Professional Documents
Culture Documents
a r t i c l e i n f o a b s t r a c t
Article history: This study demonstrates a method to prepare an immobilized cellulase by using an electrospun poly-
Received 21 December 2010 acrylonitrile (PAN) nanofibrous membrane as the support. To obtain an immobilized cellulase with
Received in revised form 14 April 2011 high hydrolytic activity, the immobilization conditions including activation time, enzyme concentra-
Accepted 16 April 2011
tion, immobilization time, and temperature were optimized. Under those conditions, the immobilized
cellulase possessed a protein loading of 30 mg/g-support and a specific activity of 3.2 U/mg-protein.
Keywords:
After immobilization, the enzymatic stability of cellulase against pH and thermal stresses was improved.
Immobilization
Fourier transform infrared spectroscopy (FTIR) measurements also revealed that the cellulase was cova-
Cellulase
Microalgae
lently bonded to the supports. The immobilized cellulase was then used to hydrolyze cell wall of
Hydrolysis microalgae for the production of reducing sugars. Analyses using response surface methodology (RSM)
show that the hydrolysis yield was affected by the reaction temperature, pH, and substrate/cellulase mass
ratio, and a hydrolysis yield of 60.86% could be obtained at 47.85 ◦ C, pH 5.82, and a substrate/cellulase
mass ratio of 40 g-substrate/g-cellulase. This result suggests that the proposed scheme for the cellulase
immobilization has great potential for the application to the reducing sugar production.
© 2011 Elsevier Inc. All rights reserved.
0141-0229/$ – see front matter © 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.enzmictec.2011.04.012
T.-C. Hung et al. / Enzyme and Microbial Technology 49 (2011) 30–37 31
PAN solution was prepared by dissolving 8% (w/w) commercial PAN in DMF at 2.6. Optimizing enzymatic hydrolysis using RSM
60 ◦ C. The resulting solution was loaded into a glass syringe equipped with an 18-
gauge stainless steel needle and extruded at 1.5 mL/h. The needle was connected to The enzymatic hydrolysis of microalgal cell wall was carried out with a self-
a DC generator located 20 cm away from the earthed collector. The applied voltage made reactor (Fig. 1), which consisted of a glass container with a porous polymethyl
was 20 kV. The produced fibrous membranes were obtained from the collector. methacrylate (PMMA) holder inside. The diameters of the reactor and holder were 37
The thicknesses of the produced membranes were measured using a micrometer and 30 mm, respectively. The heights of the reactor and holder were 70 and 45 mm,
(TECLOCK Corp., Japan), and the apparent densities (a ) of the membranes were cal- respectively. For the hydrolysis reaction, 1.25 × 4 cm2 immobilized cellulase was
culated by the weight of a given membrane sample size (2 cm × 2 cm). The porosity covered on the holder, and a 15 mL buffer solution containing microalgae substrates
was determined as (1 − a /) × 100 [25]. was added to the reactor. The reaction was carried out under constant agitation using
a magnetic stirrer.
2.3. Cellulase immobilization To optimize the hydrolysis reaction conditions, a three-level and three-factorial
Box-Behnken design (BBD) was applied to investigate the effects of reaction
The nitrile groups of the PAN fibrous membranes were activated by an amid- parameters on the hydrolysis yield. Experiments were carried out under various
ination reaction for the cellulase immobilization [21,25]. The reaction was carried reaction temperatures (40–60 ◦ C), pH (3.6–7.6), and substrate/cellulase mass ratios
out by immersing fibrous membranes in absolute ethanol and bubbling the solution (40–100 g-substrate/g-cellulase). The following quadratic equation was used to
with hydrogen chloride to produce imidoester derivatives. The activation time was develop a predictive model for the yield of the hydrolysis reaction:
ranged from 3 to 10 min. After the activation process, the fibrous membranes were Y = ˇ0 + ˇ1 A + ˇ2 B + ˇ3 C + ˇ4 A × B + ˇ5 A × C + ˇ6 B × C
removed from the solution and washed with distilled water. (1)
+ ˇ7 A2 + ˇ8 B2 + ˇ9 C 2
The activated membrane was placed in a cellulase solution buffered at pH 4.5 by
50 mM acetic acid solution. The mixture was shaken at 100 rpm. The concentration where Y is the hydrolysis yield, ˇ0 to ˇ9 are model parameters, A is the reaction
of cellulase solution, immobilization time, and temperature were ranged from 2 temperature, B is pH, and C is substrate/cellulase mass ratio. This empirical model
to 5% (w/w), 30 to 120 min, and 30 to 60 ◦ C, respectively. The cellulase immobilized can be used to predict the optimal reaction conditions. Microsoft Office Excel 2003
32 T.-C. Hung et al. / Enzyme and Microbial Technology 49 (2011) 30–37
Fig. 2. SEM micrographs of (a) raw, (b) activated, and (c) cellulase immobilized nanofibrous membranes. Magnification: 50,000×.
was used to perform regression analysis, model development, analysis of variance The surface structures of a raw, an activated, and a cellulase
(ANOVA) calculations, and the optimal level of the variable input parameters. The immobilized nanofibrous membranes were examined by SEM, as
hydrolysis yield is defined as follows:
shown in Fig. 2. The diameter and the morphology of the activated
Concentration of reducing sugars membranes (Fig. 2b) were very similar to those of the raw mem-
Hydrolysis yield (%) = (2)
Concentration of total sugar branes (Fig. 2a), which indicates that the amidination reaction did
Reducing sugars and total sugar of the microalgae were quantified by the DNS not significantly alter the structure of the membranes. Film-like
method [31] and the phenol–sulfuric method [32], respectively. substances appeared on the membranes after the immobilization
of cellulase (Fig. 2c). Protein assay verified that those substances
3. Results and discussion were the cellulase molecules immobilized on the support.
The nitrile groups of the PAN fibrous membranes were modified
3.1. Characterization of nanofibrous membranes to form imidoester derivatives [21,25]. The produced imidoester
could react specifically with the epsilon aminolysil and alpha-
The physical properties of the electrospun nanofibrous mem- amino residues of enzymes [34]. Thus, the enzymes were covalently
branes depend on the kinds of polymers, spinning voltage, flow bonded to the PAN fibrous membranes (Fig. 3). To verify this mecha-
rate, and distance between needle tip and collector [33]. Previous nism, FTIR spectra of both PAN and PAN/cellulase membranes were
research [21] shows that the optimal production conditions of spin- obtained to determine how the cellulase was immobilized on the
ning PAN nanofibrous membranes are 8% (w/w) of the PAN/DMF PAN nanofibrous membranes. The results are shown in Fig. 4. PAN
ratio, 20 kV of the spinning voltage, 1.5 mL/h of the flow rate, and exhibited prominent peaks, including a stretching vibration band
20 cm of the distance between the needle tip and collector. Under of methylene (–CH2 –) at 2930 cm−1 , a stretching vibration band
these conditions, a PAN nanofibrous membrane with a uniform of nitriles (–C N–) at 2240 cm−1 , and a bending vibration band
diameter (150–300 nm) and a porosity of 88 ± 7% could be obtained. of methylene (–CH2 –) at 1470 cm−1 . The characteristic spectra of
PAN are similar to a previous research using PAN as a supported
matrix for bromelain adsorption [35]. After the cellulase was immo-
bilized, the PAN/cellulase membrane exhibited a wider spectrum
at 3372 cm−1 due to the combination of the stretching vibration
of O–H and N–H [36]. The presence of the O–H spectrum con-
firmed the presence of cellulase. Furthermore, the present of the
bonds of C N (1652 cm−1 ) and C–N (1037 cm−1 ) indicate that the
amidination reaction and the immobilization of cellulase were both
100
(a)
95
Transmittance (%)
(b)
2240
1652
90
2930
1470
3372
85
1037
80
1000 1500 2000 2500 3000 3500 4000
-1
Wavenumber (cm )
Fig. 3. Schematic illustration of cellulase immobilization onto electrospun PAN Fig. 4. FTIR spectra of the (a) PAN nanofibrous membrane and (b) cellulase immo-
nanofibrous membrane by amidination reaction. bilized nanofibrous membrane.
T.-C. Hung et al. / Enzyme and Microbial Technology 49 (2011) 30–37 33
200 2.0
140
80
1.0
100 60
40 0.5
50 20
0 0.0
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
Enzyme concentration (wt%)
0
0.0 2.5 5.0 7.5 10.0
Fig. 6. The effect of cellulase concentration on the protein loading amount and spe-
Activation time (min) cific activity. Activation time: 7.5 min; immobilization time: 1.5 h; immobilization
temperature: 60 ◦ C.
Fig. 5. The effect of activation time on the protein loading amount. Enzyme concen-
tration: 10 wt%; immobilization time: 1.5 h; immobilization temperature: 60 ◦ C.
protein loadings of the immobilized cellulase were not changed
happened on the PAN membranes [21,37]. These results confirm significantly under various cellulase concentrations, a 2 wt% of
that the cellulase was immobilized on the PAN membranes by cova- cellulase concentration was chosen to be the optimal cellulase con-
lent bonding. centration for the immobilization process in order to minimize the
production cost.
3.2. Optimal conditions for immobilizing cellulase An appropriate immobilization time might increase the enzyme
loading on the supports [21]; therefore, various immobilization
The amount of immobilized cellulase is correlated closely to the time periods ranged from 30 to 120 min were tested. Fig. 7 shows
available active sites on the PAN membranes. The activation time that the protein loading increased slightly with the increasing of
for amidination reaction is a major parameter affecting the number immobilization time, but the specific activity decreased dramat-
of available active sites [21]. Therefore, the activation time was var- ically and then leveled off. Increasing the immobilization time
ied from 3 to 10 min to evaluate its effect on the amount of cellulase increased the chance for the suspended cellulase in the solution to
loading on the PAN membranes at the given immobilization time interact with the activated support. However, the decreasing of the
of 1.5 h, temperature of 60 ◦ C, pH of 4.6, and enzyme concentra- specific activity implies that the binding protein denatured during
tion of 10 wt%. Fig. 5 shows the relationship between the activation the immobilization process. Therefore, the optimal immobilization
time and the amount of protein loaded on the supports. The pro- time was determined based on the consideration combining both
tein loading increased with the increasing of activation time from the protein loading and the specific activity of the immobilized
2 to 7.5 min. When the activation time was longer (10 min), the cellulase; this immobilization time was determined to be 30 min.
protein loading increased slightly from 175 to 180 mg-protein/g- A high temperature might be able to provide much energy to
support. This result indicates that the activation process reached the system and speed up the immobilization process. However,
equilibrium quickly. Since a prolonged activation time might cause the high temperature might also denature the structure of pro-
some damage to the membrane structure [21], and the protein load- tein and reduce the enzymatic activity of the cellulase. Therefore,
ing amount was almost saturated at an activation time of 7.5 min, various temperatures, ranging from 30 to 60 ◦ C, were examined to
further experiments were conducted under this condition to deter- reveal the effect of temperature on the cellulase immobilization.
mine other operating parameters. Fig. 8 shows that the protein loading increased with the increas-
After most all the active sites on the PAN membranes were acti-
vated, the other immobilization parameters were then optimized.
140 2.5
The parameters include the cellulase concentration, immobiliza-
Protein loading (mg-protein/g-support)
120 4.0
100
Protein loading (mg-protein / g-support)
100 3.5
60 2.5 60
40 2.0
40
20 1.5
20
0 1.0
25 30 35 40 45 50 55 60 65
Immobilization temperature (ºC) 0
20 30 40 50 60 70 80
Fig. 8. The effect of immobilization temperature on the protein loading amount and Temperature (ºC)
relative specific activity. Activation time: 7.5 min; enzyme concentration: 2 wt%;
immobilization time: 30 min. Fig. 10. The relationship between temperature and the enzymatic stability of free
(circle) and immobilized cellulases (square).
20 Table 1
Coded values of the variables for the Box-Behnken design.
Table 2 Table 4
Experimental matrix for the three independent variables on the hydrolysis yield in Significance of the coefficients in the empirical model.
coded values and experimental results.
Model term Parameter Standard t Valuea P value
Run Variables Response estimate error
55
Fig. 11 shows a response surface curve by plotting the hydrol-
Hydrolysis yield
45 5.0
Source DFb SSb MSb F value Probability (P) > F Tem 50 4.5
pera 4.0
ture 55
Modela 9 2580.64 286.744 43.72 0.0003 (ºC)
60
Residual (error) 5 32.79 6.56
Total 14 2613.43
Fig. 11. The response surface plot of combined effects of the reaction temperature
a
Coefficient of determination (R2 ) = 0.98; adjusted R2 = 0.96. and pH on the hydrolysis yield with a given substrate/cellulase mass ratio of 40 g-
b
DF, degree of freedom; SS, sum of squares; MS, mean square. substrate/g-cellulase.
36 T.-C. Hung et al. / Enzyme and Microbial Technology 49 (2011) 30–37
70 [6] Oberoi HS, Vadlani PV, Brijwani K, Bhargav VK, Patil RT. Enhanced ethanol
production via fermentation of rice straw with hydrolysate-adapted Candida
tropicalis ATCC 13803. Process Biochem 2010;45:1299–306.
60 [7] Qi BK, Chen XR, Shen F, Su Y, Wan YH. Optimization of enzymatic hydrol-
ysis of wheat straw pretreated by alkaline peroxide using response surface
Hydrolysis yield (%)
[39] Yigitoglu M, Temocin Z. Immobilization of Candida rugosa lipase on [41] Mateo C, Palomo JM, Fernandez-Lorente G, Guisan JM, Fernandez-Lafuente R.
glutaraldehyde-activated polyester fiber and its application for hydrolysis of Improvement of enzyme activity, stability and selectivity via immobilization
some vegetable oils. J Mol Catal B Enzym 2010;66:130–5. techniques. Enzyme Microb Technol 2007;40:1451–63.
[40] Cheng ZF, Wu CM, Yang WH, Xu TW. Preparation of bromomethylated poly(2,6- [42] Jeya M, Zhang YW, Kim IW, Lee JK. Enhanced saccharification of alkali-treated
dimethyl-1,4-phenylene oxide) hollow fiber cation-exchange membranes and rice straw by cellulase from Trametes hirsuta and statistical optimization of
immobilization of cellulase thereon. J Membr Sci 2010;358:93–100. hydrolysis conditions by RSM. Bioresour Technol 2009;100:5155–61.