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
pg_0002
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.
pg_0003
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.
pg_0004
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.
pg_0005
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.
pg_0006
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
pg_0007
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.
pg_0008
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。