CelB and β-glucosidase immobilization for carboxymethyl cellulose hydrolysis

Article abstract of DOI:10.1039/c3ra43666g

Cellulose hydrolysis is an important step in the production of bioethanol from cellulosic biomass. Two key cellulase enzymes, celB and β-glucosidase, were covalently immobilised on polystyrene treated with plasma immersion ion implantation (PIII) which creates radicals that form covalent bonds. The immobilized enzymes were used to produce glucose from carboxymethyl cellulose (CMC), a solubilised form of cellulose. CelB from Caldicellulosiruptor saccharolyticus is a thermophilic cellulase with both endoglucanase and exoglucanase activity. The highest activity of the immobilised celB on PIII treated surfaces was achieved when their immobilisation is carried out at a pH in the range 5-6.5. The immobilized celB on the PIII treated surface had the same activation energy as free celB showing substrate accessibility is not affected by the presence of the surface. The V_max and K_m values of immobilized celB were found to be comparable to those of equal free celB concentrations. The areal density of immobilized celB on the PIII treated surface was estimated to be 0.3 μg cm^-2. The polystyrene surface with immobilized celB at 45 °C could be reused over four times (23 hours each) with approximately 30% total activity loss. High ratios of β-glucosidase to celB were found to enhance the activity of immobilized celB for CMC hydrolysis.

Full text of DOI:10.1039/c3ra43666g

RSC Advances
View Article Online
View Journal | View Issue
PAPER
CelB and b-glucosidase immobilization for
carboxymethyl cellulose hydrolysis
Cite this: RSC Adv., 2013, 3, 23604
*
Clara T. H. Tran, Neil J. Nosworthy, Alexey Kondyurin, David R. McKenzie
and Marcela M. M. Bilek
Cellulose hydrolysis is an important step in the production of bioethanol from cellulosic biomass. Two key
cellulase enzymes, celB and b-glucosidase, were covalently immobilised on polystyrene treated with plasma
immersion ion implantation (PIII) which creates radicals that form covalent bonds. The immobilized
enzymes were used to produce glucose from carboxymethyl cellulose (CMC), a solubilised form of
cellulose. CelB from Caldicellulosiruptor saccharolyticus is
a thermophilic cellulase with both
endoglucanase and exoglucanase activity. The highest activity of the immobilised celB on PIII treated
surfaces was achieved when their immobilisation is carried out at a pH in the range 5–6.5. The
immobilized celB on the PIII treated surface had the same activation energy as free celB showing
substrate accessibility is not affected by the presence of the surface. The Vmax and Km values of
immobilized celB were found to be comparable to those of equal free celB concentrations. The areal
density of immobilized celB on the PIII treated surface was estimated to be 0.3 mg cmꢀ2. The polystyrene
surface with immobilized celB at 45 ꢁC could be reused over four times (23 hours each) with
approximately 30% total activity loss. High ratios of b-glucosidase to celB were found to enhance the
activity of immobilized celB for CMC hydrolysis.
Received 15th July 2013
Accepted 1st October 2013
DOI: 10.1039/c3ra43666g
www.rsc.org/advances
differing units, it is clear that large quantities of enzymes are
currently required for industrial scale processes. Therefore the
1. Introduction
The production of ethanol from agricultural sources is a fast
growing industry with approximately 400% growth over the last
decade. The development of new enzymes is opening the way to
the so called “second generation” technology where ethanol is
derived from low cost cellulosic biomass sources such as
municipal waste, agricultural waste and forest waste. At the
moment, cellulosic ethanol production is limited to pilot scale
plants due to the relatively low efficiency of hydrolysing the
biomass into fermentable sugars. The hydrolysis process
comprises many transformations in which cellulose and
hemicellulose are hydrolysed into shorter fragments and nally
into the short chain sugars including hexoses and pentoses.
This process is usually implemented with mixtures of enzymes
known as cellulases. There are several commercial cellulolytic
enzymes for cellulosic biomass hydrolysis in the market such as
the “Cellic” series from Novozyme and the “Accelerase” series
from Genencor. A dosage of 50 kg of Cellic Ctec3 is required per
1000 kg of ethanol while at least 250 kg of a competing enzyme
is needed to create the same amount of ethanol.1 “Accelerase
Trio” from Genencor requires a dosage of 30–160 l per ton of
biomass.2 Although it is difficult to make a comparison due to
enzyme cost to hydrolyse the low value and abundant cellulosic
biomass at these levels is high, reducing the economic benets
of cellulosic ethanol. One possible way of reducing the effective
cost of these enzymes would be to re-use the enzymes.
Cellulases are enzyme mixtures which hydrolyse the b-1,4
glucosidic linkages in cellulose. Cellulases consist of three main
groups of enzymes: endoglucanases (these randomly cleave
internal bonds to create new chain ends), exoglucanases (these
cleave every two or four units from the ends to create cellobiose
or cellotetraose) and b-glucosidase (this cleaves cellobiose into
two glucose units). Beta-glucosidase, the last enzyme in the
hydrolysis process, is required in large quantities. A new ther-
mophilic b-glucosidase has recently been immobilized on a
plasma immersion ion implantation (PIII) treated polystyrene
surface and has been successfully reused, substantially
increasing its economic efficiency.3
PIII treatment is a method in which energetic ions bombard
the polymer surface, breaking bonds in polymer chains and
creating radicals in the surface and subsurface regions.4 The
surface exposed to air aer PIII treatment becomes more
hydrophilic with the appearance oxygen containing groups.5
The radicals on the PIII treated polystyrene surface can be
detected using electron paramagnetic resonance.6 Cells,
proteins and enzymes have been reported to be covalently
attached to the PIII treated surfaces within a short incubation
Applied and Plasma Physics, School of Physics (A28), University of Sydney, Sydney,
New South Wales 2006, Australia. E-mail: C.Tran@physics.usyd.edu.au; Tel: +61 2
91141338
23604 | RSC Adv., 2013, 3, 23604–23611
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
RSC Advances
time.3,7–10 Unlike other covalent immobilization using chemical
linker groups, these biomolecules bind to the PIII treated
surface without requiring any chemical linkers. Experiments
with copolymers containing carboxyl groups11 showed that
carboxyl groups appearing upon oxidation of the surface are not
responsible for the covalent bonding of the biomolecules. A
direct link between the surface embedded radicals and the
covalent attachment was established4 by the correspondence of
the time dependences of these processes.
Covalent binding is known to be stronger than physical
adsorption and ionic binding and can prevent enzyme desorp-
tion from the carrier.12 Because the covalent bonds are irre-
versible, it is important that the enzyme keep its activity while
being immobilized. Most polymers have hydrophobic surfaces.
The adsorption of enzymes on a polymer surface can induce a
change of enzyme conformation due to the spreading of the
hydrophobic domains of the enzyme molecule on that surface.
As a consequence, its activity decreases. In contrast, the poly-
mer surface aer the PIII treatment is more hydrophilic.5,8,11
When an enzyme molecule approaches the PIII treated surface,
less change in its conformation is required. Hence, its catalytic
activity is retained. b-Glucosidase3 and horseradish peroxide8
attached on the PIII treated polymers have been observed to
have higher activity than on the untreated surfaces.
2. Materials and methods
2.1 Expression and purication of recombinant celB from
Caldicellulosiruptor saccharolyticus
Genomic DNA from Caldicellulosiruptor saccharolyticus was a
gi from Serve Kengen. The PCR product for the protein (Uni-
ProtKB/Swiss-Prot Accession number P10474) was sub-cloned
into pET 3C (Novagen) using restriction enzymes NdeI and
BamH1, transformed into E. coli BL21 (DE3) Star (Invitrogen)
and the product was veried by DNA sequencing. The protein
was expressed with and without the putative signal peptide. For
this paper, the expressed protein was used without the putative
signal peptide according to the UniProtKB/Swiss-Prot data base
(residues 29–1039).
E. coli cells were grown in culture, harvested and resuspended
in 100 ml of 10 mM Tris, 100 mM EDTA, 100 mM EDTA pH 8 and
lysed using a Rannie homogenizer. Aer lysis the supernatant
was heated to 70 ꢁC for 30 minutes and centrifuged. The
supernatant was subsequently applied to a Q Sepharose ion
exchange column and proteins were eluted using a 0–0.5 M NaCl
gradient. The peak containing celB was pooled and precipitated
ꢁ
with 50% ammonium sulphate overnight at 4 C. The precipi-
tation was dissolved in 10 mM Tris at pH 8 and then applied to a
Sephacryl S-200HR gel ltration column and run using 25 mM
NaCl, 10 mM Tris buffer at pH 8. The peak containing celB was
pooled and applied to a Mono Q column again. The column was
washed with 10 mM phosphate buffer pH 8 and eluted with a 0–
0.5 M gradient of NaCl in 10 mM phosphate pH 8. Fractions
containing celB were pooled and dialysed against 10 mM
In this paper, celB, a thermophilic enzyme performing the
functions of both endoglucanases and exoglucanases,13 was
immobilised and demonstrated its use together with immobi-
lised thermophilic b-glucosidase in an enzymatic hydrolysis
process. The use of only two enzymes is a simplication of the
process that may lead ultimately to cost advantages. In order to
obtain the best performance, thermophilic enzymes were used
to allow the operation at higher temperatures while preserving
enzyme function.
ꢁ
phosphate buffer pH 7, aliquoted and stored frozen at ꢀ80 C.
Protein concentration was determined using the molar extinc-
tion coefficient (3280 ¼ 298 565 Mꢀ1 cmꢀ1). The yield of celB was
approximately 4 mg lꢀ1
.
It has been reported in two papers that celB consists of three
domains, Saul et al.13 label the domains A, B and C while Park
et al.14 label them GH10, CBM3 and GH5. The middle domain
(B or CBM3) is a binding domain which helps celB to bind to an
insoluble cellulose substrate. When used with the solubilised
cellulose, carboxymethyl cellulose (CMC), domain C was found
to have CMCase activity (activity to hydrolyse CMC) while
domain A was found to have exoglucanase activity. HPLC
analysis showed that cellobiose is the main product of celB
when using cellooligosaccharide15 or “Avicel” (crystalline cellu-
lose)14 as substrate. Cellobiose inhibits celB catalytic activity
when present in high concentrations.14 In the presence of
b-glucosidase, cellobiose is hydrolysed into glucose quickly,
hence reducing the inhibition effect.
The aims of this work are the identication of the optimum
parameters for the immobilization of recombinant celB onto
PIII treated polystyrene (PS) surfaces, determining the effects of
immobilisation on the thermal stability and activity; and
assessing the reusability of immobilized celB and its synergy
when used with surface immobilised b-glucosidase. This work
has important implications for the production of ethanol using
immobilised enzymes. The possibility of a fully immobilised
2.2. PIII treatment of polystyrene and celB immobilization
Polystyrene sheet (0.19 mm thick from Goodfellow) was
attached to a conductive sample holder connected to an elec-
trode with a pulsed bias applied to it. A stainless steel mesh was
placed approximately 50 mm from the sample surface to reduce
charging effects.16 The sample holder was immersed in nitrogen
plasma (generated by RF power, 100 W, 13.56 MHz) and peri-
odically bombarded by nitrogen ions under a bias of ꢀ20 kV,
50 Hz for 20 ms. The ion uence over the treatment was 5 ꢂ 1015
ions per cm2. Samples were treated on both sides and used
within 24 hours aer treatment.
Samples (PIII treated and untreated PS) were incubated in
celB solution on a rocker overnight at room temperature
(approximately 22 ꢁC). All celB solution for incubation was
prepared with a concentration of 95 mg mlꢀ1 using acetate buffer
(10 mM, pH 5.5) unless stated. Aer immobilization, samples
were transferred to a new container and washed four times with
the same acetate buffer to remove any unbound molecules.
2.3 Test of covalent binding
enzyme scheme would allow the maximum benets of immo- This experiment was performed to illustrate the different
bilisation to be realised.
mechanism of celB attachment onto PIII treated surfaces
This journal is ª The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 23604–23611 | 23605

View Article Online
RSC Advances
Paper
compared to untreated surfaces. CelB was immobilized on PIII spectrophotometer (Amersham, Pharmacia Ultrospec 3100 Pro
treated and untreated PS, washed using acetate buffer and milli- Spectrophotometer) at 540 nm. CMC solution or milliQ water
Q water and dried at room temperature before subjected to was added to DNS reagent in the same procedure and used as a
Fourier Transform Infrared Spectroscopy-Attenuated Total blank. Cellobiose and glucose were used to build standard
Reection (FTIR-ATR) analysis. Spectra from the sample curves for absorbance conversion into relevant concentration.
surfaces were recorded using a Digilab FTS 7000 FTIR spec-
trometer with germanium crystal ATR accessory (Harrick Inc.,
USA) and an incident angle of 45ꢁ. Five hundred scans were
2.5 Thermal stability of free celB in solution and
immobilized celB on PIII treated surface
done on each analysis with a resolution of 4 cmꢀ1
.
The purpose of this experiment is to investigate the changing of
the thermophilic property of celB when being immobilized and
determine the optimum temperatures of hydrolysis.
Aer that, both samples were treated with 2% Sodium
Dodecyl Sulphate (SDS) at 70 ꢁC for an hour. SDS washing is
commonly used to test for covalent binding in the litera-
ture.4,10,17 Aer the treatment, samples were washed three times
with milli-Q water and dried in a desiccator before being ana-
lysed again with FTIR-ATR.
The peaks corresponding to the PS were removed from the
protein-containing spectra by subtraction of the spectra of the
same surfaces receiving identical exposure to buffer and
atmosphere except without protein. Spectra aer subtraction
before and aer SDS treatment were normalized to account for
pressure differences between sample and ATR crystal during
measurement using the intensity of the line of PS at 1492 cmꢀ1
in the original spectrum with protein.
CelB was added to 10 ml of CMC solution (6.7 mg mlꢀ1, pH
5.5, 50 mM acetate) with a concentration of 0.38 mg mlꢀ1
.
Immobilized celB on PIII treated PS pieces (2 ꢂ 4 cm2) were
prepared as described above and added in 10 ml of CMC solu-
tion with the same concentration. The tubes were incubated at
different temperatures and samples were taken in the linear
stage to determine the reaction rate (K, mM minꢀ1) using DNS
assay. The absorbance was converted into cellobiose concen-
tration using standard curve from cellobiose. The Arrhenius
equation of the hydrolysis is determined as follow:
k ¼ A exp(ꢀEa/kBT)
(1)
where k is the rate constant (minꢀ1), A is a pre-exponential
factor, Ea is the activation energy (Joule), kB ¼ 1.38 ꢂ 10ꢀ23 J Kꢀ1
is Boltzmann constant, T is Kelvin temperature.
Rate constants (k) are proportion to the reaction rates (K, mM
minꢀ1) due to the same concentrations of substrate used at each
temperature. Logarithm values of reaction rates at different
temperatures were plotted against 1/T to determine the slope of
the linear part and subsequently calculate the activation energy
2.4 Inuence of pH on carboxymethyl cellulose (CMC)
hydrolysis using free celB and immobilized celB
Enzyme activity is affected by the pH of the buffer. In order to
determine the optimum pH for celB attachment on PIII treated
surface, the optimum pH of free celB to hydrolyse CMC in
solution was studied. CMC solutions (6.7 mg mlꢀ1) were
prepared with different buffers (pH ¼ 4.5, 5.0, 5.5, 6.0, 6.5 and
7.0) at a concentration of 50 mM. Acetate was used for buffer
with pH 4.5–5.5 and MES was used for buffers with pH 6.0–7.0.
CelB was added to 15 ml of these CMC solutions at 0.76 mg mlꢀ1
and incubated in a water bath at 55 ꢁC. Samples were taken aer
ln K ¼ ln A ꢀ Ea/kBT
(2)
24 hours to analyse reducing sugar using DNS assay. Optimum 2.6 Comparison of free celB and immobilized celB activities
pH of free celB is the pH which shows the highest activity.
The PIII treated surface has an increasing negative charge in
buffers ranging from pH 4–7.18 When celB comes to the PIII
To study the activity of immobilized celB, we compare its Vmax
and Km values with those of free celB in substrate solutions of
increasing concentration. For these comparisons, we have to
choose a concentration of free celB which is equal to the
treated surface and attaches to the surface, the charge on celB
molecules and the surface determine the enzyme's orientation
amount of immobilized celB on the polymer surface. The
and subsequently may or may not block the catalytic domain.
concentration of active celB immobilized on PIII treated PS was
Therefore, it is important to study the good range of pH for celB
estimated from the number of molecules to be tted to the
surface area of polymer. If we assume that celB molecules fold
in a spherical shape, their minimum radius Rm can be calcu-
lated from their mass as proposed by Erickson:20
immobilization. Six different buffers (pH ¼ 4.5, 5.0, 5.5, 6.0, 6.5
and 7.0) were prepared with low concentration (10 mM). PIII
treated surfaces (2 ꢂ 4 cm2) were incubated with celB in
different buffers overnight; washed four times with the same
buffers and assayed in 10 ml CMC solution (6.7 mg mlꢀ1, pH
Rm ¼ 0.066M1/3 (M is the molecular weight of protein in Dalton)
5.5, 50 mM acetate). The reducing sugars created in each pH
buffers were determined using DNS assay to choose the
optimum pH for celB attachment.
CelB has a molecular weight of 118–120 kDa,13 therefore the
DNS assay: an aliquot of 100 ml sample was added to an minimum radius is approximately 3.25 nm. However, the radius
Eppendorf tube with the same volume of 1% DNS (dini- of celB could be larger because celB has 3 domains which are
trosalicylic acid) reagent.19 The mixture was heated at 90 ꢁC in a independently folded. Therefore, we chose two concentrations
water bath for 10 minutes before adding Rochelle salt (33.3 ml) of free celB (0.96 and 0.40 mg mlꢀ1) which correspond
to stabilize the colour. Absorbance was done by to the amount of enzyme in a densely packed monolayer on a
23606 | RSC Adv., 2013, 3, 23604–23611
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
RSC Advances
2 ꢂ 4 cm2 (2 sided) surface assuming enzyme molecular radii 24, 31 and 48 hours and activities of celB and b-glucosidase were
3.1and 4.8 nm respectively. We assumed that dense packing was determined using the DNS assay and a glucose oxidase/peroxi-
achieved in a hexagonal lattice (packing density of 0.9068). Ten dase assay kit (Sigma #G3660-1CAP) respectively. The reducing
millilitres of CMC solutions were prepared in 15 ml falcon tubes sugars as determined by the DNS assay were used to calculate
with concentrations varying from 5 to 50 g lꢀ1 in acetate buffer the synergy between immobilized celB and immobilized
(50 mM, pH ¼ 5). PIII treated polystyrene samples (2 ꢂ 4 cm2) b-glucosidase. Error bars were estimated at approximately 9.2%
were incubated with celB overnight and washed as described due to the total aliquot volume of sampling.
above before being transferred to the tubes with CMC solution.
The experiments using free celB with two concentrations
(0.96 and 0.40 mg mlꢀ1) were conducted at the same time. The
hydrolysis was performed at 55 ꢁC and 5 samples (300 ml each)
were taken from each concentration to determine the reaction
rate using the DNS assay. The absorbance at 540 nm was con-
verted into cellobiose concentration by interpolating against the
standard curve of cellobiose in the same buffer. The slopes of
the straight line which correspond to the reaction rates at each
substrate concentration were plotted against CMC concentra-
tions. Vmax and Km values were calculated by tting the points
with the following equation:
3. Results and discussion
3.1 The use of PIII treatment to achieve covalent surface
immobilisation of celB
Fig. 1 shows the subtracted spectra of PIII treated and untreated
surfaces before and aer treating with SDS. The presence of
three amide peaks (3300 cmꢀ1 (Amide A), 1650 cmꢀ1 (Amide I)
and 1540 cmꢀ1 (Amide II)) on spectra before SDS indicates the
presence of a protein layer on both surfaces aer incubation
V ¼ Vmax[S]/(Km + [S])
(3)
where V (mM hꢀ1) is the reaction rate, [S] (g lꢀ1) is the substrate
concentration, Vmax (mM hꢀ1) is the highest reaction rate under
the experiment condition, Km (g lꢀ1) is the Michaelis constant.
2.7 Reuse of immobilized celB
Although the reaction rate increases with increasing tempera-
ture, the enzyme tends to be inactivated quicker. To study how
the immobilized celB is stable with temperature, the reuse test
ꢁ
was conducted with four temperatures (45, 55, 65 and 75 C).
CelB immobilized on PIII treated surface (2 ꢂ 4 cm2) was added
to 10 ml of CMC solution (6.7 mg mlꢀ1, pH 5.5, 50 mM acetate)
and incubated in water baths at those temperatures. An aliquot
was taken from each tube aer 23 hours for DNS assay. Aer
that the PIII treated samples were transferred to new tubes with
10 ml of CMC solution to repeat the hydrolysis. The activity of
immobilized celB over 3 courses of re-use was normalized with
the rst use to compare the activity loss at different
temperatures.
2.8 Simultaneously working of celB and b-glucosidase
Recombinant b-glucosidase was cloned and puried as
described by Hirsh et al.3 Two sets of experiment were con-
ducted to study the simultaneous hydrolysis of immobilized
celB and b-glucosidase on CMC. CelB and b-glucosidase were
immobilized on separate PIII treated pieces (1 ꢂ 4 cm2) in
separate containers so as to avoid competition between the two
enzymes during covalent immobilisation. The surface immo-
bilised enzymes on PIII supports were then washed and incu-
bated in 10 ml CMC solution (6.7 mg mlꢀ1, pH 5.0, 50 mM
acetate) at 55 ꢁC. The surface area ratios of PIII treated PS
immobilized with b-glucosidase to PIII treated PS immobilized
with celB vary from 0 to 10 in the rst experiment with low celB
(0.8 cm2 mlꢀ1) and from 0 to 4 in the second experiment with
Fig.
1
FTIR-ATR spectra of amide absorptions from immobilised celB on
untreated (A) and PIII treated polystyrene (PS) (B) before and after SDS treatment.
PS peaks were removed by subtraction of spectra of the same surfaces receiving
identical exposure to buffer and atmosphere except without protein. The ability
higher celB (1.6 cm2 mlꢀ1). An aliquot of 230 ml was taken at 4, 7, of PIII treated surfaces to retain the enzyme shows covalent binding.
This journal is ª The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 23604–23611 | 23607

View Article Online
RSC Advances
Paper
Fig. 3 (A) Plot of logarithm of the reaction rate constant of free celB (-) and
immobilized celB (C) as a function of inverse temperature. The linearity of the
plot indicates Arrhenius behaviour. (B) Dependence of reaction rate constant for
free celB (-) and immobilized celB (C) on temperature. The solid lines are guides
for the eye. Temperature induced conformational changes limit the activity at
temperatures above 65 ꢁC for immobilised celB and above 75 ꢁC for free celB.
Fig. 2 (A) The dependence of free celB activity on the pH of CMC solution. (B)
The dependence of immobilized celB activity on the pH of the solution used for
immobilisation.
with celB. Aer the rigorous SDS treatment, there is a signicant
reduction of amide peak intensities on the untreated surface
but only a small reduction on the PIII treated surface. It can be
concluded that the PIII treatment greatly increases the ability of
the surface to bind the enzyme covalently. The removal of
protein from the untreated surface indicates that the binding is
mainly by physical adsorption.
the surface do not substantially affect their activity. There is a
departure from the Arrhenius behaviour at 65 ꢁC for the
ꢁ
immobilised celB and at 75 C for the celB in solution. This is
attributed to the onset of unfolding which occurs somewhat
more readily on the surface. The presence of the surface can
destabilise (or lower the energy of some unfolded states of)
some enzymes.21,22 When temperature increases from 40 to
80 ꢁC, the reaction rate of free celB increases 5.5 times as
compared to 4.5 times of immobilized celB (Fig. 3B). In both
cases, the temperature of 80 ꢁC shows the highest reaction rate.
The reaction rates of immobilized celB on the surface are
comparable to those of free celB in the middle range (45–65 ꢁC),
3.2 Inuence of pH on activity of free and immobilized celB
The activity of free celB in CMC solution is high in the range pH
5–7 and decreases rapidly as the pH decreases below 5 (Fig. 2A).
The optimum pH is 5.5 which is similar to the optimum pH of
GH5 domain reported by Park et al.14 The pH at which the
immobilisation is performed also inuences the activity. The
relatively high activity of celB (Fig. 2B) aer being immobilized
in a range of pH buffers from 4.5 to 7 indicates that the orien-
tation of celB relative to the surface and relative to its neigh-
bours when immobilized is favourable so that the active sites of
the enzyme are not hindered. The optimum pH range for
attachment is 5–6.5. Therefore, pH 5.5 was chosen for all celB
immobilisation in this work.
ꢁ
but are lower at higher temperatures (70–90 C).
As can be seen from Fig. 3B, the activity of both free and
immobilized celB dramatically decreases from 80 to 90 ꢁC. The
decrease in reaction rate in this temperature range is similar for
free and immobilized celB as partially unfolded lower activity
states become dominant.
3.4 Activity of immobilized celB in comparison with free
celB in solution
3.3 Thermophilic property of immobilized celB
Fig. 4 illustrates the reaction rates as a function of CMC
concentration of free celB of two different concentrations
The temperature dependent behaviour of an enzyme reveals the
energetics of its interactions. For an enzyme catalysing a rst
order reaction with an activation energy of Ea, the rate constant
for its catalytic activity increases with temperature according to
the Arrhenius relation of eqn (1). The activation energy is
determined by the energy barrier to be overcome during the
reaction. For an immobilized enzyme it is possible that the
proximity of the surface affects the energy barrier. In this
experiment, a concentration of celB in solution was chosen to
show an activity close to the activity of the immobilized celB on
a 2 ꢂ 4 cm2 PIII treated PS surface. Fig. 3A shows a plot of the
logarithm of the reaction rate as a function of inverse temper-
ature and shows Arrhenius behaviour as per eqn (1). The acti-
vation energies obtained from the plot of free and immobilized
celB are (8.27 ꢃ 1.01) ꢂ 10ꢀ20 and (7.81 ꢃ 1.36) ꢂ 10ꢀ20 Joule
respectively and therefore equal within the uncertainty limits.
The similar activation energy of immobilized and free celB
Fig. 4 Enzymatic reaction rate of immobilized and free celB as a function of
implies the immobilization of the celB enzyme molecules on substrate concentration.
23608 | RSC Adv., 2013, 3, 23604–23611
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
RSC Advances
65 ꢁC, 70%, 60% and 40% of the activity remain aer 4 days
respectively. Therefore, there is a trade-off between the rate
constant and the total activity which should be considered
when choosing the optimum working temperature for immo-
bilized celB.
Table 1 Vmax and Km values of free and immobilized celB obtained from the
fitting of eqn (3) to experimental data
Free celB
Immobilized
Free celB
0.4 mg mlꢀ1
celB on 16 cm2
0.96 mg mlꢀ1
Vmax, mM hꢀ1
Km, g Lꢀ1
0.61 ꢃ 0.07
26.6 ꢃ 6.69
0.64 ꢃ 0.05
17.33 ꢃ 3.7
1.83 ꢃ 0.35
45.58 ꢃ 14.88
3.6 CMC hydrolysis with immobilized celB and b-
glucosidase
(0.4 and 9.6 mg mlꢀ1) and immobilized celB. The tting values of
eqn (3) are shown in Table 1. The reaction rate of the immo-
bilised enzyme falls between the values of two different
concentrations of free enzyme, allowing a calculation of the
effective concentration of the immobilised enzyme. The
measured value of 0.48 mg mlꢀ1 of this equivalent concentration
enables the areal density of active enzyme on the surface to be
calculated at 0.3 mg cmꢀ2. This number is comparable to the
areal density calculated on the basis of the molecular weight
and the assumption of a packing density of 0.9068.
CMC is a derivative of cellulose in which hydrogen atoms of the
hydroxyl groups on glucose units are replaced by sodium car-
boxymethyl groups to increase solubility. The substitutions can
occur on C2, C3 or C6 of glucose units. The CMC used in this
experiment has a molecular weight of 90 000 and a degree of
substitution (DS) of 0.7. It was found that with this level of
substitution the CMC remains susceptible to enzyme hydrolysis
while a DS higher than 1 renders it inert to cellulases.23 Enzy-
matic degradation of CMC occurs only between two adjacent
unsubstituted anhydroglucose units.23,24 Therefore the conver-
sion into glucose will become increasingly less complete with
increasing degree of substitution.
CelB hydrolyses CMC to produce shorter cello-oligosaccha-
rides and cellobiose. Cellobiose and the smaller cello-oligo-
saccharides are subsequently hydrolysed by b-glucosidase to
produce glucose. Each polymer chain has a reducing end, as do
the cellobiose and glucose molecules. The absorbance
measured in the DNS assay depends on the total number of
cello-oligosaccharide, cellobiose and glucose molecules in
solution.
Table 2 shows the total absorbance of reducing sugars in the
DNS assay and glucose concentration aer 48 hours of the
hydrolysis of 67 mg/10 ml of CMC with two different levels of
immobilized celB. At high ratios of b-glucosidase (1 : 8, 1 : 10 at
a level of 0.8 cm2 celB per ml and 1 : 4 at a level of 1.6 cm2 celB
per ml), more reducing sugars were produced than by the same
amount of celB working alone. This indicates that these high
levels of b-glucosidase can reduce cellobiose accumulation and
hence, decrease cellobiose inhibition effect on celB activity.
To calculate the enhancement effect of b-glucosidase on the
activity of celB, the following ratio was evaluated: the total
reducing sugars produced by the enzyme mixture was divided
by the sum of reducing sugars produced by celB alone. Note that
b-glucosidase acting on CMC alone produces no reducing
3.5 Reuse of celB immobilized on PIII treated polystyrene
(PS)
One advantage of the immobilized enzyme is the possibility of
reuse by transferring to a new substrate solution. This avoids
inhibition by product accumulation and allows the enzyme to
be used for longer times. The results of reuse experiments in
which PS surfaces with immobilized celB were moved to a fresh
CMC solution every day are shown in Fig. 5 for four different
temperatures. The data for each temperature are normalised to
the reducing sugars measured aer the rst use. The amount of
reducing sugars detected aer 23 hours (rst use) at 75 ꢁC is
ꢁ
double the amount detected aer the same time at 45 C. The
activity of immobilized celB at 75 ꢁC is signicantly reduced
aer the 1st use and no activity remained aer the 2nd use
(Fig. 5), indicating the complete inactivation of immobilized
enzyme on the PIII treated surface. As for samples at 45, 55 and
Table 2 Reducing end absorbance and glucose concentration of different
combinations of immobilized celB and b-glucosidase from 48 hours of CMC
hydrolysis at 55 ꢁC
CelB 0.8 cm2 mlꢀ1
CelB : b-glucosidase
1 : 0 1 : 2 1 : 4 1 : 6 1 : 8 1 : 10
1.43 1.37 1.55 1.46 1.7 1.7
Reducing end absorbance
Glucose concentration, mg mlꢀ1 0.02 0.12 0.19 0.24 0.26 0.27
CelB 1.6 cm2 mlꢀ1
CelB : b-glucosidase
1 : 0 0 : 1 1 : 0.5 1 : 1 1 : 2 1 : 3 1 : 4
Fig. 5 Comparison of celB activity measured by the DNS assay for reducing
sugars for 4 uses (23 hours each) at different temperatures. Activities of the 2nd,
3rd and 4th use were normalized to the activity at the 1st use.
Reducing end absorbance
1.80 0
1.80 1.75 1.72 1.84 2.00
0.11 0.20 0.29 0.32 0.36
Glucose concentration, mg mlꢀ1 0.01 0
This journal is ª The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 23604–23611 | 23609

View Article Online
RSC Advances
Paper
sugars. The enhancement is dependent on the composition of
the mixture and is also expected to be time dependent when one
enzyme creates substrate for another. In the case of b-glucosi-
dase and celB, the highest enhancement ratio is 1.19 obtained
at the surface area ratios b-glucosidase : celB of 8 and 10. At
higher levels of celB, the enhancement obtained at ratio b-glu-
cosidase : celB of 4 is 1.1. These levels of enhancement are
small compared to the reported values by Chir et al.25 who
studied the synergy between free cel9A and b-glucosidase in
CMC hydrolysis using the DNS assay. The CMC concentration
(10 mg mlꢀ1) in their study is 1.3 times higher than the
concentration used in this study and the concentration of cel9A
is 150 times higher than the concentration of free celB which
has equivalent activity of 1.6 cm2 per ml of immobilized celB.ꢁIn
addition, the hydrolysis in Chir et al. was conducted at 40 C
with sampling at 1 hour compared to 55 ꢁC and sampling at 48
hours in this experiment. The degree of substitution of car-
boxymethyl groups on CMC was not specied in the work of
Chir et al. They reported synergy between the two enzymes at all
ratio of the two enzymes tested and the highest synergy was 1.6
at the molar ratio of 1 : 10. The lower enhancement observed in
this study could be due to the lower enzyme concentration used
or to effects associated with being immobilised on a surface
such as reduced accessibility to substrate; hindered access to
conformation states used and proximity of the immobilized
b-glucosidase and celB to each other. In this case, hindered
access to conformation states can be ruled out for celB since the
activation energy of the immobilised enzyme is the same as that
of the molecule in solution at the temperature for which the
synergy study was carried out.
In terms of glucose production, although celB alone can
hydrolyse CMC to cellobiose and cello-oligosaccharides, only a
very small amount of the monosaccharide, glucose, was detec-
ted aer 2 days (Table 2). Beta-glucosidase alone cannot
hydrolyse CMC; hence no glucose was detected without celB.
With a xed surface area of immobilized celB, increasing
surface areas of b-glucosidase resulted in more glucose
production. Fig. 6 shows the reaction rates obtained from the
slopes glucose concentration versus reaction time. The contin-
uous line (from data with 0.8 cm2 per ml celB) increases quickly
as the enzyme area ratio changes from 0 to 6 but begins to
saturate between 6 and 10. The discontinuous line (from data
with 1.6 cm2 per ml celB) has the same increasing trend from
0 to 4.
Fig. 6 Influence of ratios between b-glucosidase and celB on glucose produc-
tion. The continuous line represents initial reaction rates (as determined by the
slope of glucose curves in Fig. 5B and D) of experiments with 0.8 cm2 per ml celB
and the discontinuous line represents reaction rates of experiments with 1.6 cm2
per ml celB. The points M, N and P are discussed in the text.
for b-glucosidase. Therefore, with the same surface area of
immobilized b-glucosidase – Fig. 6 points P (0.8 cm2 per ml of
celB and 3.2 cm2 per ml of b-glucosidase) and M (1.6 cm2 per ml
of celB and 3.2 cm2 per ml of b-glucosidase) – an increase in celB
resulted in a higher glucose production rate.
Enzyme loading can be readily increased by increasing the
area of polymer surfaces with immobilized enzyme. Since only a
monolayer of enzyme can be covalently attached to the surface,
the enzyme concentration can be maximised by maximising the
total area of immobilization surfaces in the reactor volume
whilst taking care to ensure that substrate from solution can
easily access all of the surfaces. Beads can be used to signi-
cantly increase the available surface area in a given volume,
increasingly so for small diameter beads. Bead geometry would
also be more favourable for the accessibility of substrate to the
immobilized enzyme reducing substrate depletion near the
surface. The enhancement effect may also be increased with a
fairly homogenous distribution of small sized beads.
A
homogenous distribution of celB beads and b-glucosidase
beads will reduce the travel distance of substrate–product
exchange between celB-bead surfaces and b-glucosidase bead
surfaces. The benet of large surface area of nano/microsized
hybrid materials in immobilizing enzymes has been
reviewed.12,27 The enhancement of distribution between enzyme
carriers and the substrate becomes more important when
insoluble cellulose is used as has been demonstrated from the
immobilization of cellulases and hemicellulases on 2 mm glass
porous beads.28
At the same ratio between the two enzymes, the hydrolysis to
glucose occurs more rapidly with the higher level of celB
(1.6 cm2 mlꢀ1). For example, when considering the two reac-
tions with the same ratios of enzymes, denoted M in Fig. 6
(1.6 cm2 per ml of celB and 3.2 cm2 per ml of b-glucosidase) and
N (0.8 cm2 per ml of celB and 1.6 cm2 per ml of b-glucosidase),
while the surface areas of both enzymes increase by a factor of
two, the reaction rate triples indicative of the synergy between
4. Conclusion
the two enzymes. Apart from hydrolysing cellobiose, b-glucosi- This paper shows that celB and b-glucosidase function effec-
dase from Caldicellulosiruptor saccharolyticus has been reported tively together to produce glucose from CMC when covalently
to show activity on cellotriose, cellotetraose and cellopentaose.26 immobilised on PIII treated PS. Recombinant celB is optimally
The high level of celB produces more short chain cello-oligo- immobilised in a pH range 5–6.5 and exhibits a maximum
saccharides which in addition to cellobiose are also substrates activity when used at pH 5.5. Immobilized celB shows a similar
23610 | RSC Adv., 2013, 3, 23604–23611
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
RSC Advances
activation energy to the free enzyme in CMC hydrolysis.
The areal density of immobilized celB is estimated to be
9 C. MacDonald, R. Morrow, A. S. Weiss and M. M. M. Bilek,
J. R. Soc., Interface, 2008, 5, 663–669.
0.3 mg cmꢀ2. Immobilized celB retained 70% of its initial activity 10 C. T. Tran, A. Kondyurin, S. L. Hirsh, D. R. McKenzie and
aer 4 uses of 23 hours each. High ratio of immobilized
M. M. Bilek, J. R. Soc., Interface, 2012, 9, 2923–2935.
b-glucosidase enhances the activity of immobilized celB in CMC 11 A. V. Kondyurin, P. Naseri, J. M. R. Tilley, N. J. Nosworthy,
hydrolysis.
M. M. M. Bilek and D. R. McKenzie, arXiv:1110.3125
2011.
12 E. T. Hwang and M. B. Gu, Eng. Life Sci., 2013, 13, 49–61.
13 D. J. Saul, L. C. Williams, R. A. Grayling, L. W. Chamley,
D. R. Love and P. L. Bergquist, Appl. Environ. Microbiol.,
1990, 56, 3117–3124.
14 J. I. Park, M. S. Kent, S. Datta, B. M. Holmes, Z. Huang,
B. A. Simmons, K. L. Sale and R. Sapra, Bioresour. Technol.,
2011, 102, 5988–5994.
Acknowledgements
The authors would like to thank Serve Kengen for provision of
the Genomic DNA from Caldicellulosiruptor saccharolyticus.
Many thanks are expressed to Professor Cris dos Remedios for
stimulating discussions and for access to laboratory facilities
used in this study. This work is nancially supported by the
Australian Research Council.
15 J.-D. Bok, D. A. Yernool and D. E. Eveleigh, Appl. Environ.
Microbiol., 1998, 64, 4774–4781.
16 A. Kondyurin and M. Bilek, in Ion beam treatment of polymers,
Elsevier, Amsterdam, 2008.
References
17 Y. Yin, M. M. M. Bilek, D. R. McKenzie, N. J. Nosworthy,
A. Kondyurin, H. Youssef, M. J. Byrom and W. Yang, Surf.
Coat. Technol., 2009, 203, 1310–1316.
18 C. T. H. Tran, A. Kondyurin, W. Chrzanowski, M. M. M. Bilek
and D. R. McKenzie, Colloids Surf., B, 2013, 104, 145–152.
19 G. L. Miller, Anal. Chem., 1959, 31, 426–428.
20 H. P. Erickson, Biol. Proced. Online, 2009, 11, 32–51.
21 D. B. Berkowitz and D. W. Webert, J. Immunol. Methods,
1981, 47, 121–124.
1 Novozymes, http://ethanolproducer.com/articles/8580/novoz
ymes-announces-new-advanced-biofuels-enzyme-technology.
2 Genencor, http://biosciences.dupont.com/leadmin/user_
upload/genencor/documents/AccelleraseTRIOProductLiterat
ure_120719.pdf.
3 S. L. Hirsh, N. J. Nosworthy, A. Kondyurin, C. G. dos
Remedios, D. R. McKenzie and M. M. M. Bilek, J. Mater.
Chem., 2011, 21, 17832–17841.
4 M. M. M. Bilek, D. V. Bax, A. Kondyurin, Y. Yin,
N. J. Nosworthy, K. Fisher, A. Waterhouse, A. S. Weiss,
C. G. dos Remedios and D. R. McKenzie, Proc. Natl. Acad.
Sci. U. S. A., 2011, 108, 14405–14410.
22 J. D. Whittle, N. A. Bullett, R. D. Short, C. W. Ian Douglas,
A. P. Hollander and J. Davies, J. Mater. Chem., 2002, 12,
2726–2732.
23 R. A. Gelman, J. Appl. Polym. Sci., 1982, 27, 2957–2964.
24 M. G. Wirick, J. Polym. Sci., Part A-1: Polym. Chem., 1968, 6,
1965–1974.
5 A. Kondyurin, P. Naseri, K. Fisher, D. R. McKenzie and
M. M. M. Bilek, Polym. Degrad. Stab., 2009, 94, 638–
646.
25 J. Chir, C. Wan, C. Chou and A. Wu, Biotechnol. Lett., 2011,
33, 777–782.
26 M.-R. Hong, Y.-S. Kim, C.-S. Park, J.-K. Lee, Y.-S. Kim and
D.-K. Oh, J. Biosci. Bioeng., 2009, 108, 36–40.
27 M. Verma, C. Barrow and M. Puri, Appl. Microbiol.
Biotechnol., 2013, 97, 23–39.
6 E. A. Kosobrodova, A. V. Kondyurin, K. Fisher, W. Moeller,
D. R. McKenzie and M. M. M. Bilek, Nucl. Instrum. Methods
Phys. Res., Sect. B, 2012, 280, 26–35.
7 D. V. Bax, Y. Wang, Z. Li, P. K. M. Maitz, D. R. McKenzie,
M. M. M. Bilek and A. S. Weiss, Biomaterials, 2011, 32,
5100–5111.
28 P. Mandali and B. K. Dalaly, J. ASTM Int., 2010, 7, DOI:
10.1520/JAI102580.
8 A. Kondyurin, N. J. Nosworthy and M. M. M. Bilek, Acta
Biomater., 2008, 4, 1218–1225.
This journal is ª The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 23604–23611 | 23611