Thermo-Stable Xylanases from Non Conventional Yeasts

Xylanase is a commercial enzyme that has considerable applications in different types of industries, such as the feed, food, textile and paper industries, amongst others. The main goal of the present work was to select and evaluate the enzyme production from isolated wild yeasts strains obtained from several different Brazilian regions. From a total of 349 strains, two were selected, namely LEB-AAD5 and LEB-AY10, which produced fairly stable enzymes. The characterization studies showed that the strain LEB-AAD5 produced an enzyme with the following optimal range: temperature from 57.5 to 67.5 oC and pH from 4.7 to 5.5, with a half-life of 21.33 hours at 52oC and pH 5.3, Vmax of 1.77 μmol/mL.min and Km of 0.44 g/L. The enzyme from strain LEB-AY10 showed the following optimal range: pH from 4.1 to 4.8 and temperature around 80oC, with a half-life of 11.21 hours at 72oC and pH 5.3, Vmax of 5.47 μmol/mL.min and Km of 1.37 g/L. Xylanase from LEB-AY10 strain is more Thermo-stable than that of LEB-AAD5, both of which belongs to Cryptococcus sp. After optimized using two experimental designs, xylanase production increased by 600%, reaching 11.25 IU/mL under the optimal fermentation conditions (30°C, initial pH of 6.0 and 20 g/L of xylan as substrate). Thermo-Stable Xylanases from Non Conventional Yeasts


Introduction
industrial bioprocesses, xylan being the second most abundant natural polysaccharide [1,2]. It is a heteropolysaccharide consisting of a chain of β-1,4-linked D-xylanopyranose units with substitution by acetyl, xylan requires the action of several enzymes, including endo-1,4-β-D-xylanase (EC3.2.1.8), which is crucial for xylan depolymerization. Xylanases have applications in animal feed digestion, food industries and as bleaching agents in the pulp and paper industries [3]. However, the main application of xylanases is in the pulp and paper industry, where they are used in pre-treatment prior to bleaching [4]. [2]. Xylan is the most common hemicellulosic polysaccharide in the cell walls of land plants, representing up to 30%-35% of the total dry weight [5]. Xylan is the major hemicellulose in the hardwood from accounting for approximately 15%-30% and 7%-12% of their total 4-0-methylglucuronoxylan, consisting of at least 70 β-xylopyranose residues [average degree of polymerization (DP) between 150 and 200], linked by β-1,4-glycosidic bonds. Every tenth xylose residue carries a 4-0-methylglucuronic acid attached to the 2 position of xylose. Hardwood xylans are highly acetylated (e.g. the birchwood xylan contains more than 1 mol of acetic acid per 2 mol of xylose), and acetylation is more frequent at the C-3 than at the C-2 position. solubility of xylan in water and they are readily removed when the composed of arabino-4-0-methylglucuroxylans and they have a higher 4-0-methylglucuronic acid content than hardwood xylans, the 4-0-methylglucuronic acid residues being attached to the C-2 position.
The first selection was carried out according to the enzymatic ratio (diameter of the non-dyed halo/diameter of the colony) of higher than 2.5 [9]. The next selection step was performed according to the ability of the strain to produce large amounts of the enzyme in liquid cultures.
The selected yeasts were inoculated into a liquid culture medium containing: 10 g/L birchwood xylan; 5 g/L peptone; 3 g/L yeast extract; 7 g/L KH 2 PO 4 ; 2.0 g/L K 2 HPO 4 ; 0.1 g/L MgSO 4 .7H 2 O and 1.0 g/L (NH 4 ) 2 SO 4 . The pH was adjusted to 6.0. The cultures were inoculated into 500 mL flasks containing 100 mL of medium and incubated at 30ºC and 150 rpm. Samples were taken at 12 h intervals, centrifuged at 6,000 rpm for 10 min at 4ºC and the supernatants were used to estimate the xylanase activities, as described below. Two strains, LEB-AY 10 and LEB-AAD 5 , were selected and both identified as Cryptococcus sp. strains, and their enzymes were characterized.

Enzymatic assay
The xylanase (1,4-β-D-xylanase) activity was assayed by incubating the diluted enzyme extract with a solution of 1% (w/v) birchwood xylan dissolved in 50mM pH 5.3 sodium citrate buffer at 50ºC for 5 min [10]. The amount of reducing sugars liberated was determined using the 3,5-dinitrosalicylic acid method [11]. One unit of xylanase activity (IU) was defined as the amount of enzyme necessary to produce 1 µmol of reducing sugars (D-xylose) from xylan per min at 50ºC.

Optimal activity conditions
First experimental design: A central composite design (CCD) with k = 2, was used to generate 11 treatment combinations, with the pH and temperature as the independent variables. Five levels were chosen for each variable, the upper and lower limits being set according to the literature. In the statistical model, Y denotes units of xylanase activity. Table 1 shows the real levels corresponding to the coded settings, the treatment combinations and the responses of the microorganisms LEB-AY 10 and LEB-AAD 5 . Each design is represented by a secondorder polynomial regression model, Eq. (1), to generate the surface plots: The test factors were coded according to the following regression equation: x i = (x i -x 0 )/ΔX i . Where x i is the 'coded' value and X i is the real value of the i th independent variable, X 0 is the real value at the center point, and ΔX i is the step change value. The Statistical software 5.0 [12] was used for the regression and for the graphical analysis of the data. The significance of the regression coefficients was determined using the Student t-test, and the second-order model equation was determined by the Fisher test. The variance explained by the model was given by the multiple coefficient of determination, R 2 .
i. Thermal and pH stability: The temperature and pH stability of the enzyme extract were determined by incubation at different values. Samples were taken at different intervals, depending on the temperature and pH used, and the residual enzyme activity measured.
ii. Kinetics determination: The initial reaction rates were determined at different substrate concentrations ranging from 0.5 to 30 mg of birchwood xylan/ml of 50mM pH 5.3 sodium citrate buffer at 50ºC. The kinetic constants K m and V max were estimated according to the method of Lineweaver and Burk [13].
iii. Optimizing the xylanase production: The plackett and burman design with the aim of optimizing xylanase production by the LEB-AY 10 strain, under adequate medium and fermentation conditions, a Plackett & Burman (PB) design with 12 assays was applied. The independent variables and levels can be seen in Table 2. Results represent the mean of three experiments. X 1 = T(ºC), X 2 = pH, code settings are X 1 = (temperature -65)/10, X 2 = (pH -4.8)/1.0. Xylan (g/L) 5 10 15 Yeast Extract (g/L) 1 2 3 Temperature (°C) 25 30 35 Results represent the mean of three experiments. X 1 = xylan (g/L), X 2 = T(ºC), code settings are X 1 = (xylan -20)/7, X 2 = (temperature -30)/3.5. Four central points were added to the 12 PB assays, giving a total of 16 assays. The effect of each variable and error were calculated using Statistica (Statsoft Corporation, USA) software, the enzymatic activity being the main response (dependent variable).

Second experimental design:
The PB design indicated the variables that were significant in the enzyme production, and these were assayed using the CCD and the optimal xylanase production conditions obtained from the response surface generated by the mathematical model. The independent variables and levels can be seen in Table 3.

Culture selection and growth conditions
The yeast strains were evaluated in Petri plates, however only 9 of the 349 tested showed an enzymatic ratio above 2.5, these strains being shown in Table 4. These strains were assayed for their enzyme production capacity in submerged medium, and the strains with better characteristics in terms of enzyme activity and stability, LEB-AAD 5 and LEB-AY 10 , were chosen for the following assays. Figure 1 shows the enzyme production of the microorganism LEB-AY 10 , also showing its pH behavior. The production started after practically 24 hours of fermentation, with a peak of activity at 36 hours followed by a slight decrease, although the activity can be considered similar due to the standard deviations. The enzymatic activity was about 0.67 IU/mL up to the end of fermentation, and showed good stability with respect to the effects of both pH and temperature. Figure 2 shows the pH behavior and enzyme production of the LEB-AAD 5 strain, the highest enzymatic activity of 0.73 IU/mL being produced after 60 hours of fermentation, then decreasing to about 0.50 IU/mL at the end.

Optimization of the conditions for the enzyme activities
The temperature and pH were optimized using factorial design methodology, and the results for the first experimental design are shown in Table 1. The optimal temperature and pH conditions for this enzyme were around 65ºC and 4.8, respectively (assays 9, 10 and 11). The data from Table 1 were used to obtain the coded models for the LEB-AAD 5 and LEB-AY 10 strains, equations (2) and (3), and the ANOVA (Table 5) confirmed that the models described the enzymatic activity well as a function of pH and temperature, since the values for F calculated were higher than those for F listed for both models, with R 2 values higher than 0.88. The response surfaces were built using equations (2) and (3) and are represented in Figure 3 and 4.
According to the analysis of the response surfaces and contour plots, it can be seen in Figure 3 that the highest enzymatic activity was achieved for the LEB-AAD 5 strain under the center point conditions, that is at a temperature of 65ºC and a pH of 4.8, whereas for the enzyme from LEB-AY 10 , the highest activity was achieved at a pH around the center point (4.8), but at a temperature as high as 80ºC, different from the enzyme produced by the acidophilic fungus Bispora sp. MEY-1 and reported by Luo et al. [14], which showed its optimal conditions at 65ºC and pH 2.6.     In order to confirm the effects of temperature on the LEB-AY 10 enzyme, additional assays were carried out with the pH set at 4.8 and the temperature ranging from 50 to 85ºC, and the results are expressed in Figure 5.

Strains
As can be seen in Figure 5, an increase in temperature also increased the enzyme activity, 80ºC being the optimal temperature for this enzyme, equivalent to the conditions described for Bacillus circulans BL53 in solid-state cultivation [15]. Thus the optimal conditions for the LEB-AY 10 xylanase were a temperature of 80ºC and a pH value of 4.8.

Thermal and pH stability
The thermal stability of xylanase was studied from the enzyme halflife. Enzymes from both the LEB-AAD 5 and the LEB-AY 10 strains were incubated at different temperatures and the activities measured after pre-defined time intervals. Based on this data, the half-lives (t 1/2 ) were determined as a function of temperature.
For the xylanase from the LEB-AAD 5 strain, the half-lives were determined at 52, 57, 60, 65 and 70ºC while for that from the LEB-AY 10 strain, they were determined at 70, 72, 75, 77, 80 and 85ºC. These temperatures were chosen according to the optimal temperature of each enzyme, and thus the ranges were different for each one. The results for both xylanases are shown in Figure 6, where it can be seen that the enzyme produced by LEB-AY 10 was considerably more stable.
The enzyme produced by the LEB-AY 10 strain can be considered as significantly more thermostable as compared to others found in the literature, if the values of t 1/2 are compared. Moreover, this enzyme maintained 75% of its initial enzyme activity after 5 minutes of incubation at 80ºC. On the other hand, the enzyme from Penicillium capsulatum, reported by Ryan et al. [5], lost its activity in less than 3 min at 75ºC and the enzyme from Aspergillus niveus, reported by Sudan and Bajaj [16], lost 90% of its activity after 5minutes incubation at 70ºC. However, the thermostability close to the optimal activity temperature was quite poor, a common behavior for enzymes, so they should be used at temperatures below the optimal ones, as stated by Santos et al. [17].
With respect to pH stability, a set of experiments was carried out with this proposal, incubating the enzymes in citrate buffer solution (50mM) at pH values from 3.5 to 6.0. The temperature was maintained at 52ºC for the LEB-AAD 5 strain enzyme, and at 72ºC for the LEB-AY 10 strain one. Samples were collected and their half-lives evaluated, and the data are shown in Figure 7.

LEB-AAD5
Temperature (  As can be seen in Figure 7, the variations in the half-lives of the two enzymes with pH, are significantly different. On the other hand, it can be seen that both enzymes show good stability at an equivalent range of pH values between 4.5 and 5.3, this range being close to that of optimal activity for these enzymes. Therefore, different from the effect of temperature, they can be used close to the pH for optimal activity, with practically no effect on their stability.

Kinetic parameters
Enzyme extracts for each strain were used to determine the kinetic parameters (K m and V max ), with the substrate concentration ranging from 0.5 to 30 g/L of xylan. The plots for reaction rate versus substrate concentration show that the enzymes followed the Michaelis-Menten kinetic model, with K m and V max being obtained using the linearization procedure of Lineweaver and Burk [13] (Figures 8 and 9).
The enzyme produced by the LEB-AAD 5 strain showed a K m of 0.379 g/L and a V max of 1.73 µmol/mL.min, whereas the values for the enzyme from the LEB-AY 10 strain were, respectively, 0.970 g/L and 5.0 µmol/mL.min. Sudan and Bajaj [16] observed a K m of 2.5 g/L and V max of 26 µmol/mg.min for the xylanase produced by Aspergillus niveus, while Bakir et al. [18] reported K m and V max values of 18.5 g/L and 90 µmol/mg.min, respectively, for the xylanase from Rhizopus oryzae. Both papers indicated much higher values for the K m and V max than that found in the present work.

Optimization of enzyme production
Plackett and burman design: Since the enzyme from the LEB-AY 10 strain was shown to be more stable and its production was also higher than that from the LEB-AAD 5 strain, two experimental designs were carried out in order to optimize its production, using the pre-defined medium. Table 6 shows the codified values for the PB design and the enzyme activity after 24h of fermentation.
The PB design indicated that the variables of temperature, xylan concentration and (NH 4 ) 2 SO 4 were statistically significant at 90% of confidence, so these were considered in the next CCD, except for (NH 4 ) 2 SO 4 , which showed negative effects and was therefore maintained at 1 g/L.

Second experimental design:
For the CCD, all the concentrations of the non-statistically significant components were fixed according to level 0 of the previous PB design, while the temperature and xylan concentration varied according to Table 3.
The data from the second experimental design were used to obtain     the coded model for enzymatic activity as a function of the xylan concentration (g/L) and temperature (ºC), as expressed by equation (4). The ANOVA (Table 7) confirmed that the models described the enzymatic activity well, since F calculated was higher than F listed , with a value for R 2 higher than 0.98. Thus a response surface was built according to equation (4) as represented in Figure 10.
It can be seen from the surface and the contour plots that higher values of enzymatic activity can be achieved under the center point conditions. Therefore the results from the second experiment determined the optimal conditions for maximum the enzyme activity, namely 20g/L xylan and a temperature of 30ºC (assays 9, 10 and 11). Considering the results of the previous PB, the final optimal conditions for all the variables are expressed in Table 8.

Conclusions
In the first part of this work, the strains LEB-AAD 5 and LEB-AY 10 , both identified as Cryptococcus sp. were selected from a total of 349 strains as producers of fairly stable enzymes. The enzymes were characterized and that produced by the LEB-AY 10 strain showed optimal temperature and pH values of around 80ºC and 4.5, respectively, a half-life of 11.21 hours at 72ºC and pH 5.3, a V max of 5.47 µmol/mL.min and a K m of 1.37 g/L. Since this enzyme was the more stable of the two assayed, its production was optimized in flask fermentations using two experimental designs, resulting in an increase in enzyme activity of 600%, reaching 11.25 IU/mL under the optimized fermentation conditions.  Temperature (°C) 30 Table 8: Optimized process variables for the production of xylanase by the LEB-AY 10 strain.