Plasmids and S. cerevisiae strains used in this study are summarized in Table 1. Escherichia coli strain DH5α was used for subcloning and was grown on lysogeny broth (LB) agar plates supplemented with 100 mg/L ampicillin. Defined mineral medium was used for S. cerevisiae cultivation (carbon source (glucose or xylose); mineral salts (5 g/L (NH₄)₂SO₄, 3 g/L KH₂PO_4, 0.5 g/L MgSO₄·7H₂O); vitamins and trace elements 13 and buffering agent (50 mM potassium hydrogen phthalate, pH 5.5)) 14 . Yeast strains were streaked from 15% v/v glycerol stock and grown for 2 days on yeast nitrogen base (YNB) glucose plates at 30°C. A single colony was pre-grown overnight to late exponential phase in mineral medium (20 g/L glucose) and used for subsequent inoculation. Cultivation of S. cerevisiae was performed at 30°C.

Standard molecular biology techniques were used for all cloning procedures 15 . Plasmid purification was performed using a commercial kit (GeneJet Plasmid Miniprep Kit; Fermentas, Vilnius, Lithuania) and DNA was extracted from agarose gels (QIAquick Gel Extraction Kit; Qiagen, Hilden, Germany). Enzymes including T4 DNA ligase, calf intestine alkaline phosphatase, DreamTaq polymerase and various restriction enzymes were obtained from Fermentas. Transformation of E. coli and S. cerevisiae was performed using the calcium chloride 16 and the lithium acetate 17 methods, respectively.

The SUT1 gene from P. stipitis and the At5g59250gene from A. thaliana were codon-optimized for expression in S. cerevisiae by using the Codon Adaptation Index (CAI) and the Java Codon Adaptation Tool (JCat) 18 . The optimized SUT1 and At5g59250 genes were constructed synthetically (GenScript, Piscataway, NJ, USA), resulting in plasmids pSUT1 and pAt5g59250. The SUT1 and At5g59250 open reading frames (ORF) were isolated from plasmids pSUT1 and pAt5g59250 using the restriction enzymes SpeI and SalI followed by gel purification. The YIpDR4 integrative plasmid was constructed by inserting the SUT1 ORF between the TDH3 promoter and the CYC1 terminator of the previously constructed YIpDR1 vector (Table 1). YIpDR5 was similarly constructed using the excised At5g59250 ORF. The integrative plasmids YIpDR1, YIpDR4 and YIpDR5 were linearized using EcoRV, and transformed into S. cerevisiae strain TMB 3662 19 , yielding strains TMB 3416, TMB 3418 and TMB 3419 (Table 1). The control strain TMB 3415 was constructed by integration of plasmid YIplac128, not harboring any heterologous transporter gene, in TMB 3662 (Table 1).

Initial xylose uptake rates were determined using a previously described method 20 with minor modifications. Cells were inoculated into 1 L shake flasks in 100 mL medium (20 g/L glucose, 0.5 g/L glucose or 60 g/L xylose) at an optical density at 620 nm (OD₆₂₀) of 0.35, and were harvested by centrifugation (5000 g for 2 minutes) after approximately two replications (OD₆₂₀ = 1.5). The resulting cell pellet was washed in 10 mL potassium phosphate buffer 100 mM (pH 6.8), centrifuged (5000 g for 2 minutes) and resuspended in 1 mL potassium phosphate buffer. Initial uptake rates were determined at 25°C using D-[U-¹⁴C] labeled xylose (GE Healthcare, Buckinghamshire, UK; specific radioactivity 500-1000 cpm/nmol). The reactions were initiated by mixing 20 μL of buffered cell solution with 20 μL of radioactive substrate, giving final substrate concentrations between 3 and 150 mM. The reaction was quenched after 20 seconds by addition of 3.5 mL ice-cold 100 mM sorbitol solution. The resulting solution was immediately filtered through a 2.5 cm GF/C filter (Whatman, Maidstone, UK) and washed with 2 × 50 mL ice-cold water. Washed filters were transferred to scintillation vials containing 5 mL scintillation cocktail (Optiphase Hisafe 2; Perkin Elmer, CT, USA) and the corresponding radioactivity was measured in a liquid scintillation counter (Tri-Carb; Perkin Elmer). The baseline radioactivity was obtained in parallel measurements by adding the radioactive substrate to the cell solution after the quenching solution had been added. All measurements were performed in biological duplicate for each strain and condition. Matlab software (Matlab R2007b; The MathWorks Inc., MA, USA) was used for non-linear regression analysis of kinetic data.

Yeast strains were inoculated in 25 mL xylose medium (60 g/L, 15 g/L or 4 g/L) in 250 mL baffled shake flasks at a starting OD₆₂₀ of 0.2. Cells were grown at 30°C in an orbital shaker at 200 rpm. The growth rate was followed by measuring the optical density (OD) at 620 nm. All measurements were performed in biological duplicate for each investigated strain and each condition.

Xylose transport kinetics were determined in the xylose-utilizing S. cerevisiae strain TMB 3662 19 expressing the heterologous transporters Gxf1, Sut1 and At5g59250 from C. intermedia, P. stipitis and A. thaliana, respectively. Cells were grown in 20 g/L glucose for two cell divisions, after which xylose transport was assayed as initial rate of D-[U-¹⁴C] xylose uptake. The measured xylose transport rate was plotted against substrate concentration and fitted to a single-component Michaelis-Menten equation by non-linear regression (Figure 1, Table 2). Because the measured transport rate included both heterologous and native S. cerevisiae transporters, the determined kinetic constants refer to the overall cellular transport capacity. The ratio of the maximum transport rate, V_max and the substrate affinity, K_m, has, despite some shortfalls been widely used as an overall measure of enzymatic performance 21 22 , and was used here for quantitative comparison of the different transporters (Table 2). Expression of the C. intermedia Gxf1 transporter and the P. stipitis Sut1 transporter was thus seen to increase the transport performance compared with the reference strain expressing only native transporters (Figure 1, Table 2). In particular, the Gxf1 transporter increased the overall xylose transport performance by three times with respect to the control strain, which is comparable with results from previous studies 9 23 . A slight improvement was observed for the strain expressing the At5g59250 transporter, but it was not statistically significant (Figure 1, Table 2).

It has previously been shown that the effect of increased transport performance on xylose utilization is a function of strain background and substrate concentration 8 9 24 . In the current study, isogenic strains expressing the Gxf1, Sut1 and At5g59250 transporters were used to relate the measured transport kinetics to growth performance. The aerobic xylose growth rate of strains expressing the heterologous transporters was thus investigated at different substrate concentrations.

Strains expressing the Gxf1, Sut1 and At5g59250 transporters were first grown at high xylose concentration (60 g/L). At this concentration, the strains expressing heterologous transporters and the control strain all displayed the same specific growth rate (μ), 0.16 h-¹. Similarly, at intermediate xylose concentration (15 g/L), all strains still displayed the same growth rate, but it was reduced to 0.04 h-¹. At low xylose concentration (4 g/L), the specific growth rate was further reduced to 0.002 to 0.008 h-¹. However, at this concentration, strains expressing the heterologous transporters grew significantly faster than the control strain (Figure 2). The strain expressing the Gxf1 transporter displayed the highest growth rate, while the strain expressing the Sut1 transporter showed a small but reproducible increase in growth rate. The growth rate of the strain expressing the At5g59250 transporter was the same as that of the control strain (Figure 2).

Kinetic characterization of S. cerevisiae strains expressing the Gxf1 and Sut1 transporters showed increased xylose transport performance compared with the control strain expressing only native transporters (Figure 1, Table 2). The performance of native S. cerevisiae transporters, as opposed to constitutively expressed heterologous transporters, depends on the type of transporters expressed in the cell, which in turn depends on the carbon source and its concentration 7 25 . The reported overall xylose transport affinity of native S. cerevisiae thus varies between 80 and 200 mM, depending on the cultivation conditions in the particular experiment 6 7 10 . In the current study, the effect of the cultivation condition was assayed by measuring the xylose transport kinetics of native S. cerevisiae transporters during cultivation on different concentrations of glucose and xylose.

TMB 3415, expressing only native transporters, was grown on 0.5 g/L and 20 g/L glucose and on 60 g/L xylose, and was subsequently assayed for xylose transport (Figure 3). The xylose transport performance (V_max/K_m) increased when cells were grown at 0.5 g/L glucose (28 × 10-⁷ min-¹ × mgDW-¹) and 60 g/L xylose (13 × 10-⁷ min-¹ × mgDW-¹) compared with 20 g/L glucose (10 × 10-⁷ min-¹ × mgDW-¹). Cultivation at the low glucose concentration (0.5 g/L) yielded the highest xylose transport performance, which was comparable with cells expressing the Gxf1 transporter during cultivation on 20 g/L glucose (Figure 1, Table 2). The results confirm those of previous studies, which showed increased expression of high affinity transporters and increased xylose transport during non-repressive conditions; that is, low glucose concentration 7 25 .