Bioresource Technology 182 (2015) 136–143
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Non-ionic surfactants do not consistently improve the enzymatic hydrolysis of pure cellulose Yan Zhou a, Hongmei Chen a, Feng Qi b, Xuebing Zhao a,⇑, Dehua Liu a a b
Institute of Applied Chemistry, Department of Chemical Engineering, Tsinghua University, Beijing 100084, China College of Life Sciences/Engineering Research Center of Industrial Microbiology, Fujian Normal University, Fuzhou 350108, China
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
Improvement of pure cellulose
hydrolysis by non-ionic surfactants is limited. The beneficial action of non-ionic surfactants depends on various conditions. Higher surfactant concentration inhibits enzymatic hydrolysis of pure cellulose. Surfactant–enzyme interaction might inhibit the productive adsorption of cellulases.
a r t i c l e
i n f o
Article history: Received 9 December 2014 Received in revised form 27 January 2015 Accepted 30 January 2015 Available online 7 February 2015 Keywords: Lignocellulose Pure cellulose Non-ionic surfactant Cellulase Enzyme adsorption
a b s t r a c t Non-ionic surfactants have been frequently reported to improve the enzymatic hydrolysis of pretreated lignocellulosic biomass and pure cellulose. However, how the hydrolysis condition, substrate structure and cellulase formulation affect the beneficial action of surfactants has not been well elucidated. In this work, it was found that the enzymatic hydrolysis of pure cellulose was not consistently improved by surfactants. Contrarily, high surfactant concentration, e.g. 5 g/L, which greatly improved the hydrolysis of dilute acid pretreated substrates, actually showed notable inhibition to pure cellulose conversion in the late phase of hydrolysis. Under an optimal hydrolysis condition, the improvement by surfactant was limited, but under harsh conditions surfactant indeed could enhance cellulose conversion. It was proposed that non-ionic surfactants could interact with substrates and cellulases to impact the adsorption behaviors of cellulases. Therefore, the beneficial action of surfactants on pure cellulose hydrolysis is influenced by hydrolysis condition, cellulose structural features and cellulase formulation. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Demand for ethanol as a clean and renewable secondgeneration biofuel has increased in the past few years. More than 40 million tons of lignocellulosic materials, including agricultural and forestry residues, waste paper, and energy crops, are produced every year, but much of them is just thrown away (Sanderson, 2011). Biological conversion of this discarded lignocellulose to ⇑ Corresponding author. Tel./fax: +86 10 62772310. E-mail address: [email protected] (X. Zhao). http://dx.doi.org/10.1016/j.biortech.2015.01.137 0960-8524/Ó 2015 Elsevier Ltd. All rights reserved.
ethanol via enzymatic hydrolysis attracts widespread interest due to the significant environmental and economic advantages (Zhu and Zhuang, 2012). However, the main limitation for this process is the high production cost of releasing sugar from biomass. Previous researchers have found that the addition of surfactants, especially non-ionic surfactants, could significantly enhance the cellulose hydrolysis thus reducing the enzyme loading (Tu and Saddler, 2010). However, inhibitory effects were observed with some amphoteric, anionic and cationic surfactants addition (Eriksson et al., 2002; Hemmatinejad et al., 2002). The usually used non-ionic surfactants are Tween and PEG series with
137
Y. Zhou et al. / Bioresource Technology 182 (2015) 136–143
concentrations in the range of trace to 20 g/L (Cui et al., 2011; Eriksson et al., 2002; Kim et al., 2007; Borjesson et al., 2007; Sipos et al., 2011; Wu and Ju, 1998). The mechanism of the positive effect of non-ionic surfactant on the enzymatic hydrolysis of pretreated lignocellulosic biomass is generally believed to be the prevention of the non-productive adsorption of cellulases onto the lignin fraction, which increases the amount of free enzyme that would be beneficial for the hydrolysis of cellulose substrate (Alkasrawi et al., 2003; Jorgensen et al., 2007; Sipos et al., 2011). However, lignocelluloses are different from pure cellulose in terms of surface hydrophobic properties, the primary driving force to bind cellulase, due to the presence of lignin (Nakagame et al., 2011). The effects of surfactants on the hydrolysis of pure cellulose thus may be different. Some studies have been reported on the effects of non-ionic surfactants on the hydrolysis of pure cellulose, and the results consistently indicate that non-ionic surfactant can improve pure cellulose hydrolysis by different extents (Eriksson et al., 2002; Borjesson et al., 2007; Mizutani et al., 2002; Ouyang et al., 2010; Yang et al., 2011; Kim et al., 2007; Helle et al., 1993) (see Supplemental data Table S1). For example, Kim et al. (2007) found that the enzymatic digestibility of a-cellulose and filter paper increased from 59.4% and 69.2% to 81.3% and 91.2%, respectively, by addition of 0.5% Tween 80 at the enzymatic hydrolysis stage. The conversion of high purity cellulose powders (Sigmacell 100) could be increased from 7% to 43% when 4 g/L Tween 80 was added (Helle et al., 1993). However, the effects of surfactants on the enzymatic hydrolysis of pure cellulose are dependent on not only surfactant type, but also hydrolysis conditions. For example, shaking speed was found to affect the action of surfactant on pure cellulose hydrolysis (Yang et al., 2011). In the experiments, it was found that Tween 20 and 80 at concentration of 5 g/L actually showed inhibitive action in late phase of enzymatic hydrolysis of pure cellulose (filter paper (FP) and micro-crystal cellulose (MCC)). It indicates that non-ionic surfactants do not exclusively increase cellulose hydrolysis but depending on several factors such as substrate structure, surfactant type and concentration and hydrolysis conditions. Similar conclusions were obtained by Wang et al. (2013) and Lou et al. (2014) when lignosulfonate (a type of anionic surfactant) was added to the enzymatic hydrolysis of pure cellulose. The enhancing effect of lignosulfonate varied with its properties, loading and hydrolysis pH. Inhibitive effect on cellulose saccharification was also observed using lignosulfonate with large molecular weight and low degree of sulfonation (Lou et al., 2014). In the present work, several non-ionic surfactants were tested under different operational conditions (such as shaking speed, pH, substrate concentration, cellulose crystallinity etc.) in order to figure out the effects of operational parameters on the improving action of surfactant for enzymatic hydrolysis of pure cellulose.
Triton X-100 and PEG 4000 were purchased from Xilong Chemical Co., Ltd., China, and Triton X-114 from Beijing Biodee Biotechnology Co., Ltd., China. 2.3. Hydrolysis experiments Filter paper was cut into about 0.5 0.5 cm size. Enzymatic hydrolysis was performed in 50 mL flasks at 50 °C, pH 4.8 (in 0.05 M sodium acetate buffer) with total working volume of 10 mL. The experiment was carried out with substrate concentrations of 5% (w/v) and enzyme loading of 5 FPU/g substrate. Flasks without surfactant addition were used as the control. To study the concentration of surfactant, experiments with various concentrations (ranging from critical micelle concentration (CMC) to 5 g/ L) were conducted. All the flasks were placed in an air-bath shaker at a constant shaking speed of 150 rpm. At regular intervals, 100 lL aliquots were withdrawn from each flask and centrifuged at 14,000 rpm for 5 min. The supernatant was diluted and then analyzed for monomer sugars. Experiments were performed in duplicate and results were represented as the mean values. Similarly, to investigate the effects of operational condition, substrate concentrations (5–20%), pH (4.0–6.0), shaking speed (100–200 rpm), and cellulose crystallinity (8–80%) were considered as variables. 2.4. Enzyme adsorption Enzyme adsorption on cellulose matrix was performed at substrate concentration of 5% (w/v) at pH 4.8, 25 °C and 150 rpm. After adsorption for 1 h, the protein concentration of supernatant was measured. 2.5. Tween 80 adsorption Tween 80 adsorption on cellulose matrix was performed at substrate concentration of 5% (w/v) at pH 4.8, 25 °C and 150 rpm. After adsorption for 30 min and 1 h, Tween 80 concentration of supernatant was measured at 235 nm using Microplate spectrophotometer (Multiskan GO, Thermo Scientific, MA, USA). 2.6. Analytical methods Monomer sugars were measured by HPLC (high-performance liquid chromatography, Shimadzu, Kyoto, Japan) using an Aminex HPX-87H column (Bio-Rad, Hercules, CA, USA) with 5 mM H2SO4 as mobile phase at a flow rate of 0.8 mL/min at 65 °C. The cellulose conversion was calculated as follows:
Cellulosic conversion ð%Þ ¼
Glucose produced ðgÞ 0:9 100 Initial cellulose substrate ðgÞ ð1Þ
2. Methods 2.1. Substrates and enzymes Filter paper (Hangzhou Special Paper Industry Co., Ltd., China) and microcrystalline cellulose (Sinopharm Chemical Reagent Co., Ltd., China) were used as pure cellulose substrates. Three commercial cellulase formulations were used in the hydrolysis experiments, which were kindly provided by Novozymes A/S (Denmark, Cellulase I), Erbslöh Geisenheim AG (Germany, Cellulase II) and Shandong University (China, Cellulase III). 2.2. Surfactants Five non-ionic surfactants, Tween 20, Tween 80, Triton X-100, Triton X-114 and PEG 4000 were used. Tween 20 was purchased from Sinopharm Chemical Reagent Co., Ltd. China. Tween 80,
Filter paper activity (FPA) was measured according to Adney and Baker (2008). Measurement method of Exo-b-1,4-glucanase activity with Avicel as substrate was similar to that of FPA (Dashtban et al., 2010). b-Glucosidase activity and endo-b-1,4-glucanase (CMCase) activity were measured according to Ghose (Ghose, 1987) using cellobiose and carboxymethylcellulose (CMCell) as substrates, respectively. Protein content of cellulase was measured by Coomassie Brilliant Blue dye method with BSA as a standard according to Bradford (Bradford, 1976). The crystallinity of cellulose was measured by XRD (X-ray diffraction). The diffracted intensity was measured in a 2h range between 5° and 40°, at a speed of 5 °/min. Crystallinity index (CrI) of samples were calculated by the following equation:
CrI ¼ ðI002 Iam Þ=I002 100%
ð2Þ
138
Y. Zhou et al. / Bioresource Technology 182 (2015) 136–143
where I002 is the maximum intensity of the 0 0 2 peak which represents both crystalline and amorphous material (2h 22.5) and Iam is the lowest height between the 2 0 0 and 1 1 0 peaks (2h 18.7°), which represents amorphous material only (Zhao et al., 2010). 3. Results and discussion 3.1. Effects of surfactant type and concentration Tween and PEG are the most frequently reported non-ionic surfactants to enhance cellulose conversion. These surfactants have been proved to block the non-productive adsorption of cellulase enzymes on lignin matrix thus effectively increasing the enzymatic hydrolysis of pretreated lignocellulosic biomass. In the present work, five non-ionic surfactants (Tween 20, 80, Triton X-100, X114, and PEG 4000) were added to FP and MCC hydrolysis,
respectively, with concentration ranges from critical micelle concentration (CMC, 0.015–0.2 g/L) to 5 g/L. As shown in Fig. 1(a), hydrolysis of FP was somewhat improved by addition of the surfactants at CMC, but no significant improvement was found for MCC. When surfactant concentration increased to 0.5 g/L, no significant improvement was found for Tween and Triton in cellulose conversion of FP, while PEG 4000 still showed some positive effect. However, Triton showed some improvement for MCC, while others had no significant effect on cellulose hydrolysis. When surfactant concentration was increased to 1 g/L, negative effects were observed, especially for Tween 20 and 80. This inhibitive effect became more significant at late phase of hydrolysis. For more clear comparison, the effect of surfactant concentration on enzymatic conversion of cellulose at 120 h was shown in Supplemental data Fig. S1. It is obvious that Tween 20 and 80 showed some inhibition to the hydrolysis of both FP and MCC,
Fig. 1. Time courses of enzymatic conversion of filter paper (A column) and microcrystalline cellulose (B column) with addition of surfactant at (a) CMC (g/L); (b) 0.5 g/L; (c) 1 g/L; (d) 5 g/L. FP and MCC were hydrolyzed at concentration of 5% w/v, pH 4.8, 50 °C, 150 rpm, using 5 FPU/g substrate Cellic CTec2.
139
Y. Zhou et al. / Bioresource Technology 182 (2015) 136–143
especially at high concentrations. FP conversion decreased from 74% to 49% and 54% with addition of 5 g/L Tween 20 and Tween 80, respectively. Triton 100 could enhance cellulose conversion at a low concentration (for example 1 g/L) but inhibited the process at a high concentration (for example 5 g/L). The statistical analysis results demonstrated that the inhibitive effects of Tween 20, 80 and Triton 100 at 5 g/L were significant (P < 0.05). Triton 114 showed a little improvement to the 120 h conversion of FP with 5 g/L addition (P = 0.006), but no effect on MCC hydrolysis (P = 0.76). PEG 4000 seemed no significant influence on cellulose hydrolysis (P = 0.09–0.62). These results demonstrated that non-ionic surfactant actually could not exclusively increase the enzymatic hydrolysis of cellulose. On the contrary, higher surfactant concentration might decrease cellulose conversion. More discussion will be made later. 3.2. Effects of shaking speeds Shaking speed is an important factor influencing the activity of enzymes and interaction between substrate, enzyme and surfactant. Hydrolysis experiments were performed at 100, 150 and 200 rpm with addition of 0.5 g/L Tween 80, Triton 114, and PEG 4000, respectively. It was found that the enzymatic hydrolysis of FP decreased with shaking speed (see Supplemental data Fig. S2). For example, the 120 h enzymatic cellulose conversion decreased from 82.7% to 69.2% when shaking speed increased from 100 rpm to 200 rpm. No improvement was found for FP hydrolysis at 100 rpm when surfactant was added. However, Triton 114 and PEG 4000 showed some significant improvements (P = 0.02, 0.03) at 200 rpm, and corresponding cellulose conversion were 78.0% and 77.4%, respectively. However, for MCC no significant impact (P = 0.07, 0.29) was found when Triton x-114 and Tween 80 were added. PEG showed some inhibition to MCC hydrolysis conversion (P = 0.001). It indicated substrate structure might be an important factor influencing the action of surfactant. 3.3. Effects of pH pH, as another factor during enzymatic hydrolysis, has close relationships with activity of enzymes, surface hydrophilicity of substrate, and cellulose–cellulase interactions. With consideration of the optimal pH range for cellulose hydrolysis (4.0–5.0) and the suitable range for subsequent fermentation (5.5–6.0), experiments were conducted at pHs ranging from 4.0 to 6.0. As found in the experiments (Table 1), the optimum pH for enzymatic hydrolysis of pure cellulose is 4.8–5.0. Triton X-114, Tween 80 and PEG showed no improvement on FP conversion at pH lower than 5.0, but significantly enhanced the 120 h conversion at higher pH, e.g. from 36.9% (control) to 43.8%, 42.8% and 43.6% at pH 6.0, respectively. This was probably because that most cellulase enzymes mixtures have an isoelectric point (pI) around 4.8 and increasing pH beyond the pI (=4.8) would result in repulsion between cellulase and cellulose fibers as the fibers are also negatively charged (Lou et al., 2014). Addition of surfactants may alter the surface charge properties of both cellulose substrates and cellulase enzymes, and thus pH may show some influence on the improving action of non-ionic surfactants. However, no positive effects on the hydrolysis of MCC were observed at pHs ranging from 4.0 to 6.0. 3.4. Effects of substrate concentrations To evaluate the relationship between substrate concentration and surfactant effect, hydrolysis experiments were conducted at 5%, 10%, 15%, 20% (w/v) solid concentration. The results shown in Table 1 indicated that cellulose conversion dramatically decreased with increase of solid concentration. For FP, positive effect was
Table 1 Effects of pH and substrate concentration on the action of several non-ionic surfactants for hydrolysis of FP and MCC. For pH tests, FP and MCC (5% w/v) were hydrolyzed at 50 °C, 150 rpm for 120 h, using 5 FPU/g substrate Cellic CTec2 and 0.5 g/L surfactant; for substrate concentration tests, FP and MCC (5% w/v) were hydrolyzed at 50 °C, 150 rpm for 96 h, using 5 FPU/g substrate Cellic CTec2 and 0.5 g/L surfactant. Cellulose conversion (%) Control
Triton X-114
Tween 80
PEG 4000
40.3 ± 2.1(0.87)* 75.5 ± 2.9(0.40) 74.5 ± 6.3(0.29) 75.0 ± 1.0(0.70) 61.2 ± 1.5(0.33) 43.8 ± 0.4(0.00)
40.0 ± 3.8(0.83) 75.3 ± 9.5(0.73) 70.0 ± 3.7(0.75) 72.6 ± 0.1(0.00) 61.3 ± 1.5(0.29) 42.8 ± 0.1(0.00)
43.8 ± 1.3(0.35) 75.9 ± 3.1(0.39) 76.6 ± 3.1(0.08) 77.2 ± 0.4(0.01) 66.4 ± 0.6(0.01) 43.6 ± 0.3(0.00)
Microcrystal cellulose 4.0 51.2 ± 2.5 49.2 ± 5.6(0.22) 4.5 83.3 ± 0.5 82.7 ± 1.2(0.76) 4.8 82.1 ± 7.6 86.8 ± 9.0(0.34) 5.0 86.1 ± 0.4 83.9 ± 0.8(0.08) 5.5 71.2 ± 0.8 70.0 ± 0.1(0.14) 6.0 49.3 ± 1.5 45.9 ± 0.1(0.08)
47.9 ± 1.1(0.23) 80.3 ± 1.2(0.30) 78.5 ± 1.7(0.37) 78.6 ± 0.2(0.00) 70.9 ± 2.7(0.87) 49.5 ± 1.8(0.90)
56.3 ± 0.7(0.11) 84.2 ± 1.2(0.64) 79.8 ± 1.8(0.56) 89.7 ± 2.0(0.13) 70.8 ± 0.5(0.57) 52.3 ± 2.0(0.23)
Substrate concentration (%) Filter paper 5 66.1 ± 5.7 68.5 ± 5.0(0.48) 10 30.8 ± 1.5 36.9 ± 0.1(0.01) 15 33.4 ± 2.8 36.9 ± 0.1(0.17) 20 30.2 ± 1.9 34.3 ± 2.1(0.07)
66.4 ± 4.4(0.85) 34.1 ± 0.5(0.05) 34.0 ± 0.6(0.78) 32.2 ± 2.5(0.32)
71.1 ± 3.9(0.19) 35.1 ± 1.2(0.03) 35.5 ± 0.1(0.56) 31.0 ± 0.2(0.61)
Microcrystal cellulose 5 77.7 ± 6.4 81.3 ± 9.3(0.47) 10 48.4 ± 4.7 50.8 ± 1.7(0.54) 15 39.6 ± 2.5 43.6 ± 2.6(0.15) 20 31.7 ± 3.7 37.7 ± 0.02(0.10)
75.5 ± 3.8(0.52) 49.5 ± 2.3(0.76) 37.6 ± 5.8(0.56) 37.8 ± 0.4(0.10)
76.3 ± 2.4(0.65) 50.9 ± 1.5(0.52) 43.0 ± 0.2(0.15) 37.4 ± 1.4(0.12)
pH Filter paper 4.0 40.8 ± 3.3 4.5 72.4 ± 3.1 4.8 70.8 ± 5.5 5.0 74.6 ± 0.1 5.5 59.6 ± 0.8 6.0 36.9 ± 0.5
* The data in parentheses are statistical P values towards control experimental data.
found at solid concentration of 10% (w/v), while no improvement was found at other solid concentrations. However, the improvement of cellulose conversion at 10% solid concentration is limited (3.3–6.1%). For MCC, surfactants showed slight positive effects at high solid concentration. Cellulose conversion was enhanced by approximately 6% at 20% (w/v) solid concentration. 3.5. Effects of cellulose crystallinity Crystallinity of cellulose deceases as ball-milling time increases (see Supplemental data Fig. S3 for X-ray diffraction diagrams). Crystallinities of FP samples after ball-milling for 0 h, 4 h, 8 h, 24 h were 73.2%, 48.8%, 32.7% and 25.8%, respectively. Crystallinities of MCC sample decreased rapidly after ball-milling for 4 h (from 77% to 20.6%), and turned to 8.6% after milling for 8 h. Cellulose conversion of pure cellulose increased with decrease of crystallinity; however, surfactant addition at 0.5 g/L showed no any positive effects on cellulose hydrolysis no matter what crystallinities the samples have (see Supplemental data Fig. S4). 3.6. Effects of cellulase formulation Three sources of commercial cellulases were compared for enzymatic hydrolysis of FP and MCC. These cellulases showed much difference on the activities of exo-b-1,4-glucanase, endo-b1,4-glucanase, b-glucosidase, total protein contents and filter paper activity (see Supplemental data Table S2). Therefore, surfactants showed different actions on cellulose hydrolysis by these cellulases (see Supplemental data Fig. S5). For cellulase II (Erbslöh, Germany), Triton X-114 and Tween 80, greatly enhanced the 120 h conversion
140
Y. Zhou et al. / Bioresource Technology 182 (2015) 136–143
of FP. For example, addition of 0.5 g/L above surfactants could increase the conversion from 28.7% to 40.8% (P = 0.05) and 43.1% (P = 0.01), respectively. However, surfactants showed no or even slight negative effects on the hydrolysis of MCC by cellulase II. On the contrast, surfactants accelerated MCC hydrolysis by cellulase III (Shandong University, China), but no influence on FP hydrolysis. As aforementioned results, surfactants showed no positive effects when FP and MCC was hydrolyzed by cellulase I (Novozymes, Denmark). This result demonstrates that the action of surfactant is also dependent on cellulase formulation. 3.7. Effect of non-ionic surfactant on cellulase adsorption and enzyme activity To find out the effect of surfactant on the absorption of cellulases onto substrate, the amount of free protein in supernatant was determined to evaluate the adsorption of enzyme to substrate with and without addition of Tween 80. As shown in Fig. 2(A), the cellulase adsorption onto substrate decreased when Tween 80 was added for all of the cellulase enzymes used. Similarly, Tween 80 concentration in the supernatant was further analyzed to determine the adsorption ratio of Tween 80 on substrates. As shown in Fig. 2(B), most of the surfactants were present in the supernatant, and only a small part (
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Non-ionic surfactants do not consistently improve the enzymatic hydrolysis of pure cellulose Yan Zhou a, Hongmei Chen a, Feng Qi b, Xuebing Zhao a,⇑, Dehua Liu a a b
Institute of Applied Chemistry, Department of Chemical Engineering, Tsinghua University, Beijing 100084, China College of Life Sciences/Engineering Research Center of Industrial Microbiology, Fujian Normal University, Fuzhou 350108, China
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
Improvement of pure cellulose
hydrolysis by non-ionic surfactants is limited. The beneficial action of non-ionic surfactants depends on various conditions. Higher surfactant concentration inhibits enzymatic hydrolysis of pure cellulose. Surfactant–enzyme interaction might inhibit the productive adsorption of cellulases.
a r t i c l e
i n f o
Article history: Received 9 December 2014 Received in revised form 27 January 2015 Accepted 30 January 2015 Available online 7 February 2015 Keywords: Lignocellulose Pure cellulose Non-ionic surfactant Cellulase Enzyme adsorption
a b s t r a c t Non-ionic surfactants have been frequently reported to improve the enzymatic hydrolysis of pretreated lignocellulosic biomass and pure cellulose. However, how the hydrolysis condition, substrate structure and cellulase formulation affect the beneficial action of surfactants has not been well elucidated. In this work, it was found that the enzymatic hydrolysis of pure cellulose was not consistently improved by surfactants. Contrarily, high surfactant concentration, e.g. 5 g/L, which greatly improved the hydrolysis of dilute acid pretreated substrates, actually showed notable inhibition to pure cellulose conversion in the late phase of hydrolysis. Under an optimal hydrolysis condition, the improvement by surfactant was limited, but under harsh conditions surfactant indeed could enhance cellulose conversion. It was proposed that non-ionic surfactants could interact with substrates and cellulases to impact the adsorption behaviors of cellulases. Therefore, the beneficial action of surfactants on pure cellulose hydrolysis is influenced by hydrolysis condition, cellulose structural features and cellulase formulation. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Demand for ethanol as a clean and renewable secondgeneration biofuel has increased in the past few years. More than 40 million tons of lignocellulosic materials, including agricultural and forestry residues, waste paper, and energy crops, are produced every year, but much of them is just thrown away (Sanderson, 2011). Biological conversion of this discarded lignocellulose to ⇑ Corresponding author. Tel./fax: +86 10 62772310. E-mail address: [email protected] (X. Zhao). http://dx.doi.org/10.1016/j.biortech.2015.01.137 0960-8524/Ó 2015 Elsevier Ltd. All rights reserved.
ethanol via enzymatic hydrolysis attracts widespread interest due to the significant environmental and economic advantages (Zhu and Zhuang, 2012). However, the main limitation for this process is the high production cost of releasing sugar from biomass. Previous researchers have found that the addition of surfactants, especially non-ionic surfactants, could significantly enhance the cellulose hydrolysis thus reducing the enzyme loading (Tu and Saddler, 2010). However, inhibitory effects were observed with some amphoteric, anionic and cationic surfactants addition (Eriksson et al., 2002; Hemmatinejad et al., 2002). The usually used non-ionic surfactants are Tween and PEG series with
137
Y. Zhou et al. / Bioresource Technology 182 (2015) 136–143
concentrations in the range of trace to 20 g/L (Cui et al., 2011; Eriksson et al., 2002; Kim et al., 2007; Borjesson et al., 2007; Sipos et al., 2011; Wu and Ju, 1998). The mechanism of the positive effect of non-ionic surfactant on the enzymatic hydrolysis of pretreated lignocellulosic biomass is generally believed to be the prevention of the non-productive adsorption of cellulases onto the lignin fraction, which increases the amount of free enzyme that would be beneficial for the hydrolysis of cellulose substrate (Alkasrawi et al., 2003; Jorgensen et al., 2007; Sipos et al., 2011). However, lignocelluloses are different from pure cellulose in terms of surface hydrophobic properties, the primary driving force to bind cellulase, due to the presence of lignin (Nakagame et al., 2011). The effects of surfactants on the hydrolysis of pure cellulose thus may be different. Some studies have been reported on the effects of non-ionic surfactants on the hydrolysis of pure cellulose, and the results consistently indicate that non-ionic surfactant can improve pure cellulose hydrolysis by different extents (Eriksson et al., 2002; Borjesson et al., 2007; Mizutani et al., 2002; Ouyang et al., 2010; Yang et al., 2011; Kim et al., 2007; Helle et al., 1993) (see Supplemental data Table S1). For example, Kim et al. (2007) found that the enzymatic digestibility of a-cellulose and filter paper increased from 59.4% and 69.2% to 81.3% and 91.2%, respectively, by addition of 0.5% Tween 80 at the enzymatic hydrolysis stage. The conversion of high purity cellulose powders (Sigmacell 100) could be increased from 7% to 43% when 4 g/L Tween 80 was added (Helle et al., 1993). However, the effects of surfactants on the enzymatic hydrolysis of pure cellulose are dependent on not only surfactant type, but also hydrolysis conditions. For example, shaking speed was found to affect the action of surfactant on pure cellulose hydrolysis (Yang et al., 2011). In the experiments, it was found that Tween 20 and 80 at concentration of 5 g/L actually showed inhibitive action in late phase of enzymatic hydrolysis of pure cellulose (filter paper (FP) and micro-crystal cellulose (MCC)). It indicates that non-ionic surfactants do not exclusively increase cellulose hydrolysis but depending on several factors such as substrate structure, surfactant type and concentration and hydrolysis conditions. Similar conclusions were obtained by Wang et al. (2013) and Lou et al. (2014) when lignosulfonate (a type of anionic surfactant) was added to the enzymatic hydrolysis of pure cellulose. The enhancing effect of lignosulfonate varied with its properties, loading and hydrolysis pH. Inhibitive effect on cellulose saccharification was also observed using lignosulfonate with large molecular weight and low degree of sulfonation (Lou et al., 2014). In the present work, several non-ionic surfactants were tested under different operational conditions (such as shaking speed, pH, substrate concentration, cellulose crystallinity etc.) in order to figure out the effects of operational parameters on the improving action of surfactant for enzymatic hydrolysis of pure cellulose.
Triton X-100 and PEG 4000 were purchased from Xilong Chemical Co., Ltd., China, and Triton X-114 from Beijing Biodee Biotechnology Co., Ltd., China. 2.3. Hydrolysis experiments Filter paper was cut into about 0.5 0.5 cm size. Enzymatic hydrolysis was performed in 50 mL flasks at 50 °C, pH 4.8 (in 0.05 M sodium acetate buffer) with total working volume of 10 mL. The experiment was carried out with substrate concentrations of 5% (w/v) and enzyme loading of 5 FPU/g substrate. Flasks without surfactant addition were used as the control. To study the concentration of surfactant, experiments with various concentrations (ranging from critical micelle concentration (CMC) to 5 g/ L) were conducted. All the flasks were placed in an air-bath shaker at a constant shaking speed of 150 rpm. At regular intervals, 100 lL aliquots were withdrawn from each flask and centrifuged at 14,000 rpm for 5 min. The supernatant was diluted and then analyzed for monomer sugars. Experiments were performed in duplicate and results were represented as the mean values. Similarly, to investigate the effects of operational condition, substrate concentrations (5–20%), pH (4.0–6.0), shaking speed (100–200 rpm), and cellulose crystallinity (8–80%) were considered as variables. 2.4. Enzyme adsorption Enzyme adsorption on cellulose matrix was performed at substrate concentration of 5% (w/v) at pH 4.8, 25 °C and 150 rpm. After adsorption for 1 h, the protein concentration of supernatant was measured. 2.5. Tween 80 adsorption Tween 80 adsorption on cellulose matrix was performed at substrate concentration of 5% (w/v) at pH 4.8, 25 °C and 150 rpm. After adsorption for 30 min and 1 h, Tween 80 concentration of supernatant was measured at 235 nm using Microplate spectrophotometer (Multiskan GO, Thermo Scientific, MA, USA). 2.6. Analytical methods Monomer sugars were measured by HPLC (high-performance liquid chromatography, Shimadzu, Kyoto, Japan) using an Aminex HPX-87H column (Bio-Rad, Hercules, CA, USA) with 5 mM H2SO4 as mobile phase at a flow rate of 0.8 mL/min at 65 °C. The cellulose conversion was calculated as follows:
Cellulosic conversion ð%Þ ¼
Glucose produced ðgÞ 0:9 100 Initial cellulose substrate ðgÞ ð1Þ
2. Methods 2.1. Substrates and enzymes Filter paper (Hangzhou Special Paper Industry Co., Ltd., China) and microcrystalline cellulose (Sinopharm Chemical Reagent Co., Ltd., China) were used as pure cellulose substrates. Three commercial cellulase formulations were used in the hydrolysis experiments, which were kindly provided by Novozymes A/S (Denmark, Cellulase I), Erbslöh Geisenheim AG (Germany, Cellulase II) and Shandong University (China, Cellulase III). 2.2. Surfactants Five non-ionic surfactants, Tween 20, Tween 80, Triton X-100, Triton X-114 and PEG 4000 were used. Tween 20 was purchased from Sinopharm Chemical Reagent Co., Ltd. China. Tween 80,
Filter paper activity (FPA) was measured according to Adney and Baker (2008). Measurement method of Exo-b-1,4-glucanase activity with Avicel as substrate was similar to that of FPA (Dashtban et al., 2010). b-Glucosidase activity and endo-b-1,4-glucanase (CMCase) activity were measured according to Ghose (Ghose, 1987) using cellobiose and carboxymethylcellulose (CMCell) as substrates, respectively. Protein content of cellulase was measured by Coomassie Brilliant Blue dye method with BSA as a standard according to Bradford (Bradford, 1976). The crystallinity of cellulose was measured by XRD (X-ray diffraction). The diffracted intensity was measured in a 2h range between 5° and 40°, at a speed of 5 °/min. Crystallinity index (CrI) of samples were calculated by the following equation:
CrI ¼ ðI002 Iam Þ=I002 100%
ð2Þ
138
Y. Zhou et al. / Bioresource Technology 182 (2015) 136–143
where I002 is the maximum intensity of the 0 0 2 peak which represents both crystalline and amorphous material (2h 22.5) and Iam is the lowest height between the 2 0 0 and 1 1 0 peaks (2h 18.7°), which represents amorphous material only (Zhao et al., 2010). 3. Results and discussion 3.1. Effects of surfactant type and concentration Tween and PEG are the most frequently reported non-ionic surfactants to enhance cellulose conversion. These surfactants have been proved to block the non-productive adsorption of cellulase enzymes on lignin matrix thus effectively increasing the enzymatic hydrolysis of pretreated lignocellulosic biomass. In the present work, five non-ionic surfactants (Tween 20, 80, Triton X-100, X114, and PEG 4000) were added to FP and MCC hydrolysis,
respectively, with concentration ranges from critical micelle concentration (CMC, 0.015–0.2 g/L) to 5 g/L. As shown in Fig. 1(a), hydrolysis of FP was somewhat improved by addition of the surfactants at CMC, but no significant improvement was found for MCC. When surfactant concentration increased to 0.5 g/L, no significant improvement was found for Tween and Triton in cellulose conversion of FP, while PEG 4000 still showed some positive effect. However, Triton showed some improvement for MCC, while others had no significant effect on cellulose hydrolysis. When surfactant concentration was increased to 1 g/L, negative effects were observed, especially for Tween 20 and 80. This inhibitive effect became more significant at late phase of hydrolysis. For more clear comparison, the effect of surfactant concentration on enzymatic conversion of cellulose at 120 h was shown in Supplemental data Fig. S1. It is obvious that Tween 20 and 80 showed some inhibition to the hydrolysis of both FP and MCC,
Fig. 1. Time courses of enzymatic conversion of filter paper (A column) and microcrystalline cellulose (B column) with addition of surfactant at (a) CMC (g/L); (b) 0.5 g/L; (c) 1 g/L; (d) 5 g/L. FP and MCC were hydrolyzed at concentration of 5% w/v, pH 4.8, 50 °C, 150 rpm, using 5 FPU/g substrate Cellic CTec2.
139
Y. Zhou et al. / Bioresource Technology 182 (2015) 136–143
especially at high concentrations. FP conversion decreased from 74% to 49% and 54% with addition of 5 g/L Tween 20 and Tween 80, respectively. Triton 100 could enhance cellulose conversion at a low concentration (for example 1 g/L) but inhibited the process at a high concentration (for example 5 g/L). The statistical analysis results demonstrated that the inhibitive effects of Tween 20, 80 and Triton 100 at 5 g/L were significant (P < 0.05). Triton 114 showed a little improvement to the 120 h conversion of FP with 5 g/L addition (P = 0.006), but no effect on MCC hydrolysis (P = 0.76). PEG 4000 seemed no significant influence on cellulose hydrolysis (P = 0.09–0.62). These results demonstrated that non-ionic surfactant actually could not exclusively increase the enzymatic hydrolysis of cellulose. On the contrary, higher surfactant concentration might decrease cellulose conversion. More discussion will be made later. 3.2. Effects of shaking speeds Shaking speed is an important factor influencing the activity of enzymes and interaction between substrate, enzyme and surfactant. Hydrolysis experiments were performed at 100, 150 and 200 rpm with addition of 0.5 g/L Tween 80, Triton 114, and PEG 4000, respectively. It was found that the enzymatic hydrolysis of FP decreased with shaking speed (see Supplemental data Fig. S2). For example, the 120 h enzymatic cellulose conversion decreased from 82.7% to 69.2% when shaking speed increased from 100 rpm to 200 rpm. No improvement was found for FP hydrolysis at 100 rpm when surfactant was added. However, Triton 114 and PEG 4000 showed some significant improvements (P = 0.02, 0.03) at 200 rpm, and corresponding cellulose conversion were 78.0% and 77.4%, respectively. However, for MCC no significant impact (P = 0.07, 0.29) was found when Triton x-114 and Tween 80 were added. PEG showed some inhibition to MCC hydrolysis conversion (P = 0.001). It indicated substrate structure might be an important factor influencing the action of surfactant. 3.3. Effects of pH pH, as another factor during enzymatic hydrolysis, has close relationships with activity of enzymes, surface hydrophilicity of substrate, and cellulose–cellulase interactions. With consideration of the optimal pH range for cellulose hydrolysis (4.0–5.0) and the suitable range for subsequent fermentation (5.5–6.0), experiments were conducted at pHs ranging from 4.0 to 6.0. As found in the experiments (Table 1), the optimum pH for enzymatic hydrolysis of pure cellulose is 4.8–5.0. Triton X-114, Tween 80 and PEG showed no improvement on FP conversion at pH lower than 5.0, but significantly enhanced the 120 h conversion at higher pH, e.g. from 36.9% (control) to 43.8%, 42.8% and 43.6% at pH 6.0, respectively. This was probably because that most cellulase enzymes mixtures have an isoelectric point (pI) around 4.8 and increasing pH beyond the pI (=4.8) would result in repulsion between cellulase and cellulose fibers as the fibers are also negatively charged (Lou et al., 2014). Addition of surfactants may alter the surface charge properties of both cellulose substrates and cellulase enzymes, and thus pH may show some influence on the improving action of non-ionic surfactants. However, no positive effects on the hydrolysis of MCC were observed at pHs ranging from 4.0 to 6.0. 3.4. Effects of substrate concentrations To evaluate the relationship between substrate concentration and surfactant effect, hydrolysis experiments were conducted at 5%, 10%, 15%, 20% (w/v) solid concentration. The results shown in Table 1 indicated that cellulose conversion dramatically decreased with increase of solid concentration. For FP, positive effect was
Table 1 Effects of pH and substrate concentration on the action of several non-ionic surfactants for hydrolysis of FP and MCC. For pH tests, FP and MCC (5% w/v) were hydrolyzed at 50 °C, 150 rpm for 120 h, using 5 FPU/g substrate Cellic CTec2 and 0.5 g/L surfactant; for substrate concentration tests, FP and MCC (5% w/v) were hydrolyzed at 50 °C, 150 rpm for 96 h, using 5 FPU/g substrate Cellic CTec2 and 0.5 g/L surfactant. Cellulose conversion (%) Control
Triton X-114
Tween 80
PEG 4000
40.3 ± 2.1(0.87)* 75.5 ± 2.9(0.40) 74.5 ± 6.3(0.29) 75.0 ± 1.0(0.70) 61.2 ± 1.5(0.33) 43.8 ± 0.4(0.00)
40.0 ± 3.8(0.83) 75.3 ± 9.5(0.73) 70.0 ± 3.7(0.75) 72.6 ± 0.1(0.00) 61.3 ± 1.5(0.29) 42.8 ± 0.1(0.00)
43.8 ± 1.3(0.35) 75.9 ± 3.1(0.39) 76.6 ± 3.1(0.08) 77.2 ± 0.4(0.01) 66.4 ± 0.6(0.01) 43.6 ± 0.3(0.00)
Microcrystal cellulose 4.0 51.2 ± 2.5 49.2 ± 5.6(0.22) 4.5 83.3 ± 0.5 82.7 ± 1.2(0.76) 4.8 82.1 ± 7.6 86.8 ± 9.0(0.34) 5.0 86.1 ± 0.4 83.9 ± 0.8(0.08) 5.5 71.2 ± 0.8 70.0 ± 0.1(0.14) 6.0 49.3 ± 1.5 45.9 ± 0.1(0.08)
47.9 ± 1.1(0.23) 80.3 ± 1.2(0.30) 78.5 ± 1.7(0.37) 78.6 ± 0.2(0.00) 70.9 ± 2.7(0.87) 49.5 ± 1.8(0.90)
56.3 ± 0.7(0.11) 84.2 ± 1.2(0.64) 79.8 ± 1.8(0.56) 89.7 ± 2.0(0.13) 70.8 ± 0.5(0.57) 52.3 ± 2.0(0.23)
Substrate concentration (%) Filter paper 5 66.1 ± 5.7 68.5 ± 5.0(0.48) 10 30.8 ± 1.5 36.9 ± 0.1(0.01) 15 33.4 ± 2.8 36.9 ± 0.1(0.17) 20 30.2 ± 1.9 34.3 ± 2.1(0.07)
66.4 ± 4.4(0.85) 34.1 ± 0.5(0.05) 34.0 ± 0.6(0.78) 32.2 ± 2.5(0.32)
71.1 ± 3.9(0.19) 35.1 ± 1.2(0.03) 35.5 ± 0.1(0.56) 31.0 ± 0.2(0.61)
Microcrystal cellulose 5 77.7 ± 6.4 81.3 ± 9.3(0.47) 10 48.4 ± 4.7 50.8 ± 1.7(0.54) 15 39.6 ± 2.5 43.6 ± 2.6(0.15) 20 31.7 ± 3.7 37.7 ± 0.02(0.10)
75.5 ± 3.8(0.52) 49.5 ± 2.3(0.76) 37.6 ± 5.8(0.56) 37.8 ± 0.4(0.10)
76.3 ± 2.4(0.65) 50.9 ± 1.5(0.52) 43.0 ± 0.2(0.15) 37.4 ± 1.4(0.12)
pH Filter paper 4.0 40.8 ± 3.3 4.5 72.4 ± 3.1 4.8 70.8 ± 5.5 5.0 74.6 ± 0.1 5.5 59.6 ± 0.8 6.0 36.9 ± 0.5
* The data in parentheses are statistical P values towards control experimental data.
found at solid concentration of 10% (w/v), while no improvement was found at other solid concentrations. However, the improvement of cellulose conversion at 10% solid concentration is limited (3.3–6.1%). For MCC, surfactants showed slight positive effects at high solid concentration. Cellulose conversion was enhanced by approximately 6% at 20% (w/v) solid concentration. 3.5. Effects of cellulose crystallinity Crystallinity of cellulose deceases as ball-milling time increases (see Supplemental data Fig. S3 for X-ray diffraction diagrams). Crystallinities of FP samples after ball-milling for 0 h, 4 h, 8 h, 24 h were 73.2%, 48.8%, 32.7% and 25.8%, respectively. Crystallinities of MCC sample decreased rapidly after ball-milling for 4 h (from 77% to 20.6%), and turned to 8.6% after milling for 8 h. Cellulose conversion of pure cellulose increased with decrease of crystallinity; however, surfactant addition at 0.5 g/L showed no any positive effects on cellulose hydrolysis no matter what crystallinities the samples have (see Supplemental data Fig. S4). 3.6. Effects of cellulase formulation Three sources of commercial cellulases were compared for enzymatic hydrolysis of FP and MCC. These cellulases showed much difference on the activities of exo-b-1,4-glucanase, endo-b1,4-glucanase, b-glucosidase, total protein contents and filter paper activity (see Supplemental data Table S2). Therefore, surfactants showed different actions on cellulose hydrolysis by these cellulases (see Supplemental data Fig. S5). For cellulase II (Erbslöh, Germany), Triton X-114 and Tween 80, greatly enhanced the 120 h conversion
140
Y. Zhou et al. / Bioresource Technology 182 (2015) 136–143
of FP. For example, addition of 0.5 g/L above surfactants could increase the conversion from 28.7% to 40.8% (P = 0.05) and 43.1% (P = 0.01), respectively. However, surfactants showed no or even slight negative effects on the hydrolysis of MCC by cellulase II. On the contrast, surfactants accelerated MCC hydrolysis by cellulase III (Shandong University, China), but no influence on FP hydrolysis. As aforementioned results, surfactants showed no positive effects when FP and MCC was hydrolyzed by cellulase I (Novozymes, Denmark). This result demonstrates that the action of surfactant is also dependent on cellulase formulation. 3.7. Effect of non-ionic surfactant on cellulase adsorption and enzyme activity To find out the effect of surfactant on the absorption of cellulases onto substrate, the amount of free protein in supernatant was determined to evaluate the adsorption of enzyme to substrate with and without addition of Tween 80. As shown in Fig. 2(A), the cellulase adsorption onto substrate decreased when Tween 80 was added for all of the cellulase enzymes used. Similarly, Tween 80 concentration in the supernatant was further analyzed to determine the adsorption ratio of Tween 80 on substrates. As shown in Fig. 2(B), most of the surfactants were present in the supernatant, and only a small part (
