 mg nicotinic acid, 25 mg inositol , 1 mg thiamine HCI, 1 mg pyridoxine HCI, and 0.2 mg para-aminobenzoic acid (pH 5.0) . In chemostat cultures, 0.1 g I^"1 polypropylene glycol P 2000 was added to prevent foam formation. The medium was supplemented with ergosterol (Fluka) and Tween 80 (Sigma) for anaerobic cultivation. Both components were dissolved in boiling 99.8% (v/v) ethanol and were added to the medium at a final concentration of 0.01 g I^"1 and 0.42 g I^"1, respectively. Solid media were prepared by adding 1.5% (w/v) technical Agar (Becton Dickinson). For anaerobic growth on xylose plates, population aliquots were washed twice with PBS (8 g I^"1 NaCl, 0.2 g I^"1 KCI, 1.44 g I^"1 Na₂HP0₄, 0.24 g T¹ KH₂P0₄, pH 7.0) and plated on anaerobic minimal medium containing 20 g I^"1 xylose as the sole carbon source. Plates were incubated at 30°C in sealed jars, using the GasPack Plus system (Becton Dickinson) to provide an anaerobic atmosphere, which was verified by indicator strips (Becton Dickinson).

Cloning and yeast transformation procedures

Standard molecular biology techniques were used to clone the phosphoketolase (xfp), the phosphotransacetylase (pta) and the acetaldehyde dehydrogenase (Ehadh2) genes under the strong truncated HXT7 promotor in the yeast multicopy plasmids p426HXT7, p424HXT7 and p425HXT7, respectively. The Bifidobacterium lactis xfp gene was sub-cloned by ligation of the 2.6 kbp DNA fragment, resulting from the EcoRI-Hindlll digestion of the pFPKδ plasmid, after gel extraction (QUIAEX II, QUIAGEN, Basel, Switzerland). Analogously, the Entamoeba histolytica Ehadh2 containing 3 kbp cDNA fragment was sub-cloned from the BamHI-Xbal digestion of pET3a-ehadh2 into the BamHI-Spel digested p425HXT7 plasmid. The Bacillus subtiiis pta gene was amplified with taq DNA polymerase (Promega, Madison, WI) by PCR (Cycle: lx 2 min. at 95°C; 30x (lmin at 95°C, 0.5 min at 58°C, 1.25 min at 74°C); 5 min at 74°C) from B. subtiiis genomic DNA using the following primers: Fwd: 5'-cgg gat cca tgg cag att tat ttt caa cag tg-3'; Rev: 5'-cca teg atg teg aga get gcc att gtc tcc-3'. This fragment was ligated using the BamHI-Clal restriction sites of p424HXT7.

Plasmids were transformed in S. cerevisiae by the lithium acetate method using the S.c. EasyComp transformation kit (Invitrogen, Carlsbad, CA).

Long-term selection cultures

Chemostat selections were performed in a Sixfors 6-minireactors system (Infors, Botmingen, Switzerland) at a dilution rate (D) of 0.05 h^"1 and mixing at 300 rpm. A constant working volume of 300 ml was maintained by continuously removing excess culture broth through a needle that was fixed at a predetermined height. The culture pH was maintained at 5.0 ± 0.3 by supplementing the minimal medium with 50 mM potassium hydrogen phthalate (Fluka). Aerobic conditions were installed by aeration at a rate of 0.3 I min^"1. Microaerobic conditions were installed by stepwise reduction of aeration until no measurable flow was seen in the reactor effluent gas. Microaerobic conditions were defined as < lml/min per 300 ml vessel containing medium and biomass, i.e. a severely oxygen depleted condition. Anaerobic conditions were established by slight sparging (< 1 ml min^"1) with technical N₂ (< 200 ppm 0₂; independently quantified with a Prima 600 mass spectrometer (Fisons Instruments, Uxbridge, England)). It should be noted that due to the contaminating 0₂, these conditions are not strictly anaerobic. To ensure robust long-term operation of up to 4 months, marprene tubing (Ismatech, Glattbrugg, Switzerland) was used with external peristaltic pumps for feeding and harvesting. Contamination controls were done in two weeks intervals by plating culture aliquots on YPD medium (10 g I^"1 yeast extract, 20 g I^"1 peptone, and 20 g I^"1 glucose) plates and by microscopic analysis.

Selection in serial, strict anaerobic batch cultures was done in Hungate tubes, which are 17 ml Pyrex glass tubes that are sealed with butyl rubber septa and plastic screw caps (Bellco Glass Inc., Vineland, NJ). Cultures were grown in minimal medium, containing 10 g I^"1 xylose as the sole carbon source. New cultures were inoculated when the growth rate declined, which occurred typically after about 1 week.

Growth conditions.

Aerobic cultures were grown in 500 ml baffled shake flasks with 50 ml minimal medium at 300 rpm on a rotary shaker and 30°C. To adapt TMB3001 to aerobic growth on sole xylose, it was first grown on YPX medium ( 10 g I^"1 yeast extract, 20 g I^"1 peptone, and 20 g I^'1 D- xylose), then on YNB xylose medium (6.7 g I^"1 Yeast Nitrogen Base and 20 g I^"1 D-xylose), and finally once in minimal medium with sole xylose prior to inoculation.

Fermentation performance was evaluated in anaerobic batch cultures containing 50 g I^"1 each of glucose and xylose. The concentrations of all other minimal medium components except KH₂P0₄ were doubled. To avoid major drops in pH, 100 mM citric acid buffer (pH 5.5) was added, which maintained the pH above 4.7 in all cases. Cultures were grown in 175 ml serum bottles, filled with 150 ml medium and stirred magnetically at 100 rpm and 30°C. Anaerobic (but not strictly anaerobic) conditions were maintained by slight continuous sparging (1-2 bubbles sec^"1) with technical N₂ (0₂ < 200 ppm) (PanGas, Dagmersellen, Switzerland). Inocula were prepared by growing frozen stock cultures first on YPD medium and finally in minimal medium with 20 g I^"1 glucose. Strictly anaerobic growth experiments on sole xylose were done in Hungate tubes or serum bottles sealed with butyl-rubber septa by sparging the basic salt solution of the minimal medium with pure N₂ (0₂ < 5 ppm) (PanGas) for 15 min. After autoclaving, the remaining filter-sterilized, N₂-sparged medium components and 10 g I^"1 xylose were added. To ensure strict anaerobiosis, 0.25 g I^"1 Na₂S or 0.5 g I^"1 L-cysteine were added in selected cases as reducing agents after incubation at 60°C for 5 min. Withdrawal of culture aliquots was done under purging with pure N₂. To verify strict anaerobiosis, the redox indicator resazurin was added to the medium at a final concentration of 0.0001% (w/v) before sparging with pure N₂. Strict anaerobic growth experiments on sole xylose were inoculated with cultures grown on 20 g I^"1 glucose minimal medium. Inocula were washed twice with PBS prior to inoculation to avoid glucose contamination.

Chemostat cultivations for transcriptome and intracellular metabolic flux analysis were grown on minimal medium with double amounts of all components except KH₂P0 . A dilution rate of D 0.05 h^"1 at a working volume of 1 I was maintained volumetrically in a 2 I stirred tank reactor (Bioengineering, Wald, Switzerland) by continuous removal of excess culture broth with a fixed needle. 2 M KOH was used to maintain a pH value of 5.0 and an airflow of 1 I min^"1 provided aerobic conditions, whereas continuous sparging of 0.35 I min^"1 pure N₂ (5.0, Pan Gas) was used to establish anaerobiosis. Under anaerobic conditions, the feeding medium tank was continuously maintained in an N₂ atmosphere. Stirrer speed was set at 1000 and 500 rpm under aerobic and anaerobic conditions respectively.

Stocks for strain maintenance were generated from overnight cultures grown in YPD medium by adding glycerol to a final concentration of 15% (w/v) and were then stored at - 80°C. To preserve the original clonal composition of selection chemostats, population aliquots from the selection cultures were frozen directly, without intermediate batch growth.

EMS mutagenesis

To increase genetic variability, cultures were randomly mutagenized with ethyl metha ne sulfonate (EMS) (Sigma), by harvesting population aliquots at 1,500 rpm and 4°C for 3 min from minimal medium batch cultures in early stationary phase. Pellets were washed once with PBS and resuspended in 10 ml PBS. 300 μl of EMS were added and the suspension was incubated on a rotatory shaker at 300 rpm and 30°C. After 40 min, 20 ml 5% (w/v) Na₂S₂0₃ were added to inactivate the mutagen. After centrifugation, the pellet was washed twice with 5% (w/v) Na₂S₂0₃ to remove residual EMS, resuspended in 20 ml minimal medium, and stored at -80°C after addition of 15% (w/v) glycerol. Survival rates of 5 to 30% were verified by counting colony-forming units.

Sample withdrawal and transcriptome analysis

After chemostat cultures showed constant steady state physiological parameters for three volume changes, two 50 ml samples were withdrawn for transcriptome analysis. This occurred at least six volume changes after feed start. Culture samples for micro-array analysis were harvested in liquid N₂ pre-cooled 50 ml polypropylene tubes (Greiner-bio-one, Kremsmϋnster, Austria), and immediately centrifuged for 3 min at 5000 rpm and 4 °C. The pellets were then washed twice with ice-cold AE buffer (50 mM Sodium acetate, 10 mM EDTA, pH 5.2) and rapidly frozen in liquid N₂ for storage at -80 °C. Total RNA extraction was performed by the hot-phenol-method, quantified and checked for high quality at 260 and 280 nm with a spectrophotometer. Formaldehyde containing agarose gels were used to assess RNA integrity. mRNA isolation, cDNA synthesis, in vitro transcription, cRNA fragmentation, hybridization (GeneChip YG-S98 Arrays, Affymetrix, Santa Clara, CA), array washing, staining and scanning were performed by the SWEGENE Micro-array Resource Centre (Lund, Sweden). Data acquisition, processing and comparison analysis was performed with The Microarray Suite Software, version 5.0 (Affymetrix) and the

Significance Analysis of Micro-arrays (SAM) EXCEL add-in software. Prior to comparison, the average signal from all gene features was scaled to a target value of 100 in all arrays using Microarray Suite v5.0. From the 9,335 transcript features present on YG-S98 arrays, only the 6,383 yeast open reading frames were considered for comparison analysis. The coefficient of variation (CV = standard deviation divided by the mean) was determined for all the 6,383 transcript features within duplicate experiments to quantify data quality. The highest calculated average CV of all 6,383 transcripts was 0.34. Since low transcript levels were difficult to be reliably measured, expression values lower than 20 were set to a value of 20. During the clustering process, transcript levels were considered as significantly changed if they were called as such by SAM, when the fold-change was at least two and the false positive rate was less than 4%.

Analytical methods

Cell growth was monitored by following the optical density at 600 nm (OD₆₀₀) or by determining the Klett-value with a Klettmeter (Bel-Art Products, Pequonock, NJ). Cellular dry weight (cdw) was determined from 10 ml culture aliquots that were centrifuged at 5,000 rpm for 20 min in pre-weighed glass tubes, washed once with water, and dried at 110°C for 24 h to constant weight. Commercially available kits were used for enzymatic determination of glucose (Beckman), xylose (Medichem, Steinenbronn, Germany), xylitol (R-Biopharm, Darmstadt, Germany), acetate (R-Biopharm), and glycerol (Sigma). Ethanol concentrations were determined by gas chromatography (5890E chromatograph; Hewlett- Packard) with a Permabond-CW20M-0.25 column (Macherey-Nagel) and butyrate as the internal standard. C0₂ and ethanol concentrations in the reactor offgas were determined with a Prima 600 mass spectrometer (Fisons Instruments, Uxbridge, England).

Determination of physiological parameters and intracellular metabolic fluxes In batch culture, exponential growth rates were determined by log-linear regression of OD₆₀o versus time with growth rate (μ) as the regression coefficient. The specific biomass yield (V_x/_S) was determined as the coefficient of linear regression of biomass concentration (X) versus substrate concentration (S) during the exponential growth phase. The biomass concentration was estimated from predetermined OD₆oo-to-cdw correlations during the mid- exponential growth phase of aerobic cultures on glucose for TMB3001, TMB3001C5, and TMB3001C1 (0.530, 0.581, and 0.479 g ODeoo^"1, respectively). During the exponential growth phase, specific glucose or xylose uptake rates (q_g\c or ςf_xyι) were calculated as the ratio between μ and Y_x/S. Ethanol, xylitol, acetate, and glycerol yields were calculated by linear regression of by-product concentration versus S.

In mixed substrate fermentation analysis, the specific xylose uptake rate was determined as the ratio of the linear regression coefficient of xylose concentration versus time and the average biomass concentration between the onset of xylose consumption and about 100 h after inoculation. In these cases, the OD₆oo-to-cdw correlation was determined at the end of each fermentation.

During chemostat cultivation, biomass and by-product yields were determined as the ratio of the produced molar carbon amount for a considered product and the molar amount of total carbon sources consumed during steady state. A ratio of 0.476 gC gBiomass^"1 was assumed. Specific consumption and production rates were calculated as the ratio of the considered molar production rates and the steady state biomass concentration. The amount of evaporated ethanol was measured and considered. To determine intracellular metabolic fluxes, a modified preexistent stoichiometric model was used.

Chemostat selection Evolution of a yeast strain capable of anaerobic growth on sole xylose was started with the metabolically engineered S. cerevisiae strain TMB3001, which overexpresses XR and XDH from P. stipitis and the native XK from a stable chromosomal insertion (Eliasson et al. 2000). To increase genetic variability, an EMS-mutagenized population of TMB3001 was used to inoculate all selection procedures. Direct gain-of-function selections for growth on sole xylose in batch cultures or in petri dishes proved to be unsuccessful under anaerobic conditions (data not shown). Similarly, extended selection in serial anaerobic batch cultures with xylose in combination with glucose for more than 30 generations did not yield a population capable of anaerobic growth on sole xylose (data not shown). Hence, two long- term anaerobic chemostats at a D of 0.05 h^"1 that contained a limiting concentration (1 g I^{" x}) of the growth-promoting sugars glucose or galactose and 5 g I^"1 of xylose were initiated. Galactose was chosen to avoid catabolic repression and competitive inhibition of xylose transport by glucose (Hahn-Hagerdal et al. 2001). Within about 170 generations (100 days), the steady-state biomass concentrations remained unaltered and only 5 to 10% of the supplied xylose were consumed in both cultures (data not shown), indicating the absence of any evolutionary progress toward anaerobic growth on xylose.

To facilitate sequential evolution of multi-gene changes that may be required for efficient anaerobic xylose catabolism, an aerobic chemostat culture with 5 g I^"1 xylose and 1 g I^"1 glucose was initiated (Fig. 1A). After about 30 generations, the steady-state xylose concentration declined and OD₆₀₀ increased. Within 90 generations a new steady-state was attained, during which 80% of the supplied xylose was consumed. At this stage, a culture aliquot was withdrawn, EMS-mutagenized, and used to inoculate two new aerobic chemostat cultures. Settings in the first chemostat were identical to the previous one and a comparable steady-state was attained immediately. This chemostat was then switched to anaerobiosis and OD₆₀₀ decreased to 0.4 while the residual xylose concentration increased to 4.5 g I^'1 (data not shown). This steady-state physiology was similar to the previous direct anaerobic selection on 5 g I^"1 xylose and 1 g I^"1 glucose. Since no significant improvements were observed during the following 30 generations, this culture was not followed further. The second, aerobic chemostat contained 5 g I^"1 xylose as the sole carbon source (Fig. IB). Different from the initial EMS-mutagenized TMB3001 population, a growing population, which consumed increasingly more xylose, was obtained thereby decreasing the residual xylose concentration from 1.5 g I^"1 to 0.3 g I^"1 after 60 generations. To install micro-aerobic conditions, the aeration rate was drastically reduced from 0.3 I min^{" 1} to below 1 ml min^"1 at generation 140. Within the following 20 generations, the residual xylose concentration increased and OD₆₀₀ decreased rapidly. When OD₆₀o appeared to be stable, aeration was turned off at generation 170. After an immediate rise, the residual xylose concentration decreased and OD₆₀n increased steadily for 100 generations (Fig. IB). At a residual xylose concentration of 0.4 g I^"1 and an OD₆₀₀ of 3.1 at generation 270, anaerobic conditions were established by continuous sparging with technical N₂. Soon after the onset of anaerobiosis, a stable steady-state was attained, albeit with a low OD₆oo (Fig. IB). To elucidate whether anaerobic growth on xylose could be improved further, an aliquot withdrawn at generation 310 was EMS mutagenized. Upon anaerobic batch growth on xylose as the sole carbon source, an anaerobic chemostat culture was grown for another 150 generations on xylose, during which a gradual increase in biomass formation was observed (Fig. IC). It should be noted that due to an 0₂ contamination of below 200 ppm in the N₂, conditions were not strictly anaerobic in this selection culture, as will also be shown later.

To identify the time point at which the capability for anaerobic growth on sole xylose emerged first in the 460 generations (266 days) of evolution, frozen aliquots from generations were plated on xylose minimal medium and incubated in a strictly anaerobic atmosphere. Colonies were only detected in population aliquots after 270 generations. To exclude the possibility that anaerobic growth of the evolved population was due to trace amounts of 0₂ in the technical N₂, the final evolved population was grown in batch cultures with increasing strengths of anaerobiosis. When anaerobiosis was established in serum bottles by continuous sparging with technical N₂ (0₂ < 200 ppm), the maximum biomass concentration was 0.50 ± 0.08 g_cdw I^"1 on 5 g I^"1 xylose. When anaerobiosis was established in tightly sealed, anaerobic Hungate tubes (without continuous N₂ sparging), however, the maximum biomass concentration was only 0.24 ± 0.03 g_cdw I^"1 on 10 g I^"1 xylose. This lower biomass yield under more stringent anaerobiosis clearly demonstrates the physiologically relevant role of contaminating 0₂; not only in these batch cultures but likely also during the anaerobic selection. Finally, anaerobic growth on sole xylose in Hungate tubes with the addition of Na₂S or cysteine as reducing agents and resazurin as a redox indicator confirmed the gain-of-function phenotype (data not shown). In all cases xylose was completely consumed at the end of growth. Clonal analysis

Since chemostat-evolved, asexual populations are typically heterogeneous, a population aliquot after 460 generations was plated on xylose minimal medium. Under anaerobic conditions the number of colony-forming units was 54 ± 4% of that seen on aerobic YPD plates, thus providing first evidence for population heterogeneity. The parental strain TMB3001, the evolved population at 460 generations, and 15 clones, which were isolated from anaerobic xylose plates, were then compared for their fermentation performance in anaerobic batch cultures with 50 g I^"1 glucose and 50 g I^"1 xylose. During the initial phase of exponential growth on glucose, almost no xylose was consumed but upon glucose depletion, growth ceased and xylose was consumed in a second phase (Fig. 2).

To quantitatively compare clonal fermentation performance, physiological parameters were determined during the glucose and the xylose consumption phases (Fig. 3). Since the specific xylose uptake decrease after about 100 h of fermentation (Fig. 2), parameters for the second phase were calculated for the period between glucose depletion and 100 h. Generally, two major phenotypic classes could be discerned. The first class of clones and the evolved population were similar to TMB3001 during growth on glucose, but exhibited on average an about 60% higher specific xylose uptake rate and reduced formation of the by- product xylitol during the second phase (Figs. 2A, B, C and 3). As a consequence, these clones accumulated up to 19% higher final ethanol concentrations (Fig. 3C). The more abundant second class of clones exhibited a radically different mode of growth on glucose (Figs. 2D and 3). On average, the maximum growth rate and biomass yield on glucose were reduced by 60% and 40%, respectively. Unlike TMB3001 and the class I clones, the class II clones began to consume xylose prior to glucose depletion (Fig. 2D). The specific xylose uptake rates were at least doubled in the class II clones, ranging from 0.19 g g^"1 h^"1 (TMB3001C12) to 0.31 g g^"1 h^"1 (TMB3001C1), when compared to 0.08 g g^"1 h^"1 for TMB3001. Xylitol yields (Fig. 3B) but also final ethanol accumulation (Fig. 3C) were reduced in these clones, when compared to TMB3001.

Determination of the time course of acetate and glycerol for selected clones of both classes confirmed the drastic physiological changes of the class II phenotype (Fig. 4). Compared to TMB3001 and the class I clones, the class II clones produced significantly more acetate and glycerol on both glucose and xylose. Physiological characterization

To further elucidate the phenotypic differences of the two co-evolved subpopulations, the best representatives of each phenotypic class were grown in single substrate batch cultures. The class II representative TMB3001C1 exhibited significantly slower aerobic glucose catabolism with a 36% reduced growth rate and a 48% reduced specific rate of glucose uptake, when compared to TMB3001 (Table 1). Although the efficiency of exponential growth on glucose was not affected, as judged from the maximum bioma ss yield (Y_x s), the maximum biomass concentration attained by TMB3001C1 was significantly lower.

During aerobic growth on sole xylose, both clones grew significantly faster than their parent, but TMB3001C1 grew by far the most rapid (Table 2). Surprisingly, only TMB3001C1 but not TMB3001C5 was capable of strict anaerobic growth on sole xylose in Hungate tubes (Fig. 5). Further increased strength of anaerobiosis by addition of cysteine (Table 3) or Na₂S (data not shown) had no significant impact on the growth of TMB3001C1. Compared to the evolved population, which also includes this clones, TMB3001C1 grew significantly faster on xylose under anaerobic conditions.

TABLE 1. Physiological parameters of TMB3001, TMB3001C5 (class I), and TMB3001C1 (class II) in aerobic batch culture with 5 g I^"1 glucose.

Strain μ_max q_gic Y_x/S Biomass_max

(h^"1) (g g^ h^"1) (g g^"1) (g I^"1)

TMB3001 0.44^a 3.14 ± 0.05 0.14^b 2.1^c TMB3001C5 0.41 2.61 ± 0.09 0.16 2.2 TMB3001C1 0.28 1.62 ± 0.02 0.17 1.4

^a Standard deviation (SD) from duplicate experiments, ± 0.01 ^b SD, ± 0.01. ^C SD, ± 0.02.

TABLE 2. Physiological parameters of TMB3001, TMB3001C5 (class I), and TMB3001C1 (class II) in aerobic batch culture with 5 g I^"1 xylose.

Strain μ_max C7_xyι Vχ _S Biomass_max

(h^"1) (g g^"1 !!-¹) (g g^"1) (g I^"1)

TMB3001 0.016^a' n. d.^c n. d. 2.1^d

TMB3001C5 0.064 0.13 ± 0.00 0.50 ± 0.02 1.9 TMB3001C1 0.119 0.27 ± 0.02 0.45 ± 0.04 2.0

^a Aerobic growth on sole xylose was observed only after serial growth in

YPX and YNB medium with 20 g I^'1 xylose. ^b SD from duplicate experiments, ± 0.001. ^c Not determined. ^d SD, ± 0.02 TABLE 3. Physiological parameters of TMB3001C1 and the 460-generation population in strict anaerobic batch culture with 10 g I^"1 xylose.

Yields of^a

Strain max Qxyl Biomass Ethanol Xylitol Glycerol

(h-¹) (g g^"1 h- (g g-¹) (g g^"1) (g g^"1) (g g^_1)

TMB3001C1 0.012^b 0.56° 0.021 ± 0.24^d 0.32 ± 0.044 ± 0.004 0.00 0.005

TMB3001C1 + 0.010 0.52 0.022 ± 0.21 0.37 ± 0.047±

Cys^e 0.008 0.09 0.008

460-gen. 0.004 0.23 0.018 ± 0.25 0.33 ± 0.036± population 0.006 0.01 0.001

Acetate yields were below 0.006 g g^"1 in all cases. ^b SD from duplicate experiments, ± 0.001. ^C SD, ± 0.15. ^d SD, ± 0.01. ^e Anaerobiosis was installed by addition of cystein.

To elucidate the reason for the persistence of the class I clones in the evolved population, despite their inability to grow anaerobically on sole xylose, anaerobic conditions were established akin to those in the final selection chemostat by continuous sparging (< 1 ml min^"1) with technical N₂ (< 200 ppm 0₂). Surprisingly, all three strains, TMB3001, TMB3001C1, and TMB3001C5, grew to an OD₆₀₀ of about 0.2 on minimal medium without carbon source supplementation, presumably on the 0.5 g I^"1 of ethanol that was added with the ergosterol and Tween 80 stock solution, and which provides sufficient carbon for the observed growth. Since neither TMB3001 nor TMB3001C5 grew on xylose under these not strictly anaerobic conditions, it appears that the class I clones survived in the selection by scavenging contaminating 0₂ for oxidation of the produced ethanol or possibly other metabolic by-products of the class II clones. The growth rate of TMB3001C1 on xylose was 0.07 h^"1 under these conditions, compared to 0.012 h^'1 under strict anaerobic conditions (Table 3). This indicates that also TMB3001C1 benefited from 0₂ contamination and explains why these class II clones are not washed out in the anaerobic chemostat at a D of 0.05 h^"1.

Batch culture selection Since strict anaerobic growth on xylose of the best isolated clone was still relatively slow, an EMS-mutagenized population of TMB3001C1 was grown sequentially in seven serial batch cultures, which corresponds to 40 generations. 20 isolated clones were then grown on sole xylose in strict anaerobic Hungate tubes (Fig. 6). Relative to the maximum growth rate of TMB3001C1, seven clones grew at about the same rate, ten clones grew 1.2- to 2- fold faster, and three clones grew 2- to 2.5-fold faster. The highest growth rate observed was 0.028 h^"1 for the clones 1 and 18. The three best clones were then characterized more accurately in anaerobic xylose batch cultures. While TMB3001C1 grew at the previously determined rate of 0.012 h^"1, clones 1, 14, and 18 grew at 0.027 ± 0.002 h ¹, 0.021 ± 0.002 h^"1, and 0.018 ± 0.002 h^"1, respectively.

Transcriptome and flux analysis

To elucidate the putative functional mechanisms responsible for the observed xylose catabolism improvements in general and anaerobic growth on xylose in particular, we cultivated the strains TMB3001C1, TMB3001C5 and TMB3001 in chemostat cultures at D 0.05 h^"1 for transcriptome analysis (Table 4). TMB3001C1 and TMB3001 were analyzed on 10 g I^'1 glucose and 10 g I^'1 xylose under both aerobic and anaerobic conditions, whereas TMB3001C5 showed no significant differences to TMB3001 under these conditions (data not shown), and was therefore not considered. Do to the inability of TMB3001 to attain a steady state under aerobic conditions on sole xylose at D 0.05 h^"1, we only investigated TMB3001C1 and TMB3001C5 under the latter conditions.

TABLE 4. Physiological parameters of TMB3001 (TMB), TMB3001C1 (Cl) and TMB3001C5 (C5) during chemostat cultivations (D 0.05 h^'1) for transcriptome analysis.

Aeration Aerobic Aerobic Anaerobic

10 g/L xyl 10 g/L xyl

C-sources 20 g/L xyl lO g/L GIc 10 g/L Glc

Strain TMB Cl Cl C5 TMB Cl

Biomass conc.^a (g/L) 6.773 7.973 8.723 7.782 0.982 0.855

Glucose conc.^b (g/L) 0.023 0.040 n.d. n.d. 0.047 0.075

Xylose cone.¹ ' (g/L) 4.287 0.277 0.826 0.725 7.381 4.723 q(Glucose)^b (gGlc/(gX.h)) 0.070 0.059 n.d. n.d. 0.492 0.568 q(Xylose)^b (gxyl/(gx.h)) 0.042 0.061 0.114 0.127 0.140 0.313 q(EtOH)^a (gEtOH/(gX.h)) 0.002 0.001 0.001 n.d. 0.244 0.287

Y(Xyhtol/Xylc ) se)^b (gXyhtol/gXylose) 0.004 0.003 0.005 n.d. 0.364 0.342 (mol C EtOH/mol C

Y(EtOH/S)^a sugars) 0.020 0.014 0.013 n.d. 0.504 0.424

(mol C Xyhtol/mol C

Y(Xyhtol/S)^b sugars) 0.001 0.001 0.005 n.d. 0.079 0.120

(mol C Acet./mol C

Y(Acet./S)^a sugars) 0.000 0.000 0.000 n.d. 0.002 0.003

(mol C Glyc./mol C

Y(Glycero/S)^{i 3} sugars) 0.000 0.000 0.001 n.d. 0.086 0.158 (mol C Succ./mol C

Y(Sucαn./S)^ε ' sugars) n.d. n.d. n.d. n.d. 0.000 0.000 (mol C Pyr/mol C

Y(Pyruv/S)^a sugars) n.d. n.d. n.d. n.d. 0.000 0.002

(mol C C02/mol C

Y(C02/S)^a sugars) 0.517 0.511 0.462 0.478 0.302 0.265

(mol C X/mol C

Y(X/S)^a sugars) 0.531 0.495 0.519 0.466 0.092 0.066

Carbon recovery³ (%) 106.9 102.1 100.0 94.4 106.5 103.8

^a S.D. < 5% S.D. < 10%

TMB3001C1 depleted completely the fed xylose in steady state under aerobic conditions in both presence and absence of glucose as co-substrate. Under these conditions, increased xylose consumption was correlated with growth, whereas under anaerobic conditions a doubled specific xylose uptake rate, without xylose depletion in steady state, could be observed. In the latter case, increased xylose catabohc rate was correlated with doubled glycerol yield and decreased biomass formation. As TMB3001C1, TMB3001C5 was abl e to consume completely xylose in the aerobic steady state on this sole carbon source.

To clarify the mechanisms, which are generally important for xylose catabolism under both anaerobic and aerobic conditions, in presence or absence of glucose as a co-substrate, we compared gene transcript levels from TMB3001C1 and TMB3001C5 to TMB3001 under the different cultivation conditions (Table 5). We grouped genes in two main clusters. 39 genes with significantly higher expression levels in all of the four considered comparisons composed the first cluster (Table 6), whereas the second cluster contained 19 genes with significantly decreased expression levels in all comparisons (Table 7).

TABLE 5. Comparisons performed to identify genes with important functions in xylose catabolism.

ED^a TMB3001C1, Aerobic, lOg/l glc + lOg/l xyl vs.^b TMB3001, Aerobic, lOg/l glc + lOg/l x^

AD^a TMB3001C1, Aerobic, 20g/l xyl vs. TMB3001, Aerobic, lOg/l g lc + lOg/l x^

FD TMB3001C5, Aerobic, 20g/l xyl vs. TMB3001, Aerobic, lOg/l glc + lOg/l x\

CB^a TMB3001C1, Anaerobic, lOg/l glc + lOg/l xyl vs. TMB3001, Anaerobic, lOg/l glc + lOg/l ^a comparison name vs. = Compared to

TABLE 6. List of genes with higher expression levels in all comparisons

Comparisons³ Gene ID Name (fold-changes) Gene function

YEL057C - 7.4 12.2 6.3 hypothetical protein YPL014W - 5.0 7.2 4.3 2.6 hypothetical pro ein ^ _{L ^} __

YIRO 19C MUC 1 4.0 3.4 eel I su rface f loccu I i n YOR119C RIOl 4.7 2.6 4.9 si m i la ri ty to a C . leg a ns ZK632^3 protein

YER176W EC 323.2

E1II1I

YOR329C SCD5 2.8 2.8 multicopy suppressor of clathrin deficiency and of ts mutants of IPL]

111111 YORI^OW GCYI ^ similar to mammalian aldoVketo reductases

YCL£09 UΛ 6 2.4 2.8 small ^^^^^^^- °f Acetolactate synthase m llil i

YER152C 2.4 2.3 weak similarity to E.coli hypothetical protein f470 mil fill s

YNL239W LAP3 aminopeptidase of cysteine protease family

11111111 YOR267C similarity to ser/thr protein kinases

KISsi

YJL102W MEF2 mitochondrial elongation factor G-like > rote in

SfeoIlS

^a Table 5 TABLE 7. List of genes with lower expression levels in all comparisons

Comparisons³

Gene ID Name (fold-changes) Gene function ss m

YDR461W MFA1 0.110.17 0.27 0.12 a-factor mating pheromone precursor isiEa

YJ L1 3C PH084 0.14 O-⁴² 0- 13 0 ^Inorgam protein

YMR266W RSN1 0.170.17 0.22 0.31 similarity to A.thaliana hypl protein w MgBMBmβi BMmESS ««— m *»-

>.tϊ5SSBBSSEeroew^sel

YML063W 0.230.22 0.15 0.33 Ribosomal protein SIB

YDL236W YCR018C SRD1 0.290.41 0.29 0.46 transcription regulator

YDL219W 0.340.240.35 0.41 strong similarity to S.equisimilis hypothetical protein B HB

YNR044W 0.500.47 0.41 0.18 Anchorage subunit of a-agqlutinin

³ Table 5

A more careful analysis of the first cluster (Table 6) shows strong overexpression of genes involved in the galactose catabolic pathway. From these genes only the galactose permease is correlated with xylose catabolism, because of its xylose transport ability (Hamacher et al. 2002). The generalized overexpression of the majority of the galactose catabolism genes is an indication for regulatory mutations. Furthermore, significant overexpression of pyruvate kinase 2, glycerol-3-phosphate dehydrogenase and alcohol dehydrogenase 4 indicates that the mutants try to improve intracellular metabolic fluxes in central carbon metabolism by increasing expression levels of limiting enzymes, and by lowering intracellular NADH levels, what increases xylitol dehydrogenase activity. Different regulatory proteins are listed in the first cluster.

Among genes listed in the second cluster (Table 7), strong decreased expression levels are observed for the mating type specific genes, indicating that TMB3001C1 and TMB3001C5 are both at least diploid. Beside lowered expression of some regulatory proteins, the second cluster contains also the hxt2 high affinity hexose transporter.

To approach the question about which are the mutations necessary to improve anaerobic xylose catabolism in order to achieve anaerobic growth on sole xylose, we compared intracellular metabolic fluxes and transcriptome data of TMB3001C1 and TMB3001 in chemostat culture on 10 g I^"1 glucose and 10 g I^'1 xylose under anaerobic conditions (Figure 7). Because xylose reductase was found to catalyze also the conversion of dihydroxyacetone phosphate to glycerin 3-phosphate using both NADH and NADPH, the model used to calculate intracellular metabolic fluxes was tested with both, a fully NADH and a fully NADPH dependent glycerin 3-phosphate dehydrogenase activity. This test resulted in the same flux solution for both cofactors except for the NADH specific and the NADPH specific xylose reductase catalyzed reactions (data not shown). As a consequence we unified this two reactions to one. From this comparison, it appears clearly that higher intracellular metabolic fluxes correlate well with increased expression levels of the majority of the enzymes involved in the corresponding fluxes. This observation is completed by the observation that under the considered conditions, TMB3001C1 shows tendentially increased expression levels of NADPH producing and NADH consuming enzymes. As in the more general analysis, the galactose permease seems to be the most relevant xylose transporter which is strongly overexpressed, whereas significantly changed expression profiles of different regulators were observed (data not shown). Correlations between regulators and genes with changed expression profiles remain to be established.

The first yeast strain that grows on xylose as the sole carbon source under strict anaerobic conditions is described above. Such strains were isolated from a long-term, multi-step chemostat evolution experiment, which was initiated with the metabolically engineered S. cerevisiae strain TMB3001 that overexpresses the xylose-utilization pathway of P. stipitis (Eliasson et al. 2000). The selection procedure was based on the well-known evolution of mutants with increased substrate affinity and utilization in chemostat cultures. However, the key to successful evolution was to decouple selection for aerobic and anaerobic xylose utilization (Fig. 1). Thus, the selective pressure was adjusted to the present capabilities of the evolving culture, allowing advantageous mutations to accumulate under growth permissive conditions. Although the clones described here were isolated after 460 generations or 266 days of selection, the ability to grow anaerobically on sole xylose was first detected after 270 generations, immediately after switching the culture conditions to anaerobiosis (< 200 ppm 0₂). The achieved phenotype of the best xylose-utilizing clone TMB3001C1, with a maximum specific growth rate of 0.012 h^"1 and a biomass yield of 0.021 g g^"1 during strict anaerobic growth on xylose, represents by no means a final stage of evolution. For instance, the anaerobic growth rate of TMB3001C1 could be more than doubled within 40 generations of batch culture selection (Fig. 6). Although the rate of anaerobic xylose metabolism is still relatively slow, the isolation of these improved clones argues against the view that eucaryotic xylose metabolism is necessarily tied to respiration. Our results are more consistent with the view that anaerobic growth on xylose does not naturally occur in yeasts because the rate of xylose metabolism is too slow, so that the rate of ATP production is insufficient (Kόtter and Ciriacy 1993; Hahn-Hagerdal et al. 2001). Since the strains evolved here consume xylose at a several-fold higher specific rate than for example the control strain TMB3001, one would expect that the accumulated beneficial mutations affect, at least in part, the rate of catabolism and thus ATP formation. The nature of the underlying genetic changes that cause the observed phenotypic changes remains unclear at present and are subject to further investigation in our lab. Multiple mutations were probably necessary to endow TMB3001 with the capability of strict anaerobic growth on xylose, since direct selection on plate, in batch, or chemostat culture failed. This may also explain to some extend why intense rational metabolic engineering efforts have not yet yielded such strains (Ho et al. 1999; Aristidou and Penttila 2000; Hahn-Hagerdal et al. 2001).

After 460 generations, the population consisted of at least two subpopulations with distinct phenotypes, thus evidencing population heterogeneity (or polymorphism) that is often observed during evolution experiments. The class I phenotype of the smaller subpopulation, representing one third of the isolated clones, was rather similar to the parental TMB3001 on glucose but was significantly improved on xylose. The best representative of these, TMB3001C5, exhibited a 60% higher specific xylose uptake rate and a four-fold higher aerobic growth rate on sole xylose, when compared to TMB3001 (Table 2). Consequently, this strain accumulated up to 19% more ethanol when grown anaerobically under process-like conditions in a mixture of 50 g I^"1 each of glucose and xylose (Fig. 3C). None of these class I clones, however, grew anaerobically on sole xylose; neither under strict anaerobiosis nor in the presence of contaminating 0₂. The class II phenotype of the more abundant subpopulation was characterized by an even further improved xylose metabolism and was additionally capable of strict anaerobic growth on xylose. The best representative of this subpopulation, TMB3001C1, exhibited a more than three-fold higher specific xylose uptake rate and an eight-fold higher aerobic growth rate on xylose, when compared to TMB3001 (Table 2). All class II clones grew slower and less efficiently on glucose than TMB3001 and exhibited significantly increased overflow metabolism to acetate and glycerol (Fig. 4), indicating a drastic reorganization of central metabolism.

The incapability of the class I subpopulation to grow anaerobically on sole xylose is surprising because it stably propagated in the anaerobic selection chemostat. Moreover, the maximum anaerobic growth rate of all isolated clones on xylose was significantly lower than the dilution rate in the anaerobic selection chemostat, thus these clones would be expected to wash-out. The most likely explanation for this obvious discrepancy is the 0₂ contamination (< 200 ppm) in the technical N₂ that was used to establish anaerobiosis in the bioreactor. This contamination was independently verified by mass spectroscopy (data not shown). Although incapable of anaerobic growth on xylose even in the presence of contaminating 0₂, we could show that the class I clone TMB3001C5 can grow on ethanol, and possibly other metabolic by-products of the class II clones, with the contaminating 0₂ as an external electron acceptor. Likewise, the class II clone TMB3001C1 grows significantly faster then the D of the anaerobic selection chemostat when cultivated under conditions with contaminating 0₂. This view is also consistent with the obvious absence of a strong selection pressure for high anaerobic growth rate on xylose during chemostat selection, since faster growing clones were readily selected within comparatively few generations in strict anaerobic batch culture.

The applied strategy is a fruitful combination of rational metabolic engineering to render a strain amenable for selection and evolutionary techniques. Recently, two industrial ethanol- producing strains were metabolically engineered with the same xylose-utilization pathway that was used here (Zaldivar et al. 2002). Compared to these industrial strains, the evolved strains shown here accumulate less xylitol and some clones have higher xylose consumption rates (e.g. TMB3001C1). Moreover, the engineered industrial strains produced only about 8% more ethanol than TMB3001 from a mixture of glucose and xylose (Zaldivar et al. 2002), while our best clone TMB3001C5 produced about 19% more ethanol than TMB3001 (Fig. 3). The presented evolutionary engineering of enabling or improving substrate utilization is not confined to the recombinant strain used here, but can in principle be applied to other substrates or organisms; e. g. the above industrial strains. As is the case for many pentoses in yeasts, the organism subjected to selection should have the genetic potential to utilize the new substrate. Evolution may then be used to improve substrate utilization or to improve it under novel conditions. While simpler traits may be directly selected for, more complex, multi-gene modifications require an evolution approach for step-wise improvements (Sauer 2001).

A particular aspect of the invention relates to the exploration of TMB3001C1 (DSM 15519). Strains, media, and cultivation conditions

The S. cerevisiae strain TMB3001 (CEN.PK 113-7A (MATa, his3-Δl, MAL2-8C, SUC2) /7/^'s3;;YIpXR/XDH/XK) (Eliasson et al., 2000) and its evolved mutant Cl (Sonderegger and Sauer, 2003) were used throughout. Cultures were stored in aliquots supplemented with 15% glycerol at -80°C and were revived by growth in YPD medium (10 g I^"1 yeast extract, 20 g I^"1 peptone, and 20 g I^"1 glucose). All cultures for physiological and DNA microarray experiments were grown in minimal medium (Sonderegger and Sauer, 2003; Verduyn et al., 1992). For anaerobic cultivation, the medium was supplemented with ethanol-dissolved ergosterol (Fluka) and Tween 80 (Sigma) at final concentrations of 0.01 g I^"1 and 0.42 g I^"1, respectively.

Shake flask cultures were grown in 500 ml baffled shake flasks with 50 ml medium at 30°C and 300 rpm. Carbon-limited chemostat cultures were grown in 1 I medium in a 2 I stirred tank reactor (Bioengineering, Wald, Switzerland) at a dilution (growth) rate of 0.05 h^"1. The volume was kept constant by continuous removal of excess culture broth through a sterile needle that was fixed at a predetermined height. A constant pH of 5.0 was maintained by automatic addition of 2 M KOH. Sparging with air at a rate of 1 I min^"1 established aerobic conditions, whereas sparging with pure N₂ (0₂ < 5 ppm) at a rate of 0.35 I min^"1 established anaerobiosis. Constant gas flow rates were controlled by a mass flowmeter (Inceltech, Toulouse, France). To ensure anaerobiosis, the feed medium was also maintained under a N₂ atmosphere. The stirrer speed was set to 1,000 or 500 rpm under aerobic or anaerobic conditions, respectively. Culture aliquots for metabolic flux and transcript analysis were withdrawn in physiological steady state, defined as stable cell density and rate of C0₂ evolution for at least three volume changes. Anaerobic batch experiments for metabolic flux analysis were done in Hungate tubes (Bellco Glass Inc., Vineland, NJ) that were sealed with butyl-rubber septa, using inocula that were washed twice with PBS (8 g I^"1 NaCl, 0.2 g I^"1 KCI, 1.44 g I^"1 Na₂HP0₄, 0.24 g I^"1 KH₂P0₄, pH 7.0). The basic salt solution was sparged with pure N₂ (0₂ < 5 ppm) for 15 min and, upon autoclaving, the remaining filter-sterilized, N₂-sparged medium components and 10 g I^"1 xylose or glucose were added. Where indicated, acetoin was added at a final concentration of 0.5 g I^"1. Withdrawal of culture aliquots was done under purging with pure N₂ .

RNA isolation and DNA microarray analysis

Two 50 ml culture aliquots were harvested in liquid N₂ pre-cooled polypropylene tubes (Greiner, Kremsmϋnster, Austria) and immediately centrifuged at 5,000 rpm and 4°C for 3 min. The pellets were washed twice with ice-cold AE buffer (50 mM Na-acetate, 10 mM EDTA, pH 5.2) and rapidly frozen in liquid N₂ for storage at -80°C. Total RNA was extracted by the hot-phenol-method (Schmitt et al. , 1990), and the absorbance at 260 and 280 nm was used for quantification and purity control. RNA integrity was assessed in formaldehyde- containing agarose gels (Sambrook et al., 1989). mRNA isolation, cDNA synthesis, in vitro transcription (cRNA synthesis), and cRNA fragmentation were performed according to the Affymetrix expression analysis technical manual. Hybridization, washing, staining, and scanning of the Gene Chip Yeast Genome S98 Arrays (Affymetrix) were done in a hybridization oven (Affymetrix), the Fluidics Station 400 (Affymetrix), and the GeneArray Scanner (Affymetrix).

Gene expression data were analyzed using the Microarray Suite 5.0 software (Affymetrix). The average fluorescence of each array was normalized to a common value of 100. From the 9,335 transcripts present on the YG-S98 array, only the 6,383 yeast open reading frames were considered (Boer et al. , 2003). The coefficient of variation (CV = standard deviation divided by the mean) was calculated for all 6,383 transcripts from duplicate experiments, and the average coefficient of variation over the entire array was used to assess the experimental error of gene expression analysis in each experiment. Since low transcript levels are inherently difficult to qantify, all expression values below 20 were set to a value of 20 for fold-change analysis. The fold-change indicates the relative change in transcript levels when compared to a reference culture of TMB3001, and was used to identify genes that were differentially expressed in the two mutants. Differential gene expression analysis was done with the Significance Analysis of Micro-arrays (SAM) EXCEL add-in software (Tusher er al., 2001). For this purpose, fold-changes were considered statistically significant above a factor of two at a false positive rate of 1% (Piper et al., 2002).

Analytical methods

Cell growth was monitored by following the optical density at 600 nm (OD₆₀₀). Cellular dry weight (DW) was determined from at least five 10 ml culture aliquots that were centrifuged at 5,000 rpm for 20 min in pre-weighed glass tubes, washed once with water, and dried at 110°C for 24 h to a constant weight. Commercially available kits were used for enzymatic determination of glucose (Beckman), xylose (Medichem, Steinenbronn, Germany), xylitol (R-Biopharm, Darmstadt, Germany), acetate (R-Biopharm), and glycerol (Sigma). Ethanol, acetoin, and butanediol concentrations were determined by GC as described before (Sauer er al. , 1996). Pyruvate and succinate concentrations were determined by HPLC (Perkin Elmer, Shelton, Connecticut), with a Supelco H column (Supelco, Bellefonte, PA) and 0.15% H₃P0 as the mobile phase. C0₂ and ethanol concentrations in the reactor off-gas were determined with a Prima 600 mass spectrometer (Fisons Instruments, Uxbridge, England).

Determination of physiological parameters and intracellular metabolic fluxes Maximum exponential growth rates in batch culture were determined by log-linear regression of OD₆oo versus time with growth rate (μ) as the regression coefficient. The specific biomass yield (/_x/_s) was determined as the coefficient of linear regression of biomass concentration versus substrate concentration (S) during the exponential growth phase. The biomass concentration was estimated from predetermined OD₆₀o-to-DW correlations during the mid-exponential growth phase of aerobic cultures on glucose for

TMB3001 and Cl (0.530 and 0.479 g OD₆oo respectively). During the exponential growth phase, specific glucose or xylose uptake rates were calculated as the ratio between μ and

V_{x S}. Ethanol, xylitol, acetate, glycerol, and butanediol yields were calculated by linear regression of by-product concentration versus substrate concentration, their specific production rates were calculated as the product of specific xylose or glucose uptake rate and the by-product yield.

In chemostat cultures, biomass and by-product yields were determined as the ratio of the molar carbon in the considered product and the total molar carbon in the consumed substrates in steady state, assuming a ratio of 0.476 g(C) g(biomass)^"1 (Wahlbom er al. , 2001). Specific consumption and production rates were calculated as the ratio of the considered molar production rates and the steady state biomass concentration. The fraction of evaporated ethanol, 0₂, and C0₂ in the bioreactor off-gas were determined by on-line MS analysis.

A previously developed stoichiometric model (Wahlbom er al. , 2001) was used to estimate intracellular carbon fluxes in anaerobic chemostat and batch cultures. The fluxes to ethanol and C0₂ were defined as free fluxes, whose computed values were compared with the redundant experimental ethanol and C0₂ production rates. The computed free fluxes were always within 13% of the experimental values, thus confirming the reliability of the stoichiometric model used. This model was extended with acetoin reduction to butanediol (Wahlbom and Hahn-Hagerdal, 2002) for cases where acetoin was added. The macromolecular cell composition was assumed to be 39% (w/w) polysaccharides, 50% (w/w) protein, and 6% (w/w) RNA in chemostat culture, and 40% (w/w) polysaccharides, 52% (w/w) protein, and 3% (w/w) RNA as well as 31% (w/w) polysaccharides, 56% (w/w) protein, and 9% (w/w) RNA in batch cultures on xylose and glucose, respectively (Wahlbom er al., 2001). Since the recombinant xylose reductase in our strains can also catalyze the NADPH-dependent reduction of dihydroxyacetone-P (Jeppsson er al., 2003), we assessed the effect of changing the co-substrate specificity of the dihydroxyacetone-P dehydrogenase from NADH to NADPH. The effect was strictly local with a higher fraction of NADH-dependent xylose reduction (data not shown). Since we do not know the in vivo concentrations of NADH and NADPH, we do not know the in vivo co-factor preference of xylose reductase (Eliasson et al. , 2000). Hence, we were not able to determine the proportion between the NADH- and the NADPH-dependent xylose reduction fluxes.

Enzymatic assays

Cell extracts were prepared from mid-exponential growth phase cultures in minimal medium with glucose. Cell pellets were harvested by centrifugation, washed with deionized water, and resuspended in a 0.1 M triethanolamine buffer (pH 7.0), containing 1 mM phenylmethylsulfonyl fluoride, 0.5 mM dithio-threitol, and 0.5 mM EDTA. The suspension was vortexed with glass beads (0.5 mm diameter) at 4°C for 5 min, incubated on ice for 5 min, and vortexed again for 5 min. Cell debris and glass beads were separated by centrifugation at 20,000 x g and 4°C for 5 min. In vitro activities of xylose reductase (with NADPH and NADH), xylitol dehydrogenase (with NADH), and xylulokinase were determined as described previously in the supernatant (Eliasson et al., 2000). The sole difference was the use of thriethanolamine buffer at pH 7.0 instead of glycine buffer at pH 9.0 for the xylitol dehydrogenase assay. The total protein content of the supernatant was determined with a commercially available kit (Beckman). Specific activities were expressed as units per mg of protein, where one unit is defined as reduction or oxidation of one micromole of NAD(P)H per min.

TABLE 8. Metabolite concentrations and physiological parameters of S.cerevisiae TMB3001 and Cl in steady state of chemostat cultures at a dilution rate of 0.05 h^"1

Strain Feed concentration Steady state concentrations³ Specific consumption or Carbon balance

production rates

(g i^"1) (g i^"1) (g g(biomass)^"1 h^"1) (%)

Glucose Xylose Glucos Xylose Biomass Ethanol Xylitol Glycero Xylose Ethanol C0₂ e I c

53 t i Aerobic

H *■* TMB300 9.6 9.7 ± 5.5 ± 6.5 ± 0.3 0 0 -0.03 ± ^ϋ 0.09±000 114 ± 0 e H ¹ ±0.1^b

H 0.5 0.2 0.00 W Cl 9.6 ± 9.7 ± 0.1 ± 10.0 0 0 -0.05 ± ⁰ 0.07±0.0 115 ± 1

C 3

0.1 0.1 0.0 ±0.2 0.00

PI Cl 0 20.9 H 0.8 ± 8.7 ± 0.0 0.2 0.1 ± O. l±O.O -0.11 < 0.01 0.08 100 ± 0 ±0.0 0.2 ±0.0 0.0 ±0.00 ±0.00 α w Anaerob

O ic

TMB300 IO.O±O. 10.2 O. l±O. 7.4 ± 1.0 ± 0.0 ⁴-° 1.0 ± 1.2 0.14 ± 0.24 0.28±0.0 107 ± 3

1 ±0.1 0 0.0 ±0.2 0.0 ±0.0 0.01 ±0.00^c 0

Cl IO.O±O. 10-2 O. l±O. 4.7 ± 0.8 ± 0.0 4.1 1.9 ± 2.6 0.31 ± 0.29 0.34 104 ± 2 ±0.1 0 0.0 ±0.0 0.3 ±0.1 0.00 ±0.00° ±0.00

³ Acetate was below the detection level of 0.05 g I^"1 with the exception of the anaerobic Cl culture with 0.1 ± 0.0 g I^"1. ^b Mean value and 5 standard deviation from two independent cultures.

TABLE 9. Average coefficient of variation for the microarray experiments.

Cultivation Average coefficient of

Strain conditions variation³

Aerobic glucose/xylose TMB3001 0.16 glucose/xylose Cl 0.13 xylose Cl 0.34

Anaerobic glucose/xylose TMB3001 0.17 glucose/xylose Cl 0.17

^a Is the average of the coefficients of variation (standard deviation divided by the mean from two independent cultures) for all genes.

TABLE 10. Metabolism-related genes with significantly³ increased expression levels in Cl relative to TMB3001 under at least one chemostat condition.

Fold-change of gene expression in chemostat culture

Aerob 'ic Anaero bic c ,ene Gluco Xylos Glucose

Nam se e Xylose e ID Xylose Gene description

Significant in Cl under all conditions

ADH4 YGL256 2.6 4.3 2.5 Alcohol dehydrogenase isoenzyme IV

\Λ;

ADH5 YBR145 2.6 3 2.1 Alcohol dehydrogenase isoenzyme V

W

ASN1 YPR145 4 5.1 3.8 Asparagine synthetase

W

CDC2 YOR074 2.6 3.3 2.2 Thymidylate synthase

1 C

DFRl YOR236 2.3 3.5 2 Dihydrofolate reductase

W

GALl YBR020 388 288 140 Galactokinase

W

GALl YBR019 119 87 67 UDP-glucose 4-epimerase

0 C

GAL2 YLR081 211 285 339 Galactose permease

W

GAL7 YBR018 226 248 242 Galactose-1-P uridyl transferase

C

GPD2 YOL059 2.8 2.3 2.5 Glycerol-3-P dehydrogenase

W

HIS3 YOR202 2.4 2.5 2.4 Imidazoleglycerol-P dehydratase

W

ILV6 YCL009 2.7 3.2 2.6 Acetolactate synthase, small subunit

C

MAE1 YKL029 5.5 3.8 2.4 Mitochondrial malic enzyme

C

MCT1 YOR221 3.4 3.6 2.5 Malonyl-CoA carrier protein transferase

C

MEP3 YPR138 2 2.9 2.5 Ammonia transport, high capacity low

C affinity PUT4 YOR348 5 6.6 3.4 Proline permease

C

PYK2 YOR347 9.1 12 3.1 Pyruvate kinase, glucose-repressed

C isoform SOL3 YHR163 2.2 2.6 2.6 Possible 6-phosphogluconolactonase

W

SUL2 YLR092 2 2.2 2.1 Sulfate permease

W

YEL041 3.5 4 26 NAD kinase

W YER053 2.6 2.7 2.5 Similarity to C.elegans mitochondrial

C phosphate carrier

Significant in Cl only under aerobic conditions

ACH1 YBL015 2.7 2.2 1.5 Acetyl-CoA hydrolase W

BNA2 YJR078 3 2.7 1.5 Tryptophan 2,3-dioxygenase W

CAR2 YLR438 2.6 3.1 1.3 Ornithine transaminase W

GDH YAL062 2.5 2.2 0.9 NADP-linked glutamate dehydrogenase

3 W

HEM YDR232 2 2.1 1 5-aminolevulinate synthase

1 W

HXT1 YJR158 8 4.6 1.4 Hexose permease

6 W

MSE1 YOL033 2.2 2.1 1.7 Mitochondrial glutamyl-tRNA synthetase W

MSK YNL073 3.1 3.7 1.7 Mitochondrial lysyl-tRNA synthetase

1 W

NCP1 YHR042 2.9 2.9 0.9 NADPH-ferrihemoprotein reductase W

PDC6 YGR087 2.8 3 1.3 Pyruvate decarboxylase C

PGM YMR105 3.9 4.8 1.6 Phosphoglucomutase

2 C

PMT3 YOR321 2.3 2.1 0.9 Dolichyl-P-mannose-protein W mannosyltransferase

SDH1 YKL148 2.4 2.6 0.6 Succinate dehydrogenase cytochrome b C

YATl YAR035 3.7 4.9 1.8 Carnitine o-acetyltransferase W

ZWF YNL241 2.7 3.3 1.6 Glucose-6-P dehydrogenase

1 C

Significant in Cl only under aerobic conditions oι

COR1 YBL045 2.4 1.7 1.3 Ubiquinol-cytochrome c reductase core C subunit 1

CYTl YOR065 2.1 1.6 2 Ubiquinol-cytochrome c reductase W cytochrome cl subunit

HEM YDR044 2.6 1.6 1.1 Coproporphyrinogen oxidase, aerobic

13 W

RHR2 YIL053 2.1 1.8 1.7 DL-glycerol-3-phosphatase W

THR1 YHR025 2.2 1.4 1.3 Homoserine kinase W Significant in Cl only under anaerobic conditions

ARG5 YER069 1.6 1.9 2.4 N-acetyl-γ-glutamyl-P reductase and W acetylglutamate kinase

ECM4 YMR062 1.2 1.1 2.3 Amino-acid n-acetyltransferase

0 C

MET2 YOL064 1.6 1.5 2.1 3' - 5' bisphosphate nucleotidase

2 C ^a Fold-change of at least 2 with a maximal false positive rate of 1% (SAM). ^b Compared to TMB3001 cultivated under aerobic conditions on xylose and glucose.

TABLE 11. Specific enzyme activities (U mg(protein)^"1) in the catabolic xylose pathway of S. cerevisiae TMB3001, and Cl during aerobic batch growth on glucose.

xylose reductase xylitol dehydrogenase xylulokinase

TMB3001 0.08 ± 0.00^b 0.04 ± 0.00^c 0.8 ± 0.0 0.5 ± 0.0 Cl 0.42 ± 0.03 0.24 ± 0.01 1.9 ± 0.0 1.1 ± 0.0

³ Mean value and standard deviation from two independent experiments. With NADPH as cofactor. ^c With NADH as cofactor

TABLE 12. Regulatory genes of sugar transport, osmotic stress response, central carbon and galactose metabolism with changed expression levels in Cl relative to TMB3001 under at least one chemostat condition.

Fold-change of gene expression in chemostat culture Gene Anaerobi

Aerobic c

Glucos Xylos Glucos a e e e

H ID Name Xylose Xylose

Sugar transport YDR277C MTH1 2.6 2.1 1.0 Negative regulator of HXT gene expression g Central carbon metabolism w

H Transcription factor involved in glycolytic

2.1

YBR049C REB1 1.6 0.8 gene regulation

Transcriptional repressor, mediates

2.0

YBR112C CYC8 2.2 1.5 glucose repression

Transcriptional repressor, mediates

2.6

YCR084C TUP1 1.8 1.0 glucose repression

YDR081C PDC2 1.7 1.9 1.2 Regulates transcription of PDC1 and PDC5

Positive regulator of ADH2 and

0.8

YDR216W ADR1 1.0 0.1 peroxisomal genes

Required for release from glucose

1.8

YDR477W SNF1 2.0 1.0 repression

YLR256W HAP1 2.0 1.7 1.4 Activator of CYC1 and CYP3 transcription

Transcription factor involved in glycolytic

1.9

YNL216W RAPl 2.0 1.0 gene regulation

YOR358W HAP5 2.8 2.5 2.1 Regulates respiratory functions

YPL075W GCR1 1.6 1.9 1 6 Positive regulator of glycolytic genes

Galactose metabolism

Involved in galactose induction of GAL

YDR009W GAL3 21.1 14.6 7.5 genes

Inhibits activation by Gal4p in the absence

YML051W GAL80 4.1 3.5 2.1 of galactose a YNL239W GAL6 2.6 2.7 2.7 Negative regulator of the GAL genes

H YPL248C GAL4 2.1 2.1 0.4 Positive regulator of GAL genes d

HOG-pathway (osmotic stress response) YCR073C SS/ 22 2.1 2.9 1.8 Involved in osmolaπty sensing

Regulation of meiosis and response to

H YFL033C RIM15 1.7 2.2 1.1 stress

Transcriptional activator of the SKN7 d d YGR097W AS/CZO 1.7 2.2 2.5 regulatory system p Expressed during oxidative and osmotic

YHR206W SKN7 2.0 1.8 0.8 stress

YJL128C PBS2 1.3 1.4 2.9 Involved in osmolaπty sensing

Key regulator of stress-responsive gene

YKL062W MSN4 2.4 2.6 2.1 expression

YOL116W MSN1 2.3 2.6 1.3 Expressed during hyperosmotic response

YOR208W PTP2 2.0 2.9 2.1 Involved in osmolaπty sensing

Compared to TMB3001 cultivated under aerobic conditions on xylose and glucose.

TABLE 13. Physiological parameters of TMB3001 on glucose and of Cl on xylose in anaerobic batch cultures with 10 g I^'1 carbon source (C-source).

Yields of

Strain Substrate Maximum Specific Biomass³ Ethanol^b Xylitol^b Glycerol³ Acetate³

growth substrate rate³ uptake rate^b

(h ¹) (g g(biomass) ¹ (gX g ¹) (g g ¹) (g g ¹) (g g^"1) (g g^"1) d h^"1)

53 OJ

TMB3001 glucose 0.373 4.51 0.083 0.397 0 0.085 0.007

H d Cl xylose 0.014 0.60 0.024 0.277 0.321 0.041 0.007

Cl + xylose 0.019 0.62 0.031 0.373 0.180 0.028 0.011 acetoin⁰

P ≡I

H ³ Mean values. The standard deviation was below 10%. ^b Mean values. The standard deviation vas below 5%. d ^c 0.5 g I -¹ acetoin was added. pi

9

This Cl mutant provides a unique opportunity to elucidate the molecular mechanisms that are required for eukaryotic xylose metabolism under anaerobic conditions. Using DNA microarray (Conway and Schoolnik, 2003) and metabolic flux analysis (Sauer et al., 1996; Varma and Palsson, 1994; Wahlbom et al., 2001), we identify here the key components that enable this phenotype. While the results reveal that balancing of redox equivalents is one crucial component, we provide strong evidence that it is ultimately the rate of ATP formation that limits anaerobic growth on xylose. Results Chemostat cultivation For reliable and meaningful identification of specific differences between mutants, in particular when using genome-wide methodologies, it is important to ensure identical environmental conditions and to minimize physiological differences that cause unspecific responses (Conway and Schoolnik, 2003). For this purpose, we grew S. cerevisiae TMB3001 (Eliasson et al., 2000) and the evolved Cl mutant (Sonderegger and Sauer, 2003) in carbon-limited chemostat cultures at an identical growth rate for genome-wide transcript level and metabolic flux analysis. The dilution rate of 0.05 h^'1 was chosen as the highest value that allowed reliable establishment of a physiological steady state for both strains under all conditions used. The Cl strain consumed xylose at a higher specific rate and left significantly lower residual xylose concentrations in the medium (Table 8). This higher co-utilization of xylose enabled much higher accumulation of biomass in the aerobic Cl culture. Under anaerobic conditions Cl and TMB3001 attained comparable biomass concentrations but the Cl strain generated significantly more metabolic by-products from the additionally consumed xylose, in particular glycerol and xylitol. In contrast to Cl, TMB3001 was incapable to propagate aerobically on 20 g/L xylose as the sole carbon source (Table 8).

Global gene expression analysis in chemostat culture

DNA microarray analysis was done with RNA isolated from the above chemostat cultures to elucidate molecular changes that underlie the capability to grow on xylose as the sole carbon source. For each strain and condition, transcript levels were quantified from duplicate experiments with average coefficients of variation between 0.13 and 0.34 (Table 9). In total, 577 genes exhibited greater or equal than two-fold differential expression pattern (Piper et al., 2002) in Cl when compared to TMB3001 under at least one cultivation condition, 119 of which were differentially expressed under all three conditions (supplementary material). For growth on sole xylose, differential gene expression was determined by comparison with TMB3001 under the same condition but with the additional carbon source glucose, since TMB3001 cannot grow at this rate on xylose (Hamacher et al. , 2002; Sonderegger and Sauer, 2003).

A significant portion of the differentially expressed genes encoded for metabolic functions (partly shown in Table 10), and the by far strongest up-regulation was evident for the galactose metabolism genes GALl, GAL2, GAL7, and GAL10 under all conditions (Table 10). Expression of the central metabolic genes SOL3, MAE1, GPD2, ADH4, ADH5, and in particular PYK2 was generally increased, whereas PDC6, PGM2, SDH1, ZWF1, YATl, ACH1 and RHR2 were over-expressed only under specific conditions (Table 10). Although reported to transport also xylose (Hamacher et al., 2002), the hexose transporters HXT5 and HXT2 were down-regulated under all conditions and HXT4 and STL1 under aerobic conditions (supplementary material). The HXT16 gene, in contrast, was strongly up- regulated (Table 10). Generally, we noted lower aerobic expression of several genes encoding minor or putative isoenzymes in central carbon metabolism (TKL2, SOL4, YGR043C, and ALD3), and lower anaerobic expression of the peroxysomal genes ICLl and IDP3, as well as the glycerol kinase GUT1 and the mitochondrial glycerol-3-P dehydrogenase GUT2 genes in Cl (supplementary material). This down-regulation of genes that are responsible for cytosolic glycerol consumption correlates favorably with the higher expression of the glycerol producing genes GPD2 and RHR2 and the significantly increased glycerol production in the anaerobic Cl culture (Table 8). Furthermore, increased expression of the high capacity and low affinity ammonium transporter MEP3, and concomitantly decreased expression of the low capacity and high affinity ammonium transporter MEP2, demonstrate that Cl had adapted to the nitrogen-excess and carbon- limited conditions in the evolution chemostat (Sonderegger and Sauer, 2003). Among the most consistently and strongly down-regulated genes were the mating type-specific genes MFA1 and STE2. Since decreased expression of mating type-specific genes is characteristic for diploid or polyploid strains (Haber, 1998), we verified haploidy and mating type of the Cl mutant by PCR (data not shown). Upon detailed inspection of metabolic functions specified by differentially expressed genes, we noted a general pattern : central carbon metabolism genes involved in cytosolic NADPH formation (ZWF1, YEL041W) or NADH consumption (ADH4, ADH5, GPD2, and YEL041W) were up-regulated and those involved in NADH formation (ALD3 and GUT2) were down- regulated. Consistently, increased aerobic expression of the respiratory genes SDH1, COR1, and CYT1 indicates increased NADH oxidation in Cl. In addition, we observed higher expression levels of the NADPH-producing mitochondrial malic enzyme, which has been previously hypothesized to constitute a transhydrogenation system that re-oxidizes cytosolic NADH by reduction of mitochondrial NADP⁺ (Bakker et al. , 2001). Similarly, up- regulation of the NADPH-dependent glutamate dehydrogenase GDH3 indicates operation of a transhydrogenation cycle catalyzed by GDH1, GDH2 and GDH3 (Boles et al., 1993), which may re-oxidize NADH by reducing NADP⁺. The putative transhydrogenation mechanisms would be, however, irrelevant under fermentative conditions because of the very low tricarboxylic acid cycle flux (Gombert et al., 2001; Maaheimo et al., 2001; Nissen et al., 1997).

Comparison between transcript level and intracellular carbon fluxes

The large number of differentially expressed genes in central carbon and redox metabolism suggested significant metabolic differences between TMB3001 and Cl mutant. Hence we estimated the intracellular flux distribution from the anaerobic chemostat data of both strains on the glucose/xylose mixture (Table 8) with a previously described stoichiometric model (Wahlbom et al., 2001). Enhanced xylose catabolism in the Cl culture resulted in about 20% increased fluxes throughout the entire network but more than doubled fluxes through the pentose posphate pathway and into glycerol formation (Fig. 2). Notably, the flux of glucose-6-P into the pentose phosphate pathway was strongly increased in Cl, which is consistent with the notion that more cytosolic NADPH is required in the mutant to drive xylose reduction.

Generally, increased carbon fluxes in the Cl mutant were in excellent agreement with increased transcript levels of the corresponding genes from the same chemostat culture (Fig. 2). Albeit in many cases below the two-fold significance level, it is remarkable that a consistently increased expression level is discernible for most relevant central metabolic genes that is of about the same scale as the flux increase; i.e. ZWFl, SOLI, SOL3, GNDl, XKS1, TKL1, and TALI in the pentose phosphate pathway, PFK1, PFK2, FBA1, TDH1, TDH2, TDH3, PGK1, ENOl, EN02, PYK1, and PYK2 in glycolysis, and GPD1, GPD2, GPP1, PDC1, ADH1, ADH3, ADH4 and ADH5 in the by-product pathways. Among the generally overexpressed ADH genes, only ADH2 was unchanged. This is fully consistent with the function of ADH2 in metabolizing ethanol (Young et al. , 2002) because it is thus not required for anaerobic ethanol formation. Expression levels of the ALD genes involved in acetate formation, in contrast, were not increased significantly, as would be expected from the unaltered flux through this pathway. Since the recombinant xylose reductase and xylitol dehydrogenase were not present on the YG-S98 array, we determined the in vitro enzymatic activities of the xylose catabolism enzymes from batch cultures of TMB3001 and Cl (Table 11). The Cl mutant exhibited drastically higher activities of all three enzymes, in particular a five-fold higher xylose reductase activity. Indeed, as observed in Cl, increased overexpression of the xylose reductase has been previously reported to increase xylose uptake rate, glycerol and acetate yields, as well as decrease xylitol yield (Jeppsson et al., 2003). The redox cofactor specificity of xylose reductase remained unaltered at an about 100% higher activity with NADPH. While uncontrolled xylulokinase over-expression is detrimental to S. cerevisiae (Jin et al., 2003; Johansson et al., 2001), more than doubled activity is without any apparent negative effect in Cl, presumably because evolution increased all three enzyme activities simultaneously.

The large number of differentially expressed genes in central carbon metabolism and their consistent expression level suggests few regulatory rather than many specific mutations in the Cl mutant. Supporting this view, we found increased expression of the positive regulators of glycolysis and ethanol formation GCR1, REB1, RAPl and PDC2 (Table 12). The extraordinary up-regulation of GAL3 appears to override the regulation exerted by the moderately increased expression of the galactose repressors GAL80 and GAL6 (Table 12), thus being responsible for the strong induction of the GAL genes (Table 10). Increased expression of many regulators from the high osmolarity glycerol (HOG) regulatory pathway (Wojda et al., 2003) was probably responsible, at least in part, for the up-regulation of genes involved in glycerol formation. Lastly, increased expression of the negative regulator of hexose transport, MTH1, may be the common cause for the down-regulation of HXT2, HXT5, HXT4, and STL1.

Increased catabolic rates improve anaerobic growth on xylose Directed evolution of the Cl mutant had apparently relieved the metabolic limitations for anaerobic growth on xylose (Sonderegger and Sauer, 2003), and the above identified molecular differences in the Cl mutant suggest that the coordinated changes in carbon and redox metabolism are key components of this phenotype. These analyses reveal not, however, the pivotal limitation of eukaryotic pentose metabolism for anaerobic growth. The evolved Cl mutant provided a unique opportunity to address this question experimentally. Generally, the stoichiometry of anaerobic glucose metabolism was not radically different from that of anaerobic xylose metabolism, although less biomass was generated per gram of consumed substrate in the latter case (Table 13). The maximum specific rate of growth, in contrast, was dramatically lower on xylose (Table 13). This indicates a kinetic problem, which may reside in uptake and/or catabolism of xylose or, more specifically, in an insufficient rate of ATP production to support growth, as was hypothesized previously (Rizzi et al., 1989).

To differentiate between these possibilities, we grew Cl anaerobically on xylose in the presence of acetoin, which can be reduced to butanediol by consuming NADH. This reaction reduces the redox burden of significant cytosolic NADH formation in the xylitol dehydrogenase reaction (Fig. 1) and thus increases the rate of xylose catabolism during anaerobic xylose fermentation (Wahlbom and Hahn-Hagerdal, 2002). Acetoin was quantitatively converted to butanediol in the Cl culture (data not shown) and, presumably as a consequence of the reduced NADH burden, less xylitol accumulated in the presence of acetoin (Table 13). While the specific rate of xylose uptake remained virtually identical, the rate of xylose catabolism increased significantly because ethanol was produced instead of xylitol. Notably, co-metabolism of acetoin increased the rate of anaerobic growth and the biomass yield on sole xylose by about one third (Table 13), thus clearly demonstrating that xylose uptake is not the limiting factor for anaerobic growth of yeast on sole xylose.

Metabolic flux analysis was then used to distinguish whether increased xylose catabolism per se or, more specifically, increased rate of ATP formation caused the more rapid anaerobic growth of Cl on xylose plus acetoin. As expected (Maaheimo et al., 2001; Nissen et al., 1997), less than 4% of the consumed glucose enters the pentose phosphate pathway in the anaerobic TMB3001 culture, primarily to supply pathway intermediates for biomass formation (Fig. 3A). In the Cl culture, in contrast, about 20% of the consumed xylose re-enters the oxidative pentose phosphate pathway to supply NADPH for the xylose reductase reaction (Fig. 3B; upper values). Should the primary problem reside in the cytosolic redox cofactor balance with insufficient regeneration of NADPH, one would expect that regeneration of NAD⁺ in the reduction of acetoin increases the catabolic flux of xylose through the oxidative pentose phosphate pathway to generate more NADPH. However, no increased fluxes through the oxidative pentose phosphate pathway were observed in the acetoin co-fed Cl culture (Fig. 3B; lower values). From the estimated intracellular fluxes and the known ATP-producing biochemical reactions, we then calculated the specific ATP production rates of the three cultures. The production rate of 3.7 mMol(ATP) g(biomass)^"1 h^"1 of Cl on xylose was extraordinarily low, when compared to the 38.7 mMol g(biomass)^"1 h^'1 produced by TMB3001 during growth on glucose. In the presence of acetoin, however, the production rate increased to 4.9 mMol g(biomass)^'1 h^"1, and this 32% increase correlates almost perfectly with the 35% increase in growth rate in the presence of acetoin. Discussion

Based on global gene expression and metabolic flux analysis, we identified two distinguishing characteristics that are seemingly necessary for anaerobic growth of S. cerevisiae on sole xylose. Firstly we showed that enhanced xylose consumption was accompanied by moderately increased carbon fluxes throughout the entire network of Cl, most pronounced in the pentose phosphate and glycerol pathways on a glucose/xylose mixture. These calculated changes in flux between intracellular metabolites were in excellent agreement with the 1.5 to 2-fold up-regulated expression level of most central metabolic genes in the Cl mutant compared to its parent TMB3001. Secondly, the pattern of differentially expressed metabolic genes suggested that increased cytosolic NADPH formation and NADH consumption are important to drive xylose catabolism under anaerobic conditions. From these results we conclude that organisms such as yeasts that rely on two consecutive redox reactions for pentose catabolism (Fig. 1) are only capable of anaerobic growth on this compound when redox metabolism enables sufficiently high catabolic fluxes. Since we could not detect any mutations in the open reading frames and promoter regions of the most prominently up-regulated genes GAL2, PYK2, and those encoding for xylose reductase, xylitol dehydrogenase, and xylulokinase (data not shown), it appears that altered expression of key regulator proteins is predominantly responsible for this phenotype (Table 12).

Originating from a polyploid parent strain, the randomly generated S. cerevisae mutant TMB3400 with improved co-metabolism of xylose (Wahlbom et al., 2003b) was recently subjected to global expression analysis under almost identical conditions to those used here (Wahlbom et al., 2003a). Consistent with our Cl mutant, several pentose phosphate pathway and galactose metabolism genes were up-regulated. Since the Gal2p can also transport xylose (Hamacher et al., 2002), the observation that the GAL2 gene was among the most strongly up-regulated genes in the mutant investigated here and in TMB3400 (Wahlbom et al., 2003a) suggests strongly that xylose uptake is at least partly catalyzed by Gal2p. The consistent and strong up-regulation of galactose metabolism-related genes was probably induced by the higher expression level of the dominant positive regulator GAL3 (Table 12) (Wahlbom et al. , 2003a), in particular since we were unable to detect any mutation in the upstream region of GAL2 in our mutants. In contrast to Cl, expression levels of genes involved in glycolysis, redox metabolism, and ethanol or glycerol production were not altered in TMB3400, which may explain why this mutant cannot grow anaerobically on sole xylose (Wahlbom et al., 2003b). Our conclusion that increased cytosolic NADPH formation and NADH consumption are important for anaerobic growth of yeast on xylose concurs with flux and proteome data from recombinant, xylose-fermenting S. cerevisiae during a shift from glucose to glucose/xylose containing media (Pitkanen et al. , 2003; Salusjarvi et al. , 2003). In particular, it was shown that strains overexpressing the preferentially NADPH-dependent xylose reductase from P. stipitis generated the required NADPH primarily through the oxidative pentose phosphate pathway (Jeppsson et al. , 2002). This causes lower glycolytic fluxes and the concomitantly reduced NADH re-oxidation by reactions downstream of fructose-6-P can be compensated in respiring but not in fermenting yeast (Bakker et al. , 2001); hence explaining the capacity of such recombinant strains to grow aerobically but not anaerobically on sole xylose (Eliasson et al. , 2000; Ho et al., 1998). While directed evolution of Cl had apparently relived all metabolic bottlenecks for anaerobic growth on xylose (Sonderegger and Sauer, 2003), the slow rate of growth with a doubling time of about 50 h shows that the biosynthetic components are barely generated at an appropriate rate. Hence we tested whether or not the redox cofactor imbalance imposed by the two-step oxidoreductase reaction from xylose to xylulose is the primary reason why yeasts are generally incapable of anaerobic growth on sole xylose. Adding the NADH-oxidizing compound acetoin to Cl cultures increased the anaerobic growth rate on sole xylose by about one third, which demonstrates that balancing of redox co-factors was still a growth-limiting problem for Cl. While the rate of xylose uptake remained constant, xylose catabolism increased because less of the side-product xylitol was produced, which demonstrates clearly that xylose uptake was not limiting. Flux analysis revealed that the specific cytosolic NADPH production rate remained constant but that the ATP production rate increased by the same factor as the growth rate in the acetoin co-feed culture. This result strongly suggests that the rate of ATP formation is the primary limiting factor for anaerobic growth on sole xylose, although we cannot exclude that the pentose phosphate pathway operates at its maximum, and thus cannot supply more NADPH. Moreover, Cl is able to grow on sole xylose under anaerobic conditions at a specific ATP production rate of 3.7 mMol g(biomass)^"1 h^"1, whereas no growth was previously detected for TMB3001 with a rate of 1.8 mMol ATP g(biomass)^'1 h^"1 after acetoin addition (Wahlbom and Hahn-Hagerdal, 2002). This observation supports the hypothesis that the maintenance energy for anaerobic growth on sole xylose in yeast may be either higher as the value of about 1.2 mMol ATP g(biomass)^"1 h^"1 determined for P. stipitis under oxygen limited conditions (Rizzi et al. , 1989), or that the difference between ATP production and maintenance energy is insufficient to generate visible growth in TMB3001 after acetoin addition. Within 460 generations under selective pressure for improved xylose metabolism (Sonderegger and Sauer, 2003), the Cl mutant has apparently evolved such that altered redox cofactor metabolism in a number of reactions and higher expression levels of almost all catabolic genes permits higher catabolic fluxes of xylose to ethanol, which in turn provides ATP at a sufficient rate for growth under anaerobic conditions.

Transcriptome and intracellular metabolic flux analysis show clearly that, mainly under anaerobic conditions, the improved xylose catabolism observed in the mutant able to grow anaerobically on sole xylose increases absolute intracellular metabolic fluxes in the entire central carbon metabolism. This effect is correlated with higher expression levels of the involved enzymes, but also by a generalized increased expression of the majority of the NADPH producing and NADH consuming enzymes. This latter effect can be interpreted as an attempt of the mutant cell to increase flux through xylose reductase and xylitol dehydrogenase catalyzed reactions, lowering the negative effect caused by the redox imbalance generated by the different cofactor specificity of the two enzymes. One hypothesis, which is supported by these observations, could be that anaerobic growth on sole xylose is connected to a threshold ATP production rate. If this threshold value is reached, growth is observed (Kδtter and Ciriacy 1993; Hahn-Hagerdal et al. 2001). Furthermore, it appears that the bottleneck hampering the achievement of a sufficient ATP production rate is relived not only by acting against the redox imbalance, but also by a generalized overexpression of the majority of the enzymes involved in yeast anaerobic central carbon metabolism. In this view, xylose transport seems to be favored by overexpression of the galactose permease.

FIGURE LEGENDS

FIG. 1. Evolution of S. cere visiae TMB3001 in carbon-limited chemostat cultures at D of 0.05 h^"1 under aerobic conditions with 5 g I^"1 xylose and 1 g I^"1 glucose (A); aerobic, microaerobic (light gray background), and anaerobic (dark gray background) conditions with 5 g I^"1 xylose (B); and anaerobic conditions with 5 g I^"1 xylose (C). (1) Airflow reduction from 0.3 I min^"1 to < 1 ml min^"1, (2) shut-off of airflow, and (3) onset of anaerobiosis by sparging with technical N₂. The evolving population was subjected to EMS mutagenesis prior to inoculation of the chemostats A, B, and C.

FIG. 2. Fermentation profile of TMB3001 (A), the 460-generation selection (B), clone TMB3001C5 representing the first phenotypic class (C), and clone TMB3001C1 representing the second phenotypic class (D), during anaerobic growth on 50 g I^"1 glucose and 50 g I^"1 xylose. Glucose and xylose consumption phases are highlighted by I and II, respectively. Gray shading indicates simultaneous consumption of glucose and xylose.

FIG. 3. Physiological parameters during anaerobic growth on 50 g I^"1 glucose and 50 g I^"1 xylose of TMB3001, the 460-generation population, and 15 clones isolated from this population. Maximum growth rate and biomass yield were determined during exponential growth on glucose (A). Specific xylose uptake rate and xylitol yield on xylose were determined between glucose depletion and 100 h of fermentation (B). The final ethanol concentration was determined at 180 h (C). Values for TMB3001 and the population are average values from duplicate experiments. Hairlines indicate the reference values of TMB3001.

FIG. 4. Yields of acetate (A) and glycerol (B) on glucose (black bars) and xylose (open bars) during anaerobic growth on 50 g I^"1 glucose and 50 g I^"1 xylose of TMB3001 and selected clones from both phenotypic classes. Yields on glucose were determined between inoculation and the begin of xylose uptake. Yields on xylose were determined between glucose depletion and 130 h. Values were determined from single experiments.

FIG. 5. OD₆₀o and xylose concentration during strict anaerobic growth of TMB3001C1 in minimal medium with xylose as sole carbon source.

FIG. 6. Strict anaerobic growth rates on xylose minimal medium of 20 clones that were isolated after seven serial anaerobic batch cultures on xylose. The hairline indicates the growth rate of the parental TMB3001C1 before selection.

FIG.7. Comparison of absolute intracellular metabolic fluxes (ovals, mMol gDW^"1 h^"1, S.D. < 10%) and transcript levels (Boxes, arbitrary unit) of TMB3001 (upper values) and TMB3001C1 (lower values) during anaerobic chemostat cultivation on 10 g I^"1 glucose and 10 g I^"1 xylose. Only genes with significant expression changes are shown.

REFERENCES

Aristidou A, Penttila M (2000) Metabolic engineering applications to renewable resource utilization.

Curr. Opin. Biotechnol. 11 : 187-198 Bakker, B.M., Overkamp, K.M., van Maris, A.J., Kόtter, P., Luttik, M.A., van Dijken, J.P., and Pronk, J.T. (2001) Stoichiometry and compartmentation of NADH metabolism in Saccharomyces cerevisiae. FEMS Microbiol. Rev. 25 : 15-37. Boer, V.M., de Winde, J.H., Pronk, J.T., and Piper, M.D. (2003) The genome-wide transcriptional responses of Saccharomyces cerevisiae grown on glucose in aerobic chemostat cultures limited for carbon, nitrogen, phosphorus, or sulfur. J. Biol. Chem. 278: 3265-3274. Boles, E., Lehnert, W., and Zimmermann, F.K. (1993) The role of the NAD-dependent glutamate dehydrogenase in restoring growth on glucose of a Saccharomyces cerevisiae phosphoglucose isomerase mutant. Eur. J. Biochem. 217: 469-477. Conway, T., and Schoolnik, G.K. (2003) Microarray expression profiling : capturing a genome-wide portrait of the transcriptome. Mol. Microbiol. 47 : 879-889.

Eliasson A, Christensson C, Wahlbom CF, Hahn-Hagerdal B (2000) Anaerobic xylose fermentation by recombinant Saccharomyces cerevisiae carrying XYL1, XYL2, and XKS1 in mineral medium chemostat cultures. Appl. Environ. Microbiol. 66: 3381-3386 Fiaux J, ςakar ZP, Sonderegger M, Wuthrich K, Szyperski T, Sauer U (2003) Metabolic flux profiling of the yeasts Saccharomyces cerevisiae and Pichia stipitis. Euk. Cell in press Gombert AK, dos Santos MM, Christensen B, Nielsen J (2001) Network identification and flux quantification in the central metabolism of Saccharomyces cerevisiae under different conditions of glucose repression.

J. Bacteriol. 183: 1441-1451 Haber, J.E. ( 1998) Mating-type gene switching in Saccharomyces cerevisiae.

Annu. Rev. Genet. 32: 561-599. Hahn-Hagerdal B, Wahlbom CF, Gardonyi M, van Zyl WH, Cordero Otero RR, Jδnsson U (2001) Metabolic engineering of Saccharomyces cerevisiae for xylose utilization.

Adv. Biochem. Eng. Biotechnol. 73: 53-84 Hamacher T, Becker J, Gardonyi M, Hahn-Hagerdahl B, Boles E (2002) Characterisation of the xylose-transporting properties of yeast hexose transporters and their influence on xylose utilization.

Microbiology 148:2783-2788 Ho NWY, Chen Z, Brainard AP (1998) Genetically engineered Saccharomyces yeast capable of effective cofermentation of glucose and xylose.

Appl. Env. Microbiol. 64: 1852-1859 Ho NWY, Chen Z, Brainard AP, Sedlak M (1999) Successful design and development of genetically engineered Saccharomyces yeasts for effective cofermentation of glucose and xylose from cellulosic biomass to fuel ethanol.

Adv. Biochem. Eng. Biotechnol. 65: 163-192 Jeppsson, M., Johansson, B., Hahn-Hagerdal, B., and Gorwa-Grauslund, M.F. (2002)

Reduced oxidative pentose phosphate pathway flux in recombinant xylose-utilizing

Saccharomyces cerevisiae strains improves the ethanol yield from xylose. Appl. Environ. Microbiol. 68 : 1604-1609.

Jeppsson, M., Traff, K., Johansson, B., Hahn-Hagerdal, B., and Gorwa-Grauslund, M.F.

(2003) Effect of enhanced xylose reductase activity on xylose consumption and product distribution in xylose-fermenting recombinant Saccharomyces cerevisiae.

FEMS Yeast Res. 3 : 167-175. Jin, Y.S., Ni, H., Laplaza, J.M., and Jeffries, T.W. (2003) Optimal growth and ethanol production from xylose by recombinant Saccharomyces cerevisiae require moderate

D-xylulokinase activity.

Appl. Environ. Microbiol. 69 : 495-503. Johansson, B., Christensson, C, Hobley, T., and Hahn-Hagerdal, B. (2001) Xylulokinase overexpression in two strains of Saccharomyces cerevisiae also expressing xylose reductase and xylitol dehydrogenase and its effect on fermentation of xylose and lignocellulosic hydrolysate.

Appl. Environ. Microbiol. 67: 4249-4255. Kόtter P, Ciriacy M (1993) Xylose fermentation by Saccharomyces cerevisiae. Appl. Microbiol. Biotechnol. 38:776-783

Maaheimo, H., Fiaux, J., ςakar, Z.P., Bailey, J.E., Sauer, U., and Szyperski, T. (2001)

Central carbon metabolism of Saccharomyces cerevisiae explored by biosynthetic fractional ¹³C labeling of common amino acids.

Eur. J. Biochem. 268: 2464-2479. Nissen, T., Schulze, U., Nielsen, J., and Villadsen, J. (1997) Flux distributions in anaerobic, glucose-limited continuous cultures of Saccharomyces cerevisiae. Microbiology 143: 203-218. Ostergaard S, Olsson L, Nielsen J (2000) Metabolic engineering of Saccharomyces cerevisiae.

Microbiol. Mol. Biol. Rev. 64: 34-50 Piper, M.D., Daran-Lapujade, P., Bro, C, Regenberg, B., Knudsen, S., Nielsen, J., and Pronk, J.T. (2002) Reproducibility of oligonucleotide microarray transcriptome analyses. An interlaboratory comparison using chemostat cultures of Saccharomyces cerevisiae.

J. Biol. Chem. 277: 37001-37008. Pitkanen, J.-P., Aristidou, A., Salusjarvi, L., Ruohonen, L., and Penttila, M. (2003) Metabolic flux analysis of xylose metabolism in recombinant Saccharomyces cerevisiae using continuous culture. Metabolic Eng. 5: 16-31.

Rizzi, M., Klein, C, C, S., Bui-Thanah, N.A., and Dellweg, H. (1989) Xylose fermentation by yeasts. 5. Use of ATP balances for modeling oxigen-limited growth and fermentation of yeast Pichia stipitis with xylose as carbon source. Biotechnol. Bioeng. 34: 509-514. Salusjarvi, L., Poutanen, M., Pitkanen, J.P., Koivistoinen, H., Aristidou, A., Kalkkinen, N., Ruohonen, L., and Penttila, M. (2003) Proteome analysis of recombinant xylose- fermenting Saccharomyces cerevisiae. Yeast 20: 295-314. Sambrook, T., Fritsch, E.F., and Maniatis, T. (1989) Molecular cloning: a laboratory manual.

Cold Spring Harbor, N.Y. : Cold Spring Harbor Laboratory Press. Sauer, U., Hatzimanikatis, V., Hohmann, H.-P., Manneberg, M., van Loon, A.P.G.M., and Bailey, J.E. ( 1996) Physiology and metabolic fluxes of wild-type and riboflavin- producing Bacillus subtiiis. Appl. Environ. Microbiol. 62: 3687-3696.

Sauer U (2001) Evolutionary engineering of industrially important microbial phenotypes.

Adv. Biochem. Eng. Biotechnol. 73: 129-170 Schmitt, M. E., Brown, T.A., and Trumpower, B.L. (1990) A rapid and simple method for preparation of RNA from Saccharomyces cerevisiae. Nucleic Acids Res 18: 3091-3092. Stafford DE, Stephanopoulos G (2001) Metabolic engineering as an integrating platform for strain development.

Curr. Opin. Microbiol. 4: 336-340 Toivari MH, Aristidou A, Ruohonen L, Penttila M (2001) Conversion of xylose to ethanol by recombinant Saccharomyces cerevisiae: importance of xylulokinase (XKS1) and oxygen availability.

Metabolic Eng. 3: 236-249 Tusher, V.G., Tibshirani, R., and Chu, G. (2001) Significance analysis of microarrays applied to the ionizing radiation response. Proc. Natl. Acad. Sci. USA 98: 5116-5121.

Varma, A., and Palsson, B.O. (1994) Metabolic flux balancing: Basic concepts, scientific, and practical use.

Bio/Technol. 12: 994-998. Verduyn, C, Postma, E., Scheffers, W.A., and van Dijken, J.P. (1992) Effect of benzoic acid on metabolic fluxes in yeasts: a continuous-culture study on the regulation of respiration and alcoholic fermentation.

Yeast 8: 501-517. Walfridsson M, Hallborn J, Penttila M, KerDenen S, Hahn-Hagerdal B (1995) Xylose- metabolizing Saccharomyces cerevisiae strains overexpressing the TKL1 and TALI genes encoding the pentose phosphate pathway enzymes transketolase and transaldolase.

Appl. Env. Microbiol. 61 :4184-4190 Wahlbom, C.F., Eliasson, A., and Hahn-Hagerdal, B. (2001) Intracellular fluxes in a recombinant xylose-utilizing Saccharomyces cerevisiae cultivated anaerobically at different dilution rates and feed concentrations.

Biotechnol. Bioeng. 72: 289-296. Wahlbom, C.F., and Hahn-Hagerdal, B. (2002) Furfural, 5-hydroxymethyl furfural and acetoin act as external electron acceptors during anaerobic fermentation of xylose by recombinant Saccharomyces cerevisiae. Biotechnol. Bioeng. 78: 172-178.

Wahlbom, C.F., Cordero Otero, R.R., van Zyl, W.H., Hahn-Hagerdal, B., and Jonsson, L.J.

(2003a) Molecular analysis of a Saccharomyces cerevisiae mutant with improved ability to utilize xylose shows enhanced expression of proteins involved in transport, initial xylose metabolism, and the pentose phosphate pathway. Appl. Environ. Microbiol. 69: 740-746. Wahlbom, C.F., van Zyl, W.H., Jonsson, L.J., Hahn-Hagerdal, B., and Otero, R.R. (2003b) Generation of the improved recombinant xylose-utilizing Saccharomyces cerevisiae TMB 3400 by random mutagenesis and physiological comparison with Pichia stipitis CBS 6054. FEMS Yeast Res. 3: 319-326.

Walfridsson, M., Bao, X., Anderlund, M., Lilius, G., Bulow, L., and Hahn-Hagerdal, B. (1996) Ethanolic fermentation of xylose with Saccharomyces cerevisiae harboring the Thermus thermophilus xylA gene, which expresses an active xylose (glucose) isomerase. Appl. Env. Microbiol. 62: 4648-4651.

Wojda, I., Alonso-Monge, R., Bebelman, J.P., Mager, W.H., and Siderius, M. (2003) Response to high osmotic conditions and elevated temperature in Saccharomyces cerevisiae is controlled by intracellular glycerol and involves coordinate activity of MAP kinase pathways. Microbiology 149: 1193-1204.

Young, E.T., Kacherovsky, N., and Van Riper, K. (2002) Snfl protein kinase regulates Adrl binding to chromatin but not transcription activation. J. Biol. Chem. 277: 38095-38103.

Zaldivar J et al. (2002) Fermentation performance and intracellular metabolite patterns in laboratory and industrial xylose-fermenting Saccharomyces cerevisiae.

Appl. Microbiol. Biotechnol. 59:436-442

Claims

1. New xylose-utilizing Saccharomyces cerevisiae mutant strain overexpressing the xylose-utilizing pathway by overexpressing xylose reductase (XR), xylitol dehydrogenase (XDH), and xylokinase (XK), wherein one or more of genes encoding NADPH producing enzymes are up-regulated and one or more of the genes encoding

NADH consuming enzymes are up-regulated.

2. New strain according to claim 1, wherein the genes XKSl, TALI, TKLl, ZWFl, SOLI, SOL3, GNDl, PGI1, PFK1, PFK2, FBA1, TPI1, GPD1, GPD2, TDH1, TDH2, GPM1, GPM2, EN02, PYK1, PYK2, PDC1, ADH1, ADH3, ADH4, ADH5, PYC2, GAL2, PUT4 and

GPP1, SOL2, TDH3, PGK1, ENOl, ADH6, MAE1, GPP2, HXT6, GAL2, YEL041W, and GCY1 are up- regulated.

3. New strain according to claim 1, wherein regulators of glycolysis and central carbon metabolism : GCR1, RAPl, REB1, PDC2, HAP1, HAP5, and SNF1 are over-expressed.

4. New strain according to claim 1, wherein a regulator of galactose metabolism : GAL3 is over-expressed.

5. New strain according to claim 1, wherein regulators of osmotic stress response

(HOG-pathway) : SSK22, RIM15, ASK10, SKN7, PBS2, MSN4, MSN1, and PTP2 are over-expressed.

6. New strain according to claim 1, wherein regulators with other functions: TEA1, YAP5, ACA1, CLN3, RIOl, and SNF2 are over-expressed.

7. New strain according to claim 1, wherein the genes YATl, YCR060W, YEL057C, YIL176C, MCH5, and ATM1 are expressed to improve the capability of anaerobic growth on xylose.

8. New Strain according to one or more of the preceding claims, wherein the genes encoding for xylose reductase (XR), and xylitol dehydrogenase (XDH) are over expressed to a higher degree than in the corresponding non-mutated strain TMB3001.

9. New xylose-utilizing Saccharomyces cerevisiae mutant strain deposited at DSMZ under deposition number DSM 15519.

10. New xylose-utilizing Saccharomyces cerevisiae mutant strain deposited at DSMZ under deposition number DSM 15520.

11. Method for selecting an improved xylose-utilizing Saccharomyces cerevisiae strain, which strain overexpresses the xylose-utilizing pathway by overexpressing xylose reductase (XR), xylitol dehydrogenase (XDH), and xylokinase (XK), characterized in that a starting Saccharomyces cerevisiae strain is cultured under aerobic conditions in minimal medium containing xylose as sole carbon source, is selected with regard to growth on xylose, is continued cultured under microaerobic conditions in minimal medium containing xylose as sole carbon source, is selected with regard to growth on xylose, and is finally cultured under anaerobic conditions in minimal medium containing xylose as sole carbon source to produce a xylose-utilising Saccharomyces cerevisiae strain.

12. Method according to claim 11, wherein random mutagenesis using ethyl methane sulfonate is carried out on the starting Saccharomyces cerevisiae strain during the aerobic culture phase.

13. Method according to claim 11, wherein random mutagenesis using ethyl methane sulfonate is carried out on the Saccharomyces cerevisiae strain selected after the aerobic culture phase.

14. Method according to claims 11-13, wherein selection is done using continuous culture.

15. Method according to claim 11, wherein selection is done using chemostat selection.

16. Method according to claims 11-12, wherein culturing in a continuous culture takes place at constant dilution rate.

17. Method according to claim 16, wherein the dilution rate is 0.05 h^"1.

18. Method according to claim 11, wherein the first selection is made after at least 50 generations culturing in minimal medium using xylose as only carbon source, the second selection is made after further at least 100 generations, and the third selection is made after further at least 120 generations.

19. Method according to claim 11, wherein the culturing under aerobic conditions takes place under at least 90 generations, preferably 130 generations.

20. Method according to claim 11, wherein culturing under microaerobic conditions takes place under further at least 110 generations, preferably at least 140 generations.

21. Method according to claim 11, wherein culturing under anaerobic conditions takes place under further at least 150 generations, preferably 190 generations.

22. Method according to one or more of the preceding claims, wherein single clones from anaerobic growth on xylose plates are selected, and plated on anaerobic minimal medium containing xylose as the sole carbon source, and plates are incubated in sealed environment, to provide for an anaerobic atmosphere.

23. Method according to claims 11-22, wherein a pre-selection step is carried out prior to the aerobic culture on sole xylose medium comprising an aerobic growth on a minimal medium containing glucose and xylose as carbon sources in a molar ratio of about 50: 50.