1. Microplate-Based Filter Paper Assay
to Measure Total Cellulase Activity
Zhizhuang Xiao, Reginald Storms, Adrian Tsang
Centre for Structural and Functional Genomics, Concordia University,
7141 Sherbrooke Street West, Montreal H4B 1R6 Quebec, Canada;
telephone: (514) 848-2424 extension 3405; fax: (514) 848-4504;
e-mail: tsang@ vax2.concordia.ca
Received 18 February 2004; accepted 29 July 2004
Published online 30 September 2004 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/bit.20286
Abstract: The standard filter paper assay (FPA) published creased digestibility of organic matter with high fiber con-
by the International Union of Pure and Applied Chemistry tent (Garcia et al., 2002; Bowman et al., 2002). The cellu-
(IUPAC) is widely used to determine total cellulase activity. lase market is expected to expand dramatically particularly
However, the IUPAC method is not suitable for the parallel
analyses of large sample numbers. We describe here a as the industrial scale conversion of cellulosic materials to
microplate-based method for assaying large sample num- ethanol is implemented (Cherry and Fidantsef, 2003).
bers. To achieve this, we reduced the enzymatic reaction A variety of strategies can be employed to obtain ef-
volume to 60 Al from the 1.5 ml used in the IUPAC meth- fective cellulases for industrial application. For example,
od. The modified 60-Al format FPA can be carried out in different naturally produced cellulolytic enzymes can be
96-well assay plates. Statistical analyses showed that the
cellulase activities of commercial cellulases from Tricho- produced and tested in combinations to create potent syn-
derma reesei and Aspergillus species determined with our ergistic systems that hydrolyze cellulose in concert. Strat-
60-Al format FPA were not significantly different from the egies that utilize directed evolution such as random muta-
activities measured with the standard FPA. Our results genesis by error-prone PCR and DNA shuffling provide
also indicate that the 60-Al format FPA is quantitative promising methods to create novel and improved enzymes
and highly reproducible. Moreover, the addition of excess
h-glucosidase increased the sensitivity of the assay by up (Chen and Arnold, 1993; Stemmer, 1994). Both strategies,
to 60%. B 2004 Wiley Periodicals, Inc. but particularly the latter, involve the identification of en-
Keywords: filter paper assay; total cellulase activity; h- zymes with improved characteristics within a large back-
glucosidase; DNS; high throughput screening ground of wildtype, neutral, and deleterious mutants. An
effective high-throughput assay is therefore needed to
screen for cellulases with improved characteristics.
INTRODUCTION
Since the Commission on Biotechnology of The Inter-
Cellulose is an insoluble glucose polymer linked with h-1, national Union of Pure and Applied Chemistry (IUPAC)
4-glycosidic bonds. The complete hydrolysis of cellulose proposed a number of standard procedures for the measure-
into glucose requires the synergistic action of several cel- ment of cellulase activities in 1984, the filter paper assay
lulases, which include endoglucanase (EC 3.2.1.4), h-1, 4- (FPA) has been widely used to measure the total activity of
cellobiohydrolases (EC 3.2.1.91), and h-glucosidase (EC cellulases (Wood and Bhat, 1988). Filter paper is used as
3.2.1.21) activities. the standard substrate because it is readily available,
Cellulases are useful in many industrial applications. inexpensive, and assays performed with it reliably provide
They are used to convert cellulosic biomass to glucose, reproducible results (Coward-Kelly et al., 2003). This assay
which is further fermented into ethanol (Galbe and Zacchi, is performed so that 0.5 ml of diluted enzymes releases
2002; Sun and Cheng, 2002). The addition of cellulases in about 2.0 mg of glucose equivalents in 60 min, as deter-
detergents brightens the color of cotton textiles, softens the mined by the dinitrosalicylic acid (DNS) reducing sugar
fabric, and improves the removal of particulates (Niehaus assay (Miller, 1959; Wood and Bhat, 1988).
et al., 1999). Other applications of cellulases include im- However, the IUPAC filter paper assay is time-con-
provement of the nutritional value of cellulosic materials suming, labor-intensive, and requires large amounts of re-
and forage crops, enhancement of pulp quality, and in- agents. It is also difficult to obtain adequate sensitivity and
reproducibility when characterizing new cellulase mixtures
using this method. Factors that affect sensitivity and repro-
ducibility often result from the fact that most natural cel-
Correspondence to: Adrian Tsang
Contract grant sponsors: Strategic Projects grant from the Natural
lulase complexes tend to have a shortage of cellobiase
Sciences and Engineering Research Council of Canada, Genome Quebec,
´ or h-glucosidase activity (Coward-Kelly et al., 2003; Breuil
and Genome Canada et al., 1986). Most recently, Coward-Kelly et al. (2003)
B 2004 Wiley Periodicals, Inc.

2. supplemented cellobiase into a cellulase system with low Filter Paper Assay Protocols
levels of h-glucosidase activity to increase the production
All the enzyme activity values presented are averages ob-
of reducing sugar, thereby increased the sensitivity and
tained from assays performed in triplicate. The standard
reproducibility of the FPA. Decker et al. (2003) replaced
FPA assays were performed essentially as described pre-
the traditional filter paper strips with filter paper disks,
viously (Miller, 1959; Wood and Bhat, 1988). Diluted en-
Solka-Floc, SigmaCell-20, and Avicel PH101 to suit as-
zyme, in 0.5 ml volume, was added to 25-ml test tubes
say automation. However, the enzyme values detected with
containing a 1 Â 6 cm paper strip in 1.0 ml of 50 mM NaAc
these assays differed from the standard filter paper method
buffer, pH 4.8. After 60 min of incubation in a 50jC wa-
by 4% to 60%. In the cellulase assay presented here, fil-
ter bath, 3.0 ml of DNS reagent was added to each tube,
ter paper was retained as the substrate. However, the re-
followed by vigorous boiling for 5 min. Finally, after the
action volume was reduced so that cellulase assays could
colored solution was diluted with 20 ml of H2O, 200-Al
be conducted in a 96-well microplate format suitable for
samples were transferred to a 96-well flat bottom plate
high-throughput screening. Furthermore, we show that sup-
(Corning, Corning, NY) and the absorbance at 540 nm was
plementing cellulase assays with excess h-glucosidase in-
measured with a plate reader (Power Wave HT Bio-TEK
creased assay sensitivity.
Instruments, Winooski, Vermont). The amount of enzyme
used for an assay should produce about 2.0 mg of glucose
equivalents in 60 min of incubation at 50jC (Wood and
MATERIALS AND METHODS Bhat, 1988).
The 96-Al microplate-based assay was performed as fol-
Enzymes and Chemicals lows. A 32-Al aliquot of diluted enzyme was added into
wells containing a 7-mm diameter filter paper disk and
Cellulase C-2730 from Trichoderma reesei, Cellulase C- 64 Al of 50 mM NaAc buffer, pH 4.8. After 60 min of
2605 from Aspergillus species, C-6105 Novozyme 188 incubation at 50jC, 50 Al of each reaction was transferred
(h-glucosidase) from Aspergillus niger, and all chemicals into the corresponding well of a second 96-well PCR plate
were purchased from Sigma-Aldrich (St. Louis, MO). containing 100 Al of DNS and incubated at 95jC for 5 min.
All incubations were performed in a temperature cycler
(Perkin-Elmer Applied Biosystems 9700 PCR system, Nor-
Substrates
walk, CT). Following color development, a 36-Al aliquot of
Whatman No. 1 filter paper was cut into 1.0 Â 6.0 cm strips each sample was transferred to the wells of a flat-bottom,
with a paper cutter for the standard 1.5 mL FPA and into 96-well plate containing 160 Al of H2O and the absorbance
circles with diameter of 7.0 mm with an office paper punch at 540 nm was measured in a plate reader. Relative to the
for the 96-well microplate assays. amount of glucose equivalents expected to be produced in
the standard FPA, the amount of enzyme used in the 96-Al
FPA should release about 128 Ag of glucose equivalents in
Reaction Volume of FPA in 96-Well Plate each well.
The ratio of the reaction volume to the area of the filter The 60-Al format assay was carried out as follows. A
paper strip used in the standard IUPAC FPA was main- 20-Al aliquot of diluted enzyme was added into wells con-
tained for our initial assessment of the 96-well assay plate- taining 40 Al of 50 mM NaAc buffer and a filter paper disk.
based method. In the standard FPA, 0.5 ml of enzyme was After 60 min of incubation at 50jC, 120 Al of DNS was
added into 1.0 ml of 50 mM NaAc buffer, pH 4.8, con- added into each reaction and incubated at 95jC for 5 min.
taining a filter paper strip (1 Â 6 cm = 600 mm2), while the Finally, a 36-Al aliquot of each sample was transferred to
7-mm diameter paper disk, whose area is 38.5 mm2, was the wells of a flat-bottom plate containing 160 Al of H2O,
used in the 96-well plate assay. Accordingly, the reaction and the absorbance at 540 nm was measured.
volume for assays done in 96-well plates was calculated as
(1.0 ml buffer + 0.5 ml enzyme) H 600 mm2 Â 38.5 mm2 =
0.096 ml = 96 Al. Calculation of Filter Paper Units (FPU)
Under the IUPAC recommended assay conditions, one
Determination of Evaporative Loss FPU is defined as an average of one Amole of glucose
equivalents released per min in the assay reaction. FPU/ml
To determine the evaporative loss of the buffer during was calculated using the following formula:
60 min of enzymatic action at 50jC, 96 Al of 50 mM NaAc
buffer, pH 4.8, was added into each well of a 96-well PCR
A540 sample
plate (UltiDent Scientific, St. Laurent, Quebec, Canada). FPU ml ¼ ð5:55 lmole mgÞ
A540 =mg standard
The net weight of the buffer in the plate was measured
before and after incubation at 50jC. Then the evaporative 1 1
Â
loss of buffer was determined. 60 min Xml
XIAO ET AL.: MICROPLATE-BASED CELLULASE ASSAY 833

3. where FPU/ml is the determined cellulose activity in the glucoside as the substrate. A 20-Al aliquot of varying
suitably diluted sample; A540 sample is the absorbance amounts of diluted cellulase enzyme was mixed with 40 Al
obtained from the DNS assay for each cellulase assay; of 1.0 CBU/ml Novozyme 188 in 50 mM NaAc, pH 4.8.
A540/mg standard is the absorbance for 1 mg of glucose as Different ratios of h-glucosidase units (or cellobiase units,
derived from the glucose standard curve (for example, the CBU) to FPUs were supplemented into the assay. The
curve in Figure 1A was used to calculate the A540/mg assay without the addition of h-glucosidase was used as
standard used for the standard FPA assays); 5.55 Amole/ the control.
mg is the number of Amoles of glucose in 1 mg; 60 min
is the assay incubation time, and X ml is the volume of
suitably diluted cellulase that was assayed. In the standard, RESULTS AND DISCUSSION
96-Al and 60-Al FPA assays, X ml is 0.5 ml, 0.032 ml and
0.02 ml, respectively.
Modification of the Standard Filter Paper Assay
Initially, we wanted to use the same ratio of reaction
Supplement of h-Glucosidase volume to filter paper area for the microplate-based assay
h-Glucosidase activity was determined as described pre- as is used in the standard IUPAC FPA. For this, we began
viously (Wood and Bhat, 1988) using p-nitrophenyl-h-D- by determining the weight of the 1 Â 6 cm filter paper
strips used in the standard FPA and the 7-mm circles to be
used in the microplate assay. The average weight of 100
Whatman No. 1 filter paper strips was 52.3 F 1.3 mg, with
2.5% of coefficient of variation, whereas the average
weight of 100 filter paper circles was 3.42 F 0.28 mg, with
8.2% of coefficient of variation. Since the average weight
of the filter paper substrate is directly proportional to its
area, we initially used 96 Al reaction volumes for assays
done in 96-well assay plates. This enabled us to maintain
the same ratio of substrate, enzyme, buffer, DNS reagent,
and water for the 96-Al format assay as we used in the
standard FPA. In the standard FPA, all components are
added into one 25-ml test tube, where the enzymatic hy-
drolysis, the color development with DNS reagent, and
dilution of the colored solution take place before the ab-
sorbance is measured. In the 96-Al FPA, due to the well
capacity, it was necessary to sequentially transfer the re-
action products into two other microplates to develop color
and then measure the absorbance at 540 nm.
Reducing Evaporative Loss
Previous results found that there was about 45% evap-
orative loss of the water from the wells of a 96-well
plate after 60 min of incubation at 50jC without a lid
on the plate (Decker et al., 2003). In order to minimize
evaporation during incubation we incubated the 96-well
PCR plate in a temperature cycler, which has a built-in
105jC hot lid to prevent evaporation. A plastic mat was
also used to cover the PCR plate. The combination of
hot lid and plastic mat reduced the evaporative loss to
only 4.0%.
Glucose standard curves for the IUPAC method are
generated using samples that are generally not subjected
to the 50jC incubation step. To offset the effect of evap-
orative loss on enzyme activity estimates, the glucose stan-
Figure 1. Standard DNS curves for measuring the release of glucose
during FPAs. The absorbance obtained with different amounts of glucose
dard curve was plotted using a series of glucose standards
was measured using a plate reader. The best-fit lines are shown. A: Curve that were also subjected to incubation in the temperature
for the standard FPA. B: Curve for the 96-Al FPA. cycler at 50jC for 60 min.
834 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 88, NO. 7, DECEMBER 30, 2004

4. Determination of Cellulase Activity Using Standard reagents were adjusted proportionally to that of the 96-Al
FPA and 96-Ml FPA reactions. Figure 2 shows that the standard curves obtained
The release of glucose equivalents was determined using for assays performed in both 60 Al and 135 Al reaction vol-
a standard curve produced using glucose samples having umes were linear.
concentrations of 0, 2, 4, 6, 8, 10 mg/ml in 50 mM NaAc For the 135-Al reactions, the measured enzyme activity
buffer, pH 4.8. The absorbance at 540 nm versus glucose was 20% lower than that of 96-Al FPA. However, when the
amount in the reaction was plotted using the standard for- reaction volume was reduced to 60 Al, the determined
mat (25-ml glass tubes) and microplate format methods enzyme activity was not significantly different from that
(Fig. 1). The data indicate that both methods accurately obtained with the 96-Al FPA (Table II). These results sug-
measure the quantity of glucose over the concentration gest that the amount of substrate was rate-limiting where
range tested. the reaction volume was 135 Al. Hence, increasing the
To initiate a comparison of the standard assay and our amount of enzyme used, while keeping the substrate quan-
modified assay, the C-2730 cellulase was diluted 100-fold tity constant, resulted in reduced assay sensitivity. On the
in 50 mM NaAc buffer (pH 4.8) and divided into 30 sam- other hand, decreasing the amount of enzyme used by about
ples. These 30 cellulase samples were separated into two 40%, the 60-Al reaction did not adversely affect the ability
equal groups, one of which was assayed using the standard
FPA and the other with the 96-Al FPA. The results obtained
are presented in Table I. A t-test performed on the data
showed that there was no significant difference at the 95%
confidence level between the units of enzyme activity de-
termined with the standard FPA and the 96-Al FPA. Our
96-Al FPA carried out in 96-well PCR plates therefore
accurately measured cellulase activity levels.
Effects of Different Reaction Volumes on FPA
When 7.0-mm diameter filter paper disks are used as the
substrate, a reaction volume of 96 Al maintains the same
ratio of the reaction volume to the amount of substrate as is
used in the standard IUPAC assay. We next examined the
effect of different reaction volumes on the sensitivity of the
assay. Using the 7-mm diameter filter paper disks as
substrates, we performed the assays in reaction volumes of
60 Al and 135 Al. The volumes of enzyme and other
Table I. Trichoderma cellulase C-2730 activity determined using the
standard FPA and the 96-Al FPA.
Sample Standard FPA (FPU/ml) 96-Al FPA (FPU/ml)
1 45.6 45.3
2 44.2 43.6
3 50.2 44.7
4 43.2 42.8
5 52.5 61.8
6 42.6 41.8
7 51.0 41.7
8 42.6 39.5
9 42.4 43.5
10 41.9 61.6
11 43.0 57.5
12 44.5 60.1
13 43.3 48.1
14 45.2 47.2
15 48.1 45.8
Average 45.4 F 3.3 48.3 F 7.5 Figure 2. Standard DNS curves for measuring the release of glucose
in the 60-Al FPA and the 135-Al FPA. The absorbance obtained with dif-
T-test: The data obtained from the 96-Al FPA were compared to those ferent amounts of glucose was measured using a plate reader at 540 nm.
obtained from the regular standard FPA. The calculated t-value is 1.32, The best-fit lines are shown. A: Curve for the 135-Al FPA. B: Curve for
which is below the critical t-value of 2.05 at P = 0.05 and DF = 28. the 60-Al FPA.
XIAO ET AL.: MICROPLATE-BASED CELLULASE ASSAY 835

5. Table II. Effects of reaction volumes on the measurement of cellu- Table IV. Comparison of Trichoderma C-2730 activity measured using
lase activity. the standard FPA and the 60-Al FPA.
96-Al FPA 135-Al FPA 60-Al FPA Sample Standard FPA (FPU/ml) 60-Al FPA (FPU/ml)
Sample (FPU/ml) (FPU/ml) (FPU/ml)
1 45.6 44.4
1 58.0 42.2 51.7 2 44.2 49.7
2 58.8 41.0 54.5 3 50.2 45.0
3 48.5 45.3 48.6 4 43.2 52.6
4 50.4 45.9 48.5 5 52.5 52.6
5 51.0 48.9 52.4 6 42.6 48.5
6 51.5 40.4 47.6 7 51.0 41.3
7 55.9 40.5 56.3 8 42.6 44.7
8 51.6 30.1 46.0 9 42.4 44.9
Average 53.2 F 3.6 41.8 F 5.2 50.7 F 3.4 10 41.9 45.9
11 43.0 43.9
T-test: The data obtained from the 96-Al FPA were compared in- 12 44.5 45.3
dependently with data obtained from the 135-Al FPA and the 60-Al PFA. 13 43.3 46.4
The t-value for 35-Al FPA is calculated as 4.75, which is above the critical 14 45.2 48.1
t-value of 2.14 at P = 0.05 and DF = 14. However, the t-value for the 60-Al 15 48.1 52.2
FPA is 1.35, which is below the critical t-value. Average 45.4 F 3.3 47.0 F 3.5
T-test: calculated t-value = 1.27, which is below the critical t-value of
2.05 at P = 0.05 and DF = 28.
of the assay to accurately determine the amount of cellulase
activity present.
the variation in results. We used filter paper disks with an
60-ML FPA Is Highly Reproducible
average weight of 3.4 mg, while those used by Decker et al.
In the 96-Al format FPA, we had to transfer a portion of the (2003) had an average weight of 2.65 mg. Since FPA is
reaction mixture to a second 96-well plate for the color based on a nonlinear reaction and the activity estimates are
development step. The 60-Al format assay can accommo- influenced by the size of the filter paper, the use of different
date the 120 Al of DNS reagent required to perform the filter paper disks as substrate may lead to variation in
color development step in the same 96-well plate without results. Moreover, the use of a higher temperature and
the necessity of transferring the reaction mixtures to a fresh incubation time in color development, 98jC for 10 min by
microplate. Table III showed the comparison of the com- Decker et al. (2003) versus 95jC for 5 min in our
ponents of the filter paper assays performed in 96-well procedure, may have exaggerated the enzyme activity mea-
plates and standard FPA. sured by their miniaturized assay.
Recently, Decker et al. (2003) reported a miniature-scale To verify the utility of the 60-Al assays, we performed
FPA. However, the detectable filter paper activities were assays in this format along with the standard FPA. Table IV
56% higher than that obtained with standard FPA (Decker shows that the determined enzyme activity with the 60-Al
et al.). We are not clear why their miniaturized FPA yielded FPA was comparable to that obtained with the standard
different results from ours and from the standard FPA. Two FPA. Another commercial cellulase preparation, C-2605
discernable differences in the procedures may contribute to from Aspergillus species, was also assayed using the 60-Al
Table III. Comparison of the components of the filter paper assays in different reaction volumes.
Standard FPA 96-Al FPA 60-Al FPA
2 2
Filter paper substrate 600 mm (52 mg) 38.5 mm (3.4 mg) 38.5 mm2 (3.4 mg)
Buffer volume 1.0 ml 64 Al 40 Al
Enzyme (or 0.5 ml 32 Al 20 Al
glucose standard
volume)
Enzyme reaction at Test tubes in 96-well microplates 96-well microplates
50jC for 60 min waterbath in temperature in temperature
cycler cycler
Volume of DNS 3.0 ml 50 Al reaction mixture + 120 Al
100 Al DNS
Color development 100jC for 5 min 95jC for 5 min 95jC for 5 min
Amount of H2O 20 ml 160 Al to 36 Al of 160 Al to 36 Al of
added prior to the reaction/DNS the reaction/DNS
measurement mixture mixture
836 BIOTECHNOLOGY AND BIOENGINEERING, VOL. 88, NO. 7, DECEMBER 30, 2004

6. Table V. Comparison of Aspergillus cellulase C-2605 activity using the format is a robust assay that can rapidly and accurately mea-
standard and the 60-Al FPA. sure the cellulase content of different enzyme preparations.
Sample Standard FPA (FPU/ml) 60-Al FPA (FPU/ml)
1 4.29 4.10 CONCLUSIONS
2 4.07 4.26
3 4.35 3.95 We have developed a 60-Al format cellulase assay using
4 4.11 3.18 filter paper disks as the substrate. This highly reproducible
5 3.86 4.11 and accurate assay can be used to rapidly and easily assay
6 3.89 3.79 a large number of samples. It also reduces the amount
7 3.99 3.54
of chemicals and enzyme used by 25-fold relative to the
8 4.00 3.62
9 4.21 4.39 standard IUPAC method. As in the standard method, sup-
10 4.10 4.01 plementing the enzyme with h-glucosidase dramatically
11 4.21 3.85 increases assay sensitivity. The microplate-based assay de-
12 3.97 4.13 veloped here therefore provides a useful technology plat-
13 3.95 4.66
form for the discovery and characterization of cellulases for
14 3.97 4.28
15 3.87 3.62 industrial application.
Average 4.06 F 0.15 3.97 F 0.38
We thank Dr. Xiao Zhang for comments and suggestions.
T-test: calculated t-value = 0.86, which is below the critical t-value of
2.05 at P = 0.05 and DF = 28.
References
Bowman GR, Beauchemin KA, Shelford JA. 2002. The proportion of the
FPA to verify its reproducibility and accuracy (Table V). diet to which fibrolytic enzymes are added affects nutrient digestion
Statistical analysis of the assay results showed that the 60-Al by lactating dairy cows. J Dairy Sci 85:3420 – 3429.
format assay accurately measured the cellulase content of Breuil C, Mayers P, Saddler J. 1986. Substrate conditions that influence
the assay used for determining h-glucosidase activity of cellulolytic
the C-2605 cellulase sample.
microorganisms. Biotechnol Bioeng 28:1653 – 1656.
Chen K, Arnold FH. 1993. Tuning the activity of an enzyme for un-
usual environments: sequential random mutagenesis of subtilisin E
Supplement of h-Glucosidase Increases the Cellulase
for catalysis in dimethylformamide. Proc Natl Acad Sci U S A 90:
Assay Sensitivity
5618 – 5622.
Breuil et al. (1986) reported that most natural cellulase Cherry JR, Fidantsef AL. 2003. Directed evolution of industrial enzymes:
an update. Curr Opin Biotechnol 14:438 – 443.
complexes tend to have a shortage of h-glucosidase
Coward-Kelly G, Aiello-Mazzari C, Kim S, Granda C, Holtzapple M.
activity. By adding an excess amount of cellobiase (h- 2003. Suggested improvements to the standard filter paper assay used
glucosidase) into the cellulase Spezyme CP produced by to measure cellulase activity. Biotechnol Bioeng 82:745 – 749.
Genecor, Coward-Kelly et al. (2003) showed that the filter Decker SR, Adney WS, Jennings E, Vinzant TB, Himmel ME. 2003.
paper assay response was increased by 56%. These results Automated filter paper assay for determination of cellulase activity.
Appl Biochem Biotechnol 105 – 108:689 – 703.
suggested that extra h-glucosidase facilitated the conver-
Galbe M, Zacchi G. 2002. A review of the production of ethanol from
sion of cellobiose into glucose. softwood. Appl Microbiol Biotechnol 59:618 – 628.
We measured the amount of h-glucosidase activity in the Garcia O, Torres AL, Colom JF, Pastor F, Diaz P, Vidal T. 2002. Effect of
two commercial cellulase preparations using p-nitrophenyl- cellulase-assisted refining on the properties of dried and never-dried
h-D-glucoside as substrate. The ratio of h-glucosidase units eucalyptus pulp. Cellulose 9:115 – 125.
Miller GL. 1959. Use of dinitrosalicylic acid reagent for determination of
(cellobiase unit or CBU) to filter paper unit (FPU) for the
reducing sugar. Anal Chem 31:426 – 428.
Trichoderma reesei cellulase (C-2730) was about 1.0. No Niehaus F, Bertoldo C, Kaehler M, Antranikian G. 1999. Extremophiles as
detectable h-glucosidase activity was observed in the As- a source of novel enzymes for industrial application. Appl Microbiol
pergillus cellulase (C-2605). We used the 60-Al format as- Biotechnol 51:711 – 729.
say to evaluate the effect of addition of h-glucosidase on Stemmer WP. 1994. Rapid evolution of a protein in vitro by DNA shuf-
fling. Nature 370:389 – 391.
the FPA. The addition of excess h-glucosidase to the FPA
Sun Y, Cheng J. 2002. Hydrolysis of lignocellulosic materials for ethanol
increased the amount of cellulase measured by 55% and production. Bioresour Technol 83:1 – 11.
60%, respectively, with the C-2730 and C-2605 cellulase Wood TM, Bhat KM. 1988. Methods for measuring cellulose activities.
preparations. These results further indicate that the 60-Al Methods Enzymol 160:87 – 112.
XIAO ET AL.: MICROPLATE-BASED CELLULASE ASSAY 837