Puri ﬁ cation and characterization of xylanases from Trichoderma inhamatum

Background: Two xylanases, Xyl I and Xyl II, were puri ﬁ ed from the crude extracellular extract of a Trichoderma inhamatum strain cultivated in liquid medium with oat spelts xylan. Results: Themolecular masses ofthe puri ﬁ ed enzymes estimated by SDS-PAGE and gel ﬁ ltration were, respectively, 19and14kDaforXylIand21and14.6kDaforXylII.Theenzymesareglycoproteinswithoptimumactivityat50°C enzymes.Thexylanases 462.2 protein (Xyl I) 10.7, 4.0 mg·mL 4553.7 1972.7 -1 (Xyl II) on oat spelts and birchwood xylan, respectively. The hydrolysis of oat spelts xylan released xylobiose, xylotriose, xylotetrose and larger xylooligosaccharides. Conclusions: The enzymes present potential for application in industrial processes that require activity in acid conditions, wide-ranging pH stability, such as for animal feed, or juice and wine industries.


Introduction
Xylans are a diverse and complex group of polysaccharides with the common feature of a backbone of β-(1 → 4)-linked xylose residues [1], in which side-chains are attached at the C2 and C3 positions of D-xylosyl. These substituents can normally be acetyl, 4-0-methyl-D-glucuronosyl or L-arabinosyl groups [2,3]. Xylan is the predominant hemicellulose found in plant cell walls strongly associated to cellulose microfibrils and the strength of this interaction is inversely related to the degree of substitution of the main chain by side-groups [4].
The conversion of xylan into useful products represents a significant part of the effort to achieve economical viability of the lignocellulose biomass processing and to develop different ways to produce chemicals and renewable energy as well. Owing to its complex structure, the complete degradation of xylan requires the joint of several hydrolytic enzymes acting in synergy that is known as a xylanolytic system. Endo-β-1,4-xylanases (β-1,4-D-xylan xylanohydrolases, EC 3.2.1.8) are the most important xylanolytic enzymes, cleaving internal glycoside bonds in xylan backbone, reducing the degree of polymerization of the polymer [5,6]. The cleavage carried out by these enzymes is not aleatory, i.e. side chain decorations in xylan are recognized by xylanases, and the degree of substitution in the polymer influences the product of hydrolysis. This difference in substrate specificity among different xylanases has important implications in the deconstruction of xylan [3].
Interest in xylanolytic enzymes has increased in last decades due to their industrial applications in the food, feed, and pharmaceutical industries and for sustainable production of fuels and chemicals. Besides, they can be applied in some processes in which cellulolytic activity must be absent, to preserve the vegetal fibers, in the pulp and paper industries, and in the processing of flax, hemp and jute in the textile industries [7,8,9].
Fungi are commonly used as source of xylanases and their xylanolytic systems have been widely studied [6,10]. Trichoderma spp. xylanases are among the most known enzymes, therefore, this fungal genus is suited for further examination of function and application of these enzymes [11]. Trichoderma reesei Rut C-30 is the most well-known Trichoderma strain producing several xylanases and cellulases with different biochemical properties and specificities for substrates, as predicted by genome sequence [12], and also many enzyme preparations obtained from the large-scale cultivation of this fungus have been commercialized. Recently, rational and efficient systems for the production of cocktails containing different balances between xylanolytic and cellulolytic enzymes have been investigated for the different application of these enzymes [13]. Xylanases from other Trichoderma species have also been studied as those from Trichoderma harzianum, Trichoderma lignorum, Trichoderma longibrachiatum, Trichoderma koningii, Trichoderma pseudokoningii and Trichoderma viride [14,15]. Other xylanases, including that from a psychrotrophic Trichoderma strain have been purified and characterized [16], and also xylanase encoding genes from other Trichoderma species have been isolated, cloned and expressed in Escherichia coli [17], Saccharomyces cerevisiae [18] and Pichia pastoris [19].
When a new efficient xylanase-producing microorganism is isolated, it is essential to purify and characterize the enzymes to know the action towards substrates of each component of a xylanolytic complex, its regulation and biochemical properties in order to develop more competitive processes. A Trichoderma inhamatum strain, isolated from soil in São Paulo state (Brazil), produces high levels of xylanase in the absence of cellulases [20], an important condition for some industrial applications, as stated above. The influence of some parameters affecting xylanase production by this fungal strain has already been investigated [21]. The present study aimed to purify and characterize the main xylanases produced by this fungus under previously optimized culture conditions.

Materials and methods
2.1. Fungal strain: maintenance and culture conditions T. inhamatum was deposited in the Environmental Studies Center Collection, CEA/UNESP, Brazil. The fungal strain was maintained on Vogel solid medium slants [22] with 1.5% (w/v) wheat bran, at 4°C and cultured periodically. The cultures were inoculated in the same medium with 1.5% (w/v) glucose and incubated for conidia production during 7 days at 28°C. Conidia were harvested, suspended in sterile distilled water and the concentration of the suspension was adjusted to 10 7 conidia milliliter -1 . One milliliter of this suspension was inoculated into Vogel liquid medium pH 6.0 with 1% (w/v) oat spelts xylan as sole carbon source for enzyme production. Cultures were maintained in orbital shaker (120 rpm) at 25°C for 60 h [21]. After cultivation, mycelium was removed by vacuum filtration and the crude culture filtrate was used as source of enzymes.

Enzyme assay
Xylanase activity was determined by incubating enzyme samples with 1% (w/v) birchwood xylan (Sigma, USA; xylose residues ≥90%) in 0.05 M sodium acetate buffer pH 5.5 at 50°C. At suitable intervals, the reaction was interrupted with 3,5-dinitrosalicylic acid (DNS) reagent and the released reducing sugars were measured [23], using xylose as standard. One unit of activity (U) was defined as the amount of enzyme capable to release 1 μmol of reducing groups per min. Specific activity was expressed by the relation between enzyme activity and protein content.

Determination of protein and carbohydrate
Protein was determined by the Lowry method [24], with bovine serum albumin as standard. During the chromatographic steps, proteins were detected by reading absorbance at 280 nm. Total carbohydrate was determined by the phenol-sulphuric acid method [25], with glucose as standard.

Purification of xylanases
The crude culture filtrate (200 mL) was dialyzed against 0.05 M Tris-HCl buffer pH 7.0 for 8 h, with buffer changes each 2 h, in order to exclude small molecules and obtain a buffered solution in this pH. This sample was submitted to ion exchange chromatography on a DEAE Sephadex A-50 column (1.1 × 14.5 cm), previously equilibrated with the same dialysis buffer; at 50 mL/h flow rate. Adsorbed proteins were then eluted with a 0.0-0.5 M NaCl linear gradient in the same buffer. Proteins were detected by reading absorbance at 280 nm and xylanase activity was assayed in the collected 3 mL fractions. Fractions with high xylanase activity were pooled and submitted to electrophoresis (SDS-PAGE). The sample corresponding to the not retained fractions was further dialyzed against 0.05 M ammonium acetate buffer pH 6.8 for 8 h, with buffer changes each 2 h, and then lyophilized. The sample was re-dissolved in a small volume of this buffer and applied to size exclusion chromatography on a Sephadex G-75 (1.8 × 60.5 cm) column equilibrated and eluted in the same buffer, at 19 mL/h flow rate. Proteins were detected by reading absorbance at 280 nm and xylanase activity was assayed in the collected 3 mL fractions. Fractions with high xylanase activity were pooled and submitted to electrophoresis (SDS-PAGE). All purification steps were carried out at 4°C.

Determination of molecular mass under non-denaturing conditions
Molecular masses of the proteins were estimated by gel filtration from a regression curve by plotting log of the molecular masses of the standards against the ratio between their elution volumes and the void volume (V 0 ), estimated using blue dextran. Standard proteins (Sigma, USA) were ribonuclease (15.4 kDa), chymotrypsin (25.0 kDa), ovalbumin (43.0 kDa) and bovine serum albumin (67.0 kDa).

Temperature and pH optima, stability in different temperature and pH
The best pH for activity of the purified xylanases was determined by assaying enzymatic reactions in McIlvaine buffer adjusted to various pH between 3.0 and 8.0, with 0.5 unit intervals at 50°C. For the optimal temperature, enzymatic reactions were carried out with the purified enzymes in 0.05 M sodium acetate buffer pH 5.5 at temperatures from 20 to 60°C, with 5°C intervals.
Thermal stability was determined by verifying residual activity after incubating samples of the purified enzymes without substrate at 40, 50 and 60°C during different periods. The pH stability was determined by verifying the remaining activity after incubating the purified enzymes for 24 h at room temperature. Diluted (1:2; v/v) enzyme samples were incubated with McIlvaine buffer in the pH range from 2.5 to 8.0, with 0.5 unit intervals.

Effect of substances on enzyme activity
The effect of substances was verified by assaying xylanase activity with a variety of compounds dissolved in 0.05 M sodium acetate buffer pH 5.5. The following substances were evaluated at 2 and 10 mM: lead acetate, ammonium chloride, barium chloride, calcium chloride, cobalt chloride, copper chloride, mercury chloride, iron sulfate, magnesium sulfate, manganese sulfate, zinc sulfate, tetrasodium ethylenediaminetetraacetate (EDTA), 1,4-dithiothreitol (DTT), sodium dodecyl sulfate (SDS) and phenylmethylsulfonyl fluoride (PMSF). All assays were carried out in triplicate.

Substrate specificity
The specificity of the xylanases was verified by assaying the activity against different substrates. Xylanase activity was measured on birchwood and oat spelts xylans. Endoglucanase (CMCase) and exoglucanase (Avicelase) activities were assayed in a reaction mixture with carboxymethylcellulose (CMC) and microcrystalline cellulose (Avicel), respectively. Substrates at 1% (w/v) were prepared in 0.05 M sodium acetate buffer pH 5.5 and appropriately diluted enzyme solution. Reducing sugars were quantified with the DNS acid reagent and the absorbance was measured at 540 nm. One unit of activity was defined as the amount of enzyme required to release 1 μmol of product equivalent per min in the assay conditions at 50°C.

Kinetic parameters
The enzymes were incubated with xylan from birchwood and from oat spelts (Sigma, USA; arabinose residues ≤ 10%, glucose residues ≤ 15%, xylose residues ≥ 70%) at concentrations varying from 4.0 to 30.0 mg·mL -1 . The Michaelis-Menten constant (K m ) and maximal velocity (V max ) were estimated from Lineweaver-Burk reciprocal plots [27]. Three experiments were carried out for each substrate concentration in triplicate and the straight line plotted was calculated by linear regression (Microsoft Office Excel, version 12.0) using mean values obtained from each experimental point.

Determination of hydrolysis products
The products of the enzymatic hydrolysis of oat spelts xylan were analyzed by thin-layer chromatography (TLC) on silica-gel G-60 plates (10 × 15 cm), using xylose and xylobiose solutions (1 mg/mL) as standards [28]. The mobile phase was ethyl acetate/acetic acid/formic Table 1 Purification of the xylanases from T. inhamatum.

Purification of T. inhamatum xylanases
Two xylanases were purified from the crude filtrate after growth of T. inhamatum in liquid cultures with xylan, under optimized culture conditions. Xyl II was purified to electrophoretic homogeneity by a single step of ion exchange chromatography, while Xyl I purification required a subsequent molecular exclusion chromatography with Sephadex G-75. The first step (Fig. 1) revealed two protein peaks with xylanolytic activity: Xyl I, corresponding to the not retained fraction, presented 62.7% of the activity, and Xyl II, the retained fraction was eluted with a NaCl gradient and presented 3.7% of the activity.
The sample corresponding to the first xylanase peak, considered the main xylanase produced by T. inhamatum, was subsequently subjected to molecular-exclusion chromatography (Fig. 2), giving rise to two protein peaks, one of them with xylanase activity, representing 12.0% of the initial activity. Xyl I and Xyl II from T. inhamatum were obtained with final specific activities of 3464.2 and 1216.4 U·mg prot -1 and 5.3 and 1.9-fold purification, respectively ( Table 1).
The occurrence of multiple enzyme forms is a common phenomenon and can be considered a specialized function of microorganisms to achieve a more effective hydrolysis of heterogeneous substrates in the nature [2]. Many xylanase forms are produced by T. reesei Rut C-30, for example. The two major xylanases produced by that fungus are Xyn1 and Xyn2, with the latter representing more than 50% of the total xylanolytic activity and both of them are responsible for more than 90% of the specific xylanase activity produced when the fungi was grown on cellulose or xylan [29]. Besides, the presence of several minor xylanases has also been demonstrated [30].

Physico-chemical properties of T. inhamatum xylanases
SDS-PAGE revealed a single band in each sample after the purification steps. The molecular masses estimated by this method were 19 kDa for Xyl I and 21 kDa for Xyl II (Fig. 3). The molecular masses estimated under non-denaturing conditions with Sephadex G-75 were 14.0 and 14.6 kDa for Xyl I and Xyl II, respectively. These values of molecular mass are similar to those observed for many Trichoderma xylanases [14]. The purified xylanases were highly glycosylated, presenting 79% and 62% of carbohydrates for Xyl I and Xyl II, respectively, which justifies some distortion observed in SDS-PAGE.
The activity profiles of Xyl I and Xyl II (Fig. 4a) showed that both enzymes present activity in slightly acid region with optimal activity in pH 5.0-5.5. Both enzymes showed more than 50% of the maximum activity in the pH range from 4.5 to 6.5, and the activity decreased sharply from this range. Optimal activity was observed at 50°C for Xyl I and at 45°C-55°C for Xyl II (Fig. 4b), the former presented more than 50% of the activity between 40°C and 55°C, and the latter between 35°C-60°C. Comparatively, the optima pH and temperature observed for the T. inhamatum xylanases are in accordance with those from many other xylanases from mesophilic Trichoderma strains that are commonly observed in the pH range from 3.5 to 6.0 and in temperatures from 45 to 60°C [14].
Both enzymes were stable at 40°C (Fig. 5), i.e. Xyl I retained 71% and Xyl II retained 77.5% of the activity after 4.5 h of incubation. Xyl I exhibited half-life of 4 min at 50°C and of 40 s at 60°C, while the half-life of Xyl II was 18 min at 50°C and 46 s at 60°C. The purified Trichoderma sp. K9301 xylanase was very stable at 50°C but also rapidly lost activity when incubated at 60°C [15]. The purified xylanases exhibited distinct stabilities in pH from 3.0 to 8.0 (Fig. 6). After 24 h incubation, residual activity of 90% or more was detected in pH from 4.5 to 6.5 for Xyl I, and from 4.0 to 8.0 for Xyl II.
The effect of organic compounds, metallic ions and a chelating agent on the activity of the purified xylanases from T. inhamatum is presented in Table 2. The ion Hg 2+ was a strong inhibitor of the xylanases even at 2 mM, while Cu 2+ also inhibited both enzymes, but this effect was more pronounced only at 10 mM. The inhibition by Hg 2+ seems to be a general property of xylanases, indicating the presence of cysteine thiol groups near or in the active site of the enzyme [31]. The denaturation caused by the detergent SDS resulted in loss of the activity and, in low concentration (2 mM), the effect was more pronounced on Xyl I. The    2+ and Fe 2+ and moderately inhibited by Zn 2+ only at 10 mM·Pb 2+ inhibited the enzymes with more pronounced effect on the Xyl II at 10 mM. EDTA at both concentration and PMSF at 10 mM slightly decreased the activity of both enzymes. The effect of EDTA on both enzymes suggests that they may require divalent ion for catalysis. DTT increased the activity of both enzymes, which can be explained by the prevention of the oxidation of sulfhydryl groups in the presence of this agent or by the reduction of disulfide bridges, restoring their native structure in some specific region or even of the catalytic site.

Substrate specificity and kinetic parameters
The purified enzymes hydrolyzed exclusively xylans with no activity on Avicel or CMC. The activities of Xyl I and Xyl II on oat spelts xylan, a branched arabinoxylan were, respectively, 8% and 16% higher than those on birchwood xylan, which is as less branched xylan with 94% xylose residues [32], indicating the preference of the enzymes for branched and heterogeneous xylan. Both enzymes exhibited Michaelis-Menten kinetics and the corresponding apparent constant values were calculated ( Table 3). The K m values indicated that both enzymes had higher affinity for birchwood than for oat spelts xylan.
With birchwood xylan as substrate, Xyl I presented the lower K m and therefore higher affinity, while with oat spelts xylan, Xyl II had more affinity. For both substrates, the higher value of k cat was that of Xyl II; for both enzymes, higher values of k cat and V max were observed with oat spelts xylan. The ratio k cat /K m for birchwood xylan was 19.9 and 21.7 s -1 ·mM -1 for Xyl I and Xyl II respectively, demonstrating similar efficiencies to hydrolyze this substrate. However, with oat spelts xylan, this ratio was about 2.5-fold greater for Xyl II than for Xyl I, indicating that the former enzyme is much more efficient in degrading this xylan. Two xylanases from Aspergillus giganteus [33] have similar catalytic efficiencies on these substrates, while the Penicillum capsulatum xylanase is 2.7-fold more efficient in hydrolyze oat spelts than birchwood xylan [34].
The mode of action was investigated by identifying the main products of oat spelts xylan hydrolysis by the purified xylanases. Since both enzymes gave products with the same R f values over various incubation intervals, only the TLC profile obtained with Xyl I is shown in Fig. 7. Both enzymes released xylobiose and larger xylooligosaccharides and, thus, they may be classified as endoxylanases. Interestingly, not even after 17 h, release of xylose was verified, in contrast to the xylose release observed after long-term incubation [16]. According to the Table 2 Effect of different substances on Xyl I and Xyl II from T. inhamatum.  relative mobility to xylose [28], the two spots with lower mobility than xylobiose corresponded to xylotriose and xylotetrose.

Conclusions
This manuscript presents the first report about the purification and properties of two xylanases from T. inhamatum by a simple and inexpensive procedure. These enzymes were stable over a wide range of pH and the optimal conditions for their activities were around 50°C and pH 5.0 very similar to each other and also to the characteristics observed for the crude enzyme [21]. These two xylanases appear to be differentially modified products from the same gene because they have similar hydrolytic and physico-chemical properties, and differential glycosylation may explain the differences in molecular masses, the capacity to bind to DEAE-anion exchanger and thermal stabilities. The glycosylation explains some cases, but do not completely elucidate the functional and genetic basis for the multiplicity of these enzymes. Furthermore, a comparison of amino acid compositions indicates that three xylanases from Trichoderma harzianum E58 are products of distinct genes [35]. The results indicate possible employment of such enzymes in some industrial processes, which require activity in acid conditions, wide-ranging pH stability, such as for animal feed, or juice and wine industries.