Available online at www.sciencedirect.

com

Bioresource Technology 99 (2008) 5602–5609

Application of a statistical design to the optimization of parameters

and culture medium for a-amylase production by Aspergillus
oryzae CBS 819.72 grown on gruel (wheat grinding by-product)
Radhouane Kammoun *, Belgacem Naili, Samir Bejar
Centre de Biotechnologie de Sfax, Laboratoire d’Enzymes et Métabolites des Procaryotes, Route de sidi Mansour km 4, BP”K” 3038 Sfax, Tunisia

Received 29 September 2006; received in revised form 22 October 2007; accepted 24 October 2007
Available online 3 January 2008

Abstract

The production optimization of a-amylase (E.C.3.2.1.1) from Aspergillus oryzae CBS 819.72 fungus, using a by-product of wheat
grinding (gruel) as sole carbon source, was performed with statistical methodology based on three experimental designs. The optimisa-
tion of temperature, agitation and inoculum size was attempted using a Box–Behnken design under the response surface methodology.
The screening of nineteen nutrients for their influence on a-amylase production was achieved using a Plackett–Burman design. KH2PO4,
urea, glycerol, (NH4)2SO4, CoCl2, casein hydrolysate, soybean meal hydrolysate, MgSO4 were selected based on their positive influence
on enzyme formation.
The optimized nutrients concentration was obtained using a Taguchi experimental design and the analysis of the data predicts a the-
oretical increase in the a-amylase expression of 73.2% (from 40.1 to 151.1 U/ml). These conditions were validated experimentally and
revealed an enhanced a-amylase yield of 72.7%.
Ó 2007 Elsevier Ltd. All rights reserved.

Keywords: Aspergillus oryzae; a-Amylase; Box–Behnken design; Plackett–Burman design; Taguchi design

1. Introduction 2002) and special peptides for children diets (Kammoun

Several starch by-products, generated from many indus-
trial processes, represent one of the most abundant carbon
resources in nature. The use of such resources in the culti-
vation process is a promising way to reduce the cost price
of produced molecules particularly those of high added
value like antibiotics, enzymes, etc.
Such resources are particularly attractive as they provide
an inexpensive industrial substrate and contribute in solv-
ing pollution problems. Gruel (by-product of wheat grind-
ing) is one such by-product that has gained attention as a
substrate for the production of functional bioproducts like
biopesticides from Bacillus thuringiensis (Zouari et al.,
*
Corresponding author. Tel.: +216 98418570; fax: +216 74440451.
E-mail address: radhouan.kammoun@cbs.rnrt.tn (R. Kammoun).
et al., 2003).
With the advent of new frontiers in biotechnology, the
amylase family enzyme finds potential application in a num-
ber of industrial processes such as bread making, brewing,
starch processing, pharmacy, textile, paper industries.
a-Amylases (E.C.3.2.1.1.: endo a-1-4 glucan-4-glucanohy-
drolase) are extra-cellular enzymes that randomly cleave the
1, 4-a-D-glucosidic linkages between adjacent glucose units
in the linear amylose chain. a-Amylase is secreted as a pri-
mary metabolite of microorganisms and its production is a
growth related process (Spohr et al., 1998; Sudo et al., 1994).
Aspergillus oryzae has been largely used in the produc-
tion of food such as soy sauce (Manabe et al., 1984), organic
acid such as citric and acetic acids and commercial enzymes
including a-amylase (Ramachandran et al., 2004). For the
production of this enzyme by culture of A. Oryzae, many
by-products are tested like spent brewing grains (Francis

0960-8524/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2007.10.045
 R. Kammoun et al. / Bioresource Technology 99 (2008) 5602–5609
et al., 2003), coconut oil cake (Ramachandran et al., 2004)
and orange waste water (Djekrif-Dakhmouche et al., 2006).
Many works (Yabuki et al., 1977; Erratt et al., 1984; Tada
et al., 1991; Lachmund et al., 1993) reported that the a-amy-
lase production in A. oryzae is induced by carbohydrates
having a-1,4 glucosidic bonds, e.g., maltose or starch. The
study of the influence of many carbon sources on alpha
amylase production by this strain, realised by Carlsen and
Nielsen (2001), showed that maltodextrins are the most
inducer of alpha amylase expression. In addition, Arnesen
et al. (1998) and Nirmala and Muralikrishna (2003)
reported respectively that Tween-80 and calcium ions
enhance a-amylase activity production.
In the development of any fermentation process, the
optimization of the process variables, particularly physical
and chemical parameters, is of primary importance due to
its impact on the economy and feasibility of the process.
Classical optimization method involves single factor varia-
tion, keeping the other factors constant. This method is
unsuited for multifactor optimisation not only because it
is time-consuming but also because it is unable of detecting
the true optimum due especially to the interactions between
the factors (Liu and Tzeng, 1998; Weuster-Botz, 2000). The
limitation of such method is avoided by using a statistical
design (named also factorial experimental design): a collec-
tion of experimental strategies, mathematical methods and
statistical inferences. Statistical designs allow explaining
interactions between the different factors (Elibol, 2004;
Liu and Tzeng, 1998). These designs have been applied to
fermentation processes optimization precisely in the culture
of bacteria (Ahuja et al., 2004; Weuster-Botz, 2000), animal
cell (Ganne and Mignot, 1991) and fungus (Djekrif-Dakh-
mouche et al., 2006; Francis et al., 2003). The most popular
choices are the Plackett–Burman design (Djekrif-Dakh-
mouche et al., 2006; Viswanathan and Surlikar, 2001),
the central composite design (Dey et al., 2001; Vohra and
Satyanarayana, 2002), the Box–Behnken design (Sarra
et al., 1993) and the Graeco–Latin squares (Haltrich
et al., 1994).
Some works already reported; deal with the influence of
crucial operational parameters and media optimisation on
alpha amylase production by A. oryzae strains. The effects
of agitation on fragmentation of a recombinant strain of A.
oryzae and its consequential effects on protein production
have been investigated by Amanullah et al. (1999). Gigras
et al. (2002) reported the optimization of production of an
alpha-amylase from A. oryzae using a central composite
design. However, in our knowledge, no works concerning
the optimisation of both thermo and hydro dynamics
parameters and media composition were already cited. In
the present study, we reported the optimization of alpha
amylase production by A. oryzae using gruel based culture
media. Values of hydro and thermodynamics parameters
(temperature and agitation) and inoculums size concentra-
tion were selected by Box–Behnken design with three levels
and using response surface methodology (RSM). The
nutrients composing the medium for amylase production
5603
were selected using Plackett–Burman design and their con-
centrations were optimized using Taguchi design.
2. Methods
2.1. Microorganism, maintenance of culture
A fungal strain of A. oryzae CBS 819.72 (purchased
from Centraalbureau voor Schimmelcultures, The Nether-
lands) was used in this study. It was propagated on Potato–
Dextrose–Agar (PDA) (Fluka, France) medium plate at
30 °C. The plates were grown for five to seven days then
stored at 4 °C.
2.2. Inoculum Preparation
A. oryzae, seven days old, were harvested from plates by
adding ten millilitres of distilled water containing 0.1%
Tween-80. The spores were dislodged using the inoculation
needle under aseptic conditions. After centrifugation at
9000g during 20 min at 4 °C and elimination of superna-
tant, glycerol was added to the suspension. Then it was,
appropriately diluted for the required density of spores
and used as the master suspension. The number of viable
spores in the inoculum was determined by the counting
technique using the Thomas cell.
2.3. Substrate and culture media
Gruel, obtained from a local semolina factory process-
ing durum wheat, was used as substrate. It contains starch
(60%), other carbohydrates (5%), cellulose (1.5%), gluten
(12%) and total nitrogen (2.1%) (Bejar et al., 1999). Into
500 ml Erlenmeyer flask, two grams and half of Gruel were
mixed with an enzyme production medium (100 ml), con-
taining (g/l): peptone 10.0, yeast extract 5.0, NaCl 10,
KH2PO4 0.5, MgSO4 0.5, CaCl2 0.5 and distilled water.
The contents were thoroughly mixed and the initial pH
was adjusted to 5.0. The cotton plugged flasks were auto-
claved at 121 °C (15 psi) for 20 min. After cooling the med-
ium, they were inoculated with an appropriate volume of
the master spore suspension then kept on rotary shaker
at permanent conditions for 72 h.
2.4. Enzyme assay
a-Amylase was assayed by the addition of 50 ll enzyme
of culture supernatant to 0.5 ml of 1% (w/v) soluble starch
made in 0.1 M acetate buffer (pH 5.6) and incubation of the
reaction mixture at 60 °C for 30 min. Reducing sugars
released were measured by the 3, 5-dinitrosalicylic acid
method (Miller, 1959). A separate blank was set up for
each sample to correct the non-enzymatic release of sugars.
One unit of a-amylase was defined as the amount of
enzyme that released reducing sugars equivalent to 1 lmol
glucose per minute under the standard assay conditions.
5604 R. Kammoun et al. / Bioresource Technology 99 (2008) 5602–5609

2.5. Experimental design and statistical analysis Table 2

Nutrient supplements for screening using Plackett–Burman design
Optimization of the parameters (temperature, agitation Nutrient Nutrient (+) level () level Bibliographic
and inoculum size) for overproduction of amylase was code name (%) g/g (%) reference
base g/g base
done by Box–Behnken design especially made to require substrate substrate
three levels coded as (), (0) and (+) (N = 13) (13 experi-
A K2HPO4 0.2 0.1 Francis et al. (2003)
ments and three factors at three levels) under the response B KH2PO4 0.2 0.1 Spohr et al. (1998),
surface methodology. Table 1 shows the different levels of Dey et al. (2001)
each of the parameters. The basic points for the design C NaCl 0.1 0.05 Spohr et al. (1998)
were selected from a preliminary study (data not shown). D MgSO4 0.1 0.05 Djekrif-Dakhmouche
Box–Behnken design is a fractional factorial design et al. (2006)
E Yeast 0.2 0.1 Djekrif-Dakhmouche
obtained by combining two-level factorial designs with extract et al. (2006),
incomplete block designs. The response surface methodol- Dey et al. (2001)
ogy (RSM) was used to analyse the experimental design F Corn steep 0.2 0.1 Djekrif-Dakhmouche
data. In order to be correlated to the independent vari- liquor et al. (2006)
ables, the response variable was fitted by a second order G Peptone 0.2 0.1 Ramachandran et al.
(2004)
model. The general form of the second degree polynomial H Urea 0.2 0.1 Ramachandran et al.
equation is (2004)
X X X I Soybean 0.2 0.1 Spohr et al. (1998),
y i ¼ b0 þ bi xi þ bii x2i þ bij xi xj meal Mulimani et al. (2000)
hydrolysate
where Yi is the predicted response, x i, xj are input variables J Casein acid 0.2 0.1 Mulimani et al. (2000)
which influence the response variable Y; b0 is the offset hydrolysate
term; bi is the ith linear coefficient; bii the ith quadratic K Glycerol 0.2 0.1 Tanyildizi et al. (2005)
coefficient and bij is the ijth interaction coefficient. L (NH4)2SO4 0.2 0.1 Mulimani et al. (2000),
Ramachandran et al.
To select the nutrients to add to gruel, 19 components (2004)
have been chosen to evaluate their effect on a-amylase pro- M CoCl2 0.2 0.1 Francis et al. (2003)
duction. These components have been used in some works N NH3 0.2 0.1 Francis et al. (2003)
for the production of a-amylase by Aspergillus species (see O CaCl2 0.1 0.05 Nirmala and
Table 2). The selected compounds were screened using Muralikrishna (2003)
P ZnCl2 0.1 0.05 Francis et al. (2003)
Plakett–Burman design (N = 20) [20 experiments, 19 nutri- Q FeCl3 0.1 0.05 Djekrif-Dakhmouche
ments and one error]. This design is a fractional factorial et al. (2006)
plan which allows the investigation of up to N  1 vari- R MnSO4 0.1 0.05 Djekrif-Dakhmouche
ables with N experiments (N multiple of 4 and less or equal et al. (2006)
to 100). The Plakett–Burman design assumes that there are S Tween 80 0.1 0.05 Arnesen et al. (1998)
no interactions between the different media constituents, xi
in the range of variables under consideration (Plackett and
Burman, 1946). A linear approach is considered to be suf-
measurements made at the high (+) and the low () levels
ficient for screening.
of the factors. This coefficient notifies the main effect of the
Y ¼ b0 þ bixiði ¼ 1; . . . ; kÞ studied factor. Table 2 shows the bibliographic reference
and the two levels (+) and () of each of the constituents
where Y is the estimated target function and bi are the
used in the Plackett–Burman design.
regression coefficients. The contrast coefficient, noted b,
For the optimization of nutrients, a Taguchi orthogonal
was calculated as the difference between the average of
array (OA) experimental design (DOE) methodology
Table 1 (N = 18) [18 experimental trails, seven factors at three lev-
Parameters values for Box–Behnken design els and one factor (Mg SO4) at two levels] was used. This
Factor Basic Variation Value of the Coded method implies the study of any given system by a set of
level interval factor value independent variables (factors) above a specific area of
Temperature (°C) 30 5 25  interest (levels) (Krishna Prasad et al., 2005). The Taguchi
30 0 design makes it possible to establish the performance at the
35 + optimum levels obtained with some well defined experi-
Agitation (rpm) 200 50 150  mental sets (Krishna Prasad et al., 2005).
200 0 In the present study, the three levels of factor variations
250 + coded as (1), (2) and (3) were considered (Table 3). The
Log10 (Inoculum size) 7 0.5 6.5  total degree of freedom is equal to the number of trails
(Spore/ml) 7 0 minus one i.e., 17. In the orthogonal array design, each col-
7.5 +
umn consists of a number of conditions depending on the
 R. Kammoun et al. / Bioresource Technology 99 (2008) 5602–5609 5605

Table 3 The regression model is determined by the SPSS proce-

Levels of selected nutrient supplements using Taguchi design dure that considers initially all the factors and then elimi-
Nutrient Nutrient name Level (1) Level (2) Level (3) nates step-by-step those that don’t have any effect. So,
code g/g base g/g base g/g base insignificant model terms were excluded. The model equa-
substrate substrate substrate
tion fitted by regression analysis is given by
X1 MgSO4 – 0.1 0
X2 KH2PO4 0.2 0.1 0 ActivityðU=mlÞ ¼ 999:942  3:643 T þ 0:108 N
X3 Urea 0.5 0.25 0
X4 Soybean meal hydrolysate 0.5 0.25 0 þ 302:697 S  22:618 S 2  0:00359 T :N
X5 Casein acid hydrolysate 0.5 0.25 0
X6 Glycerol 0.5 0.25 0 þ 0:564 T :S
X7 (NH4)2SO4 0.1 0.05 0
X8 CoCl2 0.1 0.05 0
where Y is the a-amylase activity (U/ml of fermentation
supernatant), T the temperature (°C), N the agitation
(rpm) and S the log10 (spores/ml culture media).
levels assigned to each factor. Table 3 gives the variation The results of the second order response surface model
levels of the components added to gruel. in the form of analysis of variance (ANOVA) are realised.
All experiences are carried out at pH 5.0 for 72 h in The Fisher F-test [F(6, 6) = 10.85] with a very low proba-
duplicates to reach an optimum combination of the bility value (Pmodel > F = 0.005269) demonstrate a high
parameters. significance for the regression model (Khuri and Cornell,
1987). The goodness of fit of the model was checked by
3. Results and discussion the determination coefficient (R2). In this case, the value
of the determination coefficient (R2 = 0.957) indicates that
3.1. Optimization of operational parameters only 4.43% of the total variations are not explained by the
model. A higher value of the correlation coefficient,
Contrary to the pH generally employed in the range of 5 R(=0.9782), justifies an excellent correlation between the
and 5.5, the temperature, the agitation and the size of inoc- independent variables (Khuri and Cornell, 1987). At the
ulum were recognized as being the most controlling param- same time a relatively lower value of the coefficient of var-
eters on the production of a-amylase by A. oryzae. The iation (CV = 2.22%) indicates a better precision and reli-
experimental Box–Behnken design was applied to analyze ability of the experiments carried out (Khuri and Cornell,
the interactions between these parameters and to determine 1987).
the optimum conditions for production. The design and From the model equation, it can be seen that the vari-
results of experiments carried out are given in Table 4. ables with the largest effect were, the linear term of the size
The average of amylase activity of triplicate values of inoculum (S) and to a less extent the quadratic term of
obtained was taken as the dependent variable or response the inoculum size (S2). The great importance of inoculum
Y. The results were analyzed using SPSS (Version 11.0.1 size on the production of a-alpha amylase was already
2001, LEAD Technologies, Inc., USA) statistical software reported. Francis et al. (2003) and Jiff et al. (1998) reported
and the response surface was generated using EXCEL that inoculum size was a parameter which affects greatly
(Version 2003, Microsoft office, Inc., USA) software. the production of a-amylase in A. oryzae. Similarly, Muli-
mani et al. (2000) reported that inoculums volume per
gram of substrate was among the most effective parameters
Table 4 in stimulating amylase production by Gibberella fujikuroi.
Box–Behnken experimental design used to optimize the physical param- Otherwise, the temperature is the second significant term.
eters for the production of a-amylase on gruel by A. oryzae CBS 819.72
In fact, the temperature shows a negative effect on a-amy-
Fermentation Temperature Agitation Log10 a-Amylase lase production of A. oryzae. This is in agreement with the
Code (°C) (rpm) (Inoculum/size) activity (U/ml)
results reported by Francis et al. (2003), showing a decrease
(spore/ml)
in production at temperatures outside the mesophilic
1 25 150 7 21,81
range.
2 35 150 7 17,52
3 25 250 7 31,31 The speed of agitation shows limited positive effect on
4 35 250 7 16,36 the enzyme production. An increase of the agitation, on
5 25 200 6.5 11,61 such limits, causes an enzyme production enhancement
6 35 200 6.5 15,30 since the aeration is an important factor for the growth
7 25 200 7.5 15,94
of aerobic strains. This result slightly contrast with the
8 35 200 7.5 12,24
9 30 150 6.5 11,12 studies undertaken by Amanullah et al. (1999, 2000,
10 30 250 6.5 15,93 2001), and Bennamoun et al. (2004). Indeed, Amanullah
11 30 150 7.5 24,00 et al. (2001), showed that protein production (a-amylase
12 30 250 7.5 16,26 and amyloglucosidase) was independent of agitation speed
13 30 200 7 20,23
although significant changes in mycelial morphology.
5606 R. Kammoun et al. / Bioresource Technology 99 (2008) 5602–5609

These authors suggested that rapid transients of morpho-

logical parameters in response to a speed change from
1000 to 550 rpm probably resulted from aggregation and 25
they conclude that mycelial morphology does not directly 22.5
affect protein production. Bennamoun et al. (2004) indicate 20
that limited changes in the speed of agitation have no effect

Activity (U.ml )
17.5
on the production of proteins while its effect on biomass is

α−amylase
15
clearly positive. 12.5
The 3D response surfaces are the graphical representa- 10
tions of the regression equation for a-amylase yields. They 7.5

-1
5
depict the effect of pair wise interaction of the parameters,
2.5
when the third parameter is kept constant Figs. 1–3 show 0
the effect of interaction respectively of temperature and agi-

0
15

7.
0
17

6
tation, agitation and inoculum size and temperature and

7.
0

4
19

7.
inoculum size on a-amylase production. From these plots, N

2
0
-1

7
l )

21
(r.
p.m .m

0
it is very easy and convenient to understand the interac-

6.
re

23
.) po

8
22.5-25

6.
(S

25
tions between two parameters and also to locate their opti-

6
gS 20-22.5

6.
Lo

4
mum levels. 17.5-20
15-17.5
In Figs. 1 and 3, it can be seen that the yield of a-amy-
lase increased by decreasing the temperature of 35–25 °C. Fig. 2. a-Amylase production (U/ml) observed as a response to the
On the other hand, the a-amylase output increased with interaction of agitation (rpm) and inoculum size (Log S) (spore/ml) as
increasing the agitation from 150 to 250 rpm. The effect variables and temperature (°C) at central point.
of the temperature is more pronounced with a higher agita-
tion (Fig. 1). This result indicates that A. oryzae tolerate a
high agitation and a mesophilic incubation temperature.
The literature review reported that most amylase produc- 25
tion studies have been done with mesophilic fungi having 22.5
a temperature range of 25–37 °C (Gupta et al., 2003). 20
These observations are similar to those done for A. Niger 17.5

Activity (U.ml-1)
α−amylase
ATCC16404 (Djekrif-Dakhmouche et al., 2006). The liter- 15

ature reported also that agitation intensity influences the 12.5

10
mixing and oxygen transfer rates in many fungal fermenta-
7.5
tion and thus controls mycelial morphology and a-amylase 5
formation (Gupta et al., 2003). As observed in Fig. 1, the 2.5
0
24
26
7.
6

28
7.
4

30
7.

-1 22.5-25
l )
2

32
7

T (° .m 20-22.5
34 ore
6.

C) 17.5-20
(Sp
8

36
6.

gS 15-17.5
6
6.

25 Lo 12.5-15
4

22.5
20 Fig. 3. a-Amylase production (U/ml) observed as a response to the
17.5 interaction of inoculum size (Log S) (spore/ml) and temperature (°C) as
Activity (U.ml )

15 variables and agitation (rpm) at central point.

α-amylase -1

12.5
10 agitation and the temperature can significantly affect
7.5 a-amylase production. The contours were slightly inclined
5
to the horizontal showing that there was a significant inter-
2.5
0
action between the two parameters in particular at higher
0
15 agitation. The interaction between these two parameters,
0 36
17 34
19
0
32
often studied separately, was overlooked by classical exper-
N 0
(r.
p.m
21 30 ) imental strategy.
23
0 28 °C
.) 0 26 T( 22.5-25 From Fig. 2, we can observe that the contours were par-
25 20-22.5
24 17.5-20 allel to the two axes suggesting that the two parameters
15-17.5 were quite independent of each-other. Age and size of the
Fig. 1. a-Amylase production (U/ml) observed as a response to the inoculum are usually found to relate with time-course of
interaction of temperature (°C) and agitation (rpm) as variables and incubation rather than agitation (Francis et al., 2003).
inoculum size at central point. Fig. 3 shows a decrease in production at high temperatures.
 R. Kammoun et al. / Bioresource Technology 99 (2008) 5602–5609 5607

The contours were slightly inclined to the horizontal show-
ing a significant interaction between the two parameters.
3.2. Screening of nutrient supplements
Several experimental designs have been considered for
choosing media supplements, and we selected the Plak-
ett–Burman design proposed by Francis et al. (2003). The
analysis showed that mediums ‘17’ and ‘19’ (45.32 and
45.14 U/ml) gave the maximum yield followed by mediums
‘13’ and ‘18’ (44.12 and 43.87 U/ml). (NH4)2SO4, CoCl2,
and two nitrogen sources among urea, soybean meal
hydrolysate and casein acid hydrolysate, were present in
higher titres in these media. The analysis of the contrast
coefficient (b) showed that eight compounds: CoCl2, casein
acid hydrolysate, (NH4)2SO4, glycerol, urea, soybean meal
hydrolysate (b vary between 7.29 and 3.33) and at less
extend, KH2PO4, MgSO4, MnSO4 and peptone had pro-
nounced influence on the a-amylase production (b vary
between 1.78 and 0.5). According to bibliographic studies,
the effects of P and Cl ions on the stability of a-amylase
have been well recognized (Chessa et al., 1999; Pandey
et al., 2000). Otherwise, the organic nitrogen source, like
casein acid hydrolysate and soybean meal hydrolysate,
have traditionally a positive effect over inorganic ones since
they are also carbon source and contain trace of minerals
and ions that could enhance the enzyme secretion.
The optimum combinations of the eight compounds
were further analyzed by a Taguchi design (we ignored
the effect of MnSO4 and peptone).
3.3. Optimization of nutrients levels
For this study, an L18 Taguchi design proposed by
Krishna Prasad et al. (2005) was employed. Production lev-
els were found to be very much dependent on the culture
conditions (0.25–106.7 U/ml). The affect of the level, the
yield difference between level 1 and 2, the contribution of
each factor and the resulted optimal conditions (value
and level) are shown in Table 5. Individually at level stage,
cobalt source (CoCl2) has the highest affect in level 1
whereas urea and casein acid hydrolysate has higher affects,
in level 2 and 3 respectively, on the a-amylase yield. The
difference between level 2 and level 1 (L2-L1) of each factor
indicates the relative influence of the affect. The larger the
difference (L2-L1), the stronger is the influence of the factor
level. It can be seen from Table 5, that among the factors
studied, urea showed stronger influence, compared to other
factors followed by (NH4)2SO4, soybean meal hydrolysate,
glycerol, casein acid hydrolysate, CoCl2, MgSO4 and
KH2PO4. Gupta et al. (2003) indicated that, a-amylase
activity in fungal cultures can be increased by the addition
to the media of different compounds as inducers. Our study
revealed that using many nitrogen sources: urea, casein
acid hydrolysate, soybean meal hydrolysate, (NH4)2SO4,
seems to be a better way to produce a-amylase. On the
other hand, the study showed that an increase of N leads
to an enhancement of amylase expression. This result
might indicate a kind of coupling between N and C supply.
This corroborate the results of New and Stevens (2004)
which indicate that enzyme yield could be increased by
using more N-source in the submerged fermentation med-
ium. Other wise, literature indicated that urea is the best
nitrogen source tested for the production of many fungal
bio products (Gupta et al., 2003). Nevertheless, there are
no data for the production of amylase in the presence of
enhanced supply of N-source.
Increase in concentration of factors such as KH2PO4,
casein acid hydrolysate and (NH4)2SO4 has resulted in
increase in enzyme production. While increase in urea, soy-
bean meal hydrolysate or glycerol concentration has
resulted in higher a-amylase expression up to level 2 and
subsequent increase resulted in decrease in the a-amylase
yield. CoCl2 concentration at level 1 showed higher yield
of a-amylase. Increased to level 2 then to level 3 concentra-
tions, it showed progressive increase in yield. This may be
reasoned due to the constitutive effect of the other culture
media compounds.
Phosphate and magnesium seem to play important roles
on the expression of alpha amylase. This result confirm the
literature purpose which indicate that phosphate has a
main regulatory role in the synthesis of primary and sec-
ondary metabolites in microorganisms and likewise it
affects the growth of the organism and production of

Table 5
Main effect of factors, difference of level (2) and (1), contribution and optimum levels and values of factors
Nutrient code Level (1)a Level (2)a Level (3)a L2-L1ab Optimum valuesc Optimum level Contribution (%)
X1 35.19 45.62 – 10.43 0.1 0 5.21
X2 31.94 38.73 50.55 6.79 0.2 + 10.14
X3 21.17 61.65 38.40 40.48 0.5 + 21.24
X4 30.56 50.60 40.07 20.04 0.25 0 10.19
X5 25.32 35.93 59.97 10.61 0.5 + 19.57
X6 35.91 53.90 31.42 17.99 0.25 0 13.49
X7 24.91 47.80 48.52 22.89 0.1 + 8.11
X8 63.14 23.18 34.90 39.96 0  22.73
Total contribution from all factors = 110.66 U/ml, current grand average performance = 40.41 U/ml, expected result at optimum condition = 151.1 U/ml.
a
The value is given in term of activity (U/ml).
b
Level (2)–level (1).
c
g/g base substrate.
5608 R. Kammoun et al. / Bioresource Technology 99 (2008) 5602–5609

Table 6 tion by a factorial experimental design leading to a

Comparison between the original and optimized media substantial increase in a-amylase production yield. This
Media Nutrient name Nutrient concentration a-Amylase strategy was proved to be useful and powerful tool for
(%) (g/g base substrate) activity (U/ml) screening, optimisation and modelling of fermentation pro-
Original Peptone 4.0 31.3 cess. Our study showed that gruel (maltodextrins source)
Yeast extract 2.0 constitutes a good carbon source for the growth of A. ory-
NaCl 4.0
KH2PO4 0.2
zae and a best inducer for the production of alpha amylase.
MgSO4 0.2 Statistical analysis showed that inoculum size is among the
CaCl2 0.5 0.2 most important factors affecting alpha amylase release. The
Optimised Urea 0.5 148.0 significant achievement of the present work lies in the fact
Casein acid 0.5 that using enhanced quantities of many nitrogen sources
hydrolysate leads to an increase of alpha amylase expression. In this
Soybean meal 0.25 study, the experimental results clearly showed that phos-
hydrolysate
phate and magnesium play an important role on the
Glycerol 0.25
KH2PO4 0.2 enzyme expression. These minerals might be essentials for
(NH4)2SO4 0.1 the maintenance of mould, enzyme production and stabil-
MgSO4 0.1 ity. Using the optimized conditions, the produced activity
reaches 148 U/ml and 5920 U/g of gruel. The results show
a-amylase by A. oryzae strains (Gupta et al., 2003). Liter- a close concordance between the expected and obtained
ature indicates also that production was reduced to 50% activity level. This is the first report about production of
when Mg2+ was omitted from the medium. On the other alpha amylase by A. oryzae CBS 819.72 using gruel based
hand, we demonstrated that the CaCl2 is not important medium.
for the amylase production. This corroborates the results This paper proposes a low cost medium formulation
of Kundu et al. (1973). In fact these authors demonstrate that could be of industrial value and could serve as a basal
that Ca2+ was inhibitory to amylase production by A. ory- media for further optimization studies of this and similar
zae EI 212 although Ca2+ has been reported as essential for strains. Furthermore, this study serves as another example
alpha amylase stability (Gupta et al., 2003). for the application of the statistical methodology to biolog-
ical systems. In addition, after further optimization is car-
ried out using optimized medium for batch and fed batch
3.4. Optimum validation systems and after optimizing the fermentation parameters
(aeration, agitation speed, dilution rate, etc.), the enzyme
The optimum combination was found to be (g/g dry base activity is expected to increase.
substrate): urea 0.5, casein acid hydrolysate 0.5, soybean
meal hydrolysate 0.25, Glycerol 0.25, KH2PO4 0.2, Acknowledgements
(NH4)2SO4 0.1, MgSO4 0.1 and CoCl2 0.0. The model
showed that CoCl2 was not essential for a-amylase produc- This work was supported by funds from the Tunisian
tion. This result may be explained by the fact that agro nitro- Government (‘‘Contract-Programme CBS-L.E.M.P.”).
gen source like casein acid hydrolysate and soybean meal We thank Mr. Moncef Affes for critically reading the
hydrolysate can contain trace of ions like Co++ and Cl. manuscript.
Further, to validate the proposed experimental method-
ology, fermentation experiments were performed for a- References
amylase production by employing the obtained optimized
culture conditions. Table 6 gives the comparison between Ahuja, S.K., Ferreira, G.M., Morreira, A.R., 2004. Application of
the yield of a-amylase from original and optimized media Plackett and Burman design and response surface methodology to
under the experimental conditions. The obtained yield achieve exponential growth of aggregated shipworm bacterium.
148.0 U/ml is very close to expected result 151.1 U/ml. Biotechnol. Bioeng. 85, 666–675.
Amanullah, A., Blair, R., Nienow, A.W., Thomas, C.R., 1999. Effects of
The results of this study have clearly demonstrated the note- agitation intensity on mycelial morphology and protein production in
worthy increase in yield (148.0 U/ml). Use of Gruel has chemostat cultures of recombinant Aspergillus oryzae. Biotechnol.
been beneficiary, as the yield of a-amylase has been high Bioeng. 62 (4), 434–446.
in comparison to a-amylase production reported under sub- Amanullah, A., Jüsten, P., Davies, A., Paul, G.C., Nienow, A.W.,
merged fermentation using other by-products, such as Thomas, C.R., 2000. Agitation induced mycelial fragmentation of
Aspergillus oryzae and Penicillium chrysogenum. Biochem. Eng. J. 5,
orange waste water (Djekrif-Dakhmouche et al., 2006). 109–114.
Amanullah, A., Leonildi, E., Nienow, A.W., Thomas, C.R., 2001.
Dynamics of mycelial aggregation in cultures of Aspergillus oryzae.
4. Conclusions
Bioprocess Biosyst. Eng. 24, 101–107.
Arnesen, S., Eriksen, S.H., Olsen, J., Jensen, B., 1998. Increased
In the present work, we have demonstrated the optimi- production of alpha amylase from Thermomyces lanuginosus by the
zation of A. oryzae culture conditions and media composi- addition of Tween-80. Enzyme Microb. Technol. 23, 249–252.
 R. Kammoun et al. / Bioresource Technology 99 (2008) 5602–5609
Bejar, S., Ghorbel, R., Ben Amar, R., Kammoun, R., Ben Messaoud, E.,
Belguith, K., Gargouri, R., Ellouz, R., Procédé de fabrication de sirop
de glucose. 1999; Brevet d’invention No. 17235 INNORPI-TUNSISIE.
Bennamoun, L., Zahia Meraihi, Z., Dakhmouche, S., 2004. Utilisation de
la planification expérimentale pour l’optimisation de la production de
l’a-amylase par Aspergillus oryzae Ahlburg (Cohen) 1042.72 cultivée
sur milieu à base de déchets d’oranges. J. Food Eng. 64, 257–264.
Carlsen, M., Nielsen, J., 2001. Influence of carbon source on a-amylase
production by Aspergillus oryzae. Appl. Microbiol. Biotechnol. 57,
346–349.
Chessa, J.P., Feller, G., Gerday, C., 1999. Purification and characteriza-
tion of the heat-labile a-amylase secreted by the psychrophilic
bacterium TAC 240B. Can. J. Microbiol. 45 (6), 452–457.
Dey, G., Mitra, A., Banerjee, R., Maiti, B.R., 2001. Enhanced production
of amylase by optimization of nutritional constituents using response
surface methodology. Biochem. Eng. J. 7, 227–231.
Djekrif-Dakhmouche, S., Gheribi-Aoulmi, Z., Meraihi, Z., Bennamoun,
L., 2006. Application of a statistical design to the optimization of
culture medium for a-amylase production by Aspergillus niger ATCC
16404 grown on orange waste powder. J. Food Eng. 73 (2), 190–197.
Elibol, M., 2004. Optimization of medium composition for actinorhodin
production by Streptomyces coelicolor A3(2) with response surface
methodology. Process Biochem. 39, 1057–1062.
Erratt, J.A., Douglas, P.E., Moranelli, F., Seligy, V.L., 1984. The
induction of a-amylase by starch in Aspergillus oryzae: evidence for
controlled mRNA expression. Can. J. Biochem. Cell Biol. 62, 678–690.
Francis, F., Sabu, A., Nampoothiri, K.M., Ramachandran, S., Ghosh, S.,
Szakacs, G., Pandey, A., 2003. Use of response surface methodology
for optimizing process parameters for the production of a-amylase by
Aspergillus oryzae. Biochem. Eng. J. 15, 107–115.
Ganne, V., Mignot, G., 1991. Application of statistical design of
experiments to the optimization of factor VIII expression by CHO
cells. Cytotechnology 6, 233–240.
Gigras, P., Sahai, V., Gupta, R., 2002. Statistical media optimization and
production of ITS alpha-amylase from Aspergillus oryzae in a
bioreactor. Curr. Microbiol. 45 (3), 203–208.
Gupta, R., Gigras, P., Mohapatra, H., Goswami, V.K., Chauhan, B.,
2003. Microbial a-amylases: a biotechnological perspective. Process
Biochem. 38, 1599–1616.
Haltrich, D., Laussamayer, B., Steiner, W., 1994. Xylanase formation by
Sclerotium rolfii: effect of growth substrates and development of a
culture medium using statistically designed experiments. Appl. Micro-
biol. Biotechnol. 42, 522–530.
Jiff, B., Van Leeuwen, H.J., Patel, B., Yu, Q., 1998. Utilisation of starch
processing wastewater for production of microbial biomass protein
and fungal a-amylase by Aspergillus oryzae. Biores. Technol. 66, 201–
206.
Kammoun, R., Bejar, S., Ellouz, R., 2003. Protein size distribution and
inhibitory effect of wheat hydrolysates on NeutraseÒ. Biores. Technol.
90, 249–254.
Khuri, A.I., Cornell, J.A., 1987. Response Surfaces: Design and Analysis.
Marcel Dekker, Inc., New York.
Krishna Prasad, K., Venkata Mohana, S., Sreenivas Raob, R., Ranjan
Pati, B., Sarma, P.N., 2005. Laccase production by Pleurotus ostreatus
1804: Optimization of submerged culture conditions by Taguchi DOE
methodology. Biochem. Eng. J. 24, 17–26.
Kundu, A.K., Das, S., Gupta, T.K., 1973. Influence of culture and
nutritional conditions on the production of amylase by the submerged
culture of Aspergillus oryzae. J. Ferment. Technol. 51, 142–150.
Lachmund, A., Urmann, U., Minol, K., Wirsel, S., Ruttkowski, E.,
1993. Regulation of a-amylase formation in Aspergillus oryzae
5609
and Aspergillus nidulans transformants. Curr. Microbial. 26,
47–51.
Liu, B.L., Tzeng, Y.M., 1998. Optimization of growth medium for
production of spores from Bacillus thuringiensis using response surface
methodology. Bioprocess. Eng. 18, 413–418.
Manabe, M., Tanaka, K., Goto, T., Matsura, S., 1984. Producing
capability of kojic acid and aflatoxin by koji mould. Dev. Food Sci. 7,
4–14.
Miller, G.L., 1959. Use of dinitrosalicylic acid reagent for determination
of reducing sugars. Anal. Chem. 31, 426–428.
Mulimani, V.H., Patil, G.N., Ramalingam, 2000. a-Amylase production
by solid state fermentation: a new practical approach to biotechnology
courses. Biochem. Educat. 28, 161–163.
New, N., Stevens, W.F., 2004. Effect of urea on fungal chitosan
production in solid substrate fermentation. Process Biochem. 39,
1639–1642.
Nirmala, M., Muralikrishna, G., 2003. Three amylases from malted finger
millet (Ragi, Eleusine coracana, Indaf-15)–purification and partial
characterization. Phytochemistry 62, 21–30.
Pandey, A., Nigam, P., Soccol, C.D., Singh, D., Soccol, V.T., 2000.
Mohan, Advances in microbial amylases. Biotechnol. Appl. Biochem.
31, 135–152.
Plackett, R.L., Burman, J.P., 1946. The design of optimum multifactorial
experiments. Biometrika 33, 305–325.
Ramachandran, S., Patel, A.K., Nampoothiri, K.M., Francis, F., Nagy,
V., Szakacs, G., Pandey, A., 2004. Coconut oil cake – a potential raw
material for the production of a-amylase. Biores. Technol. 93, 169–
174.
Sarra, M., Redin, I., Ochin, F., 1993. Application of factorial design to the
optimization of medium components in batch cultures of Streptomyces
lividans TK21 producing a hybrid antibiotic. Biotechnol. Lett. 15,
559–564.
Spohr, A., Carlsen, M., Nielsen, J., Villadsen, J., 1998. a-Amylase
production in recombinant Aspergillus oryzae during Fed-batch and
continuous cultivation. J. Ferment. Bioeng. 86 (1), 49–56.
Sudo, S., Ishikawa, T., Sato, K., Oba, T., 1994. Comparison of acid stable
a-amylase production by Aspergillus kawachii in solid-state and
submerged cultures. J. Ferment. Bioeng. 77 (5), 483–489.
Tada, S., Gomi, K., Kitamoto, K., Takahashi, K., Tamura, G., Hara, S.,
1991. Construction of a fusion gene comprising the Taka-amylase A
promoter and the Escherichiu coli p-glucuronidase gene and analysis of
its expression in Aspergillus oryzae. Mol. Gen. Genet. 229, 301–
306.
Tanyildizi, M.S., Dursun Özer, D., Elibol, M., 2005. Optimization of
a-amylase production by Bacillus sp. Process Biochem. 40, 2291–2296.
Viswanathan, P., Surlikar, N.P., 2001. Production of a-amylase with
Aspergillus flavus on amaranthus grains by solid-state fermentation. J.
Basic Microbiol. 41 (1), 57–64.
Vohra, A., Satyanarayana, T., 2002. Statistical optimization of medium
components by response surface methodology to enhance phytase
production by Pichia anomala. Process Biochem. 37, 999–1004.
Weuster-Botz, D., 2000. Experimental design for fermentation media
development: statistical design or Global random search? J. Biosci.
Bioeng. 90 (5), 473–483.
Yabuki, M., Ono, N., Hoshino, K., Fukui, S., 1977. Rapid induction of a-
amylase by non growing mycelia of Aspergillus oryzae. Appl. Environ.
Microbial. 34, 1–6.
Zouari, N., Ben Sik Ali, S., Jaoua, S., 2002. Production of delta-
endotoxins by Bacillus thuringiensis strains exhibiting various insecti-
cidal activities towards Lepidoptera and diptera in gruel and fish meal
media. Enzyme Microb. Techxnol. 31, 411–418.