Botanical Studies (2006) 47: 167-174

*

Corresponding author: E-mail: Moustafa64@Yahoo.com

Application of Plackett-Burman factorial design to

improve citrinin production in Monascus ruber batch

cultures

Abdulaziz Q. M. AL-SARRANI

1

and Moustafa Y. M. EL-NAGGAR

2,

*

1

Biology Department, Microbiology Division, Faculty of Science, Taibah University, P.O. Box 344, Al-Madinah Al-

Munawarah, Kingdom of Saudi Arabia

2

Botany Department, Microbiology Division, Faculty of Science, Alexandria University, Moharram Bay 21511,

Alexandria, Egypt

(Received August 17, 2005; Accepted November 28, 2005)

ABSTRACT.

A fungal strain previously isolated from a hot sauce of red pepper putatively identified as

Monascus ruber was investigated for its potential to produce an antibacterial substance in batch cultures.

This fungus was grown in three liquid media. The highest antibacterial activity was recorded for the synthetic

medium No. 2. The antibacterial substance was extracted, purified and identified as the mycotoxin citrinin.

Plackett-Burman experimental design was applied to optimize the components of medium No. 2 in an attempt

to improve citrinin production for non-food applications. As a result a medium of the following formula was

predicted to be near the optimum for producing an extracellular citrinin in the culture filtrate of M. ruber (g/l):

glucose, 45; NaNO

3

, 3.75; KH

2

PO

4

, 0.2; MgSO

4

·7H

2

O, 0.1 and ZnSO

4

·7H

2

O, 0.01; MnSO

4

·H

2

O, 0.006 and

FeSO

4

·7H

2

O, 0.003. The improvement of the antibacterial activity amounted to 1.75-fold. On the other hand,

the dry weight of the fungus in the optimized medium was 1.86 times the figure recorded for the basal setting.

The HPLC determination of citrinin concentration in the culture supernatant amounted to 220 mg/l and 400

mg/l before and after medium optimization, respectively.

Keywords: Batch cultures; Citrinin; Monascus ruber; Optimization; Plackett-Burman experimental design.

INTRODUCTION

Monascus is an ascomycetous fungus discovered by

van Tieghem (1884) traditionally used for the produc-

tion of food colouring, fermented foods and beverages

(Martinkova and Patakova, 1999). Other components of

Monascus pigments have anti-inflammatory activity and

have been reported to suppress skin cancer caused by

tumour promoters in experimental animals (Yasukawa et

al., 1994 and 1996), in addition to their clinical benefits

in treating high blood pressure in humans (Kushiro et al.,

1996) In Saudi Arabia, Monascus ruber has been isolated

from the hot sauce of red pepper (Al-Sarrani, 1998). In

Greece, the same fungus has been isolated from the brine

of thermally-processed green olives of the Conservolea

variety (Panagou et al., 2003). It was shown that M. ru-

ber is able to produce the mycotoxin citrinin (Blanc et al.,

1995a). This mycotoxin has antibiotic activities against

Gram-positive bacteria, but its nephrotoxic properties

(Blanc et al., 1995a) have limited its use as an antibiotic

for therapeutic purposes. Consequently, the production of

citrinin together with the red pigments rules out the use of

M. ruber as a producer of natural colourants for food tech-

nology (Blanc et al., 1995a; Hajjaj et al., 1997).

Nutritional and fermentation conditions have a great

influence on antibacterial activity (El-Naggar et al., 2001).

The application of statistically based experimental designs

to optimize fermentation media is an efficient approach

to studying the effects of several factors and to improve

product yields. The conventional practice of single factor

optimization by maintaining other factors involved at an

unspecified constant level does not depict the combined

effect of all factors involved (Elibol, 2004). Factorial de-

sign considers the statistical interaction between variables

to obtain a maximum of inferences for minimum of tests,

reducing process variability, time of development and

overall costs.

This work is meant to optimize the fermentation me-

dium composition of Monascus ruber for citrinin (Figure

1) production to be used for toxicological studies and as a

reference for analytical purposes. Factors affecting these

objectives were evaluated by the application of a two-level

factorial Plackett-Burman design (Plackett and Burman,

1946).

PHYSIOLOGY

168

Botanical Studies, Vol. 47, 2006

MATERIALS AND METHODS

Microorganism and growth conditions

The filamentous fungus used in this study was previ-

ously isolated from the hot sauce of red pepper Capsium

annum (Al-Sarrani, 1998). This fungus was identified as

Monascus ruber Tieghem according to the taxonomy pro-

posed by Hawksworth and Pitt (1983). The identification

was confirmed by the Commonwealth Mycological Insti-

tute, Great Britain (code number IMI 364278). Monascus

ruber was kept on slants of potato dextrose agar (PDA,

Difco Laboratories, Detroit, Mich., USA). Spores were

prepared by growing the fungus on PDA slants for 10 days

at 30°C. Spores were washed with sterile distilled water

and a spore suspension of 2.5 × 10

7

spores was used to

inoculate each of the Erlenmeyer flasks (250 ml) which

contained 50 ml culture medium for the production of

citrinin. The flasks were then incubated at 30°C, 240 rpm

using shaker incubator for 5 days. A time scale experiment

was carried out, which revealed that maximum growth of

this fungus and mycotoxin production were attained at day

5 (data not shown). In the present work, three fermenta-

tion media were originally used to study their effect on the

production of citrinin by M. ruber:

Medium 1. The synthetic medium classically used

for the production of red pigments (Fabre et al., 1993)

composed of monosodium glutamate, 5 g; K

2

HPO

4

,

5 g ; K H

2

PO

4

, 5 g; MgSO

4

·7H

2

O, 0.5 g; CaCl

2

, 0.5g;

FeSO

4

·7H

2

O, 0.5 g; ZnSO

4

·7H

2

O, 0.01 g, MnSO

4

·H

2

O,

0.03 g; ethanol, 20 g/l deionized water.

Medium 2. The synthetic medium designated for

the production of pigments (El-Naggar et al., 2000)

composed of glucose, 40 g; NaNO

3

, 3 g; KH

2

PO

4

, 0.1 g;

MgSO

4

·7H

2

O, 0.05 g; FeSO

4

·7H

2

O, 0.001 g; ZnSO

4

·7H

2

O,

0.008 g, MnSO

4

·H

2

O, 0.003 g/l deionized water.

Medium 3. A modified yeast extract-sucrose (YES)

medium, specifically used for the production of fungal

toxins (Davis et al., 1975) composed of yeast extract, 40 g

and sucrose, 160 g/l deionized water.

The initial pH of each fermentation medium was

adjusted to 6.5 with NaOH or HCl.

Antibacterial activity

A culture broth (10 ml) was centrifuged in order to sep-

arate the mycelium and the supernatant. The supernatant

was concentrated 10-fold and used for the determination

of antibacterial activity. Thirty μl were placed in each hole

(6 mm of diameter) in the Mueller-Hinton agar medium

in a Petri-dish inoculated with 0.1 ml bacterial suspension

(3 × 10

6

cfu/ml) of one of the selected indicator bacterial

strains (Bacillus subtilis ATCC 6633, B. cereus PX MU-

COB, Streptococcus lactis, or Pseudomonas fluorescens).

S. lactis and P. fluorescens are both local isolates from the

Hope Hospital, Salford, England. Petri dishes were kept

for 2 h in the refrigerator and then incubated for 12 h at

35

o

C and the inhibition zone diameter was then measured.

This experiment was performed in triplicates and repeated

twice, and the mean was considered the response.

Dry weight determination

The fungal biomass (mg/ml) was determined after

filtration the sample culture through pre-weighed membrane

filter (45-μm Millipore, Millipore Corp., Beford, Mass.,

USA) and washed with sterile distilled water. The mycelia

were dried at 80oC to a constant weight.

Plackett-Burman experimental design

The Plackett-Burman experimental design, a fractional

factorial design, was used in this work to demonstrate the

relative importance of medium components on citrinin

production and growth of M. ruber. Seven independent

variables (Table 1) in eight combinations were organized

according to the Plackett-Burman design matrix (Table

2). For each variable, a high (+1) and low (-1) level was

tested. All trials were performed in triplicate, and each

experiment was repeated twice, with the mean considered

the response. The main effect of each variable was deter-

mined according to the following equation:

E

xi

= (

Σ

M

i+

-

Σ

M

i-

) / N

where E

xi

is the variable main effect, M

i+

and

M

i-

are the

response percentage in trials, in which the independent

variable (xi) was present in high and low concentrations,

respectively, and N is the half number of trials. Using

Microsoft Excel, statistical t-values for equal unpaired

samples were calculated for the determination of variable

significance.

Isolation, purification and characterization of

citrinin

The culture supernatant (10 litres) was separated from

the mycelia using a Chilspin MSE Fisons centrifuge at

4

o

C and 5000 rpm for 15 min. The supernatant was then

acidified to pH 5.0 and extracted with ethyl acetate. The

aqueous layer was removed and the organic layer was

concentrated and applied to a silica gel 60 preparative

TLC plate (using ethyl acetate/acetone/water, 4:4:1 v/v/v)

as a solvent system. The plates were examined under ul-

traviolet light at 350 nm according to Blanc et al. (1995a).

Figure 1. Structure of citrinin.

AL-SARRANI and EL-NAGGAR — Improvement of citrinin production by

M. ruber

169

A fluorescent pale yellow spot (R

f

=0.6) was detected.

The active compound was removed from the plate and

dissolved in methanol and again purified with a Hewlett-

Packard 1090A HPLC instrument. A Spherisorb C18, 5

μm (25 cm by 4.6 mm) column was eluted with metha-

nol /water (20:80, v/v) at a flow rate of 1.0 ml/min. The

concentration of citrinin was also measured in the culture

supernatant by the same method to make sure that citrinin

was responsible for the antibacterial activity.

Nuclear magnetic resonance (NMR) spectroscopy

1

H NMR spectrum (CD

3

OD) was recorded on Varian

Oxford 300 MHz NMR spectrophotometer.

RESULTS

The antibacterial activity in M. ruber culture

After 5 days of M. ruber incubation (a time scale ex-

periment revealed that maximum growth of this fungus

and mycotoxin production were attained at day 5, data not

shown), the antibacterial activity of the crude supernatant

of M. ruber from the three culture media was determined

using B. subtilis, B. cereus, S. lactis and P. fluorescens as

selected test organisms (Table 3). Monascus ruber showed

a varied antibacterial activity pattern (measured as inhibi-

tion zone diameter in mm) when cultured in media 1, 2

and 3, but medium No. 2 recorded the highest activity.

When the concentration of citrinin was estimated in the

culture filtrate using HPLC instrument, medium No. 1

recorded 160 mg/l, and medium No. 2 recorded 220 mg/l,

compared to 180 mg/l for medium No. 3. Consequently,

medium No. 2 will be further optimized using Plackett-

Burman experimental design in an attempt to improve

citrinin production by M. ruber in batch cultures.

Optimization of medium components for citrinin

production by M. ruber

The application of a complete factorial design would

require 2

n

experiments if n factors have to be investi-

gated. Thus, seven variables would lead to 128 trials, a

huge number. However, using the factorial design without

losing information about the main effect of variables could

reduce the number of experiments. Seven levels of culture

variables were examined in the Plackett-Burman design

matrix with 9 different trials. Trial No. 9 represented the

original medium composition. The highest citrinin activity

Table 1. Experimental range and levels of independent variables in the Plackett-Burman experiment.

Variable

Level (g/l)

-1

0

1

Glucose

35.00

40.00

45.00

NaNO

3

2.75

3.00

3.75

KH

2

PO

4

0.05

0.10

0.2

MgSO

4

·

7H

2

O

0.025

0.05

0.1

ZnSO

4

·

7H

2

O

0.006

0.008

0.01

MnSO

4

·

H

2

O

0.001

0.003

0.006

FeSO

4

·

7H

2

O

0.0008

0.001

0.003

Table 2. The Plackett-Burman design matrix representing the coded values for 7 independent variables.

Trial No.

Factor

Glucose

NaNo

3

KH

2

PO

4

MgSO

4

ZnSO

4

MnSO

4

FeSO

4

1

-1

-1

-1

1

1

1

-1

2

1

-1

-1

-1

-1

1

1

3

-1

1

-1

-1

-1

-1

1

4

1

1

-1

-1

1

-1

-1

5

-1

-1

1

1

-1

-1

1

6

1

-1

1

1

1

-1

-1

7

-1

1

1

-1

-1

1

-1

8

1

1

1

1

1

1

1

9

0

0

0

0

0

0

0

0 represents the original concentration of each component in the medium No. 2 before optimization (trial No. 9). -1 represents the

low concentration level and +1 represents the high concentration level for each component.

170

Botanical Studies, Vol. 47, 2006

was recorded for trial No. 8 while the lowest activity was

recorded for trials No. 2 and 3. On the other hand, maxi-

mum fungal dry weight was also obtained for trial No. 8,

and trial No. 1 recorded the lowest fungal biomass (Table

4).

The main effect (percentage) of each variable upon

inhibition zone diameter as well as the fungal dry weight

was also calculated (Table 5). The main effect figure with

a positive sign indicates that the high concentration of this

variable is nearly optimum and a negative sign indicates

that the low concentration of this variable is nearly opti-

mum. The data obtained showed a range of positive main

effect values indicating that the presence of high levels

of glucose, NaNO

3

, KH

2

PO

4

, MgSO

4

and ZnSO

4

in the

growth medium positively affects citrinin production by

M. ruber. On the other hand, the presence of MnSO

4

and

FeSO

4

at their lowest levels would result in higher citrinin

activity. The presence of high levels of glucose, MgSO

4

and ZnSO

4

in the medium with low levels of NaNO

3

,

K

2

HPO

4

, MnSO

4

and FeSO

4

resulted in an increase in the

fungal dry weight.

Statistical analysis (t-values) demonstrated that glu-

cose, NaNO

3

, K

2

HPO

4

, had significant positive influences

on the citrinin production with main effects of 3.25, 2.25

and 3.75% while MgSO

4

and ZnSO

4

had significant posi-

tive influences on fungal growth with main effects of 3.19

and 8.30%, respectively.

Based on these results obtained from the Plackett-

Burman experiment, a medium of the following formula

was predicted to be near optimum for producing an extra-

cellular citrinin in the culture filtrate of M. ruber (g/l): glu-

cose, 45; NaNO

3

, 3.75; KH

2

PO

4

, 0.2; MgSO

4

·7H

2

O, 0.1;

ZnSO

4

·7H

2

O, 0.01; MnSO

4

·H

2

O, 0.006 and FeSO

4

·7H

2

O,

0.003. Since, the trial No. 8 is a composition of all factors

with their maximum dosage in the experiment, care was

taken to insure that the predicted formula in trial No. 8

was the optimal one. The optimized composition of this

trial was used again as a base for the construction of a new

Plackett-Burman matrix, considering concentrations high-

er than that in trial No. 8. The results obtained from the

new Plackett-Burman experiment (data not shown) proved

that the formula predicted earlier (trial No. 8) in the first

Plackett-Burman experiment was still the optimal one.

In order to evaluate the accuracy of the Plackett-

T

able 3. Antibacterial activity of the mycotoxin citrinin detected in the M. ruber culture supernatant of media 1-3.

Microorganism

Inhibition zone diameter in mm

Medium 1

Medium 2

Medium 3

B. cereus PX MUCOB

13±1.1

17±1.4

15±1.3

B. subtilis ATCC 6633

13±1.1

16±1.3

14±0.0

S. lactis

10±0.2

12±1.3

11±0.5

P. fluorescens

10±0.1

15±1.5

13±0.3

MUCOB: Manchester University Collection of Bacteria, England.

Table 4. Influence of medium components on citrinin production by and growth of M. ruber according to the Plackett-Burman

design.

Trial No.

Response (± Standard deviation)

Activity

Dry weight (mg/ml)

Citrinin concentration (mg/ml)

1

13±1.3

10.62±0.419

185±5.0

2

10±0.0

1.82±0.088

143±7.0

3

10±0.0

2.44±0.101

143±8.0

4

14±0.0

13.74±0.478

200±9.0

5

12±1.3

3.52±0.049

171±4.0

6

15±1.7

10.77± 0.342

214±9.0

7

13±1.8

1.58± 0.061

185±3.0

8

25±2.5

17.46± 0.193

375±10

9

16±1.2

10.45± 0.021

220±10

B. subtilis ATCC 6633 was used as a test organism. The activity was measured as inhibition zone diameter in mm.

AL-SARRANI and EL-NAGGAR — Improvement of citrinin production by

M. ruber

171

Burman test, a verification experiment was undertaken, in

which the predicted optimum levels of independent vari-

ables were investigated and compared to the basal setting.

The average of inhibition zone diameter and dry weight

are shown in Table 6. Citrinin production in the optimized

medium expressed as an inhibition zone diameter reached

about 28 mm, approximately 1.75-fold that obtained from

the basal medium (16 mm). On the other hand, the dry

weight of the fungus in the new medium was 1.86 times

the figure recorded for the basal medium. HPLC quanti-

fication of citrinin by M. ruber in the optimized culture

supernatant confirmed that an improvement in citrinin

production occurred and amounted to 400 mg/ml, 1.8-fold

the basal setting before optimization (220 mg/ml).

It was interesting to note the growth pattern in the cul-

ture medium, where micropellets (2-4 mm in diameter)

were formed in trials 8 and 9 and large pellets (more than

10 mm) were observed for trials 2, 3, 5 and 7. On the other

hand, an irregular growth pattern was observed for trials 1,

4 and 6. The colour of culture filtrate was dark yellow for

trial No. 8, yellow for 9, and pale yellow for the rest of tri-

als.

Finally, the antibacterial substance produced in culture

supernatant of M. ruber was isolated, purified and charac-

terized by NMR technique. The

1

H NMR spectrum (Figure

2) of the isolated antibacterial substance matched the au-

thentic sample of citrinin (Sigma Chemical Co., Mo.) and

also matched the

1

H NMR trace published by Blanc et al.

(1995a).

DISCUSSION

Statistically based experimental designs have proven to

be valuable tools in optimizing culture conditions (Zhang

et al., 1996; Ooijkaas et al., 1999; El-Helow and El-

Ahawany, 1999; El-Helow et al., 2000; Elibol, 2004). One

Table 6. Experimental verification of the combined effect of optimized medium on M. ruber growth and citrinin production.

Medium No. 2

Inhibition zone diameter (mm) Citrinin concentration (mg/l)

Dry weight (mg/ml)

Basal setting

16±1.2

220±10

10.45±0.02

Optimized

28±0.9

400±9

19.45±0.12

Table 5. The main effect (%) and statistical analys is (t-values ) for equal unpaired samples for the determination of variable

significance in the Plackett Burman experiment.

Factor

Inhibition Zone diameter

Dry weight

Main effect

*

t-value Significance level Main effect

*

t-value Significance level

Glucose

3.25

2.82

0.01

3.84

0.0034

NaNO

3

2.25

1.385

0.1

-0.37

-0.8448

KH

2

PO

4

3.75

1.4084

0.1

-1.32

-0.4905

MgSO

4

·

7H

2

O

4.25

0.0018

3.19

1.8636

0.05

ZnSO

4

·

7H

2

O

4.75

0.0004

8.30

5.9377

0.005

MnSO

4

·

H

2

O

-7.75

-0.0312

-2.24

-0.2363

FeSO

4

·

7H

2

O

-0.25

-0.8709

-5.36

-0.0019

*

Main effect is expressed as the response percentage.

Figure 2.

1

H NMR (CD

3

OD) spectrum of the mycotoxin citrinin

produced by M. ruber.

172

Botanical Studies, Vol. 47, 2006

of the advantages of applying multi-factorial experiments

is that such an approach considers the interaction between

the non-linear nature of the responses in short experiments

(Gresham and Inamine, 1986). In the present study, the

Plackett-Burman design has been demonstrated to be an

efficient approach to optimizing the medium components

affecting the production of citrinin by M. ruber.

The optimization results indicated the importance of

phosphate for efficient citrinin production. This observa-

tion can be interpreted mainly as an effect of phosphate

functional groups on medium pH (Bonthrone et al., 2000).

Zinc has been shown to be essential for growth of several

fungi. One overall effect of zinc in the fungus metabolism

is a reduction in the economic coefficient (sugar con-

sumed/weight of fungus). It has been established that the

addition of zinc sulphate reduces the biomass (Lin and

Demain, 1991). Similarly, for M. purpureus, the addition

of zinc reduced the economic coefficient from 13.1 to 6.1,

indicating that zinc enables the fungus to utilize D-glucose

for growth (Johnson and McHan, 1975).

Utilization of the carbon source is an important factor

for citrinin production by M. ruber. 136 mg/ml was pro-

duced in agitated Erlenmeyer flasks containing synthetic

medium No. 1, which included ethanol (28 g/l as a carbon

source) and 226 mg/ml was produced in the same syn-

thetic medium which contained glucose (45 g/l) as a car-

bon source (Blanc et al., 1995b). In this work, significant

improvement in the production of citrinin by M. ruber was

achieved and the yield was 400 mg/l when glucose (45 g/l)

was used. Although M. ruber utilized glucose as a carbon

source, other Monascus strains grew well on galactose or

maltose (Yoshimura et al., 1975; Lin et al., 1992). This re-

sult is in agreement with the fact that utilization of carbon

source by this genus appears to be strain dependent (Fabre

et al., 1993).

Regarding nitrogen source utilization, ammonium

chloride has been reported to be a better inorganic nitro-

gen source than sodium nitrate for biomass and pigment

production by M. purpureus (Juzlova et al., 1996; Chen

and Johns, 1993), but monosodium glutamate and yeast

extracts were preferred by other M. ruber strains for pig-

ment and citrinin production (Blanc et al., 1995a, b). Al-

though Blanc et al. (1995a) obtained a high citrinin titre

(370 mg/l) in M. ruber culture that utilized yeast extract

as a nitrogen source (YES, medium 3), the present report

demonstrates that sodium nitrate (400 mg/ml) is a better

nitrogen source than monosodium glutamate (160 mg/l) or

yeast extract (180 mg/l) for citrinin production. This is in

line with the role of nitrate as a terminal electron acceptor

for this aerobic fungus. This again appears to reinforce the

strain-dependence concept for nitrogen source utilization.

Changes in the nature and concentration of carbon and

nitrogen sources and phosphate concentration have been

reported to affect polyketide biosynthesis through which

citrinin is synthesized (Doull and Vining, 1989). Medium

2 is thus sufficiently improved over the medium 1 and

3 used by the same species (Blanc et al., 1995a) though

the reduction in some ingredients and the weight of some

others may also provide another biotechnological advan-

tage to the industrial sector. Furthermore, a complete de-

fined medium for citrinin production may provide another

advantage for downstream processing.

It was shown that small micropellets formed in trials

No. 8 and 9 were related to the highest activity and dry

weight, while large pellets observed for trials 2, 3, 5, 7,

resulted in the lowest values for both activity and growth.

Since the citrinin yield proved to be directly related to the

oxygenation conditions (Hajjaj et al., 1999), the large pel-

lets may have reduced the mass and oxygen transfer due

to the internal diffusion limitation which in turn lowered

citrinin yield.

The results of the present work clearly demonstrate the

effectiveness of factorial design in optimizing medium

composition and improving the growth of M. ruber. Its

usefulness in designing an efficient fermentation process

for the production of citrinin by Monascus ruber to be

used for non-food applications has also been proven.

LITERATURE CITED

Al-Sarrani, A.Q. 1998. Amylases, proteases and lipases produc-

tion by Monascus ruber Tieghem. Int. J. Exp. Bot. 63: 1-9.

Blanc, P.J., J.P. Laussac, J. Le Bars, P. Le Bars, M.O. Loret,

A. Pareilleux, D. Prome, J.C. Prome, A.L. Santerre, and

G. Goma. 1995a. Characterization of monascidin A from

Monascus as citrinin. Int. J. Food Microbiol. 27: 201-213.

Blanc, P.J., M.O. Loret, and G. Goma. 1995b. Production of ci-

trinin by various species of Monascus. Biotechnol. Lett. 17:

291-294.

Bonthrone, K.M., J. Quarmby, C.J. Hewitt, V.J.M. Allam, M.

Paterson-Beedle, J.F. Kennedy, and L.E. Macaskie. 2000.

The effect of the growth medium on the composition and

metal binding behaviour of the extracellular polymeric

material of a metal-accumulating Citrobacter sp. Environ.

Technol. 21: 123-134.

Chen, M.H. and M.R. Johns. 1993. Effect of pH and nitrogen

source on pigment production by Monascus purpureus.

Appl. Microbiol. Biotechnol. 40: 132-138.

Davis, N.D., D.K. Dalby, U.L. Diener, and G.A. Sansing. 1975.

Medium-s cale production of citrinin by Penicillium ci-

trinum in a semi-synthetic medium. Appl. Microbiol. 29:

118-120.

Doull, J.L. and L.C. Vining. 1989. Culture conditions promoting

dispersed growth and biphasic production of actinorhodin

in shaken cultures of Streptomyces coelicolor A3 (2). FEMS

Microbiol Lett. 65: 265-268.

El-Helow, E.R. and A.M. El-Ahawany. 1999. Lichenase produc-

tion by catabolites repression-resistant Bacillus subtilis mu-

tants: optimization and formulation of an agro-industrial by-

product medium. Enzyme Microb. Technol. 24: 325-331.

El-Helow, E.R., S.A. Sabry, and R.M. Amer. 2000. Cadmium

bios orption by a c admium res is tant st ra in of Bacillus

AL-SARRANI and EL-NAGGAR — Improvement of citrinin production by

M. ruber

173

thuringiensis: regulation and optimization of cell surface

affinity for metal cations. BioMetals 13: 273-280.

Elibol, M. 2004. Optimization of medium composition for ac-

tinorhodin production by Streptomyces coelicolor A3(2)

with response surface methodology. Process Biochem. 39:

1057-1062.

El-Naggar, M.Y., M.A. Hassan, A.H. El-Dakkak, and S.A. El-

Aassar. 2000. Improvement of pigment production by algi-

nate-immobilized Monascus purpureus cultures. Adv. Food.

Sci. (CMTL) 22: 22-30.

El-Naggar, M.Y., M.A. Hassan, and W.Y. Said. 2001. Isolation

and characterization of an antimicrobial substance produced

by Streptomyces violatus. Proceedings of the first interna -

tional Conference of Egyptian British Biological Society:

Biodiversity and conservation of natural heritage. Vol. 3:

11-21.

Fabre, C.E., A.L. Santerre, M.O. Loret, R. Baberian, A. Pareil-

leux, G. Goma, and P.J. Blanc. 1993. Production and food

application of the red pigments of Monascus ruber. J. Food.

Sci. 58: 1099-1110.

Gresham, R. and E. Inamine. 1986. Nutritional improvement of

processes. In A. L. Demain and N. A. Solomon (eds.), Man-

ual of Industrial Microbiology and Biotechnology. ASM.

Washington, USA, pp. 41-48.

Hajjaj, H.A., A. Klaebe, M.O. Loret, T. Tazedakis, G. Goma,

and P.J. Blanc.1997. Production and identification of N-glu-

cosylrubropunctamine and N-glucosylmonascorubramine

from Monascus ruber and occurrence of electron donor-

acceptor complexes in these red pigments. Appl. Environ.

Microbiol. 63: 2671-2678.

Hajjaj, H.A., P.J. Blanc, E. Groussac, G. Goma, J.L. Uribelarrea,

and P. Loubiere.1999. Improvement of red pigments/citrinin

production ratio as a function of environmental conditions

by Monascus ruber. Biotechnol. Bioeng. 64: 497-505.

Hawks worth, D.L. and J .L. Pitt. 1983. A new taxonomy for

Monascus species based on the cultural and microscopical

characters. Aust. J. Bot. 31: 51-61.

Johnson, G.T. and F. McHan. 1975. Some effects of zinc on the

utilization of carbon sources by Monascus purpureus. My-

cologia 67: 806-815.

Juzlova, P., L. Martinkova, and V. Kren. 1996. Secondary me-

tabolites of the fungus Monascus: a review. J. Ind. Micro-

biol. 16: 163-167.

Kushiro, T., J. Hashida, H. kawamura, H. Mitsubayashi, and T.

Saito. 1996. Clinical effects of beni-koji in mild essential

hypertension. A multi-center double-blind comparison with

placebo. Jpn. J. Nephrol. 38: 625-633.

L in, T.F. a nd A.L . De ma in. 1 991. E ffec t o f nut rit ion o f

Monascus sp. on formation of red pigments. Appl. Micro-

biol. Biotechnol. 36: 70-75.

Lin, T.F., K.Yakushijin, G.H. Buchi, and A.L. Demain. 1992.

Formation of water-soluble Monasucs red pigments by bio-

logical and semi-synthetic processes. J. Ind. Microbiol. 9:

173-179.

Martinkova, L. and P. P atakova. 1999. Monascus. In R.K.

Robinson, C.A. Batt, and P.D. P atel (eds.), Encyclope-

dia of Food Microbiology. Academic Press. London, pp.

1481-1487.

Ooijkaas, L.P., E.C. Wilkinson, J. Tramper, and R.M. Buite-

laar. 1999. Medium optimization for spore production of

Conithyrium minitans using statistically-based experimental

designs. Biotechnol. Bioeng. 64: 92-100.

Panagou, E.Z., P.N. Skandamis, and G. Nychas. 2003. Model-

ing the combined effect of temperature, pH and a

w

on the

growth rate of Monascus ruber, a heat- resistant fungus

isolated from green table olives. J. Appl. Microbiol. 94:

146-156.

Plackett, R.L. and J.P. Burman 1946. The design of optimum

multifactorial experiments. Biometrika 33: 305-325.

Van Tieghem, P. 1884. Monascus, genre nouveau de l’ordre des

Ascomycetes. Bull. Soc. Bot. 31: 226-231.

Yasukawa, K., M. Takaha shi, S. Nat ori, K. Kawai , and M.

Yamazaki. 1994. Azaphilones inhibit tumor promotion by

12-O-tetradecanoylphorbol-13-acetate in two-stage carcino-

genesis in mice. Oncology 51: 108-112.

Yasukawa, K., M. Takahashi, S. Yamanouchi, and M. Takido.

1996. Inhibitory effect of oral administration of Monascus

pigment on tumor promotion in two-stage carcinogenesis in

mouse skin. Oncology 53: 247-249.

Yoshimura, M., S. Yamanada, K. Mitsugi, and Y. Hirose. 1975.

Production of Monascus pigments in a submerged culture.

Agr. Biol. Chem. 39: 1789-1795.

Zhang, J., C. Marcin, M.A. S hifflet, P. Salmon, T. Brix, R.

Greasham, B. Buckland, and M. Chartrain. 1996. Develop-

ment of a defined medium fermentation process for phy-

sotigmine production by Streptomyces griseofuscus. Apl.

Microbiol. Biotechnol. 44: 568-575.

174

Botanical Studies, Vol. 47, 2006

應用 Plackett-Burman 因子設計以改善 Monascus ruber 批次

培養時 citrinin 之產量

Abdulaziz Q. M. AL-SARRANI

1

and Moustafa Y. M. EL-NAGGAR

2

1

Biology Department, Microbiology Division, Faculty of Science, Taibah University,

P.O. Box 344, Al-Madinah Al-Munawarah, Kingdom of Saudi Arabia

2

Botany Department, Microbiology Division, Faculty of Science, Alexandria University,

Moharram Bay 21511, Alexandria, Egypt

　　先前由紅椒汁單離所得之黴菌定性為 Monascus ruber 者，被用來試驗其在批次培養時生產抗菌物

½之潛能。此黴以三種液態培養基供試。長在合成培養基 No. 2 者可得最高抗菌活性。該抗菌物½經抽

取，純化，及定性為黴毒素 citrinin。Plackett-Burman 實驗設計被用來最適化培養基 No. 2 之成份以求得

改善 citrinin 之產量以供非食品應用。結果得出下述配方可得濾液中接近最佳之細胞外生產 citrinin 之條

件（克/升）：葡萄糖，45；碳酸鈉，3.75；磷酸二氮鉀，0.2；硫酸鎂（7個水合物），0.1 及硫酸鋅（7

個水合物）0.01；硫酸錳，0.006；硫酸鐵，0.003。改善可達 1.75 倍產量。另一方面，黴之乾物量經改

善後可達控制組之 1.86 倍。以高性能液態層析法 (HPLC) 測上澄液之 citrinin 濃度，經改善者及控制組

分別為：400 及 220 毫克/升。

關鍵詞：批次培養；Citrinin；Monascus ruber；最適化；Plackett-Burman 實驗設計。