properties of a novel glucose-enhanced β-glucosidase purified from streptomyces sp. (atcc 11238)

E L S E V I E R Biochimica et Biophysica Acta 1251 (1995) 145-153

BB Biochi~ic~a et Biophysica A~ta

Properties of a novel glucose-enhanced fl-glucosidase purified from Streptomyces sp. (ATCC 11238)

Josep A. Ptrez-Pons, Xavier Rebordosa, Enrique Querol *

Institut de Biologia Fonamental and Departament de Bioqulmica i Biologia Molecular, Universitat Autbnoma de Barcelona, 08193 Bellaterra, Barcelona, Spain

Received 9 December 1994; revised 5 April 1995; accepted 13 April 1995

Abstract

An inducible intracellular /3-glucosidase (EC 3.2.1.21) from Streptomyces sp. QM-B814 (ATCC 11238) has been purified and characterized. The purified polypeptide is monomeric with a relative molecular mass of 62 kDa by SDS-PAGE and 42 kDa by size-exclusion chromatography; its isoelectric point is 4.2. The difference in the molecular mass values can be attributed to the glycosylated nature of the protein. The purified enzyme has a pH optimum of 6.0-6.5. The temperature optimum for activity is 50°C; at this temperature the enzyme is stable for 1 h. The enzyme hydrolyzes mainly aryi-/3-glucosides but also presents significant activity against fl-linked disaccharides and maltose. The enzyme displays an unusual kinetic behavior and biphasic Lineweaver-Burk and Eadie-Hofstee plots for p-nitrophenyl-fl-D-glucoside and cellobiose were obtained. The enzyme presents /3-glycosyltransferase activity and an exoglycosidase-type action on cellodextrins. It is inhibited by 6-gluconolactone (K i 0.44 mM) but, remarkably, glucose in the range 25-200 mM enhances the rate of p-nitrophenyl-/3-D-glucoside hydrolysis.

Keywords: /3-Glucosidase; Purification; Characterization; Glucose-enhanced glucosidase; (Streptomycete)

1. Introduction

/3-Glucosidases (/3-glucoside glucohydrolase, EC 3.2.1.21) are enzymes that catalyze the transfer of glucosyl groups between oxygen nucleophiles. In physiological conditions, such a transfer reaction generally results in the hydrolysis of a /3-D-glucosidic bond. /3-Glucosidases are widely distributed in microorganisms, plants, and mam- mals, playing different metabolic roles from the utilization of primary carbon sources to the resistance against phy- topathogens, or preventing the accumulation of certain compounds, such as glucocerebrosides, in tissues. How- ever, fl-glucosidases have been traditionally referred as one of the three enzyme classes composing the 'cellulase' system [1-3]. In cellulolytic processes, /3-glucosidases catalyze the hydrolysis of cellobiose and cellooligosaccharides to glucose, and usually represent the bottleneck be-

Abbreviations: DMAB, 3-dimethylaminobenzoic acid; MBTH, 3- methyl-2-benzothiazolinone hydrazone; MUG, 4-methylumbelliferyl-fl- D-glucopyranoside; pNPC, p-nitrophenyl-fl-D-cellobioside; pNPG, p- nitrophenyl-/3-D-glucopyranoside; pNPX, p-nitrophenyl-fl-D-xyloside.

* Corresponding author. Fax: +34 3 5812011.

0167-4838/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0167-4838(95)00074-7

cause of their low amount in crude cellulase preparations [4]. Moreover, their substrates are inhibitors of the exo- and endo-glucanases. In turn, most microbial /3-glucosidases are subjected to glucose inhibition [5]. To this respect, the availability of enzymes insensitive to product inhibition can represent an important goal to attain the enzymatic conversion of lignocellulosics becomes a com- mercially interesting process [6].

Actinomycetes are bacteria well adapted to grow in soil and, therefore, possess multiple enzyme systems for the hydrolysis of natural polymers [7-9]. Although species of Streptomyces, an actinomycete genus, appeared in the early studies on cellulolytic microorganisms [ 10,11], their ligno- cellulase systems only recently have received much atten- tion (reviewed in [12] and [13]). Streptomyces sp. QM- B814 (ATCC 11238) is a cellulolytic organism which presents a whole glycohydrolytic enzyme set we have studied [14]. Its fl-glucosidase activity is inducible, intracellular, and presents more than one component as it has been found in other streptomycete species [15-17] and other bacteria, such as Cellulomonas [18], Ther- momonospora [19], Bacillus [20], or Clostridium [21]. In relation to their substrate specificity or preference different

146 J.A. P~rez-Pons et al. / Biochimicu et Biophysica Acta 1251 (1995) 145-153

types of fl-glucosidases have been proposed [22]. The mode of attack and the hydrolysis of reduced cellodextrins have been studied in fungal glucanases [23] and glucosidases [24] but little information is available about enzymes of bacterial origin.

In a previous paper we described the purification and characterization of a cloned /3-glucosidase (Bgl3) from QM-B814 strain as expressed in a fl-glucosidase-negative mutant of S. lividans [25]. That enzyme displays kinetic properties different from those of the fl-glucosidase (Bgl l) herein reported, which would represent a second inducible intracellular /3-glucosidase whose main feature is to be insensitive to glucose-inhibition.

instructions but bubbling N 2 to avoid sugar oxidation during the coupling reaction performed at pH 12-12.5. The cellobiose-Sepharose column (1 × 5 cm) was equilibrated with 50 mM sodium acetate buffer (pH 5.5; starting buffer). Sample was loaded, column washed thoroughly with starting buffer until A280 was constant and specific elution carried out with 20 ml 25 mM cellobiose in starting buffer at pH 6.5 at a flow rate of 5 ml/h. Fractions of 1 ml were collected and elution monitored at 280 nm. After each run, the column was washed with 20 ml starting buffer containing 2 M NaCI.

2.3. Determination of M r and pl

2. Materials and methods

2.1. Organism and growth conditions

Streptomyces sp. (ATCC 11238) was obtained from the Colecci6n Espafiola de Cultivos Tipo (CECT No. 3145) and maintained on 7 to 10-day-old sporulated slant-cultures of potato infusion-agar. Spore suspensions were pre- pared from slants with sterile 0.85% (w/v) NaC1, containing 1% (v /v ) Tween-80 and, either used to inoculate liquid cultures or kept frozen at -20°C in 20% (w/v ) glycerol. To induce the fl-glucosidase activity, cultures were performed in two steps: 50 ml (per 250 ml erlen- meyer flask) basal medium (BM) [26], supplemented with 0.5% glycerol as a carbon source, were incubated at 30°C on a rotatory shaker (240 rpm) for 24-30 b. These cultures were used as a 5-10% (v /v ) inoculum for a second culture in BM, containing 0.25% glycerol. Secondary cultures (1 liter) were incubated for 20-24 h as described above, then induced by adding cellobiose at a final concentration of 0.5-1% (w/v) , and the incubation continued for 6-8 h before collecting mycelia.

2.2. Protein purification

Mycelia from cellobiose-induced cultures were used as enzyme source. Purification of Bgll glucosidase was as previously described [25], and included selective ammonium sulphate precipitation and two anion exchange chromatographic steps. Hydroxyapatite and affinity chromatographic setups were tested as alternative purification proce- dures. Hydroxyapatite (Bio-Gel HTP, Bio-Rad) chromatography was done at 4°C on a column (0.8 X 14 cm) equilibrated with 15 mM sodium phosphate buffer (pH 6.8), and elution of the adsorbed proteins was performed by applying a 15-250 mM linear gradient (50 ml) of the former buffer at a flow rate of 20 ml/h. Fractions of 1 ml were collected and elution monitored at 280 nm. Affinity chromatography was done at 4°C using cellobiose as a ligand coupled to an epoxy-activated Sepharose matrix (Pharmacia Biochemicals) following the manufacturer's

The relative molecular mass (M r) was determined either by SDS-PAGE and by analytical size-exclusion HPLC. Discontinuous SDS-PAGE in reducing conditions was performed according to the procedure of Laemmli [27]. The M r was estimated by densitometric analysis of Coomassie brilliant blue R-250 stained gels including M r standards (kit Combithek, Boehringer, Mannheim, Germany), and using a Beckman DU-8B Slab-Gel spectrophotometer. An- alytical size-exclusion HPLC was performed at room temperature on a Spherogel-TSK G3000SW column (300 mm X 7.5 mm), equilibrated with 20 mM Tris-HC1 buffer (pH 6.8) containing 0.2 M NaCi, at a flow rate of 0.5 ml/min. Column calibration was done using as M r standards: fer- ritin/apoferritin (480-440 kDa), catalase (220 kDa), BSA (60.3 kDa), ovalbumin (43.5 kDa), chymotrypsinogen (24 kDa) and bovine RNAase A (13.7 kDa), all of them from Sigma (St. Louis, MO, USA), except RNAase, which was from Calbiochem.

Narrow range isoelectric focusing, pH 4.0-6.5, was performed in a 5% (w/v ) polyacrylamide gel containing carrier ampholytes (Pharmalyte, Pharmacia, Uppsala, Swe- den), according to the method of Robertson et al. [28]. The pH gradient was measured by a surface electrode pH-me- ter. After running, gels were stained with Coomassie brilliant blue R-250 and fl-glucosidase activity was detected by overlaying the gel with a solution containing 2 mM 4-methylumbelliferyl-fl-D-glucoside (MUG; Sigma, St. Louis, MO, USA) in 50 mM sodium phosphate buffer (pH 6.5), and incubating the overlay for 15-30 min at 37°C. Activity bands were visualized under UV light; fluores- cence was enhanced by soaking the gel in 1 M Na2CO 3 solution.

2.4. Amino-acid analysis

The amino-acid analysis was performed using the PICO TAG method (Millipore Waters Associates, Milford, MA, USA). For the analysis, triplicates of about 75-100 pmol of pure enzyme were dried by vacuum centrifugation, dissolved in 25 ml 6 N HC1 (Sequanal grade, Pierce, Rockford, IL, USA), containing 0.1% phenol, and left to hydrolyze for 24 h at l l0°C. As standard was used a

J.A. P~rez-Pons et al . /Biochimica et Biophysica Acta 1251 (1995) 145-153 147

mixture containing 250 pmol of each amino acid (Amino acid Standard H, Pierce).

2.5. Enzyme assays

/3-Glucosidase activity was measured using p- n i t rophenyl- /3-D-glucoside (pNPG; Boehr inger , Mannheim), cellobiose (Fluka, Buchs, Switzerland) and p-nitrophenyl-/3-D-cellobioside (pNPC; Sigma) as substrates. Assays were performed incubating aliquots of the sample in a total volume of 0.6 ml 50 mM sodium phosphate buffer (pH 6-6.5) containing substrates at either final concentration of 5 mM pNPG, 10 mM cellobiose or 2 mM pNPC. Mixtures were incubated at 50°C for up to 10 min. When using pNPG, 0.5 ml 1 M Na2CO 3 were added and the p-nitrophenol produced was calculated assuming an extinction coefficient of 18800 cm ~ M -~ at 400 nm. The release of glucose from cellobiose and pNPC was determined using the Glucose UV-method kit (Boehringer, Mannheim) after stopping reactions by heating at 100°C for 5 min. One unit (U) of /3-glucosidase activity was defined as the amount of enzyme that releases 1 /zmol p-nitrophenol or glucose per min. For the enzyme characterization, substrates (see Table 2) were used at a concentration of 10 mM, except for pNPG, pNPC, and p- nitrophenyl-/3-D-xyloside (pNPX, Sigma) that were 5 mM, 2 mM, and 5 mM, respectively. Reaction mixtures were incubated at 50°C for up to 10 min. The p-nitrophenol released from pNPG and pNPX, and the glucose released from all the other substrates were measured as above described. Protein concentration was determined by the dye-binding method of Bradford [29] using BSA as a standard.

Kinetic constants were measured using different amounts of pNPG (0.1-100 mM) or cellobiose (0.25-250 mM), and performing the standard activity assays above described. Data were analyzed using fitting software and double reciprocal plots. The effect of potential inhibitors on enzyme activity such as 6-gluconolactone (0-5 mM) and glucose (0-500 mM) was tested by co-incubation with pNPG (1-10 mM). All incubations were done for 5 min at 50°C and pH 6.0 following measurement of the p- nitrophenol released. Data were analyzed by plotting the double reciprocals of hydrolysis rate and substrate concentration at each inhibitor concentration.

Thin-layer chromatography (TLC) analyses of pNPG and pNPC hydrolysis reactions were performed on silica- gel plates using ethyl acetate/methanol/water (4:1.5:2, v / v ) as a solvent. Plates were developed for 10-15 min at 120°C after spraying with phenol-sulfuric acid.

2.6. Effect of pH, temperature and chemicals

The pH optimum for /3-glucosidase activity was determined by using either pNPG (5 mM) and cellobiose (10 mM) as substrates and performing the enzyme assay as

above described. Temperature optimum was determined by running the enzyme assay at different temperatures (30- 80°C) with pNPG (5 mM) as substrate.

The effect of several metal ions and compounds was assayed by including the different chemicals at different final concentrations in the standard reaction mixture used for the determination of enzyme activity with pNPG (5 mM).

2.7. HPLC analysis of the ~-glucosidase activity

Time-course hydrolysis of cellooligosaccharides was followed by HPLC. Aliquots of purified /3-glucosidase were incubated with 10 mM of either cellotetraose, cellotriose, or cellobiose in 50 mM sodium phosphate buffer (pH 6.5) at 50°C. Aliquots were withdrawn at different times and reactions were inactivated by heating at 100°C for 5 min. Samples were centrifuged at 12000 × g for 5 min and 10 /xl portions of the supernatants were injected onto a Bio-Rad Aminex HPX-42A column heated at a constant 85°C. Elution was performed by using water at a flow rate of 0.6 ml/min and detection was by differential refractometry (Waters R401 refractometer). Sugar elution patterns were registered and integrated on a Waters 746 data module. Identity of the products was determined using glucose, cellobiose, cellotriose, and cellotetraose as a standards.

The /3-glycosyltransferase activity of the purified enzyme was also analyzed by HPLC. Reaction mixtures containing 6 /zg pure enzyme, 0.4 M cellobiose, and 10% (v /v ) ethanol in 50 mM sodium phosphate buffer (pH 6.5) were incubated at 50°C for 1 h and stopped by heating at 100°C for 5 min before the analysis; portions of 10 /xl were analyzed by HPLC as above described.

2.8. Protein glycosylation

Protein glycosylation was checked by incubating different amounts (0.2-1.5 /xg) of purified /3-glucosidase with specific glycosidases (endoglycosidase F, endoglycosidase H and neuraminidase; Boehringer, Mannheim), according to the manufacturer's specifications. Parallel aliquots were run without glycosidase. Samples were applied into wells of a microtiter plate and left to adsorb overnight at room temperature in a wet chamber. Then, plates were developed by means of specific lectin affinity (concanavalin A specific for mannose and N-acetylglucosamine, the endoglycosidases F and H substrates, and lectin from Limulus polyphemus for sialic and glucuronic acids, the neuraminidase substrates; Boehringer, Mannheim), following a standard ELISA sandwich assay: biotin-labeled lectin, peroxidase-avidin conjugate, and DMAB and MBTH as chromogenic peroxidase substrates (according to Boehringer's specifications). The colour developed was measured with a Dynatech MR250 apparatus, equipped

148 J.A. P£rez-Pons et al. / Biochimica et Biophysica Acta 1251 (1995) 145-153

with a 600 nm filter. Glycosylated and non-glycosylated pure fractions of RNAase were used as positive and negative controls, respectively.

170

M a b c d M

3. Results and discussion 97.4 - -

3.1. Enzyme purification 55.4 ' ,~

An intracellular /3-glucosidase was purified from cell- free extracts of Streptomyces sp. QM-B814 following selective ammonium sulfate precipitation and two anion- exchange chromatographic steps. Table 1 summarizes the results of the purification. Data obtained using pNPG, cellobiose, and pNPC as activity substrates are presented. The enzyme was purified around 95-fold with an overall yield of 15%. The ratios pNPGase/cellobiase/pNPCase were practically constant throughout purification steps thus indicating that a single protein is responsible for the three activities at least under the growth conditions used for the enzyme preparation. However, two minor active peaks (representing less than 5% of the total activity) were detected along the elution profile corresponding to the first anion exchange chromatography. The electrophoretic analysis of pure enzyme showed an homogeneous protein band (Fig. 1). In order to check the presence of multiple fl-glucosidase forms, hydroxyapatite and affinity chromatographic runs were performed as alternative purification steps. Purification factors and yields were 72-fold and 7% for hydroxyapatite chromatography, and 76-fold and 4% for affinity chromatography, respectively. In both cases, the same single protein band as above was associated to the eluted activity peak upon analysis by SDS-PAGE (not

36.5 , ~

20.1

J Fig. 1. Electrophoretic analysis of Bgll fl-glucosidase at various stages of purification. Separation was performed on a 12% (w/v) polyacrylamide-SDS gel. Lanes: (a) purified fl-glucosidase from Mono-Q chromatography at pH 6.4; (b) pooled fractions from Mono-Q at pH 7.1; (c) 35-65% ammonium sulfate fraction; (d) crude cell extract; (M) include the following M r markers: reduced oL2-macroglobulin (170 kDa), phos- phorylase b (97.4 kDa), glutamate dehydrogenase (55.4 kDa), lactate dehydrogenase (36.5 kDa), and trypsin inhibitor (20.1 kDa). Gel was stained with Coomassie brilliant blue.

shown). Activity ratios were also similar. According to the results obtained, the standard protocol used for the prepara- tive enzyme purification includes two anion-exchange chromatographic runs (Table 1).

Table 1 Purification of the Bgll fl-glucosidase

Fraction Volume Total activity [Protein] Specific activity Yield Purification factor (ml) (U) (mg/ml) ( U / m g protein) (%)

Activity substrate: pNPG Crude cell extract 62 Ammonium sulfate 10 Mono-Q (pH 7.1) 2.5 Mono-Q (pH 6.4) 0.8

Activity substrate: cellobiose

24.3 8.7 0.045 100 1 15.8 31.0 0.051 65 1.1 8.34 6.65 0.502 34.3 11.2 3.42 1.16 3.690 14. I 82

Crude cell extract 62 12.9 8.7 0.024 100 1 Ammonium sulfate 10 8.35 31.0 0.027 65 1.1 Mono-Q (pH 7.1) 2.5 5.02 6.65 0.302 38.8 12.6 Mono-Q (pH 6.4) 0.8 2.15 1.16 2.311 16.6 96.6

Activity substrate: pNPC Crude cell extract 62 11.1 8.7 0.021 100 1 Ammonium sulfate 10 8.44 31.0 0.027 76 1.3 Mono-Q (pH 7.1 ) 2.5 3.73 6.65 0.224 33.6 10.9 Mono-Q (pH 6.4) 0.8 1.92 1.16 2.066 17.3 100.4

Details of the protocol are described in (see section 2). One unit (U) of fl-glucosidase activity is defined as the amount of enzyme that releases 1 /zmol of p-nitrophenol (from pNPG) or glucose (from cellobiose and pNPC) per min at 50°C.

J.A. Pdrez-Pons et al./Biochimica et Biophysica Acta 1251 (1995) 145-153 149

The pure enzyme was kept at 4°C for short periods. The enzyme was stable for two years when stored at - 20°C in 45-50% (w/v ) glycerol; lyophilization causes the loss of the enzyme activity.

3.2. Molecular mass, isoelectric point, and amino-acid composition

The M r of the purified fl-glucosidase was 62 kDa and 42 kDa when determined by SDS-PAGE (Fig. 1) and analytical size exclusion HPLC, respectively. The enzyme was found to be monomeric. The discrepancy between M r values obtained with both methods could be explained by the presence of sugar residues in the purified fl-glucosidase (see below). Similar differences have been reported for fungal [30] and bacterial fl-glucosidases [31-34], as commonly attributed to a decreased binding of SDS to glycoproteins [35]. In these cases, the preferred M r value for the purified enzyme is that obtained by size-exclusion chromatography (42 kDa). Such a value is of the same order as those reported for other bacterial [4] and streptomycete /3-glucosidases [16,17,33,36]. To check the nature of the glycosylation, fl-glucosidase samples treated with neuraminidase and developed in the presence of Limulus polyphemus lectin, which specifically recognizes the gly- cosylic substrates for neuraminidase, showed a significa- tive decrease in A60 o as compared to untreated samples (not shown). Thus, these results prompt the suggestion that the glycosyl moiety would be mainly constituted by sialic, uronic or glucuronic acids. In contrast, upon ELISA assay no differences between endoglycosidases F and H-treated and untreated enzyme samples and further developing using concanavalin A lectin were observed.

The isoelectric point determined by narrow-range isoelectric focusing was 4.2. The enzyme appeared as a joined double band by Coomassie brilliant blue staining and both bands were active following activity staining using MUG as substrate (not shown), likely due to the presence of protein isoforms. The fi-glucosidases are mostly acidic proteins (ranging from 3.5 to 5.5), and the observed pI value was nearly identical to those reported for other fl-glucosidases from the genus Streptomyces [16,17,33,36].

As expected from the pI value, the amino-acid analysis showed a higher content of Asx and Glx residues compared to Arg and Lys residues. Taking the amino-acid content and 42 kDa as the molecular mass of the fl-glucosidase, the number of residues can be estimated to be about 370. The comparison of the amino-acid composition between the enzyme herein reported and the Bgl3 /3-glucosidase from the same Streptomyces strain showed some small differences in composition. Bgl3 is slightly richer in Arg and His residues [25]. Comparison with other streptomycete fl-glucosidases cannot be done due to the lack of reported sequences or amino-acid composition data.

Table 2 Substrate specificity of the purified 13-glucosidase

Substrate Linkage of Activity glycosyl group (% of that on pNPG)

pNPG flglc 100 pNPC /3 glc/3 ( 1-4)glc 52 Cellobiose [3( 1-4)glc 55 Cellotriose /3(1-4)glc 60 Cellotetraose /3( 1-4)glc 73 Sophorose 13 ( 1-2)glc 57 Laminaribiose 13(1-3)glc 45 Gentiobiose 13(1-6)glc 0 Lactose /3( 1-4)gal 33 Sucrose fructo a ( 1-2)glc 0 Maltose e~ ( 1-4)glc 26 S alicin /3 glc 13 pNPX 13 xyl 2

Details of the protocol are described in (see section 2). The absolute value for pNPG was 3.8_+0.2 U/nag of protein. Data are the means from triplicate assays

3.3. Substrate specificity

As shown in Table 2, maximum activity was obtained with pNPG, indicating that Bgll enzyme displays a great preference for aryl-/3-glucosides. Nevertheless, a significant activity was measured towards fl-disaccharides, such as sophorose ( r - l ,2 ) , cellobiose ( r - l ,4 ) , and laminaribiose ( r-1,3); no activity was detected towards gentiobiose, a fl-l,6-1inked disaccharide. Usually, among fl- glucosidases, the latter linkage is less susceptible to hydrolysis [22,32,37] or it is not hydrolyzed at all [38]. Accord- ing to general observations [22], the hydrolysis rate of soluble cellodextrins by Bgll enzyme increases with the chain length, as observed using cellotriose and cellotetraose as substrates. The purified fl-glucosidase also hydrolyzes lactose to a significant extent, thus indicating an enzyme tolerance for the hydroxyl group at C4 position. The purified enzyme showed some activity on maltose (a- l ,4) indicating a relaxed specificity for c~- or r-lin- kages between glucose moieties. This characteristic is uncommon in fi-glucosidases, and only few cases have been reported [39]. Sucrose was not hydrolyzed, so the enzyme does not possess invertase activity.

3.4. Effect of pH, temperature, metal ions and chemicals on the activity

The pH/activity profiles obtained using pNPG and cellobiose as substrates showed a maximum at pH 6.0 and 6.5, respectively. The purified Bgll /3-glucosidase retained about 20% of its activity at pH 9-10, while a rapid declining was observed at pH < 5.0. The pK l and pK 2 values derived from the corresponding pH/activity profile were 4.97 _+ 0.14 and 8.00 __ 0.13, respectively. Most gly- cohydrolases use Glu/Asp residues as nucleophile/proton donor on the basis of their acid-base catalytic mechanism

150 J.A. P~rez-Pons et al. / Biochirnica et Biophysica Acta 1251 (1995) 145-153

[40]. In these cases, the pK value of the residue acting as proton donor it is assumed to be slightly increased. The temperature optimum for the hydrolysis of pNPG was 50°C. At this temperature, the enzyme retained 100% activity for 1 h; at 60°C, the slopes of product formation started to decline after about 30 min.

Enzyme activity (pNPGase) was tested in the presence of several metal ions and chemicals. A slight (10%) stimulation of the activity was observed with Mn 2+ and Ca2+; moderate (50%) and strong (100%) inhibition was detected in the presence of Cu 2+ and Hg 2+, respectively, while Mg 2+ and Zn 2- practically had no effect; all of them at 1 mM concentration. Strong inhibition was also observed with thiol-specific reagents, such as p-chloromercuri- benzoate, at 0.1 mM concentration. These results, together with a stimulation (5-15%) of the enzyme activity observed in the presence of reducing agents (2-mercapto-

ethanol and dithiothreitol) at concentrations ranging from 1 to 50 raM, suggest the existence of thiol groups essential for activity. Both effects, the inhibitory by Hg 2-- and compounds specifically reacting with thiol groups and the stimulatory by reducing agents, have been described for microbial /3-glucosidases [31,32,37,41], as well for streptomycete enzymes [16,17]. Glycerol (10-25% w /v ) and ethanol (1-10% v /v ) have a stimulatory (5-30%) effect on the enzyme activity. Such an effect could be explained by the /3-glycosyltransferase activity of the enzyme (see below). In this sense, both compounds, acting as glycosyl acceptors competing with water, would increase the rates of p-nitrophenol release. The inhibition observed at higher ethanol concentrations (i.e., 20%) was probably due to denaturation or to conformational changes of the enzyme in apolar environments. The enzyme is fully inactivated in the presence of 1% (w/v ) SDS.

S=pNPG 4 ~ t ~

6"

5"

E 4 -

2"

(a) 7 -

' ' I ' ' ' I ' ' ' I ' ' I ' ' ' I

20 40 60 80 1 O0

[S] (raM)

(b) 7

6

5

4

3

2 -

1-

0

0 ' ' ' ' l . . . . I . . . . I ~ ' ' ; I ' ' ' ' I

1 2 3 4 5 v/IS]

o11 S=cellobiose ~ 7

E 3- 3 -

-

> 2- 2 -

1 1

0 ' ~ ' ' I ' ' ' ~ I ' ' ~ ' I ' ' ' ' I ' ' ' ' I 0

0 50 100 1 S0 200 250 0

[S] (mM)

' ' ' I ' I ' ' ' I ' I '

0.2 0.4 0.6 0.8

v/IS]

' I

1

Fig. 2. Effect of substrate concentration on Bgll fl-glucosidase activity. Initial hydrolysis rates at different concentrations of p-nitrophenyl-fl-D-glucoside (pNPG) (a,b) and cellobiose (c,d). Data are displayed as Michaelis-Menten (a,c) and Eadie-Hofstee (b,d) plots. Each point represents the mean value of at least four determinations, and the standard deviation was less than 10%. [S], substrate concentration; v, initial velocity expressed as izmol of p-nitrophenol (from pNPG) or glucose (from cellobiose) released per min and mg of purified enzyme. Assays were run in 50 mM sodium phosphate buffer, pH 6.0 (pNPG) and 6.5 (cellobiose), at 50°C.

J.A. P£rez-Pons et al . /Biochimica et Biophysica Acta 1251 (1995) 145-153 151

3.5. Kinetic characterization

The effect of substrate (pNPG and cellobiose) concentration on the enzyme activity was analyzed by means of different plots of the initial velocities and substrate concentrations data. Michaelis-Menten diagram followed an hy- perbolic behavior but Lineweaver-Burk and Eadie-Hofs- tee data diagrams gave clearly non-linear plots using the ENZ~qTTER software [42]. Michaelis-Menten and Eadie- Hofstee plots for pNPG and cellobiose are shown in Fig. 2. Biphasic plots for fl-glucosidases have been elsewhere reported [32,41,43] and explained alternatively by ho- motropic negative co-operativity [44], hysteretic co-operativity [43], and by the presence of isoenzymes with different kinetic behavior [45]. In the present work, the latter possibility seems more feasible since pure enzyme preparations showed a double band in isoelectric focused gels, being both bands active towards MUG and not distinguishable by SDS-PAGE. Further, the analysis by TLC of enzymatic reactions towards pNPG and pNPC showed the presence of products likely arising from the fl-glycosyltransferase activity (see below). These products, affecting the catalytic reaction, might alter the hydrolysis measurements. Upon the assumption of the equimolecular presence of two isoenzymes, two apparent K m and Vma x values for pNPG and cellobiose (Table 3) were estimated by the procedure of Spears et al. [45]. The difference between low and high apparent K m values is greater than one order of magnitude for both substrates. Our data are consistent with the general finding that K m (cellobiose) > K m (pNPG). Substrate inhibition was not observed, lead- ing to a relevant kinetic difference between the enzyme herein reported and the Bgl3 glucosidase from the same strain, in which a clear inhibition at pNPG concentrations greater than 0.5 mM was observed [25].

Table 3 Properties of the Streptomyces sp. QM-B814 fl-glucosidases

Property Bgll ~ Bgl3 ~

Molecular mass (nucleotide sequence) N.A. 52.3 kDa Molecular mass (SDS-PAGE) 62 kDa 54 kDa Molecular mass (size-exclusion HPLC) 42 kDa 59 kDa Isoelectric point p l 4.2 p l 4.4 Glycoprotein yes no pH optimum pH 6.0-6.5 pH 6.5 Temperature optimum 50°C 50°C K m (pNPG) 0.12/8.7 mM 0.27 mM Vm~ x (pNPG) 0.45/7.1 U 5.7 U K m (cellobiose) 0.19/29.9 mM 7.9 mM Vma x ( ce l l ob iose ) 0.16/7.4 U 9.3 U K i (glucose) N.D. 65 mM K i (&gluconolactone) 0.44 mM 0.08 mM /3-Glycosyltransferase activity yes yes

Bgll indicates the enzyme from this work; the Bgl3 enzyme corresponds to a cloned /3-glucosidase as expressed in S. lividans [25]. a N.A., not available; N.D., not determined; Vma x units are expressed in U / m g protein.

..---

g

o

o ,A

8-

6-

4"

2 i i ( i i

0 100 200 300 400 500

Glucose concentration [mM]

Fig. 3. Effect of glucose on Bgll /3-glucosidase activity. Activity assays were performed as described in (see section 2) using 5 mM pNPG as substrate in the presence of the indicated glucose concentrations. Assays were run by triplicate; error bars are depicted.

Time-course HPLC analysis of cellotetraose hydrolysis indicated that glucose units were split from the tetrasaccha- ride since only cellotriose and glucose were produced at first. The enzyme, therefore, has an exo-glycosidase nature. The release of glucose would occur from the non-reducing end of the oligosaccharide, as indicated by TLC analysis of pNPC hydrolysis in which can be observed that pNPG and glucose are the primary products. The effect of glucose and &gluconolactone on the enzyme activity was assayed using different concentrations of pNPG as a substrate. From double reciprocal plots, it was concluded that 6-gluconolactone is a competitive inhibitor with a K i value of 0.44 mM. &Gluconolactone is considered one of the most effective reversible inhibitors for /3-glucosidases [46]. On the contrary, glucose in the range about 25-200 mM stimulated the rate of pNPG hydrolysis, reaching a 2-fold factor at approx. 100 mM glucose (Fig. 3). Only at glucose concentrations greater than 300 mM a slight decrease in the pNPG hydrolysis was observed. An activity enhancement by glucose has been elsewhere reported for a cellobiase from the fungus Trichoderma viride [47] and recently for /3-glucosidases from actinomycetes [17,48], although, as far as we know, no reliable explanations have been afforded. In this sense, an extracellular fl-glucosidase isolated from Trichoderma reesei showed an activation of its pNPG hydrolyzing activity upon binding of a cell-wall polysaccharide [49]. The authors state the possibility that upon glycan binding to the glucosidase, the polysaccharide alters the enzyme conformation in such a way as to render the active center better accessible to the substrates. In any case, such a feature can be interesting in order to use these enzymes in industrial processes for the enzymatic hydrolysis of lignocellulosics since glucose usually represents an inhibitory end-product for the fl-glucosidase activity [6,50].

152 J.A. P~rez-Pons et aL / Biochimica et Biophysica Acta 1251 (1995) 145-153

Thus, glucose-enhanced glucosidases would display syn- ergy with exo-glucanases or even with glucose-sensitive /3-glucosidases.

The /3-glycosyltransferase activity of Bgll was analyzed by HPLC of reaction mixtures containing high cellobiose concentration. The chromatograms showed the presence of a putative transglycosylation product with a retention time consistent to that of a trisaccharide as compared to the cellotriose standard (not shown). The /3-glycosyltransferase activity is a common property of the /3-glucosidases so far analyzed, varying the transglycosylation products they form [25,32,51,52].

In conclusion, upon its physico-chemical and enzymatic characterization, the purified Bgll glucosidase displays properties distinct from those of the elsewhere reported Bgl3 glucosidase [25]. Table 3 summarizes the physico- chemical and enzymatic properties of both /3-glucosidase forms isolated from Streptomyces sp. QM-B814. The latter enzyme was subclassified as a cellobiase or exo-glucosidase with broad substrate specificity while Bgll would mainly be an aryl-glucosidase according to its relative activity against different substrates (Table 2). However, by comparison of the ratios of the pseudo-second-order constant (Vmax/K m) for pNPG and cellobiose, it can be observed that Bgl3 has a several-fold higher relative specificity for both substrates. The fact that activity ratios were constant during the progress of the purification (Table 1) suggests that Bgl3 is poorly, if ever, expressed in the culture conditions used for the induction of Bgll. To this respect, immediately upstream of the Bgl3 coding region was found a repeated sequence which could be the target for a positive activator as it has been demonstrated in cellulase-coding genes [53]. The cloning of Bgll gene and the analysis of its coding and promoter sequences will afford further comparisons between both enzymes.

Acknowledgements

We thank M. Salv~ for performing the amino-acid composition analysis. We thank M. Osset for providing control samples for the glycosylation analysis. We thank C. Malet and A. Planas for providing cellotriose and cellotetraose. We thank A. Planas for helpful discussions and suggestions on the manuscript. X.R. is a fellowship recipient from the PFPI of the Ministerio de Educaci6n y Ciencia (Spain). This work has been supported by grants BIO91-0477 and BIO94-0912 from the CICYT (Ministerio de Educaci6n, Spain) and by the Centre de Refer~ncia de R + D de Biotecnologia de la Generalitat de Catalunya.

References

[1] Berghem, L.E.R. and Pettersson, L.G. (1974) Eur. J. Biochem. 46, 295 -305.

[2] Knowles, J., Lehtovaara, P. and Teeri, T. (1987) Trends Biotechnol. 5, 255-261.

[3] B~guin, P. and Aubert, J.P. (1994) FEMS Microbiol. Rev. 13, 25-58.

[4] Woodward, .1. and Wiseman, J. (1982) Enzyme Microb. Technol. 4, 73-79.

[5] Wood, T.M. and McCrae, S.I. (1982) J. Gem Microbiol. 128, 2973-2982.

[6] Klyosov, A.A. (1990) Biochemistry 29, 10577-10585. [7] Ball, A.S. and McCarthy, A.J. (1988) J. Gem Microbiol. 134,

2139-2147. [8] Adhi, T.P., Korus, R.A. and Crawford, D.L. (1989) Appl. Environ.

Microbiol. 55, 1165-1168. [9] Wang, P., Clark Mason, J. and Broda, P. (1993). J. Gem Microbiol.

139, 1987-1993. [10] Reese, E.T., Siu, R.G.H. and Levinson, H.S. (1950)J. Bacteriol. 59,

485-497. [l 1] Enger, M.D. and Sleeper, B.P. (1965) J. Bacteriol. 89, 23-27. [12] McCarthy, A.J. (1987) FEMS Microbiol. Rev. 46, 145-163. [13] McCarthy, A.J. and Williams, S.T. (1992) Gene 115, 189-192. [14] P~rez-Pons, J.A., Rebordosa, X. and Querol, E. (1994) Biotechnol.

Lett. 16, 959-964. [15] Chatterjee, S. and Vining, L.C. (1982) Can. J. Microbiol. 28,

593-599. [16] Kusama, S., Kusakabe, I. and Murakami, K. (1986) Agric. Biol.

Chem. 50, 2891-2898. [17] Ozaki, H. and Yamada, K. (1991) Agric. Biol. Chem. 55, 979-987. [18] Wakarchuck, W.W., Kilburn, D.G., Miller Jr., R.C. and Warren,

R.A.J. (1984) J. Gen. Microbiol. 30, 1385-1389. [19] Bernier, R. and Stutzenberger, F. (1988) Lett. Appl. Microbiol. 7,

103-107. [20] Gonzfi]ez-Candelas, L., Aristoy, M.C., Polaina, J. and Flors, A.

(1989) Appl. Environ. Microbiol. 55, 3173-3177. [21] Grabnitz, F., Seiss, M., Rucknagel, K.P. and Staudenbauer, W.L.

(1991) Eur. J. Biochem. 200, 301-309. [22] Shewale, J.G. (1982) Int. J. Biochem. 14, 435-443. [23] Shoemaker, S.P. and Brown, R.D., Jr. (1978) Biochim. Biophys.

Acta 523, 133-146. [24] Chirico, W.J. and Brown, R.D., Jr. (1987) Eur. J. Biochem. 165,

343-351. [25] Prrez-Pons, J.A., Cayetano, A., Rebordosa, X., Lloberas, J., Guasch,

A. and Querol, E. (1994) Eur. J. Biochem. 223, 557-565. [26] Ishaque, M. and Kluepfel, D. (1980)Can. J. Microbio[. 26, 183-189. [27] Laemmli, U.K. (1970) Nature 277, 680-685. [28] Robertson, E.F., Dannelly, H.K., Malloy, P.J. and Reeves, H.C.

(1987) Anal. Biochem. 167, 290-294. [29] Bradford, M.M. (1976) Anal. Biochem. 72, 248-254. [30] McHale, A. and Coughlan, M.P. (1982) J. Gen. Microbiol. 128,

2327-2331. [31] Ohmiya, K., Shirai, M., Kurachi, Y. and Shimizu, S. (1985) J.

Bacteriol. 161,432-434. [32] Patchett, M.L., Daniel, R.M. and Morgan, H.W. (1987) Biochem. J.

243, 779-787. [33] Mihoc, A. and Kluepfel, D. (1990) Can. J. Microbiol. 36, 53-56. [34] Katayeva, I.A., Golovchenko, N.P., Chuvilskaya, N.A. and Aki-

menko, V.K. (1992) Enzyme Microb. Technol. 14, 407-412. [35] Pitt-Rivers, R. and Impiombato, F.S.A. (1968) Biochem. J. 109,

825-830. [36] Heupel, C., Schlochtermeier, A. and Schrempf, H. (1993) Enzyme

Microb. Technol. 15, 127-132. [37] A'it, N., Creuzet, N. and Cattaneo, J. (1982) J. Gen. Microbiol. 128,

569-577. [38] Ham Y.W. and Srinivisan, V.R. (1969) J. Bacteriol. 100, 1355-1363. [39] Gond~, P., Ratomahenina, R., Arnaud, A. and Galzy, P. (1985) Can.

J. Biochem. Cell. Biol. 63, 1160-1166. [40] Sinnott, M.L. (1990) Chem. Rev. 90, 1171-1202.

J.A. P~rez-Pons et al. / Biochimica et Biophysica Acta 1251 (1995) 145-153 153

[41] Gabelsberger, J., Liebl., W. and Schleifer, K.H. (1993) Appl. Micro- biol. Biotechnol. 40, 44-52.

[42] Leatherbarrow, R.J. (1987) ENZ.PII'I~R: A non-linear regression data analysis program for the IBM-PC, Elsevier/Biosofl, Cambridge, UK.

[43] Nari, J., Noat, G. and Ricard, J. (1984) Eur. J. Biochem. 145, 319-322.

[44] Bodenmann, J., Heiniger, U. and Hohl, H.R. (1985) Can. J. Micro- biol. 31, 75-82.

[45] Spears, G., Sneyd, J.G.T. and Loten, E.G. (1971) Biochem. J. 125, 1149-1151.

[46] Dale, M.P., Ensley, H.E., Kern, K., Sastry, K.A.R. and Byers, L.D. (1985) Biochemistry 24, 3530-3539.

[47] Maguire, R.J. (1977) Can. J. Biochem. 55, 19-26. [48] Wright, R.M., Yablonsky, M.D., Shalita, Z.P., Goyal, A.K. and

Eveleigh, D.E. (1992) Appl. Environ. Microbiol. 58, 3455-3465. [49] Messner, R., Hagspiel, K. and Kubicek, C.P. (1990) Arch. Micro-

biol. 154, 150-155. [50] Kadam, S.K. and Demain, A.L. (1989) Biochem. Biophys. Res.

Commun. 161,706-711. [51] Painbeni, E., Valles, S., Polaina, J. and Flors, A. (1992) J. Bacteriol.

174, 3087-3091. [52] Watanabe, T., Sato, T., Yoshioka, S., Koshijima, T. and Kuwahara,

M. (1992) Eur. J. Biochem. 209, 651-659. [53] Lin, E. and Wilson, D.B. (1988) J. Bacteriol. 170, 3843-3846.

properties of a novel glucose-enhanced β-glucosidase purified from streptomyces sp. (atcc 11238)

Documents