Non-specific depolymerization of chitosan by pronase and characterization of the resultant products Acharya B. Vishu Kumar 1 , Lalitha R. Gowda 2 and Rudrapatnam N. Tharanathan 1 1 Department of Biochemistry and Nutrition, 2 Department of Protein Chemistry and Technology, Central Food Technological Research Institute, Mysore, India Pronase (type XXV serine protease from Streptomyces griseus) efficiently depolymerizes chitosan, a linear bfi1,4- linked polysaccharide of 2-amino-deoxyglucose and 2-amino-2-N-acetylamino- D -glucose, to low-molecular weight chitosans (LMWC), chito-oligomers (degree of poly- merization, 2–6) and monomer. The maximum depolymeri- zation occurred at pH 3.5 and 37 °C, and the reaction obeyed Michaelis–Menten kinetics with a K m of 5.21 mgÆmL )1 and V max of 138.55 nmolesÆmin )1 Æmg )1 .Themolecularmassof the major product, LMWC, varied between 9.0 ± 0.5 kDa depending on the reaction time. Scanning electron microscopy of LMWC showed an approximately eightfold decrease in particle size and characterization by infrared spectroscopy, circular dichroism, X-ray diffractometry and 13 C-NMR revealed them to possess a lower degree of acety- lation, hydration and crystallinity compared to chitosan. Chitosanolysis by pronase is an alternative and inexpensive methodtoproduceavariety ofchitosandegradationproducts that have wide and varied biofunctionalities. Keywords: chitosan; chito-oligomers; low-molecular weight chitosan; pronase; structure. Chitosan is the de-N-acetylated derivative of chitin, a linear polysaccharide of b1fi4-linked 2-deoxy-2-acetamido- D -glu- cose units [1], that constitutes the exoskeleton of inverte- brates and is one of the components of the cell walls of fungi. It has wide and varied applications in medicine, agriculture, pharmaceuticals and the food industry [1,2], which is attributed to the biofunctionality of the amine moiety that confers both cationic (polyelectrolyte) and chelating properties. Despite being biocompatible, nontoxic and multifunctional, the use of chitosan in vivo is hampered by its high-molecular mass and high viscosity even at low concentrations [2]. Therefore, a prerequisite for efficient utilization of chitosan is its depolymerization to low- molecular weight chitosans (LMWC), chito-oligomers and monomer. The depolymerized products find additional applications as hypo-cholesterolemic, antitumorigenic, antimicrobial, immuno-enhancing agents, and also in the treatment of osteoarthritis, gastritis, etc. [3–5]. A 9 kDa LMWC suppressed Escherichia coli activity whereas that of LMWC 5 kDa showed antihyperlipemic and hypocholes- terolemic effects [2]. Chito-oligomers with a degree of polymerization (DP) > 6 showed antitumor activity towards Sarcoma-180 and Meth-A tumors, and caused activation of defence responses in plants. Chitotriose exhibited maximum inhibitory effect towards angiotensin converting enzyme (ACE) [4]. Chitosan can be depolymerized by acid or enzymatic hydrolysis. The former is harsh, time consuming, modifies the products and forms a large quantity of monomers [6]. In contrast, enzymatic hydrolysis produces specific products as the reaction can be precisely controlled. Chitosanase, the enzyme of choice due to its specificity, degrades chitosan to chito-oligomers, but is, however, very expensive and unavailable in bulk for commercial exploitation [7]. The earlier concept of a ratio of one enzyme to one substrate/group of related substrates is no more a reality in most of hydrolases as evidenced in recent literature. b1fi4 Glucanase, although specific for b1fi4-linked glucans, can hydrolyze mannans and cellobiose [8]. Chitosanase from Myxobacter A-1, Streptomyces griseus HUT 6037, Bacillus sp.7-M and Bacillus megaterium degrades carboxymethyl- cellulose. A b1fi3/1fi4 glucanase from Bacillus circulans WL-12 depolymerizes chitosan to low-molecular mass products [9–12]. Susceptibility of chitosan to various nonspecific enzymes like wheat germ lipase, lysozyme, papain, cellulase, hemicellulase, b-glucosidase, etc., has been reported earlier, although enzyme purity was in doubt and could be contaminated with chitosanase [13,14]. Recently, we have shown that a homogeneous isozyme of Aspergillus niger pectinase could depolymerize chitosan quantitatively to yield LMWC and chito-oligomers [15]. The objective of the present study was to demonstrate yet another example of enzyme nonspecificity in catalyzing the cleavage of completely unrelated substrates. An electropho- retically pure pronase preparation was used to depolymerize chitosan to LMWC, chito-oligomers and monomer, and structural characterization of the depolymerized products is presented herein. Correspondence to R. N. Tharanathan, Department of Biochemistry and Nutrition, Central Food Technological Research Institute, Mysore ) 570 013, India. Fax: + 91 821 2517233, Tel.: + 91 821 2514876, E-mail: tharanathan@yahoo.co.uk Abbreviations: DP, degree of polymerization; LMWC, low-molecular weight chitosan. Enzyme: Pronase (type XXV serine protease from Streptomyces griseus; EC 3.4.24.31) (Received 30 October 2003, revised 12 December 2003, accepted 22 December 2003) Eur. J. Biochem. 271, 713–723 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2003.03975.x Materials and methods Materials Chitosans (molecular mass, 150–600 kDa) and phenyl- methylsulfonyl fluoride were obtained from Fluka Chemika (Buchs, Switzerland). Pronase (type XXV protease from Streptomyces griseus, EC 3.4.24.31), Sephadex G15 and Celite were obtained from Sigma. Shrimp chitin was obtained from the CFTRI Regional Center at Mangalore, India. Other chemicals used were of highest purity available. Preparation of chitosan Shrimp chitin was subjected to heterogeneous N-deacety- lation to obtain chitosan [16], which was further purified by dissolving in 1% acetic acid, filtered through a plug of glass-wool to remove suspended particles and finally precipitated with 2% sodium carbonate. The precipitate was water-washed, freeze-dried and stored at ambient temperature until further use. Criteria of enzyme homogeneity To establish the purity, the stock pronase solution (10 mgÆmL )1 ) was subjected to capillary electrophoresis (Prince 550 system, Prince Technologies, Emmen, the Netherlands) using fused silica capillary (80 cm length · 75 lm i.d.) connected to an UV detector (280 nm) and Tris/glycine buffer (pH 8.8) at 10 kV, 100 mbar, 26 ± 1 °C. The data acquisition and control were performed using DAX software. The enzyme solution after suitable dilution (30 lg) was subjected to SDS/ PAGE (12.5% T, 2.7% C using Tris/glycine buffer of pH 8.8) according to the method of Laemmli followed by Coomassie Brilliant Blue staining [17]. For further assay, the stock solution was suitably diluted. Enzyme assay Proteolytic activity of pronase was evaluated using casein (1%) as the substrate at optimum conditions (pH 7.5 and 37 °C) and estimating the trichloroacetic acid soluble peptides released (specific activity, unit ¼ absorbance at 280 nm/reaction time · mg protein in the reaction mix- ture). For chitosanolytic activity, 1% chitosan in 1% aqueous acetic acid was treated with pronase in the ratio 100 : 1 (w/w), at optimum conditions (pH 3.5 and 37 °C). To terminate the reaction, the mixture was heated for 5 min followed by the addition of an equal volume of 2 M NaOH and centrifugation (1000 g)ofthereaction mixture. The supernatant was analyzed for reducing groups (specific activity, unit ¼ lmoles of reducing equi- valents released per minute per mg of protein at optimum conditions) [18]. Energy of activation (E a ) was determined by the slope (–E a /2.3R, where R is the gas constant) of Arrhenius plots obtained by plotting logarithm of maximum enzyme velocity determined at different temperatures (20–37 °C) under standard reaction conditions against the reciprocal of absolute temperatures (T ¼ 273 + A, where A is the temperature in °C) [19]. Kinetics of the chitosanolytic activity The pH optimum was determined using chitosan solution (1%) at a pH value between 1.5 and 6.0 (pH was adjusted using 0.1 M HCl/NaOH and above pH 6.5, chitosan was insoluble) and 1% casein solution of pH 4.0–11.0. The temperature optimum was determined by carrying out the reaction between 20 and 60 °C. pH and temperature stability measurements were carried out by preincubating the enzyme at different pH values/ temperatures followed by determination of the residual chitosanolytic activity (%) as described above by taking the aliquots of the enzyme at regular intervals. The Michaelis constants, K m and V max were evaluated from the double reciprocal plot of the initial velocity versus substrate concentrations. Isolation of the products Chitosan solution (1%, dissolved in 1% acetic acid with pH adjusted to 3.5) was treated with pronase in the ratio, 100 : 1 (w/w), incubated for different periods at optimum conditions followed by arresting the reaction using an equal volume of 2 M NaOH. The precipitate (LMWC) obtained after centrifugation at 1000 g, was dialyzed against water using a membrane with a molecular mass cut-off of 2 kDa (Sigma Chemicals Co.) and freeze-dried. The supernatant containing chito-oligomers and mono- mer as well as the heat denatured enzyme, was passed through a charcoal-Celite column. Unadsorbed saccha- rides (GlcN, GlcN-rich oligomers), enzyme and excess alkali added while arresting the reaction were collected by eluting with distilled water (Fraction IA), and the adsorbed saccharides (GlcNAc, GlcNAc-rich oligomers) were recovered using 60% ethanol as the eluant (Fraction II). Fraction IA was re-N-acetylated and subjected to a second charcoal-Celite chromatography phase wherein the water-washing removed the enzyme and alkali, and elution with 60% ethanol resulted in the recovery of re-N-acetylated GlcN and GlcN-rich oligomers (Fraction I). The fractions were concentrated by flash treatment at ambient temperature. Simultaneously, after neutralization, the supernatant was passed through Sephadex G15 column (62 · 0.8 cm; bed volume, 78 mL) to remove salt and denatured enzyme, and the fraction containing reducing groups (chito-oligomers + monomer) were pooled, followed by its freeze-drying. Determination of the molecular mass of chitosan and the depolymerization products The molecular mass was measured using three techniques: (a) viscometric measurements; (b) gel permeation chroma- tography (GPC) and HPLC. Viscometric measurements. The viscosity of chitosan and LMWC dissolved in sodium acetate buffer (0.5 M acetic acid + 0.2 M sodium acetate, pH 4.5) was measured using an Ostwald viscometer [20]. The average molecular mass was deduced using the Mark–Houwink’s equation g ¼ K·(molecular mass) a where, g is the intrinsic visco- sity, K ¼ 3.5 · 10 4 and a ¼ 0.76. 714 A. B. Vishu Kumar et al. (Eur. J. Biochem. 271) Ó FEBS 2004 Gel permeation chromatography. The molecular mass of chitosan was determined by GPC on a Sepharose CL-4B column (Sigma Chemicals Co; bed volume, 180 mL) equilibrated in sodium acetate buffer (pH 4.5) [21]. The column was precalibrated with dextrans (of known mole- cularmass)and chitosanmolecularmassstandards.Fractions I and II obtained after charcoal-Celite column chromatogra- phy were subjected to GPC using a Biogel P2 column (100 · 0.8 cm, Bio-Rad laboratories) equilibratedwith water and calibrated with chito-oligomers and GlcNAc [18]. High performance liquid chromatography. The molecular mass of LMWC was determined using E-linear and E-1000 columns in series. The mobile phase used was acetate buffer (0.5 M acetic acid + 0.2 M sodium acetate, pH 4.5) at a flow rate of 0.8 mLÆmin )1 and the detection was performed using an RI detector (Shimadzu Corp., Kyoto, Japan). The columns were calibrated with dextran standards. The use of acetate buffer minimized the interaction between the column material and -NH 2 groups. Fractions I and II were subjected to HPLC on an aminopropyl column (3.9 · 300 mm, Waters Associates, Ireland) using acetonitrile:water (70 : 30, v/v) as the mobile phase at a flow rate of 1.0 mLÆmin )1 and detected using an RI detector. Chito-oligomers and GlcNAc were used as standards. Further characterization of chitosan and the depolymerization products Further characterization of chitosan and the depolymeriza- tion products was performed using a variety of techniques as detailed below. Environmental scanning electron microscopy (ESEM). Freeze dried samples were spread on a double sided conducting adhesive tape pasted onto a metallic stub. Samples were observed under SEM (LEO 435 VP, LEO Electron Microscopy Ltd, Cambridge, UK) at 20 kV and variable pressure. Infrared (IR) spectroscopy. IR spectral studies were performed on a Perkin Elmer 2000 spectrometer under dry air at room temperature using KBr pellets. Chitosan, LMWC and freeze-dried chito-oligomers + monomer (4 mg samples) were mixed thoroughly with 200 mg KBr and 40 mg of the mixture was pelletized. The reproducibility of the spectra was verified on two preparations and the degree of acetylation (DA) was determined using the formula (A 1655 Æcm )1 /A 3450 Æcm )1 ) · 100 / 1.33, where A is the absorb- ance at these wavelengths, calculated from the baseline [22]. Potentiometric titration. Chitosan, LMWC and chito- oligomers + monomer (200 mg) were dissolved in 25 mL of 0.1 M HCl and the volume was made up to 100 mL with distilled water and the ionic strength was adjusted to 0.1 M using KCl. 0.1 M NaOH containing 0.1 M KCl was used as the titrant. Initially, the solution pH was brought to pH 2.0 by adding the titrant, after which it was added in stepwise increments (0.5 mL, each time). The titration was termin- ated when the solution pH reached 6.0. For each sample, five replicates were performed. A graph of pH as a function of the volume of titrant was plotted, which gave a titration curve having two inflexion points, whose difference along the abscissa corresponded to the amount of acid required for the protonation of the amine groups of samples. The number of equivalents of acid groups was calculated using the formula: [NH 3 + ](mequiÆkg )1 ) ¼ 1000 · molarity (moles/L) · volume of titrant (mL)/mass of sample (g) and the level of deacetylation was calculated by comparison between the number of free amino groups (per unit weight of the sample) and the equivalent weight of GlcN [23]. Circular Dichroism (CD). CD spectra for native chitosan and LMWC (5 mgÆmL )1 in 0.1 M perchloric acid; path length, 1 cm) were recorded on a Jasco J-810 automatic recording spectropolarimeter, continuously purged with N 2 before and during the experiment (Japan Spectroscopic, Tokyo, Japan). Slits were programmed to yield 10 A ˚ band width at each wavelength so that the resolution was more or less constant. The spectra were recorded between 200 and 240 nm (far UV region) and baseline was obtained using 0.1 M perchloric acid. After accumulation of scans, the spectra were standardized to mean residual ellipticity expressed as h in degree · cm 2 per residue using the mean residual weight of GlcNAc [24]. X-ray diffractometry. X-ray diffraction studies were car- ried out on an EG-7 G solid state germanium, liquid nitrogen cooled detector Scintag XDS-2000 diffractometer equipped with a h–h goniometer, 30 kV + 25 mA with CuKa radiation at 1.5414 nm (Enraf Nonius Co., Bohemin, NY). The relative intensity was recorded in a scattering range (2h)of0–45°. The crystallinity index (CrI, %) was determined using the formula (I 110 – I am ) · 100/I 110 ,where I 110 is the maximum intensity at 20° and I am is the intensity of amorphous diffraction at 16° [25]. Solid-state CP-MAS 13 C-NMR spectroscopy. 13 C-NMR spectra were obtained with a Bruker dsx 300 spectrometer at 75 MHz. The cross polarization pulse sequence was utilized for all samples, which were spun at the magic angle at 6.2 kHz for native chitosan and 7 kHz for LMWC and chito-oligomers + monomer. A contact time of 1 ms and a pulse repetition time of 5 ms were used, and more than 2000 scans were accumulated for each spectrum. Approximately 300 mg of freeze-dried samples were inserted into a 7 mm ceramic rotor. The DA was calculated using the equation, I CH3 /(I C1 + I C2 + … I C6 )/6, where I is the intensities of C1–C6 as well as methyl carbons [26]. Liquid-state 13 C-NMR spectroscopy. Chitosan, LMWC and freeze-dried chito-oligomers + monomer (50 mg each) were dissolved in 1 mL solvent mixture of D 2 O-DCl (0.98 + 0.02 mL, respectively). After ensuring complete dissolution of the samples, the spectra were recorded with a Bruker amx 400 spectrometer at 100 MHz and 27 °C. Results and Discussion Characterization of chitosan The molecular mass of purified chitosan was 71 ± 2 kDa as determined by both viscometry and GPC. The solid-state Ó FEBS 2004 Chitosanolysis by pronase (Eur. J. Biochem. 271) 715 CP-MAS 13 C-NMR and IR spectra indicated chitosan to have a b-conformation, as evidenced by the absence of splitting of signals corresponding to C3-C5 carbons (Fig. 1) and appearance of a single broad peak around 3371 cm (Fig. 2), in contrast to the presence of two peaks in case of a-conformation. The degree of acetylation (DA) determined by IR and 13 C-NMR was in good agreement with each other (Table 1). The crystallinity index (CrI) determined by X-ray diffraction pattern was 70% (Table 1) and indicated chemical homogeneity, and this was in close correspondence with chitosans of Euphausia superba (68.4%) [27]. Effect of different solvents, molecular mass and DA on chitosanolysis by pronase Preliminary studies indicated that chitosan could be degra- ded nonspecifically by proteases such as pepsin, papain, protease, etc. It was observed that chitosanolysis by pronase was dependent upon the solubility, molecular mass and DA of chitosan. Muzzarelli et al. (1994) compared acetic acid with lactic acid as a chitosan solvent during the nonspecific chitosanolysis by papain and reported lactic acid to be a better solvent owing to the hydrolyzing action of the former [28]. Contrary to this, the specific activity of pronase indicated that chitosan dissolved in aqueous acetic acid (1%) and was the best solvent when compared to formic and lactic acids. In acetic acid, being a weak acid (see pK a values, Table 2), chitosan was much less decomposed, [29] thus, resulting in better enzymic depolymerization. Increase in the molecular mass of chitosan from 71fi600 kDa and DA from 15fi26% resulted in increased specific activity (Table 2). Ibrahim et al. (2002) reported the effect of molecular mass and DA on lipase loaded chitosan bead characteristics, in which increases in molecular mass and DA resulted in efficient loading and minimization of the release of entrapped lipase [30]. The observed increase in specific activity with increasing molecular mass suggests the preference of pronase towards higher molecular mass chitosans. Improved chitosanolysis with higher DA indica- ted that the affinity of enzyme is more towards chitosan with higher DA. Homogeneity and specific activity of pronase Capillary zone electrophoresis and SDS/PAGE of pronase showed the presence of a single homogeneous protein (Fig. 3 and the inset) of molecular mass  20 kDa as determined using protein markers. The enzyme notably showed a much higher specific activity towards proteolysis of casein (4.14 U) as compared to 1.15 U towards chitosan depolymerization. Although the specificity of pronase was much lower ( 43-fold less) in comparison with that of chitosanase (from Streptomyces griseus,50UÆmg )1 ,Sigma Chemical Co.), the advantage of pronase catalyzed chitos- anolysis was the production of LMWC in higher yields (>70%), which was lacking in chitosanase as a result of its specificity towards the formation of mono- and oligomers (DP 2–3), rather than LMWC. Enzyme kinetics The pH optimum of pronase towards proteolysis and chitosanolysis were 7.5 and 3.5, respectively, whereas the temperature optimum towards chitosanolysis was 37 °C (Fig. 4A,B). Investigation of the pH and temperature stabilities indicated enzyme activation in the initial hour, which could be due to prior attainment of an active conformation by the enzyme, remaining stable up to 240 min (4 h) after which there was a decline (Fig. 4C,D), indicating enzyme stability in the pH range 2.0–5.0 and 45–60 °C in the initial hours. The effect of varying chitosan concentration on the initial velocity showed that chitosano- lysis by pronase follows Michaelis–Menten kinetics. The use of substrate concentration greater than 20 mgÆmL )1 was limited owing to increased viscosity of the solutions, affecting the enzyme penetration. The K m and V max values obtained from a double reciprocal plot were 5.21 mgÆmL )1 and 138.55 nmolesÆmin )1 Æmg )1 , respectively. Activation energy calculated for pronase towards proteolysis and chitosanolysis were  15.5 and 5.12 kcalÆmol )1 , respectively, further supporting the fact that pronase prefers proteins over chitosan as the substrate, which was in accordance with the specific activities of pronase. The maximum chitosano- lysis was observed at 240 min (4 h) beyond which there was no further appreciable increase. Fig. 1. Solid-state CP-MAS 13 C-NMR spectra of chitosan, LMWC and freeze-dried chito-oligomers + monomer. 716 A. B. Vishu Kumar et al. (Eur. J. Biochem. 271) Ó FEBS 2004 Characterization of LMWC Chitosan showed b-conformation as evidenced by both 13 C- NMR spectra and IR (Figs 1 and 2, respectively) [27], and accordingly was more susceptible to the hydrolytic proces- ses, due to individual chains arranged in a parallel manner [1], resulting in a weaker hydrogen-bonding network between the chains and thus enabling enzyme penetration. The percentage yield of LMWC, chito-oligomers and monomer (GlcNAc) was dependent on the reaction time (Table 1). There was approximately eightfold decrease in the molecular mass following depolymerization, which occurred in the initial hour beyond which the decrease was not appreciable (Table 1). Unless stated otherwise, further characterization was performed with the LMWC obtained after 1 h reaction time. Environmental scanning electron microscopy. Although chitosan is a linear polymer, its freeze-drying results in the aggregation of the chains by inter- and intrachain hydrogen bonding network resulting in the formation of particle- like structure [1]. ESEM of LMWC (obtained after 5 h Fig. 2. Infrared (IR) spectra of chitosan and LMWC. Table 1. Characteristics of chitosan and low-molecular weight chitosan (LMWC), and the percentage yield of the depolymerization products. Percentage yield is dependent on the incubation time (1–5 h) and is approximate. Molecular mass (kDa) Degree of acetylation (DA; %) Crystallinity index (CrI; %) Yield (%) IR 13 C-NMR (solid state) Chitosan Native 71 ± 2 25.7 26.3 70 – LMWC 8.5–9.5 a 19.0 18.6 62 74–80 Depolymerization products Chito-oligomers – – – – 10–12 GlcNAc – – – – 6–8 a Dependent on the incubation time (1–5 h) . Ó FEBS 2004 Chitosanolysis by pronase (Eur. J. Biochem. 271) 717 incubation) indicated decreased particle size compared to chitosan (Fig. 5) in accordance with the lower molecular mass. The continuous fibrous/flat ribbon type appearance of chitosan disappeared upon depolymerization and resul- ted in the formation of discontinuous smaller pieces. IR spectra. In both chitosan and LMWC, a well defined band around 2922 cm )1 due to the -CH vibration mode represented a good internal reference for comparison of band absorbance. Absence of a sharp and convoluted spectral band around 3600–3000 cm )1 in the spectra of both chitosan and LMWC indicated the absence of free -OH groups and the involvement of both -OH…3and -CH 2 OH…6 in intra- and intermolecular hydrogen bonding (Fig. 2) [31]. Line width and frequency position of the band above 3000 cm )1 corresponds to the intermolecular crystal lattice of a molecule, which depends on the hydrogen bonding network, and results in the incorporation of water molecule into molecular structure. The latter promotes a perfect crystal structure and increases CrI. In chitosan, this band appeared at 3371 cm )1 and there was a slight shifting ofthesameinLMWC(3368cm )1 ), indicating a decrease in the ordered structure. This is probably due to the decrease in the DA (Table 1) of LMWC, which results in lower degree of hydration of the molecule affecting the crystallinity index and thus the orderliness. This was further confirmed by a decrease in the peak height at 1320 cm )1 in LMWC compared to chitosan, which corresponds to the GlcNAc residues [32]. The region between 1420 and 1435 cm )1 , considered to be polysaccharide conformation sensitive, is assigned to -CH 2 bending, which depends on the confor- mation of the primary -OH group (C6) in a favorable orientation [33]. The shift of the same to 1458 cm )1 in the LMWC would probably indicate a change in the environ- ment of primary -OH groups and thus the overall hydrogen- bonding network. Potentiometric titration. Potentiometric titration is a well- known analytical tool for quantifying acidic functional groups and is independent of sample molecular mass [23]. The DA (%) calculated by titration (graphs, not shown) were found to be 26.3, 18.8 and 32.3, respectively, which were in good agreement with that calculated using IR spectra. CD-Spectra. Compared to chitosan, in LMWC, there was a decrease in the peak height near 211 nm (due to gfip* transition) (Fig. 6) corresponding to the acetyl content, which is independent of conformation, chain length, ionic strength and pH [24]. This decrease in the peak height continued with increased time of enzyme digestion. X-ray diffractometry. The X-ray diffraction (Fig. 7) pat- tern of chitosan was that of a typical hydrated form with a peak in the range 9–11° and 020 reflection appearing at 10.32° (d-spacing, 8.571 A ˚ ), characteristic of a ÔtendonÕ type polymorph [34]. The LMWC showed a shift of the same to 9.94° (d-spacing, 8.898 A ˚ ), which is probably due to larger d-spacing as a result of increase in the unit cell dimension that results in the incorporation of fewer water molecules. Decrease in the intensity of 020 reflections in LMWC was also indicative of a decrease in the acetamido group [33,34]. The observed decrease in CrI (Table 1) was in good agreement with the fact that decreased DA results in less hydrated crystals that affects the CrI. Solid-state CP-MAS 13C-NMR. This technique is known to be very sensitive to changes in the local order structure. Chitosan showed higher crystalline nature as evidenced by narrow line width of the peaks (Fig. 1). Peak broadening in LMWC indicates decrease in crystalline nature (Fig. 7), in support of its low CrI. The splitting of peaks corresponding to C1, 3, 4 and 5 ring carbons supports a rather different hydrogen-bonding network and thus the existence of multiple conformations in LMWC. Chemical shift values for ring and methyl carbon atoms are given in Table 3. The observed difference in chemical shift values of C1 and C4 indicated a conformational change of the glycosidic linkage in LMWC. Multiplicity of the peak corresponding to C4 is independent of DA and is associated with deacetylation Table 2. Effect of acids, molecular mass and DA on the chitosanolytic activity of pronase. Effect pH Activity (l M reducing sugar releasedÆmin )1 Æmg )1 ) Effect of different acids (1%) as the solvent Acetic acid (pK a , 4.76) 3.4 1.10 3.0 0.89 Formic acid (pK a , 3.74) 3.4 1.08 Lactic acid (pK a , 3.73) 3.0 0.59 Effect of chitosans of different molecular mass (DA,  23–26%) 71 (kDa) – 1.15 150 (kDa) – 1.60 400 (kDa) – 1.67 600 (kDa) – 1.76 Effect of DA (%) on chitosanolysis by pronase  15 (83 kDa) – 0.70  19 (75 kDa) – 0.86  25 (71 kDa) – 1.15 Fig. 3. Capillary electropherogram of pronase in Tris/glycine buffer. pH 8.3 detected at 280 nm. Inset: SDS/PAGE of protein markers (lane A) and pronase (lane B). 718 A. B. Vishu Kumar et al. (Eur. J. Biochem. 271) Ó FEBS 2004 temperature [27]. Nevertheless, it could also be due to increased mobility of the individual chains resulting from decrease in chain length, as evidenced by decreased molecular mass of LMWC. The DA calculated using 13 C-NMR spectra was in accordance with the one calcula- ted by IR-spectra (Table 1). The peak corresponding to -CH 3 showed a slight decrease in the height in case of LMWC further supporting a decrease in the DA [33] (Table 3 and Fig. 1). In LMWC, the -C¼O peak showed splitting and considerable difference in chemical shift value (3.467 p.p.m.) compared to native chitosan, further con- firming the conformational inhomogeneity. Liquid-state 13C-NMR. 13 C-NMR in the liquid state (Fig. 8) was performed as the peaks corresponding to -CH 3 and -C¼O were not obvious in solid-state spectra. There was a decrease in the peak height corresponding to -CH 3 in LMWC compared to chitosan, whereas chito- oligomers + monomer showed slight increase in the same. The DA (%) calculated using the spectra was 26.02 and 18.97%, respectively, for chitosan and LMWC [22], which were in good agreement with those calculated before (Table 1). Solubility of LMWC. The solubility study was performed according to the method described by Qin et al. (2003) [35]. Although the molecular mass of LMWC was <10 kDa, it was not readily soluble in aqueous medium (water-solubility of LMWC, 1 h sample )72%, 5 h )63%), instead it required a very dilute acidic medium for complete solubilization (LMWC, 100% solubility compared to 13% solubility of chitosan in 0.01% acetic acid, chitosan required 1% acetic acid for complete dissolution). There was no evidence for annealed polymorphism in LMWC, as indicated by the absence of a 15° reflection in the X-ray diffraction pattern [25]. This observation was contrary to the earlier reports that decrease in the molecular mass of chitosan is associated with increased solubility, attributed to decreased intermolecular interactions, such as van der Walls forces and hydrogen bonds, provided that the DA did not change after degradation [35]. However, chitosanolysis by pronase was associated with decreased DA as evidenced by the spectral data and hence the observed decreased solubility (in water) could probably be due to the exposure and thus conversion of -NH 2 groups on LMWC to the sodiated form (Na + , added to terminate the reaction), which could result in a different molecular conformation. This was obvious from the d-spacing of LMWC (8.898 A ˚ ), which was neither typical of tendon nor of L-2 polymorph. Compared to the 1 h sample, LMWC obtained after 5 h incubation showed much lower solubility as there was a further decrease in the DA (Table 1), exposing more -NH 2 groups for sodiation. Cheng et al. (2000) made use of two volumes of acetonitrile [36], instead of an equal volume of 2 N NaOH as in the present study, to arrest the catalytic reaction. Similar use of acetonitrile in the present study Fig. 4. Effect of (A) pH and (B) temperature on pronase activity and the stability studies (C & D). Ó FEBS 2004 Chitosanolysis by pronase (Eur. J. Biochem. 271) 719 resulted in LMWC that was readily soluble in water, thus, supporting the formation of sodiated LMWC when alkali was added. The drawbacks of using acetonitrile were, its high cost and flash treatment of the supernatant at elevated temperature, which may result in Maillard reaction products limiting one step production of LMWC and oligomers + monomer. Although 0.01% acetic acid was necessary for LMWC solubilization, the pH value of resulting solution was near neutrality ( 6.8). Nevertheless, the decrease in molecular mass, DA and CrI, and molecular inhomogeneity of LMWC did not hinder its biofunctionality. LMWC exhibited potent anti- microbial activity towards Bacillus cereus [10 6 colony forming units (CFU)] and Escherichia coli (10 3 CFU), with a minimum inhibitory concentration of 0.01 and 0.03% (w/v), respectively. The inhibitory activity of LMWC was much superior over that of chitosan, and resulted in a progressive lysis of the bacterial cells as evidenced by SEM (unpublished observations). Characterization of the mono-oligomeric mixture Chito-oligomers + monomer, due to their affinity, bind to charcoal through a weak van der Wall’s force and their binding capacity is dependent on charge/mass ratio. The affinity decreases with increase in the charge/mass ratio of the binding molecules. The bound fractions, due to their varying solubility, are desorbed using organic solvent gradient [37,38]. In the present study, use of 60% ethanol resulted in the elution of all the bound chito-oligomers + monomer. Fraction I (obtained after re-N-acetylation), by both GPC (Biogel P2 column) and HPLC (aminopropyl column) (Figs 8 and 9, respectively) showed the existence of dimer and trimer in abundance along with higher oligosac- charides (tetra- to hexamer), whereas Fraction II showed mainly dimer and tetramer along with GlcNAc. Fig. 5. Environmental scanning electron microscopy (3500) of (A) chitosan and (B) LMWC. Fig. 6. Circular dichroic (CD) spectra of chitosan and LMWC. Fig. 7. X-ray diffractograms of chitosan and LMWC. 720 A. B. Vishu Kumar et al. (Eur. J. Biochem. 271) Ó FEBS 2004 IR-spectrum of the freeze-dried chito-oligomers + monomer showed a DA value of 31.2% (spectrum not shown), which was in accordance with that calculated using 13 C-NMR ( 30.83%, Fig. 1). GPC-HPLC profiles of Fraction I and II are depicted in Figs 9 and 10, respectively. In the solid-state 13 C-NMR spectra (Fig. 1), appearance of peaks at 25.831 and 173 600 p.p.m. corresponding to -CH 3 and -C¼O groups, respectively, indicated the release of GlcNAc/GlcNAc-rich oligomers; this was further con- firmed by liquid-state 13 C-NMR (Fig. 8). The DA value calculated using liquid-state spectrum was 31.33%, which was in good agreement with that calculated by solid-state spectrum (Table 1). The monomeric residue sequence in chitosan is of four types, -GlcN-GlcN-, -GlcN-GlcNAc-, -GlcNAc-GlcN- and -GlcNAc-GlcNAc-, of which the first is the major type and the last one results from the heterogeneous de-N- acetylation. Formation of chito-oligomers as one of the products and a drastic decrease in molecular mass of chitosan indicates an endo-type activity of pronase. Addition of hexosaminidase (specific for the release of GlcNAc from the nonreducing end) to LMWC and Fraction II did not result in the release of reducing groups indicating action of pronase on -GlcNAc-GlcN-linkage resulting in the products (LMWC and oligomers) with GlcNAc at reducing ends. In addition, the presence of GlcNAc in Fraction II and the absence of GlcN in Fraction I demonstrates the exo-type action of pronase and this was further confirmed by using (GlcNAc) 2 and (GlcNAc) 3 as substrates, where the activities were found to be 0.09 and 0.18 units, respectively. From these data, Table 3. Solid-state CP-MAS 13 C-NMR chemical shift values of ring and methyl carbons of native chitosan, LMWC and chito-oligomers. Sample Chemical shift values (p.p.m.) CH 3 C2/C6 C3/C5 C4 C1 -C¼O Chitosan 26.843 60.906 78.703 84.703 108.103 176.524 LMWC (1 h) 25.521 62.942 77.878 87.332 107.025 173.057 Chito-oligomers + monomer 25.831 61.025 78.051 87.140 107.562 173.600 Fig. 8. Liquid state 13 C-NMR spectra of chitosan, LMWC and freeze dried chito-oligomers + monomer. Fig. 9. Chromatographic profile of the chito-oligomeric Fractions (I and II) on Biogel P2 column (Bed volume, 80 mL). Ó FEBS 2004 Chitosanolysis by pronase (Eur. J. Biochem. 271) 721 along with that of hexosaminidase, it could be concluded that during exo-type of activity, pronase prefers GlcNAc at the nonreducing ends, i.e. -GlcNAc-GlcNAc-linkage, the action on which results in the release of GlcNAc and products with GlcNAc at the reducing ends. The observed decrease in DA value of LMWC is due to the release of GlcNAc and GlcNAc-rich oligomers as some of the products. An examination of DA values indicates that the total DA of the products is equal to that of native chitosan (19% for LMWC and 31.2% for chito-oligomers + monomer, the average of these is 25.1%, in agreement with DA of native chitosan, Table 1). From these data, it was also evident that in the reaction catalyzed by pronase, only glycosidic bonds were cleaved leaving the N-acetyl groups intact. Evaluation of the ionization constants of pronase towards chitosanolysis The ionization constants of enzymes that obey Michaelis– Menten kinetics can be further evaluated by analyzing the curves of log (V max /K m ) vs. pH, which yields pK E values, and often provides valuable clues as to the identities of the amino acid residues essential for enzymatic activity [39]. In the present study, the calculated pK E was  4.0, suggesting the involvement of aspartic (Asp, pK a 3.90)/glutamic (Glu, pK a 4.07) acid residues for the catalysis. Moreover, the pH optima of pronase towards proteolysis and chitosanolysis were at two extremes (Fig. 2A), suggesting possible varia- tions in the protein conformation. Use of 1 m M phenyl- methanesulfonyl fluoride in the reaction medium brought about 100% inhibition of proteolysis, whereas, the chitos- anolytic activity was inhibited by only 30%, which sugges- ted the involvement of a serine residue in this nonspecific catalysis. As a serine protease, pronase bears a triad consisting Asp, His and Ser residues in the catalytic site [39,40]. Watanabe et al. [41], by site directed mutagenesis, identified the involvement of crucial Glu and Asp residues during the chitinolytic activity of chitinase A1 from Bacillus circulans W-12. Replacement of SerfiAla decreased the catalytic activity without affecting the K m , and the mutant retained only 10% of the wild-type activity. It was concluded that Ser might have an important role in maintaining the structural features of the catalytic site. On similar lines, it may be tempting to speculate that Asp/ Glu + Ser are probably involved during the pronase- catalyzed chitosanolysis. While screening the proteases for chitosanolysis, highest activity was shown by pepsin, followed by papain, pronase and a protease from Aspergillus niger (4.98, 1.78, 1.16 and 0.058 units, respectively). Though, pepsin and papain were inexpensive, the former showed a decrease in the activity during the course of time due to its auto-catalytic property and the latter, due to its higher initial velocity, resulting in the formation of LMWC of  4–5 kDa and below. As pronase overcomes these drawbacks, it could be considered as a better chitosanolytic agent in comparison with other proteases screened. In conclusion, depolymerization of chitosan using pro- nase, though unusual and nonspecific, is a viable alternative way of catalysis to produce LMWC, chito-oligomers (DP 2–6) and monomer (GlcNAc), quantitatively and econom- ically with multiple applications. Although chitosanase is Ôthe enzymeÕ for chitosanolysis, it is expensive ( 15 $ per unit), unavailable in bulk and results in more monomer and oligomers (DP 2–3), whereas pronase is relatively inexpen- sive ( 0.13 $ per unit), easily available and LMWC of any desired molecular mass can be custom made by manipula- ting the reaction conditions. Use of pronase in place of chitosanase is still feasible even though pronase quantity has to be increased by  40-fold to mimic the specific activity of chitosanase. Thus, the results add value to otherwise commercially available raw materials, viz. chitosan from the offal of marine food processing industry (and pronase), and show the biofunctionalities of these materials. Acknowledgements Authors thank Dr S. Subramanian (Indian Institute of Science, Bangalore) for X-ray analysis, Dr A.G. Appu Rao and Dr Sridevi Annapurna Singh (Department of Protein Chemistry and Technology, CFTRI, Mysore) for CD measurements. A.B.V.K. thanks CSIR (New Delhi) for the senior research fellowship. References 1. Tharanathan, R.N. & Kittur, F.S. (2003) Chitin – The undisputed biomolecule of great potential. Crit.Rev.FoodSci.Nutr.43, 61–87. Fig. 10. HPLC profiles of chito-oligomeric Fractions I and II separated on charcoal/Celite column (values in the parenthesis indicate retention time of the individual peaks in minutes). 722 A. B. Vishu Kumar et al. (Eur. J. Biochem. 271) Ó FEBS 2004 [...]... in relation with chitin and chitosan characterization Polymer 42, 3569–2580 33 Kittur, F.S., Vishu Kumar, A.B & Tharanathan, R.N (2003) Low molecular weight chitosans – preparation by depolymerization with Aspergillus niger pectinase, and characterization Carbohydr Res 338, 1283–1290 34 Ogawa, K (1991) Effect of heating an aqueous suspension of chitosan on the crystallinity and polymorphs Agric Biol... Molecular weight dependency of antimicrobial activity by chitosan oligomers In Food Hydrocolloids: Structure, Properties and Functions, (Nishinari, K & Doi, E., eds), pp 71–76, Plenum Press, New York 22 Baxter, A., Michael, D & Taylor, K.D.A (1992) Improved method for i.r determination of the degree of N-acetylation of chitosan Int J Biol Macromol 14, 166–169 Chitosanolysis by pronase (Eur J Biochem 271)... activity of the enzyme previously reported as b)1,3/1,4 glucanase from Bacillus circulans WL12 Biosci Biotechnol Biochem 62, 2107–2114 13 Pantaleone, D., Yalpani, M & Scollar, M (1992) Unusual susceptibility of chitosan to enzymic hydrolysis Carbohydr Res 237, 325–332 14 Zhang, H & Neau, S.H (2001) In vitro degradation of chitosan by a commercial enzyme preparation: effect of molecular weight and degree of. .. A.B., Gowda, L.R & Tharanathan, R.N (2003) Chitosanolysis by an isozyme of Aspergillus niger – A non-specific activity Carbohydr Polym 53, 191–196 16 Kittur, F.S., Kumar, K.R & Tharanathan, R.N (1998) Functional packaging properties of chitosan films Z Lebensm Unter Forsch A 206, 44–47 17 Laemmli, U.K (1970) Cleavage of structural proteins during the assembly of head of Bacteriophage T4 Nature 227, 680–685... applications of chitin and chitosans J Food Sci Technol 10, 37–51 3 Ikeda, I., Sugano, M., Yashida, K., Sasaki, E., Iwamoto, Y & Hatono, K (1995) In Dietary Fibre in Health and Disease, (Kritchevsky, D & Bonefield, C., eds), pp 95, Eagan Press, USA 4 Shon, D.H (2001) Chitosan oligosaccharides for functional foods and microbial enrichment of chitosan oligosaccharides in soypaste Proceedings of the International... (1996) Compositional heterogeneity of heterogeneously deacetylated chitosans Carbohydr Polym 29, 17–24 27 Focher, B., Naggi, A., Tori, G., Cosani, A & Terbojevich, M (1992) Chitosans from Euphausia superba 2: Characterization of solid-state structure Carbohydr Polym 18, 43–49 28 Muzzarelli, R.A.A., Tomasetti, M & Hari, P (1994) Depolymerization of chitosan with the aid of papain Enz Microb Technol 16,... Natural Products, Japan, pp 56–66, The Committee on Science and Technology in Developing Countries (COSTED) and The Science Council of Japan, Tokyo, Japan 5 Sukwattanasinitt, M., Zhu, H., Sashiwa, H & Aiba, S (2002) Utilization of commercial non-chitinase enzymes from fungi for preparation of 2-acetamido-2-deoxy-D-glucose from b-chitin Carbohydr Res 337, 133–137 6 Rupley, J.A (1964) The hydrolysis of chitin... Effect of hemicellulase on the molecular weight and structure of chitosan Polym Degradation Stability 80, 435–441 36 Cheng, C.Y & Li, Y.K (2000) An Aspergillus chitosanase with potential for large-scale preparation of chitosan oligosaccharide Biotechnol Appl Biochem 32, 197–203 37 Allan, G.G., Altman, L.C., Bensinger, R.E., Ghosh, D.K., Hirabayashi, Y., Neogi, A.N & Neogi, S (1984) In Chitin, Chitosan and. .. Degree of deacetylation of chitosan using conductometric titration and solid state NMR Carbohydr Res 246, 331–336 24 Domard, A (1987) pH and c.d measurements on a fully deacetylated chitosan: application to CuII–polymer interactions Int J Biol Macromol 9, 333–336 25 Focher, B., Beltrame, P.L., Naggi, A & Tori, G (1990) Alkaline N-deacetylation of chitin enhanced by flash treatments: Reaction kinetics and. .. 227, 680–685 18 Imoto, T & Yagishita, K (1971) A simple activity measurement of lysozyme Agric Biol Chem 33, 1154–1157 19 Whitaker, J.R (1972) Principles of Enzymology for the Food Sciences Marcel Dekker, New York 20 Rinaudo, M., Milas, M & Dung, L.P (1993) Characterization of chitosan – influence of ionic strength and degree of acetylation on chain expansion Int J Biol Macromol 15, 281–285 21 Sekiguchi, . Non-specific depolymerization of chitosan by pronase and characterization of the resultant products Acharya B. Vishu Kumar 1 , Lalitha R. Gowda 2 and. of chitosan and LMWC. Table 1. Characteristics of chitosan and low-molecular weight chitosan (LMWC), and the percentage yield of the depolymerization products. Percentage