Characterization of a thermostable extracellular

SAVE OFFLINE

Biochimica et Biophysica Acta 1472 (1999) 314^322 www.elsevier.com/locate/bba

Characterization of a thermostable extracellular L-galactosidase from a thermophilic fungus Rhizomucor sp. S.A. Shaikh, J.M. Khire, M.I. Khan * Division of Biochemical Sciences, National Chemical Laboratory, Pune 411 008, India Received 23 March 1999; received in revised form 26 July 1999; accepted 28 July 1999

Abstract An extracellular L-galactosidase from a thermophilic fungus Rhizomucor sp. has been purified to homogeneity by successive DEAE cellulose chromatography followed by gel filtration on Sephacryl S-300. The native molecular mass of the enzyme is 250 000 and it is composed of two identical subunits with molecular mass of 120 000. It is an acidic protein with a pI of 4.2. Purified L-galactosidase is a glycoprotein and contains 8% neutral sugar. The optimum pH and temperature for enzyme activity are 4.5 and 60³C, respectively. The enzyme is stable at 60³C for 4 h, and has a t1=2 of 150 min at 70³C which is one of the highest reported for fungal L-galactosidases. Substrate specificity studies indicated that the enzyme is specific for L-linked galactose residues with a preference for p-nitrophenyl-L-D-galactopyranoside (pNPG). The Km and Vmax values for the synthetic substrates pNPG and o-nitrophenyl-L-D-galactopyranoside (oNPG) were 0.66 mM and 1.32 mM; and 22.4 mmol min31 mg31 and 4.45 mmol min31 mg31 , respectively, while that for the natural substrate, lactose, was 50.0 mM and 12 mmol min31 mg31 . The end product galactose and the substrate analogue isopropyl thiogalactopyranoside (ITPG) inhibited the enzyme with Ki of 2.6 mM and 12.0 mM, respectively. The energy of activation for the enzyme using pNPG and oNPG were 27.04 kCal and 9.04 kCal, respectively. The active site characterization studies using group-specific reagents revealed that a tryptophan and lysine residue play an important role in the catalytic activity of the enzyme. ß 1999 Published by Elsevier Science B.V. All rights reserved. Keywords: L-Galactosidase; Characterization; Thermophilic fungus; Active site; Rhizomucor sp.

1. Introduction L-Galactosidase

(L-D-Galactohydrolase,

EC

Abbreviations: pNPG, p-nitrophenyl-L-D-galactopyranoside; oNPG, o-nitrophenyl-L-D-galactopyranoside; L-ME, L-mercaptoethanol; IPTG, isopropyl thiogalactopyranoside ; PEG, polyethylene glycol-8000; HNBB, 2-hydroxy-5-nitrobenzyl bromide; NBS, N-bromosuccinimide ; TNBS, 2,4,6,-trinitrobenzenesulfonic acid; DEPC, diethylpyrocarbonate; NAI, N-acetylimidazole ; DTNB, 5,5P-dithiobis(2-nitrobenzoic acid); EDAC, 1-ethyl-3-(3dimethylaminopropyl) carbodiimide * Corresponding author. Fax: +91-20-5884032; E-mail: [email protected]

3.2.1.23) is among the most common industrial enzymes due to its various applications in dairy and food industry [1,2]. The enzyme is clinically important in the preparation of lactose-free milk and milk products for patients su¡ering from lactose intolerance [2]. The occurrence of this enzyme in nature is diverse, i.e. in plants, animals, and microorganisms [3]. The microorganisms have the advantage of high enzyme production and among them fungi are preferred due to the extracellular localization, acidic pH optima and broad stability pro¢les of the enzymes produced by them. [4]. L-Galactosidase is mainly an intracellular enzyme in bacteria and yeast, but

0304-4165 / 99 / $ ^ see front matter ß 1999 Published by Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 4 1 6 5 ( 9 9 ) 0 0 1 3 8 - 5

BBAGEN 24887 1-10-99

S.A. Shaikh et al. / Biochimica et Biophysica Acta 1472 (1999) 314^322

the enzyme produced by fungi is generally extracellular. The L-galactosidase from Escherichia coli is the most extensively studied enzyme and its properties, reaction mechanism and structure have been determined [5,6]. Though extracellular L-galactosidase from mesophilic fungi, viz. Fusarium moniliforme, Penicillium notatum and Aspergillus sp. [7^9], has been puri¢ed and characterized, only a few reports exist on the enzyme from thermophilic fungi [10,11]. We have previously reported production and increased yields by solid-state fermentation of an extracellular L-galactosidase from a thermophilic fungus Rhizomucor sp. [12]. Here we report the puri¢cation and characterization of the enzyme and identi¢cation of active site residues. 2. Materials and methods 2.1. Materials p-Nitrophenyl-L-D-galactopyranoside (pNPG), onitrophenyl-L-D-galactopyranoside (oNPG), lactose, 2-hydroxy-5-nitrobenzyl bromide (HNBB), N-bromosuccinimide (NBS), 2,4,6-trinitrobenzenesulfonic acid (TNBS), pNP-L-D-N-acetylglucosaminide, pNPK-D-galactopyranoside, pNP-L-D-glucopyranoside, pNP-K-D-mannopyranoside, pNP-L-D-mannopyranoside, HEPES, MES, polyethylene glycol-8000, L-mercaptoethanol and ampholines (pH 3^10) were obtained from Sigma Chemical Co. (St. Louis, MO, USA). Sephacryl S-300 and DEAE-cellulose were obtained from Pharmacia (Uppsala, Sweden). All other reagents used were of the highest purity available from commercial sources. 2.2. Methods 2.2.1. Culture conditions and production of L-galactosidase The growth and maintenance of the thermophilic fungus Rhizomucor sp. and production of extracellular L-galactosidase by solid-state fermentation were carried out as described earlier [12]. 2.2.2. Enzyme assays L-Galactosidase activity was assayed by incubating 25 Wl of suitably diluted enzyme with 50 Wl of 6.6

315

mM pNPG or oNPG and 925 Wl of 20 mM citratephosphate bu¡er, pH 4.5 at 60³C for 30 min. The reaction was terminated by adding 1.0 ml of 0.5 M Na2 CO3 and the p-nitrophenol or o-nitrophenol released was determined by reading the absorbance at 405 or 410 nm, respectively. When lactose was used as a substrate, the liberated glucose was determined using a glucose oxidase-peroxidase kit [13]. One unit of L-galactosidase activity (U) is expressed as the amount of enzyme that releases 1.0 Wmol of product (p-nitrophenol/o-nitrophenol/D-glucose) per minute under the assay conditions. Inhibition studies with glucose, galactose, sucrose and IPTG were carried out by pre-incubating the enzyme with the respective saccharides (5^50 mM) for 15 min and the substrate pNPG subsequently added to determine the enzyme activity. The inhibition constant (Ki ) was determined using a minimum of three di¡erent concentrations of respective inhibitor. 2.2.2.1. Other glycosidase activities. The enzyme was assayed for other glycosidase activities (K-galactosidase, L-N-acetylglucosaminidase, L-glucosidase, K- and L-mannosidase, L-L-fucosidase and L-D-xylosidase) using their respective pNP substrates. 2.2.3. Puri¢cation of L-galactosidase All puri¢cation steps were carried out at 4³C unless otherwise mentioned. Routine assays of the enzyme during puri¢cation steps were based on the hydrolysis of pNPG. Protein estimations of column e¥uents were monitored by absorbance at 280 nm. Total protein estimation of pooled fractions during puri¢cation steps was monitored according to the method of Lowry et al. using BSA as standard [14]. 2.2.3.1. Ammonium sulfate precipitation. The crude enzyme extract (in citrate phosphate bu¡er, 100 mM, pH 4.5 containing 150 mM NaCl), was brought to 90% saturation by adding solid ammonium sulfate under stirring and left overnight at 4³C. The precipitated protein was collected by centrifugation (8000Ug, 15 min), dissolved in a minimum amount of distilled water and dialyzed exhaustively against phosphate bu¡er, 20 mM, pH 7.2. The undissolved material was removed by centrifugation

BBAGEN 24887 1-10-99

316

S.A. Shaikh et al. / Biochimica et Biophysica Acta 1472 (1999) 314^322

(5000Ug, 10 min) and the clear supernatant was fractionated by DEAE-cellulose chromatography.

method of Vesterberg [17] using ampholines in the pH range 3^10.

2.2.3.2. DEAE-cellulose chromatography. The enzyme obtained from the above step was loaded on a DEAE-cellulose column (2.5U50 cm) pre-equilibrated with 20 mM phosphate bu¡er, pH 7.2. The column was then washed with the same bu¡er until the £ow-through fractions showed an absorbance below 0.05 at 280 nm. The bound enzyme was eluted using a step gradient of NaCl (0.1^0.5 M) in 20 mM phosphate bu¡er, pH 7.2. Fractions (5 ml) showing L-galactosidase activity were pooled, concentrated by lyophilization, dialyzed against the same bu¡er and rechromatographed on a fresh DEAE-cellulose column under identical conditions. The active fractions were collected, concentrated and dialyzed against citrate-phosphate bu¡er, 20 mM, pH 4.5.

2.2.4. Characterization of L-galactosidase

2.2.3.3. Polyethylene glycol (PEG) treatment. The enzyme obtained after the DEAE-cellulose step was treated with polyethylene glycol-8000 (10% w/v) and sodium citrate (30% w/v). The mixture was left at room temperature for 30 min, and centrifuged (5000Ug, 10 min). The color-free protein (lower layer) was collected and dialyzed extensively against distilled water followed by 20 mM citratephosphate bu¡er, pH 4.5. The dialysate was concentrated by lyophilization and used for further puri¢cation. 2.2.3.4. Gel ¢ltration. The above partially puri¢ed enzyme (8.0 mg) was loaded on a Sephacryl S300 column (1.2U160 cm) equilibrated with 20 mM citrate phosphate bu¡er pH 4.5. The active fractions (1.0 ml each) were pooled and subjected to re-chromatography on an S-300 column under identical conditions. Active fractions were pooled, concentrated, dialyzed and stored at 320³C until further use. No loss of activity was observed under these conditions. 2.2.3.5. Electrophoresis. Native polyacrylamide gel electrophoresis (PAGE) was carried out in 7.5% (w/v) acrylamide gels in Tris-glycine system at pH 8.9 [15], and the gel was run in the absence of L-mercaptoethanol (L-ME). The gels were visualized by staining with silver nitrate [16]. Isoelectric focusing was done in 7.5% (w/v) acrylamide gels according to the

2.2.4.1. Determination of Mr . The native molecular mass of the enzyme was determined by gel ¢ltration on a Sephacryl S-300 column (1.2U100 cm). The column was equilibrated with 100 mM citratephosphate bu¡er, pH 4.5 and calibrated using blue dextran (2000 kDa), thyroglobulin (669 kDa), apoferritin (443 kDa), alcohol dehydrogenase (150 kDa) and bovine serum albumin (66 kDa). The subunit molecular mass was determined by SDS-PAGE [18], in 10% (w/v) acrylamide gels using high molecular weight markers (Sigma SDS-6H) as reference standards. The sample was treated with L-ME and heated for a few minutes in a boiling water bath before electrophoresis. The running bu¡er did not contain any reducing agent. The gels were stained with Coomassie brilliant blue R-250. 2.2.4.2. Amino acid analysis. The amino acid composition of puri¢ed L-galactosidase was determined on a liquid phase automated amino acid analyzer (Hewlett Packard Ti series 1050, equipped with a £uorescence detector) after hydrolysis in 6 N HCl for 20 h at 110³C. Total cysteine content was determined by the method of Habeeb [19] and total tryptophan was determined according to Spande and Witkop [20]. 2.2.4.3. Total carbohydrate. The total neutral sugar content of the enzyme was determined by the phenol-sulfuric acid method [21] using mannose as the standard. 2.2.4.4. E¡ect of pH on L-galactosidase. The optimum pH of the enzyme was determined by carrying out the enzyme assay in universal bu¡ers (pH 3^9, 100 mM) [22]. The stability of the enzyme at various pHs was studied by incubating the enzyme in the respective bu¡ers (pH 3^9) for 24 h at 28³C. The enzyme was then assayed for residual L-galactosidase activity under standard assay conditions. 2.2.4.5. E¡ect of temperature on L-galactosidase. Temperature dependence was studied by

BBAGEN 24887 1-10-99

S.A. Shaikh et al. / Biochimica et Biophysica Acta 1472 (1999) 314^322

monitoring the enzyme activity on pNPG in 100 mM citrate-phosphate bu¡er (pH 4.5), at various temperatures (25^75³C). Stability studies were carried out by incubating the enzyme at di¡erent temperatures (25^75³C) in the absence of the substrate and by monitoring the activity at de¢nite time intervals (0^ 240 min) using pNPG. The thermal stability of the chemically modi¢ed enzyme using group-speci¢c reagents was also determined similarly. To determine the energy of activation of the enzyme, the kinetic rate constants, Km and Vmax of Lgalactosidase were determined at various temperatures (37^70³C) with oNPG and pNPG. Suitably diluted enzyme (2 U ml31 ) was incubated with varying concentrations of the substrates oNPG/pNPG (25^ 350 Wg). Ln Vmax was plotted versus the reciprocal of the absolute temperature to obtain an Arrhenius plot for both the substrates. Energy of activation was calculated using the equation Eact = 3[slope/0.219] in Cal mol31 .

317

tryptophan residues modi¢ed was determined using a molar extinction coe¤cient of 5500 M31 cm31 [20].

2.3. Chemical modi¢cation studies

2.3.1.2. Reaction with HNBB. HNBB was prepared fresh in dry acetone. Puri¢ed L-galactosidase (2 WM, 500 Wl) in 50 mM sodium acetate bu¡er, pH 4.0, was incubated with 5.0^20 mM HNBB. The reaction mixture was rapidly mixed and allowed to react for 10^20 min. After the incubation period, the reaction was terminated by the addition of a 10-fold excess of L-tryptophan (200 mM stock). The reaction mixture was passed through Sephadex G-25 (0.5U10 cm) to remove the excess reagent and the residual activity was measured under standard assay conditions. Enzyme samples incubated in the absence of reagent served as control. The number of tryptophan residues modi¢ed was determined spectrophotometrically at 410 nm, using a molar absorption coe¤cient of 18 450 M31 cm31 [23]. The acetone concentration in the reaction mixture did not exceed 5% (v/v) and had no e¡ect on the activity and stability of the enzyme during the incubation period.

2.3.1. Modi¢cation of tryptophan residues

2.3.2. Modi¢cation of lysine residues

2.3.1.1. Reaction with NBS. The enzyme solution (2 WM) in 100 mM sodium acetate bu¡er, pH 5.5 was titrated with increasing concentrations of freshly prepared NBS (5^100 WM). The reagent was added in 10 additions of 10 Wl each. After each addition an aliquot was removed and quenched with 20 Wl L-tryptophan (25 mM) and the residual activity was determined under standard assay conditions. Enzyme samples incubated in the absence of NBS served as control. The NBS-mediated inactivation was also monitored spectrophotometrically, by measuring the decrease in absorbance at 280 nm. The number of

2.3.2.1. Reaction with TNBS. The enzyme (2 WM), in 4% sodium bicarbonate, pH 8.5 was incubated with TNBS in the dark at 37³C for 2 h. Aliquots were withdrawn at regular intervals of time and the reaction was terminated by adjusting the pH to 4.5. The residual activity was determined under standard assay conditions. Enzyme samples incubated in the absence of TNBS served as control. The number of amino groups modi¢ed was quantitated spectrophotometrically, using a molar absorption coe¤cient of 9950 M31 cm31 at 335 nm [24] for the trinitrophenylated lysine.

Table 1 Puri¢cation of L-galactosidase from Rhizomucor sp. Step

Protein (mg)

Activity (U)

Speci¢c activity (U mg31 )

Fold puri¢cation

Recovery (%)

Culture ¢ltratea (NH4 )2 SO4 precipitation DEAE-cellulose-I DEAE-cellulose-II Sephacryl S-300-I Sephacryl-S-300-II

750 209 70 50 4.6 4.0

2500 2200 1800 1650 900 800

3.3 10.5 25.71 33.0 195 200

3.18 7.8 10.0 59.0 60.0

100 88.0 72 66.0 36 32.0

a

From an extract of 500 ml.

BBAGEN 24887 1-10-99

318

S.A. Shaikh et al. / Biochimica et Biophysica Acta 1472 (1999) 314^322

Arginine and histidine residues were modi¢ed with phenylglyoxal and diethylpyrocarbonate (DEPC), respectively [25,26]. The modi¢cation of carboxylate and sulfhydryl groups was carried out by treating the enzyme with 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDAC) and 2,2P-dithiobis nitrobenzoic acid (DTNB), respectively [27,19]. Tyrosine was modi¢ed with N-acetylimidazole (NAI) [28]. The residual activity on modi¢cation was estimated after removal of excess reagent by dialysis. 2.3.3. Substrate protection studies In all the chemical modi¢cation studies the e¡ect of substrate was studied by incubating the enzyme with excess amounts of lactose (20 mM) followed by treatment with various modifying reagents. 3. Results and discussion 3.1. Puri¢cation and properties of L-galactosidase from Rhizomucor sp. The results of a typical procedure for the puri¢cation of L-galactosidase from Rhizomucor sp. are given in Table 1. The enzyme was puri¢ed 60-fold with an overall yield of 32% and a speci¢c activity of 200 U mg31 . Although most of the coloring impurities and other contaminating proteins were bound to the DEAE-cellulose, it was observed that the enzyme eluted from the DEAE-cellulose chromatography step contained considerable amounts of colored impurities, which could be removed by treatment with PEG. The enzyme obtained after gel ¢ltration moved as a single band in native and SDS-PAGE indicating its homogeneity (Fig. 1A,B). The properties of L-galactosidase from Rhizomucor sp. are summarized in Table 2. The molecular mass of the enzyme was found to be 250 000 as determined by gel ¢ltration on Sephacryl S-300. The enzyme moved as a single band both in the presence and in the absence of L-mercaptoethanol on SDS-PAGE indicating that the enzyme was a dimer of identical subunits with a relative subunit molecular mass of 120 000 (Fig. 1B), which do not contain any intersubunit S-S bridges. The enzyme contained 8% neutral sugar and focussed at pH 4.2. This observation is consistent with that of the mesophilic fungal L-ga-

Fig. 1. Electrophoresis of L-galactosidase from Rhizomucor sp. (A) Native PAGE (7.5% w/v acrylamide gel, pH 8.9). Lane 1, ammonium sulfate precipitate ; lane 2, enzyme eluted after DEAE-cellulose (II); lane 3, enzyme eluted after Sephacryl S300 (II). (B) SDS-PAGE (10% w/v acrylamide gel, pH 8.9). The molecular masses are indicated beside the gel. Lane 1, puri¢ed L-galactosidase; lane 2, marker proteins (Sigma SDS-6H) carbonic anhydrase (29 000), egg albumin (45 000), bovine serum albumin (66 000), phosphorylase b (97 000) and myosin (205 000); lane 3, puri¢ed L-galactosidase treated with L-mercaptoethanol.

lactosidases reported so far [4,29], the only exception being the L-galactosidase from Kluyveromyces lactis which showed an unusually high carbohydrate content of 45% [31]. L-Galactosidases with identical subunits have been reported for Aspergillus oryzae (Mr 220 kDa), Thermomyces lanuginosus (Mr 220 kDa) and K. lactis (Mr 228 kDa) [10,11,30]. Variations in the molecular mass of L-galactosidases have also Table 2 Properties of puri¢ed L-galactosidase from Rhizomucor sp. Property

Value

Molecular mass by Sephacryl S-300 by SDS-PAGE Carbohydrate content pI Optimum pH Optimum temperature Temperature stability Inhibitors Speci¢city

250 kDa 120 kDa 8% neutral sugar 4.2 4.5 60³C t1=2 150 min at 70³C galactose, IPTG, Hg2‡ , Cu2‡ speci¢c for L-galactopyranosides

BBAGEN 24887 1-10-99

S.A. Shaikh et al. / Biochimica et Biophysica Acta 1472 (1999) 314^322 Table 3 Amino acid composition of L-galactosidase from Rhizomucor sp. Amino acid

Number of residues mol31

Gly Ser Cysa Ala Pro Val As(x) Thr Gl(x) Ile Leu His Lys Met Phe Tyr Arg Trpb

241 153 6 246 107 155 305 150 177 105 148 21 106 33 132 107 50 54

a

Total cysteine was determined spectrophotometrically by the method of Habeeb [20]. b Total tryptophan was determined by the method of Spande and Witkop [21].

been observed, e.g. the enzyme from Mucor pusillus and the phytopathogenic fungus Ophiostoma novoulmi exhibited a monomeric L-galactosidase with Mr 131 kDa and 135 kDa, respectively [30,32], whereas the E. coli L-galactosidase is a homotetramer with subunit Mr 116 kDa [6]. The amino acid composition (Table 3) showed a high content of aromatic amino acids (11% of the total residues) and a low content of cysteine residues (0.3% of the total residues). The L-galactosidase from Rhizomucor sp. also shows a low Arg/Lys ratio (0.5). A similar low Arg/Lys ratio and low cysteine content were observed in the thermostable L-galactosidase of Sulfolobus solfataricus [32], indicating that the Rhizomucor L-galactosidase could also be a thermostable enzyme. 3.1.1. E¡ect of pH and temperature on Lgalactosidase The optimum pH for the enzyme activity was 4.5 and it was stable over a pH range of 3.5^7.5 ( s 90%). The optimum pH for fungal L-galactosidases was found to be 2.5^5.0. In this respect the L-galactosidase from Rhizomucor sp. compares well

319

with the other reported fungal L-galactosidases [7^9]. The pH stability pro¢le and acidic pH optimum suggest the potential use of the enzyme in the hydrolysis of whey as well as in the treatment of milk lactose. The optimum temperature for the L-galactosidase from Rhizomucor sp. was 60³C (Table 2). At its optimum temperature (60³C) the enzyme was stable for 4 h and the t1=2 of the enzyme at 70³C was 2.5 h. The thermal stability of the L-galactosidase was found to be superior to that of the L-galactosidases from F. moniliforme (t1=2 6 h at 50³C) [7], A. oryzae (60³C for 2 h) [8] and Sclerotinia sclerotium (t1=2 of 30 min at 65³C) [33]. The enzyme from a bacterial source, viz. Bacillus stearothermophilus, was stable for less than 10 min at 60³C [34]. The Arrhenius plot of L-galactosidase from Rhizomucor sp. exhibited a straight line for the synthetic substrates, oNPG and pNPG (Fig. 2A,B). The energy of activation (Eact ) was 27.3 kCal mol31 for pNPG and 9.06 kCal mol31 for oNPG (Fig. 2A,B). Linear Arrhenius plots were also observed in the case of the L-galactosidases from F. moniliforme (Eact 8500 Cal mol31 for oNPG) [7] and M. pusillus (Eact 26 kJ mol31 for oNPG) [29]. The substrate speci¢city studies indicated that the enzyme hydrolyzed pNPG and oNPG besides the natural substrate lactose. The Km and Vmax for pNPG and oNPG were 0.66 mM and 1.32 mM; and 22.5 mmol min31 mg31 and 4.45 mmol min31 mg31 , respectively. The Km for lactose was observed to be 50 mM indicating that the enzyme shows a higher a¤nity for the synthetic substrates (Table 4). A similar preference for synthetic substrates has been observed in the case of the enzyme from Aspergillus fonsecaeus [9] and M. pusillus [30]. The Km values were also comparable to those reported in the case of L-galactosidase from other fungal sources [10,11]. Besides lactose, galactooligosaccharides and fragments of plant cell wall polysaccharides may also act as biological substrates of some of the L-galactosidases from fungi [35]. The enzyme derived from Table 4 Substrate speci¢city of L-galactosidase from Rhizomucor sp. Substrate

Km (mM)

Vmax (mmol min31 mg31 )

pNPG oNPG Lactose

0.66 1.32 50.0

22.14 4.45 12.0

BBAGEN 24887 1-10-99

320

S.A. Shaikh et al. / Biochimica et Biophysica Acta 1472 (1999) 314^322

Fig. 2. Arrhenius plot for L-galactosidase from Rhizomucor sp. (A) Using saturating concentrations of oNPG as substrate (25^350 WM). (B) Using saturating concentrations of pNPG as substrate (25^350 WM). Data points are the average of three independent sets of experiments þ S.D. 6 5%. The energy of activation was calculated from the slopes of each linear segment.

Tritrichomonos foetus hydrolyzed lacto-N-biose 1 (GalL1-3-GlcNAc) and N-acetyllactosamine (GalL14-GlcNAc) [36]. Associated activities like L-D-glucosidase and K-L-arabinosidase activity along with the L-galactosidase activity are also known to be present in some cases [37], but the puri¢ed L-galactosidase from Rhizomucor sp. did not possess any other glycosidase activity as tested against pNP-L-D-Glc, pNPK-D-GlcNAc, pNP-L-D-Man and pNP-L-D-Xyl. Galactose and IPTG inhibited the enzyme competitively with a Ki value of 2 mM and 12.0 mM, respectively. 3.2. E¡ect of metal ions on L-galactosidase activity The L-galactosidase from Rhizomucor sp. does not require alkaline or divalent metal ions for its activity. No inactivation was observed as a result of extensive dialysis of the enzyme solution against bu¡ers containing EDTA or by adding chelating agents directly to the assay mixture indicating that it is not a metal requiring enzyme nor a metalloprotein. Divalent metal ions like Hg2‡ and Cu2‡ inhibited the enzyme, while none of the other metal ions had any signi¢cant e¡ect on the enzyme activity. It is known that Mg2‡ is required for enzyme activity in most L-galactosidases, but those derived from fungi, especially the thermostable enzymes, do not seem to require metal ions for their activity [38].

3.3. Chemical modi¢cation studies Treatment of the enzyme with NAI, EDAC, phenylglyoxal, DEPC and DTNB did not lead to any loss of enzyme activity indicating that tyrosine, carboxyl, arginine, histidine or sulfhydryl groups do not have a role in the activity of the enzyme (data not shown). The thermal stability of the enzyme treated with various group-speci¢c reagents was also determined. The native enzyme has a t1=2 of 150 mi. at 70³C, while the DEPC-modi¢ed enzyme has a t1=2 of 125 min at 70³C. The L-galactosidase from Rhizomucor sp. modi¢ed with EDAC, PHMB, phenylglyoxal and NAI showed a t1=2 of 130 min, 140 min, 130 min and 140 min at 70³C, respectively. Therefore the modi¢cation of histidine, carboxylate groups, cysteine, arginine or tyrosine residues does not a¡ect the thermal stability of the enzyme. Puri¢ed L-galactosidase on incubation with a maximum of 20 WM NBS at pH 5.5 and 30³C lost all of its activity. However, no loss of activity was observed in the control samples. The NBS-mediated inactivation of the enzyme was concentration-dependent and was accompanied by a decrease in absorbance of the modi¢ed enzyme at 280 nm. Based on a molar absorption coe¤cient of 5500 M31 cm31 at 280 nm [21], and the subunit Mr of the enzyme of 120 kDa, the total number of tryptophan residues

BBAGEN 24887 1-10-99

S.A. Shaikh et al. / Biochimica et Biophysica Acta 1472 (1999) 314^322

321

Fig. 3. Plot of percent residual activity versus the number of residues modi¢ed. (A) Modi¢cation of L-galactosidase from Rhizomucor sp. with NBS. (B) Modi¢cation of L-galactosidase from Rhizomucor sp. with TNBS.

modi¢ed was found to be 8. However, the plot of percent residual activity versus the number of residues modi¢ed showed that the loss of activity resulted from the modi¢cation of a single tryptophan residue per monomer of the enzyme (Fig. 3A). Similar results were obtained when the modi¢cation was carried out with a tryptophan-speci¢c reagent, viz. HNBB. The NBS- as well as HNBB-mediated inactivation of L-galactosidase could be prevented signi¢cantly ( s 75%) by incubating the enzyme with excess amounts of the natural substrate, lactose (50 mM), prior to the modi¢cation reaction. Determination of the number of tryptophan residues modi¢ed by NBS/ HNBB in the absence and presence of lactose gave values of 4 and 3, respectively (calculated on the basis of subunit Mr of 120 kDa), suggesting the involvement of one tryptophan per monomer, in the activity of the enzyme (Table 5). Puri¢ed L-galactosidase, when incubated with 1 mM TNBS at pH 8.0 and 37³C, lost 60% of its initial activity and the inactivation was concentration-dependent. No loss of activity was observed in the control samples. The plot of the number of amino groups modi¢ed versus the percent residual activity revealed that the loss of activity resulted from the modi¢cation of one residue per monomer of the en-

zyme (Fig. 3B). The inhibition of the enzyme activity by TNBS was prevented to a signi¢cant extent by incubating the enzyme with excess lactose (50 mM) prior to the modi¢cation reaction. Determination of the number of lysine residues modi¢ed by TNBS in the absence and presence of lactose gave values of 4 and 3, respectively (calculated on the basis of subunit Mr of 120 kDa), suggesting the involvement of one lysine per monomer, in the catalytic activity of the enzyme (Table 5). Table 5 In£uence of modi¢cation of tryptophan (A) and lysine (B) on activity of L-galactosidase from Rhizomucor sp. Incubation mixture

(A) Control Enzyme+NBS (5 WM) Enzyme+lactose (50 mM)+NBS Enzyme+HNBB (5 mM) Enzyme+lactose (50 mM)+HNBB (B) Enzyme+TNBS (3 WM) Enzyme+lactose (50 mM)+TNBS a

Residual activity (%)

Number of residues modi¢eda

100 30 73 30 75

0 4 3 3 2

27 71

4 3

Number of residues modi¢ed is calculated based on a subunit Mr of 120 kDa.

BBAGEN 24887 1-10-99

322

S.A. Shaikh et al. / Biochimica et Biophysica Acta 1472 (1999) 314^322

Investigations on the active site characterization of L-galactosidase from E. coli have shown the involvement of tyrosine and carboxyl groups in the catalytic activity of the enzyme. A comprehensive study of the active site components and mechanism of action of L-galactosidase from E. coli showed that Trp and Glu are important for the activity of the enzyme [5]. However, a single tryptophan and a single lysine residue were observed to be important for the activity of L-galactosidase from Rhizomucor sp. In conclusion, the high optimum temperature, acidic pH optimum and its ability to hydrolyze Llinked galactose residues suggest the potential use of this enzyme in the hydrolysis of whey. The stability of the enzyme at pH 6^7 also suggests its potential use in the hydrolysis of milk lactose. Acknowledgements The authors are grateful to Dr. V. Shankar for his critical suggestions during the preparation of the manuscript. The ¢nancial grant from the Council of Scienti¢c and Industrial Research (CSIR), India, to S.A. Shaikh is gratefully acknowledged. References [1] H.H. Nijpels, in: G.G. Brich, N. Blakebrough, N.J. Parker (Eds.), Enzymes and Food Processing, Applied Science Publishers, London, 1981, pp. 89^104. [2] T.P. Shukla, CRC Crit. Rev. Food Technol. 1 (1975) 325^ 356. [3] V. Gekas, M. Lopez-Leiva, Process Biochem. 20 (1985) 2^ 12. [4] B.K. Lonsane, N.P. Ghildyal, in: H.W. Doelle, D.A., Mitchell, C.E. Rolz (Eds.), Solid Substrate Cultivation, Elsevier Applied Sciences, London, 1992, pp. 191^209. [5] R.E. Huber, M.N. Gupta, S.K. Khare, Int. J. Biochem. 26 (1994) 309^318. [6] R.H. Jacobson, X.-J. Zhang, R.F. DuBose, B.W. Matthews, Nature 369 (1994) 761^766. [7] B.J. Macris, P. Markakis, Appl. Environ. Microbiol. 41 (1981) 956^958. [8] J. Fiedurek, A. Gromada, J.J. Jamroz, Basic Microbiol. 36 (1996) 27^32. [9] R.R. Gonzalez, P. Monsan, Enzyme Microbiol. Technol. 13 (1991) 349^352.

[10] L. Fischer, C. Scheckermann, F. Wagner, Appl. Environ. Microbiol. 61 (1995) 1497^1501. [11] J.-L. Berger, B.H. Lee, C. Lacroix, Biotechnol. Appl. Biochem. 25 (1997) 29^41. [12] S.A. Shaikh, J.M. Khire, M.I. Khan, J. Ind. Microbiol. Biotechnol. 19 (1997) 239^245. [13] P. Tinder, J. Clin. Pathol. 22 (1969) 158^161. [14] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, J. Biol. Chem. 193 (1951) 265^275. [15] B.J. Davis, Ann. NY Acad. Sci. 121 (1964) 404^427. [16] H. Blum, H. Beier, H.J. Gross, Electrophoresis 8 (1987) 93^ 99. [17] O. Vesterberg, Biochim. Biophys. Acta 257 (1972) 11^19. [18] U.K. Laemmli, Nature 227 (1970) 680^685. [19] A.F.S.A. Habeeb, Methods Enzymol. 25 (1972) 457^464. [20] T.F. Spande, B. Witkop, Methods Enzymol. 11 (1967) 498^ 532. [21] M. Dubois, K.K. Gills, J.K. Hamilton, P.A. Rebers, F. Smith, Anal. Chem. 28 (1956) 350^356. [22] S.J. Rinderle, I.J. Goldstein, K.L. Matta, R.M. Ratcli¡e, J. Biol. Chem. 264 (1989) 16123. [23] H.R. Horton, D.E. Koshland Jr., Methods Enzymol. 25 (1972) 468^477. [24] A.F.S.A. Habeeb, Anal. Biochem. 14 (1966) 328^336. [25] K. Takahashi, J. Biol. Chem. 243 (1968) 6171^6179. [26] J. Ovadi, S. Libor, P. Elodit, Acta Biochim. Biophys. (Budapest) 2 (1967) 455^458. [27] D.B. Pho, C. Rouston, A.N.T. Tot, L.A. Pradel, Biochemistry 16 (1977) 4533^4537. [28] J.F. Riordan, W.E.C. Wacker, B.L. Valle, Biochemistry 4 (1965) 1758^1765. [29] S.A. Ismail, S.S. Mabrouk, R.R. Mahoney, J. Food Biochem. 21 (1997) 145^162. [30] D. Cavaille, D. Combes, Biotechnol. Appl. Biochem. 22 (1995) 55^64. [31] T. Binz, C. Gremaud, G. Canevascini, Can. J. Microbiol. 43 (1997) 1011^1016. [32] F.M. Pisani, R. Rella, C.A. Raia, C. Rozzo, R. Nucci, A. Gambacorta, M. DeRosa, M. Rossi, Eur. J. Biochem. 187 (1990) 321^328. [33] C. Riou, G. Freyssinet, M. Fever, FEMS Microbiol. Lett. 95 (1992) 37^42. [34] R.E. Goodman, D.M. Pederson, Can J. Microbiol. 22 (1976) 817^825. [35] H. Nakano, S. Takenishi, Y. Watanabe, Agric. Biol. Chem. 24 (1987) 2267. [36] M. Vella, P. Greenwell, Glycoconjug. J. 14 (1997) 883^ 887. [37] N. Onoshi, T. Tanaka, Lett. Appl. Microbiol. 24 (1997) 82^ 86. [38] R.E. Huber, N.J. Royh, H.J. Bahl, Protein Chem. 15 (1996) 621^629.

BBAGEN 24887 1-10-99