Accepted Manuscript Temperature and pH influence adsorption of cellobiohydrolase onto lignin by changing the protein properties Xianqin Lu, Can Wang, Xuezhi Li, Jian Zhao PII: DOI: Reference:
S0960-8524(17)31453-0 http://dx.doi.org/10.1016/j.biortech.2017.08.139 BITE 18748
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
27 June 2017 21 August 2017 22 August 2017
Please cite this article as: Lu, X., Wang, C., Li, X., Zhao, J., Temperature and pH influence adsorption of cellobiohydrolase onto lignin by changing the protein properties, Bioresource Technology (2017), doi: http:// dx.doi.org/10.1016/j.biortech.2017.08.139
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Temperature and pH influence adsorption of cellobiohydrolase onto lignin by changing the protein properties Xianqin Lu, Can Wang, Xuezhi Li, Jian Zhao* State Key Laboratory of Microbial Technology, Shandong University, Jinan City, Shandong Province, 250100, China
* Corresponding author: Jian Zhao Tel.: +86-531-88364690 E-mail: [email protected]
Abstract Non-productive adsorption of cellulase onto lignin restricted the movement of cellulase and also hindered the cellulase recycling in bioconversion of lignocellulose. In this study, effect of temperature and pH on adsorption and desorption of cellobiohydrolase (CBH) on lignin and its possible mechanism were discussed. It found that pH value and temperature influenced the adsorption and desorption behaviors of CBH on lignin. Different thermodynamic models suggested that the action between lignin and CBH was physical action. More CBH was adsorbed onto lignin, but lower initial adsorption velocity was detected at 50 oC comparing with 4 oC. Elevating pH value could improve desorption of cellulase from lignin. The changes of hydrophobicity and electric potential on protein surface may partially explain the impact of environmental
conditions on the adsorption and desorption behaviors of CBH on lignin, and comparing to electrical interaction, the hydrophobicity may be the dominating factor influencing the behaviors.
Keywords: CBH, Lignin, Adsorption, Desorption, Thermodynamic models, Protein properties
1 Introduction Bioconversion of lignocelluloses to liquid fuels such as bioethanol was a possible and good way to resolve the energy crisis. Production of bioethanol included three main steps: pretreatment, saccharification and fermentation (Tan et al., 2013). Pretreatment was a promising way to overcome natural recalcitrance of lignocellulose, destroying the lignocellulose structure by dissolving hemicellulose and lignin, and breaking the cellulose chain to improve the cellulase access to cellulose and enzymatic digestibility of lignocellulose (Kumar & Wyman, 2009). In the saccharification stage, cellulase, as the main enzymes with a complex enzyme system, was used for converting cellulose to fermentable sugar. The cost of cellulase was one of main factor influencing bioethanol commercial production, in spite many studies were carried out to improve the enzyme activity and optimize the cellulase cocktails (Lee et al., 1995; Akimkulova et al., 2016; Sun et al., 2015). Recycling of enzyme was also reported as a valuable strategy to reduce the cost of cellulase (Lee et al., 1995; Rodrigues et al., 2015). But the existence
of the residue lignin in pretreated lignocellulose made enzymatic hydrolysis yield decreased by physical blocking cellulase access to cellulose and nonproduction adsorbing cellulase onto lignin (Lee et al., 1995). The non-productive adsorption of cellulase onto biomass substrate also hindered the cellulase recycling after enzymatic saccharification. Reducing non-productive adsorption of cellulase to lignin was a main factor influencing enzymatic hydrolysis and enzyme recycling after hydrolysis (Akimkulova et al., 2016; Haven & Jørgensen, 2013). Many studies reported that addition of surfactant and BSA (Bovine serum albumin) in enzymatic hydrolysis system could improve enzymatic hydrolysis yield and increase enzyme concentration after hydrolysis, by reducing the amount of cellulase adsorbed to lignin (Haven & Jørgensen, 2013; Li et al., 2016). Supplying lignosulfonate, metal ions and heaving the pH were also reported to reduce cellulase adsorption to lignin during enzymatic hydrolysis (Akimkulova et al., 2016; Rodrigues et al., 2015; Li et al., 2016). It was reported that hydrophobicity, electrical potential of lignin surface and structural conformability, stability of cellulase synergistically influenced the cellulase-lignin interaction (Zhu et al., 2009). Many researchers reported that structural properties of lignin were a main factor influencing the adsorption of cellulase onto lignin (Nakagame et al., 2010). Lignin adsorption affinity to cellulase could be reduced by increasing lignin surface charge (negative), reducing the hydrophobicity of lignin and making lignin more hydrophilic (Lou et al., 2013; Li et al., 2015). In addition to the physical properties of lignin, it was
also reported by many researches that the chemical groups of lignin could also influence the adsorption of cellulase onto lignin, for example, carboxylic acid, free hydroxyl groups and phenolic hydroxyl groups (Nakagame et al., 2011; Yang & Pan, 2016). Except the characteristics of lignin, cellulase properties also played an important role in the adsorption of cellulase onto lignin. Different components of cellulase, even for same component, were showed different adsorption behavior to lignin (Lu et al., 2016). Researches about the interaction of enzyme-lignin had pointed that hydrophobicity and electrostatic interaction were the main force to drive cellulase to lignin (Kellock et al., 2017; Sammond et al., 2014; Haarmeyer et al., 2016). The hydrophobic clusters on enzymes surfaces was reported to correlate to the amount of protein adsorbed to lignin(Sammond et al., 2014). Haarmeyer et al also reported a correlation between interaction of protein-lignin and net charge on protein by computational redesign of the protein potential (Haarmeyer et al., 2016). Except original properties of different cellulases, reaction conditions during adsorption also affected the adsorption of cellulase onto lignin by influencing properties of cellulase. For example, Rahikainen et al. suggested that non-productive adsorption of celulase onto lignin was highly pH and temperature dependent, and they also indicated that unfolding or surface denaturation of enzyme was caused during adsorption which may be the reason for the changes of adsorption(Rahikainen et al., 2013a; Rahikainen et al., 2013b). It was reported that the surroundings not only affected adsorption of enzyme onto lignin, but also affected protein desorption from lignin (Wang et al., 2012; Huang et al., 2016). To our
knowledge, however, how the reaction conditions influenced the properties of cellulase and the corresponding impact on adsorption of cellulase onto lignin was still unclear. It was important to understanding the change rule of cellulase properties under different reaction conditions and how the changes influenced lignin-cellulase interaction, which was helpful to engineer weak-lignin binding cellulase, optimize enzymatic hydrolysis processes and improve cellulase recycling. Penicillium oxalicum (P. oxalicum) JU-A10-T is a cellulase hyper-production strain screened and obtained by our laboratory (Liu et al., 2013). The strain could produce cellulase with a complex whole enzyme system, which effectively degraded cellulose to glucose. Cellobiohydrolase (CBH) was main enzyme component occupying about 30% of the total protein in the cellulase mixtures from the P. oxalicum JU-A10-T, and was a rate-limiting factor during enzymatic hydrolysis of cellulose (Lu et al., 2016; Guo et al., 2014). Our previous study showed that the CBH had higher adsorption affinity to lignin than endo-beta-1, 4-glucanases (EGs) and β-glucosidases (BGLs) and Xylanase (Lu et al., 2016; Guo et al., 2014). Cellulose binding domain (CBD) played an essential role in CBH adsorption onto lignin, and enzyme activity and adsorption affinity of CBH to lignin were obviously reduced without CBD (Palonen et al., 2004; Palonen et al., 1999; Arslan et al., 2016). Nonpolar amino acids in CBD structure were responsible for cellulase adsorption affinity to lignin, as well as hydrophobic interaction was confirmed as main force between CBH-lignin (Gao et al., 2014). Kathryn et al also reported changing the hydrophobicity and electrical potential
of CBD of CBH could make an influence on the adsorption of CBH to lignin. In this study, CBH was expressed in Penicillium oxalicum and purified by Ni column, and the adsorption behavior of the purified CBH onto lignin was analyzed. Conventional adsorption experiments only could show the amount of protein adsorbed by lignin, but hardly gave other information about adsorption process. Using appropriately thermodynamic models of adsorption, some information about adsorption details could be obtained and analyzed (Sugimoto et al., 2012; Kiran et al., 2006). In this paper, thermodynamic behaviors were investigated using different kinetic and isotherm models at different reaction temperatures. The desorption behaviors of CBH from lignin at different temperatures and pH values were studied to assess the feasibility of enhancement of cellulase recycling. And the possible mechanism about how reaction conditions influenced the adsorbed/desorbed behaviors of CBH was also explored by studying the changes of surface hydrophobicity and charges of CBH at different conditions. 2. Materials and Methods 2.1 Materials Enzymatic residual lignin (EHL) from corn stover pretreated by liquid hot water (LHW) at severity of 4.2 was isolated as described before and was characterized in our pervious study (Lu et al., 2016). CBH I (Cel7A-2) of P. oxalicum 114 was expressed in expression system with low-background host ∆15A as described by Hu et al. (Hu et al., 2015), and purified by His Trap TM FF crude column and column superdex 100 (GE
Healthcare, Sweden) according to method described in literature (Zou et al., 2013). It was analyzed that the molecular weight of the CBH was 74.2 kDa and the isoelectric point value was 5.0 (Wei X., 2011). Cellulase mixture used in hydrolysis was from the liquid fermentation of Penicillium oxalicum JU-A10-T stored in our laboratory (Lu et al., 2016; Guo et al., 2014). 2.2 Adsorption of CBH Adsorption studies were performed at 4℃ and 50℃ respectively because 50℃ was the typical temperature used in enzymatic hydrolysis for the cellulase from P. oxalicum, and the protein could retain original structure in low temperature of 4℃, and all experiments were performed in triple. To analyze the kinetic adsorption progress of CBH to lignin, lignin and protein was added to a final concentration of 5 mg protein / g lignin (pH 4.8 NaAc-HAc buffer) in a 1 mL reaction mixture and incubated in 4℃ and 50℃ for different times (Lu et al., 2016). After incubation, the supernatant was centrifuged (10000 rpm for 10 min at 4℃) and the concentration of protein was measured by Bradford method with BSA (Sigma Aldrich) as the standard (Bradford, 1976). To obtain the isotherm of CBH adsorption to lignin at different temperatures, various concentrations of protein and lignin were mixed in a 1 mL reaction system and respectively incubated at 4℃ and 50℃ for 24 hours (Sugimoto et al., 2012). After adsorption, the supernatant and the residues were separated as described above and the concentration of protein was measured by Bradford method.
2.3 Desorption of CBH Before desorption, CBH was firstly adsorbed by lignin at 50℃ for 24 hours in a 1 mL reaction system with the concentration of 5mg protein / g lignin to reach the adsorption equilibrium. Then the precipitation was collected by centrifugation. Desorption of CBH was performed as following: firstly, precipitation was washed by buffer (pH 4.8 NaAc-HAc buffer) for 3 times to remove the free CBH permeated among lignins and then the washed precipitation was incubated in buffer with different pH values and temperatures for 1 hour. After incubation, the supernatant was separated by centrifugation, and the concentration of protein in the supernatant was measured as described above. 2.4 Enzyme recycling after hydrolysis Hydrolysis of LHW pretreated corn stover was performed in 10 mL reaction system containing 0.05M NaAc-HAc buffer (pH4.8) at 1% (w/v) solids loading with enzyme loading (5FPU/g dry solid substrate) at 50℃ for 72 hours(Lu et al., 2016). After hydrolysis, pH value of the slurry was adjusted to pH 6.5 with NaAc, and to pH 8 with NaOH respectively, then the slurry was incubated at 40℃ for 1 hours to release the protein bound to lignin. After incubation, the supernatant was separated from hydrolysis residuals by centrifugation, and the protein in the supernatant was measured by Bradford methods as described above. The protein amount in the supernatant of slurry after hydrolysis was used as control. 2.5 Zeta potential analysis
To determine the impact of pH on the zeta potential of CBH, protein was dissolved in buffers (0.05M NaAc-HAc) with pH 4.8 and 6.5 respectively and incubated in ice. The impact of temperature on the zeta potential of CBH was conducted by dissolving the protein in buffer of pH 4.8, 0.05M NaAc-HAc, and respectively incubated in 4℃, 40℃ and 50℃ for 2 hours. After incubation, the zeta potential of CBH was measured by Zetasizernano ZS (Malvern instrument) as described by Jachimska et al. (Jachimska et al., 2008). 2.6 Hydrophobicity analysis Hydrophobicity of protein surfaces undergoing structural changes were measured by ANS (1-anilino-8-naphthalene-sulfonate) (Tang & Yang, 2013; Schonbrunn et al., 2000). Firstly, 2 mM ANS was dissolved in specific buffer at 50 ℃ for about 12 hours to promote ANS dissolving. Then ANS solution was filtered by 0.2 um membrane and the filtrate was prepared to ANS binding assay. The protein used in the assay was 0.05 mg/mL protein. The binding assay was performed as followings: To determine the temperature impact on the hydrophobicity of CBH, ANS filtrate and protein respectively incubated in 4℃, 40℃ and 50℃ for 2 hours, then the ANS was added to 2.5 mL protein solution in a 5 mL test tubes. Fluorescence of ANS was excited at wavelength of 375 nm and emission spectra were recorded from 400 to 700 nm by Hitachi F-4500 FL spectrophotometer (Tokyo, Japan). To determine the pH value impact on the hydrophobicity of CBH, ANS and protein
were respectively dissolved in buffers with pH 4.8 and 6.5. Then ANS was bond to protein and analyzed as described above. For data evaluation, the emission spectra of ANS with different treatment were subtracted from the specific ANS/protein spectra giving the fluorescence intensities (Schonbrunn et al., 2000). 3 Results and Discussion: 3.1 Kinetic and isotherm analysis of CBH adsorption to lignin CBH was the main component in cellulase mixture, and it showed high nonproductive adsorption affinity to lignin (Lu et al., 2016). However, the adsorption mechanism and the factors determining the adsorption affinity were not yet fully understood. In this study, kinetics and isotherm of CBH adsorption to lignin at 4℃ and 50℃were investigated and various models were used and the adsorbed parameters were also deduced, which could give some details about temperature-induced changes of adsorption of protein to lignin (Kiran et al., 2006). 3.1.1 Kinetic analysis The dependence of the amount of CBH adsorbed onto lignin on the time was plotted in the Fig 1A. It was observed that there was obvious difference in the adsorption curves between 4 ℃ and 50 ℃.They were respectively fitted to different kinetic models such as First-order model, Second-order model and Elovich model. The respective correlation coefficient, R2, was calculated and presented in Table 1. It was found that the second-order model showed a best correlation with the data obtained
from adsorption experiments. The second- order kinetic model can be expressed as follows: (1) Where Qt was the adsorption amount at time t, Qe was the adsorption amount at equilibrium. K2 was the rate constant. The two-order rate equation can be expressed linearly (Nethaji et al., 2013; Duranoğlu et al., 2012): (2) The value of Qe and k2 were calculated by the slope of the t/Qt to t. Fitting curve of the experimental data to second-order model was showed in Fig 1B and the various parameters obtained by above steps were listed in Table 2. The adsorption amount at equilibrium, Qe, monitored from the model was close to experimental value, which confirmed that the adsorption progress followed the second-order kinetic model. The Qe value at 50 ℃ was obviously higher than that at 4 ℃, which indicated that more CBH protein could be adsorbed onto lignin at 50 ℃ than at 4 ℃ at equilibrium. From the model, the amount of protein adsorption to lignin was increased at high temperature, which was consistent to the reported results (Rahikainen et al., 2013a; Rahikainen et al., 2011). At high temperature, the structure of protein was unfolded which triggered the enzyme denaturation and promoted enzyme irreversibly bind on the lignin surface (Rahikainen et al., 2011). It explained that the inhibitory of lignin on hydrolysis was increased at elevated temperature during
hydrolysis (Rahikainen et al., 2013a). The initial adsorption velocity, H, was also calculated from k2 and Qe by the formulation (3) (Duranoğlu et al., 2012): (3) It was showed that the H value at 50 ℃ was lower than that at 4 ℃, which was also tested by the lower absorption amounts at 50 ℃ relative to 4 ℃ in the initial 5 hours in adsorption in Fig 1A. The result that there was lower initial adsorption velocity at higher temperature was not reported before. Adsorption studies suggested that the conformational of protein structure may be altered during protein binding on the lignin surface. The changing of protein surface was a slower process and was temperature dependent (Rahikainen et al., 2011). The lower initial adsorption velocity but higher total adsorption amount of protein to lignin at a higher temperature (50 ℃) compared to 4 ℃ suggested that CBH structure properties may be slowly altered when it was incubated at high temperature, which was preferable to adsorb to lignin, but influenced the initial adsorption velocity. 3.1.2 Isotherm analysis To further study changes on adsorption behavior of CBH onto lignin along with temperature, isotherm analysis was also performed in this paper. Adsorption experiments of CBH onto lignin was conducted at 4 ℃ and 50 ℃ for 24 hours with different protein contents, and different adsorption models such as Langmuir, Freundlich, Dubinin-Radushikevich (D-R) and Hill model were constructed (Kiran et al.,
2006). Table 3 gave the correlation coefficients (R2) of the models, and it was found that the correlation coefficients of Freundlich and D-R models were relatively higher at 50 ℃, while D-R and Hill model displayed higher R2 value at 4 ℃. However, we found that the maximal adsorption ability (Qm) predicted by Freundlich model at 50 ℃ was not close to experimental value, which suggested that the Freudlich model was not optimal adsorption model at 50 oC. In the earlier reports, Langmuir model was frequently used to analyze the adsorption behavior of cellulase to lignin (Lu et al., 2016) , but in this study, D-R model and Hill model showed relative higher fitting degree and the Qm values predicted by the two models were close to the tested value, which indicated that the two models should be more suitable for simulating the adsorption behavior of CBH protein onto lignin at 50 oC and 4 oC. Comparing to Langmuir model, D-R model was more generally used because it did not assume surface homogeneous. The linear equation of D-R model (Kiran et al., 2006) was shown as follows: Ln Qe=Ln Qm –Kd ·ε2
(4)
Where Qe was the adsorption ability at equilibrium (mg/g), Qm was the maximal adsorption ability (mg/g), Kd was the adsorption constant related to the mean free energy (E) of adsorption, ε was adsorption potential: (5) in which, R was the gas constant 8.314 J mol-1 K-1, T was the temperature (K), Ce was the protein concentration in the supernatant after adsorption.
E was the mean free energy of adsorption per gram of protein when it was moved to the surface of lignin from infinity in the adsorption system, and E can be calculated from Kd: (6) D-R model can be used to distinguish the type of adsorption behaviors as physical or chemical interaction. The value of E (mean free energy) from D-R model between 8 and 16 indicated the adsorption was a chemical process, while the value of E
S0960-8524(17)31453-0 http://dx.doi.org/10.1016/j.biortech.2017.08.139 BITE 18748
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
27 June 2017 21 August 2017 22 August 2017
Please cite this article as: Lu, X., Wang, C., Li, X., Zhao, J., Temperature and pH influence adsorption of cellobiohydrolase onto lignin by changing the protein properties, Bioresource Technology (2017), doi: http:// dx.doi.org/10.1016/j.biortech.2017.08.139
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Temperature and pH influence adsorption of cellobiohydrolase onto lignin by changing the protein properties Xianqin Lu, Can Wang, Xuezhi Li, Jian Zhao* State Key Laboratory of Microbial Technology, Shandong University, Jinan City, Shandong Province, 250100, China
* Corresponding author: Jian Zhao Tel.: +86-531-88364690 E-mail: [email protected]
Abstract Non-productive adsorption of cellulase onto lignin restricted the movement of cellulase and also hindered the cellulase recycling in bioconversion of lignocellulose. In this study, effect of temperature and pH on adsorption and desorption of cellobiohydrolase (CBH) on lignin and its possible mechanism were discussed. It found that pH value and temperature influenced the adsorption and desorption behaviors of CBH on lignin. Different thermodynamic models suggested that the action between lignin and CBH was physical action. More CBH was adsorbed onto lignin, but lower initial adsorption velocity was detected at 50 oC comparing with 4 oC. Elevating pH value could improve desorption of cellulase from lignin. The changes of hydrophobicity and electric potential on protein surface may partially explain the impact of environmental
conditions on the adsorption and desorption behaviors of CBH on lignin, and comparing to electrical interaction, the hydrophobicity may be the dominating factor influencing the behaviors.
Keywords: CBH, Lignin, Adsorption, Desorption, Thermodynamic models, Protein properties
1 Introduction Bioconversion of lignocelluloses to liquid fuels such as bioethanol was a possible and good way to resolve the energy crisis. Production of bioethanol included three main steps: pretreatment, saccharification and fermentation (Tan et al., 2013). Pretreatment was a promising way to overcome natural recalcitrance of lignocellulose, destroying the lignocellulose structure by dissolving hemicellulose and lignin, and breaking the cellulose chain to improve the cellulase access to cellulose and enzymatic digestibility of lignocellulose (Kumar & Wyman, 2009). In the saccharification stage, cellulase, as the main enzymes with a complex enzyme system, was used for converting cellulose to fermentable sugar. The cost of cellulase was one of main factor influencing bioethanol commercial production, in spite many studies were carried out to improve the enzyme activity and optimize the cellulase cocktails (Lee et al., 1995; Akimkulova et al., 2016; Sun et al., 2015). Recycling of enzyme was also reported as a valuable strategy to reduce the cost of cellulase (Lee et al., 1995; Rodrigues et al., 2015). But the existence
of the residue lignin in pretreated lignocellulose made enzymatic hydrolysis yield decreased by physical blocking cellulase access to cellulose and nonproduction adsorbing cellulase onto lignin (Lee et al., 1995). The non-productive adsorption of cellulase onto biomass substrate also hindered the cellulase recycling after enzymatic saccharification. Reducing non-productive adsorption of cellulase to lignin was a main factor influencing enzymatic hydrolysis and enzyme recycling after hydrolysis (Akimkulova et al., 2016; Haven & Jørgensen, 2013). Many studies reported that addition of surfactant and BSA (Bovine serum albumin) in enzymatic hydrolysis system could improve enzymatic hydrolysis yield and increase enzyme concentration after hydrolysis, by reducing the amount of cellulase adsorbed to lignin (Haven & Jørgensen, 2013; Li et al., 2016). Supplying lignosulfonate, metal ions and heaving the pH were also reported to reduce cellulase adsorption to lignin during enzymatic hydrolysis (Akimkulova et al., 2016; Rodrigues et al., 2015; Li et al., 2016). It was reported that hydrophobicity, electrical potential of lignin surface and structural conformability, stability of cellulase synergistically influenced the cellulase-lignin interaction (Zhu et al., 2009). Many researchers reported that structural properties of lignin were a main factor influencing the adsorption of cellulase onto lignin (Nakagame et al., 2010). Lignin adsorption affinity to cellulase could be reduced by increasing lignin surface charge (negative), reducing the hydrophobicity of lignin and making lignin more hydrophilic (Lou et al., 2013; Li et al., 2015). In addition to the physical properties of lignin, it was
also reported by many researches that the chemical groups of lignin could also influence the adsorption of cellulase onto lignin, for example, carboxylic acid, free hydroxyl groups and phenolic hydroxyl groups (Nakagame et al., 2011; Yang & Pan, 2016). Except the characteristics of lignin, cellulase properties also played an important role in the adsorption of cellulase onto lignin. Different components of cellulase, even for same component, were showed different adsorption behavior to lignin (Lu et al., 2016). Researches about the interaction of enzyme-lignin had pointed that hydrophobicity and electrostatic interaction were the main force to drive cellulase to lignin (Kellock et al., 2017; Sammond et al., 2014; Haarmeyer et al., 2016). The hydrophobic clusters on enzymes surfaces was reported to correlate to the amount of protein adsorbed to lignin(Sammond et al., 2014). Haarmeyer et al also reported a correlation between interaction of protein-lignin and net charge on protein by computational redesign of the protein potential (Haarmeyer et al., 2016). Except original properties of different cellulases, reaction conditions during adsorption also affected the adsorption of cellulase onto lignin by influencing properties of cellulase. For example, Rahikainen et al. suggested that non-productive adsorption of celulase onto lignin was highly pH and temperature dependent, and they also indicated that unfolding or surface denaturation of enzyme was caused during adsorption which may be the reason for the changes of adsorption(Rahikainen et al., 2013a; Rahikainen et al., 2013b). It was reported that the surroundings not only affected adsorption of enzyme onto lignin, but also affected protein desorption from lignin (Wang et al., 2012; Huang et al., 2016). To our
knowledge, however, how the reaction conditions influenced the properties of cellulase and the corresponding impact on adsorption of cellulase onto lignin was still unclear. It was important to understanding the change rule of cellulase properties under different reaction conditions and how the changes influenced lignin-cellulase interaction, which was helpful to engineer weak-lignin binding cellulase, optimize enzymatic hydrolysis processes and improve cellulase recycling. Penicillium oxalicum (P. oxalicum) JU-A10-T is a cellulase hyper-production strain screened and obtained by our laboratory (Liu et al., 2013). The strain could produce cellulase with a complex whole enzyme system, which effectively degraded cellulose to glucose. Cellobiohydrolase (CBH) was main enzyme component occupying about 30% of the total protein in the cellulase mixtures from the P. oxalicum JU-A10-T, and was a rate-limiting factor during enzymatic hydrolysis of cellulose (Lu et al., 2016; Guo et al., 2014). Our previous study showed that the CBH had higher adsorption affinity to lignin than endo-beta-1, 4-glucanases (EGs) and β-glucosidases (BGLs) and Xylanase (Lu et al., 2016; Guo et al., 2014). Cellulose binding domain (CBD) played an essential role in CBH adsorption onto lignin, and enzyme activity and adsorption affinity of CBH to lignin were obviously reduced without CBD (Palonen et al., 2004; Palonen et al., 1999; Arslan et al., 2016). Nonpolar amino acids in CBD structure were responsible for cellulase adsorption affinity to lignin, as well as hydrophobic interaction was confirmed as main force between CBH-lignin (Gao et al., 2014). Kathryn et al also reported changing the hydrophobicity and electrical potential
of CBD of CBH could make an influence on the adsorption of CBH to lignin. In this study, CBH was expressed in Penicillium oxalicum and purified by Ni column, and the adsorption behavior of the purified CBH onto lignin was analyzed. Conventional adsorption experiments only could show the amount of protein adsorbed by lignin, but hardly gave other information about adsorption process. Using appropriately thermodynamic models of adsorption, some information about adsorption details could be obtained and analyzed (Sugimoto et al., 2012; Kiran et al., 2006). In this paper, thermodynamic behaviors were investigated using different kinetic and isotherm models at different reaction temperatures. The desorption behaviors of CBH from lignin at different temperatures and pH values were studied to assess the feasibility of enhancement of cellulase recycling. And the possible mechanism about how reaction conditions influenced the adsorbed/desorbed behaviors of CBH was also explored by studying the changes of surface hydrophobicity and charges of CBH at different conditions. 2. Materials and Methods 2.1 Materials Enzymatic residual lignin (EHL) from corn stover pretreated by liquid hot water (LHW) at severity of 4.2 was isolated as described before and was characterized in our pervious study (Lu et al., 2016). CBH I (Cel7A-2) of P. oxalicum 114 was expressed in expression system with low-background host ∆15A as described by Hu et al. (Hu et al., 2015), and purified by His Trap TM FF crude column and column superdex 100 (GE
Healthcare, Sweden) according to method described in literature (Zou et al., 2013). It was analyzed that the molecular weight of the CBH was 74.2 kDa and the isoelectric point value was 5.0 (Wei X., 2011). Cellulase mixture used in hydrolysis was from the liquid fermentation of Penicillium oxalicum JU-A10-T stored in our laboratory (Lu et al., 2016; Guo et al., 2014). 2.2 Adsorption of CBH Adsorption studies were performed at 4℃ and 50℃ respectively because 50℃ was the typical temperature used in enzymatic hydrolysis for the cellulase from P. oxalicum, and the protein could retain original structure in low temperature of 4℃, and all experiments were performed in triple. To analyze the kinetic adsorption progress of CBH to lignin, lignin and protein was added to a final concentration of 5 mg protein / g lignin (pH 4.8 NaAc-HAc buffer) in a 1 mL reaction mixture and incubated in 4℃ and 50℃ for different times (Lu et al., 2016). After incubation, the supernatant was centrifuged (10000 rpm for 10 min at 4℃) and the concentration of protein was measured by Bradford method with BSA (Sigma Aldrich) as the standard (Bradford, 1976). To obtain the isotherm of CBH adsorption to lignin at different temperatures, various concentrations of protein and lignin were mixed in a 1 mL reaction system and respectively incubated at 4℃ and 50℃ for 24 hours (Sugimoto et al., 2012). After adsorption, the supernatant and the residues were separated as described above and the concentration of protein was measured by Bradford method.
2.3 Desorption of CBH Before desorption, CBH was firstly adsorbed by lignin at 50℃ for 24 hours in a 1 mL reaction system with the concentration of 5mg protein / g lignin to reach the adsorption equilibrium. Then the precipitation was collected by centrifugation. Desorption of CBH was performed as following: firstly, precipitation was washed by buffer (pH 4.8 NaAc-HAc buffer) for 3 times to remove the free CBH permeated among lignins and then the washed precipitation was incubated in buffer with different pH values and temperatures for 1 hour. After incubation, the supernatant was separated by centrifugation, and the concentration of protein in the supernatant was measured as described above. 2.4 Enzyme recycling after hydrolysis Hydrolysis of LHW pretreated corn stover was performed in 10 mL reaction system containing 0.05M NaAc-HAc buffer (pH4.8) at 1% (w/v) solids loading with enzyme loading (5FPU/g dry solid substrate) at 50℃ for 72 hours(Lu et al., 2016). After hydrolysis, pH value of the slurry was adjusted to pH 6.5 with NaAc, and to pH 8 with NaOH respectively, then the slurry was incubated at 40℃ for 1 hours to release the protein bound to lignin. After incubation, the supernatant was separated from hydrolysis residuals by centrifugation, and the protein in the supernatant was measured by Bradford methods as described above. The protein amount in the supernatant of slurry after hydrolysis was used as control. 2.5 Zeta potential analysis
To determine the impact of pH on the zeta potential of CBH, protein was dissolved in buffers (0.05M NaAc-HAc) with pH 4.8 and 6.5 respectively and incubated in ice. The impact of temperature on the zeta potential of CBH was conducted by dissolving the protein in buffer of pH 4.8, 0.05M NaAc-HAc, and respectively incubated in 4℃, 40℃ and 50℃ for 2 hours. After incubation, the zeta potential of CBH was measured by Zetasizernano ZS (Malvern instrument) as described by Jachimska et al. (Jachimska et al., 2008). 2.6 Hydrophobicity analysis Hydrophobicity of protein surfaces undergoing structural changes were measured by ANS (1-anilino-8-naphthalene-sulfonate) (Tang & Yang, 2013; Schonbrunn et al., 2000). Firstly, 2 mM ANS was dissolved in specific buffer at 50 ℃ for about 12 hours to promote ANS dissolving. Then ANS solution was filtered by 0.2 um membrane and the filtrate was prepared to ANS binding assay. The protein used in the assay was 0.05 mg/mL protein. The binding assay was performed as followings: To determine the temperature impact on the hydrophobicity of CBH, ANS filtrate and protein respectively incubated in 4℃, 40℃ and 50℃ for 2 hours, then the ANS was added to 2.5 mL protein solution in a 5 mL test tubes. Fluorescence of ANS was excited at wavelength of 375 nm and emission spectra were recorded from 400 to 700 nm by Hitachi F-4500 FL spectrophotometer (Tokyo, Japan). To determine the pH value impact on the hydrophobicity of CBH, ANS and protein
were respectively dissolved in buffers with pH 4.8 and 6.5. Then ANS was bond to protein and analyzed as described above. For data evaluation, the emission spectra of ANS with different treatment were subtracted from the specific ANS/protein spectra giving the fluorescence intensities (Schonbrunn et al., 2000). 3 Results and Discussion: 3.1 Kinetic and isotherm analysis of CBH adsorption to lignin CBH was the main component in cellulase mixture, and it showed high nonproductive adsorption affinity to lignin (Lu et al., 2016). However, the adsorption mechanism and the factors determining the adsorption affinity were not yet fully understood. In this study, kinetics and isotherm of CBH adsorption to lignin at 4℃ and 50℃were investigated and various models were used and the adsorbed parameters were also deduced, which could give some details about temperature-induced changes of adsorption of protein to lignin (Kiran et al., 2006). 3.1.1 Kinetic analysis The dependence of the amount of CBH adsorbed onto lignin on the time was plotted in the Fig 1A. It was observed that there was obvious difference in the adsorption curves between 4 ℃ and 50 ℃.They were respectively fitted to different kinetic models such as First-order model, Second-order model and Elovich model. The respective correlation coefficient, R2, was calculated and presented in Table 1. It was found that the second-order model showed a best correlation with the data obtained
from adsorption experiments. The second- order kinetic model can be expressed as follows: (1) Where Qt was the adsorption amount at time t, Qe was the adsorption amount at equilibrium. K2 was the rate constant. The two-order rate equation can be expressed linearly (Nethaji et al., 2013; Duranoğlu et al., 2012): (2) The value of Qe and k2 were calculated by the slope of the t/Qt to t. Fitting curve of the experimental data to second-order model was showed in Fig 1B and the various parameters obtained by above steps were listed in Table 2. The adsorption amount at equilibrium, Qe, monitored from the model was close to experimental value, which confirmed that the adsorption progress followed the second-order kinetic model. The Qe value at 50 ℃ was obviously higher than that at 4 ℃, which indicated that more CBH protein could be adsorbed onto lignin at 50 ℃ than at 4 ℃ at equilibrium. From the model, the amount of protein adsorption to lignin was increased at high temperature, which was consistent to the reported results (Rahikainen et al., 2013a; Rahikainen et al., 2011). At high temperature, the structure of protein was unfolded which triggered the enzyme denaturation and promoted enzyme irreversibly bind on the lignin surface (Rahikainen et al., 2011). It explained that the inhibitory of lignin on hydrolysis was increased at elevated temperature during
hydrolysis (Rahikainen et al., 2013a). The initial adsorption velocity, H, was also calculated from k2 and Qe by the formulation (3) (Duranoğlu et al., 2012): (3) It was showed that the H value at 50 ℃ was lower than that at 4 ℃, which was also tested by the lower absorption amounts at 50 ℃ relative to 4 ℃ in the initial 5 hours in adsorption in Fig 1A. The result that there was lower initial adsorption velocity at higher temperature was not reported before. Adsorption studies suggested that the conformational of protein structure may be altered during protein binding on the lignin surface. The changing of protein surface was a slower process and was temperature dependent (Rahikainen et al., 2011). The lower initial adsorption velocity but higher total adsorption amount of protein to lignin at a higher temperature (50 ℃) compared to 4 ℃ suggested that CBH structure properties may be slowly altered when it was incubated at high temperature, which was preferable to adsorb to lignin, but influenced the initial adsorption velocity. 3.1.2 Isotherm analysis To further study changes on adsorption behavior of CBH onto lignin along with temperature, isotherm analysis was also performed in this paper. Adsorption experiments of CBH onto lignin was conducted at 4 ℃ and 50 ℃ for 24 hours with different protein contents, and different adsorption models such as Langmuir, Freundlich, Dubinin-Radushikevich (D-R) and Hill model were constructed (Kiran et al.,
2006). Table 3 gave the correlation coefficients (R2) of the models, and it was found that the correlation coefficients of Freundlich and D-R models were relatively higher at 50 ℃, while D-R and Hill model displayed higher R2 value at 4 ℃. However, we found that the maximal adsorption ability (Qm) predicted by Freundlich model at 50 ℃ was not close to experimental value, which suggested that the Freudlich model was not optimal adsorption model at 50 oC. In the earlier reports, Langmuir model was frequently used to analyze the adsorption behavior of cellulase to lignin (Lu et al., 2016) , but in this study, D-R model and Hill model showed relative higher fitting degree and the Qm values predicted by the two models were close to the tested value, which indicated that the two models should be more suitable for simulating the adsorption behavior of CBH protein onto lignin at 50 oC and 4 oC. Comparing to Langmuir model, D-R model was more generally used because it did not assume surface homogeneous. The linear equation of D-R model (Kiran et al., 2006) was shown as follows: Ln Qe=Ln Qm –Kd ·ε2
(4)
Where Qe was the adsorption ability at equilibrium (mg/g), Qm was the maximal adsorption ability (mg/g), Kd was the adsorption constant related to the mean free energy (E) of adsorption, ε was adsorption potential: (5) in which, R was the gas constant 8.314 J mol-1 K-1, T was the temperature (K), Ce was the protein concentration in the supernatant after adsorption.
E was the mean free energy of adsorption per gram of protein when it was moved to the surface of lignin from infinity in the adsorption system, and E can be calculated from Kd: (6) D-R model can be used to distinguish the type of adsorption behaviors as physical or chemical interaction. The value of E (mean free energy) from D-R model between 8 and 16 indicated the adsorption was a chemical process, while the value of E
