Enzyme and Microbial Technology 49 (2011) 30–37

Contents lists available at ScienceDirect

Enzyme and Microbial Technology

journal homepage: www.elsevier.com/locate/emt

Immobilization of cellulase onto electrospun polyacrylonitrile (PAN) nanofibrous

membranes and its application to the reducing sugar production from microalgae
Tien-Chieh Hung a , Chun-Chong Fu b , Chia-Hung Su c,∗ , Jing-Yi Chen d , Wen-Teng Wu d , Yu-Sheng Lin c
a
Department of Biological and Agricultural Engineering, University of California, One Shields Av., Davis, CA 95616, USA
b
Intellectual Property Office, Ministry of Economic Affairs, No. 185, Hsinhai Road, Sec. 2, Taipei 10637, Taiwan
c
Graduate School of Biochemical Engineering, Ming-Chi University of Technology, No. 84 Gung-Juan Road, Taipei 24301, Taiwan
d
Department of Chemical Engineering, National Cheng Kung University, No. 1, Ta-Hsueh Road, Tainan 701, Taiwan

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.

1. Introduction ical homogenizing pretreatment [7]. Enzymatic hydrolysis using

Glucose is widely used as a carbon source in microbial fermen-
tations since it can be easily assimilated by many microbes [1–3].
However, the production cost of refined glucose is high and limits
its applications to bioenergy production [4,5]. Fermentable sugars
derived from lignocellulosic biomass are alternative carbon sources
which can be used for microbial fermentations. Many works have
been done and shown that microbes are capable to convert those
sustainable sugars into bioenergy [6]. Lignocellulosic biomass can
be obtained from agricultural wastes like wheat straw and sugar
cane bagasse, thus reducing the production cost of bioenergy [7,8].
However, the amount of bioenergy produced based on these alter-
native sugars cannot meet the worldwide demand [9,10]. Previous
research [11] shows that microalgae might be another promising
sugar source due to its rapid growth rate compared to other plants.
The cell wall of microalgae, which consists primarily of cellu-
lose [12,13], can be hydrolyzed to form reducing sugars. Traditional
hydrolysis methods usually involve a chemical catalytic process
at critical temperature and pressure conditions after a mechan-
∗ Corresponding author. Tel.: +886 2 29089899x4665; fax: +886 2 29083072.
E-mail address: chsu@mail.mcut.edu.tw (C.-H. Su).
cellulase, however, does not require such critical conditions,
thus decreasing the energy demand during the hydrolysis pro-
cess [14–16]. The cellulase complex consists of endoglucanase
(E.C. 3.2.1.4), cellobiohydrolase (EC 3.2.1.91), and cellobiase (beta-
glucosidase, EC 3.2.1.21) [5]. During the hydrolysis process, the
endoglucanase and cellobiohydrolase first degrade the crystal cel-
lulose into soluble cellodextrins and amorphous cellulose. These
intermediates are then hydrolyzed into the reducing sugars by
the cellobiase [9]. Due to the consideration of the accessibility of
enzyme to the cellulosic substrates, free cellulase is commonly used
in this process. However, the cost of using free biocatalysts is a
major concern for their industrial applications. Immobilization of
the cellulase complex could be an effective solution to improve the
stability and reusability of the biocatalyst and to reduce the cost.
Many studies have been conducted to immobilize cellulase in
an attempt to enhance the enzyme stability and reusability. Var-
ious materials, such as chitin [17], chitosan [18], nylon [17], and
polyvinyl alcohol (PVA) [19] have been used as supports of the
immobilized cellulase. In addition, in order to reduce the diffusional
limitation of substrates to the immobilized enzyme, materials with
nanoscale have been widely used recently. Among all the meth-
ods used for the production of nanoscale supports, electrospinning
is one of the methods that have been successfully developed. The

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
electrospun nanofibrous PVA membranes have been developed for
the cellulase immobilization [19]. However, the high water dis-
solution and poor mechanical strength of the nanofibrous PVA
limit its applications [20]. To overcome the aforementioned prob-
lems, polyacrylonitrile (PAN), a synthetic polymer, can be used as
an alternative material for the fabrication of electrospun nanofi-
brous membranes, because of its water proof property and strong
mechanical strength [21]. The electrospun fibrous PAN has also
been widely used as high performance filters, reinforcements of
dental materials, scaffolds used in tissue engineering, and sup-
ports of chemical catalysts [22,23]. To immobilize enzymes on
this nanofibrous membrane, additional functional groups, such as
reactive carboxyl groups, are introduced into the polymer back-
bone of PAN, followed by covalently binding enzyme molecules
[24]. However, tedious and complicated steps are required for
this immobilization process, thus increasing the production cost
of immobilized enzyme. Direct activation of nitrile groups of the
electrospun nanofibrous PAN by amidination reaction is a desirable
method for directly conjugating enzyme molecules onto the sup-
port surface. The activated PAN nanofibrous membranes have been
used as an alternative support for lipase immobilization [21,25],
but there is very limited study focusing on their applications to the
cellulase immobilization.
This study uses the electrospun PAN nanofibrous membrane as a
support for the cellulase immobilization. Immobilization parame-
ters such as activation time, enzyme concentration, immobilization
time, and temperature have been optimized to achieve a high
hydrolytic activity. The immobilized cellulase was further used to
hydrolyze cell wall of microalgae to produce the reducing sugars.
Chlorella sp., a strain of microalgae, is commonly used as a model
strain since its cell wall content is known to be 13.6% (dry weight)
of the whole cell [26], which indicates the potential for the produc-
tion of reducing sugars. The reaction conditions for the hydrolysis
reaction were also optimized using response surface methodology
(RSM).
2. Materials and methods
2.1. Materials
PAN (average molecular weight of 1.5 × 105 Da and density () of
1.18 g/cm3 ) was purchased from Scientific Polymer Products, Inc. (USA). N,N-
dimethylformamide (DMF, 99.8%) was provided by ECHO Chemical Co., Ltd. (Miaoli,
Taiwan). Aspergillus niger cellulase and carboxymethyl cellulose (CMC) were
purchased from Sigma–Aldrich Co. (Missouri, USA). The enzymatic activity of the
cellulase declared by the vendor is at least 0.3 U/g. All other chemicals used in
this study were analytical grade. The microalga, Chlorella sp. was harvested from a
marine environment reported previously [27].
2.2. Preparation of PAN nanofibrous membranes
PAN solution was prepared by dissolving 8% (w/w) commercial PAN in DMF at
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
a DC generator located 20 cm away from the earthed collector. The applied voltage
was 20 kV. The produced fibrous membranes were obtained from the collector.
The thicknesses of the produced membranes were measured using a micrometer
(TECLOCK Corp., Japan), and the apparent densities (a ) of the membranes were cal-
culated by the weight of a given membrane sample size (2 cm × 2 cm). The porosity
was determined as (1 − a /) × 100 [25].
2.3. Cellulase immobilization
The nitrile groups of the PAN fibrous membranes were activated by an amid-
ination reaction for the cellulase immobilization [21,25]. The reaction was carried
out by immersing fibrous membranes in absolute ethanol and bubbling the solution
with hydrogen chloride to produce imidoester derivatives. The activation time was
ranged from 3 to 10 min. After the activation process, the fibrous membranes were
removed from the solution and washed with distilled water.
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
of cellulase solution, immobilization time, and temperature were ranged from 2
to 5% (w/w), 30 to 120 min, and 30 to 60 ◦ C, respectively. The cellulase immobilized
31
Fig. 1. Schematic diagram of a self-made reactor for the hydrolysis reaction.
membrane was then removed from the solution and washed with the buffer solution
several times to remove any unbound enzyme.
The amount of cellulase used for the immobilization reaction and the residual
cellulase (unbound protein) in the supernatant after immobilization were deter-
mined by the Bradford method using bovine serum albumin as a standard [28]. The
amount of cellulase immobilized on the PAN nanofibrous membrane was deter-
mined by subtracting the unbound protein from the protein of the cellulase used for
the immobilization process. The protein loading efficiency was determined by the
ratio of the protein on the membrane to the membrane weight.
2.4. Cellulase characterization
Cellulase activity was determined by measuring the amount of glucose produced
from the hydrolysis of CMC [19]. A CMC solution was prepared by dissolving 2% (w/v)
commercial CMC in a 50 mM acetic acid buffer solution (pH 4.6). The hydrolysis reac-
tion was carried out at 50 ◦ C for 30 min. The glucose produced during the reaction
was measured by a spectrophotometer (␮Quant, BioTek Instruments Inc., USA) at
540 nm with 3,5-dinitrosalicylic acid (DNS) as an indicator [29]. The specific activ-
ity of cellulase (U/mg-protein) was defined as the amount of cellulase required to
produce 1 mg of glucose per minute under the hydrolysis conditions.
Nanofibrous membranes with and without cellulase immobilization were
coated with gold in a vacuum at 13.3 Pa. The morphologies of these samples were
analyzed with a JSM-6700F field scanning electron microscopy (SEM) (JEOL Ltd.,
Japan) at an accelerating voltage of 10 kV. The PAN nanofibrous membranes before
and after cellulase immobilization were characterized with an FT-730 Fourier trans-
form infrared spectroscopy (FTIR) (Horiba Ltd., France).
2.5. Enzymatic stability
The enzymatic stabilities of both free and immobilized cellulase under different
pH and thermal stresses were determined using the methods described by Zhou [30].
Residual activity, which is defined as the percentage of specific activity remained
after stress-treatments, is used to express the enzymatic stability. For the pH tests,
free and immobilized cellulases were pretreated with buffer solutions with different
pH (3.0–7.0) at 25 ◦ C for 3 h. On the other hand, the effect of temperature on enzy-
matic stability was studied by pretreating the cellulases at temperatures ranging
from 30 to 70 ◦ C at pH 5 for 3 h.
2.6. Optimizing enzymatic hydrolysis using RSM
The enzymatic hydrolysis of microalgal cell wall was carried out with a self-
made reactor (Fig. 1), which consisted of a glass container with a porous polymethyl
methacrylate (PMMA) holder inside. The diameters of the reactor and holder were 37
and 30 mm, respectively. The heights of the reactor and holder were 70 and 45 mm,
respectively. For the hydrolysis reaction, 1.25 × 4 cm2 immobilized cellulase was
covered on the holder, and a 15 mL buffer solution containing microalgae substrates
was added to the reactor. The reaction was carried out under constant agitation using
a magnetic stirrer.
To optimize the hydrolysis reaction conditions, a three-level and three-factorial
Box-Behnken design (BBD) was applied to investigate the effects of reaction
parameters on the hydrolysis yield. Experiments were carried out under various
reaction temperatures (40–60 ◦ C), pH (3.6–7.6), and substrate/cellulase mass ratios
(40–100 g-substrate/g-cellulase). The following quadratic equation was used to
develop a predictive model for the yield of the hydrolysis reaction:
Y = ˇ0 + ˇ1 A + ˇ2 B + ˇ3 C + ˇ4 A × B + ˇ5 A × C + ˇ6 B × C
(1)
+ ˇ7 A2 + ˇ8 B2 + ˇ9 C 2
where Y is the hydrolysis yield, ˇ0 to ˇ9 are model parameters, A is the reaction
temperature, B is pH, and C is substrate/cellulase mass ratio. This empirical model
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
(ANOVA) calculations, and the optimal level of the variable input parameters. The
hydrolysis yield is defined as follows:
Concentration of reducing sugars
Hydrolysis yield (%) =
Concentration of total sugar
Reducing sugars and total sugar of the microalgae were quantified by the DNS
method [31] and the phenol–sulfuric method [32], respectively.
3. Results and discussion
3.1. Characterization of nanofibrous membranes
The physical properties of the electrospun nanofibrous mem-
branes depend on the kinds of polymers, spinning voltage, flow
rate, and distance between needle tip and collector [33]. Previous
research [21] shows that the optimal production conditions of spin-
ning PAN nanofibrous membranes are 8% (w/w) of the PAN/DMF
ratio, 20 kV of the spinning voltage, 1.5 mL/h of the flow rate, and
20 cm of the distance between the needle tip and collector. Under
these conditions, a PAN nanofibrous membrane with a uniform
diameter (150–300 nm) and a porosity of 88 ± 7% could be obtained.
Fig. 3. Schematic illustration of cellulase immobilization onto electrospun PAN
nanofibrous membrane by amidination reaction.
The surface structures of a raw, an activated, and a cellulase
immobilized nanofibrous membranes were examined by SEM, as
shown in Fig. 2. The diameter and the morphology of the activated
membranes (Fig. 2b) were very similar to those of the raw mem-
(2)
branes (Fig. 2a), which indicates that the amidination reaction did
not significantly alter the structure of the membranes. Film-like
substances appeared on the membranes after the immobilization
of cellulase (Fig. 2c). Protein assay verified that those substances
were the cellulase molecules immobilized on the support.
The nitrile groups of the PAN fibrous membranes were modified
to form imidoester derivatives [21,25]. The produced imidoester
could react specifically with the epsilon aminolysil and alpha-
amino residues of enzymes [34]. Thus, the enzymes were covalently
bonded to the PAN fibrous membranes (Fig. 3). To verify this mecha-
nism, FTIR spectra of both PAN and PAN/cellulase membranes were
obtained to determine how the cellulase was immobilized on the
PAN nanofibrous membranes. The results are shown in Fig. 4. PAN
exhibited prominent peaks, including a stretching vibration band
of methylene (–CH2 –) at 2930 cm−1 , a stretching vibration band
of nitriles (–C N–) at 2240 cm−1 , and a bending vibration band
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. 4. FTIR spectra of the (a) PAN nanofibrous membrane and (b) cellulase immo-
bilized nanofibrous membrane.
 T.-C. Hung et al. / Enzyme and Microbial Technology 49 (2011) 30–37 33

200 2.0
140

Protein loading (mg-protein / g-support)

Protein loading (mg-protein/g-support)

Specific activity (U/mg-protein)

120
1.5
150
100

80
1.0
100 60

40 0.5

50 20

0
0.0 2.5 5.0 7.5
Activation time (min)
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.
happened on the PAN membranes [21,37]. These results confirm
that the cellulase was immobilized on the PAN membranes by cova-
lent bonding.
3.2. Optimal conditions for immobilizing cellulase
The amount of immobilized cellulase is correlated closely to the
available active sites on the PAN membranes. The activation time
for amidination reaction is a major parameter affecting the number
of available active sites [21]. Therefore, the activation time was var-
ied from 3 to 10 min to evaluate its effect on the amount of cellulase
loading on the PAN membranes at the given immobilization time
of 1.5 h, temperature of 60 ◦ C, pH of 4.6, and enzyme concentra-
tion of 10 wt%. Fig. 5 shows the relationship between the activation
time and the amount of protein loaded on the supports. The pro-
tein loading increased with the increasing of activation time from
2 to 7.5 min. When the activation time was longer (10 min), the
protein loading increased slightly from 175 to 180 mg-protein/g-
support. This result indicates that the activation process reached
equilibrium quickly. Since a prolonged activation time might cause
some damage to the membrane structure [21], and the protein load-
ing amount was almost saturated at an activation time of 7.5 min,
further experiments were conducted under this condition to deter-
mine other operating parameters.
After most all the active sites on the PAN membranes were acti-
vated, the other immobilization parameters were then optimized.
The parameters include the cellulase concentration, immobiliza-
0 0.0
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
Enzyme concentration (wt%)
10.0
Fig. 6. The effect of cellulase concentration on the protein loading amount and spe-
cific activity. Activation time: 7.5 min; immobilization time: 1.5 h; immobilization
temperature: 60 ◦ C.
protein loadings of the immobilized cellulase were not changed
significantly under various cellulase concentrations, a 2 wt% of
cellulase concentration was chosen to be the optimal cellulase con-
centration for the immobilization process in order to minimize the
production cost.
An appropriate immobilization time might increase the enzyme
loading on the supports [21]; therefore, various immobilization
time periods ranged from 30 to 120 min were tested. Fig. 7 shows
that the protein loading increased slightly with the increasing of
immobilization time, but the specific activity decreased dramat-
ically and then leveled off. Increasing the immobilization time
increased the chance for the suspended cellulase in the solution to
interact with the activated support. However, the decreasing of the
specific activity implies that the binding protein denatured during
the immobilization process. Therefore, the optimal immobilization
time was determined based on the consideration combining both
the protein loading and the specific activity of the immobilized
cellulase; this immobilization time was determined to be 30 min.
A high temperature might be able to provide much energy to
the system and speed up the immobilization process. However,
the high temperature might also denature the structure of pro-
tein and reduce the enzymatic activity of the cellulase. Therefore,
various temperatures, ranging from 30 to 60 ◦ C, were examined to
reveal the effect of temperature on the cellulase immobilization.
Fig. 8 shows that the protein loading increased with the increas-
140 2.5
Protein loading (mg-protein/g-support)

tion time, and temperature. Protein loading and specific activity

120
Specific activity (U/mg-protein)

were used to evaluate the efficiency of cellulase immobilization. 2.0

Since the amount of cellulase used is highly related to the 100
production cost, the effect of cellulase concentration on the
immobilization procedure was investigated. Various cellulase con- 1.5
80
centrations (2–5 wt%) were tested. Fig. 6 shows that the amount
of immobilized protein increased with the increasing of cellulase 60
1.0
concentration. This is because the contact frequency between the
cellulase and the active sites on the support increased with the 40
enzyme concentration [38]. The figure also shows that the specific 0.5
activity of the immobilized cellulase did not change significantly 20
(P > 0.05) under different cellulase concentrations. These results
indicate that the cellulase molecules were immobilized on the 0 0.0
0 30 60 90 120
active sites of the PAN membrane rather than stacked onto the
cellulase molecules that already are immobilized. The stacking of Immobilization time (min)
cellulase leads to a decrease of the specific activity of immobilized
Fig. 7. The effect of immobilization time on the protein enzyme loading amount and
cellulase because some active sites exposed to the substrate are specific activity. Activation time: 7.5 min; enzyme concentration: 2 wt%; immobi-
blocked [38]. Based on the fact that the specific activity and the lization temperature: 60 ◦ C.
34 T.-C. Hung et al. / Enzyme and Microbial Technology 49 (2011) 30–37

120 4.0
100
Protein loading (mg-protein / g-support)

100 3.5

Specific activity (U/mg-protein)

80

Residual activity (%)

80 3.0

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).

ing of temperature. However, the specific activity of the cellulase

decreased when the temperature was higher than 40 ◦ C. Protein relatively higher cellulase loading amount. These results indicate
denaturalization might lead to this result [39]. Since the maximal that the effect of immobilization methods (cross-linking or cova-
specific activity of immobilized cellulase reached 3.2 U/mg-protein lent bonding) on the retained activity of immobilized cellulases
at 40 ◦ C, this temperature was considered to be the optimal condi- using supports with nano-scale surface structures may not be as
tion for cellulase immobilization. significant as that on those supports without.
Using cross-linkers to immobilize enzyme may cause the loss
of enzymatic activity due to the enzyme aggregation or the occu- 3.3. Enzymatic stability
pation of catalytic sites responsible for the immobilization [40].
The retained enzyme activity (specific activity of immobilized cel- The results of the enzymatic stability for both free and immo-
lulase/specific activity of free cellulase) can be used to represent bilized cellulases treated with different pH are shown in Fig. 9.
the percentage of enzymatic activity of an enzyme remained after The optimal stability for both cases was obtained at pH 5.0, and
immobilization processes. For the cases using glutaraldehyde as a significant decreasing of stability was found when pH was higher
a cross-linker to immobilize cellulase on different supports such than 5 and lower than 4. This result is similar to those of previous
as chitin, chitosan microspheres, chitosan sponges, and polyvinyl reports using other supports for cellulase immobilization [30,40].
alcohol (PVA) casting films, the retained enzyme activities are The results also show that the immobilized cellulase has a better
generally low (25.6%, 60.1%, 65.6%, and 34.2–80.8%, respectively) stability than the free cellulase.
[17–19]. However, Wu et al. [19] showed that PVA nanofibrous Fig. 10 shows the enzymatic stability of free and immobi-
enzymes could have a high retained enzyme activity (65.5–99%) lized cellulases under different temperatures. The residual activity
due to a much less diffusion resistance between the substrate and of both free and immobilized cellulases remained high until the
catalytic sites. In the present study, the retained enzyme activity is temperature increased to 50 ◦ C and then decreased with the
86.23% (the specific activity of free cellulase is 3.71 U/mg-protein). increasing of temperature. The immobilized cellulase showed a
This high retained enzyme activity agrees with Wu’s observation. better resistance to the thermal stress, which might be due to the
In addition, the observed enzyme stacking phenomenon in this immobilization processes making the enzyme molecule more rigid
study (Fig. 6) is very similar to Wu’s results when they have a to prevent the occurrence of denaturalization [41].

3.4. Application of immobilized cellulase

100
By carrying out the immobilization process under the optimal
immobilization conditions, an immobilized cellulase with a pro-
Residual activity (%)

80 tein loading amount of 30 mg/g-support and a specific activity of

3.2 U/mg-protein could be obtained. The immobilized cellulase was
then employed to hydrolyze the cell wall of microalgae, Chlorella sp.,
60
to produce reducing sugars. The effects of reaction temperature, pH,
and substrate/cellulase mass ratio on the hydrolysis performance
40
were studied using Box-Behnken RSM design with three central
points. Table 1 shows the experimental parameters and their lev-

20 Table 1
Coded values of the variables for the Box-Behnken design.

0 Variables Symbols Variable levels

2 3 4 5 6 7 8
−1 0 1
pH
Temperature (◦ C) A 40 50 60
pH B 3.6 5.6 7.6
Fig. 9. The relationship between pH and the enzymatic stability of free (circle) and
Substrate/cellulase mass ratio C 40 70 100
immobilized cellulases (square).
 T.-C. Hung et al. / Enzyme and Microbial Technology 49 (2011) 30–37 35

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

A B C Y (%) ˇ0 41.88 1.47 28.33 0.000b

ˇ1 −0.98 0.91 −1.08 0.328
1 −1 −1 0 19.32 ˇ2 −0.13 0.91 −0.14 0.892
2 −1 1 0 18.33 ˇ3 −12.92 0.91 −14.27 0.000b
3 1 −1 0 17.99 ˇ4 1.19 1.28 0.93 0.394
4 1 1 0 21.77 ˇ5 3.55 1.28 2.77 0.039b
5 0 −1 −1 42.55 ˇ6 −3.59 1.28 −2.80 0.039b
6 0 −1 1 25.52 ˇ7 −8.25 1.33 −6.19 0.002b
7 0 1 −1 47.82 ˇ8 −16.24 1.33 −10.71 0.000b
8 0 1 1 16.43 ˇ9 5.48 1.33 4.11 0.009b
9 −1 0 −1 58.88
a
10 1 0 −1 46.8 t˛/2,n–p = t0.025,5 = 2.571.
b
11 −1 0 1 24.32 P value less than 0.05 indicates model terms are significant.
12 1 0 1 26.44
13 0 0 0 40.38
14 0 0 0 43.11 [42]. In this plot, a maximal response is observed, and the optimal
15 0 0 0 42.15
reaction conditions can be predicted by the empirical model. The
optimal hydrolysis yield was determined to be 60.86% with a tem-
perature of 47.85 ◦ C, a pH of 5.82, and a substrate/cellulase mass
els. The reaction temperature, pH, and substrate/cellulase mass ratio of 40 g-substrate/g-cellulase. An experiment was carried out
ratio were variable input parameters while the hydrolysis yield was under these conditions for verification. An actual hydrolysis yield
the measured response. The reaction time was fixed at 24 h for all of 63.15% was obtained, indicating the agreement of the empirical
experiments. model. Therefore, this model could be used to predict the hydrol-
Table 2 shows the experimental results of enzymatic hydrol- ysis yield as well as to optimize the reaction conditions for further
ysis by a Box-Behnken RSM design. From the middle runs of the applications.
experiment (runs 13–15), the coefficient of variance was 3.31% Reusability is a major advantage of using immobilized cellulases,
on the repeated measures, indicating the reproducibility of the especially in industrial applications. The immobilized cellulase
experiments. The measured responses were fitted with a quadratic developed in this study has been used to hydrolyze microal-
empirical model, shown as following: gal cell walls under the optimized reaction conditions obtained
above. Fig. 12 shows the reusability of the developed immobi-
Hydrolysis yield %, Y = 41.88 − 12.92C + 3.55A × C − 3.59B lized cellulase. The immobilized cellulase can be reused at least five
times with a gradual reduction on the hydrolytic performance. The
× C − 8.25A2 − 16.24B2 + 5.48C 2 (3)
hydrolysis yield at the fifth cycle was only about 55.12% of its orig-
inal performance. Further studies could be conducted to improve
The statistical significance of the above model was examined the reusability performance of the immobilized cellulase.
by the analysis of variance (ANOVA), as Table 3 shows. The small Besides the enhancing of recycling and enzymatic stability,
value of probability (P) > F indicated that the model was statistically the immobilized cellulase developed in this study also shows its
significant. The goodness-of-fit measure of the model was evalu- specific and promising application potential. Unlike hydrolyzing
ated using the coefficient of determination (R2 ). The high value of liquid-based substrates, cellulose hydrolysis requires agitation to
R2 (R2 = 0.98) indicates a high reliability of the model in predict- reduce the diffusion resistance between the substrates and immo-
ing the hydrolysis yield. Table 4 shows the results of T tests and
corresponded P values along with the variable input parameters.
A smaller P values indicate that the corresponding parameters are
highly significant. The linear terms, A and B, and an interaction term 65

of A × B were irrelevant parameters for this empirical model. This

60
empirical model can be used to plot response surface curves and to
determine the optimal conditions for maximizing responses.
(%)

55
Fig. 11 shows a response surface curve by plotting the hydrol-
Hydrolysis yield

ysis yield against the temperature and pH variables with a given 50

substrate/cellulase mass ratio at its −1 level. The hydrolysis yield
increased with the increasing of temperature. However, a further 45

increase in the reaction temperature reversed this trend. A similar

40
trend in hydrolysis yield appeared after varying pH from 3.6 to 7.6.
The trends along with temperature and pH variables are similar 35
to a previous work using cellulase for the hydrolysis of rice straw
7.5
30 7.0
6.5
Table 3 25 6.0
40 5.5
Analysis of variance (ANOVA) for the empirical model.
pH

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
60
Hydrolysis yield (%)
50
40
30
20
10
0 Ultrastructure, chemical composition and biosynthesis of the cell wall in
1 2 3 4 5 Koliella antarctica (Klebsormidiales, Chlorophyta). Eur J Phycol 2000;35:
Reuse number (times)
Fig. 12. Reusability test of the immobilized cellulase.
bilized cellulase. By installing the PAN nanofibrous membranes
on a bioreactor (Fig. 1) instead of suspending in it, the loss of
immobilized cellulase due to the breaking down caused by the
agitation could be reduced. Also, only a simple separation step is
needed for removing the immobilized cellulase from the reactant.
In addition, the high enzyme loading and low diffusion resis-
tance of a nanofibrous enzyme could reduce the required reaction
time [19].
4. Conclusions
This study presents a promising method of immobilizing cellu-
lase onto electrospun PAN nanofibrous membranes. FTIR results
indicated the immobilization mechanism between the cellulase
and PAN supports was covalent bonding. The optimal condi-
tions for preparing an immobilized cellulase were studied, and
an activation time of 7.5 min, an enzyme concentration of 2 wt%,
an immobilization time of 30 min, and a temperature of 40 ◦ C
were obtained. An immobilized cellulase with a protein loading of
30 mg/g-support with a specific activity of 3.2 U/mg-protein could
be obtained under these immobilization conditions. The enzy-
matic stability of cellulase against pH and thermal stresses was
improved by immobilization. The obtained immobilized cellulase
was then used to hydrolyze the cell wall of microalgae to produce
reducing sugars. The enzymatic hydrolysis was optimized by RSM.
The optimal reaction conditions were determined, and a hydrol-
ysis yield of 63.15% was obtained. This study provides a novel
process for cellulase immobilization and demonstrates an alter-
native process for sugar production by hydrolyzing the cell wall of
microalgae.
References
[1] Fu N, Peiris P, Markham J, Bavor J. A novel co-culture process with Zymomonas
mobilis and Pichia stipitis for efficient ethanol production on glucose/xylose
mixtures. Enzyme Microb Technol 2009;45:210–7.
[2] Hang HF, Ye XH, Guo MJ, Chu J, Zhuang YP, Zhang M, et al. A simple fermentation
strategy for high-level production of recombinant phytase by Pichia pastoris
using glucose as the growth substrate. Enzyme Microb Technol 2009;44:185–8.
[3] Golovleva L, Golovlev E. Microbial cellular biology and current problems of
metabolic engineering. J Mol Catal B Enzym 2000;10:5–21.
[4] Cervero JM, Skovgaard PA, Felby C, Sorensen HR, Jorgensen H. Enzymatic
hydrolysis and fermentation of palm kernel press cake for production of
bioethanol. Enzyme Microb Technol 2010;46:177–84.
[5] Wang YG, Wang XN, Tang RT, Yu SS, Zheng BS, Feng Y. A novel thermostable cel-
lulase from Fervidobacterium nodosum. J Mol Catal B Enzym 2010;66:294–301.
[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.
[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
methodology. Ind Eng Chem Res 2009;48:7346–53.
[8] Geddes CC, Peterson JJ, Roslander C, Zacchi G, Mullinnix MT, Shanmugam
KT, et al. Optimizing the saccharification of sugar cane bagasse using dilute
phosphoric acid followed by fungal cellulases. Bioresour Technol 2010;101:
1851–7.
[9] Binod P, Sindhu R, Singhania RR, Vikram S, Devi L, Nagalakshmi S, et al.
Bioethanol production from rice straw: an overview. Bioresour Technol
2010;101:4767–74.
[10] Girio FM, Fonseca C, Carvalheiro F, Duarte LC, Marques S, Bogel-Lukasik
R. Hemicelluloses for fuel ethanol: a review. Bioresour Technol 2010;101:
4775–800.
[11] Fu CC, Hung TC, Chen JY, Su CH, Wu WT. Hydrolysis of microalgae cell
walls for production of reducing sugar and lipid extraction. Bioresour Technol
2010;101:8750–4.
[12] Piro G, Lenucci M, Dalessandro G, La Rocca N, Rascio N, Moro I, et al.
331–7.
[13] Spolaore P, Joannis-Cassan C, Duran E, Isambert A. Commercial applications of
microalgae. J Biosci Bioeng 2006;101:87–96.
[14] Piccolo C, Wiman M, Bezzo F, Liden G. Enzyme adsorption on SO2 cat-
alyzed steam-pretreated wheat and spruce material. Enzyme Microb Technol
2010;46:159–69.
[15] Varnai A, Siika-Aho M, Viikari L. Restriction of the enzymatic hydrolysis of
steam-pretreated spruce by lignin and hemicellulose. Enzyme Microb Technol
2010;46:185–93.
[16] Xu F, Ding HS, Osborn D, Tejirian A, Brown K, Albano W, et al. Partition of
enzymes between the solvent and insoluble substrate during the hydrolysis of
lignocellulose by cellulases. J Mol Catal B Enzym 2008;51:42–8.
[17] Vaillant F, Millan A, Millan P, Dornier M, Decloux M, Reynes M. Co-immobilized
pectinlyase and endocellulase on chitin and nylon supports. Process Biochem
2000;35:989–96.
[18] Mao XP, Guo GJ, Huang JF, Du ZY, Huang ZS, Ma L, et al. A novel method to
prepare chitosan powder and its application in cellulase immobilization. J Chem
Technol Biotechnol 2006;81:189–95.
[19] Wu LL, Yuan XY, Sheng J. Immobilization of cellulase in nanofibrous PVA mem-
branes by electrospinning. J Membr Sci 2005;250:167–73.
[20] Koski A, Yim K, Shivkumar S. Effect of molecular weight on fibrous PVA pro-
duced by electrospinning. Mater Lett 2004;58:493–7.
[21] Li SF, Chen JP, Wu WT. Electrospun polyacrylonitrile nanofibrous membranes
for lipase immobilization. J Mol Catal B Enzym 2007;47:117–24.
[22] Sun W, Cai Q, Li P, Deng XL, Wei Y, Xu MM, et al. Post-draw PAN-PMMA
nanofiber reinforced and toughened bis-GMA dental restorative composite.
Dent Mater 2010;26:873–80.
[23] Wang Y, Serrano S, Santiago-Aviles JJ. Conductivity measurement of electro-
spun PAN-based carbon nanofiber. J Mater Sci Lett 2002;21:1055–7.
[24] Ye P, Xu ZK, Wu J, Innocent C, Seta P. Nanofibrous membranes containing reac-
tive groups: electrospinning from poly(acrylonitrile-co-maleic acid) for lipase
immobilization. Macromolecules 2006;39:1041–5.
[25] Li SF, Wu WT. Lipase-immobilized electrospun PAN nanofibrous membranes
for soybean oil hydrolysis. Biochem Eng J 2009;45:48–53.
[26] Northcote DH, Goulding KJ, Horne RW. The chemical composition and structure
of the cell wall of Chlorella pyrenoidosa. Biochem J 1958;70:391–7.
[27] Hsieh CH, Wu WT. A novel photobioreactor with transparent rectangular cham-
bers for cultivation of microalgae. Biochem Eng J 2009;46:300–5.
[28] Bradford MM. A rapid and sensitive method for the quantitation of micro-
gram quantities of protein utilizing the principle of protein–dye binding. Anal
Biochem 1976;72:248–54.
[29] Ghose K. Measurement of cellulase activities. Pure Appl Chem 1987;59:
257–68.
[30] Zhou JQ. Immobilization of cellulase on a reversibly soluble–insoluble support:
properties and application. J Agric Food Chem 2010;58:6741–6.
[31] Miller GL. Use of dinitrosalicylic acid reagent for determination of reducing
sugar. Anal Chem 1959;31:426–8.
[32] DuBois M, Gilles KA, Hamilton JK, Rebers PA, Smith F. Colorimetric method
for determination of sugars and related substances. Anal Chem 1956;28:
350–6.
[33] Wang ZG, Wan LS, Liu ZM, Huang XJ, Xu ZK. Enzyme immobilization on electro-
spun polymer nanofibers: an overview. J Mol Catal B Enzym 2009;56:189–95.
[34] Basri M, Ampon K, Yunus WMZ, Razak CNA, Salleh AB. Amidination of lipase
with hydrophobic imidoesters. J Am Oil Chem Soc 1992;69:579–83.
[35] Zhang HT, Nie HL, Yu DG, Wu CY, Zhang YL, White CJB, et al. Surface modifica-
tion of electrospun polyacrylonitrile nanofiber towards developing an affinity
membrane for bromelain adsorption. Desalination 2010;256:141–7.
[36] Skoog DA, Holler FJ, Crouch SR. Principles of instrumental analysis. 6th ed.
California: Thomson Higher Education; 2007.
[37] Stuart B. Biological applications of infrared spectroscopy. West Sussex: John
Wiley & Sons, Ltd; 1997.
[38] Ting WJ, Tung KY, Giridhar R, Wu WT. Application of binary immobilized
Candida rugosa lipase for hydrolysis of soybean oil. J Mol Catal B Enzym
2006;42:32–8.
 T.-C. Hung et al. / Enzyme and Microbial Technology 49 (2011) 30–37 37

[39] Yigitoglu M, Temocin Z. Immobilization of Candida rugosa lipase on
glutaraldehyde-activated polyester fiber and its application for hydrolysis of
some vegetable oils. J Mol Catal B Enzym 2010;66:130–5.
[40] Cheng ZF, Wu CM, Yang WH, Xu TW. Preparation of bromomethylated poly(2,6-
dimethyl-1,4-phenylene oxide) hollow fiber cation-exchange membranes and
immobilization of cellulase thereon. J Membr Sci 2010;358:93–100.
[41] Mateo C, Palomo JM, Fernandez-Lorente G, Guisan JM, Fernandez-Lafuente R.
Improvement of enzyme activity, stability and selectivity via immobilization
techniques. Enzyme Microb Technol 2007;40:1451–63.
[42] Jeya M, Zhang YW, Kim IW, Lee JK. Enhanced saccharification of alkali-treated
rice straw by cellulase from Trametes hirsuta and statistical optimization of
hydrolysis conditions by RSM. Bioresour Technol 2009;100:5155–61.