OUP user menu

High-level functional expression of a fungal xylose isomerase: the key to efficient ethanolic fermentation of xylose by Saccharomyces cerevisiae?

Marko Kuyper, Harry R. Harhangi, Ann Kristin Stave, Aaron A. Winkler, Mike S.M. Jetten, Wim T.A.M. de Laat, Jan J.J. den Ridder, Huub J.M. Op den Camp, Johannes P. van Dijken, Jack T. Pronk
DOI: http://dx.doi.org/10.1016/S1567-1356(03)00141-7 69-78 First published online: 1 October 2003


Evidence is presented that xylose metabolism in the anaerobic cellulolytic fungus Piromyces sp. E2 proceeds via a xylose isomerase rather than via the xylose reductase/xylitol-dehydrogenase pathway found in xylose-metabolising yeasts. The XylA gene encoding the Piromyces xylose isomerase was functionally expressed in Saccharomyces cerevisiae. Heterologous isomerase activities in cell extracts, assayed at 30°C, were 0.3–1.1 μmol min−1 (mg protein)−1, with a Km for xylose of 20 mM. The engineered S. cerevisiae strain grew very slowly on xylose. It co-consumed xylose in aerobic and anaerobic glucose-limited chemostat cultures at rates of 0.33 and 0.73 mmol (g biomass)−1 h−1, respectively.

  • Xylose isomerase
  • Hemicellulose
  • Fermentation
  • Pentose
  • Yeast
  • Bioethanol

1 Introduction

Cost-effective production of fuel ethanol from renewable carbohydrate feedstocks (i.e. hydrolysates of plant biomass) requires the efficient ethanolic fermentation of the hemicellulose fraction, which is rich in pentose sugars. For this reason, ethanolic fermentation of the pentose sugar xylose has received considerable attention [13]. So far, however, no suitable microorganism is available for xylose fermentation, and ethanolic fermentation of lignocellulose is not yet carried out on a commercial scale.

Anaerobic metabolism of xylose by bacteria as well as anaerobic fungi invariably involves a mixed-acid fermentation with ethanol, formate, lactate, acetate, CO2 and hydrogen as products [4]. Bacteria such as Zymomonas spp. that can carry out homoethanolic fermentation do not ferment xylose, unless they are genetically modified [1].

Also, facultatively fermentative yeasts suffer from one or more disadvantages with respect to conversion of xylose into ethanol [57]. Three situations are encountered: (1) xylose cannot be metabolised (e.g. Saccharomyces cerevisiae); (2) xylose can be assimilated but cannot be fermented (e.g. Candida utilis); (3) xylose can be (slowly) fermented but anaerobic growth is not possible, not even on glucose (e.g. Pichia stipitis).

In bacteria, xylose metabolism is initiated by its isomerisation to xylulose (Fig. 1). Although xylose isomerase has been reported to occur in some fungi [811], metabolism of xylose in eukaryotes has so far been shown to proceed via xylose reductase and xylitol dehydrogenase (Fig. 1).

Figure 1

Conversion of xylose into ethanol via redox-neutral routes (I and II) or via NADPH-linked xylose reductase and NAD-linked xylitol dehydrogenase (III). Route III leads to a surplus of 1 NADH per xylose consumed. 1, Xylose isomerase; 2, NADH-linked xylose reductase; 3, xylitol dehydrogenase; 4, NADPH-linked xylose reductase; 5, xylulokinase.

The inability of most facultatively fermentative yeasts to convert xylose anaerobically into ethanol is due to an impairment of the redox balance resulting from the different coenzyme specificities of xylose reductase and xylitol dehydrogenase [12]. Only in cases where a reduced nicotinamide adenine dinucleotide (NADH)-dependent xylose-reductase activity is present in addition to the more common reduced nicotinamide adenine dinucleotide phosphate (NADPH)-linked activity (Fig. 1) can yeast species ferment xylose to ethanol in the absence of oxygen [13,14]. In facultatively fermentative yeasts that lack NADH-dependent xylose-reductase activity, slow anaerobic ethanolic fermentation of xylose can be accomplished when xylose isomerase (EC is added to the medium [12]. This clearly demonstrates that conversion of xylose into xylulose forms a bottleneck in pentose fermentation by yeasts.

Since the advent of metabolic engineering, many groups have tried to confer the ability to ferment xylose upon S. cerevisiae. This yeast is genetically well accessible, displays a high ethanol tolerance and is the only yeast species known that exhibits fast anaerobic growth on hexoses. Wild-type strains of S. cerevisiae show low rates of d-xylulose fermentation [15]. Engineered S. cerevisiae strains expressing heterologous genes encoding xylose reductase and xylitol dehydrogenase were first constructed in 1993 by Kötter and Ciriacy [16]. Such strains, however, exhibit very low rates of xylose metabolism. Despite more than a decade of intensive research in many laboratories, involving many additional metabolic-engineering steps, conversion rates of xylose to ethanol by S. cerevisiae strains based on this approach are still low.

An interesting alternative approach to realise ethanolic xylose fermentation by S. cerevisiae is the expression of a heterologous gene encoding aldose isomerase in this yeast. Hitherto, all attempts to express a prokaryotic xylose-isomerase gene in S. cerevisiae have met with disappointing results ([17] and references cited therein).

Much to our surprise, we found that in Piromyces sp. E2 the typical eukaryotic mode of xylose metabolism did not occur. This obligatory anaerobic cellulolytic fungus [4] appeared to lack the route via xylose reductase and xylitol dehydrogenase. This paper describes the cloning of the Piromyces xylose-isomerase gene and its functional expression in S. cerevisiae.

2 Materials and methods

2.1 Microorganisms and maintenance

The anaerobic fungus Piromyces sp. E2 (ATCC 76762), isolated from faeces of an Indian elephant, was maintained anaerobically under N2/CO2 (80%/20%) at 39°C in medium M2 [4]. The S. cerevisiae strains used and constructed in this study are congenic members of the CEN.PK family [18]. Stock cultures were grown at 30°C in shake flasks on YPD medium (yeast extract 10 g l−1 (BD Difco, Erembodegem, Belgium), peptone 20 g l−1 (BD Difco), and glucose 20 g l−1). When stationary phase was reached, 30% (v/v) sterile glycerol was added, and 2-ml aliquots were stored in sterile vials at −80°C.

2.2 Cultivation and media

Complex (YP: 10 g l−1 yeast extract (BD Difco), 20 g l−1 peptone (BD Difco)) and synthetic medium [19] were supplemented with glucose (2%) or xylose (2%) as carbon source and 1.5% agar in the case of plates. Plasmids were amplified in Escherichia coli strain XL-1 blue (Stratagene, La Jolla, CA, USA). Transformation was performed according to Inoue et al. [20]. E. coli was grown on LB (Luria–Bertani) plates or in liquid TB (Terrific Broth) medium for the isolation of plasmids [21].

2.3 Molecular biology techniques

Genomic DNA and total RNA were isolated from Piromyces sp. E2 cultures according to Brownlee et al. [22] and Chigwin et al. [23], respectively. Random sequencing was performed on a Piromyces sp. E2 cDNA library [24]. An aliquot of this library was converted to pBluescript SK-clones by mass excision with the ExAssist helper phage (Stratagene). Randomly selected clones were sequenced with the M13 reverse primer to obtain 5′ partial sequences. Incomplete cDNAs were used to synthesise probes to rescreen the library. To obtain full-length sequences, subclones were generated in pUC18. Sequence analysis was performed with the ABI Prism 310 automated sequencer using the dRhodamine terminator cycle sequencing ready-reaction DNA-sequencing kit (Perkin-Elmer Applied Biosystems, Foster City, CA, USA). For the preparation of poly(A)+ RNA, an mRNA purification kit (Pharmacia, Peapack, USA) was used according to the manufacturer's protocol. Southern and Northern analyses were performed as described previously [24]. For quantification, Northern blots were analysed with the Bio-Rad Gel Doc 1000 system, using Multi-Analyst software (Bio-Rad Laboratories, Hercules, CA, USA). Restriction endonucleases (New England Biolabs, Beverly, MA, USA and Roche, Basel, Switzerland) and DNA ligase (Roche) were used according to the manufacturers’ specifications. Plasmid isolation from E. coli was performed with the Qiaprep spin miniprep kit (Qiagen, Hilden, Germany). DNA fragments were separated on a 0.8% agarose (Sigma, St. Louis, MO, USA) gel in 1×TBE [21]. Isolation of fragments from gel was carried out with the Qiaquick gel extraction kit (Quiagen). Amplification of the Piromyces XylA gene was done with VentR DNA polymerase (New England Biolabs) according to the manufacturer's specification with the following settings: 30 cycles of 1 min annealing at 60°C, 3 min extension at 75°C and 1 min denaturing at 94°C. The polymerase chain reaction (PCR) was performed in a Biometra TGradient Thermocycler (Biometra, Göttingen, Germany).

2.4 Sequence analysis

DNA sequences were compared to sequences in the GenBank and EMBL databases using the BLAST algorithm (The Netherlands Centre for Molecular and Biomolecular Informatics, Nijmegen). A selection of sequences with the highest similarity was made and used to create an alignment with the PILEUP programme [25]. Alignments were checked manually. Phylogenetic analysis was done using the PHYLIP 3.5c software package [26]. The sequence data of the Piromyces sp. E2 XylA gene have been deposited in the DDJB/EMBL/GenBank databases under accession number AJ249909.

2.5 Plasmid construction

For expression of the Piromyces sp. E2 xylose isomerase in S. cerevisiae, the XylA gene was amplified by PCR using the primer combination 5′-GTGGCCCAGCCGGCCAAAATGGCTAAGGAATATTTC-3′ and 5′-AATCTAGATATTGGTACATGGCAACAATAG-3′. The PCR product was cloned into the pYES2 and pPICZα expression vectors (Invitrogen, Carlsbad, CA, USA), yielding pYES2XylA and pPICZαXylA, respectively. To replace the galactose-inducible GAL1 promoter with the constitutive TPI1 promoter, pYES2XylA was digested with SpeI–EcoRI, excising the GAL1 promoter, and ligated to an NheI–EcoRI fragment containing the TPI1 promoter, resulting in pAKX001. The TPI1 promoter fragment was obtained from pYX012-Aat (A.A. Winkler, unpublished), a derivative of pYX012 (R&D Systems, Minneapolis, MN, USA). To restore the original stop codon, a PCR was performed on pPICZαXylA with the following primers: 5′-CGGGATCCGAATTCAAAATGGCTAAGGAATATTTCC-3′ and 5′-GACTAGTTATTGGTACATGGCAACAAT-3′. The resulting PCR fragment was digested with EcoRI and SpeI and ligated to pAKX001 digested with EcoRI and XbaI, resulting in plasmid pAKX002 (Fig. 2). Plasmids pAKX001 and pAKX002 were used to transform S. cerevisiae CEN.PK113-5D MATa ura3–52. Transformation of yeast was done according to Gietz and Woods [27].

Figure 2

Expression vector pAKX002. The Piromyces sp. E2 xylose isomerase (XylA) gene expressed from the S. cerevisiae TPI1 promoter.

2.6 Batch cultivation

Piromyces sp. E2 was grown on xylose (0.5%, w/v) anaerobically under N2/CO2 (80%/20%) at 39°C in medium M2 as described previously [4]. Batch cultivation of S. cerevisiae strains was performed in shake flasks at 30°C on a synthetic medium [19]. The pH of the medium was adjusted to 6.0 with 2 M KOH prior to sterilisation. Precultures were prepared by inoculating 100 ml medium containing 20 g l−1 glucose in a 500-ml shake-flask with a frozen stock culture. After 24 h incubation at 30°C in an orbital shaker (200 rpm), this culture was used to inoculate either shake-flask cultures or chemostat cultures.

2.7 Chemostat cultivation

Chemostat cultures were grown in 1-l working-volume laboratory fermenters as described previously [28], at a temperature of 30°C, pH=5.0 and at a dilution rate of 0.10 h−1. This reference also cites the procedures that were used for gas analysis, analysis of biomass dry weight, biomass protein content and control of culture purity. For aerobic growth the dissolved oxygen tension was kept above 30% of air saturation. Anaerobic conditions were maintained by flushing the culture vessel with nitrogen gas. Some diffusion of oxygen was allowed via the inflowing medium and the silicone tubing on the reactor. In this way the addition of the anaerobic growth factors Tween 80 and ergosterol, which need to be dissolved in ethanol, could be circumvented, thus facilitating a more accurate construction of carbon balances.

2.8 Carbon balances

Ethanol evaporation from cultures was determined as follows. A synthetic medium containing an appropriate amount of ethanol was fed to a non-inoculated chemostat for five volume changes. Thereafter the ethanol concentrations in the feed and in the reactor were determined. These assays showed that 7.5% of the ethanol present in the reactor fluid evaporated under the stirring and gassing conditions maintained in anaerobic steady-state cultures. Therefore, all steady-state ethanol concentrations were corrected for evaporation by multiplying with 1.075. Carbon recoveries were calculated as carbon in products formed divided by the total amount of sugar carbon consumed. This calculation method is more informative than dividing ingoing carbon by outgoing carbon, as a large fraction of the xylose was not consumed. For carbon balancing, a carbon content of biomass of 48% was used.

2.9 Metabolite analysis

Xylose, xylitol, organic acids, glycerol and ethanol were detected by high-performance liquid chromatography (HPLC) analysis on a Waters Alliance 2690 HPLC (Waters, Milford, MA, USA) containing a Bio-Rad HPX 87H column. The column was eluted at 60°C with 0.5 g l−1 H2SO4 at a flow rate of 0.6 ml min−1. Detection was by means of a Waters 2410 refractive-index detector and a Waters 2487 UV detector. Glucose in reservoir media was determined enzymically using a hexokinase/glucose-6-phosphate-dehydrogenase kit (Boehringer, Basel, Switzerland, cat. no. 716 251). Xylulose was also determined enzymically. The reaction mixture consisted of 100 mM Tris–HCl buffer (pH 7.5) with 10 mM MgCl2, 0.30 mM NADH and an adequate amount of sample (1 ml total volume); the assay was started by the addition of sorbitol dehydrogenase (0.2 U per assay) (Sigma). The xylulose concentration was calculated using an absorption coefficient of 6.3 mM−1 cm−1 for NADH.

2.10 Preparation of cell extracts

For preparation of Piromyces E2 cell extracts, cells were harvested immediately after the cessation of hydrogen production, either by centrifugation (15 000×g) or by filtration over nylon gauze (30-μm pore size). After washing with demineralised water, cells were frozen in liquid nitrogen and ground in a mortar with 0.1-mm diameter glass beads. Tris–HCl buffer (0.10 M, pH 7.0) was added to the powder (1:1, w/v) and after 15 min thawing, the suspension was centrifuged (15 000×g, 15 min). The supernatant was used as cell extract.

Cell extracts of S. cerevisiae cultures were prepared by sonication as described previously [28]. Protein concentrations in cell extracts (0.5–4 mg ml−1) were determined by the Lowry method. Bovine serum albumin (fatty acid-free; Sigma) was used as a standard.

2.11 Enzyme assays

Assays were performed immediately after preparation of extracts, using a Hitachi model 100-60 spectrophotometer and 1-ml quartz cuvettes. An extinction coefficient at 340 nm of NADH of 6.3 mM−1 cm−1 was used to calculate specific activity. Reaction rates were linearly proportional to the amount of cell extract added. Specific activity of enzymes is expressed as Units (mg protein)−1. One unit equals 1 μmol of substrate converted per minute under the conditions of the assay.

Xylose isomerase (EC activity in cell extracts was determined spectrophotometrically at 37°C (Piromyces sp. E2) or at 30°C (S. cerevisiae) using a method modified from Kersters-Hiderson et al. [29]. The xylose isomerase assay mixture contained: Tris–HCl buffer (pH 7.5) 100 mM, MgCl2 10 mM, NADH 0.15 mM, sorbitol dehydrogenase (Sigma) 2 U, and cell extract. The reaction was started with 500 mM xylose.

Xylose reductase and xylitol dehydrogenase activities in cell extracts of Piromyces sp. E2 were assayed as described previously [30]. Xylulokinase assays were performed as described by Witteveen et al. [31].

3 Results

3.1 Involvement of xylose isomerase in xylose metabolism by Piromyces

Cell extracts of xylose-grown Piromyces sp. E2 lacked detectable activities of xylose reductase and xylitol dehydrogenase, the key enzymes of eukaryotic xylose metabolism. Instead, xylose-isomerase activity was detected (0.08 U (mg protein)−1). Also xylulokinase activity was detected during growth on xylose (0.2 U (mg protein)−1) Both enzymes were also present during growth on fructose. These results strongly indicate that xylose metabolism in Piromyces sp. E2 follows the ‘bacterial’ xylose isomerase pathway.

3.2 Identification and isolation of the structural gene encoding the Piromyces xylose isomerase

To identify highly expressed genes in the anaerobic fungus Piromyces sp. E2, a cDNA library was constructed from cells grown on fructose [24]. Among the 150 randomly selected clones three exhibited high sequence similarity with xylose-isomerase genes. Clone pR3 was analysed in detail and contained an insert of 1669 bp, which included an open reading frame (ORF) encoding a predicted protein of 437 amino acids with high similarity to xylose isomerases. Although the 5′ non-translated region comprised only 4 bp, the presumed N-terminal methionine residue fitted well into an alignment of known xylose isomerase sequences. The 3′ non-translated region was 351 bp in length and had a high AT content, which is typical of anaerobic fungi. The ORF contained conserved amino acids that, in other xylose isomerases, are involved in interaction with the substrate (His 102, Asp 105, Asp 340 and Lys 235) and binding of bivalent cations (Glu 232) [32,33]. The polypeptide deduced from the XylA sequence has a predicted molecular mass of 49 395 Da and a calculated pI of 5.2.

3.3 Expression of Piromyces sp. E2 xylose isomerase gene in S. cerevisiae

The Piromyces sp. E2 XylA gene was originally isolated by PCR amplification from a cDNA library and cloned into pYES2 (Invitrogen) and pPICZα. In order to get high constitutive xylose isomerase expression in S. cerevisiae, we replaced the GAL1 promoter in pYES2XylA with the TPI1 promoter, resulting in plasmid pAKX001. When CEN.PK113-5D was transformed with pAKX001, xylose-isomerase activity was detected in cell extracts, but the activity was low (0.025 U (mg protein)−1).

In pYES2XylA, and therefore also in pAKX001, the original stop codon of the XylA gene was replaced by an XbaI site. Consequently, the resulting protein was seven amino acids longer than the original Piromyces xylose isomerase. To investigate whether this might explain the low xylose-isomerase activity in the transformed S. cerevisiae strain, a new expression plasmid was made. Amplification of the XylA gene with primers that restored the stop codon to its original place and replacement of the XylA construct in pAKX001 with this new XylA construct (see Section 2 for details) resulted in plasmid pAKX002.

When S. cerevisiae CEN.PK113-5D was transformed with pAKX002, a much higher xylose-isomerase activity was found in cell extracts (1.1 U (mg protein)−1).

3.4 Preliminary characterisation of Piromyces xylose isomerase expressed in S. cerevisiae

Xylose isomerases from different organisms differ greatly in their affinity for xylose. Km values for d-xylose range from 0.08 to 33 mM for Streptomyces species and from 1 to 10 mM for Bacillus species [34]. In cell extracts prepared from chemostat cultures, the heterologously expressed Piromyces enzyme exhibited a Km of 20±1 mM for d-xylose under the assay conditions described in Section 2. Enzyme activity depended on the presence of bivalent cations, as has been observed for most xylose isomerases [34]. Omission of magnesium from the enzyme assay led to a decrease of the enzyme activity by 33%. When 2 mM EDTA replaced magnesium sulphate in the assay, activity decreased to below the detection limit. When magnesium in the assay was replaced with zinc, iron, nickel, copper, cobalt or manganese, only the addition of cobalt fully restored activity.

3.5 Metabolic consequences of xylose-isomerase expression on xylose metabolism by S. cerevisiae

Expression of the Piromyces xylose-isomerase gene in S. cerevisiae allowed very slow growth on 2% xylose on synthetic medium in shake-flask cultures (μ=0.005 h−1; Fig. 3). As this growth rate on xylose in batch cultures is too low to allow for quantitative analysis of xylose consumption and product formation, the physiology of the engineered strain was further analysed in aerobic and anaerobic chemostat cultures fed with mixtures of glucose and xylose.

Figure 3

Growth of S. cerevisiae RWB 202 (●) (CEN.PK113-5D with pAKX002), expressing Piromyces xylose isomerase, and the reference strain CEN.PK113-7D (○) in shake-flask cultures on synthetic medium with 2% (w/v) xylose as the sole carbon source.

In chemostat cultures of the engineered strain, growing at D=0.10 h−1, xylose-isomerase activities in cell extracts ranged from 0.32 to 1.1 U (mg protein)−1, the highest activities being observed in aerobic cultures (Table 1). In cultures grown on glucose as the sole carbon source, expression of xylose isomerase did not affect biomass yields or product formation (Tables 1 and 2).

View this table:
Table 1

Sugar concentrations in medium and supernatant, ethanol production, ethanol yield on glucose and on total sugar, biomass yield on glucose and xylose-isomerase activity of wild-type S. cerevisiae CEN.PK113-7D and the XylA-expressing strain RWB 202

Sugar in medium (g l−1)Glucose in (mM)Glucose out (mM)Xylose in (mM)Xylose out (mM)Ethanol out (mM)Biomass yield on glucose (g g−1)Ethanol yield on glucose (g g−1)Ethanol yield on total sugars (g g−1)Xylose isomerase (U (mg prot)−1)
RWB 202
7.5 glucoseaerobic39.1±1.4<0.0500<0.050.48±0.01000.42±0.01
5 glucose+2.5 xyloseaerobic27.9±0.7<0.0518.1±0.18.2±1.9<0.050.60±0.02001.10±0.12
25 glucoseanaerobic139.9±2.10.3±0.100217.1±2.80.08±0.000.40±0.000.40±0.000.33±0.06
20 glucose+10 xyloseanaerobic107.0±3.10.6±0.071.3±0.258.1±0.1181.4±2.00.10±0.000.44±0.020.39±0.010.55±0.02
25 glucoseanaerobic139.2±1.40.2±0.100217.0±3.70.09±0.000.40±0.000.40±0.00<0.01
20 glucose+10 xyloseanaerobic109.0±2.10.3±0.171.2±0.569.2±0.3167.7±1.40.09±0.000.39±0.010.39±0.01<0.01
  • Chemostat cultures were grown aerobically or anaerobically at a dilution rate of 0.10 h−1 as described in Section 2. Data are the average of two independent steady-state cultures.

View this table:
Table 2

Specific sugar-consumption and product-formation rates by aerobic and anaerobic cultures of wild-type S. cerevisiae CEN.PK113-7D and the XylA-expressing strain RWB 202

Sugar in medium (g l−1)q Glucoseq Xyloseq Ethanolq Glycerolq Xylitolq Xyluloseq CO2Carbon recovery (%)
RWB 202
7.5 glucoseaerobic1.17±0.020.00±0.000.00±0.000.00±0.000.00±0.000.00±0.002.89±0.1199.4±4.0
5 glucose+2.5 xyloseaerobic0.94±0.030.33±0.060.00±0.000.00±0.000.02±0.000.01±0.002.95±0.1197.8±3.6
25 glucoseanaerobic6.62±0.090.00±0.0010.29±0.160.85±0.000.00±0.000.00±0.0010.97±0.3596.6±1.3
20 glucose+10 xyloseanaerobic5.86±0.200.73±0.009.98±0.060.81±0.000.05±0.000.20±0.0010.60±0.0196.5±3.3
25 glucoseanaerobic6.07±0.070.00±0.009.48±0.160.81±0.000.00±0.000.00±0.0010.53±0.5598.5±1.6
20 glucose+10 xyloseanaerobic5.85±0.130.11±0.019.02±0.050.75±0.000.04±0.000.00±0.009.91±0.1596.7±1.9
  • Cultures were grown carbon-limited on glucose or glucose/xylose mixtures. Data are the average of two independent steady-state cultures. Specific rates (q) are expressed as mmol (g biomass)−1 h−1.

In cultures fed with mixtures of glucose and xylose (2:1 on a mass basis), xylose consumption by the isogenic reference strain CEN.PK113-7D was negligible (Table 1). In contrast, the strain expressing the Piromyces xylose isomerase consumed 20–50% of the xylose in the feed in aerobic as well as anaerobic cultures. In aerobic cultures, xylose consumption by the engineered strain resulted in an increased biomass formation relative to the reference strain (Table 1). The biomass yield on xylose, estimated by comparing the difference in biomass yield of aerobic glucose-limited cultures with the mixed-substrate cultures, was 0.4 g biomass (g xylose)−1. This figure corresponds closely with the biomass yield of the yeast C. utilis in aerobic, xylose-limited chemostat cultures (0.42 g g−1[35]). In anaerobic cultures, xylose co-consumption amounted to less than 10% of the total sugar consumption, which precluded accurate estimates of biomass and ethanol yields on co-consumed xylose. However, the apparent ethanol yield on glucose was significantly higher in anaerobic xylose-consuming cultures of the engineered strain than in the two reference situations (reference strain grown on mixed substrate and engineered strain grown on glucose). The ethanol yield on total sugar was the same in all three situations (Table 1).

Wild-type S. cerevisiae strains, although unable to metabolise xylose, are capable of very slow growth on xylulose [36,37]. This indicates that the xylulokinase encoded by the S. cerevisiae XKS1 gene can mediate the introduction of xylulose into the pentose-phosphate pathway at low rates. Furthermore, the sugar transporters encoded by the HXT4, HXT5, HXT7 and GAL2 genes are able to transport xylose, albeit with poor affinity [38]. The slow growth of the engineered strain on xylose and the observed co-metabolism of xylose along with glucose (Table 1) were therefore not unexpected. These properties are also displayed by strains expressing heterologous aldose reductase and xylitol dehydrogenase [16,39,40]. However, in such strains imbalances in redox metabolism readily occur due to the preferential use of NADPH in the xylose reductase reaction [13,30,41]. Under anaerobic conditions, such strains accumulate substantial amounts of xylitol during co-consumption of glucose and xylose, with up to 42% of the consumed xylose being reduced to xylitol [42]. Xylitol formation was also observed in anaerobic chemostat cultures of the xylose-isomerase-expressing strain (Table 2). However, the specific rates of xylitol formation were very low and accounted for less than 7% of the specific rate of xylose consumption. Interestingly, in line with the presence of active heterologous xylose isomerase, substantial amounts of xylulose were detected in the supernatants of the mixed-substrate cultures (Table 2).

In the anaerobic cultures, the flux towards glycerol was approximately 8% of the flux towards ethanol, as is generally observed during anaerobic growth of S. cerevisiae in mineral media [43]. Neither this ratio nor the production of the minor metabolites succinate and lactate (results not shown) were significantly affected by the co-consumption of xylose (Table 2).

4 Discussion

4.1 Conversion of xylose to xylulose in fungi

Glucose metabolism in microorganisms is nearly always initiated by a phosphorylation. In all cases encountered thus far, xylose is first isomerised to xylulose and then phosphorylated at the C5 position. Bacterial isomerisation of xylose invariably proceeds via a xylose isomerase. In fungi, however, isomerisation generally proceeds via a two-step reaction involving a reduction to xylitol with NADPH followed by dehydrogenation of the polyol with NAD as cofactor. Although xylose isomerase activity has been detected in some moulds and yeasts [811], its low activity compared to the oxidoreductases suggests that it is not the main enzyme in xylose metabolism. Our finding that a xylose isomerase is a key enzyme in xylose metabolism in Piromyces sp. E2 constitutes the first report of the ‘bacterial’ xylose-isomerase route in a eukaryote.

4.2 Heterologous expression of Piromyces sp. E2 xylose isomerase in S. cerevisiae

S. cerevisiae strain RWB 202 exhibited xylose-isomerase activities exceeding 1 U (mg protein)−1 (Table 1). If this activity, assayed in vitro, accurately reflects the capacity of the enzyme in vivo, it can accommodate xylose fluxes that are as high as the specific rates of sugar consumption in anaerobic, glucose-grown S. cerevisiae cultures [28].

Previous attempts to express xylose-isomerase genes from a variety of bacteria and Archaea in S. cerevisiae have failed. A notable exception involved xylose isomerases from thermophiles [44]. These, however, exhibited only very low activities at temperatures that are permissive to yeast growth. The lack of success of attempts to functionally express the Streptomyces rubiginosus xylose isomerase in S. cerevisiae has recently been attributed to incorrect protein folding [17]. The successful expression of the Piromyces xylose isomerase in S. cerevisiae may reflect that conditions in the cytosol of this fungus and/or mechanisms for protein (re-)folding are more similar between these two eukaryotes than between S. cerevisiae and the prokaryotes and Archaea that were previously used as xylose-isomerase gene donors. Research addressing this topic is in progress.

4.3 Xylose metabolism by a xylose-isomerase-expressing S. cerevisiae strain

Similar to S. cerevisiae strains that simultaneously express xylose reductase and xylitol dehydrogenase from P. stipitis[16,39,40], a S. cerevisiae strain expressing the Piromyces xylose-isomerase gene grew, albeit very slowly, on xylose as the sole carbon source (Fig. 3).

In shake-flask culture the engineered S. cerevisiae strain grew at a specific growth rate of 0.005 h−1. Assuming an aerobic yield of 0.4 g biomass (g xylose)−1, this corresponds to a specific xylose consumption rate of 0.08 mmol (g biomass)−1 h−1. Such a low rate is barely sufficient for the generation of maintenance energy, estimated at 0.5 mmol ATP (g biomass)−1 h−1[45,46]. Despite the slow growth and incomplete xylose consumption, our data confirm earlier conclusions that S. cerevisiae is capable of xylose transport and that d-xylulose can be fed into the pentose-phosphate pathway via the native S. cerevisiae xylulokinase (Fig. 1) [15,37].

In aerobic as well as anaerobic chemostat cultures grown on glucose–xylose mixtures, xylose consumption was incomplete, whereas glucose was completely consumed. Obviously, the macroscopic balances of the chemostat cultures (Tables 1 and 2) cannot identify the enzyme-catalysed reaction(s) that control the rate of xylose metabolism in the engineered strain. However, the observation that significant concentrations of d-xylulose were detected in culture supernatants (Table 2) suggests that these reactions occur downstream from xylose isomerase (i.e. in xylulokinase, the pentose-phosphate pathway and/or glycolysis).

4.4 Towards xylose-fermenting S. cerevisiae strains for ethanol production from lignocelluloses

Bacteria as well as yeasts have been studied as possible catalysts for the conversion of xylose into ethanol. Both groups of microorganisms, however, have major disadvantages. Xylose-fermenting bacteria show a mixed-acid fermentation in which, in addition to ethanol, (toxic) by-products such as formic, acetic and lactic acid are formed. In E. coli, this problem has been overcome by metabolic engineering of pyruvate metabolism, resulting in homoethanolic fermentation of xylose [47]. Wild-type isolates of Zymomonas spp. naturally perform a homoethanolic fermentation of glucose, but cannot ferment xylose. Also, this bacterium has been metabolically engineered for xylose fermentation by introducing genes required for the conversion of d-xylose into pentose phosphates [48]. A major disadvantage of bacteria such as E. coli and Zymomonas is their susceptibility to the harsh environment of lignocellulose hydrolysates, in which a low pH and compounds like acetic acid and furfural severely inhibit growth [49]. Most workers in the ethanol field have turned to yeasts as catalysts for ethanol production from pentoses, as yeasts are more acid- and ethanol-resistant.

Among the yeasts that are capable of ethanolic fermentation of glucose, 200 species are also capable of utilising xylose as carbon and energy source for growth [5]. However, in nearly all such yeasts ethanolic fermentation of xylose is either absent or extremely slow. Only a few yeasts, notably Pachysolen tannophilus, P. stipitis, Candida shehatae, Pichia segobiensis and some Candida tenuis strains are able to slowly convert xylose into ethanol under anaerobic conditions [5,13]. However, these yeasts cannot grow under strictly anaerobic conditions, either on glucose or on xylose [7].

Successful expression of xylose isomerase in S. cerevisiae is an important step towards a yeast that exhibits anaerobic xylose fermentation at a rate sufficient for industrial application. The main challenge in this respect is to achieve anaerobic growth of an engineered S. cerevisiae strain on xylose as the sole carbon source, a target that has not yet been met. The importance of the present work is that metabolic engineering strategies for xylose fermentation based on xylose isomerase circumvent the intrinsic redox problems associated with the xylose-reductase/xylitol-dehydrogenase approach. Many of the concepts developed in the extensive, painstaking research on S. cerevisiae strains expressing heterologous xylose reductase and xylitol dehydrogenase should be directly applicable to improve xylose fermentation in xylose-isomerase expression strains. For example, it has been established that enhancing the xylulokinase activity and one or more enzymes of the pentose-phosphate pathway leads to increased rates of xylose consumption [1,50]. Furthermore, deletion of the aldose reductase encoded by the GRE3 gene leads to a strong reduction of xylitol formation [51]. Also, the concept of Evolutionary Engineering (for a review see [52]), which has recently yielded interesting results with a xylose-reductase/xylitol dehydrogenase-expressing strain [53], may be applicable to xylose isomerase-expressing strains.

In further metabolic engineering of xylose metabolism, the xylose transport step should receive special attention. Native S. cerevisiae monosaccharide transporters catalyse facilitated diffusion and exhibit a poor affinity for xylose [38]. The rather high Km of the Piromyces xylose isomerase implies that achieving high rates of xylose fermentation may require the introduction of a heterologous high-affinity xylose transporter that catalyses xylose-proton symport [5458]. In this way, intracellular xylose concentrations can be maintained that saturate the isomerase, even at low extracellular xylose concentrations. This situation is reminiscent of maltose metabolism by S. cerevisiae in which maltose-proton symport [59] serves to supply a low-affinity α-glucosidase (‘maltase’) with its substrate.

Although the successful expression of a xylose isomerase in S. cerevisiae provides evidence that xylose metabolism without redox problems is possible, simple implementation of previously proposed concepts into a xylose-isomerase-expressing strain is no guarantee for success. In S. cerevisiae, the pentose-phosphate pathway is predominantly an assimilatory pathway. To sustain very high rates of xylose fermentation, many regulatory mechanisms may have to be bypassed, leading to imbalances in the overall metabolic network. In this respect, it is of interest that Saccharomyces species exist which, unlike S. cerevisiae strains, can employ the pentose-phosphate pathway as a catabolic route for anaerobic, fermentative growth. In Saccharomyces bulderi, δ-gluconolactone fermentation by anaerobic cultures involves equal contributions of the pentose-phosphate pathway and glycolysis [60]. Metabolic engineering of S. bulderi may therefore provide an interesting alternative for S. cerevisiae-based approaches.


We thank Sander Sieuwerts for his contribution to the optimisation of the isomerase assay. Part of this work was sponsored by the Dutch Foundation for Sustainable energy (SDE) and the Kluyver Centre for Genomics of Industrial Fermentation.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
  34. [34].
  35. [35].
  36. [36].
  37. [37].
  38. [38].
  39. [39].
  40. [40].
  41. [41].
  42. [42].
  43. [43].
  44. [44].
  45. [45].
  46. [46].
  47. [47].
  48. [48].
  49. [49].
  50. [50].
  51. [51].
  52. [52].
  53. [53].
  54. [54].
  55. [55].
  56. [56].
  57. [57].
  58. [58].
  59. [59].
  60. [60].
View Abstract