Bioresource Technology 96 (2005) 491–499

Production of cellulase by Trichoderma reesei from dairy manure

Zhiyou Wen *, Wei Liao, Shulin Chen
Department of Biological Systems Engineering, Center of Multiphase Environmental Research, Washington State University, Pullman, WA 99164, USA
Received 10 October 2003; received in revised form 10 May 2004; accepted 12 May 2004

Abstract
Cellulase production by the fungi Trichoderma reesei was studied using dairy manure as a substrate. Data showed that T. reesei
RUT-C30 had higher cellulase production than T. reesei QM 9414 and that a homogenized manure, treated by a blender to reduce
fiber size, led to higher cellulase production. The cellulase production was further optimized by growing T. reesei RUT-C30 on
homogenized manure. The effects of manure concentration, pH, and temperature on cellulase production were investigated with
optimal parameter values determined to be 10 g/l manure (dry basis), 25.5 °C, and pH 5.7, respectively. Elimination of CaCl2 ,
MgSO4 , nitrogen sources (NHþ 2þ 2þ
4 and urea) and trace elements (Fe , Zn , Co
2þ
and Mn2þ ) from the original salt solution had no
negative influence on the cellulase production, while phosphate elimination did reduce cellulase production. Based on above results,
the final medium composition was simplified with manure additives being KH2 PO4 , tween-80 and CoCl2 only. Using this medium
composition and a reaction time of 6–8 days, a maximum cellulase production activity of 1.74 IU/ml of filter paper activity, 12.22
IU/ml of CMCase activity, and 0.0978 IU/ml of b-glucosidase was obtained. This filter paper activity is the highest ever reported in
cellulase production from agricultural wastes.
Ó 2004 Elsevier Ltd. All rights reserved.

Keywords: Cellulase; Trichoderma reesei; Animal manure; Lignocellulose; Value-added products

1. Introduction 2002), while limited efforts have been focused on utiliz-

Environmental friendly disposal and utilization of
animal manure are significant challenges to the livestock
industries. It is estimated that about 160 million tons of
animal manure (dry basis) are produced annually in the
United States alone (Council for Agricultural Science
and Technology, 1995). Currently, the disposal of the
manure is predominately through direct land applica-
tion. This management practice, however, is coming
under environmental and regulatory scrutiny due to the
limited amount of land available for manure disposal,
which may result in surface and ground water contam-
ination.
Animal manure represents a large potential bio-
resource for producing bio-based chemicals, materials,
and energy. There have been some reports about man-
ure nutrients recovery (mainly N and P) by micro-
organisms (Potter et al., 2001; Wilkie and Mulbry,
*
Corresponding author. Tel.: +1-509-335-6239; fax: +1-509-335-
2722.
E-mail address: zwen@wsu.edu (Z. Wen).
ing the lignocellulosic components of manure.
One possible approach to manure lignocellulose uti-
lization is to hydrolyze the materials into fermentable
saccharides, which can then be converted into value-
added products or bioenergy, as in the case dairy man-
ure anaerobic digestion (AD). A typical AD process,
however, is limited by the microbes’ inability to utilize
the fiber as the lignocellulosic component of the fiber is
very resistant to biodegradation (Colberg, 1988; Nielsen
et al., 2004). To effectively convert lignocellulose into
reducing sugars, commercial cellulase enzymes could be
used (Wen et al., 2004). The process, however, is con-
sidered non-economical because the cost of commercial
cellulase enzymes remains very high (von Sivers and
Zacchi, 1995). If cellulase could be produced directly
from manure though, and then be applied to AD to
further degrade the manure cellulose, the cost of cellu-
lase could be significantly reduced which at the same
enhancing the methane yield and potential profitability
of AD of dairy manure.
As a potentially less expensive alternative, cellulolytic
enzyme could be produced by a number of bacteria and

0960-8524/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2004.05.021
492 Z. Wen et al. / Bioresource Technology 96 (2005) 491–499
fungi. The cellulolytic fungi Trichoderma viride and
Trichoderma reesei have been extensively studied for
their cellulase production (Mandels and Weber, 1969;
Mandels et al., 1971; Montenecourt and Eveleigh, 1979;
Gadgil et al., 1995; Velkovska et al., 1997; Domingues
et al., 2000). To enhance the cellulase titer, various
mutants of Trichoderma have been developed, among
which T. reesei RUT C30 is of industrial interest because
of its high cellulase production level (Montenecourt and
Eveleigh, 1979) as well as its ability to grow on waste
cellulosic material (Reczey et al., 1996; Ju and Afolabi,
1999; Domingues et al., 2000).
Cellulolytic fungi can use cellulose as a primary car-
bon source. Pure, crystalline cellulose, such as Solka
Floc, Avicel, and cotton are good cellulase inducers, but
are expensive. To keep costs down it is therefore
important to use a substrate that is less expensive. Many
cellulosic materials such as wood (Duff and Murray,
1996; Reczey et al., 1996), wastepaper (Ju and Afolabi,
1999), bagasse (Rajoka and Malik, 1997; Ogel et al.,
2001), wheat straw (Romero et al., 1999; Kalogeris et al.,
2003), corncob (Xia and Cen, 1999), wheat bran (Smits
et al., 1996), and fruit pomace (Umikalsom et al., 1997a;
Haddadin et al., 2001) have been studied as potential
substrates for production of cellulase. There is, however,
a lack of investigation on the cellulase production from
manure cellulosics. The aim of the present work is to
study the potential of using animal manure for cellulo-
lytic enzyme production by the fungi T. reesei The
information would be useful for the development of a
cost-effective process for cellulase production and sub-
sequent enzymatic hydrolysis of manure lignocellulose.
2. Methods
2.1. Manure collection and characterization
Fresh manure was collected from the Dairy Center at
Washington State University in Pullman, WA and
stored for later use in a freezer. The diet of the cows
consisted of (DM/animal/day): 17 lbs alfalfa hay, 16 lbs
alfalfa haylage, 7 lbs cottonseed, 7 lbs wheatmill run,
and 20 lbs grain. The grain portion included 32% corn,
19% wheat, 17% barley, 15% peas, 4.5% soybean meal,
4.5% corn gluten and 8% other additives such as
molasses, limestone, sodium bicarbonate, and vitamins.
The manure was mixed with water (ratio 2:1, w/w) and
homogenized by an Osterizerâ blender (Model 6698,
Sunbeam Product Inc., Boca Raton, FL). The homo-
genized samples were used for analyzing various nutri-
ents including total carbon, total nitrogen, ammonium,
phosphorus, potassium, calcium, magnesium, sodium,
sulfur, and trace elements (iron, manganese, zinc, co-
balt, copper). Also, dry matter (DM) and lignocellulose
content including neutral-detergent fiber (NDF), acid-
detergent fiber (ADF), and acid-detergent lignin (ADL)
were analyzed. All these parameters were then converted
to those of raw manure (as it was collected) by con-
sidering volume changes.
2.2. Microorganism, medium and culture conditions
Two fungal strains, T. reesei RUT-C30 (ATCC 56765)
and T. reesei QM 9414 (ATCC 26921) were used in the
study. The fungi were maintained in potato dextrose agar
slants at 4 °C. To prepare the inoculum, the spores in the
slant were suspended in 2 ml medium (106 –107 spores/ml)
and transferred into a 250 ml Erlenmeyer flask contain-
ing 50 ml of medium. The subculture medium was a salt
solution with 2 ml/l tween-80, 1 g/l peptone, and 10 g/l
glucose added (Table 1). The initial pH of the medium
was adjusted to 4.8 before being autoclaved at 121 °C for
15 min. Fungal cells were sub-cultured in an orbital
shaker (175 rpm) at 30 °C for 1–2 generations (two days
each generation) and then used for inoculum.
Five milliliter (5 ml) of exponential cells was inoculated
into 50 ml of medium containing manure as a substrate.
In the feasibility study (Section 3.2), both untreated
manure (as it was collected) and homogenized manure (as
it was analyzed for characterization) were used and
manure concentration was adjusted to 6.7 g/l (dry basis).
After the effects of manure concentration were studied
(Section 3.3.1), 10 g/l (DM) of homogenized manure was
used in the following studies (Sections 3.3.2 and 3.3.3).
The medium composition was the same as the subculture
medium (Table 1), except that peptone was eliminated
and glucose was replaced with manure (Table 1).
2.3. Analysis
Manure dry matter (DM) was determined by drying
the samples at 105 °C for
48 h or until the weight
became constant. NDF, ADF, and ADL were deter-
mined using the gravimetric method (Goering and Van
Soest, 1970). The apparent values of ADF and ADL
were corrected by subtracting acid-detergent nitrogen,
which was also determined by the gravimetric method
(Goering and Van Soest, 1970). Total carbon and total
nitrogen were measured using automatic combustion
(LECO CNS-2000). Ammonium was determined by
titrimetric method (Eaton et al., 1995). EPA method
3050/6010 was used to analyze other elements (potas-
sium, phosphorus, calcium, magnesium, sodium, sulfur,
iron, manganese, zinc, cobalt, copper). Here, the 3050
method is a nitric/hydrochloric acid digest and 6010
indicates metal analysis by inductively coupled argon
plasma spectroscopy (ICP).
The activities of total cellulase (filter paper activity,
FPA), endo-b-1,4 glucanase (CMCase), and b-glucosi-
dase were determined according to standard IUPAC
procedures and expressed as an international unit (IU)
 Z. Wen et al. / Bioresource Technology 96 (2005) 491–499 493

Table 1
Medium composition for subculture and cellulase production of the fungi T. reesei
Components Unit Concentration
Subculture Cellulase productiona Cellulase productionb
Salt solution
KH2 PO4 g/l 2.0 2.0 2.0
CaCl2 Æ 2H2 O g/l 0.4 0.4 –
MgSO4 Æ 7H2 O g/l 0.3 0.3 –
(NH4 )2 SO4 g/l 1.4 1.4 –
Urea g/l 0.3 0.3 –

Trace elements
FeSO4 Æ 7H2 O mg/l 5.0 5.0 –
MnSO4 Æ H2 O mg/l 1.6 1.6 –
ZnSO4 Æ 7H2 O mg/l 1.4 1.4 –
CoCl2 mg/l 2.0 2.0 2.0

Tween-80 ml/l 2.0 2.0 2.0

Peptone g/l 1.0 – –
Glucose g/l 10 – –
Manure (DM) g/l – 6.7 or 10 10
a
The medium composition was used for feasibility study (Section 3.2, manure concentration: 6.7 g/l); effects of manure concentration (Section
3.3.1), effects of temperature and pH (Section 3.3.2, manure concentration: 10 g/l), and nutrients elimination test (run 4 of Section 3.3.3, manure
concentration: 10 g/l).
b
This is the nutrient-reduced medium after nutrient elimination test (Section 3.3.3) was finished, this composition was used to obtain the result of
Fig. 4.

(Ghose, 1987). One unit of FPA and CMCase activity
were defined as the amount of enzymes which release 1
lmol of glucose equivalents from Whatman No. 1 filter
paper and carboxymethyl cellulose (CMC) in 1 min,
respectively. One unit of b-glucosidase activity was de-
fined as the amount of enzyme converting 1 lmol of
cellobiose to produce 2 lmol of glucose in 1 min. The
glucose concentrations in the cellubiose hydrolysates
were measured using the enzyme assay kit from Sigma
(Product No. GAGO-20).
2.4. Experimental design and data analysis
A central composite design (CC0211) was used to
optimize temperature (T ) and pH for cellulase produc-
tion (Haaland, 1989). The central composite design
matrix was a 22 factorial design combined with one
central points and four axial points where one variable
was set at an extreme level (±1.41) while the other var-
iable was set at its central point (Table 2). The true
values of the variables are also given in Table 2. Coding
of variable i was done as follows:
xi ¼ ðXi  Xcp Þ=DXi ; i ¼ 1; 2;
where xi is the coded level, Xi is the real value, Xcp is the
real value at central point, and DXi is the step change of
variable i, respectively.
For cellulase production, the filter paper activity can
be written as the function of the independent variables
by second-polynomial, i.e.,
Y ¼ a0 þ a1  x1 þ a2  x2 þ a11  x21 þ a22  x22 þ a12  x1  x2 ;
ð1Þ

Table 2
Central composite design of temperature (T ) and pH with the filter paper activity (FPA) as response
Runa Variables Response
Coded unit Real value FPA (IU/l) Coefficient of
variation (%)
T pH T (°C) pH
1 )1 )1 22 3.8 0.083 ± 0.004 4.80
2 )1 +1 22 5.8 1.387 ± 0.024 1.73
3 +1 )1 32 3.8 0.085 ± 0.002 2.35
4 +1 +1 32 5.8 0.869 ± 0.031 3.56
5 )1.41 0 20 4.8 0.743 ± 0.019 2.56
6 +1.41 0 34 4.8 0.087 ± 0.002 2.29
7 0 )1.41 27 3.4 0.061 ± 0.001 1.69
8 0 +1.41 27 6.2 1.263 ± 0.048 3.80
9 0 0 27 4.8 1.258 ± 0.020 1.60
a
Runs 1–8 were performed in duplicate, while run 9 (central point) was performed in triplicate.
494 Z. Wen et al. / Bioresource Technology 96 (2005) 491–499

where Y is the predicted response (filter paper activity), Table 3

x1 and x2 are the coded levels of the two variables (T and Major composition of freshly collected dairy manurea
pH), and a0 , a1 , a2 , a11 , a22 , a12 are coefficients. This Dry matter (DM) 14.60 ± 0.25% (w/w)
equation was used to evaluate the linear, quadratic and Lignocellulosics (% of DM)
interactive effects of T and pH. NDF 49.10 ± 1.30
ADFb 37.83 ± 1.01
The response and variables (in coded unit) were
ADLb 11.24 ± 1.02
correlated by the ‘‘Response Surface Analysis’’ function Hemicellulose (NDF-ADF) 11.27 ± 0.90
of the NCSS software (NCSS Statistical Software Inc., Cellulose (ADF-ADL) 26.59 ± 0.28
UT, 2000) to obtain the coefficients of Eq. (1). The Lignin (ADL) 11.24 ± 1.02
correlation coefficient (R2 ) and significance of the model Elements (% of DM)
were also obtained by the NCSS software. The signifi- Carbon 50.51 ± 1.22
cance of the model was evaluated by F -test, with a P Total nitrogen 3.03 ± 0.58
(probability) value less than 0.05 being regarded as sig- NH4 -N 0.44 ± 0.029
Potassium 1.24 ± 0.017
nificant.
Phosphorus 0.810 ± 0.054
The central point (run 9) was repeated three times to Calcium 2.41 ± 0.184
obtain the standard deviation by using Microsoft Excel Magnesium 0.966 ± 0.061
2000. Although it has been reported that the standard Sodium 0.243 ± 0.019
deviation could be used to evaluate the experimental Sulfur 0.496 ± 0.021
Iron 0.134 ± 0.012
error of the design (Haaland, 1989; Montgomery, 1997),
Manganese 0.015 ± 0.001
consideration that the deviation may vary under differ- Zinc 0.013 ± 0.001
ent conditions required us to perform runs 1–8 in Cobalt 0.0002 ± 0.000
duplicates and the deviation of each individual run was Copper 0.0046 ± 0.000
estimated. Acid-detergent nitrogen (1.57% of DM) was formed via the non-
enzymatic browning reaction when nitrogen-enriched manure is heated
above 50 °C (Goering and Van Soest, 1970). The value of acid-deter-
gent nitrogen was determined by the gravimetric method (Goering and
3. Results and discussion
Van Soest, 1970).
a
Data is expressed as mean ± SD of three replicate samples.
3.1. Manure characterization b
Data of ADF, ADL listed are the true values corrected by sub-
tracting acid-detergent nitrogen from the apparent values (Goering
The chemical characteristics of dairy manure are and Van Soest, 1970).
given in Table 3. Raw dairy manure contained 14.6%
dry matter, half of which are lignocellulosics (hemicel- 1996; Ju and Afolabi, 1999; Domingues et al., 2000). In
lulose, cellulose and lignin). In terms of element com- this work, the two mutants were respectively grown in
ponents, carbon is the most abundant, followed by medium containing manure as substrate. Both the un-
nitrogen, calcium and potassium. The manure also treated manure and homogenized manure were used.
contained phosphorus, magnesium, sodium, sulfur, iron, It was found that both of the fungi could produce
manganese, zinc, cobalt and copper, comprising less cellulase in medium containing untreated or homoge-
than 1% of dry weight. These components are notably nized manure (Fig. 1A). The un-treated manure resulted
ideal nutrients for fungi culture. Some microorganisms in a lower FPA than the homogenized manure. The
have been used to recover the manure nutrients (such as average length of fiber size was about 10 mm in un-
nitrogen and phosphorus) to produce nutritional bio- treated manure and less than 2 mm for homogenized
mass (Mulbry and Wilkie, 2001; Wilkie and Mulbry, manure (data not shown). One possible explanation is
2002) or value-added product (Potter et al., 2001). that in the longer fiber, the accessibility of fungi cells to
However, the utilization of manure lignocellulosics, a cellulose was lower due to reduced specific surface area;
major composition of the manure, was rarely attempted. ultimately leading to lower cellulase production (Umi-
kalsom et al., 1997a; Romero et al., 1999).
3.2. Feasibility of using manure as a substrate for cellulase The time courses of cellulase production by T. reesei
production in medium containing homogenized manure are shown
in Fig. 1B. The patterns of cellulase production were
Manure has cellulosic components that can induce similar for the two mutants. Cellulase activity increased
the production of cellulase when used as carbon sources during the first five days, reached the maximum level
for fungi growth. Of all the celluloytic fungi, T. reesei between day 5 and day 8, and then decreased at the end
has been the most extensively studied, with the mutants of cultivation.
T. reesei RUT-C30 and T. reesei QM 9414 having been The results in Fig. 1 also show that the mutant T.
identified as possessing improved filter paper activity reesei RUT-C30 produced higher cellulase activity than
(Gadgil et al., 1995; Smits et al., 1996; Reczey et al., T. reesei QM-9414, and in addition, the homogenized
 Z. Wen et al. / Bioresource Technology 96 (2005) 491–499 495

1 1.5

RUT-C30
0.8 1.2
QM-9414

FPA (IU/ml)
FPA (IU/ml)

0.9
0.6

0.6
0.4

0.3
0.2
0
0 3.35 6.7 10 13.38 16.75
Un-treated manure Homogenized manure (A) Manure concentration (g DM/L)
(A) Different manures
800
1
700

Yield (IU/g cellulose)

600
0.8
500
FPA (IU/ml)

0.6 400

300
0.4
200

0.2
0 (B)
0 2 4 6 8 10
(B) Time (day)
Fig. 1. Cellulase production by two mutants of T. reesei using different
dairy manures as a substrate. (A) Filter paper activity (FPA) after six
days of cultivation in medium containing untreated manure and
homogenized manure. (B) Time course of FPA in medium containing
homogenized manure. h, T. reesei RUT-C30; n, T. reesei QM 9414.
Data are means of three replicates and error bars show standard
deviation.
manure resulted in higher cellulase activity as well. The
fungi T. reesei RUT-C30 and homogenized manure were
therefore, used for optimizing fungi cellulase produc-
tion.
3.3. Optimization of cellulase production by T. reesei
RUT-C30
3.3.1. Effects of manure concentration
Cellulase production by T. reesei RUT-C30 investi-
gated at different manure concentrations showed that
FPA increased with manure concentration from 3.35 to
10 g/l (dry basis), with the highest FPA of 1.2 IU/ml
being obtained for the range from 10 to 13.38 g/l of
manure (Fig. 2A). FPA decreased when manure con-
centration was over 13.38 g/l. The variation of manure
100
0
3.35 6.7 10 13.38 16.75
Manure concerntation (g DM/L)
Fig. 2. Effects of manure concentration on (A) cellulase production
and (B) yield of cellulase based on cellulose in the medium. Data are
means of three replicates and error bars show standard deviation.
concentration resulted in different cellulose levels in the
media, therefore the cellulase yield based on added cel-
lulose (FPA per gram of cellulose added) was presented
(Fig. 2B). The highest cellulase yield, 708 IU/g cellulose,
was obtained at the lowest manure level. The yield
monotonically decreased with increasing manure/cellu-
lose concentration. The cellulase yield based on added
cellulose reflects the efficiency of the fungal culture,
however, as the substrate (manure cellulose) is a non-
value material, the total cellulase titer (total fermenta-
tion yield) was considered more important than the
efficiency.
It had been reported that high cellulose concentration
could result in a higher cellulase activity with specific
studies including Chaetomium globosum on fruit fiber
(Umikalsom et al., 1997b) and Neurospora crassa on
wheat straw (Romero et al., 1999). Additionally, Reczey
et al. (1996) also reported that the cellulase yield in-
creased with cellulose level in the medium when using
wood as a substrate. The results obtained in this work
were similar with those reports. However, the effects of
496 Z. Wen et al. / Bioresource Technology 96 (2005) 491–499

manure concentration on cellulase production did not

solely depend on the amount of cellulose, but also per-
haps depended on other nutrients or ions. For example,
the lower FPA at high manure concentration (16.75 g/l,
Fig. 2A) was probably attributed to the inhibitory effects
caused by high mineral salts and lignin content (Reczey
et al., 1996). Currently, we are trying to use scanning
electron microscopy technology to observe the fungi
growth status at different manure concentration to fur-
ther confirm this assumption.
Ultimately though, since 10 g/l (dry basis) was the
optimal level for FPA, this manure level was used in the
optimization of temperature and pH and well as for
elimination of the addition of external mineral salts to
the salt solution.

3.3.2. Effects of temperature and pH

The effects of temperature (T ) and pH on cellulase
production were investigated by a central composite
design. This statistical method has proved efficient in
the optimization of various fermentation processes
(Haaland, 1989). The cellulase activity of runs 1–8 were
the means of two duplicates, while the central point was
run in triplicate (run 9) and its standard deviation was
0.02 IU/ml. The results showed a high reproducibility as
the deviation of each run was less than 5%.
Filter paper activity in Table 2 was correlated as a
function of T and pH (coded value) (Eq. (1)) by the
NCSS 2000 software. The resultant equation was:
FPA ¼ 1:252  0:181  T þ 0:474  pH  0:401  T 2
 0:277  pH2  0:130  T  pH: ð2Þ
The correlation coefficient (R2 ) of the above equation
was 0.984. F -tests showed that the model had a signifi-
cance P < 0:01. All these indicated that the model was
reliable in reflecting the relationship between T and pH
with cellulase production.
Eq. (2) was then used for deriving the optimal values
of T and pH. Using the NCSS 2000 software, the exact
optimal T and pH values (in coded unit) were obtained
as )0.285 and 0.922, which corresponded to real values
of 25.5 °C and pH 5.7, respectively. Eq. (2) predicted the
maximum FPA as 1.51 IU/ml, and was verified by
the experimental data (1.59 IU/ml) obtained under the
optimal conditions.
The response surface of FPA as a function of T and pH
is shown in Fig. 3. The plot is hump shaped with a clear
peak within the experimental range investigated. It was
also found that the value of response did not fall steeply
when the variables changed slightly from their best values
(Fig. 3). This is a desired property for the fungal culture
process because it means that the process will be robust to
slight fluctuation in T and pH during operation.
The effects of temperature and pH on cellulase pro-
duction have been extensively studied, and the influences
Fig. 3. Three-dimension surface plot of filter paper activity (FPA)
versus temperature (T ) and initial medium pH.
depended on the fungal species used. Generally, the
optimal temperature for cellulase production by T.
reesei is within the range of 25–28 °C (Duff and Murray,
1996), which is in agreement with the results obtained in
this work. It is noticeable that the optimal pH for
enzymatic hydrolysis is 4.8 (Ghose, 1987), while the
cellulase production (FPA) at pH 4.8 was relatively low
(Table 2). At the same time, relatively high levels of
reducing sugars were detected under this pH level (data
not shown). The inhibition of end-products may be an
explanation for why cellulase activity is low at pH 4.8.
As a result, it has been suggested that the fungi should
be grown in sub-optimal pH values for enzymatic
hydrolysis in order to reduce the efficiency of cellulase
and thereby reduce the accumulation of extracellular
sugars and therefore mitigate the impact of catabolite
repression of enzyme synthesis (Sternberg, 1976).
3.3.3. Effects of eliminating nutrients in salt solution
The medium used in this work is a salt solution with
manure as substrate. As manure is a complex mixture
consisting not only of cellulosic materials but also other
elements (Table 3), it was logical to test if the nutrients
in the salt solution could be replaced by the corre-
sponding manure nutrients.
As shown in Table 4, each nutrient (or nutrients
group) contained in the salt solution was respectively
eliminated from the medium (runs 1–5). The medium
containing full nutrients was the control (run 6). The
cellulase production from each run was determined. It
was found that T. reesei produced much lower cellulase
in KH2 PO4 -eliminated medium (run 1). However, the
cellulase production in the other runs was almost the
same (runs 4 and 5) or even better (runs 2 and 3) than
 Z. Wen et al. / Bioresource Technology 96 (2005) 491–499 497

Table 4
Experimental design for eliminating various nutrients from salt solution in Table 1 and corresponding cellulase activitya
Run Nutrients FPA (IU/ml)
KH2 PO4 CaCl2 MgSO4 (NH4 )2 SO4 and Trace-elements
Urea
1 ) + + + + 0.797 ± 0.085
2 + ) + + + 1.740 ± 0.037
3 + + ) + + 1.710 ± 0.057
4 + + + ) + 1.610 ± 0.123
5 + + + + ) 1.580 ± 0.109
6 + + + + + 1.590 ± 0.066
a
The symbol ) represents the corresponding nutrient is eliminated from salt solution while + represents the corresponding nutrient is included in
salt solution.

that of control (Table 4). Further statistical analysis by
the software SAS (version 8.0) indicated that the dif-
ference between runs 3, 4, 5 and run 6 was non-signifi-
cant (p > 0:1), while the difference between runs 2 and
run 6 were significant (p < 0:02). In all experiments,
tween-80 was kept in the medium because the surfactant
makes the fungal membranes more permeable, and thus,
facilitates the fungal cellulase release into aqueous phase
(Duff and Murray, 1996).
A further analysis of the nutrient distribution in
manure and in the salt solution is presented in Table 5.
It was found that the amounts of calcium, magnesium,
iron, manganese and zinc contained in the manure are
much higher than those in salt solution. As T. reesei
could utilize those nutrients from manure, the elimina-
tion of nitrogen, calcium, magnesium and trace elements
from salt solution had no negative influence on the
cellulase production. It is noticeable that the cobalt level
in manure is low, suggesting cobalt is probably not a
crucial element for cellulase production. Although
nitrogen from manure is lower than that from the salt
solution, it may be sufficient for fungi culture, therefore,
elimination of a nitrogen source also had no negative
influence.
The contributions of potassium and phosphorus from
the manure were much lower than those from the salt
solution (Table 5). This may be the reason why cellulase
production in KH2 PO4 -eliminated medium was much
lower than the other cases; because potassium and
phosphorus within the manure are insufficient to sup-
port the cellulase production by the fungi. Another
reason may be that part of manure phosphorus is in the
form of organic phosphate and polyphosphates (for
example, phytate as contained in most corn grain),
which makes the utilization more difficult (Fontenot
et al., 1983; Henze et al., 1996; Moller et al., 2002).
A further experiment was conducted by growing the
fungi in medium containing manure (10 g/l) added with
KH2 PO4 (2 g/l), CoCl2 (2 mg/l), and tween-80 (2 g/l).
Nitrogen, calcium, magnesium, and trace elements
(except cobalt) were all eliminated from the salt solu-
tion. The medium containing manure added with full-
nutrients (as in salt solution in Table 1) was taken as
control. To give a detailed cellulase profile of the fungi
under this condition, the time course of filter paper
activity, CMCase activity, and b-glucosidase activity
were monitored, respectively. As shown in Fig. 4, the
three enzymes were in a similar pattern and increased in
parallel with incubation time. The filter paper activity
and b-glucosidase activity reached their highest levels at
day 6, while the highest CMCase production occurred
near day 8. It was also found that the activities of all
the three enzymes were higher than the control, sug-
gesting the medium with reduced nutrients could suffi-
ciently support a high cellulase production by T. reesei
(Fig. 4).

Table 5
The distribution of nutrients in manure and salt solutiona
Nutrient From manure From salt solution Total concentration in Nutrient ratio of manure
medium to salt solution
Calcium (g/l) 0.241 0.11 0.350 2.19:1
Magnesium (g/l) 0.097 0.029 0.126 3.34:1
Iron (mg/l) 13.4 1.0 14.4 13.4:1
Manganese (mg/l) 1.5 0.521 2.021 2.88:1
Zinc (mg/l) 1.3 0.317 1.617 4.10:1
Cobalt (mg/l) 0.02 0.908 0.928 0.02:1
Nitrogen (g/l) 0.303 0.436 0.739 0.69:1
Potassium (g/l) 0.124 0.573 0.697 0.22:1
Phosphorus (g/l) 0.081 0.456 0.537 0.18:1
a
The calculation was based on 10 g/l (DM) of manure and the composition of salt solution in Table 1.
498 Z. Wen et al. / Bioresource Technology 96 (2005) 491–499

0.20 2 16

0.16 1.6
12

β -glucosidase (IU/ml)

CMCase (IU/ml)
FPA (IU/ml)
0.12 1.2

8
0.08 0.8

4
0.04 0.4

0 0 0
0 2 4 6 8 10
Time (day)

Fig. 4. Time course of filter paper activity (FPA), CMCase activity, and b-glucosidase activity in medium containing manure added with KH2 PO4 ,
CoCl2 , and tween-80. Symbols: d, FPA; m, CMCase activity; j, b-glucosidase activity. The open symbols are the control medium containing
manure added with full-nutrients (as in salt solution in Table 1). Data are means of three replicates and error bars show standard deviation.

Table 6
Comparison of cellulase production by different fungal species and substrates
Fungal species Substrate FPA (IU/ml) Reference
Lignocellulosic substrate
Trichoderma reesei RUT C30 Steam-treated willow 0.66 Reczey et al. (1996)
Trichoderma reesei RUT C30 Steam-treated willow 1.55 Szengyel and Zacchi (2000)
Trichoderma reesei RUT C30 Wastepaper 0.30 Ju and Afolabi (1999)
Trichoderma reesei RUT C30 Dairy manure 1.72 This work
Chaetomium globosum Kunze Oil palm fruit fiber 0.95 Umikalsom et al. (1997b)
Neurospora crassa 4335 (cell-1) Wheat straw 1.33 Romero et al. (1999)
Scytalidium thermophilum 3-A Apple pomace 0.39 Ogel et al. (2001)
Scytalidium thermophilum 3-A Lentil bran 0.23 Ogel et al. (2001)
Scytalidium thermophilum 3-A Bagasse 0.21 Ogel et al. (2001)
Pure cellulose or reducing sugar
Trichoderma reesei QM 9414 Acid-swollen cellulose 0.54 Gadgil et al. (1995)
Trichoderma reesei RUT NG14 Acid-swollen cellulose 15 Montenecourt and Eveleigh
(1979)
Trichoderma reesei RUT C30 Acid-swollen cellulose 15 Montenecourt and Eveleigh
(1979)
Trichoderma reesei RUT C30 Solka floc (cellulose) 4.65 Velkovska et al. (1997)
Trichoderma reesei RUT C30 Solka floc (cellulose) 2.10 Domingues et al. (2000)
Trichoderma reesei RUT C30 Lactose 1.30 Domingues et al. (2000)
Trichoderma viride QM 6a Solka floc (cellulose)a 3.3 Mandels and Weber (1969)
a
0.2% peptone was added to the culture medium.

3.4. Comparison of cellulase production
The highest cellulase production value (based on filter
paper activity) achieved was 1.7 IU/ml and was obtained
in medium which contained 10 g/l (dry basis) homo-
genized manure supplemented with 2 g/l KH2 PO4 , 2 mg/
l CoCl2 , and 2 g/l tween-80 at pH 5.7 and 25.5 °C. An
overall comparison of cellulase production by different
fungal species and substrates is presented in Table 6.
Pure substrate (cellulose or lactose) resulted in a much
higher cellulase activity than lignocellulosic substrates,
however, considering the high cost of raw materials
used, lignocellulosic wastes are still preferred as a sub-
strate for cellulase production. Among different ligno-
cellulosic residues, the cellulase production obtained in
this work is higher than previous reports, suggesting
dairy manure is a good source for cellulase production.
4. Conclusion
The present work showed that dairy manure was a
suitable substrate for cellulase production by T. reesei.
The optimal culture conditions were determined as fol-
lows: solid fraction of pre-treated manure as substrate
medium containing 10 g/l manure (dry basis), 2 g/l
KH2 PO4 , 2 mg/l CoCl2 , and 2 g/l tween-80; and initial
medium pH of 5.7 and temperature of 25.5 °C. The filter
 Z. Wen et al. / Bioresource Technology 96 (2005) 491–499
paper activity under these conditions achieved 1.72 IU/
ml, which is much higher than results obtained using
other lignocellulosics residues (Table 6). Further work is
needed to enhance the production of b-glucosidase and
this is under way in our lab.
Acknowledgements
This work was supported by U.S. Department of
Energy (grant number: DE-FC36-01GO11048). The
authors gratefully acknowledge Pacific Northwest Na-
tional Laboratory (PNNL), Ms. Yan Liu and Mr. Frear
Craig for their kind suggestions and technical assistance.
References
Colberg, P.J., 1988. Anaerobic microbial degradation of cellulose,
lignin, oligonols, and monoaromatic lignin derivatives. In: Zehn-
der, A.J.B. (Ed.), Biology of Anaerobic Microorganisms. John
Wiley and Sons, New York, pp. 333–372.
Council for Agricultural Science and Technology, 1995. Waste
management and utilization in food production and processing.
Task force Rpt. No. 124. CAST, 4420 West Lincoln way, Ames IA.
Domingues, F.C., Queiroz, J.A., Cabral, J.M.C., Fonseca, L.P., 2000.
The influence of culture conditions on mycelial structure and
cellulase production by Trichoderma reesei Rut C-30. Enzyme
Microb. Technol. 26, 394–401.
Duff, S.J.B., Murray, W.D., 1996. Bioconversion of forest products
industry waste cellulosics to fuel ethanol: a review. Bioresour.
Technol. 55, 1–33.
Eaton, A.D., Clesceri, L.S., Greengerg, A.E., 1995. Standard Methods
for the Examination of Water and Wastewater, 19th ed. American
Public Health Association, Washington, DC.
Fontenot, J.P., Smith, L.W., Sutton, A.L., 1983. Alternative utilization
of animal waters. J. Animal Sci. (Suppl. 2), 221.
Gadgil, N.J., Daginawala, H.F., Chakrabarti, T., Khanna, P., 1995.
Enhanced cellulase production by a mutant of Trichoderma reesei.
Enzyme Microb. Technol. 17, 942–946.
Ghose, T.K., 1987. Measurement of cellulase activities. Pure Appl.
Chem. 59, 257–268.
Goering, H.K., Van Soest, P.J., 1970. Forage Fiber Analysis (Appa-
ratus, Reagent, Procedures and Some Application). In: Agricul-
tural Handbook No. 379. Agricultural Research Service, United
State Department of Agriculture, Washington, DC. pp. 1–20.
Haaland, P.D., 1989. Experimental Design in Biotechnology. Marcel
Dekker Inc., New York.
Haddadin, M.S.Y., Abu-Reesh, I.M., Haddadin, F.A.S., Robinson,
R.K., 2001. Utilisation of tomato pomace as a substrate for the
production of vitamin B12––a preliminary appraisal. Bioresour.
Technol. 78, 225–230.
Henze, M., Harremoes, P., Jansen, J.L.C., Arvin, E., 1996. Wastewater
Treatment: Biological and Chemical Process, second ed. Apringer,
Berlin.
Ju, L.K., Afolabi, O.A., 1999. Wastepaper hydrolysate as soluble
inducing substrate for cellulase production in continuous culture of
Trichoderma reesei. Biotechnol. Prog. 15, 91–97.
Kalogeris, E., Christakopoulos, P., Katapodis, P., Alexiou, A.,
Vlachou, S., Kekos, D., Macris, B.J., 2003. Production and
characterization of cellulolytic enzymes from the thermophilic
fungus Thermoascus aurantiacus under solid state cultivation of
agricultural wastes. Process Biochem. 38, 1099–1104.
Mandels, M., Weber, J., 1969. The production of cellulases. Adv.
Chem. Ser. 95, 391–414.
499
Mandels, M., Weber, J., Parizek, R., 1971. Enhanced cellulase
production by a mutant of Trichoderma viride. Appl. Microbiol.
21, 152–154.
Moller, H.B., Sommmer, S.G., Ahring, B.K., 2002. Separation
efficiency and particle size distribution in relation to manure type
and storage conditions. Bioresour. Technol. 85, 189–196.
Montenecourt, B.S., Eveleigh, D.E., 1979. Selective screening methods
for the isolation of high yielding cellulase mutants of Trichoderma
reesei. Adv. Chem. Ser. 181, 289–301.
Montgomery, D.C., 1997. Design and analysis of experiments, fourth
ed. John Wiley and Sons, New York.
Mulbry, W.W., Wilkie, A.C., 2001. Growth of benthic freshwater
algae on dairy manure. J. Appl. Phycol. 13, 301–306.
Nielsen, H.B., Mladenovska, Z., Westermann, P., Ahring, B.K., 2004.
Comparison of two-stage thermophilic (68 °C/55 °C) anaerobic
digestion with one-stage thermophilic (55 °C) digestion of cattle
manure. Biotechnol. Bioeng. 86, 291–300.
Ogel, Z.B., Yarangumeli, K., Dundar, H., Ifrij, I., 2001. Submerged
cultivation of Schtalidium thermophilum on complex lignocellulosic
biomass for endoglucanase production. Enzyme Microb. Technol.
28, 689–695.
Potter, M., Oppermann-Sanio, F.B., Steinbuchel, A., 2001. Cultivation
of bacteria producing polyamino acids with liquid manure as
carbon and nitrogen source. Appl. Environ. Microbiol. 67, 617–
622.
Rajoka, M.I., Malik, K.A., 1997. Cellulase production by Cellulo-
monas biazotea cultured in medium containing different cellulosic
substrate. Bioresour. Technol. 59, 21–27.
Reczey, K., Szengyel, Z., Eklund, R., Zacchi, G., 1996. Cellulase
production by T. reesei. Bioresour. Technol. 57, 25–30.
Romero, M.D., Aguado, J., Gonzalez, L., Ladero, M., 1999. Cellulase
production by Neurospora crassa on wheat straw. Enzyme Microb.
Technol. 25, 244–250.
Smits, J.P., Rinzema, A., Tramper, J., van Sonsbeek, H.M., Knol, W.,
1996. Solid-state fermentation of wheat bran by Trichoderma reesei
QM9414: substrate composition changes, C balance, enzyme
production, growth and kinetics. Appl. Microbol. Biotechnol. 46,
489–496.
Sternberg, D., 1976. Production of cellulase by Trichoderma. Biotech-
nol. Bioeng. Symp. 6, 35–53.
Szengyel, Z., Zacchi, G., 2000. Effect of acetic acid and furfural on
cellulase production of Trichoderma reesei RUT C30. Appl.
Biochem. Biotechnol. 89, 31–42.
Umikalsom, M.S., Ariff, A.B., Shamsuddin, Z.H., Tong, C.C., Hassan,
M.A., Karim, M.I.A., 1997a. Production of cellulase by a wild
strain of Chaetomium globosum using delignified oil palm empty-
fruit-bunch fiber as substrate. Appl. Microbiol. Biotechnol. 47,
590–595.
Umikalsom, M.S., Ariff, A.B., Zulkifli, H.S., Tong, C.C., Hassan,
M.A., Karim, M.I.A., 1997b. The treatment of oil palm empty fruit
bunch fiber for subsequent use as substrate for cellulase production
by Chaetomium globosum kunze. Bioresour. Technol. 62, 1–9.
Velkovska, S., Marten, M.R., Ollis, D.F., 1997. Kinetic model for
batch cellulase production by Trichoderma reesei RUT C30. J.
Biotechnol. 54, 83–94.
von Sivers, M., Zacchi, G., 1995. A techno-economical comparison of
three processes for the production of ethanol. Biorec. Technol. 51,
43–52.
Wen, Z., Liao, W., Chen, S., 2004. Hydrolysis of animal manure
lignocellulosics for reducing sugar production. Bioresour. Technol.
91, 31–39.
Wilkie, A.C., Mulbry, W.W., 2002. Recovery of dairy manure
nutrients by benthic freshwater algae. Bioresour. Technol. 84, 81–
91.
Xia, L., Cen, P., 1999. Cellulase production by solid state fermentation
on lignocellulosic waste from the xylose industry. Process Biochem.
34, 909–912.