Carbohydrate Research, 41 (1975) 273-283 © Elsevier Scientific Publishing Company, Amsterdam - Printed in Belgium

STUDIES ON THE STRUCTURE AND MECHANISM OF AN EXO-(1-+3)-P-D-GLUCANASE FROM BASIDIOMYCETE QM806 DENNIS

R.

PETERSON AND SAMUEL KIRKWOOD

Department of Biochemistry, College of Biological Sciences, University of Minnesota, St. Paul, Minnesota 55108 (U. S. A.) (Received July 10th, 1974; accepted in revised form, November 4th, 1974)

ABSTRACT

A method for the large-scale production of a (1-+3)-P-D-glucan glucohydrolase (EC 3.2.1.58) from the culture filtrate of Basidiomycete QM806 is described. The final preparation is homogeneous by disc electrophoresis under non-dissociating and denaturing conditions, by ultracentrifugation, and by isoelectric focusing. Various physical and chemical characteristics of the enzyme have been determined, including terminal amino acid residues, extinction coefficient, and stability to pH extremes. The N-terminal amino acids are leucine and serine (Sanger's method) and the C-terminal amino acids are alanine, serine, and glycine (hydrazinolysis). pH profile studies show that no group titrating in the region 2.5-8 is directly involved with substrate binding and that a single group having a pKa of 6.5 is involved in the catalysis. Photooxidation of the enzyme caused rapid inactivation. The pH-dependence of this photooxidation, and amino acid analysis of the photooxidized enzyme, indicate that decomposition of histidine is probably responsible for the loss of activity. Other chemical modifications performed were: treatment with hydrogen peroxide under acidic conditions, esterification with diphenyldiazomethane, and oxidation with N-bromosuccinimide. Oxidation with N-bromosuccinimide indicated that a tryptophan side-chain is involved in, but not necessary for, the catalytic activity. INTRODUCTION

The exo-(1-+3)-P-D-glucan glucohydrolase used in this study was originally discovered by Reese and Mandels in the culture filtrate of a Basidiomycete 1. The mode of attack, specificity, and some physical and chemical characteristics of the enzyme have been reported by a number of workers, including Reese and Mandels!, Chesters and Bu1l 2 , and Nelson et al. 3 ,4. A purification scheme that produces homogenous enzyme has been reported, together with a determination of molecular weight, amino acid composition, and other physical and chemical properties 5. The present report describes a method for production of the pure enzyme in large quantity. The larger amounts of enzyme available through this procedure have allowed the initiation of studies into the nature of the active site.

274

D.R.PETERSON,S.KIRKVVOOD

RESULTS AND DISCUSSION

The scheme for enzyme production and purification presented in the Experimental section, and in Table I, is rep' oducible on a routine basis and gives yields of enzyme, per unit volume of culture filtrate, that are 20-fold those reported previously and a specific activity that is twice that reported previouslys. The increased yield results from studies of the culture conditions that favor increased production of enzyme. The yield of enzyme per unit volume is shown to be a function of the mycelial mass produced and is increased, for a given mycelial mass, by shortening the time required to reach the stage of maximum growth. The level of D-glucose used produced mycelial growth of maximum density, and the level of yeast extract used resulted in the production of this growth density in the minimum time. The end result of both effects il, the production of maximum enzyme yields per unit volume. Streptomycin was added to the culture medium to minimize the chance of bacterial contamination of the cultures, which is otherwise difficult to avoid under the conditions of growth used. The increased specific activity of the final enzyme-preparation results from elimination of the preparative-electrophoresis step used in the previous procedures. This step partially inactivates the enzyme without changing its electrophoretic mobility. Elimination of the preparative-electrophoresis step also greatly increases the enzyme capacity of the procedure, and O.5-g amounts of pure enzyme can be produced in a single preparation. TABLE I PURIFICATION SCHEME

Step

Total protein (mg)

Total units (x 10- 3)

Specific activity (units/mg protein)

Crude filtrate (8.1 liters) After concentration and dialysis (NH 4hS04 fractionation, 25-50% Chromatography on O-(diethylaminoethyl)cellulose" 5 Chromatography on O-(carboxymethyl)cellulose b

25,500 20,800 7,600

980 890 755

38 43 100

91 77

1.1 2.6

1,530

475

310

48

8.1

655

380

570

39

15.0

1 2 3 4

Activity yield (%)

Purification (fold)

"The O-(diethylaminoethyl)cellulose column (4.5 x 45 cm) was equilibrated with 2roM, pH 7.5, phosphate buffer and the enzyme added in the same buffer (l g in 40 mI). Irrigation was commenced with the same buffer and, after a protein peak had been eluted with the hold-up volume, a linear, 2-1 gradient was begun between the starting buffer and 50mM phosphate buffer, pH 7.5, containing 0.2M NaCI. The enzyme emerged as a skewed peak at about 1.3 I. bThe O-(carboxymethyl)cellulose column (4.5 x 45 cm) was equilibrated with roM, pH 6.0 phosphate buffer. The enzyme was placed on the column in the same buffer (0.5 g in 30 mI) and eluted with the same buffer until a protein peak eluted with the hold-up volume. A 2-1 linear gradient was then begun between the starting buffer and 50mM phosphate buffer, pH 6.0. The enzyme emerged as a single symmetrical peak near the end of the gradient.

Exo-(1-+3)-P-D-OLUCANASE FROM ABASIDIOMYCETE

275

Enzyme prepared by this procedure was homogenous in an ultracentrifuge, upon acrylamide-gel disc electrophoresis at both acid and alkaline pH, and upon cellulose acetate electrophoresis at pH 7. It also gave a single band upon sodium dodecyl sulfate-acrylamide-gel electrophoresis and thus apparently has no subunit structure. It gave a single peak by isoelectric focusing over the pH range 5-8 and had an isoelectric point at pH 6.5. The A~~':n value over the pH range 3-7 was 17.2. Analyses of the terminal amino acids by the Sanger procedure gave 0.64 moles ofleucine and 0.71 moles of serine per mole of enzyme, after correction for hydrolytic and chromatographic losses. Analysis for carboxyl-terminal amino acids by the hydrazinolysis procedure gave 0.73 moles of serine, 0.63 moles of glycine, and 0.64 moles of alanine per mole of enzyme, after correction for losses. Treatment of the enzyme with carboxypeptidase supported the carboxyl-terminal analysis by the hydrazinolysis procedure. Incubation of the denatured enzyme with carboxypeptidase A under the conditIOns described in the Experimental section for 20 min yielded serine, glycine, and alanine (the same terminal amino acids revealed by hydrazinolysis). Incubation for 0.5 h yielded threonine in addition to the aforementioned amino acids, and still longer incubation periods released additional amino acids. The amount of serine released after incubation for 24 h was quantitated as described for the Nterminal amino acids, and was found present at the level of 0.5 moles/mole of enzyme (after correction for chromatographic losses). These results are puzzling. The most obvious interpretation of both terminalgroup analyses is that the enzyme consists of more than one polypeptide chain. This is certainly possible, as it contains two disulfide linkages and these account for all of the cystine in the molecule 6 (it contains no cysteine). The problem lies in the fact that the amino-terminal data indicate two chains, linked (presumably) by disulfide linkages, whereas the carboxyl-terminal data indicate three chains. It is possible that there are three chains, in which case one of the N-terminal amino acids is not detected by the Sanger procedure. This would be the case if it were derivatized or if it were a proline or cysteine residue, neither of which are detectable by the procedure used here. Another possibility is that there are two chains and one of the supposed carboxylterminal residues is an artifact of the hydrazinolysis procedure. The results with carboxypeptidase do not permit a decision, as the method always gives a small amount of the penultimate as well as the ultimate residue, no matter how it is employed. There are some indications in the literature that hydrazinolysis can produce artifacts, although in a carefully controlled study of 15 proteins. Niu and Fraenkel-Conrat showed that none were produced 7 • However, the a-amylase from Aspergillus oryzae gives nearly equivalent amounts of alanine, glycine, and serine by hydrazinolysis, even though it is accepted as having only one N-terminal amino acids. A branched structure for the enzyme was proposed to account for the anomaly. Similarly, hydrazinolysis of the poD-galactosidase from Escherichia coli shows lysine, glycine, and traces of serine at the carboxyl terminal, even though there is but one N-terminal amino acid 9 • These latter authors suggest that the glycine and serine are artifacts generated by the hydrazinolysis procedure. These are not the only literature examples of discrepancy

276

D. R. PETERSON,S. KIRKVVOOD

between the number ofN- and carboxyl-terminal amino acids in proteins. The present N- and carboxyl-terminal data do not permit the question of the number of chains to be resolved. A study has been made of the effect of pH on the activity of the pure enzyme. It is completely stable over the pH range 3-7 for 24 h at 3r. Beyond this range it becomes markedly unstable; for instance, at pH values of 2 and 8, its half life is '" 15 min. However, by monitoring the time-course of the reaction, so that the initial rate can be determined by extrapolation, it is possible to obtain useful initial-rate data over the pH range 2.5-8. The effect of pH upon Km and Vmax is shown in Fig. 1. It is evident from Fig. 1 that no group titrating over this range is concerned with binding ofthe substrate, and it appears that a single group having a pKa value in the region of 6.5 is involved in the catalysis. This group is probably the protonated imidazole ling of the amino acid histidine, and the photoinactivation studies reported here lend support ot this idea. If a carboxylate group were involved in the catalysis, as is the case with lysozyme 10 , it must have a pK value considerably below 2.5. To investigate this possibility further, the effect of the esterifying agent, diphenyldiazomethane, was investigated. The enzyme is essentially insensitive to this reagent, although prolonged treatment results in significant esterification (1.6 carboxyl groups per enzyme molecule). This observation makes it unlikely that a carboxylate ion is part of the catalytic mechanism.

3

! .. S2

.~

>E

'"

.2 I L....--.L.._ _- ' - _ - - '_ _"7'::_ _""--_-::-'. 2.0 4.0 6.0 8.0 pH

Fig. l. Variation in the Km and Vmax values of the enzyme with pH. The initial rates at each pH value were determined at a number of substrate levels by monitoring the time-course of the reactions. All hydrolyses were performed in citrate-phosphate buffer. The initial rate-data were then used to determine graphically the Km and Vmax values at each pH value: ., log Vmax ; 0, pKm • VVhen the two legs of the log Vmax plot are extended, their intersect at pH 6.5 indicates that a group having a pK in this region is involved in the catalysis.

As a histidyl side-chain is probably the catalytic group that is titrated at pH 6.5, the effect of dye-activated (Rose Bengal) photoinactivation on the enzyme was investigated. Photooxidation at pH 7.0 caused rapid, total loss of enzyme activity. Rose Bengal is bleached under the experimental conditions used and, if the enzyme inactivation is corrected for dye bleaching, it is immediately evident (a) that there is

Exo-(1-+3)-P-D-GLUCANASE FROM A BASIDIOMYCETE

277

a rapid, first-order loss in enzyme activity over five orders of magnitude and (b) that substrate affords very considerable protection against inactivation (Fig. 2). This firstorder loss in activity is strong evidence that the decomposition of only a single type of functional group is occurring during the course of photoinactivation 11.

Fig. 2. Logarithmic plots of photooxidation of (l-+3)-P-o-glucanase in the presence and absence of substrate. The points have been corrected for a linear bleaching of the dye over the time-period investigated. Do is the initial dye absorbance and k is the rate constant for the decrease in dye absorbance (= 0.015 min-i). A o is the enzyme activity at t = 0 and A is the enzyme activity after time t. The reactions were illuminated with a focused beam from a 500-watt tungsten lamp and maintained in a constant-temperature (25°) water bath. The reactions were all performed in the presence of 0.003 % Rose Bengal at pH 7.0 in O.IM sodium phosphate buffer containing 0.5M sodium chloride. 0, enzyme (0.3tlM); e, enzyme (0.3tlM) + substrate (0.8mM); 6, enzyme (lOnM) + substrate (4mM).

There can be no doubt from these results that a group essential for the catalysis is very susceptible to photooxidation and that this group is in the region of the active site, as it is protected by the substrate. Although photooxidation at pH 7.0 is reasonably specific for histidine, and we have determined that the pH dependence of the photoinactivation is similar to that expected for histidine and totally different from that of the other photosensitive amino acids 11, the possibility exists that the loss in enzyme activity is due to decomposition of one of the other photosensitive amino acids, namely cysteine, cystine, methionine, tryptophan, and tyrosine. Cysteine can immediately be ruled out as the enzyme does not contain this amino acid and is completely insensitive to sulfhydryl reagents (iodoacetate, iodoacetamide). In order to investigate the effect of photooxidation on the other sensitive amino acids, the photooxidation was performed on large (3 mg) samples of the enzyme for various periods of time (Fig. 3), and these samples were then submitted to amino acid analysis after hydrolysis in acid and base. As the total quantity of protein in each sample was not known, it was estimated by the normalization procedure of Ray and Koshland 12. The ratio of tryptophan to tyrosine was determined by the method of Goodwin and Morton 13, and the tryptophan content by amino acid analysis. Only tyrosine,

278

D. R. PETERSON, S. KIRKWOOD

~

+0.4

I

OOL.!--'--~! --'---"c__--'-~'C__--'--__:!.

o

2

4

6

/iF 8

PHOTO OX IDATION TIME (min)

Fig. 3. Logarithmic plots of the loss in sensitive amino acid residues upon photooxidation. Conditions are the same as those described in Fig. 2, except that 3-mg samples of enzyme were photooxidized to determine each point and the points have not been corrected for dye bleaching. Samples were freed of dye by absorption on O-(diethylaminoethyl)cellulose and then hydrolyzed and subjected to amino acid analyses. See text for details. 0, enzyme activity; •• histidine; t::., methionine; ",", tryptophan; D. tyrosine.

8,..--r------,.--------,.------.---.-......,.---..--,

lii 6

9

o....,::II!!i!i._--'-_-'----''----'-_--:-_.L.--;!

o

2

4

6

8

PHOTPOXIDATION TIME (min)

Fig. 4. Residues decomposed by photooxidation. The actual number of residues decomposed in the experiment described in Fig. 3 are presented. 0, enzyme activity; •• histidine; t::., methionine; . , tryptophan; D, tyrosine.

Exo-(1--+3)-fJ-D-GLUCANASE FROM A BASIDIOMYCETE

279

tryptophan, methionine, and histidine were found to be affected, all of them decreasing with duration of photooxidation (Figs. 3 and 4). Cystine was not affected; apparently the cystine residues in this protein are not available to the active agent generated in the photooxidation procedure. Histidine is decomposed at a rate far greater than the other three amino acids, and the rate of decomposition (Fig. 4) is great enough to account for the observed inactivation. However, these data do not exclude the other three amino acids as in all cases an appreciable fraction of one mole of amino acid is decomposed by the time the enzyme is 50% inactivated. The fact that six moles of histidine have disappeared by this time does not necessarily mean that it is the key amino acid in the phenomenon. An attempt was made to rule out methionine and tryptophan through the use of specific reagents. Treatment of the enzyme with hydrogen peroxide at pH 3.0 would be expected to oxidize all accessible methionine residues and, at this pH, hydrogen peroxide is highly specific toward the oxidation of methionine 14 • Treatment of the enzyme with 0.22M hydrogen peroxide at pH 3.3 and 23° caused 20% of inhibition after one h, and there was still 50% of the enzyme activity left after 24 h. Thus, although the enzyme activity is slowly affected by acid peroxide, it is considered that this experiment rules out any possibility that the rapid and complete photoinactivation is due to decomposition of methionine. An attempt was made to determine the role of tryptophan by the use of N-bromosuccinimide, which is reasonably specific for this amino acid. Addition of 3 moles of the reagent per mole of enzyme caused a loss of 90% of the enzyme activity with the loss of between one and two particularly sensitive tryptophan residues (Fig. 5). However the enzyme retains some activity even after 25 moles of reagent have been added, at which point seven of the eight tryptophan residues in the enzyme have

80

(;l

60

z

Z -

l-

S; 20

i=

~

oL-==:::~~~~~ o 4 8 12 16 20 24 MOLES NBS ADDED/ MOLE ENZYME

Fig. 5. Loss in enzyme activity with addition of N-bromosuccinimide. The reaction was performed on 0.31 mg of enzyme in 1.0 m1 of 0.05M sodium acetate buffer, pH 4.0, at room temperature. Small (2-10 }lm1) aliquots of 3.1mM N-bromosuccinimide were added and samples were removed and assayed for enzyme activity to obtain each point.

280

D. R. PETERSON, S. KIRKWOOD

been decomposed. Similar results have been observed by Hayashi et al. 15 with the use of this reagent on lysozyme. Thus photooxidation of tryptophan residues is not responsible for the rapid, total loss in activity that occurs upon irradiation, but there can be no doubt that tryptophan has some function in the catalysis. This could well be a role in the binding of the substrate, as has been established for lysozyme 16 • This idea is supported by the observation that, over the range of 0-3 moles of added N-bromosuccinimide per mole of enzyme, the presence of substrate affords complete protection against inactivation. It is not possible to determine, by the use of derivatizing agents, which of the remaining pair of photosensitive amino acids (histidine and tyrosine) accounts for the sensitivity of the enzyme to photooxidation. Reagents that react with one react approximately equally with the other. Tetranitromethane, often considered specific for tyrosine, inactivates the enzyme. However, we have shown, by amino acid analysis of tetranitromethane-treated preparations of the enzyme, that both tyrosine and histidine are decomposed by this reagent 6. It is our opinion that a protonated, imidazole side-chain is the best candidate for the group with pK 6.5 that is involved in the catalysis. Tyrosine would require its pK value decreased by three full pH units if it were to function in this manner. It appears, on the basis of the foregoing data, that the Basidiomycete glucanase acts by a mechanism different from the employed by lysozyme 16. This is not surprising as there is no reason to believe that the lysozyme mechanism is the prototype for all glycosidases. In fact, there has been good reason for some time to consider that there are at least two distinct glycosidase mechanisms in Nature. It has long been known that glycosidases can be divided into two classes, those that invert configuration when they act and those that do not. It is evident that a mechanism that leads to inversion of configuration must be significantly different from one that does not, and since lysozyme does not invert configuration 1 7 whereas the Basidiomycete enzyme does 18, they must have different mechanisms of catalysis. It is useful to contrast these two mechanisms in the light of the present state of knowledge. In the Basidiomycete glucanase, the proton donOl in the acid catalysis is probably a plOtonated imidazole side-chain rather than the undissociated glutamic acid side-chain that serves this function in lysozyme 10. We have no data that rule out the possibility that the group in our enzyme titrating at pH 6.5 is a carboxyl group having a raised pKa value, as with glutamic acid-35 in lysozyme 1o . However, a histidine residue functioning in the hydrolytic mechanism would titrate at the same point, and as we have considerable evidence indicating involvement of histidine, we favor it as the proton donor at the present state of the evidence. The fact that the enzyme is insensitive to the action of diphenyldiazomethane could be interpreted as indicating that no carboxyl group of any kind is involved in the mechanism. However, the reagent could well be excluded from the active site for steric reasons, and so this evidence is not conclusive. If there is a basic group functioning to assist, as does the dissociated aspartate-52 in lysozyme, then it has the interesting property that it cannot be detected by titration over the pH range 2.5-8 and cannot be derivatized with diphenyldiazomethane. The binding

Exo-(1-+3)-{J-O-GLUCANASE FROM A BASIDIOMYCETE

281

phenomena observed by Huotari et al. 5 could well serve to increase the rate of hydrolysis, as do the binding forces exerted by lysozyme 10. These binding phenomena are much more extensive in the case of the Basidiomycete enzyme. It was observed that substrate binding (as measured by apparent Michaelis constants) increased steadily as the degree of polymerization of the substrate increased from two to 15 o-glucose units. This extensive binding could well playa more prominent role in the catalysis than is the case with lysozyme, thus making an assisting base unnecessary in the catalytic mechanism of the Basidiomycete enzyme. EXPERIMENTAL

Enzyme assay. - The assay procedure has been described previously 5, and one unit is defined as that amount of enzyme catalyzing the formation of one Jimole of o-glucose per min from the substrate (laminaran) under the conditions of the assay. Enzyme production and purification. - The enzyme was obtained from culture filtrates of Basidiomycete QM 806 by modifications of the procedure described previously 5. The modifications result in much larger yields of enzyme per unit volume of the medium and the purification scheme yields enzyme of higher specific activity and lends itself to the preparation of much larger amounts of pure enzyme. The organism was grown on a medium composed of 4% ofo-glucose, 1% of yeast extract and 0.3% of dipotassium hydrogen phosphate, all dissolved in tap water containing 0.003% of steptomycin sulfate. Fermentor vats (8.1 liters) were inoculated with lOO-ml inocula of 5-day old mycelia grown in the same medium in shake flasks. The vats were incubated at 30°, mechanically stirred, aerated at 10 I· min - 1, and the filtrates were harvested at the point of maximum enzyme concentration, which varied between 7 and 10 days. Initial stages of the enzyme purification were performed as previously described 5 , except that the ammonium sulfate fraction was taken between 25 and 50% of saturation. The DEAE-cellulose column step was similar to that previously described, except that the material used was Bio-Rad Cellex D. The final step, chromatography on 0-(carboxymethyl)cellulose, was performed on Whatman CM 11 material. Chemical and physical characterization. - Ultracentrifugation was performed on a Beckman-Spinco model E ultracentrifuge, An-D rotor, equipped with a 12-mm cell. The rotor speed was 59,840 r.p.m. and the enzyme was dissolved in 0.05M sodium acetate buffer, pH 4.8, at a concentration of 10 mg'ml- 1. Analytical disc-electrophoresis was performed by the method of Reisfeld, Lewis, and Williams 19 at the two pH values of 9.0 and 4.3. Isoelectric focusing was performed on an LKB model 8101 instrument with ampholine specified for the pH range 5-8, according to the directions supplied by the manufacturer. Amino acid composition was determined by means of an automatic amino acid analyzer (Beckman-Spinco model 120). Amino-terminal amino acids were determined by a modification of the procedure of Sanger 20 • After precipitation of the derivatized enzyme with 5% trichloro-

282

D. R. PETERSON, S. KIRKWOOD

acetic acid, it was suspended in 6M hydrochloric acid and hydrolyzed in a sealed tube for 16 h at HO°. After extraction and purification, the derivatized N-terminal amino acids were identified by ascending thin-layer polyamide chromatography as described by Wang and Wang 2 1, using Brinkman Instruments MN-Polygram polyamide-UV plates. Two solvent systems were used, 4:1 benzene-acetic acid and 1:1 formic acid-water. The derivatized amino acids were identified by co-chromatography in both solvent systems with authentic standards. They were quantitated on the basis of extinction coefficients given by Fraenkel-Conrat et al. 2 2, and losses during hydrolysis, elution, and chromatography were corrected for by use of N-(2,4-dinitrophenyl)amino acid standards throughout the procedure. Further identification of N-(2,4dinitrophenyl)leucine was made by the basic hydrolytic procedure of Roland and Gross 23 • Carboxyl-terminal amino acids were determined by the hydrazinolysis method, essentially as described by Niu and Fraenkel-Conrat 7 • Decomposition of C-terminal amino acids was corrected for by using the data of Niu and Fraenkel-Conrat 7 • The nature of the carboxyl-terminal amino acids was also investigated by hydrolysis with crystalline carboxypeptidase A, diisopropylphosphofluoridatetreated, from Worthington Biochemicals. The hydrolysis was performed on Basidiomycete glucanase that had been denaturated by heating for 10 min at 60° in 0.5% sodium lauryl sulfate, and the carboxypeptidase was then added at a level of 1 mg per 55 mg of (l-3)-P-D-glucanase in a total volume of 60 ml that was O.IM with respect to Tris buffer, pH 7.5, and Mwith respect to sodium chloride. Aliquots (10 ml) were taken 0,2.5, 5, 10,20, and 40 min and 24 h after the addition of the carboxypeptidase. The protein was precipitated with trichloroacetic acid and the liberated amino acids were derivatized with 2,4-dinitrofluorobenzene and identified as described for the N-terminal amino acids. ACKNOWLEDGMENTS

This paper is No. 8159, Scientific Series, Minnesota Agricultural Experiment Station. The present study was supported by Grant GM 09293, National Institutes of Health. REFERENCES

1 2 3 4 5 6 7 8 9 10 11

T. REESE AND M. MANDELS, Can. J. Microbiol., 5 (1959) 173. G. C. CHESTERS AND A. T. BULL, Biochem. J., 86 (1963) 31 and 38. T. E. NELSON, J. V. SCALETfI, F. SMITH, AND S. KIRKWOOD, Can. J. Chem., 41 (1963) 1671. T. E. NELSON, J. JOHNSON, E. JANTZEN, AND S. KIRKWOOD, J. BioI. Chem., 244 (1969) 5972. F. I. HUOTARI, T. E. NELSON, F. SMITH, AND S. KIRKWOOD, J. BioI. Chem., 243 (1968) 952. R. JEFFCOAT AND S. KIRKWOOD, unpublished results. C. I. Nro AND H. rRAENKEL-CONRAT, J. Amer. Chem. Soc., 77 (1955) 5882. T. IKENEKA, J. Biochem. (Tokyo), 43 (1956) 255. J. L. BROWN, S. KooRAJIAN, J. KATZE, AND I. ZABIN, J. BioI. Chem., 241 (1966) 2826. D. C. PHILLIPS, Proc. Nat. Acad. Sci. U.S., 57 (1957) 484. E. W. WESTHEAD, Biochemistry, 4 (1965) 2139. E. C.

Exo-(1-.3)-P-D-GLUCANASE FROM A BASIDIOMYCETE

12 13 14 15 16 17 18 19 20 21 22 23

J. Bioi. Chem., 237 (1962) 2493. A. MORTON, Biochem. J., 40 (1946) 628. N. P. NEUMANN, Methods Enzymol., 11 (1967) 485. K. HAYASHI, T. IMOTO, AND M. FuNATSU, J. Biochem. (Tokyo), 54 (1963) 381. L. M. JOHNSON AND D. C. PHILLIPS, Nature, 206 (1965) 761. J. A. RUPLEY AND V. GATES, Proc. Nat. Acad. Sci., U.S., 57 (1967) 496. T. E. NELSON, J. Bioi. Chem., 245 (1970) 869. R. A. REISFELD, V. J. LEWIS, AND D. E. WILLIAMS, Nature, 195 (1962) 281. F. SANGER, Biochem. J., 39 (1945) 507. K. T. WANG AND I. S. Y. WANG, J. Chromatogr., 27 (1967) 318. H. FRAENKEL-CONRAT, J. I. HARRIS, AND A. L. LEVY, Methods Biochem. Anal., 2 (1955) 359. J. F. ROLAND AND A. M. GROSS, Anal. Chem., 26 (1954) 502. W. J.

T.

RAy

AND

D. E.

KOSHLAND,

W. GOODWIN AND R.

283

Studies on the structure and mechanism of an exo-(1 yields 3)-beta-D-glucanase from Basidiomycete QM806.

Carbohydrate Research, 41 (1975) 273-283 © Elsevier Scientific Publishing Company, Amsterdam - Printed in Belgium STUDIES ON THE STRUCTURE AND MECHAN...
655KB Sizes 0 Downloads 0 Views