The yeasts used in this study were originally isolated from soil samples from King George island, the major island of the South Shetland Archipelago (sub-Antarctic region) (Carrasco et al., 2012), and are listed in Table 1. Yeasts were grown in YM medium (0.3% yeast extract, 0.3% malt extract, 0.5% Bacto peptone) supplemented with 1% glucose for GOD and ALP studies, and with 1% sucrose for Inv studies. The incubation temperature corresponded to that for the optimal growth of each yeast species (Table 1).

Yeast cultures at the early stationary phase of growth (100 ml) were centrifuged at 6,000 × g for 10 min at 4°C, and the supernatant was filtered through a polyvinylidene fluoride membrane of 0.45 mm pore-size (Millipore, Billerica, MA, USA). Proteins in the cell-free supernatant were precipitated by adding ammonium sulfate at 80% saturation, followed by incubation at 4°C for 2 h with constant agitation and centrifugation for 15 min at 10,000 × g at 4°C. The pellet was suspended in 4 ml of Tris-HCl buffer (0.1 M, pH 7.0) and dialyzed in tubes with 10 kDa molecular weight cut-off against 1,000 ml of Tris-HCl buffer (0.1 M, pH 7.0) for 12 h at 4°C, changing the buffer two to three times. Proteins were stored at −20°C until analysis. The protein content in samples was quantified using a BCA Kit Assay (Thermo Scientific, IL, USA) according to the manufacturer's instructions, and visualized by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). The protein Mr was estimated from SDS-PAGE gels using the PageRuler Plus Prestained Protein Ladder (Thermo Scientific, IL, USA) as a commercial protein reference standard.

To determine GOD activity, 50 μl of each protein sample was mixed with 50 μl of phosphate-citrate buffer with 10% glucose and incubated for 1 h. The reaction was then mixed with 98 μl of 0.2 M phosphate buffer, 0.17 mM o-dianisidine, and 2 μl peroxidase (6 U ml⁻¹), incubated at 25°C for 30 min, and the absorbance at 500 nm was determined. To measure Inv activity, 50 μl of protein sample was mixed with 50 μl of a solution containing phosphate-citrate buffer and 0.2 M sucrose, and incubated for 1 h. The release of reducing sugars was quantified by the DNS method (Miller, 1959). The ALP activity was determined using an Alkaline Phosphatase Assay kit (Abnova, Taipei, Taiwan) according to the manufacturer's instructions. The temperature and pH in the enzyme assays varied from 4 to 84°C and from pH 2.6 to 11.5, respectively.

Proteins from cell-free supernatants were differentially precipitated by adding ammonium sulfate at increasing saturation (20–80%). In each precipitation step, the sample was incubated at 4°C for 2 h and centrifuged at 10,000 × g at 4°C for 15 min. Protein pellets were suspended in 2 ml of potassium phosphate buffer (20 mM and pH 7.0) and desalted using a HiTrap Desalting column (GE, Schenectady, New York, USA). Proteins were separated using molecular exclusion or anion exchange chromatography. For molecular exclusion chromatography, Superdex 75 10/300 GL and Superdex 200 10/300 GL columns were used with a mobile phase of 50 mM phosphate, 150 mM NaCl, pH 7.0, at 0.2 ml min⁻¹ flow rate. For anion exchange chromatography, a HiPrep 16/10 QFF column was used with a mobile phase of 20 mM Tris-HCl, pH 8.0, and a NaCl gradient from 0 to 1 M at 0.5 ml min⁻¹ flow rate. The columns were operated using ÄKTAprime plus equipment (General Electrics, New York, USA).

To evaluate several physical-chemical parameters on enzyme activities, a two-level Plackett-Burman design was applied as detailed in Table S1. In each assay, the progression of the enzyme reaction was determined and the calculated specific reaction rates were used to determine the effect of each factor. Then, taking into account the two main factors that affected their activity, the conditions for highest activity of each enzyme were determined using a central composite design, and used for the determination of kinetic parameters. The initial reaction velocities at different substrate concentrations were calculated, and V_max and K_m were calculated according to the Michaelis-Menten model. To evaluate the irreversible thermal inactivation of enzymes, their activities were determined at temperatures from 4 to 84°C, incubating for 1 to 8 h. The inactivation rate constant (k) at different temperatures was obtained from the slope of plots of the natural logarithm of percentage of activity vs. the reciprocal of the absolute temperature. The activation energy (E_a) was obtained from the slope of the plot of the natural logarithm of k vs. the reciprocal of the absolute temperature. The amount of GOD, Inv and ALP used in the assays were 15.5 ug, 39.0 ug and 39.0 ug, respectively, in 100 μl final volume of the reaction. The concentration of glucose varied from 0.006 to 11 mM, of sucrose from 1 to 100 mM and of p-Npp from 0.001 to 2 mM.

The extracellular protein samples obtained from cultures of different yeast species were evaluated for GOD, Inv, and ALP activities at several temperatures and pH values. The results obtained from samples that displayed at least one enzyme activity are shown in Figure 1. GOD activity was detected in protein samples from five yeast species and, in general, the highest activity was detected at 30 to 37°C and pH 5.6 to 6.6 (Figure 1A). The exception was the protein sample from Leucosporidium creatinivorum, which displayed the maximum GOD activity at pH 2.6. The highest GOD activity was found in the protein sample from Goffeauzyma gastrica at 37°C and pH 5.6. Protein samples from three yeast species displayed Inv activity, which was highest at 30–37°C and pH 3.6–6.6. The highest Inv activity was detected in the protein sample from Wickerhamomyces anomalus at 37°C and pH 5.6 to 6.6 (Figure 1B). ALP activity was assayed at pH 10.0 and was detected in protein samples from W. anomalus, Candida sake, and Mrakia psychrophila with a maximum activity at 30 to 37°C. Goffeauzyma gilvescens and Dioszegia sp. displayed the highest ALP activity at 15°C and ≤ 22°C, respectively (Figure 1C).

As the protein samples from G. gastrica and W. anomalus displayed the highest GOD and Inv activity, respectively, these species were selected for further analyses to characterize the respective enzymes. The highest ALP activity was observed in samples from C. sake and Dioszegia sp.; the latter was selected for additional studies because it showed the highest enzyme activity at a lower temperature (15°C).

Proteins from yeast culture supernatants were fractioned by precipitation with ammonium sulfate from 20 to 80% saturation; each fraction was analyzed for protein content and enzyme activity. As shown in Figure S1, the Inv and ALP activities were detected in the 80% ammonium sulfate fraction of protein samples from W. anomalus and Dioszegia sp., respectively, while GOD activity was detected in the 60% ammonium sulfate fraction of the G. gastrica sample. However, multiple protein bands were observed in fractions having enzyme activity, so these fractions were subjected to additional purification steps by exclusion or ionic interchange chromatography (Figure 2). Three main peaks were observed in all chromatograms, and in each particular case, peaks 3, 2, and 1, coincided with the fractions having GOD, Inv, and ALP activity, respectively. Fractions in which the enzyme activity was detected and had a unique protein band were pooled together and concentrated in order to obtain the purified sample of each enzyme (Figure 2). According to the SDS-PAGE analysis, the GOD, Inv, and ALP activity would correspond to a protein band of Mr 91,000, 101,000, and 39,000, respectively. The purified enzyme samples were used to determine the physical-chemical factors that mainly influence the activity of each enzyme. The temperature and pH were tested for all enzymes, and the concentration of glucose, FeSO₄, and CaCl₂ was tested for GOD, the concentration of H₂O₂, urea, and ethanol was tested for Inv, and the concentration of ZnS0₄, MgCl₂, and L-phenylalanine was tested for ALP. According to this analysis, pH and temperature were the second most important factors affecting the activity of the three studied enzymes (under the ranges tested), with the addition of CaCl₂, ethanol, and MgCl₂ being the main factors affecting GOD, Inv, and ALP, respectively (Table S1). However, by analyzing the results of each assay, enzyme activity only increased in the case of adding CaCl₂, contrary to the addition of ethanol or MgCl₂ in which cases the enzyme activity decreased.

A central composite design with two levels was applied to determine the conditions at which each enzyme had the highest activity considering the two-main physical-chemical factors that affected the activity of each one. GOD showed the highest activity at 64°C and with 2 M CaCl₂, Inv at 60°C and pH 5.0, and ALP at 47°C and pH 9.0 (Figure 3). These conditions were used to determine the enzymes' kinetic parameters using glucose, sucrose, and p-nitrophenylphosphate at different concentrations as substrates for GOD, Inv, and ALP, respectively. All enzymes displayed a Michaelis-Menten kinetic (Figure S2) and the calculated parameters are listed in Table 2. The calculated catalytic efficiency (k_cat/K_M) for GOD, Inv, and ALP was 1.4 × 10⁷ M⁻¹ s⁻¹, 4.3 × 10⁷ M⁻¹ s⁻¹, and 2.4 × 10⁶ M⁻¹ s⁻¹, respectively, indicating that these enzymes are highly efficient.

The thermal effect on the enzymes' activity was assayed at 4 to 84°C, considering as 100% the activity of each enzyme at its optimal temperature. In all cases, the greatest activity decay was observed at higher rather than at lower temperatures (Figure 4A). At 30°C, the percentage of GOD and Inv activity was 28%, while it was 88% for ALP. ALP was the enzyme that displayed the highest percentage of activity at lower temperatures with 67% at 10°C, while the percentages of GOD and Inv activities were 13% and 18%, respectively, at the same temperature. At 4°C, the percentage of ALP activity was 43%.

Enzymes were also incubated for 1 h at different temperatures before performing the enzymatic assay at their optimal temperatures for activity. All of them retained at least 80% activity when they were previously incubated at temperature ranging from 4 to 54°C, 4 to 55°C, and 4 to 47°C for GOD, Inv, and ALP, respectively (Figure 4A). The percentage of activity strongly declined after 1 h of incubation at higher temperatures. In incubations for longer times, the GOD activity decayed linearly across 30, 47, and 54°C, maintaining 79, 68, and 11% of activity, respectively, after 6 h of incubation (Figure 4B). After 6 h of incubation, the percentage of Inv activity was 85, 71, and 52% when incubated at 22, 37, and 47°C, respectively, displaying a faster decay at 60 and 70°C (Figure 4C). Similarly, at 22, 37, and 47°C, ALP retained 87, 57, and 40% of activity, respectively, after 6 h of incubation, which declined faster at 57 and 60°C (Figure 4D). The calculated activation energy from the Arrhenius plots (Figure S3) was 35, 25, and 8 kJ mol⁻¹ for GOD, Inv, and ALP, respectively. The enzyme inactivation rates increased as the temperature increased, as indicated by the increment of the inactivation constant (k), and, accordingly, the time required to reduce the enzyme activity to 10% (D-value) decreased as the temperature increased (Figure 5).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.