Characterization of a Cellobiose Phosphorylase from a Hyperthermophilic Eubacterium, Thermotoga Maritima MSB8

Biosci. Biotechnol. Biochem., 66 (12), 2578–2586, 2002

Characterization of a Cellobiose Phosphorylase from a Hyperthermophilic Eubacterium, Thermotoga maritima MSB8

Eranna RAJASHEKHARA,MotomitsuKITAOKA,† Yeon-Kye KIM,andKiyoshiHAYASHI

Enzyme Laboratory, National Food Research Institute, 2-1-12 Kannondai, Tsukuba, Ibaraki 305-8642, Japan

Received May 27, 2002; Accepted July 25, 2002

The cepA putative gene encoding a cellobiose phos- ing b-1,4 glycosidic bonds, but has a relaxed speciˆci- phorylase of Thermotoga maritima MSB8 was cloned, ty with respect to the reducing sugars that function as expressed in Escherichia coli BL21-codonplus-RIL and glucosyl acceptors in the synthetic reaction. The ab- characterized in detail. The maximal enzyme activity solute recognition of the b-anomeric hydroxyl group was observed at pH 6.2 and 809C. The energy of activa- of the acceptor molecule was conˆrmed for the en- tion was 74 kJWmol. The enzyme was stable for 30 min zyme from Cellvibrio gilvus by using pseudo-glu- at 709C in the pH range of 6–8. The enzyme phos- coses.9) However, b-glucosides are not accepted as phorolyzed cellobiose in an random-ordered bi bi the acceptor by any reported cellobiose phosphory- mechanism with the random binding of cellobiose and lases. Syntheses of various disaccharides and bran- phosphate followed by the ordered release of D-glucose ched trisaccharides using the synthetic reaction of 10–14) and a-D-glucose-1-phosphate. The Km for cellobiose and cellobiose phosphorylase have been reported. phosphate were 0.29 and 0.15 mM respectively, and the Direct synthesis of cellobiose from sucrose by the －1 kcat was 5.4 s . In the synthetic reaction, D-glucose, D- combined action of three enzymes including cel- mannose, 2-deoxy-D-glucose, D-glucosamine, D-xylose, lobiose phosphorylase was also reported,15) and hence and 6-deoxy-D-glucose were found to act as glucosyl ac- indicated the possible commercial importance of the ceptors. Methyl-b-D-glucoside also acted as a substrate enzyme. The enzyme was used in the selective colori- for the enzyme and is reported here for the ˆrst time as a metric measurment of cellobiose in the presence of substrate for cellobiose phosphorylases. D-Xylose had glucose and cellooligosaccharides,16) a useful tool in －1 the highest (40 s ) kcat followed by 6-deoxy-D-glucose investigation of biomass degradation. (17 s－1) and 2-deoxy-D-glucose (16 s－1 ). The natural TheenzymefromCell. gilvus 17) and Cellulomonas －1 8) substrate, D-glucose with the kcat of 8.0 s had the uda catalyzed the phosphorolytic reaction in an or- 4 －1 －1 highest (1.1×10 M s ) kcat WKm compared with other dered bi bi mechanism as follows: the enzyme at ˆrst glucosyl acceptors. D-Glucose, a substrate of cellobiose bound cellobiose and then phosphate followed by the phosphorylase, acted as a competitive inhibitor of the sequential release of glucose and G-1-P. Recently, we other substrate, a-D-glucose-1-phosphate, at higher con- found that the enzyme from Clostridium thermocel- centrations. lum YM-4 also followed an ordered bi bi mechanism, but the order of the substrate binding was reverse.18) Key words: hyperthermostable enzyme; cellobiose The type strain of Thermotoga maritima,strain phosphorylase; Thermotoga maritima; MSB8, is a hyperthermophilic fermentative bacteri- phosphorolytic reaction; synthetic reac- um that grows optimally at 809C and its whole tion genome was sequenced for functional genomic studies.19) Both T. maritima and T. neapolitana are able Cellobiose, a common intermediate in enzymatic to use various carbohydrates for growth, and their cleavage of cellulose, can be degraded by hydrolysis cellulolytic enzymes are reported to be stable at ex- or inorganic-phosphate-dependent phosphorolysis. treme temperatures. Due to the extreme thermal Cellobiose phosphorylase (EC 2.4.1.20) catalyzes the resistance, there is an increasing interest for applica- reversible phosphorolysis of cellobiose into a-D-glu- tion of hyperthermostable enzymes in biotechnologi- cose 1-phosphate (G-1-P) and D-glucose with the in- cal processes.20–25) version of the anomeric conˆguration, but does not Recently, a cellobiose phosphorylase from a hyper- phosphorolyze cellotriose or higher cello-oligosac- thermophilic bacterium, T. neapolitana, was charac- charides. The presence of this enzyme was ˆrst terized,26) but not in detail. Because of the commer- demonstrated in Clostridium thermocellum1) and cial importance of the enzyme and for fundamental subsequently studied in several organisms.2–8) The en- understanding, a detailed functional analyses, phos- zyme is absolutely speciˆc for cleaving and synthesiz- phorolytic- and synthetic reaction of the recombinant

† To whom correspondence should be addressed. Fax: +81-298-38-7321; E-mail: mkitaoka＠nfri.aŠrc.go.jp Cellobiose Phosphorylase from Thermotoga maritima 2579 cellobiose phosphorylase from another hyperther- shaker in Luria-Bertani (LB) broth (100 ml) with mophilc bacterium, T. maritima, was studied and 30 mgWml of kanamycin to produce seed culture. This reported in this paper. 100-ml seed culture was used as the inoculum for 900 ml of LB broth supplemented with 30 mgWml of Materials and Methods kanamycin. When the OD600 of the broth reached approximately 0.6, the target gene was induced by the Genomic DNA, plasmids and bacterial strains. addition of isopropyl thio-galactopyranoside (IPTG) The genomic DNA of the type strain T. maritima to a ˆnal concentration of 1 mM. Cultivation of the MSB8 was kindly provided by Prof. Dr. K. O. Stetter cells was continued for another 5 h at 309C. Then the (Lehrstuhl Fuer Mikrobiologie, University of Regen- cells were harvested by centrifugation at 4,000 g for sburg, Germany). A plasmid pCR 2.1-TOPO vector 20 min at 49C. (TOPO TA cloning kit, Invitrogen, USA) was used to insert the PCR-ampliêd product of the cepA Puriˆcation of recombinant enzyme. Approxi- gene described later. TOP 10 electrocompetant mately 2.7 g (wet weight) of cells were suspended in Escherichia coli cells F„ merAD (mrr-hsdRMS- 20 mM sodium phosphate buŠer, pH 8.0, containing mcrBC)q80lacZDM15DlacX74deoRrecA1araD139D 300 mM NaCl and 10 mM imidazole (buŠer A) and (ara-leu)7697galUgalKrpsL(strR)-endA1nupG ]were the cells were lysed by soniˆcation. The soniêd sam- used as the host for the transformed by pCR2.1 plas- ple was centrifuged at 10,000 g for 30 min to remove mids harboring the cellobiose phosphorylase gene the cell debris. The supernatant containing crude en- (cepA)ofT. maritima. The plasmid vector pET zyme (71 ml) was mixed with 3 mL of fresh NiNTA 28a(+) from Novogen, USA, was used for the sub- resin (Qiagen, Germany) and kept on ice under gentle cloning of the gene orienting the (His)6 epitope tag at shaking for 15–20 min. The enzyme-bound resin was the gene's N-terminal end. The E. coli strain then packed into a column and eluted the enzyme BL21-Codonplus-RIL competent cells [BF„ompT with a linear gradient of 10–200 mM of imidazole in hsdS(rB„mB„)dcm+Tetr gal endA Hte [argU ileY buŠer A. The active fractions were pooled and dia- leuW Cam r ] were used for the heterologous expres- lyzed overnight at 49Cagainst5mM 2-[N-mor- sion of the cepA gene. All the recombinant DNA pholino]ethane sulfonic acid (MES) buŠer, pH 6.2. techniques followed were as described by Sambrook The dialyzed enzyme sample was further puriêd by et al.27) an anion exchange column, Source 30 Q (Pharmacia, Sweden), of column volume 4 ml. The enzyme pro- Cloning, Sequencing of cepA, and Expression. tein was eluted with a linear gradient of 0–100 mM of The ORF TM1848 (AE001822) which codes for the NaCl in 20 mM MES buŠer, pH 6.2. putative cellobiose phosphorylase of T. maritima MSB8 was ampliêd from the genomic DNA by a 25 Protein analyses. Protein concentrations were cycle PCR (989C, 30 s—559C, 30 s—729C, 150 s) us- measured using Coomassie Plus Protein assay rea- ing KOD-DASH DNA polymerase (Toyobo, Osaka, gent kit (Pierce, Rockford, USA) with bovine serum Japan), a forward primer, 5?CAT ATG GTG CGA albumin as a standard. Sodium dodecyl sulfteWpoly- TTC GGT TAT TTT GAT GAC GTC, and a reverse acrylamide gel electrophoresis (SDS-PAGE) of the primer, 5?GAG CTC TCA GCC CAT CAC AAC enzyme samples was done and the protein bands were TTC AAC CCT GTG with the incorporation of NdeI made visible by staining the gel with Coomassie brilli- (N-terminal end) and SacI (C-terminal end) restric- ant blue.28) A 10-kDa ladder (Life technologies, USA) tion sites (underlined) respectively. The amplicon was was used as a standard molecular mass protein mar- cloned in pCR 2.1-TOPO and transferred to elec- ker. The puriêd enzyme (approximately 0.2 mg) trocompetent E. coli TOP 10 cells. The nucleotide se- from the Mono Q fraction was put on a gel ˆltration quence of the cloned gene was conˆrmed using a Big column of Superose 6 HR10W30 previously equili- dye terminator cycle sequencing kit (Perkin-Elmer, brated with 50 mM MES buŠer, pH 6.2, containing Applied Biosystems, CA) and GENETIX program 0.1 M NaCl, at a ‰ow rate of 0.1 ml per min to meas- (Software Development Co., Tokyo, Japan). The ure its native molecular mass with MW-Marker target gene, cepA from the pCR 2.1-TOPO-TmcepA HPLC (Oriental Yeast Co., Osaka, Japan). was digested with NdeI and SacI restriction enzymes and ligated to pET 28a(+) that had been previously Enzyme assays. The reaction of cellobiose digested with the above restriction enzymes. phosphorylase was done at 609Cin25mM MES Ligation-High T4 DNA ligase (Toyobo) was used for buŠer, pH 6.2. In the phosphorolytic reaction the gene-vector ligation during subcloning. The pET amount of a-D-glucose 1-phosphate (G-1-P) or D-glu-

28a(+) harboring cepA with (His)6 residues at its N- cose formed was measured. G-1-P was estimated by terminal end was transferred to electrocompetent E. the phosphoglucomutase-glucose-6-phosphate de- coli BL21-Codonplus-RIL cells. The transformed hydrogenase system.29) D-Glucose was measured by cells were cultured overnight at 379C over a rotary the glucose oxidase-peroxidase method using a Glu 2580 E. RAJASHEKHARA et al.

Table 1. Summary of Puriˆcation of T. maritima Cellobiose Phosphorylase

Total Total Speciˆc Activity Puriˆcation Puriˆcation step activity protein activity fold recovery (units) (mg) (UWmg) (z) Crude extract 92.3 208.7 0.44 1 100 NiNTA 54.4 22.6 2.44 5.4 59 Source 30 Q 19.7 7.4 2.66 6 21

ICII kit (Wako Pure Chemicals, Osaka, Japan). In the synthetic reaction the product, phosphate was estimated by the method of Lowry and Lopez.30) One unit of activity was deˆned as the amount of enzyme that produced one mmol of phosphate per minute Fig. 1. SDS-PAGE Analysis of Recombinant T. maritima Cel- from 10 mM G-1-P and 10 mM glucose at 609C. lobiose Phopshorylase at DiŠerent Puriˆcation Steps. lane M, molecular mass markers; lane 1, crude extract; lane 2, Optimum pH and temperature. To ˆnd the opti- NiNTA fraction, lane 3, ``Source 30 Q'' fraction. mum pH, the phosphorolytic reaction was done at 609Cfor30minwith10mM cellobiose and 10 mM phosphate in 50 mM of various buŠers prepared at tivity at 609C was estimated by measuring the D-glu- 259C such as sodium citrate (pH 2.2 to 4.1), sodium cose formed. The inhibition eŠects of the products, acetate(pH3.7to5.7),MES(pH5.1to7.2), D-glucose (0–1 mM)andG-1-P(0–2mM), against the 3-[N-morpholino]propane sulfonic acid (MOPS) substrates, cellobiose and phosphate, were also stud- (pH 6.4 to 8.8), N-[2-hydroxyethyl]piperzine-N?- ied at several ˆxed concentrations of one substrate [2-ethanesulfonic acid] (HEPES) (pH 6.5 to 8.6), (2.5,5,20mM) and varying the concentration of Tris[hydroxymethyl]aminomethane hydrochloride) other substrate (0.25–2.5 mM)bymeasuringG-1-P (TrisWHCl) (pH 7.1 to 9.1), 2-[N-cyclohex- and D-glucose, respectively, to ˆnd the orders of both yamino]ethane sulfonic acid (CHES) (pH 8.2 to substrates binding and products release. 10.3), and 3-[cyclohexylamino]-1-propane sulfonic acid (CAPS) (pH 9.6 to 11.5), and the amount of D- Synthetic reaction. Various sugars and their glucose formed was measured as mentioned earlier. derivatives mentioned in Table 2 were examined for To ˆnd the optimum temperature, the phosphoro- the glucosyl acceptor of the cellobiose phosphorylase lytic reaction was done at temperatures ranging from at 10 mM concentration with 10 mM G-1-P at 609Cin 25 to 959Cfor30minin25mM MES buŠer with 25 mM MES (pH 6.2) buŠer. The kinetic parameters 10 mM cellobiose and 10 mM phosphate. The amount were measured for those glucosyl acceptors that of D-glucose formed after the reaction was measured showed a rate higher than 1 mmoleWminWmg. and used for evaluating the average activity for 30 min. Kinetic analyses. Kinetic parameters were calculated by regressing the experimental data to the ap- pH and thermal stability. The puriêd enzyme propriate rate formula based on the curve ˆt method (13.3 mgWml) was incubated for 30 min at 709Cin by using a computer program, Graˆt-Ver.4 (Erithaus various buŠers as mentioned above and the residual Software, UK). enzyme activity was measured at 609Cfor30minin MES buŠer, pH 6.2 with 10 mM of cellobiose and Results 10 mM phosphate. To measure the thermal stability, theenzyme(13.3mgWml) was incubated for 30 min in Puriˆcation and properties of the enzyme 25 mM MES buŠer (pH 6.2) at diŠerent temperatures The recombinant cellobiose phosphorylase ranging from 30 to 1009C and the residual enzyme was puriêd 6-fold to a ˆnal speciˆc activity of activity was measured at 609Cfor30mininMES 2.66 unitsWmgproteinwithanactivityrecoveryof buŠer, pH 6.2 containing 10 mM of cellobiose and 21z through two puriˆcation steps (Table 1). Ap- 10 mM phosphate. proximately 7.4 mg of pure enzyme was obtained from 1000 ml of culture. The puriêd cellobiose Reaction mechanism. In the phosphorolytic direc- phosphorylase gave a single protein band on SDS- tion, the reaction mechanism was measured using PAGE gel corresponding to a molecular mass (Mr) diŠerent combinations of cellobiose and phosphate of approximately 85 kDa (Fig. 1), which agreed with concentrations from 0.25 to 2.5 mM. The enzyme ac- the size calculated (94,322) on the basis of the prima- Cellobiose Phosphorylase from Thermotoga maritima 2581

Fig. 2. EŠects of pH on the Activity of the T. maritima Cel- Fig. 4. EŠects of pH on the Stability of the T. maritima Cel- lobiose Phosphorylase. lobiose Phosphorylase. The activity was measured at 609Cfor30minin25mM The stability of the enzyme was examined by treating the en- buŠers of various pH levels. Values are shown as a percentage of zyme at 709C for 30 min in buŠers at various pH levels. The the maximum activity, which is deˆned as 100z.BuŠersused residual enzyme activity was measured at 609Cin25mM MES were sodium citrate (), sodium acetate (), MES (), MOPS buŠer of pH 6.2. Values are shown as percentages of the original (), HEPES (#), TrisWHCl ($), CHES (%),andCAPS(&). activity (measured in 25 mM MES, pH 6.2, without incubation at 709C) and this is deˆned as 100z. BuŠers used were sodium citrate (), sodium acetate (), MES (), MOPS (), HEPES (#), TrisWHCl ($), CHES (%),andCAPS(&).

The optimum pH for the maximal enzyme activity was found at 6.2 and, the enzyme showed more than 80z of the maximal activity between pH 4.5 to 7.5 (Fig. 2). The optimum temperature for the maximal enzyme reaction for 30 min was approximately 809C (Fig. 3). From the linear part of the Arrhenius plot between 25 and 809C, the activation energy for the phosphorolytic reaction was found to be 74 kJWmol. The enzyme retained more than 90z of its maximal activity in the pH range of 5 to 9 after 30 min of heating at 709C (Fig. 4). The enzyme retained its original activity at temperatures lower than 709C, but approximately 25z of the activity was lost upon incubating at 909C and the activity was completely lost at 1009C after a 30-min incubation period Fig. 3. EŠects of Temperature on the Activity of the T. maritima (Fig. 5). It is notable that the reaction rate at 959C Cellobiose Phosphorylase. for 30 min (35z of the maximum) was much lower The activity was measured at various temperatures for 30 min than the value expected from the stability data (71z in 25 mM MES buŠer, pH 6.2. Values are shown as a percentage remaining after 30 min). The fact may suggest some of the maximum activity, which is deˆned as 100z. reversible change in the enzyme structure to an inac- tive form at higher temperatures. The course of the thermal inactivation of the en- ry structure of the enzyme.19) Thenativemolecular zyme at 60 and 809C was studied in the absence of mass was estimated to be 150 kDa by gel ˆltration, any substrate or stabilizer. At 609C the enzyme was indicating that the recombinant cellobiose phos- found to be stable for a period of 5 h, while at 809C phorylase existed in a dimer form. The enzyme phos- there was a 45z reduction in the enzyme activity af- phorolyzed cellobiose, but did not phosphorolyze ter 60 min with the enzyme displaying a half-life of cellotriose or cellotetrose to form G-1-P even with a 70 min. long incubation time. 2582 E. RAJASHEKHARA et al.

Fig. 5. EŠect of Temperature on the Stability of the T. maritima Fig. 7. Detailed Reaction Mechanisms in Sequential bi bi Cellobiose Phosphorylase. Mechanisms. The enzyme was incubated in 25 mM ˆnal concentration of (1) The typical sequential bi bi mechanisms appeared in MES, pH 6.2 buŠer at diŠerent temperatures for 30 min. Resid- Cleland's report.31) (2) Inhibition pattern and postulated ual enzyme activities were measured at 609C for 30 min. The mechanism of T. maritima cellobiose. phosphorylase. residual activity are shown as percentages of the maximum ac- Left column, scheme of each mechanism; center column, name tivity (measured at 609C for 30 min without a prior incubation of each mechanism; right column, table of inhibition pattern of step) taken as 100z. each product against each substrate. A and B are substrates and P and Q are products. In the table, c, m, and u mean competitive, mixed-type, and uncompetitive inhibitions, respectively. Inhibition patterns out of and in the parenthesis mean the inhibition pattern observed with unsaturated and saturated concentrations of the other substrate, respectively. The concentrations of the other substrate used in the table of (2) are 2.5, 5 mM (unsaturated), and 20 mM (saturated).

quential bi bi mechanism were calculated by regressing the experimental data with the following equation as shown below:

[＝kcat[E][A][B]W(KiA KmB+KmA [B]+KmB [A] +[A][B]) where [A] is the initial concentration of cellobiose, Fig. 6. Double Reciprocal Plot of the Phosphorolytic Reaction [B] is the initial concentration of phosphate, and [E] of Cellobiose. is the enzyme concentration. The enzyme activity was measured as a function of the con- „1 centration of cellobiose at various concentrations of phosphate kcat＝5.4 s ; KmA＝0.29 mM; KiA＝1.7 mM; at 609C for 30 min. Concentrations of phosphate were , KmB＝0.15 mM. 0.25 mM; ,0.5mM; ,1.0mM; ,1.5mM; #,2.5mM. The inhibition patterns of the products, G-1-P and D- glucose, against the substrates, D-cellobiose and Reaction mechanism phosphate, are summarized in the table of Fig. 7(2). Kinetic analyses of the phosphorolytic reaction G-1-P acted as a competitive inhibitor and glucose as were undertaken to discover the reaction mechanism a mixed-type inhibitor against both cellobiose and of the enzyme. When the concentration of cellobiose phosphate at all of the three concentrations of the opposite substrates examined. This result did not was varied, the 1Wv„1W[cellobiose] plot obtained with selected phosphate concentrations showed the match with any of three possible mechanisms in the 31) lines crossing at a point in the upper left-hand quad- sequential bi bi mechanism as shown in Fig. 7(1). A rant (Fig. 6), indicating that the phosphorolytic reac- possible explanation of the result is the ``random- tion occurred in a sequential bi bi mechanism. The ordered bi bi mechanism'' as described in Fig. 7(2), kinetic parameters in the rate equation for the se- where the sequence of the substrate binding is ran- Cellobiose Phosphorylase from Thermotoga maritima 2583

Table 2. Substrate Acceptor Speciˆcity of T. maritima Cellobiose Phosphorylase in the Synthetic Reaction

Substrate Rate ( mmolWminWmg) Substrate Rate ( mmolWminWmg)

D-Glucose 8.7 D-Glucono-d-lactone — D-Mannose 1.1 Methyl-b-D-glucoside 1.1 2-Deoxy-D-glucose 2.9 Methyl-a-D-glucoside — D-Glucosamine 2.9 D-Cellobiose — D-Mannosamine — D-Glucitol — N-Acetyl-D-glucosamine — D-Mannitol — N-Acetyl-D-mannosamine — D-Arabinose — D-Allose — L-Arabinose — D-Galactose — D-Ribose — D-Fucose — L-Sorbose — 6-Deoxy-D-glucose 13.8 L-Glucose — D-Xylose 15.0 L-Fucose — D-Glucuronamide — L-Xylose — D-Fructose —

Values were measured using the concentrations of 10 mM G-1-P and 10 mM acceptor substrate at 609C. —, not detected (less than 0.6).

Table 3. Kinetic Parameters for Various Glucosyl Acceptors of T. maritima Cellobiose Phosphorylase

Km kcat kcat WKm Substrate „1 „1 „1 (mM) (s ) (M s )

a 4 D-Glucose 0.69 8.0 1.1×10 b 1 D-Mannose 67 4.4 6.5×10 c 2 D-Glucosamine 5.7 5.2 9.1×10 b 2 2-Deoxy-D-glucose 47 16 3.4×10 c 3 6-Deoxy-D-glucose 4.1 17 4.3×10 3 D-Xylosed 14 40 2.8×10 e 1 Methyl-b-D-glucoside 135 6.5 4.8×10

Concentrations used: a, 0.5–5 mM; b, 10–100 mM; c, 1–10 mM; d, 5–50 mM; e, 20–200 mM. dom and the product release is ordered (glucose ˆrst and then G-1-P).

Substrate speciˆcity for the synthetic reaction

Out of 27 compounds tested, seven were found to Fig. 8. The v-[D-glucose] Plot of the Synthetic Reaction of the act as glucosyl acceptors for the cellobiose phos- Cellobiose Phosphorylase at Fixed Concentrations of G-1-P. phorylase (Table 2). It is notable that the enzyme Concentrations of G-1-P were ,1mM; ,2mM; ,5mM. used b-methyl-D-glucoside as the acceptor. The The theoretical lines were obtained by regressing the experimental data to the formula of competitive substrate inhibition.12) product from G-1-P and b-methyl-D-glucoside was identiˆed as b-methyl-cellobioside by NMR analysis (data not shown). Detailed kinetic studies were done which one substrate, namely D-glucose, competitively by using these substrates, and their kinetic inhibited the other substrate, G-1-P.12) Therefore, the parameters are presented in Table 3. D-Xylose had possibility of inhibition by D-glucose was studied at „1 the highest kcat (40 s ) followed by 6-deoxy-D-glu- concentrations of D-glucose ranging from 1 mM to cose (17 s„1) and 2-deoxy-D-glucose (16 s„1). The 100 mM with several ˆxed concentrations of G-1-P. „1 natural substrate, D-glucose gave the kcat of 8.0 s , As shown in Fig. 8, the [D-glucose]-v curves gave 4 „1 „1 but it had the highest kcat WKm (1.1×10 M s )com- maxima around a D-glucose concentration of 10 mM, pared with other glucosyl acceptors. demonstrating that a similar inhibition also occurred in the synthetic reaction by the cellobiose phosphory- Inhibition pattern of D-glucose lase from T. maritima. The theoretical lines obtained A substrate inhibition by D-glucose in the reverse by regressing the experimental data with the formula synthetic reaction was observed previously for the explaining the competitive substrate inhibition,12) as cellobiose phopshorylases from Cellv. gilvus 12) and shown below, well explain the result (Fig. 8). Clost. thermocellum.18,32) These inhibitions were v＝k [E][Q][P] sK K +(K +K K K )[P] explained as ``competitive substrate inhibition'', in cat W iQ mP mQ iQ mP W I1 2584 E. RAJASHEKHARA et al. the glucosyl acceptors for the enzyme. The results in- +KmP [Q]+[Q][P] dicated that the hydroxyl groups at positions 3 and 4 2 +(KiQKmP WKI1 KI2+KmQ WKI1)[P] were essential for the recognition of acceptors. The 3 +(KmQ WKI1KI2 )[P] t kinetic parameters (Table 3) showed that the natural substrate of the enzyme, D-glucose, was found to where [Q] is the initial concentration of G-1-P, [P] is have the lowest Km (0.69 mM)followedby6-deoxy-D- the initial concentration of D-glucose,and[E]isthe glucose (4.1 mM). The substitution or deletion of enzyme concentration. hydroxyl group at C1, C2 or C6 conˆgurations of D- „1 kcat＝13 s ; KmQ＝0.29 mM; KiQ＝1.7 mM; glucose resulted in signiˆcant increases in Km,indicating that these hydroxyl groups were also recog- KmP＝0.15 mM; KI1＝7.3 mM; KI2＝190 mM. nized by the enzyme. D-Xylose acted as a good „1 Discussion substrate with the maximum kcat value of 40 s as compared to D-glucose (8.0 s„1). The cellobiose The genome sequence results indicated the phosphorylases from diŠerent organisms that we presence of a putative cellobiose phosphorylase in T. studied strictly recognized the C1 conˆguration of b- 8,9,12) maritima.19) According to the primary structure, cel- D-glucopyranose and it was believed that the en- lobiose phosphorylase, cellodextrin phosphorylase, zymes did not recognize b-glucosides as the acceptor. and chitibiose phosphorylase are grouped in the However, the cellobiose phosphorylase of T. mariti- 33) ma wasfoundtoacceptmethyl-b-D-glucoside as a family 36 of glycosyl transferases. (http:WWafmb- ¿ substrate in the synthetic reaction to produce methyl- mrs.frWpedroWCAZYWgtf.html) The percent amino acid similarities among cellobiose phosphorylases (-D-cellobioside. This unique speciˆcity has not been found increasing from mesophilic (61z in Cellv. reported for any other cellobiose phosphorylases. gilvus, AB010707) to thermophilic (73z, Clost. The substitution of a hydroxyl group at C1 of D-glu- thermocellum, AB013109, and Clost. stercorarium cose with a methoxy group resulted in a 196-fold in- U56424), and hyperthermophilic (92z, T. crease in the Km values compared with that of D-glu- neapolitana AF039487 and Z99777) organisms. With cose (0.69 mM) without signiˆcant diŠerence in the the cellodextrin phosphorylase of Clost. stercorarium kcat values obtained. (U60580) the amino acid sequence similarity was Competitive inhibition by the natural substrate, found to be approximately 40z. D-glucose (Fig. 8), was again observed for the The phosphorolytic reaction by the cellobiose cellobiose phosphorylase from T. maritima.The phosphorylase of T. maritima was found to proceed phenomenon is supposed to be common in cellobiose in a sequential mechanism. However, the product in- phosphorylases. hibition pattern did not match with the possible three In summary, the characterization of the cellobiose sequential mechanisms presented by Cleland.31) Judg- phosphorylase from T. maritima revealed a high ing from the inhibition pattern, a possible explana- degree of thermal stability and a distinct substrate tion of the reaction is a random-ordered bi bi speciˆcity for reactions in the synthetic direction. mechanism. In this mechanism, the binding sequence The mechanism of the reaction was slightly diŠerent of cellobiose and phosphate is random, and the from other cellobiose phosphorylases reported. In release sequence of the glucose and G-1-P is ordered. the future, this hyperthermophilic cellobiose phos- The reaction mechanisms of Cellv. gilvus,17) Cellu. phorylase may ˆnd application in the syntheses of Uda,8) and Clost. thermocellum18) cellobiose phos- various oligosaccharides. phorylases were reported to be ordered bi bi mechanisms. The ˆrst two enzymes bind cellobiose ˆrst and Acknowledgments then phosphate followed by the sequential release of glucose and G-1-P. However, the Clost. thermocel- WethankProf.K.O.StetteroftheUniv.ofRegen- lum YM-4 enzyme binds phosphate before cellobiose sberg, Germany for providing the genomic DNA of and releases glucose and G-1-P with the same se- T. maritima. We also thank Dr. P. Selvakumar and quence of the other two enzymes.18) The random- Dr. Kshamata Goyal for their suggestions during this ordered mechanism of the T. maritima cellobiose work. The STA fellowship of Japan awarded to the phosphorylase seems intermediate between the above senior author is gratefully acknowledged. This work two types of ordered bi bi mechanisms because the was supported in part by a grant from the Program diŠerence in the mechanism is only in the sequence of for Promotion of Basic Research Activities for In- the substrate binding. novative Biosciences in Japan. In the synthetic reaction, out of 27 sugars or their derivatives tested, only D-glucose, D-xylose, D-man- References nose, D-glucosamine, 2-deoxy-D-glucose, 6-deoxy-D- 1) Sih, C. J., and McBee, R. H., A cellobiose phos- glucose, and methyl-b-D-glucopyranoside served as phorylase in Clostridium thermocellum. Proc. Mon- Cellobiose Phosphorylase from Thermotoga maritima 2585 tana Acad. Sci., 15, 21–26 (1955). 17) Kitaoka, M., Sasaki, T., and Taniguchi, H., Phos- 2) Ayres, W., Phosphorylation of cellobiose and glucose phorolytic reaction of Cellvibrio gilvus cellobiose by Ruminococcus ‰avifaciens. J. Bacteriol., 76, phosphorylase. Biosci. Biotechnol. Biochem., 56, 515–517 (1958). 652–655 (1992). 3) Hulcher, F. H., and King, K. W., Metabolic basis for 18) Kim, Y.-K., Kitaoka, M., Krishnareddy, M., Mori, disaccharide preference in Cellvibrio. J. Bacteriol., Y., and Hayashi, K., Kinetic studies of a recombinant 76, 571–577 (1958). cellobiose phosphorylase (CBP) of Clostridium ther- 4) Hutterman, A., and Volger, C., Cellobiose phos- mocellum YM4 strain expressed in Escherichia coli. J. phorylase in Fomes annosus. Nature New Biol., 245, Biochem. (Tokyo), 132, 197–203 (2002). 64 (1973). 19) Nelson,K.E.,Clayton,R.A.,Gill,S.R.,Gwinn,M. 5) Matheron, C. A., Delrot, M., Gaudet, G., and L.,Dodson,R.J.,Haft,D.H.,Hickey,E.K., Forano, E., In vivo 13CMR study of glucose and Petrson,J.D.,Nelson,W.C.,Ketchum,K.A., cellobiose metabolism by four cellulolytic strains of McDonald,L.,Utterback,T.R.,Malek,J.A., the genus, Fibrobacter. Biodegradation, 9, 451–461 Linher,K.D.,Garrett,M.,Stewart,A.M.,Cotton, (1998). M. D., Pratt, M. S., Phillips, C. A., Richardson, D., 6) Schimz, K.-L., Broll, B., and John, B., Cellobiose Heiderberg,J.,Sutton,G.G.,Fleischmann,G.G., phosphorylase of Cellulomonas: Occurrence, induc- Eisen,J.A.,White,Salzberg,S.L.,Smith,H.O., tion and its role in cellobiose metabolism. Arch. Venter, J. C., and Fraser, C. M., Evidence for lateral Microbiol., 135, 241–249 (1983). gene transfer between archaea and bacteria from ge- 7) Sasaki, T., Tanaka, T., Nakagawa, S., and Kainuma, nome sequence of Thermotoga maritima. Nature, K., Puriˆcation and properties of Cellvibrio gilvus 399, 323–329 (1999). cellobiose phosphorylase. Biochem. J., 209, 803–807 20) Bronnenmeir, K., Kern, A., Liebl, W., and (1983). Staudenbauer, W. T., Puriˆcation of Thermotoga 8) Nidetzky, B., Eis, C., and Albert, M., Role of non- maritima enzymes for degradation of cellulosic covalent enzyme-substrate interactions in the reaction materials. Appl. Environ. Microbiol., 61, 1399–1407 catalysed by cellobiose phosphorylase from Cellumo- (1995). nas uda. Biochem. J., 351, 649–659 (2000). 21) Evans, R. B., Gliman, A. K., Cordray, K., and 9) Kitaoka, M., Ogawa, S., and Taniguchi, H., A cel- Woodward, J., Mechanism of substrate hydrolysis by lobiose phosphorylase from Cellvibrio gilvus recog- a endoglucanse from Thermotoga maritiam. nizes only the b-D-form of 5a-carba-glucopyranose. Biotechol. Lett., 22, 735–740 (2000). Carbohydr. Res., 247, 355–359 (1993). 22) Goyal, K., Selvakumar, P., and Hayashi, K., Charac- 10) Alexander, J. K., Synthesis of 4-O-b-D- terization of a thermostable b-glucosidase (BglB) glucopyranosyl-D-xylose, 4-O-b-D-glucopyranosyl-D- from Thermotoga maritima showong transglycosyla- arabinose, 4-O-b-glucopyranosyl-2-deoxy-D-glucose, tion activity. J. Mol. Catalysis. B. Enzymatic., 15, 4-O-b-glucopyranosyl-D-mannose and 4-O-b-D- 45–53 (2001). glucopyranosyl-D-mannoseamine by cellobiose phos- 23) Huber, H., and Stetter, K. O., Hyperthermophiles phorylase from Clostridium thermocellum. Arch. and their possible potential in biotechnology. J. Biochem. Biophys., 123, 240–246 (1968). Biotechnol., 64, 39–52 (1998). 11) Kitaoka, M., Taniguchi, H., and Sasaki, T., Produc- 24) Liebl, W., Cellulolytic enzymes from Thermotoga tion of glucosyl-xylose using Cellvibrio gilvus cells spp. In ``Methods in Enzymology'' vol. 330, eds. and its properties. Appl. Microbiol. Biotechnol., 34, Adams, M. W. W., and Kelly, R. M., Academic 178–182 (1990). Press, San Diego, pp. 290–300 (2001). 12) Kitaoka, M., Sasaki, T., and Taniguchi, H., Synthet- 25) Liebl, W., Brem, D., and Gotschlich, A., Analysis of ic reaction of Cellvibrio gilvus cellobiose phosphory- the gene for b-fructosidase (invertase, inulinase) of lase. J. Biochem. (Tokyo), 112, 40–44 (1992). the hyperthermophilic bacterium Thermotoga mariti- 13) Percy, A., Ono, H., and Hayashi, K., Acceptor ma and characterization of the enzyme expressed in speciˆcity of Cellobiose phosphorylase from Escherichia coli. Appl. Microbiol. Biotechnol., 50, Cellvibrio gilvus: synthesis of three branched trisac- 55–64 (1998). charides. Carbo. Res., 308, 423–429 (1998). 26) Yernool, D. A., McCarthy, J. K., Eveleigh, D. E., 14) Tariq, M. A., and Hayashi, K., Synthesis of three and Bok, J.-D., Cloning and characterization of the heterodisaccharides, 4-O-b-D-glucopyranosyl-6-deox- glucooligosaccharide catabolic pathway b-glucan y-D-glucose, 4-O-b-glucopyranosyl-D-mannoseamine, glucohydrolase and cellobiose phosphorylase in the 4-O-b-D-glucopyranosyl-D-mannose and conˆrma- marine hyperthermophilic Thermotoga maritima. J. tion of their structures by C13 NMR and MS. Bacteriol., 182, 5172–5179 (2000). Biochem. Biophy. Res. Comm., 214, 568–575 (1995). 27) Sambrook, J., Fritsch, E. F., and Maniatis, T., In 15) Kitaoka, M., Sasaki, T., and Taniguchi, H., Conver- ``Molecular cloning: A laboratory manual, 2nd ed.'', sion of sucrose into cellobiose using sucrose Cold Spring Harbor Laboratory, New York (1989). phosphorylase, xylose isomerase and cellobiose phos- 28) Laemmli, U. K., Cleavage of structural proteins phorylase. Denpun Kagaku, 39, 281–283 (1992). during the assembly of the head of bacteriophage T4. 16) Kitaoka, M., Aoyagi, C., and Hayashi, K., Colori- Nature, 227, 680–685 (1970). metric quantiˆcation of cellobiose employing cel- 29) Michael, G., In ``Methods in Enzymatic Analysis, lobiose phosphorylase. Anal. Biochem., 292, 163–166 2nd ed.'' vol. 4, ed. Bergmeyer, H. U., Academic (2001). Press, San Diego, pp. 185–191 (1974). 2586 E. RAJASHEKHARA et al. 30) Lowry, O. H., and Lopez, J. A., J. Biol. Chem., 162, phosphorylase from Clostridium thermocellum. J. 421–428 (1946). Ferment. Bioeng., 79, 212–216 (1995). 31) Cleland, W. W., The kinetics of enzyme-catalyzed 33)Campbell,J.A.,Davies,G.J.,Bulone,V., reactions with two or more substrates or products. and Henrissat, B., A classiˆcation of nucleotide- Biochim. Biophys. Acta, 67, 104–137 (1963). diphospho-sugar glycosyltransferases based on 32) Tanaka, K., Kawaguchi, T., Imada, Y., Ooi, T., and aminoacid sequence similarities. Biochem. J., 326, Arai, M., Puriˆcation and properties of cellobiose 929–939 (1997).