- Email: [email protected]
Comparison of the hydrolytic activity and fluorescence of native, guanidine hydrochloride-treated and renatured cellobiohydrolase I from Trichoderma reesei
Biochimica et Biophysica Acta, 1037 (1990) 81-85 Elsevier
81
BBAPRO 33532
Comparison of the hydrolytic activity and fluorescence of native, guanidine hydrochloride-treated and renatured cellobiohydrolase I from Trichoderma reesei J o n a t h a n W o o d w a r d , N o r m a n E. Lee, J e f f r e y S. C a r m i c h a e l *, S t e v e n L. M c N a i r * * a n d J o h n M. W i c h e r t * * * Chemical Technology Division, Oak Ridge National Laboratory +, Oak Ridge, TN (U.S.A.) (Received 28 June 1989)
Key words: Cellulase; Cellobiohydrolase I; Denaturation; Guanidine hydrochloride; Renaturation; Cellulose
Guanidine hydrochloride (GdnHCl) is an effective agent for the elution of cellulase protein from unhydrolyzed cellulosic residues, but once eluted the enzyme is inactive. The studies described in this paper examine the effect of GdnHCI on the hydrolytic activity and tryptophan fluorescence of cellobiohydrolase I (CBH I) from Trichoderma reesei. CBH I was found to be completely inactivated by 0.25 M GdnHCi, but higher concentrations of GdnHC! were required to partially unfold this enzyme, as determined from the measurement of a decrease in its tryptophan fluorescence. Binding of CBH 1 to microcrystalline cellulose was prevented by 4 M GdnHCI, suggesting that a conformational change of CBH I resulted in the loss of substrate binding. Removal of the denaturant from CBH I by dialysis or gel filtration allowed the kinetics of the reactivation of CBH I, after 4 M GdnHCI treatment, to be studied. The fluorescence and specific hydrolytic activity of native and renatured CBH I were comparable. It is concluded, therefore, that GdnHC! may be used to elute cellulase components, such as CBH I, adsorbed on undigested cellulosic substrates since this component can easily be renatured and subsequently reused.
Introduction
Fungal cellulases that catalyze the hydrolysis of insoluble cellulose to glucose are composed of cellobiohydrolases (1,4-fl-D-glucan cellobiohydrolase, EC 3.2.1.91), endoglucanases (1,4-(1,3;1,4)-fl-o-glucan 4glucanohydrolase, EC 3.2.1.4) and fl-glucosidases (fl-Dglucoside glucohydrolase, EC 3.2.1.21) [1]. They can be used for the hydrolysis of lignocellulosic materials to glucose; however, under industrial conditions, there is incomplete hydrolysis of the substrate, and much of the
+ Operated by Martin Marietta Energy Systems, Inc., under Contract DE-AC05-84OR21400 with the U.S. Department of Energy. * Oak Ridge Science and Engineering Research Semester Participant from Slippery Rock University, Slippery Rock, PA. * * Technology Intern from Pellissippi State Technical Community College, Knoxville, TN. *** Oak Ridge Science and Engineering Research Semester Participant from the University of Nebraska, Lincoln, NE.
Correspondence: J. Woodward, Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37831-6194 (USA).
enzyme remains adsorbed to the lignocellulosic residue [2-6]. Recovery and reuse of adsorbed cellulase components are recognized as important issues for energy conservation and reduction in costs associated with the enzymatic hydrolysis of cellulose [7]. Reese [2] has tested many compounds for their ability to elute cellulase protein components that are bound to microcrystalline cellulose. Sodium hydroxide, guanidine hydrochloride (GdnHCI) and urea were the best eluants (> 90% of bound protein eluted), but the disadvantage to their use was that cellulase activity was inactivated by these eluants. Since high concentrations of GdnHCI (4 M) and urea (6 M) were required to effect elution of cellulase components from cellulose, the eluted components were, presumably, denatured and hence inactivated. Although many GdnHCl-denatured proteins are capable of spontaneous refolding [8], the conditions under which GdnHCl-denatured cellulase enzyme components are capable of renaturation have not been studied. However, the fact that a step in the purification of bacterial cellulase involves its elution from a cellulose affinity column with 8 M GdnHCI suggests that renaturation of cellulase activity is likely [9].
0167-4838/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
82 The cellobiohydrolase I (CBH I) component comprises as much as 60% of the total protein in a crude cellulase preparation. Thus, the majority of cellulase protein bound to a cellulosic residue, at the completion of hydrolysis, will be CBH I, especially since it possesses greater affinity for cellulose than do other cellulase components [1,2]. The recovery and subsequent reuse of CBH I are, therefore, particularly important, but the use of denaturants as eluants requires that the eluted enzyme can be subsequently renatured with, hopefully, full recovery of activity. In this study, the hydrolytic activity and fluorescence of purified T. reesei CBH I in its native, GdnHCl-denatured and renatured states have been compared. Materials and Methods
Purification and assay of CBH I Commercial cellulase from T. reesei (Celluclast 250S) was obtained from Novo Laboratories, Inc., Wilton, CT 06897. CBH I was purified to homogeneity using ionexchange chromatography and isoelectric focusing as described in Ref. 10. The isoelectric focusing procedure was carried out using the Pharmacia FPLC apparatus equipped with a Mono P H R 5 / 2 0 column. CBH I was assayed using p-nitrophenyl-fl-D-cellobioside (PNPC) as the substrate [11]. In the standard assay, 50 #1 of CBH I (0.7 m g / m l protein) was incubated with 0.5 ml of 10 mM PNPC at 23°C in 50 mM sodium acetate buffer (pH 5.0). At time intervals, an 0.1-ml aliquot of the reaction mixture containing PNP was added to 1.0 M sodium hydroxide; the resulting absorbance at 402 nm was measured within 1 min using a Perkin-Elmer Lamda Array spectrophotometer interfaced with an IBM personal computer. Purified CBH I was also a generous gift from Sharon P. Shoemaker, Genencor, Inc., San Francisco, CA 94080. Fluorescence spectrum The fluorescence spectrum of CBH I was recorded using a Perkin-Elmer LS-5 fluorescence spectrophotometer. T h e sample was excited at 280 nm, and the fluorescence emitted was measured at 350 nm, the point at which tryptophan fluorescence of CBH I dominated [121. Inactivation, denaturation and renaturation of CBH I by GdnHCI The effect of GdnHC1 on the activity of CBH I was determined by assaying the enzyme in the presence of different concentrations of GdnHC1. Denaturation, or unfolding of the CBH molecules, was monitored by measuring the decrease in fluorescence of a solution of CBH (5 #1) in 1.0 ml of GdnHC1 (0-6 M, in 5 mM sodium acetate buffer (pH 5.0) at 23 o C) at 350 nm over a given time interval. Renaturation of CBH I (250 #1)
after its denaturation with 8 M GdnHC1 (250 #1) at 23°C for 60 rain was achieved by dialysis at 4 ° C or gel filtration at 23°C using a PD-10 Sephadex G-25 column (Pharmacia) equilibrated with 50 mM sodium acetate buffer (pH 5.0) to effect removal of the denaturant.
Binding to and hydrolysis of Avicel by CBH 1 Avicel PH 105 (microcrystalline cellulose) was obtained from FMC Corporation, Philadelphia, PA, U.S.A. Samples of native, denatured and renatured CBH I were tested for their ability to bind to and hydrolyze Avicel, as described in Ref. 13. In the hydrolysis of Avicel, approx. 50 #g of native or renatured CBH was incubated with 1% ( w / v ) Avicel at pH 5.0 and 50°C, in a total volume of 5.0 ml. After the reaction had proceeded for 4.0 h, the reaction mixture was supplemented with 0.5 ml of fl-glucosidase, 3.2 u n i t s / m l (purified from A. niger crude cellulase Grade 1, as described in Ref. 14). Binding of CBH I to Avicel was determined by mixing a given concentration of CBH I protein with 1% ( w / v ) Avicel in a 400-#1 volume for 2 min at room temperature and extrapolating the bound protein from that remaining in the supernatant. Analytical procedures Protein was determined using the Coomassie Blue reagent (Bio-Rad Laboratories, Richmond, CA, U.S.A.) according to the method of Bradford [15]. Glucose was measured by using the hexokinase assay reagent (Sigma Chemical Company, St. Louis, MO). The homogeneity of CBH I was determined by SDS-gel electrophoresis and analytical liquid chromatography using an H P 1090 liquid chromatograph employing a 7.5 × 7.5 mm Bio-Gel TSK-DEAE-5W Column (Bio-Rad). Plots of absorbance against wavelength and time presenting the data in three-dimensional form were achieved using the HP 1040A data-evaluation pack II (EVALU2) interfaced with the HP 1090 liquid chromatograph. Results and Discussion
Characterization of purified CBH I from T. reesei Purified CBH I eluted from the Mono P column between pH 3.5-4.0. This indicates its low isoelectric point which is in agreement with this measurement by others [16-18]. It was homogeneous, as judged by SDSgel electrophoresis ( M r 63 340) and by analytical HPLC. The enzyme hydrolyzed only the aglycon bond of PNPC and possessed no ability to hydrolyze cellobiose (10 mM) or reduce the viscosity of a 1% ( w / v ) solution of carboxymethylcellulose (CMC), indicating it was free of fl-glucosidase and endoglucanase activities, respectively. Denaturation and inactivation of CBH I by GdnHCI The fluorescence spectrum of native CBH I that was excited at 280 nm (Fig. 1) indicates a peak of maximum
83 100
I
I
i
100
I
80
B 8O
60
Z t.~ to
w t~
40
60
20
q
0
0.1
0
~.
40
0.2
GUANIDINE
E---
HYDROCHLORIDE
0.3 (M)
Fig. 3. The effect of GdnHCI on the PNPCase activity of CBH I. CBH I (50 /~1) was incubated with 5 0 0 / t l 10 mM PNPC in 50 mM sodium acetate buffer (pH 5.0), containing 0 to 0.25 M GdnHC1 at 23 o C. At intervals of time, a 0.1-ml aliquot was removed and assayed for PNP concentration.
5 Q~
20
0 I J I 280 320 360 400 440 480
the addition of a reducing agent such as dithiothreitol or B-mercaptoethanol (data not shown). Guanidine hydrochloride inactivated the PNPC-hydrolyzing activity of CBH I (Fig. 3), but the loss in activity did not correlate with the decrease in fluorescence or partial unfolding of the enzyme. This is explained by the almost complete inactivation of CBH I that occurred at 0.25 M GdnHC1. It appears, therefore, that low concentrations of GdnHC1 inhibit CBH I, resulting in an apparent loss in activity rather than partial unfolding or denaturation, which requires much higher concentrations of this denaturant. 50% inhibition of Avicelase (microcrystalline cellulose hydrolyzing activity) of T. reesei cellulase by 0.1 M GdnHCI has also been noted previously [2]. The possibility that low concentrations of GdnHC1 cause a small localized change in the conformation of CBH I (indicated by an increase in tryptophan fluorescence), resulting in its inactivation, cannot be completely discounted. Further studies on the
WAVELENGTH OF EMISSION
(rim)
Fig. 1. Fluorescence spectrum of native CBH I. The solution of CBH I (3.5 ~ g / m l 5 mM sodium acetate buffer (pH 5.0)) was excited at 280 nm. (A) Buffer only; (B) CBH I in buffer.
emission at 350 nm. Excitation of a solution of GdnHC1 at 280 nm did not exhibit any emission at 350 nm that could have masked changes in the fluorescence of CBH I caused by its denaturation with GdnHC1. The effect of 4.5 M GdnHC1 on the fluorescence of CBH I is shown in Fig. 2A. These data show that only a partial or local unfolding of CBH I occurs, even at high concentrations of denaturant (Fig. 2B), and that the extent of unfolding is largely complete after 30 rain. This would seem likely, since this enzyme has been reported to contain 12 disulfide bridges [19J. An extensive decrease in fluorescence or the unfolding of CBH I requires
120
I
I
I
I
I
I
I
l
I
I
I
B
A 110 w
o
100
LIJ (J W n,~ 0 LL
90 80
I,I > D I.--
70
J Ld tY
60 50 40
I
I
I
I
I
I
I
I
I
I
I
10
20
30
40
50
60
1
2
3
4
5
TIME (rain)
6
GUANIDINE HYDROCHLORIDE (M)
Fig. 2. The effect of GdnHC1 on the fluorescence of CBH I. Time course of the decrease in tryptophan fluorescence at 350 nm of CBH I by 4.5 M GdnHCI: (A), effect of GdnHCI concentration; (B), fluorescence measured 1 h after incubation with CBH I. The closed triangles represent the fluorescence of 4 M GdnHCl-treated CBH I after its renaturation by gel filtration.
84 "~
E ~s
6
I
I
I
[
I
110
I ]
I
"1
I
I
lOO 90
0
m 5 z
E
Q- 1 ml
c~ 0
/
80
•
O
I
I
I
I
I
2
4
6
8
10
I 1
12
CBH I PRO'rEIN ADDED (/.z.g//lO0 ,u,L)
Fig. 4. The adsorption of native and renatured CBH I to (1% w/v) microcrystalline cellulose (Avicel). for details, see Material and Methods. Native (o) or renatured (4) CBH I.
z 0 .< > (.3 <
70 60 50
/
40 3o 20 lO
kinetics of the inhibition of CBH I by G d n H C I are in progress. It should also be noted that GdnHC1 did not affect the measurement of PNP. The adsorption kinetics for the binding of CBH I to Avicel, given in Fig. 4, were completely abolished in the presence of 4 M GdnHC1. Consequently, there was no hydrolysis of Avicel by C B H I in the presence of this concentration of GdnHC1. The extent of binding of CBH I to Avicel decreased as the concentration of GdnHC1 present in the reaction mixture increased; binding was completely prevented by 2 M GdnHC1. The finding that 4 M GdnHC1 prevents the binding of CBH I to Avieel suggests that partial unfolding or denaturation of C B H I, resulting in decreased tryptophan fluorescence, is the mechanism whereby binding is prevented. Since an essential tryptophan residue has been implicated in the binding of some ceUulases to cellulose [20], altering the position of this residue in the enzyme molecule could, therefore, affect binding to the substrate. On the other hand, the inhibition of CBH I by the positively charged guanidinium group of G d n H C I is probably due to its interaction with a carboxylate group present in the active site of C B H I (such a group has been implicated in the active site of several glycosidases including cellulase [21-23]). It is apparent that fungal and bacterial cellulases possess catalytic and binding domains, the latter protruding from the former [9,24], giving the molecule a tadpole-like shape [25]. The interaction of the essential groups in the catalytic domain of C B H I with low concentrations of GdnHC1 does not appear to affect the conformation of the binding domain deleteriously even though a small increase in fluorescence of CBH I was noticed. It has been reported that reduction of the binding domain of C B H I b y mercaptoethanol abolished cellulose binding, suggesting the requirement of the native conformation for binding [26]. Contrary to this is the finding that mercaptoethanol is ineffective in eluting cellulase from cellulose [2].
o o
I
i
i
i
1
2
3
4
Tilde
OF
DIALYSIS
5
(h)
Fig. 5. Kinetics of the reactivation of GdnHCl-denatured CBH I. GdnHCI-denatured CBH I (350 #g/ml) was dialyzed against 50 mM sodium acetate buffer (pH 5.0). At the indicated times, 100-~1 samples were withdrawn from the dialyzate and assayed for their PNPChydrolyzing activity.
Renaturation and reactivation o f C B H I
After treatment of C B H I with 4 M GdnHC1, the solution was dialysed overnight at 4 ° C to remove the denaturant. The dialysed, denatured enzyme was assayed for its ability to hydrolyze P N P C and Avicel to determine whether renaturation had occurred. The observed kinetics of the reactivation of G d n H C l - d e n a tured C B H I (Fig. 5) indicate that m a x i m u m recovery of activity is obtained after 3 h. Also, there was no obvious correlation between the yield of reactivation and the concentration of CBH I between 0.15 and 1.0 m g / m l . Yields of reactivation of some denatured proteins are dependent upon their concentration [8]. The K m and the Avicel-binding capability of the renatured enzyme were also measured. The data in Table I indicate that there is little difference between the specific activity and K m of the native and renatured CBH I with respect to P N P C hydrolysis. Purified C B H
TABLE I Comparison of the specific activity and Km of native and renatured CBH I with respect to PNPC hydrolysis
Specific activity (nmol PNP/ min per mg protein) K m (raM)
Native
Renatured
Native a
52 2.2
51 2.9
66 n.d. b
a Source of CBH I: Genencor. b n.d., not determined.
85 12 ,---. "5
11
~ z
9 8
~0 ~ 0 ~: Q_
7 6 5 4
oo ~, o
3 2 1
0
I
I
I
I
ak
Acknowledgments The authors thank Drs. D.A. Graves and G.W. Strandberg for reviewing the manuscript and Ms. Debbie Weaver and Ms. Janice Pruitt for secretarial assistance. This work was supported by the Office of Basic Energy Sciences, U.S. Department of Energy, under contrast DE-AC05-84OR21400 with Martin Marietta Energy Systems, Inc., and by the Solar Energy Research Institute. References
0
I
I
I
I
1
2
3
4
TIME ( h )
Fig. 6. Hydrolysis of Avicel (1%, w/v) by 50 #g of native and renatured CBH I. For details, see Materials and Methods. Native (o) or renatured (A) CBH I. The arrow indicates the time at which fl-glucosidase was added.
I obtained from Genencor also possessed a specific activity similar to that of the CBH I prepared for this study. Renatured CBH I was similar to the native enzyme in its Avicel-binding and hydrolysis properties (Figs. 4 and 6). It appears, therefore, that in terms of their catalytic activity, native and renatured CBH I are very similar to each other. The relative fluorescence of renatured CBH I (after denaturation with 4 M GdnHC1) was approx. 95% of the native enzyme (Fig. 2B), suggesting that the original conformation had been restored. Conclusions
It has been shown that the purified CBH I component of 7". reesei cellulase is partially unfolded at high concentrations (> 2 M) of GdnHC1. However, lower concentrations (0.25 M) are only required to abolish activity. Since this loss in CBH I activity by treatment with GdnHC1 cannot be correlated with a decrease in its intrinsic tryptophan fluorescence, inhibition of CBH I does not appear to be related to a gross conformational change in this enzyme. However, the loss of Avicel-binding capability of CBH I clearly can be correlated with at least a partial unfolding of its tertiary structure. The fact that CBH I can be renatured with full recovery of activity and conformation suggests that GdnHCI can be used to elute cellulase from undigested cellulosic residues. The eluted enzyme can be subsequently reactivated by removal of the denaturant and hence reused. In related experiments, cellulase that is adsorbed on a lignocellulosic residue, such as steam-exploded aspen wood, can also be successfully eluted by treatment with 4 M GdnHCI (data not included).
1 Wood, T.M. (1985) Biochem. Soc. Trans. 13, 407-410. 2 Reese, E.T. (1982) Process Biochem. 17, 2-8. 3 Fujishima, S., Yaku, F. and Koshijama, T. (1987) Mokuzai Gakkaishi 33, 992-993. 4 Mes-Hartree, M., Hogan, C.M. and Saddler, J.N. (1987) Biotechnol. Biceng. 30, 558-564. 5 Tatsumoto, K., Baker, J.O., Tucker, N.P., Oh, K.K., Mohagheghi, A., Grohmann, K. and Himmel, M.E. (1988) Appl. Biochem. Biotechnol. 18, 159-174. 6 Chernoglazov, V.M., Ermolova, O.V. and Klyosov, A.A. (1988) Enzyme Micro& Technol. 10, 503-507. 7 Woodward, J. and Bales, J.C. (1989) in Bioproducts and Bioprocesses (Fieehter, A., Okada, H. and Tanner, R., eds.), pp. 87-101, Springer, Berlin. 8 Jaenicke, R. (1987) Prog. Biophys. Mol. Biol. 49, 117-237. 9 Gilkes, N.R., Warren, R.A.J., Miller, R.C., Jr., and Kilburn, D.G. (1988) J. Biol. Chem. 263, 10401-10407. 10 Wood, T.M., McCrae, S.I. and MacFarlane, C.C. (1980) Biochem. J. 189, 51-65. 11 Deshpande, M.V., Pettersson, L.G. and Eriksson, K.-E. (1988) Methods Enzymol. 160, 126-130. 12 Bagshaw, C.R. and Harris, D.A. (1987) in Spectrophotometry and Spectrofluorimetry, a Practical Approach (Harris, D.A. and Bashford, C.L., eds.), pp. 91-113, IRL Press, Oxford. 13 Woodward, J., Hayes, M.K. and Lee, N.E. (1988) Bio/Technology 6, 301-304. 14 Woodward, J., Marquess, H.J. and Picker, C.S. (1986) Prep. Biochem. 16, 337-352. 15 Bradford, M.M. (1976) Anal. Biochem. 72, 248-254. 16 Shoemaker, S., Walt, K., Tsitovsky, G. and Cox, R. (1983) Bio/Technology 1,687-690. 17 Hayn, M. and Esterbauer, H. (1985) J. Chromatogr. 329, 379-387. 18 Enari, T.-M. and Niku-Paavola, M.-L. (1987) Crit. Rev. Biotechnol. 5, 67-87. 19 Bhikhabhai, R. and Pettersson, G. (1984) Biochem. J. 222, 729-736. 20 Clarke, A.J. (1987) Biochim. Biophys. Aeta 912, 424-431. 21 Yaguchi, M., Roy, C., Rollin, C.F., Paice, M.G. and Jurasek, L. (1982) Biochem. Biophys. Res. Commun. 116, 408-411. 22 Woodward, J. and Wiseman, A. (1982) in Developments in Food Carbohydrate 3 (Lee, C.K. and Lindley, M.G., eds.), pp. 1-21, Applied Science Publications, London. 23 Clarke, A.J. and Yaguchi, M. (1985) Eur. J. Biochem. 149, 233-238. 24 Tomme, P., Van Tilbeurgh, H., Pettersson, G., Van Damme, J., Vandekerckhove, J., Knowles, J., Ten'i, T. and Claeyssens, M. (1988) Eur. J. Biochem. 170, 575-581. 25 Schmuck, M., Pilz, I., Hayn, M. and Esterbauer, H. (1986) Biotechnol. Lett. 8, 397-402. 26 Johansson, G., St~thlberg, J., Lindeberg, G., Engstr~m, A. and Pettersson, G. (1989) FEBS Left. 243, 389-393.
81
BBAPRO 33532
Comparison of the hydrolytic activity and fluorescence of native, guanidine hydrochloride-treated and renatured cellobiohydrolase I from Trichoderma reesei J o n a t h a n W o o d w a r d , N o r m a n E. Lee, J e f f r e y S. C a r m i c h a e l *, S t e v e n L. M c N a i r * * a n d J o h n M. W i c h e r t * * * Chemical Technology Division, Oak Ridge National Laboratory +, Oak Ridge, TN (U.S.A.) (Received 28 June 1989)
Key words: Cellulase; Cellobiohydrolase I; Denaturation; Guanidine hydrochloride; Renaturation; Cellulose
Guanidine hydrochloride (GdnHCl) is an effective agent for the elution of cellulase protein from unhydrolyzed cellulosic residues, but once eluted the enzyme is inactive. The studies described in this paper examine the effect of GdnHCI on the hydrolytic activity and tryptophan fluorescence of cellobiohydrolase I (CBH I) from Trichoderma reesei. CBH I was found to be completely inactivated by 0.25 M GdnHCi, but higher concentrations of GdnHC! were required to partially unfold this enzyme, as determined from the measurement of a decrease in its tryptophan fluorescence. Binding of CBH 1 to microcrystalline cellulose was prevented by 4 M GdnHCI, suggesting that a conformational change of CBH I resulted in the loss of substrate binding. Removal of the denaturant from CBH I by dialysis or gel filtration allowed the kinetics of the reactivation of CBH I, after 4 M GdnHCI treatment, to be studied. The fluorescence and specific hydrolytic activity of native and renatured CBH I were comparable. It is concluded, therefore, that GdnHC! may be used to elute cellulase components, such as CBH I, adsorbed on undigested cellulosic substrates since this component can easily be renatured and subsequently reused.
Introduction
Fungal cellulases that catalyze the hydrolysis of insoluble cellulose to glucose are composed of cellobiohydrolases (1,4-fl-D-glucan cellobiohydrolase, EC 3.2.1.91), endoglucanases (1,4-(1,3;1,4)-fl-o-glucan 4glucanohydrolase, EC 3.2.1.4) and fl-glucosidases (fl-Dglucoside glucohydrolase, EC 3.2.1.21) [1]. They can be used for the hydrolysis of lignocellulosic materials to glucose; however, under industrial conditions, there is incomplete hydrolysis of the substrate, and much of the
+ Operated by Martin Marietta Energy Systems, Inc., under Contract DE-AC05-84OR21400 with the U.S. Department of Energy. * Oak Ridge Science and Engineering Research Semester Participant from Slippery Rock University, Slippery Rock, PA. * * Technology Intern from Pellissippi State Technical Community College, Knoxville, TN. *** Oak Ridge Science and Engineering Research Semester Participant from the University of Nebraska, Lincoln, NE.
Correspondence: J. Woodward, Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge, TN 37831-6194 (USA).
enzyme remains adsorbed to the lignocellulosic residue [2-6]. Recovery and reuse of adsorbed cellulase components are recognized as important issues for energy conservation and reduction in costs associated with the enzymatic hydrolysis of cellulose [7]. Reese [2] has tested many compounds for their ability to elute cellulase protein components that are bound to microcrystalline cellulose. Sodium hydroxide, guanidine hydrochloride (GdnHCI) and urea were the best eluants (> 90% of bound protein eluted), but the disadvantage to their use was that cellulase activity was inactivated by these eluants. Since high concentrations of GdnHCI (4 M) and urea (6 M) were required to effect elution of cellulase components from cellulose, the eluted components were, presumably, denatured and hence inactivated. Although many GdnHCl-denatured proteins are capable of spontaneous refolding [8], the conditions under which GdnHCl-denatured cellulase enzyme components are capable of renaturation have not been studied. However, the fact that a step in the purification of bacterial cellulase involves its elution from a cellulose affinity column with 8 M GdnHCI suggests that renaturation of cellulase activity is likely [9].
0167-4838/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
82 The cellobiohydrolase I (CBH I) component comprises as much as 60% of the total protein in a crude cellulase preparation. Thus, the majority of cellulase protein bound to a cellulosic residue, at the completion of hydrolysis, will be CBH I, especially since it possesses greater affinity for cellulose than do other cellulase components [1,2]. The recovery and subsequent reuse of CBH I are, therefore, particularly important, but the use of denaturants as eluants requires that the eluted enzyme can be subsequently renatured with, hopefully, full recovery of activity. In this study, the hydrolytic activity and fluorescence of purified T. reesei CBH I in its native, GdnHCl-denatured and renatured states have been compared. Materials and Methods
Purification and assay of CBH I Commercial cellulase from T. reesei (Celluclast 250S) was obtained from Novo Laboratories, Inc., Wilton, CT 06897. CBH I was purified to homogeneity using ionexchange chromatography and isoelectric focusing as described in Ref. 10. The isoelectric focusing procedure was carried out using the Pharmacia FPLC apparatus equipped with a Mono P H R 5 / 2 0 column. CBH I was assayed using p-nitrophenyl-fl-D-cellobioside (PNPC) as the substrate [11]. In the standard assay, 50 #1 of CBH I (0.7 m g / m l protein) was incubated with 0.5 ml of 10 mM PNPC at 23°C in 50 mM sodium acetate buffer (pH 5.0). At time intervals, an 0.1-ml aliquot of the reaction mixture containing PNP was added to 1.0 M sodium hydroxide; the resulting absorbance at 402 nm was measured within 1 min using a Perkin-Elmer Lamda Array spectrophotometer interfaced with an IBM personal computer. Purified CBH I was also a generous gift from Sharon P. Shoemaker, Genencor, Inc., San Francisco, CA 94080. Fluorescence spectrum The fluorescence spectrum of CBH I was recorded using a Perkin-Elmer LS-5 fluorescence spectrophotometer. T h e sample was excited at 280 nm, and the fluorescence emitted was measured at 350 nm, the point at which tryptophan fluorescence of CBH I dominated [121. Inactivation, denaturation and renaturation of CBH I by GdnHCI The effect of GdnHC1 on the activity of CBH I was determined by assaying the enzyme in the presence of different concentrations of GdnHC1. Denaturation, or unfolding of the CBH molecules, was monitored by measuring the decrease in fluorescence of a solution of CBH (5 #1) in 1.0 ml of GdnHC1 (0-6 M, in 5 mM sodium acetate buffer (pH 5.0) at 23 o C) at 350 nm over a given time interval. Renaturation of CBH I (250 #1)
after its denaturation with 8 M GdnHC1 (250 #1) at 23°C for 60 rain was achieved by dialysis at 4 ° C or gel filtration at 23°C using a PD-10 Sephadex G-25 column (Pharmacia) equilibrated with 50 mM sodium acetate buffer (pH 5.0) to effect removal of the denaturant.
Binding to and hydrolysis of Avicel by CBH 1 Avicel PH 105 (microcrystalline cellulose) was obtained from FMC Corporation, Philadelphia, PA, U.S.A. Samples of native, denatured and renatured CBH I were tested for their ability to bind to and hydrolyze Avicel, as described in Ref. 13. In the hydrolysis of Avicel, approx. 50 #g of native or renatured CBH was incubated with 1% ( w / v ) Avicel at pH 5.0 and 50°C, in a total volume of 5.0 ml. After the reaction had proceeded for 4.0 h, the reaction mixture was supplemented with 0.5 ml of fl-glucosidase, 3.2 u n i t s / m l (purified from A. niger crude cellulase Grade 1, as described in Ref. 14). Binding of CBH I to Avicel was determined by mixing a given concentration of CBH I protein with 1% ( w / v ) Avicel in a 400-#1 volume for 2 min at room temperature and extrapolating the bound protein from that remaining in the supernatant. Analytical procedures Protein was determined using the Coomassie Blue reagent (Bio-Rad Laboratories, Richmond, CA, U.S.A.) according to the method of Bradford [15]. Glucose was measured by using the hexokinase assay reagent (Sigma Chemical Company, St. Louis, MO). The homogeneity of CBH I was determined by SDS-gel electrophoresis and analytical liquid chromatography using an H P 1090 liquid chromatograph employing a 7.5 × 7.5 mm Bio-Gel TSK-DEAE-5W Column (Bio-Rad). Plots of absorbance against wavelength and time presenting the data in three-dimensional form were achieved using the HP 1040A data-evaluation pack II (EVALU2) interfaced with the HP 1090 liquid chromatograph. Results and Discussion
Characterization of purified CBH I from T. reesei Purified CBH I eluted from the Mono P column between pH 3.5-4.0. This indicates its low isoelectric point which is in agreement with this measurement by others [16-18]. It was homogeneous, as judged by SDSgel electrophoresis ( M r 63 340) and by analytical HPLC. The enzyme hydrolyzed only the aglycon bond of PNPC and possessed no ability to hydrolyze cellobiose (10 mM) or reduce the viscosity of a 1% ( w / v ) solution of carboxymethylcellulose (CMC), indicating it was free of fl-glucosidase and endoglucanase activities, respectively. Denaturation and inactivation of CBH I by GdnHCI The fluorescence spectrum of native CBH I that was excited at 280 nm (Fig. 1) indicates a peak of maximum
83 100
I
I
i
100
I
80
B 8O
60
Z t.~ to
w t~
40
60
20
q
0
0.1
0
~.
40
0.2
GUANIDINE
E---
HYDROCHLORIDE
0.3 (M)
Fig. 3. The effect of GdnHCI on the PNPCase activity of CBH I. CBH I (50 /~1) was incubated with 5 0 0 / t l 10 mM PNPC in 50 mM sodium acetate buffer (pH 5.0), containing 0 to 0.25 M GdnHC1 at 23 o C. At intervals of time, a 0.1-ml aliquot was removed and assayed for PNP concentration.
5 Q~
20
0 I J I 280 320 360 400 440 480
the addition of a reducing agent such as dithiothreitol or B-mercaptoethanol (data not shown). Guanidine hydrochloride inactivated the PNPC-hydrolyzing activity of CBH I (Fig. 3), but the loss in activity did not correlate with the decrease in fluorescence or partial unfolding of the enzyme. This is explained by the almost complete inactivation of CBH I that occurred at 0.25 M GdnHC1. It appears, therefore, that low concentrations of GdnHC1 inhibit CBH I, resulting in an apparent loss in activity rather than partial unfolding or denaturation, which requires much higher concentrations of this denaturant. 50% inhibition of Avicelase (microcrystalline cellulose hydrolyzing activity) of T. reesei cellulase by 0.1 M GdnHCI has also been noted previously [2]. The possibility that low concentrations of GdnHC1 cause a small localized change in the conformation of CBH I (indicated by an increase in tryptophan fluorescence), resulting in its inactivation, cannot be completely discounted. Further studies on the
WAVELENGTH OF EMISSION
(rim)
Fig. 1. Fluorescence spectrum of native CBH I. The solution of CBH I (3.5 ~ g / m l 5 mM sodium acetate buffer (pH 5.0)) was excited at 280 nm. (A) Buffer only; (B) CBH I in buffer.
emission at 350 nm. Excitation of a solution of GdnHC1 at 280 nm did not exhibit any emission at 350 nm that could have masked changes in the fluorescence of CBH I caused by its denaturation with GdnHC1. The effect of 4.5 M GdnHC1 on the fluorescence of CBH I is shown in Fig. 2A. These data show that only a partial or local unfolding of CBH I occurs, even at high concentrations of denaturant (Fig. 2B), and that the extent of unfolding is largely complete after 30 rain. This would seem likely, since this enzyme has been reported to contain 12 disulfide bridges [19J. An extensive decrease in fluorescence or the unfolding of CBH I requires
120
I
I
I
I
I
I
I
l
I
I
I
B
A 110 w
o
100
LIJ (J W n,~ 0 LL
90 80
I,I > D I.--
70
J Ld tY
60 50 40
I
I
I
I
I
I
I
I
I
I
I
10
20
30
40
50
60
1
2
3
4
5
TIME (rain)
6
GUANIDINE HYDROCHLORIDE (M)
Fig. 2. The effect of GdnHC1 on the fluorescence of CBH I. Time course of the decrease in tryptophan fluorescence at 350 nm of CBH I by 4.5 M GdnHCI: (A), effect of GdnHCI concentration; (B), fluorescence measured 1 h after incubation with CBH I. The closed triangles represent the fluorescence of 4 M GdnHCl-treated CBH I after its renaturation by gel filtration.
84 "~
E ~s
6
I
I
I
[
I
110
I ]
I
"1
I
I
lOO 90
0
m 5 z
E
Q- 1 ml
c~ 0
/
80
•
O
I
I
I
I
I
2
4
6
8
10
I 1
12
CBH I PRO'rEIN ADDED (/.z.g//lO0 ,u,L)
Fig. 4. The adsorption of native and renatured CBH I to (1% w/v) microcrystalline cellulose (Avicel). for details, see Material and Methods. Native (o) or renatured (4) CBH I.
z 0 .< > (.3 <
70 60 50
/
40 3o 20 lO
kinetics of the inhibition of CBH I by G d n H C I are in progress. It should also be noted that GdnHC1 did not affect the measurement of PNP. The adsorption kinetics for the binding of CBH I to Avicel, given in Fig. 4, were completely abolished in the presence of 4 M GdnHC1. Consequently, there was no hydrolysis of Avicel by C B H I in the presence of this concentration of GdnHC1. The extent of binding of CBH I to Avicel decreased as the concentration of GdnHC1 present in the reaction mixture increased; binding was completely prevented by 2 M GdnHC1. The finding that 4 M GdnHC1 prevents the binding of CBH I to Avieel suggests that partial unfolding or denaturation of C B H I, resulting in decreased tryptophan fluorescence, is the mechanism whereby binding is prevented. Since an essential tryptophan residue has been implicated in the binding of some ceUulases to cellulose [20], altering the position of this residue in the enzyme molecule could, therefore, affect binding to the substrate. On the other hand, the inhibition of CBH I by the positively charged guanidinium group of G d n H C I is probably due to its interaction with a carboxylate group present in the active site of C B H I (such a group has been implicated in the active site of several glycosidases including cellulase [21-23]). It is apparent that fungal and bacterial cellulases possess catalytic and binding domains, the latter protruding from the former [9,24], giving the molecule a tadpole-like shape [25]. The interaction of the essential groups in the catalytic domain of C B H I with low concentrations of GdnHC1 does not appear to affect the conformation of the binding domain deleteriously even though a small increase in fluorescence of CBH I was noticed. It has been reported that reduction of the binding domain of C B H I b y mercaptoethanol abolished cellulose binding, suggesting the requirement of the native conformation for binding [26]. Contrary to this is the finding that mercaptoethanol is ineffective in eluting cellulase from cellulose [2].
o o
I
i
i
i
1
2
3
4
Tilde
OF
DIALYSIS
5
(h)
Fig. 5. Kinetics of the reactivation of GdnHCl-denatured CBH I. GdnHCI-denatured CBH I (350 #g/ml) was dialyzed against 50 mM sodium acetate buffer (pH 5.0). At the indicated times, 100-~1 samples were withdrawn from the dialyzate and assayed for their PNPChydrolyzing activity.
Renaturation and reactivation o f C B H I
After treatment of C B H I with 4 M GdnHC1, the solution was dialysed overnight at 4 ° C to remove the denaturant. The dialysed, denatured enzyme was assayed for its ability to hydrolyze P N P C and Avicel to determine whether renaturation had occurred. The observed kinetics of the reactivation of G d n H C l - d e n a tured C B H I (Fig. 5) indicate that m a x i m u m recovery of activity is obtained after 3 h. Also, there was no obvious correlation between the yield of reactivation and the concentration of CBH I between 0.15 and 1.0 m g / m l . Yields of reactivation of some denatured proteins are dependent upon their concentration [8]. The K m and the Avicel-binding capability of the renatured enzyme were also measured. The data in Table I indicate that there is little difference between the specific activity and K m of the native and renatured CBH I with respect to P N P C hydrolysis. Purified C B H
TABLE I Comparison of the specific activity and Km of native and renatured CBH I with respect to PNPC hydrolysis
Specific activity (nmol PNP/ min per mg protein) K m (raM)
Native
Renatured
Native a
52 2.2
51 2.9
66 n.d. b
a Source of CBH I: Genencor. b n.d., not determined.
85 12 ,---. "5
11
~ z
9 8
~0 ~ 0 ~: Q_
7 6 5 4
oo ~, o
3 2 1
0
I
I
I
I
ak
Acknowledgments The authors thank Drs. D.A. Graves and G.W. Strandberg for reviewing the manuscript and Ms. Debbie Weaver and Ms. Janice Pruitt for secretarial assistance. This work was supported by the Office of Basic Energy Sciences, U.S. Department of Energy, under contrast DE-AC05-84OR21400 with Martin Marietta Energy Systems, Inc., and by the Solar Energy Research Institute. References
0
I
I
I
I
1
2
3
4
TIME ( h )
Fig. 6. Hydrolysis of Avicel (1%, w/v) by 50 #g of native and renatured CBH I. For details, see Materials and Methods. Native (o) or renatured (A) CBH I. The arrow indicates the time at which fl-glucosidase was added.
I obtained from Genencor also possessed a specific activity similar to that of the CBH I prepared for this study. Renatured CBH I was similar to the native enzyme in its Avicel-binding and hydrolysis properties (Figs. 4 and 6). It appears, therefore, that in terms of their catalytic activity, native and renatured CBH I are very similar to each other. The relative fluorescence of renatured CBH I (after denaturation with 4 M GdnHC1) was approx. 95% of the native enzyme (Fig. 2B), suggesting that the original conformation had been restored. Conclusions
It has been shown that the purified CBH I component of 7". reesei cellulase is partially unfolded at high concentrations (> 2 M) of GdnHC1. However, lower concentrations (0.25 M) are only required to abolish activity. Since this loss in CBH I activity by treatment with GdnHC1 cannot be correlated with a decrease in its intrinsic tryptophan fluorescence, inhibition of CBH I does not appear to be related to a gross conformational change in this enzyme. However, the loss of Avicel-binding capability of CBH I clearly can be correlated with at least a partial unfolding of its tertiary structure. The fact that CBH I can be renatured with full recovery of activity and conformation suggests that GdnHCI can be used to elute cellulase from undigested cellulosic residues. The eluted enzyme can be subsequently reactivated by removal of the denaturant and hence reused. In related experiments, cellulase that is adsorbed on a lignocellulosic residue, such as steam-exploded aspen wood, can also be successfully eluted by treatment with 4 M GdnHCI (data not included).
1 Wood, T.M. (1985) Biochem. Soc. Trans. 13, 407-410. 2 Reese, E.T. (1982) Process Biochem. 17, 2-8. 3 Fujishima, S., Yaku, F. and Koshijama, T. (1987) Mokuzai Gakkaishi 33, 992-993. 4 Mes-Hartree, M., Hogan, C.M. and Saddler, J.N. (1987) Biotechnol. Biceng. 30, 558-564. 5 Tatsumoto, K., Baker, J.O., Tucker, N.P., Oh, K.K., Mohagheghi, A., Grohmann, K. and Himmel, M.E. (1988) Appl. Biochem. Biotechnol. 18, 159-174. 6 Chernoglazov, V.M., Ermolova, O.V. and Klyosov, A.A. (1988) Enzyme Micro& Technol. 10, 503-507. 7 Woodward, J. and Bales, J.C. (1989) in Bioproducts and Bioprocesses (Fieehter, A., Okada, H. and Tanner, R., eds.), pp. 87-101, Springer, Berlin. 8 Jaenicke, R. (1987) Prog. Biophys. Mol. Biol. 49, 117-237. 9 Gilkes, N.R., Warren, R.A.J., Miller, R.C., Jr., and Kilburn, D.G. (1988) J. Biol. Chem. 263, 10401-10407. 10 Wood, T.M., McCrae, S.I. and MacFarlane, C.C. (1980) Biochem. J. 189, 51-65. 11 Deshpande, M.V., Pettersson, L.G. and Eriksson, K.-E. (1988) Methods Enzymol. 160, 126-130. 12 Bagshaw, C.R. and Harris, D.A. (1987) in Spectrophotometry and Spectrofluorimetry, a Practical Approach (Harris, D.A. and Bashford, C.L., eds.), pp. 91-113, IRL Press, Oxford. 13 Woodward, J., Hayes, M.K. and Lee, N.E. (1988) Bio/Technology 6, 301-304. 14 Woodward, J., Marquess, H.J. and Picker, C.S. (1986) Prep. Biochem. 16, 337-352. 15 Bradford, M.M. (1976) Anal. Biochem. 72, 248-254. 16 Shoemaker, S., Walt, K., Tsitovsky, G. and Cox, R. (1983) Bio/Technology 1,687-690. 17 Hayn, M. and Esterbauer, H. (1985) J. Chromatogr. 329, 379-387. 18 Enari, T.-M. and Niku-Paavola, M.-L. (1987) Crit. Rev. Biotechnol. 5, 67-87. 19 Bhikhabhai, R. and Pettersson, G. (1984) Biochem. J. 222, 729-736. 20 Clarke, A.J. (1987) Biochim. Biophys. Aeta 912, 424-431. 21 Yaguchi, M., Roy, C., Rollin, C.F., Paice, M.G. and Jurasek, L. (1982) Biochem. Biophys. Res. Commun. 116, 408-411. 22 Woodward, J. and Wiseman, A. (1982) in Developments in Food Carbohydrate 3 (Lee, C.K. and Lindley, M.G., eds.), pp. 1-21, Applied Science Publications, London. 23 Clarke, A.J. and Yaguchi, M. (1985) Eur. J. Biochem. 149, 233-238. 24 Tomme, P., Van Tilbeurgh, H., Pettersson, G., Van Damme, J., Vandekerckhove, J., Knowles, J., Ten'i, T. and Claeyssens, M. (1988) Eur. J. Biochem. 170, 575-581. 25 Schmuck, M., Pilz, I., Hayn, M. and Esterbauer, H. (1986) Biotechnol. Lett. 8, 397-402. 26 Johansson, G., St~thlberg, J., Lindeberg, G., Engstr~m, A. and Pettersson, G. (1989) FEBS Left. 243, 389-393.