INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Cellulose, chitin, and chitosan consist of -1,4-linked glucopyranoses, and their differences are in functional groups at the C-2 positions of their constituent sugars, i.e., the hydroxyl, acetamido, and amino groups, respectively. Chitin is one of the most abundant forms of biomass next to cellulose (9). On the other hand, chitosan, a partially or fully deacetylated form of chitin, has been found only in the cell walls of limited groups of fungi in nature (2). Chitosan has had various applications, e.g., as a carrier of immobilized enzymes and a metal-removal and cohesive reagent for purification of waste streams (27). In commercial use, chitosan is obtained by chemical deacetylation of chitin. It also has biological activities. One such activity is to elicit plant defense reactions. This defense system includes formation of fungal cell wall-degrading enzymes such as endo--1,3-glucanase and chitinase (19) and production of phytoalexin (10). The other biological activity of chitosan is growth inhibition of bacteria and fungi (15).

Chitosanases have been found in a variety of microorganisms, including bacteria and fungi (1, 6, 12, 24, 26, 30, 31). Furthermore, plant chitosanases, which also provide defensive reactions to attacks by fungal pathogens, were recently reported (5). Most purified chitosanases have been characterized as endo-type enzymes which cleave chitosans at random, and their reaction velocities are highly dependent on the degree of acetylation (D.A.) of the chitosan. On the other hand, the purification and characterization of an exo-type chitosanase called exo--D-glucosaminidase, which releases glucosamine (GlcN) continuously from the nonreducing end of the substrate, have so far been reported only for an actinomycete, Nocardia orientalis (21). Biological degradation of naturally occurring chitin in a partially deacetylated form is thought to be carried out by a two-step process (26). First, endo-type enzymes such as chitinase and chitosanase hydrolyze the chitinous material to oligosaccharides consisting of N-acetylglucosamine (GlcNAc) and GlcN. Second, the resulting oligomers are degraded completely to GlcNAc and GlcN by two exo-type enzymes, exo--D-N-acetylglucosaminidase and exo--D-glucosaminidase. However, the latter enzyme has not been studied at all except for that in N. orientalis. On the other hand, the former enzyme is distributed widely from animals to microorganisms, and its enzymological properties are well-characterized.

The genus Trichoderma, which belongs among deuteromycetes, is known as a high-cellulase producer. Trichoderma reesei secretes at least two cellobiohydrolases (exo type; EC 3.1.2.91), four endoglucanases (EC 3.1.2.4), and two -glucosidases (EC 3.1.2.20). These enzymes have already been purified or their genes have been cloned (22). Trichoderma harzianum is known as a mycoparasite and secretes multiple chitin-degrading enzymes, including endochitinase (EC 3.1.2.13), exochitinase, and exo--D-N-acetylhexosaminidase, and some of their genes have been cloned (4, 8, 11, 25). We found that T. reesei secretes multiple chitosanolytic enzymes into a culture medium under cellulase-noninducible conditions. In this paper, we describe the identification, purification, and characterization of the exo--D-glucosaminidase from the hyper-cellulolytic fungus T. reesei PC-3-7. We also discuss the catalytic mechanism of exo--D-glucosaminidase on the basis of ¹H nuclear magnetic resonance (NMR) spectroscopy of the hydrolysate. To our knowledge, this is the first report on the exo--D-glucosaminidase from eukaryotes.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Fungal strain and culture conditions. The strain used in this study was T. reesei PC-3-7, a hyper-cellulase-producing mutant that has an enhanced response to L-sorbose used as an inducer (17), and was obtained from Kyowa Hakko Kogyo Co. Ltd. (Tokyo, Japan). This strain was maintained on a potato dextrose agar (Difco) slant, and the conidia were obtained from a potato dextrose agar plate culture. For exo--D-glucosaminidase production, 10⁶ conidia were inoculated into 100 ml of a basal medium (17) containing GlcNAc (0.3%) rather than glucose as the carbon source and incubated for 72 h at 28°C with vigorous shaking (220 rpm).

Purification of enzymes. All operations were done at 4°C. The crude enzyme from 3 liters of the culture filtrate of T. reesei PC-3-7 was precipitated with ammonium sulfate (65% saturation). The precipitate was dissolved in 20 ml of 50 mM sodium acetate buffer (pH 6.0), and the solution was passed through a Bio-Gel P-6 (Bio-Rad) column (2.5 by 25 cm) previously equilibrated with the same buffer to remove remaining ammonium sulfate. The eluate containing protein fractions was applied onto a Q-Sepharose FF (Pharmacia) column (2.5 by 19 cm) which had been equilibrated with the same buffer. The column was washed with the buffer, and the eluate was obtained with a linear gradient of the buffer containing 0 to 500 mM NaCl at a flow rate of 60 ml/h. The chitosanase activity of the eluate was separated into three peaks. The last peak of chitosanolytic activity that adsorbed strongly to the column was used for further purification. The pooled chitosanase fraction was desalted and concentrated by ultrafiltration. To the sample was added ammonium sulfate to 30% saturation, and the mixture was applied onto a butyl-Sepharose FF (Pharmacia) column (2.0 by 3.5 cm) which had already been equilibrated with the same buffer containing a 30% saturation of ammonium sulfate. The column was washed with the solution, and the eluate was obtained with a linear gradient of buffer (pH 6.0) containing 30 to 0% saturated ammonium sulfate at a flow rate of 20 ml/h. The resulting active fraction was concentrated, desalted, and used as the purified enzyme preparation throughout this study.

Enzyme assay. Chitosanase activity was measured by the method of Imoto and Yagishita (16) as the concentration of reducing sugar liberated during the hydrolysis of completely deacetylated chitosan (chitosan 10B; Funakoshi, Tokyo, Japan) unless otherwise stated. Each 1 ml of the reaction mixture contained 0.5 ml of 0.2% chitosan in 50 mM sodium acetate buffer (pH 4.0) and 0.5 ml of enzyme solution. After incubation at 37°C for 30 min, the reaction was terminated by immersing the test tube in boiling water for 10 min. One unit of activity was defined as the amount of enzyme that liberated 1 µmol of reducing sugar from the substrate per min with GlcN as the standard. The same method was applied to a chitinase assay using colloidal chitin or ethylene glycol chitin as a substrate. The GlcNAc was used as the standard in chitinase assay. Exo--D-N-acetylglucosaminidase activity was measured by monitoring the release of p-nitrophenol from p-nitrophenyl--D-N-acetylglucosaminide (pNP-GlcNAc) at 430 nm. The enzyme solution (100 µl) was added to 900 µl of 1 mM pNP-GlcNAc dissolved in 50 mM sodium acetate buffer (pH 4.0). After incubation for 5 to 30 min at 37°C, the reaction was terminated by the addition of 2 ml of 1.0 M sodium carbonate (20). One unit of activity was defined as the amount of enzyme that liberated 1 µmol of pNP from the substrate per min.

Time course of hydrolysis of chitosan analogs by exo--D-glucosaminidase. Completely deacetylated chitosan (0.1%) or chitosan with a D.A. of 30% were incubated with the exo--D-glucosaminidase (17 mU) in 1 ml of 25 mM sodium acetate buffer (pH 4.0) at 37°C. The reaction was stopped at intervals by immersing the test tube in boiling water for 10 min. The mixture was centrifuged to remove insoluble materials, if necessary. The amount of reducing sugars liberated was determined as described above. For chitosan with a D.A. of 30%, the partially purified exo--D-N-acetylglucosaminidase (30 mU) was added to the reaction mixture after a 15-min incubation.

Analytical methods. Protein concentration was determined by the method of Bradford (3) with immunoglobulin as a standard. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was carried out by following the method of Laemmli (18) to examine the enzyme purity. The proteins in the gel were stained with Coomassie brilliant blue R-250. The hydrolysates of the substrates were analyzed by thin-layer chromatography (TLC) according to the method of Sakai et al. (26). Each reaction mixture (30 µl) consisted of 50 mM sodium acetate buffer (pH 4.0), 0.44 µmol of chitohexaose (GlcN₆), and 44 mU of the purified exo--D-glucosaminidase. After incubation at 37°C, the reaction was terminated by immersing the reaction tube in boiling water for 10 min. Amino sugars were detected by the ninhydrin reaction.

Chemicals. Chitosan 10B (D.A., 0%), chitosan 9B (D.A., 10%), chitosan 8B (D.A., 20%), and chitosan 7B (D.A., 30%) were obtained from Funakoshi Co., Ltd. Chitin, glycol chitosan, ethylene glycol chitin, chitobiose (GlcN₂), chitotriose (GlcN₃), chitotetraose (GlcN₄), chitopentaose (GlcN₅), GlcN₆, and N,N',N'',N''',N''''-pentaacetylchitopentaose (GlcNAc₅) were purchased from Wako Junyaku Co. Ltd. Colloidal chitin was prepared by the method of Shimahara and Takiguchi (28).

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Chitosanolytic enzyme production by T. reesei PC-3-7. Preliminary experiments to prepare chitosanase from T. reesei PC-3-7 in a liquid culture with different carbon sources showed that neither detectable growth nor chitosanase secretion was observed when chitin or chitosan (0.3%) was used as a carbon source. The fungus secreted a chitosanase into the medium containing GlcN or GlcNAc as the carbon source. The maximum chitosanase activity in the liquid medium with GlcNAc reached 1.5 U/mg of protein after a 72-h incubation, which was 3.5-fold higher than that with GlcN (data not shown).

Purification of chitosanase. When GlcNAc was used as the carbon source, the culture filtrate of T. reesei PC-3-7 contained at least three chitosanase activities. These activities were separated through Q-Sepharose FF column chromatography. We selected the largest peak of the activities adsorbed to the column for further purification by butyl-Sepharose FF column chromatography. The enzyme purification procedures are summarized in Table 1. The chitosanase was purified about 18-fold to a specific activity of 27.3 U/mg of protein against completely deacetylated chitosan (chitosan 10B). It apparently exhibited a single band by SDS-PAGE (Fig. 1). The molecular mass of the enzyme was estimated to be 93,000 Da by SDS-PAGE and about 92,000 Da by Sephacryl S-300 (Pharmacia) gel filtration, indicating that the enzyme is monomeric.

View larger version (108K): [in this window] [in a new window]   FIG. 1.   Homogeneity of purified chitosanase. Proteins containing chitosanase fractions from each purification step were detected by SDS-PAGE. Lane 1, marker proteins containing soybean trypsin inhibitor (20 kDa), carbonic anhydrase (30 kDa), ovalbumin (43 kDa), bovine serum albumin (67 kDa), and phosphorylase (94 kDa); lane 2, culture broth (5.9 µg of protein); lane 3, (NH4)2SO3 precipitate (7.7 µg of protein); lane 4, Bio-Gel P-6 column mixture (4.9 µg of protein); lane 5, Q-Sepharose column mixture (1.5 µg of protein); lane 6, butyl-Sepharose column mixture (0.5 µg of protein).

Effects of pH and temperature on activity. The 93-kDa chitosanase from T. reesei PC-3-7 had a pH optimum of 4.0 for the hydrolysis of chitosan 10B (Fig. 2). The enzyme was stable over the pH range of 6.0 to 9.0 for 1 h at 37°C. This enzyme was stable up to 30°C at pH 4.0 for a 1-h incubation, and the optimum temperature was approximately 50°C (data not shown).

View larger version (21K): [in this window] [in a new window]   FIG. 2.   Effects of pH on 93-kDa chitosanase. The chitosanase activity () was assayed at 37°C for 20 min in 50 mM acetate buffers with various pHs (3.0 to 6.0) and with chitosan 10B (0.1%) as the substrate. The residual activities of the enzyme after incubation at 37°C for 1 h at various pHs between 3.0 and 11.0 were measured. Buffers used were 50 mM citrate buffer (; pH 3.0 to 8.0) and 50 mM borate buffer (; pH 7.0 to 11.0).

Characterization of 93-kDa chitosanase. (i) Viscosimetric assay. To investigate the cleavage pattern of completely deacetylated chitosan, we first carried out a viscosimetric assay of the enzyme reaction. In the hydrolysis of chitosan 10B (0.05%), the 93-kDa chitosanase did not reduce the viscosity of the reaction mixture, while the amount of reducing sugar increased with the elapse of the reaction time (Fig. 3). On the other hand, the crude proteins from the ammonium sulfate precipitation step decreased the viscosity extensively in the early phase of the reaction, with only a small amount of reducing sugar being liberated (Fig. 3). The lack of change of the viscosity of the reaction mixture with the purified enzyme indicated that this enzyme hydrolyzed chitosan in an exo-type fashion. Furthermore, the decreased viscosity of the reaction mixture with the crude protein fraction suggests that the crude preparation possesses other chitosanolytic activities of the endo-type manner.

View larger version (25K): [in this window] [in a new window]   FIG. 3.   Relationship between reduction in viscosity of and liberation of reducing sugars from a chitosan solution with the 93-kDa chitosanase and crude enzyme. Reductions in viscosity were determined with an Ostwald viscosimeter, and amounts of reducing sugar were assayed by the method of Imoto and Yagishita (16). Relative specific viscosity is described in Materials and Methods. Twenty milliunits of the purified exo--D-glucosaminidase () or crude enzyme () was used.

(ii) Analysis of the reaction products. The hydrolysates of GlcN₆ with the 93-kDa chitosanase were analyzed by TLC (Fig. 4). GlcN₆ appeared to be hydrolyzed to GlcN₅ and GlcN at the initial stage of the reaction. The resulting chitooligosaccharides were changed to smaller chitooligosaccharides and GlcN with each successive enzymatic reaction, and the final reaction product was GlcN (Fig. 4, lane 7). When chitosan 10B was used as the substrate, it was hydrolyzed in the same manner (data not shown). These results of the viscosimetric assay and the TLC analysis of the hydrolysates suggested that the 93-kDa chitosanase may be an exo--D-glucosaminidase.

View larger version (61K): [in this window] [in a new window]   FIG. 4.   Analysis of enzymatic hydrolysates by TLC. Enzymatic hydrolysis of GlcN6 was performed in 50 mM acetate buffer (pH 4.0) at 37°C for various times. Lane S, standards containing GlcN and chitooligosaccharides from GlcN2 to GlcN6; lane 1, unhydrolyzed substrate; lanes 2, 3, 4, 5, 6, and 7, hydrolysates obtained after 2 min, 5 min, 30 min, 1 h, 10 h, and 15 h of reaction, respectively.

(iii) Determination of an anomeric form of the hydrolysate. We used 2-H resonances of GlcN and chitooligosaccharide for the determination of anomeric forms of the hydrolysate because the resonance of H₂O arising from the enzyme solution in the reaction mixture overlaid the -anomeric proton (1-H) resonances of GlcN and internal GlcN residues of oligosaccharides (data not shown). The 2-H resonances were identified from the cross-peaks in a phase-sensitive two-dimensional COSY spectrum. The assignment of the 2-H chemical shifts of GlcN₂ agreed well with that of the earlier report (33). As shown in Table 2, the 2-H resonances of the -form were divided into four classes; those of the reducing ends, those of the internal ends, and those of the terminal (nonreducing) ends of chitooligosaccharides, and those of GlcN.

View larger version (24K): [in this window] [in a new window]   FIG. 5.   Time-dependent 1H-NMR spectra in hydrolysis of GlcN6. The enzyme (320 mU) was mixed with 630 µl of 10 mM acetate buffer (pH 4.0) containing 5.2 µmol of GlcN6 and 2.9 µmol of DSS as the standard. The reaction was performed directly in an NMR tube at 30°C. The signals derived from 2-H protons were assigned by two-dimensional COSY. H2R, -form reducing-end residue of the oligomer; H2R, -form reducing-end residue of the oligomer; H2I, -form internal residue of the oligomer; H2N, -form nonreducing-end residue of the oligomer; H2M, -form of GlcN; H2M, -form of GlcN.

View larger version (23K): [in this window] [in a new window]   FIG. 6.   Time course of 2-H signals during GlcN6 degradation. The relative peak areas of the 2-H signals to the standard DSS peak were determined from the NMR spectra and plotted against reaction times. , the -form reducing-end residue of the oligomer; , the -form reducing-end residue of the oligomer; , the -form internal residue of the oligomer; , the -form nonreducing-end residue of the oligomer; , the -form of GlcN; and , the -form of GlcN.

Substrate specificity of exo--D-glucosaminidase. The data described above showed that the exo--D-glucosaminidase cleaves the GlcN-(14)-GlcN glycosidic link in a retaining fashion. The enzyme was specific for chitosan degradation. We observed no hydrolysis of various types of chitin, glycol chitosan, carboxymethyl cellulose, phosphorous-swollen cellulose, N,N'-diacetylchitobiose, and pNP-GlcNAc. Does the exo--D-glucosaminidase cleave the GlcN-(14)-GlcNAc glycosidic link? To answer this question, we performed a time course degradation of chitosan with a D.A. of 30% (chitosan 7B) with exo--D-glucosaminidase and partially purified exo--D-N-acetylglucosaminidase without any chitosanase activity against chitosan 10B (data not shown). TLC analysis showed that the exo--D-N-acetylglucosaminidase hydrolyzes GlcNAc₅ completely to GlcNAc (data not shown).

View larger version (23K): [in this window] [in a new window]   FIG. 7.   Time course of chitosan hydrolysis with exo--D-glucosaminidase. The hydrolysis pattern of completely deacetylated chitosan (chitosan 10B) or chitosan with a D.A. of 30% (chitosan 7B) with exo--D-glucosaminidase was observed. Exo--D-N-acetylglucosaminidase was added to the reaction mixture of chitosan 7B after 15 min of reaction. The arrow indicates the addition of exo--D-N-acetylglucosaminidase. , chitosan 10B with exo--D-glucosaminidase; , chitosan 7B with exo--D-glucosaminidase; , chitosan 7B with exo--D-glucosaminidase after addition of exo--D-N-acetylglucosaminidase.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

We found at least three chitosanolytic activities which were secreted into the medium by a hyper-cellulase-producing mutant strain of T. reesei PC-3-7 and purified one chitosanase to homogeneity as judged by SDS-PAGE. This enzyme was determined to be an exo-type chitosanase called exo--D-glucosaminidase with the aid of a viscosimetric assay, TLC analysis of the hydrolysate, and time-dependent ¹H-NMR spectroscopy of the enzymatic hydrolysis. The ¹H-NMR analysis using 2-H resonances of glycosides clearly showed that the exo--D-glucosaminidase possesses a retaining catalytic mechanism and also revealed that the enzyme releases continuously one GlcN residue from the nonreducing end of the substrate. Although ¹H-NMR spectroscopy with the resonance for the anomeric proton (1-H) at the C-1 position of glycosides has been used to determine the anomeric form of the glycoside produced by glycosyl hydrolase, there has been no report of the proton resonance at C-2 so far. On the basis of the ¹H-NMR data presented here, it appears that this 2-H resonance also provides an excellent means for evaluating other catalytic factors such as cleavage fashion (endo or exo type) and releasing site (reducing or nonreducing end in exo-type hydrolases) in addition to determining the anomeric form without complete substitution of D₂O for H₂O.

Several chitosanolytic enzymes have been identified in filamentous fungi (1, 6, 31). All of the enzymes were characterized as of the endo type, and their final products were chitooligosaccharides. To our knowledge, this is the first report on an exo--D-glucosaminidase from filamentous fungi. Exo--D-glucosaminidase of bacterial origin was previously purified and characterized only from an actinomycete, N. orientalis (21). These two exo--D-glucosaminidases share some properties. They are the monomeric enzyme having a molecular mass of approximately 93 to 97 kDa. They do not hydrolyze chitin, cellulose, carboxymethyl cellulose, and glycol chitosan, showing that they have a strict substrate specificity. Furthermore, these exo--D-glucosaminidases released only GlcN residues from the nonreducing end of the chitosan polymer and cleaved GlcN-(14)-GlcN and GlcN-(14)-GlcNAc bonds but not the GlcNAc-(14)-X bond.

There is a structural similarity between cellulose, chitin, and chitosan. Moreover, some chitosanases possess cellulase activity (12, 25) and some chitinases show activities against chitosan as well as chitin (32). These findings imply that a definition of these enzymes, especially the distinction between chitinase and chitosanase, based on catalytic activity is difficult. A classification of glycosyl hydrolases based on a hydrophobic cluster analysis for deduced amino acid sequences has been proposed (13, 14). According to this classification, the chitosanases belong to family 46 and are clearly distinguished from the families of chitinases or cellulases. Furthermore, Streptomyces sp. N174 chitosanase in this family has been determined to be an inverting enzyme (7). Recently, a new chitosanase gene was cloned from the phytopathogenic fungus Fusarium solani (29), which has no homology with family 46 bacterial chitosanases. This indicates that chitosanases, like cellulases and chitinases, can be classified into several families. Therefore, we are continuing the cloning and sequencing of the chitosanase gene.

In the chitosanolytic system of T. reesei, the rapid reduction of viscosity of a chitosan solution with a crude enzyme preparation allowed the presumption that it contained endo-type chitosanases. In addition to these enzymes, the culture supernatant of the fungus also contained another chitinolytic activity (unpublished data). We presume that T. reesei PC-3-7 may secrete an enzyme system which makes it possible to completely degrade chitinous polymers with a wide range of D.A. values (from 0% [chitosan] to 100% [chitin]). The physiological role of the T. reesei chitosanases, including exo--D-glucosaminidase, is still unknown. Further detailed studies of this role are required.