Bioresource Technology 167 (2014) 582–586
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Factors affecting cellulose hydrolysis based on inactivation of adsorbed enzymes Zhuoliang Ye 1, R. Eric Berson ⇑ Department of Chemical Engineering, University of Louisville, Louisville, KY 40292, USA
h i g h l i g h t s Factors affecting hydrolysis were examined based on an enzyme inactivation model. The activation energy for inactivation was within 10% of that for hydrolysis. Increasing temperature is effective only to improve the initial hydrolysis rate. Increasing the surface binding area of substrate can improve the hydrolysis rate.
a r t i c l e
i n f o
Article history: Received 19 May 2014 Received in revised form 18 June 2014 Accepted 19 June 2014 Available online 26 June 2014 Keywords: Cellulose hydrolysis Kinetic modeling Activation energy Temperature Substrate surface area
a b s t r a c t The rate of enzymatic hydrolysis of cellulose reaction is known to decrease signiﬁcantly as the reaction proceeds. Factors such as reaction temperature, time, and surface area of substrate that affect cellulose conversion were analyzed relative to their role in a mechanistic model based on ﬁrst order inactivation of adsorbed cellulases. The activation energies for the hydrolytic step and inactivation step were very close in magnitude: 16.3 kcal mol1 for hydrolysis and 18.0 kcal mol1 for inactivation, respectively. Therefore, increasing reaction temperature would cause a signiﬁcant increase in the inactivation rate in addition to the catalytic reaction rate. Vmax,app was only 20% or less of the value at 72 h compared to at 2 h as a result of inactivation of adsorbed cellulases, suggesting prolonged hydrolysis is not an efﬁcient way to improve cellulose hydrolysis. Hydrolysis rate increased with corresponding increases in available substrate surface binding area. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Enzymatic conversion of cellulose substrate is slow and presents one of the key bottlenecks that hamper the industrial development of ethanol from biomass. There have been numerous studies that analyzed factors affecting the hydrolysis rate. For example, Ferreira et al. (2009) used Response Surface Methodology to optimize reaction conditions, which was based on empirical models. Some other studies have focused on addition of surfactant or protein such as BSA to improve cellulose hydrolysis (Brethauer et al., 2011; Ouyang et al., 2010). These, however, do not provide insight into mechanistic details of the process. It would be more valuable to investigate how the hydrolysis is affected by intrinsic rate limiting steps, such as inactivation of adsorbed enzyme ⇑ Corresponding author. Address: University of Louisville, Department of Chemical Engineering, Ernst Hall 106, Louisville, KY 40292, USA. Tel.: +1 (502) 852 1567. E-mail address: [email protected]
(R.E. Berson). 1 Current address: Energy Biosciences Institute, University of Illinois, Urbana, IL 61801, USA. http://dx.doi.org/10.1016/j.biortech.2014.06.070 0960-8524/Ó 2014 Elsevier Ltd. All rights reserved.
(Ye and Berson, 2011), or slow dissociation of inactive enzyme from substrate (Cruys-Bagger et al., 2012). We previously developed a mechanistic kinetic model that accurately accounts for cellulose hydrolysis when considering ﬁrst order inactivation of adsorbed cellulases (Ye and Berson, 2011). The inactivation as proposed in that model is supported by the demonstration of ‘‘trafﬁc jams’’ of cellobiohydrolase on cellulose strips using AFM imaging (Igarashi et al., 2011). The effects of incubating time, reaction temperature and accessible surface area of substrate on hydrolysis, as they relate to adsorbed cellulase inactivation, are explored here.
2. Methods 2.1. Cellulose substrate and enzyme Cellulose substrates used in these experiments were: Sigmacell Type 20 from Sigma–Aldrich St Louis, MO; microcrystalline cellulose, from Alfa Aesar Ward Hill, MA; and dewaxed cotton from
Z. Ye, R.E. Berson / Bioresource Technology 167 (2014) 582–586
Johnson & Johnson New Brunswick, NJ. All substrates were used to study the effect of surface area on cellulose hydrolysis. Sigmacell was used to study the effect of reaction time and reaction temperature on enzymatic hydrolysis. The substrates were hydrolyzed using Spezyme CP cellulase enzyme from Genencor International, Inc. (Rochester, NY, USA) [lot # 3016295230].
Using the modeling procedure as previously reported (Ye and Berson, 2011), the rate equation of cellulose hydrolysis can be written as: kf dP ðSÞ kr ¼ V r ¼ k2 ðEÞ0 þ exp½ðkf þ kr Þ t kf þ kr kf þ kr dt K m þ ðSÞ
ð3Þ 2.2. Cellulose hydrolysis and glucose measurements Cellulose was hydrolyzed as reported previously (Ye and Berson, 2011), at 150 rpm in 250 mL ﬂasks in an Innova 4230 incubator shaker. The pH of each ﬂask was adjusted to 4.8 with citrate buffer. To prevent bacterial growth, 3 lL/mL of cycloheximide and 4 lL/mL of tetracycline was added to the slurry. The total operating volume of each test was 100 mL. 0.1, 0.2, 0.4, 1.2, and 2 g of cellulose substrate and 0.6 mL of Spezyme CP cellulases (50 FPU (ﬁlter paper unit)/mL cellulases activity) were incubated for up to 3 days at three temperatures 20, 35 and 50 °C. Spezyme CP contained sufﬁcient b-glucosidase activity, which can convert 106 lM cellobiose per min using 1 mg protein (Ximenes et al., 2011). Furthermore, substrate concentration was intentionally kept below levels known to cause product inhibition and oligomer sugar accumulation. 1.5 mL samples were removed to determine the glucose concentration at incubation times of 2, 4, 8, 16, 24, 48, and 72 h. Experiments were performed with four samples and duplicate measurements were recorded for each sample at each time point. Released glucose was assayed as reported previously (Ye and Berson, 2011). 2.3. Accessible surface area and pore size measurements The accessible surface area and pore size of various cellulose substrates were measured by nitrogen gas adsorption and desorption isotherms as reported elsewhere (Choi et al., 2007) using an adsorption apparatus (Micromeritics Instrument Corporation, Tristar 3000). Measurement of accessible surface area by nitrogen gas adsorption has been applied in studies of enzyme adsorption and enzymatic hydrolysis of cellulose by many other researchers (Rahikainen et al., 2011; Lee et al., 2007; Choi et al., 2007). The operating conditions were as follows: sample mass was 0.14– 0.21 g; temperature was 77.300 K; equilibration interval was 5 s. The ranges of speciﬁc surface area, pore size, and total pore volume for various cellulose substrates were determined from nitrogen adsorption and desorption isotherms, respectively, using the BJH model. 3. Kinetic modeling to determine activation energy In order to examine effects of temperature on cellulose hydrolysis, activation energies of both the rate limiting step and enzyme inactivation need to be determined. The rate constants used in the Arrhenius equation,
ln ðkÞ ¼
Ea 1 þ ln ðAÞ R T
were determined from the mechanism in Eq. (2). The mechanism represents cellulose hydrolysis as previously reported (Ye and Berson, 2011), and describes enzyme binding to substrate with association and dissociation rate constants k1 and k1. Some active enzyme–substrate complexes produce product with an apparent hydrolysis rate, k2, while others becomes inactive with an inactivation rate constant kf and reactivation rate constant kr. k1
E þ S $ ESactive ! E þ P k1 kf l kr ESactive
where, Vr is the real hydrolysis rate, k2 is the rate of breakdown of the enzyme–substrate complex, (E)0 is initial enzyme concentration, (S) is substrate concentration, kf is the inactivation rate constant for adsorbed enzyme, kr is the reactivation rate constant, t is reaction time, and Km, which was derived from the Langmuir adsorption model, is deﬁned in our previous model (Ye and Berson, 2011) as:
K d þ ðEÞ0 Amax
Amax is the maximum adsorption sites per unit substrate (g/g). Kd (g/L) is the equilibrium constant of dissociation, which for a simple binding mechanism is given by k1/k1. Parameters in the model were determined using procedures as reported previously (Ye and Berson, 2011) for 50, 35 and 20 °C. We compared this model with other models in a previous manuscript (Ye and Berson, 2011). While there are models that correlate enzyme deactivation to reaction temperatures, there is little in the current literature that incorporates activation energy into deactivation modeling. For example, Newman et al. (2013) developed an enzyme deactivation model to account for hydrolysis rate reduction as functions of incubating time and reaction temperatures. However, their model did not include activation energy associated with any step because it is a semi-mechanistic model, compared to our mechanistic model for describing the hydrolysis rate that is based on inactivation of adsorbed enzyme.
4. Results and discussion 4.1. Calculation of rate constants for determining activation energy The Sigmacell substrate was hydrolyzed at three different temperatures 50, 35 and 20 °C. Rate constants were regressed from the experimental product–time (P–t) curves in Fig. 1 using the procedure reported in our previous modeling (Ye and Berson, 2011). The constants are listed in Table 1. Theoretically predicted (P–t) curves were generated from rate Eq. (3) using the parameters in Table 1, and are compared to the experimental results in Fig. 1. The theoretically predicted results were generally within one standard deviation of experimental results, implying that the theoretical modeling, with the determined parameters, can accurately account for hydrolysis at the three temperatures. Activation energies for hydrolysis (k2) and inactivation (kf) were determined using Arrhenius plots (Fig. 2a and b). Activation energy is equal to (slope R), where R (1.985 cal K1 mol1) is the gas constant. The activation energies for the k2 and kf steps were determined to be 16.3 and 18.0 kcal K1 mol1, respectively. The activation energy for the hydrolysis step (k2) is within the typical range of 4–20 kcal mol1 for enzymatic reactions (Shuler and Kargi, 1992). The activation energy for the inactivation step (kf) is close to that of the hydrolysis step (k2), implying that increasing the reaction temperature may cause a corresponding increase in the inactivation rate. This is not unexpected; if jamming of cellobiohydrolases is the main reason for inactivation, the faster moving rate of cellobiohydrolases will correlate with higher frequency of inactivation.
Z. Ye, R.E. Berson / Bioresource Technology 167 (2014) 582–586 Table 1 Parameters for the cellulose hydrolysis model (Eq. (3)) at three different temperatures. Temperature (°C)
k2 (h1) kf (h1) kr (h1) Km (g/L)
2000 0.38 0.02 6.23
721 0.169 0.0280 9.51
150 0.0231 0.0009 5.81
y = -8200x + 33.05 R² = 0.9956
7.0 6.5 6.0
1/T (1/ C) 0.0 -0.5 y = -9057x + 27.25 R² = 0.9776
-1.0 -1.5 -2.0 -2.5
-3.0 -3.5 -4.0
1/T (1/ C) 3.0
50 C o 35 C o 20 C
2.0 1.5 1.0 0.5 0.0 0
t (h) Fig. 1. Comparison of experimental and predicted glucose released for Sigmacell. (a) 50 °C; (b) 35 °C; (c) 20 °C.
Fig. 2. Arrhenius Plots for determining activation energies for (a) k2 and (b) kf, and (c) Vmax,app as a function of incubating time at three different temperatures for Sigmacell.
Z. Ye, R.E. Berson / Bioresource Technology 167 (2014) 582–586
4.2. Effect of reaction temperature on cellulose hydrolysis Vmax,app, the apparent maximum hydrolysis rate, as a function of time can be simulated using Eq. (5) (Ye and Berson, 2011):
V max;app ¼ k2 ðEÞ0 kf kr þ exp½ðkf þ kr Þ t kf þ kr kf þ kr
From Fig. 2c, Vmax,app was about seven times higher at 2 h and a temperature of 50 °C than that at the same time at 20 °C, and two times higher than at 35 °C. As a result, much more glucose is released at higher reaction temperature in the initial few hours. For example, 8.9, 6.2 and 2.8 g/L of glucose were released after 24-h of incubation at 50, 35 and 20 °C, respectively. However, Vmax,app decreased signiﬁcantly as the reaction proceeded, and the decrease was more signiﬁcant at higher temperatures. Vmax,app approached about the same value of 0.15 (g/(L h)) for all three temperatures by about 24 h, and subsequently leveled off. The increment in glucose released between 24 and 72 h was only 3.1, 3.8 and 2.5 g/L at reaction temperatures of 50, 35 and 20 °C, respectively. Interestingly, the glucose increment was higher at 35 °C than at 50 °C. The modeling and results here can explain previous experimental observations by Eklund et al. (1990) and Börjesson et al. (2007). Eklund et al. (1990) found that more glucose was released between 24 and 96 h at 40 °C compared to that at 50 °C. Börjesson et al. (2007) also found increasing yield when the incubation temperature was lowered from 50 to 40 °C. 4.3. Effect of incubation time on cellulose hydrolysis As can be seen in Fig. 1, hydrolysis was fast initially, but slowed down after about 24 h. Similarly, Vmax,app decreased signiﬁcantly throughout the reaction. The value at 72 h was only about 20% or less as compared to at 2 h for all three temperatures (Fig. 2c). It is therefore not recommended to increase cellulose conversion with prolonged hydrolysis (e.g. for weeks), since productivity became much lower after initial few hours. However, if substrate cost is a signiﬁcant factor for economic processing, it is meaningful to extend hydrolysis time to minimize cost associated with substrate recycle. For example, conversion can increase another 15% between 24-h and 72-h incubating period for hydrolysis of 20 g/L cellulose at 50 °C. 4.4. Effect of surface area of substrate on cellulose hydrolysis Surface area of substrate is an important parameter affecting hydrolysis of cellulose. It affects the maximum adsorption capability of enzyme (Amax), which is inversely proportional to Km as deﬁned in Eq. (4). Particularly, the surface area accessible to a molecule the size of the enzyme (51 A) (Grethlein et al., 1984) plays an important role in the hydrolysis of cellulosic substrate (Grethlein et al., 1984; Stone et al., 1969). The surface area accessible to a molecule greater than the size of an enzyme (51 A) for different cellulose substrates were determined and found to increase in the order of Cotton ﬁber (0.22–0.29 m2/g), Microcrystalline cellulose (0.46–0.52 m2/g) and Sigmacell Type-20 (0.91–1.1 m2/g). At a solid concentration of 1 g/L, substrate conversions from low to high were cotton ﬁber, microcrystalline cellulose, and Sigmacell Type-20 (Fig. 3). Conversions correlate with an increase in the surface area. Hydrolysis of cotton ﬁber showed much lower conversion than the other substrates. Untreated cotton ﬁber has been reported to have low accessible surface area to enzyme (Stone et al., 1969), with which our ﬁnding agrees. The correlation of conversion with accessible surface area has also been demonstrated between other cellulose and lignocellulosic substrates (Sinitsyn
Fig. 3. Comparison of hydrolysis between substrates of varying binding surface area at 1 g/L loading.
et al., 1991), as well as substrates following various pretreatments (Stone et al., 1969; Rollin et al., 2011). Therefore, effective pretreatments to increase the surface binding area of substrate are believed good practice for improvement of the cellulose hydrolysis rate (Zhu et al., 2009; Kim et al., 2011). 5. Conclusions Factors to consider for fast hydrolysis of cellulose were discussed here based on a mechanistic model considering ﬁrst order inactivation of adsorbed cellulases. Increasing reaction temperature increased the inactivation rate as much as it increased the catalytic reaction rate since activation energies for both steps were very close in value, within 10%. Higher reaction temperature is only effective in improving the initial hydrolysis rate, which suggests prolonged hydrolysis time is not an efﬁcient way to improve cellulose conversion due to inactivation of adsorbed enzyme. Increasing the surface binding area of substrate can improve the hydrolysis rate. Acknowledgements We thank Dr. Andrew N. Lane of the James Graham Brown Cancer Center, University of Louisville and Dr. Moises A. Carreon of the Chemical and Biological Engineering Department, Colorado School of Mines, for their helpful discussions and technical advice, and Genencor International, Inc. for providing the Spezyme CP cellulase. This work was funded by the United States Department of Energy, award number DE-FC36-046014221. References Brethauer, S., Studer, M.H., Yang, B., Wyman, C.E., 2011. The effect of bovine serum albumin on batch and continuous enzymatic cellulose hydrolysis mixed by stirring or shaking. Bioresour. Technol. 102, 6295–6298. Börjesson, J., Peterson, R., Tjerneld, F., 2007. Enhanced enzymatic conversion of softwood lignocellulose by poly(ethylene glycol) addition. Enzyme Microb. Technol. 40, 754–762. Choi, I.G., Lee, J.W., Gwak, K.S., Park, J.Y., Park, M.J., Choi, D.H., Kwon, M., 2007. Biological pretreatment of softwood Pinus densiﬂora by three white rot fungi. J. Microbiol. 45, 485–491. Cruys-Bagger, N., Elmerdahl, J., Praestgaard, E., Tatsumi, H., Spodsberg, N., Borch, K., Westh, P., 2012. Pre-steady-state kinetics for hydrolysis of insoluble cellulose by cellobiohydrolase Cel7A. J. Biol. Chem. 287, 18451–18458. Eklund, R., Galbe, M., Zacchi, G., 1990. Optimization of temperature and enzyme concentration in the enzymatic sacchariﬁcation of steam-pretreated willow. Enzyme Microb. Technol. 12, 225–228. Ferreira, S., Duarte, A.P., Ribeiro, M.H.L., Queiroz, J.A., Domingues, F.C., 2009. Response surface optimization of enzymatic hydrolysis of Cistus ladanifer and Cytisus striatus for bioethanol production. Biochem. Eng. J. 45, 192–200. Grethlein, H.E., Allen, D.C., Converse, A.O., 1984. A comparative study of the enzymatic hydrolysis of acid-pretreated white pine and mixed hardwood. Biotechnol. Bioeng. 26, 1498–1505.
Z. Ye, R.E. Berson / Bioresource Technology 167 (2014) 582–586
Igarashi, K., Uchihashi, T., Koivula, A., Wada, M., Kimura, S., Okamoto, T., Penttilä, M., Ando, T., Samejima, M., 2011. Trafﬁc jams reduce hydrolytic efﬁciency of cellulase on cellulose surface. Science 333, 1279–1282. Kim, Y., Mosier, N.S., Ladisch, M.R., Ramesh Pallapolu, V., Lee, Y.Y., Garlock, R., Warner, R.E., 2011. Comparative study on enzymatic digestibility of switchgrass varieties and harvests processed by leading pretreatment technologies. Bioresour. Technol. 102, 11089–11096. Lee, J., Gwak, K., Park, J., Park, M., Choi, D., Kwon, M., Choi, I., 2007. Biological pretreatment of softwood Pinus densiﬂora by three white rot fungi. J. Microbiol. 45, 485–491. Newman, R.H., Vaidya, A.A., Campion, S.H., 2013. A mathematical model for the inhibitory effects of lignin in enzymatic hydrolysis of lignocellulosics. Bioresour. Technol. 130, 757–762. Ouyang, J., Dong, Z., Song, X., Lee, X., Chen, M., Yong, Q., 2010. Improved enzymatic hydrolysis of microcrystalline cellulose (Avicel PH101) by polyethylene glycol addition. Bioresour. Technol. 101, 6685–6691. Rahikainen, J., Mikander, S., Marjamaa, K., Tamminen, T., Lappas, A., Viikari, L., Kruus, K., 2011. Inhibition of enzymatic hydrolysis by residual lignins from softwood – study of enzyme binding and inactivation on lignin-rich surface. Biotechnol. Bioeng. 108, 2823–2834.
Rollin, J.A., Zhu, Z., Sathitsuksanoh, N., Zhang, Y.H.P., 2011. Increasing cellulose accessibility is more important than removing lignin: a comparison of cellulose solvent-based lignocellulose fractionation and soaking in aqueous ammonia. Biotechnol. Bioeng. 108, 22–30. Shuler, M.L., Kargi, F., 1992. Bioprocess Engineering: Basic Concepts. Prentice-Hall, Englewood Cliffs, NJ, USA. Sinitsyn, A.P., Gusakov, A.V., Vlasenko, E.Y., 1991. Effect of structural and physicochemical features of cellulosic substrates on the efﬁciency of enzymatic hydrolysis. Appl. Biochem. Biotechnol. 30, 43–59. Stone, J.E., Scallan, A.M., Donefer, E., Ahlgren, E., 1969. Digestibility as a simple function of a molecule of similar size to a cellulase enzyme. Cellulases and Their Applications. American Chemical Society, Washington DC, pp. 219–241. Ximenes, E., Kim, Y., Mosier, N., Dien, B., Ladisch, M., 2011. Deactivation of cellulases by phenols. Enzyme Microb. Technol. 48, 54–60. Ye, Z., Berson, R.E., 2011. Kinetic modeling of cellulose hydrolysis with ﬁrst order inactivation of adsorbed cellulase. Bioresour. Technol. 102, 11194–11199. Zhu, Z., Sathitsuksanoh, N., Vinzant, T., Schell, D.J., McMillan, J.D., Zhang, Y.H.P., 2009. Comparative study of corn stover pretreated by dilute acid and cellulose solvent-based lignocellulose fractionation: enzymatic hydrolysis, supramolecular structure, and substrate accessibility. Biotechnol. Bioeng. 103, 715–724.