21030243

7/28/2019 21030243

http://slidepdf.com/reader/full/21030243 1/7

O R I G I N A L A R T I C L E

Response surface methodology for optimizingthe fermentation medium of alpha-galactosidasein solid-state fermentation

C.Q. Liu, Q.H. Chen, B. Tang, H. Ruan and G.Q. He

Department of Food Science and Nutrition, Zhejiang University, Hangzhou 310029, China

Introduction

The enzymatic hydrolysis of alpha-1, 6-linked alpha-gal-

actoside residues from simple oligosaccharides and from

polymeric galactomannans with alpha-galactosidase

(E.C.3Æ2Æ1Æ22) constitutes the basis of the most promising

biotechnological processes currently being developed to

decrease the level of raffinose series oligosaccharides in

soybean meal and other legumina (Cruz and Park 1982;

Gote et al. 2004). The industrial development of such

processes is required for soybean meal frequently used inmonogastric animal diets worldwide, and should decrease

feed cost, because monogastric animals, including

humans, lack the ability to synthesize sufficient alpha-gal-

actoside in their intestinal systems to hydrolyse such

oligosaccharides (OR; Gitzelmann and Auricchio 1965).

Solid-state fermentation (SSF) holds tremendous

potential for the production of enzymes, for the facility of

equipment and technique, low capital investment, super-

ior productivity, less energy requirements and effluent

Keywords

alpha-galactosidase, Aspergillus foetidus,

medium optimization, response surface

methodology, solid-state fermentation.

Correspondence

G. Q. He, Department of Food Science and

Nutrition, Zhejiang University, Hangzhou

310029, China. E-mail: [email protected]

2006 ⁄ 1285: received 13 September 2006,

revised 23 March 2007 and accepted 2 April

2007

doi:10.1111/j.1472-765X.2007.02173.x

Abstract

Aims: Alpha-galactosidase is applied in food and feed industries for hydroly-

sing raffinose series oligosaccharides (RO) that are the factors primarily respon-

sible for flatulence upon ingestion of soybean-derived products. The objective

of the current work was to develop an optimal culture medium for the produc-

tion of alpha-galactosidase in solid-state fermentation (SSF) by a mutant strain Aspergillus foetidus.

Methods and Results: Response surface methodology (RSM) was applied to

evaluate the effects of variables, namely the concentrations of wheat bran, soy-

bean meal, KH2PO4, MnSO4ÆH2O and CuSO4Æ5H2O on alpha-galactosidase

production in the solid substrate. A fractional factorial design (FFD) was firstly

used to isolate the main factors that affected the production of alpha-galactosi-

dase and the central composite experimental design (CCD) was then adopted

to derive a statistical model for optimizing the composition of the fermentation

medium. The experimental results showed that the optimum fermentation

medium for alpha-galactosidase production by Aspergillus foetidus ZU-G1 was

composed of 8Æ2137 g wheat bran, 1Æ7843 g soybean meal, 0Æ001 g MnSO4ÆH2O

and 0Æ001 g CuSO4Æ5H2O in 10 g dry matter fermentation medium.

Conclusions: After incubating 96 h in the optimum fermentation medium,

alpha-galactosidase activity was predicted to be 2210Æ76 U g)1 dry matter in

250 ml shake flask. In the present study, alpha-galactosidase activity reached

2207Æ19 U g)1 dry matter.

Significance and Impact of the Study: Optimization of the solid substrate was

a very important measure to increase enzyme activity and realize industrial

production of alpha-galactosidase. The process of alpha-galactosidase produc-

tion in laboratory scale may have the potential to scale-up.

Letters in Applied Microbiology ISSN 0266-8254

206 Journal compilation ª 2007 The Society for Applied Microbiology, Letters in Applied Microbiology 45 (2007) 206–212

ª 2007 The Authors

7/28/2019 21030243


generation (Chahal 1985; Nigam and Singh 1994). Cruz

and Park (1982), Kotwal et al. (1997), Li et al. (2001) and

Wang et al. (2004) had reported alpha-galactosidase pro-

duction in SSF. Little is known concerning the culture

medium by SSF for formation of this enzyme by Aspergil-

lus foetidus.

The objective of this work was to develop an optimalculture medium using Response Surface Methodology

(RSM) for the production of alpha-galactosidase in SSF

by a mutant strain Aspergillus foetidus to decrease produc-

tion cost, increase enzyme activity and realize industrial

production of alpha-galactosidase. Fractional factorial

design (FFD) was applied to identify the most important

components in the media makeup, then focused on the

critical subset of media components, and finally used cen-

tral composite design (CCD) to examine and optimize

the culture media for alpha-alactosidase production by

Aspergillus foetidus ZU-G1 in solid state substrate on

wheat bran. The culture medium on alpha-galactosidase

production in laboratory scale may have a potentiality of

scaling-up.

Materials and methods

The Aspergillus sp. strain ( Aspergillus foetidus ZU-G1)

was isolated from the Chinese traditional soy sauce

mash and identified as Aspergillus foetidus according to

Wei (1979) and Qi (1997). Then used 60Co c-ray and

low energy N+ ion beam implantation to mutate the

strain, and obtained a positive mutant strain with high

alpha-galactosidase activity. It was preserved at the

China General Microbiological Culture Collection Center(CGMCC no.1628).

The seed medium was PDA. The basic fermentation

medium in alpha-galactosidase production was composed

of (w w )1, on dry basis): 83Æ21% wheat bran (WB),

16Æ64% soybean meal (SM), 0Æ05% FeSO4Æ7H2O and 0Æ1%

K2HPO4. Distilled water was added in such a way that

the final substrate moisture content was 60%. After auto-

claved sterilization at 121°C for 20 min, the medium was

cooled down for further inoculation. Cultivation in SSF

was carried out in 250 ml shake flask and an incubator

where temperature could be controlled automatically,

incubating at 28°C for 96 h.

Crude enzyme extraction

After cultivation, the fermented medium was stirred. One

gram of the fermented medium ground sample was

soaked with 50 ml of 0Æ1 mol l)1 McIlvaine buffer (pH

5Æ0), and disrupted by a vibrator in water bath at 40°C

for 30 min. Then it was filtered by filter paper for further

analysis.

Enzyme assays

Alpha-galactosidase assay was carried out in test tubes by

the modified version of the method of Garroa et al.

(2004). The reaction mixture contained: 0Æ01 mol l)1

pNPG, 50 ll; 0Æ1 mol l)1 McIlvaine buffer pH 5Æ0, 50 ll;

cell-free extract, 100 ll; final volume: 200 ll. The mixturewas incubated at 50°C for 10 min, and the reaction was

stopped by adding 3 ml of sodium carbonate

(0Æ25 mol l)1). One enzyme unit (U) was defined as the

amount of enzyme that released 1Æ0 lmol of pNP from

its substrate pNPG per min under the given assay condi-

tions. The results are expressed as U g)1 dry matter.

Experimental design

To obtain a suitable medium for enzyme production in

SSF, a series of statistically designed studies were conduc-

ted to investigate the effect of various media components

on alpha-galactosidase activity. The optimization process

firstly entails identifying the preferable nutrient (carbon

sources, nitrogen sources and essential elements) for

alpha-galactosidase production by varying one factor at a

time while keeping the others constant, then focuses on

the most important components in the media ingredients

using FFD, and then focuses on the critical subset of

media components, finally a CCD was derived to opti-

mize the critical components and maximize the alpha-ga-

lactosidase activity.

In the FFD, a 26)2 FFD leading to 16 sets of experi-

ments, performed in duplicate, was used to verify the

most significant factors affecting the alpha-galactosidaseactivity. The variables were coded according to the fol-

lowing equation:

x i ¼ X iÀ X 0

D X ið1Þ

where x i is the coded value of an independent variable, X ithe real value of an independent variable, X 0 the real

value of an independent variable at the center point, and

D X i is the step change value.

The range and the level of the variables both coded val-

ues and natural values investigated in this study were

given in Table 1. The alpha-galactosidase activity was

considered as the dependent variable or response (Y i).

The first-order model was obtained from FFD. To fit the

empirical second-order polynomial model, a CCD with

five coded levels was performed. The model proposed for

the response (Y ) was:

y ¼ b0 þX

bi x i þX

bii x i2 þX

bij x j x j ð2Þ

where y is the response variable, b0, bi, bii, bij are the

regression coefficients variables, for intercept, linear,

C.Q. Liu et al. Optimization of alpha-galactosidase in SSF

ª 2007 The Authors

Journal compilation ª 2007 The Society for Applied Microbiology, Letters in Applied Microbiology 45 (2007) 206–212 207

7/28/2019 21030243


quadratic and interaction terms, respectively, and x i and

x j are independent variables. Data were analysed using the

response surface regression (RSREG) procedure (SAS

Institute Inc., Cary, NC, USA) and x is the coded level of

the independent variable.

Results

Effect of different carbon sources on alpha-galactosidase

production

The carbon source employed in microbial enzyme pro-

duction is one of the most important factors. To study

the effect of carbon sources on alpha-galactosidase pro-

duction, cultivations were performed with 66Æ57% wheat

bran, 16Æ64% soybean meal, 0Æ1% K2HPO4, 0Æ05% FeS-

O4Æ7H2O and 16Æ64% various carbon (lactose, glucose,

sucrose, corn flour and rice flour) sources. As shown in

Fig. 1, the alpha-galactosidase activity varied in a range of

1072Æ53$1772Æ34 U g)1. The maximal alpha-galactosidase

activity by A. foetidus ZU-G1 was found with wheat bran.Corn flour, glucose and rice flour were clearly inferior.

The lowest values of alpha-galactosidase activity were

obtained when lactose was used.

Effect of different nitrogen sources on

alpha-galactosidase production

The type of nitrogen source to be used depends essen-tially upon the nutritional requirement of the organism.

The effects of 16Æ64% organic and inorganic nitrogen

sources were further studied at constant carbon source

(83Æ21% wheat bran) and results illustrated in Fig. 2. The

alpha-galactosidase activity varied in a range of

57Æ42$1777Æ56 U g)1.It was found that soybean meal was

most effective for alpha-galactosidase production

(1777Æ56 U g)1). This result is identical with Kotwal et al.

(1997) report. Urea, (NH4)2HPO4, and NaNO3 were poor

nitrogen sources for alpha-galactosidase production.

Effect of essential elements on enzyme production

Apart from C and N sources, many other essential ele-

ments such as magnesium, calcium and trace elements

may be required in the medium to support active cellular

function. The effects of essential elements on the produc-

tion of alpha-galactosidase by Aspergillus foetidus ZU-G1

under the best carbon and nitrogen sources were illustra-

ted in Fig. 3. Results demonstrated 0Æ1% KH2PO4, 0Æ05%

MnSO4ÆH2O and 0Æ05% CuSO4Æ5H2O were favourable for

alpha-galactosidase production (1654Æ17, 1815Æ45 and

1616Æ05 U g)1, respectively), and MnSO4ÆH2O was the

most effective. But 0Æ1% CaCl2Æ2H2O was negative effect

on alpha-galactosidase production for this strain(1327Æ59 U g)1). Thus, wheat bran, soybean meal,

Table 1 The coded and uncoded values of factors in FFD

Independent variables

Level

)1 0 +1

Wheat bran (g 10 g)1, X 1) 3Æ89 8Æ23 12Æ57

Soybean meal(g 10 g)1, X 2) 0Æ77 1Æ65 2Æ523

KH2PO4(g 10 g)1

, X 3) 0 0Æ01 0

Æ02

MnSO4ÆH2O(g 10 g)1, X 4) 0Æ01 0Æ05 0Æ09

CuSO4Æ7H2O(g 10 g)1, X 5) 0Æ01 0Æ05 0Æ09

2000

1800

1600

1400

1200

A l p h a - g a l a c t o s i d a s e a c t i v i t y ( U

g – 1 )

1000

800

600

400200

0 L a c t o s e

G l u c o s e

S u c r o s e

C o r n f l o u r

R i c e f l o u r

C o n t r o l

Figure 1 Effect of various carbon sources on alpha-galactosidase pro-

duction. The medium consisted of wheat bran 66 Æ57%, soybean flour

16Æ64%, K2HPO4 0Æ1%, FeSO4Æ7H2O 0Æ05% and various carbon

16Æ64%

2000

1800

1600

1400

1200

A l p h a

- g a l a c t o s i d a s e a c t i v i t y ( U

g – 1 )

1000

800

600

400200

0S o y a b e a n f l o u r

P e p t o n e

U r e a N a N O

3

( N H 4 ) 2 H

P O 4

C o n t r o l

Figure 2 Effect of nitrogen source on alpha-galactosidase produc-

tion. The medium consisted of wheat bran 66Æ57%, K2HPO4 0Æ1%,

FeSO4Æ7H2O 0Æ05% and nitrogen source 16Æ64%.

Optimization of alpha-galactosidase in SSF C.Q. Liu et al.


ª 2007 The Authors

7/28/2019 21030243


KH2PO4, MnSO4ÆH2O and CuSO4Æ5H2O were selected for

the design of FFD.

The optimization of culture medium for alpha-galactosi-

dase production by Aspergillus foetidus ZU-G1

To obtain maximal alpha-galactosidase yield, factorial

design approach was used. The factorial design approach

to medium development relies on three stages of experi-

mentation: screening, optimization and verification.

Screening aims at problem reducing a determination as to

which few process variables have the greatest impact on

performance. Optimization experiments are designed to

provide in-depth information about a few variables iden-

tified during screening as having the greatest impact on

performances. Finally, verification experiments are used

to validate the results under specific experimental condi-

tions.

Screening experiments: the FFD and analysis

The screening experiments were designed to evaluate the

impact of five factors, concentrations of wheat bran ( x 1),

soybean meal ( x 2), KH2PO4 ( x 3), MnSO4ÆH2O ( x 4), and

CuSO4Æ5H2O ( x 5). A two-level FFD was employed and

the results of the FFD are shown in Tables 1 and 2.

Results suggest that these variables significantly affect

alpha-galactosidase activity. The anova result showed

that wheat bran ( x 1) and soybean meal ( x 2) were signifi-

cant factors at probability level of 95% for alpha-galac-

tosidase production. However, x 3, x 4 and x 5 factors were

found to be insignificant at probability of 90% for alpha-

galactosidase production. Consequently, the wheat branand soybean meal were selected to be further optimized

in the following experiment. There is no evidence of any

interactions. The predicted regression equation was as fol-

lows (Eqn. 2):

y ¼ 938Á59 À 200Á14 x 1 þ 164Á03 x 2 À 85Á53 x 3 À 20Á95 x 4

þ 23Á49 x 5 ð2Þ

The results of t -test for variance between average of

observation of two-level experiment and centre point

showed that the difference was significant (P < 0Æ01). This

result indicated that optimum point was in the domain

of our experiment.

The CCD and response surface analysis

A response surface design is appropriate once the optimal

range for running the process has been identified. Further

optimization of medium components was carried out

using a Box–Wilson CCD with four-star points and five

replicates at centre point for each of wheat bran ( X 1) and

20000·10%

0·05%

CK1800

1600

1400

1200

A l p h a - g a l a c t o s i d a s e a c t i v i t y ( U

g – 1 )

1000

800

600

400

200

0 K H 2 P O

4

C a C l 2 · 2 H

2 O

N a 2 H P O

4

M n S O

4 · H 2 O

Z n S O 4 · 7 H

2 O

M g S O

4 · 7 H 2 O

C u S O 4 · 5 H

2 O

F e S O 4 · 7 H

2 O

C o n t r o l

Figure 3 Effect of essential elements on

alpha-galactosidase production.The medium

consisted of wheat bran 66Æ57% , soybean

flour 16Æ64%.


ª 2007 The Authors


7/28/2019 21030243


soybean meal ( X 2). The coded and uncoded values of fac-

tors in CCD are showed in Table 3. Experimental design

and results are shown in Table 4.

Three-dimensional response surface plot of wheat bran

and soybean meal against alpha-galactosidase can further

explain the results of the statistical and mathematical ana-

lysis. The response surface open its mouth downwards

demonstrated there must be a maximum in the stable

range (Fig. 4). In light of multi-regressive-analysis of the

central composite experiments showed in Table 5, the

second-order polynomial prediction model was obtained

as eqn (3). Table 5 and eqn. (3) showed positive effects

of x 2 and x 1 x 2, and negative effects of x 1, x 12 and x 2

2

against enzyme production.

Table 2 The design and results of FFD for alpha-galactosidase production

Run

x 1

(WB)

x 2

(SM)

x 3

(KH2PO4)

x 4

(MnSO4ÆH2O)

x 5

(CuSO4Æ7H2O)

Alpha-galactosidase activity

(U g)1 dry matter)

1 +1 (12Æ57) +1 (2Æ523) )1 (0) +1 (0Æ09) )1 (0Æ01) 1773Æ19

2 )1 (3 Æ89) +1 (2Æ523) )1 (0) +1 (0Æ09) +1 (0Æ09) 1679Æ30

3 )1 (3 Æ89) )1 (0Æ77) +1 (0Æ02) +1 (0Æ09) +1 (0Æ09) 1797Æ50

4)

1 (3Æ89) +1 (2

Æ523) +1 (0

Æ02)

)1 (0

Æ01) +1 (0

Æ09) 1966

Æ47

5 +1 (12Æ57) +1 (2Æ523) +1 (0Æ02) +1 (0Æ09) +1 (0Æ09) 2150Æ08

6 +1 (12Æ57) )1 (0Æ77) )1 (0) +1 (0Æ09) +1 (0Æ09) 1873Æ69

7 )1 (3 Æ89) )1 (0Æ77) +1 (0Æ02) )1 (0Æ01) )1 (0Æ01) 1933Æ99

8 )1 (3 Æ89) +1 (2Æ523) +1 (0Æ02) +1 (0Æ09) )1 (0Æ01) 2585Æ93

9 +1 (12Æ57) )1 (0Æ77) +1 (0Æ02) )1 (0Æ01) +1 (0Æ09) 2174Æ17

10 +1 (12Æ57) )1 (0Æ77) )1 (0) )1 (0Æ01) )1 (0Æ01) 2287Æ66

11 )1 (3 Æ89) +1 (2Æ523) )1 (0) )1 (0Æ01) )1 (0Æ01) 1882Æ95

12 +1 (12Æ57) )1 (0Æ77) +1 (0Æ02) +1 (0Æ09) )1 (0Æ01) 1572Æ32

13 +1 (12Æ57) +1 (2Æ523) +1 (0Æ02) )1 (0Æ01) )1 (0Æ01) 1973Æ01

14 +1 (12Æ57) +1 (2Æ523) )1 (0) )1 (0Æ01) +1 (0Æ09) 1876Æ16

15 )1 (3 Æ89) )1 (0Æ77) )1 (0) +1 (0Æ09) )1 (0Æ01) 1047Æ17

16 )1 (3 Æ89) )1 (0Æ77) )1 (0) )1 (0Æ01) +1 (0Æ09) 1429Æ58

17 0 (8Æ23) 0 (1Æ65) 0 (0Æ01) 0 (0Æ05) 0 (0Æ05) 2100Æ40

18 0 (8Æ23) 0 (1

Æ65) 0 (0

Æ01) 0 (0

Æ05) 0 (0

Æ05) 2324

Æ01

19 0 (8Æ23) 0 (1Æ65) 0 (0Æ01) 0 (0Æ05) 0 (0Æ05) 2117Æ34

20 0 (8Æ23) 0 (1Æ65) 0 (0Æ01) 0 (0Æ05) 0 (0Æ05) 2226Æ92

x 1 = ( X 1)8Æ23) ⁄ 4Æ34; x 2 = ( X 2)1Æ65) ⁄ 0Æ876; x 3 = ( X 3)0Æ01) ⁄ 0Æ05; x 4 = ( X 4)0Æ05) ⁄ 0Æ04; x 5 = ( X 5)0Æ05) ⁄ 0Æ04

Table 3 The coded and uncoded values of factors in CCD

Independent variables

Level

)1Æ414 )1 0 +1 +1Æ414

Wheat bran (g 10 g)1, X 1) 0Æ87 3Æ02 8Æ23 13Æ43 15Æ59

Soybean meal(g 10 g)1, X 2) 0Æ53 0Æ86 1Æ65 2Æ44 2Æ77

Table 4 Experimental design and the results of CCD

Run x 1 (WB) x 2 (SM) Y (U g)1 dry matter)

1 0 (8Æ23) 0 (1Æ65) 2196Æ19

2 0 (8Æ23) 0 (1Æ65) 2214Æ90

3 +1 (13Æ43) )1 (0Æ86) 1720Æ88

4 )1 (3Æ02) )1 (0Æ86) 1929Æ87

5 0 (8Æ23) )1Æ414 (0Æ53) 1872Æ98

6 0 (8Æ23) 0 (1Æ65) 2210Æ80

7 + 1 (13Æ43) + 1 (2Æ44) 1982Æ81

8 0 (8Æ23) + 1Æ414 (2Æ77) 1930Æ88

9 )1Æ414 (0Æ87) 0 (1Æ65) 1981Æ84

10 0 (8Æ23) 0 (1

Æ65) 2200

Æ80

11 1Æ414 (15Æ59) 0 (1Æ65) 1712Æ33

12 0 (8Æ23) 0 (1Æ65) 2209Æ09

13 )1 (3Æ02) +1 (2Æ44) 1822Æ49

x 1 = ( X 1)8Æ23) ⁄ 5Æ21; x 2 = ( X 2)1Æ65) ⁄ 0Æ79.

2210·39

2081·20

1952·02

1822·83

1693·65

Y ( U

g – 1

d r y m a t t e r )

1·000·50

0·00–0·50

–1·00 –1·00–0·50

0·00

X1(wheat bran)X2(soybean flour)

0·501·00

Figure 4 The response surface plot of wheat bran ( X 1) and soybean

meal ( X 2) against alpha-galactosidase activity in solid-state fermenta-

tion.



ª 2007 The Authors

7/28/2019 21030243


The statistical significance of the second-order model

equation was checked by an F-test (anova) and data

showed in Table 6. The fit value, termed R2 (determinant

coefficient), of the polynomial model was calculated to be

0Æ9654, indicating that 96Æ54% of the variability in the

response could be explained by the second-order polyno-

mial prediction equation given below (Eq. 3).

Y ¼ 2206Á36 À 53Á73 x 1 þ 29Á56 x 2 À 182Á29 x 12 þ 92Á33 x 1 x 2

À 154Á86 x 22 ð3Þ

The anova results showed that this model was appropri-ate. It was also suggested that the quadratic and linear

terms of wheat bran and soybean meal of the model pri-

marily determined alpha-galactosidase production by

Aspergillus foetidus ZU-G1.

The three-dimensional graph obtained from the calcu-

lated response surface plot is shown in Fig. 4. It is evident

from the plot that alpha-galactosidase activity reached

maximum at a combination of coded level )0Æ094248

( x 1), and 0Æ039393 ( x 2). This was a reconfirmation that

the fitted surface had a maximum point, which was

8Æ2137 g wheat bran and 1Æ7843 g soybean meal in 10 g

dry matter. The model predicted a maximum response of

alpha-galactosidase for this point, and alpha-galactosidase

activity reached 2210Æ76 U g)1 dry matter after 96 h incu-

bation in the optimum fermentation medium, namely

8Æ2137 g wheat bran, 1Æ7843 g soybean meal, 0Æ001 g

MnSO4ÆH2O and 0Æ001 g CuSO4Æ5H2O in 10 g dry matter

fermentation medium. To confirm these results, verified

experiments were performed using a medium representing

this maximum point, and a mean value of 2207Æ19 U g)1

dry matter (n = 3) was obtained. It suggests that the opti-

mized model is proper and effective for explaining the

actual cultivation process.

The time course of alpha-galactosidase production by

Aspergillus foetidus ZU-G1 cultivated in medium with

optimized compositions

The time course of alpha-galactosidase production by

Aspergillus foetidus ZU-G1 cultivated in optimized med-

ium is shown in Fig. 5. Sampling per 12 h was prepared

for assay. From 36 h onwards there was no substantial

increase in alpha-galactosidase activity. After 36 h, its

activity increased and reached its maximum value up to

the 144th hour (2572Æ53 U g

)1

), then decreased. Withextending cultivation time the yield of alpha-galactosidase

had decreasing trend. The concentration of residual

reduced sugar (RRS) sharply increased from the begin-

ning of the fermentation and reached to maximum at

24 h (586Æ7 mg g)1), then maintained somewhat changed

level (135$85 mg g)1) till the end of fermentation.

Discussion

Response surface methodology was proved to be a power-

ful tool in optimizing the fermentation medium for

alpha-galactosidase synthesis from Aspergillus foetidus ZU-

G1 in SSF. In the present study, the experimental results

clearly showed that wheat bran, soybean meal,

MnSO4ÆH2O and CuSO4Æ7H2O affect alpha-galactosidase

production. It is also evident from the experimental

results that wheat bran and soybean meal are very crucial

ones. Wheat bran is a better carbon source probably due

to the presence of various suitable nutrients in wheat

bran and ⁄ or due to its most suitable particle size and

consistency required for anchorage and enzyme excretion

Table 5 Results of parameter estimate for Y of CCD using RSREG

processing

Parameter Estimate Std error of estimate Pr>|T|

Intercept 2206Æ36 20Æ65 0Æ0000

x 1 )53Æ73 16Æ33 0Æ0133

x 2 29Æ56 16Æ33 0Æ1132

x 12 )182

Æ29 17

Æ51 0

Æ0000

x 2 x 1 92Æ33 23Æ09 0Æ0052

x 22

)154Æ86 17Æ51 0Æ0000

Table 6 Analysis of variance (ANOVA) for the second-order model

Regression Freedom Sum of squares R-square F-ratio P > F

Linear 2 30078 0Æ0697 7Æ051 0Æ0210

Quadratic 2 352652 0Æ8168 82Æ673 0Æ0000

Crossproduct 1 34096 0Æ0790 15Æ987 0Æ0052

Total regress 5 416826 0Æ9654 39Æ087 0Æ0001

700 3000

2500

2000

1500

1000

500

0

600

500

400

300

R

R S ( m g g – 1 )

200

100

0

0 24 48 72 96

Time(h)

RRS Alpha-galactosidase

A l p h a - g a l a c t o s i d a s e ( U

g – 1 )

120 144 168 192

Figure 5 The change in residual reduce sugar (RRS) and alpha-galac-

tosidase activity using optimized culture medium in solid-state fermen-

tation.


ª 2007 The Authors


7/28/2019 21030243


(Singh and Soni 2001). Wheat bran gives the substrate

good porosity. Moreover, it is cheaper that was chose as

carbon source for alpha-galactosidase production. Enzyme

production was the maximum when soybean meal was

used as nitrogen source. It probably was due to abundant

galactooligosaccharides in SM induced alpha-galactosidase

biosynthesis (Cruz and Park 1982). 0Æ1% KH2PO4 was

favourable for alpha-galactosidase production but no sig-

nificant impact when combined with MnSO4ÆH2O and

CuSO4Æ5H2O, this may be considered as that culture

medium contains enough nutrient for alpha-galacosidase

production.

In the present study, alpha-galactosidase activity

reached 2207Æ19 U g)1 dry matter in 250 ml shake flask

after incubating for 96 h in the optimum fermentation

medium, The optimized fermentation medium for alpha-

galactosidase production by Aspergillus foetidus ZU-G1

composed of 8Æ2137 g wheat bran, 1Æ7843 g soybean meal,

0Æ001 g MnSO4ÆH2O a nd 0Æ001 g CuSO4Æ5H2O in 10 g

dry matter fermentation medium. The maximum alpha-

galactosidase production was at 144 h cultivation

(2572Æ53 U g)1) under the optimized fermentation med-

ium. The result was the same as that for submerged fer-

mentation (data not shown). This result confirmed that

highest activity of enzyme would not occur until a suffi-

cient biomass was formed, and that further extending

incubation time had no effect on improving the enzyme

yield because of nutrient utilization by fungi with the

process.

To explore whether the optimized model from RSM

could be applied to a large scale synthesis, the reaction

volume was increased by a factor of 10. Results gave2452Æ74 U g)1 of alpha-galactosidase activity at 144 h cul-

ture broth, which indicated no significant difference

between large-scale and small-scale synthesis. The trend

of RRS concentration was coincident with the results of

submerged fermentation (data not published) and the He

et al. (2003) report, but different from those reported by

Rao et al. (2006). This may be aroused by micro-organ-

ism hydrolysing wheat bran and produced RRS.

References

Chahal, D.S. (1985) Solid-state fermentation with Trichoderma

reesei for cellulase production. Am Soc Microbiol 49,

205–210.

Cruz, R. and Park, Y.K. (1982) Production of fungal alpha-ga-

lactosidase and its application to the hydrolysis of galac-

tooligosaccharides in soybean milk. J Food Sci 47, 1973–

1975.

Garroa, M.S., Valdeza, G.F. and Gioria, G.S. (2004) Tempera-

ture effect on the biological activity of Bifidobacterium lon-

gum CRL 849 and Lactobacillus fermentum CRL 251 in

pure and mixed cultures grown in soymilk. Food Microbiol

21, 511–518.

Gitzelmann, R. and Auricchio, S. (1965) The handling of soy

alpha-galactosidase by a normal and galactosemic child.

Pediatrics 36, 31–235.

Gote, M., Umalkar, H., Khan, I. and Khire, J. (2004) Thermo-

stable alpha-galactosidase from the Bacillus stearothermo-

philus (NCIM 5146) and its application in the removal of

flatulence causing factors soymilk. Process Biochem 39,

1723–1729.

He, G.Q., Chen, Q.H., Zhang, L. and Liu, X.J. (2003) Influence

of medium components on elastase production using

crude sources by Bacillus sp.EL31410. J Zhejiang Univ Sci

4, 142–155.

Kotwal, S.M., Gote, M.M., Sainkar, S.R., Khan, M.I. and Khire,

J.M. (1997) Production of alpha-galactosidase by thermo-

philic fungus Humicola sp. in solid-state fermentation and

its application in soyamilk hydrolysis. Process Biochem 33,

337–343.

Li, X.H., Chen, S.M., Jia, X.M., Yao, X.H., Xu, S.C. and Fei,

D.B. (2001) Production of alpha-galactosidase from

Penicillum sp. by solid-state fermentation. Acta Agric

Zhejiangensis 5, 305–308.

Nigam, P. and Singh, D. (1994) Recent process developments

in solid-state substrate systems and their applications in

biotechnology solid-state fermentation. J Basic Microbiol

34, 405–423.

Qi, T.Z. (1997) Flora Fungorum Sinicorum (Vol.5)-Aspergilluset Teleomorphi Cognati Chinese Edition with Latin Name

Index . Beijing: Science Press of China.

Rao, Y.K., Lu, S.C., Liu, B.L. and Tzeng, Y.M. (2006)

Enhanced production of an extracellular protease from

Beauveria bassiana by optimization of cultivation proces-

ses. Biochem Engin J 28, 57–66.

Singh, H. and Soni, S.K. (2001) Production of starch-gel

digesting amyloglucosidase by Aspergillus oryzae HS-3 in

solid-state fermentation. Process Biochem 37, 453–459.

Wang, C.L., Li, D.F., Lu, W.Q., Wang, Y.H. and Lai, C.H.

(2004) Influence of cultivating conditions on the alpha-ga-

lactosidase biosynthesis from a novel strain of Penicillium

sp. in solid-state fermentation. Lett Appl Microbiol 39

,369–375.

Wei, J.C. (1979) Manual of Fungi Identification. Shanghai:

Shanghai Scientific & Technical Publishers.



ª 2007 The Authors

21030243

Documents