Communications to the Editor

Transcription

1 Communications to the Editor Synergism in Cellulose Hydrolysis & y Endoglucanases and Exoglucanases Purified from Trichoderma viride G. Beldman, A. G. J. Voragen. F. M. Rombouts, and W. Pilnik Department of Food Science, Agricultural University, De Dreyen 12, 6703 BC Wageningen, The Netherlands Accepted for publication December 30, 1986 One of the most intriguing phenomena in cellulose hydrolysis by the cellulases of several fungi is the synergistic action between the individual components of these enzyme mixtures. The enzymes which play a main role in synergism are the endoglucanase [ 1,4P~glucan glucanohydrolase (EC )] and the exoglucanase [ 1,CP~glucan cellobiohydrolase (EC 3.2. l.91)]. Both types of enzymes were found to exist in multiple forms in the culture filtrates of Trichoderma viride, Trichoderma Koningii, 4,5 Sporotrichum pulverulentum, Penicillium funiculosum,7~s Fusarium solani, and Talaromyces emersonii. lo Synergism in cellulose hydrolysis by the cellulase complex of several of these fungi has been demonstrated by recombination experiments of the purified endo and exogl~canases.~.~ Also cross synergism has been observed between the endo and exoacting enzymes from different microbial origins.7 Synergism in the enzymatic degradation of crystalline cellulose is not simply explained by the sequential attack of endoglucanases and exoglucanases. For instance it was found by Wood and McCrae that one of the four endoglucanases from T. Koningii, which was the most random acting enzyme in cellulose attack, and therefore expected to create relatively many chainends for an exoglucanase to act on, showed only very little synergistic effect when working together with the cellobiohydrolase from this fungus. These authors state that the formation of a complex of endoglucanase and cellobiohydrolase on the surface of the cellulose chain is essential for synergism. From the mechanistic model described by Okazaki and MooYoung, it was concluded that the degree of synergistic effect (DSE), which is defined as the ratio of the observed activity of the combined endo and exoglucanase enzymes, to the sum of the observed individual activities, greatly depends on the ratio of endoglucanase and exoglucanase, as well as on the degree of polymerization of the substrate. The model was used to predict synergistic effects, using kinetic data obtained from incubations with soluble substrates. The value of this model to predict a degree of synergistic effect during hydrolysis of insoluble crystalline cellulose is questionable. An explanation for synergism from another point of view was given by Ryu and coworkers, 2 who studied the sequential adsorption of the endoglucanase and exoglucanase components of T. reesei and found that synergism was gradually introduced into the nonadsorbed enzyme fraction by competitive adsorption of one of the glucanases. These investigators state that synergism of the cellulase complex remains undisturbed during adsorption, but show no data on synergism between cellulases in the adsorbed fraction. The work was done with partially purified enzymes, which still contained several isoenzymes. In two previous reports, we described the purification and biochemical characterization of all endoglucanases and exoglucanases of a commercial enzyme preparation from T. viride, as well as the adsorption and kinetic behaviors of these enzymes. l3 This communication presents the results of the recombination experiments with these purified enzymes. The synergistic effects which were observed could partially be explained in the light of the reported adsorption behaviors of these cellulases. MATERIALS AND METHODS Enzyme Preparations Maxazyme CL is a commercial cellulase preparation from Trichoderma viride origin and was provided by Gist Brocades (Delft, the Netherlands). From this preparation six endoglucanases (Endo I, I1 111, IV, V, and VI) and two exoglucanases (Ex0 I1 and III) were isolated as described in a previous report. Enzymes were used as such; No stabilizing agents such as bovine serum albumin were added because the cellulases retained their full activity for at least 24 h at 30 C. Biotechnology and Bioengineering, Vol. 31, Pp (1988) John Wiley & Sons, Inc. CCC /88/ $04.00

2 Recombination Experiments Endoglucanases in Combination with Ex0 Ill on Avicel Cellulose From the endoglucanase preparations (Endo I, 11,111, IV, V, and VI), a fixed amount of 90 pg enzyme was combined with an amount of Exo 111 ranging from 6 to 200 pg protein. Incubation with 0.5% Avicel cellulose (Type SF, Serva, Heidelberg, FRG) took place under continuously mixing on a roller drum during 20 h at 30 C in 0.05M sodium acetate buffer, ph 5.0. The final volume was 1 ml. In a parallel experiment, the endoglucanases and Exo I11 were incubated separately using the same conditions as in the combination experiment. After incubation the cellulose was centrifuged off (10 min at 3000 g) and the supematant was analyzed on released reducing sugars by the method of SomogyiL4 as well as on total sugars by the method of Dubois et al. Endo IV in Combination with Ex0 Ill on H3POKSwollen Cellulose Recombination of Endo IV with Exo 111 was carried out as described above, using 0.5% H3P0,swollen cellulose instead of Avicel. The H3P04swollen cellulose was prepared by the method of Wood.I6 Fixed Amounts of Endoglucanases in Combination with Ex0 I/ and Ex0 Ill on Avicel Cellulose Endoglucanases (25 pg) were combined with 25 pg of Exo I1 or Exo 111, respectively. Incubation of the combinations as well as of the separate enzymes took place during 20 h at 30 C in a volume of 1 ml of 0.05M sodium acetate, ph 5.0, with 0.5% Avicel cellulose. After removal of residual cellulose, the amount of reducing sugars in the supernatant was determined. Endo Ill in Combination with Endo I on Avicel Cellulose Endo 111 was recombined with Endo I, using 90 pg of both enzymes, and incubated at 30 C with 0.5% Avicel for 20 h. The final volume was 1 ml of O.05M sodium acetate, ph 5.0. The same incubation was carried out with the separate enzymes. Products were analyzed as described above. Endoglucanases and Ex0 Ill in Combination with Maxazyme CL on Avicel Cellulose Maxazyme CL, in a final concentration of 500 pg/ml, was combined with the individual endoglucanases, respectively, and Exo 111, and incubated at 30 C in 0.05M sodium acetate buffer, ph 5.0. The purified glucanases were used in a concentration of 77 pg/ml. After incubation for 20 h, the cellulose was centrifuged and the amount of reducing sugars in the supernatant was determined and compared to the values obtained from separate incubations of Maxazyme CL and purified cellulases, using the same conditions. RESULTS Endoglucanases in Combination with Exo 111 Recombination of the purified exoglucanase Exo 111 and endoglucanases Endo I, II, 111, IV, V, and VI resulted in all cases in a synergistic effect after incubation with Avicel crystalline cellulose. The degree of synergistic effect (DSE) could be calculated by comparing the activity of the combinations to the sum of the activities of the individual enzymes. By using a fixed concentration of endoglucanase (90 pg/ml) and a varying concentration of Exo 111 it was found that the DSE depends upon the endoglucanase to exoglucanase ratio. Figure 1 shows that for all combinations of endoglucanase with Exo 111 a maximum DSE was observed at a specific ratio of these enzymes. The maximal value of DSE was dependent upon the methods of analysis of the hydrolysis products. Analysis of released reducing sugars, which gives a value for every glucosidic bond which is enzymatically cleaved, resulted in maximal DSE values varying between 1.60 [Fig. l(c)] and 3.45 [Fig. 1(E)]. Determination of total sugars by the method of Dubois represents a solubilization effect of the enzymatic breakdown of cellulose and gives, with the exception of the combination of Endo 111 with Exo 111 [Fig. l(c)], lower maximal DSE values between 1.75 and Solubilization values of Avicel cellulose by the action of endoglucanase and Exo 111 alone, as well as by the combination of enzymes, are displayed in Table I. When working alone, the purified enzymes can only scarcely attack Avicel cellulose. The highest value was obtained for Endo 111, in accordance to its relatively high specific activity. Using the combined enzyme preparations, much more solubilization was observed, ranging % in 20 h, depending on which endoglucanase was used. These values are of the same magnitude as for long term hydrolysis of cotton by the combined action of a cellobiohydrolase and an endoglucanase from T. koningii, or as observed in a similar experiment using fliter paper or microcrystalline cellulose as substrates and enzymes from Talaromyces emersonii. The weight ratio of endoglucanase to exoglucanase, where maximal DSE occurs, is given in Table 11. Taking into account the molecular masses of these enzymes, the molar ratios could also be calculated. In our previous study on adsorption of purified endoglucanases and exoglucanases, we showed that Endo I, 111, and V adsorb strongly on Avicel cellulose. Especially, these enzymes show their maximal DSE at a much lower endoglucanase to exoglucanase ratio than Endo 11, IV, and VI when working in concert with Exo 111. Synergistic effects in hydrolysis of amorphous H3P04 swollen cellulose was investigated for the combination of Endo IV and Exo 111. From Figure 1(D) it can be seen that 174 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 31, FEBRUARY 1988

3 LL 10 I0 l Endo /Ex0 ( g /g) Figure 1. Degree of synergistic effect (DSE), obtained after recombination of Ex0 111 with various amounts of (A) Endo I, (B) Endo 11, (C) Endo 111, (D) Endo IV, (E) Endo V, and (F) Endo VI, during hydrolysis of (AF) Avicel and (D) H3P0,swollen cellulose. Products were measured as reducing sugars (RS) as well as total sugars (TS):(A) Avicel, RS; (X) Avicel, TS; (0) H,PO,swollen cellulose, RS and; (5) H,PO,swollen celluiose, TS. Table I. Solubilization of Avicel crystalline cellulose by endoglucanases and Exo 111 alone, as well as in combination. The endoglucanases and exoglucanase concentrations were 90 and 100 pg/ml respectively. For other experimental details, see the text. Alone Percent solubilization In combination with Exo 111 Enzyme ("/.I ("/.I Endo I 4.1 Endo I1 3.0 Endo I Endo IV 1.2 Endo V 3.0 Endo VI 1.7 Ex0 I the combination of these enzymes results in DSE values, which are much lower than those values obtained in hydrolysis of Avicel cellulose. Also, a much broader maximum was observed, at a lower endoglucanase to exoglucanase ratio. Table 11. Initial endoglucanase to Exo 111 ratios, at which maximal DSE was observed under the experimental conditions. Exo 111 in combination with Endo I" Endo I1 Endo 111" Endo IV Endo V" Endo VI Mass ratio Adsorb strongly on Avicel cellulose (ref. 13). Endo/Exo 111 Molar ratio Endoglucanases in Combination with Exo I1 and Exo 111 The two cellobiohydrolases Exo I1 and 111, which were isolated from Maxazyme CL, were compared with respect to their capacities to act synergistically when working together with the endoglucanases on Avicel cellulose (Table 111). DSE values measured for the combinations of Exo I1 with the endoglucanases were found to be lower than the values COMMUNICATIONS TO THE EDITOR 175

4 Table 111. DSE values obtained for the combination of endoglucanases with Exo I1 and Exo 111 using a mass ratio of 1. In combination with Endoglucanase Ex0 I1 Ex0 111 Endo I Endo I Endo 111 I.o I.2 Endo IV Endo V Endo VI obtained in a similar experiment using Ex Actually no synergism was found for the combination of Exo I1 with Endo I and 111. The enzyme concentration used in these experiments was 25 Fg/mL for each enzyme (Endo/ Exo = 1). At this protein concentration, a combination of an endoglucanase with Exo 111 exhibited a lower DSE value than observed in the previous experiment in which a protein concentration of 90 pg/ml was used for both enzymes. Endo 111 in Combination with Endo I In a previous report, some doubts remained on the endo/exoglucanase nature of Endo 111. In order to investigate whether or not Endo 111 is a true endoglucanase, the ability of this enzyme to act synergistically with Endo I was studied. Endo I was shown to be a true endoglucanase. A DSE value of 0.97 was found, indicating that these two enzymes are not able to act synergistically and Endo I11 cannot be classified as an exoglucanase. Endoglucanases and Exo 111 in Combination with Maxazyme CL The synergistic effects which could be obtained with a combination of two purified cellulases, was also studied in a much more complex mixture of these types of enzymes. Maxazyme CL, the original cellulase preparation from which the endo and exoglucanases were purified, was supplemented with the individual enzymes and investigated on its Avicel hydrolyzing capacity, in comparison with the activities obtained in separate incubations. Figure 2 shows that endoglucanases 11,111, IV, V, and VI are able to increase the activity of the original cellulase preparation, by acting synergistically with it. The DSE values obtained varied between 1.06 for a combination with Endo I1 to 1.24 when Endo 111 was combined with Maxazyme CL. Endo I and Exo 111, however, gave a negative synergistic effect when supplemented to the original cellulase preparation: DISCUSSION In a previous article, we demonstrated the different adsorption characteristics of endoglucanases and exoglucanases from T. vide on Avicel cell~lose. ~ The same substrate was used in this study in order to investigate whether or not these adsorption characteristics could be related to the synergistic behaviors of the enzymes. It is shown that all six endoglucanases (Endo I, 11,111, IV, V, and VI) which were isolated from the commercial cellulase preparation Maxazyme CL, can act synergistically with Exo 111, obtained from the same preparation, in hydrolysis of crystalline Avicel cellulose. The degree of synergistic effect (DSE) for a specific combination of endo and exoglucanase, is at least dependent upon two main factors: firstly the ratio in which both enzymes are combined and secondly the nature of the subslrate which is used (in these experiments crystalline Avicel or amorphous H3P04 swollen cellulose). Although Avicel is not a completely native cellulose, such as cotton fiber, and comparatively results in some lower DSE values upon incubation with recombined cellulases, it is commonly used as a model substrate to investigate hydrolysis of insoluble crystalline cellulose ~17 The fact that the DSE obtains a maximal value at a specific endoglucanase to exoglucanase ratio is in good agreement with the theoretical model of Okazaki, who stated that enzymatic cellulolysis is strongly affected by this ratio, as well as by the degree of polymerization of the substrate. However, these authors used their model on the basis of soluble substrates, assuming homogeneous reaction conditions. When insoluble crystalline cellulose is used, the reaction is heterogeneous and another factor is introduced: the selective adsorption characteristics of the individual endoand exoglucanases. The results, presented in this paper, show that for hydrolysis of Avicel cellulose the optimal initial endoglucanase to exoglucanase ratio is greatly dependent upon the adsorption behaviors of these enzymes. Endo I, Endo 111, and Endo V, which enzymes are known to adsorb strongly on Avicel,I3 exhibited their maximal synergistic effect with Ex0 111 (which also adsorbs strongly on Avicel) at a much lower initial endoglucanasetoexoglucanase ratio than in those incubations where Endo 11, Endo IV, and Endo VI were used. It was shown in our previous work that these latter endoglucanases adsorb only slightly on Avicel. The observations can be explained by assuming that only those cellulases which are adsorbed on the insoluble cellulose chains, participate in the formation of a complex of endo and exoglucanase, which is responsible for the synergistic effect on these crystalline substrates. In agreement to the findings of Wood and M~Crae,~ the activity of an exoglucanase in this complex can only find expression after the formation of a new chainend by the endoglucanase. On the other hand, the action of an endoglucanase can only be expressed if it is immediately followed by the removal of a cellobiose molecule, as the result of the action of an exoglucanase, in order to prevent the reformation of the first broken bond. Endoglucanases such as Endo 11, IV, and VI need to be added in a relatively high concentration to create a driving force which is strong enough for the adsorption of at least a part of the available enzyme molecules and the formation of the endoglucanaseexoglucanase complex. The results indicate that maximal DSE occurs at a specific optimal ratio 176 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 31, FEBRUARY 1988

5 DSE 1.4 Exo ID EndoI Endon Endom Endo IP Endo Y E ndom Figure 2. Degree of synergistic effect (DSE), obtained after recombination of the original cellulase preparation Maxazyme CL with purified endogiucanases and Exo 111. of adsorbed endoglucanases and exoglucanases. It is very likely that the endo and exoglucanase need to be present in such a complex in a 1 : 1 ratio, in order to obtain a maximal DSE. It is clear that this optimum depends more upon the adsorption characteristics, than upon the hydrolytic nature (specific activity, degree of randomness, oligosaccharide composition of the product) of the individual enzymes. These latter properties of the enzymes determine a.0. the different DSE values at optimal ratio of the various combinations as well as the differences between DSE values obtained by measurement of the product as reducing sugars (bondbreaking activity) or total sugars (solubilization activity) for one specific combination. The fact that Endo 111, in combination with Exo 111, gives a relatively low DSEvalue at the optimal ratio may be related to the low degree of randomness in cellulose attack of this enzyme. We suggested that this enzyme acts according to a single chain multiple attack mechanism, which is in accordance to the fact that the DSE value, measured as bondbreaking activity and as solubilization activity, differ only slightly for the combination of this enzyme with Exo 111. This may be explained by the fact that Endo I11 is more or less restricted to the same cellulose chain in a crystalline matrix and solubilization parallels hydrolytic activity. On the contrary, the action of the other endoglucanases (Endo I, 11, IV, V, and VI), which jump from chain to chain (multiplechain attack), will not always lead to solubilization, because the chain is fixed in a crystalline matrix. However, it can result in a new chainend for an exoglucanase to act on, creating new measurable reducing endgroups. The fact that synergism between Endo IV and Exo 111 on amorphous H3P04swollen cellulose is less striking than on crystalline cellulose, is in good agreement with data from the literature. Synergism on this susceptible substrate is essentially explained by the fact than an endoglucanase creates a new chainend for an exoglucanase to act on and that this latter action leads directly to soluble products which can be measured. Besides, we also found that DSE on this type of amorphous cellulose was to a lesser extent determined by the ratio of both enzymes, which is expressed in a broader maximum [Fig. l(d)]. Probably more encounters of one of these enzymes and this type of more or less soluble substrate lead to the release of a measurable product. This effect is only slightly increased by the addition of the complementary enzyme. This type of synergism takes place in solution (soluble synergism) while synergism on Avicel especially takes place on the crystalline surface (insoluble synergism). The endoglucanase nature of Endo I11 was discussed in our previous study since this enzyme exhibited a relatively low activity towards CMcellulose and a high activity towards Avicel, compared with the other endoglucanases. Through the fact that synergism was positive with Exo I11 and negative in combination with Endo I, we confirm its endoglucanase nature. Synergism can also be observed after addition of purified cellulases to a more complex mixture of these enzymes. Supplementing Maxazyme CL with the endoglucanases 11, 111, IV, V, and VI resulted in a higher activity than the sum of activities obtained in the separate incubations. One can conclude that Maxazyme CL is deficient in these endoglucanases. However, this effect cannot be generalized since addition of Endo I gave a negative synergistic effect, indicating again that synergism is more than simple succes COMMUNICATIONS TO THE EDITOR 177

6 sive action of endoglucanase and exoglucanase, but rather related to other properties, such as adsorption nature of these enzymes. During hydrolysis of cellulose by complex mixtures of cellulases, such as culture filtrates and commercial enzyme preparations, there will be a competition between the individual enzymes to occupy the available adsorption sites and to form the endoglucanaseexoglucanase complex, mentioned above. Taking into account that, in the case of T. vide, at least eight enzymes with different adsorption parameters are involved, one can conclude that it is very difficult to predict the optimal composition of an enzyme mixture in order to obtain maximal synergism. References 1. G. Beldman, M. F. Searlevan Leeuwen, F. M. Rombouts, and F. G. J. Voragen, Eur. J. Biochem., 146, 301 (1985). 2. S. P. Schoemaker and R. D. Brown, Biochem. Biophys. Actu. 523, 133 (1978). 3. S. P. Schcernaker and R. D. Brown, Biochem. Biophys. Acru, 523, 147 (1978). 4. T. M. Wood and S. I. McCrae, Proceedings ofrhe Bioconversion Symposium (IIT, New Delhi, 1977), pp T. M. Wood and S. I. McCrae, Biochem. J., 171, 61 (1978). 6. K. E. Eriksson and B. Pettersson, Eur. J. Biochem., 51, 193 (1975). 7. T. M. Wood, S. I. McCrae, and C. C. Macfarlane, Biochem. J., 189, 51 (1980). 8. T. M. Wood and S. I. McCrae, Curbohydr. Res., 110, 291 (1982). 9. T. M. Wood and S. I. McCrae, Curbohydr. Res., 57, 117 (1977). 10. A. McHale and M. P. Coughlan, FEBS Letr., 117, 319 (1980). 11. M. Okazaki and M. MooYoung, Biotechnol. Bioeng., 20, 637 ( 1978). 12. D. D. Y. Ryu, C. Kim, and M. Mandels, Biotechnol. Bioeng., 26,488 (1984). 13. G. Beldman, A. G. J. Voragen, F. M. Rombouts, M. F. Searlevan Leeuwen, and W. Pilnik, Biotechnol. Bioeng., 30, 251, (1987). 14. M. Somogyi, J. Bid. Chem., 195, 19 (1952). 15. M. Dubois, K. A. Gilles, J. K. Hamiltion, F. A. Rebers, and F. Smith, Anal. Chem., 28, 350 (1956). 16. T. M. Wood, Biochem. J., 121, 353 (1971). 17. M. Streamer, K. E. Eriksson, and B. Pettersson, Eur. J. Biochem., 59, 607 (1975). 178 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 31, FEBRUARY 1988