99

JOURNAL OF BIOSCIENCE AND BIOENGINEERING

Vol. 94, No. 2, 99105. 2002
Importance of Carbon Source Feeding and pH Control
Strategies for Maximum Kojic Acid Production
from Sago Starch by Aspergillus flavus
ROSFARIZAN MOHAMAD,
1,2,3
ARBAKARIYA ARIFF,
2,3
MOHD ALI HASSAN,
2
MOHAMED ISMAIL ABDUL KARIM,
2,3
HIROSHI SHIMIZU,
1
AND SUTEAKI SHIOYA
1
*
Department of Biotechnology, Graduate School of Engineering, Osaka University, 2-1 Yamadaoka, Suita-shi, Osaka 565-0871,
Japan,
1
Department of Biotechnology, Faculty of Food Science and Biotechnology, Universiti Putra Malaysia, 43400 UPM,
Serdang, Selangor, Malaysia,
2
and Fermentation Technology Unit, Enzyme and Microbial Technology Laboratory,
Institute of Bioscience, Universiti Putra Malaysia, 43400 UPM, Serdang, Selangor, Malaysia
3
Received 12 March 2002/Accepted 9 May 2002
To maximize kojic acid production by Aspergillus flavus Link 44-1 using gelatinized sago starch
as a carbon source, the performance of different fermentation modes (batch and fed-batch with
different feeding modes) was investigated in an 8-l stirred tank fermentor. The addition of a large
volume of concentrated sago starch (140 g/l) 2 d after the start with an initial starch concentration
of 60 g/l produced the maximum concentration of kojic acid (16.43 g/l), which was about 4 times
higher than that for batch fermentation of 100 g/l sago starch. Further improvement of kojic acid
production was obtained by adding small amounts of concentrated starch (140 g/l) intermittently
at 2-d intervals to the culture. Using this technique of fermentation, the dissolved oxygen tension
(DOT) can be controlled at high levels (4050% saturation) during the active growth phase, which
is required for the enhanced secretion of = == =-amylase used for saccharification of starch and also
for the production of mycelia with higher ability in synthesizing kojic acid. Maintaining the cul-
ture pH at 3 throughout the intermittent fed-batch fermentation reduced kojic acid production
(7.26 g/l) significantly. Very little kojic acid (0.93 g/l) was produced when the pH was kept at 4
during the growth phase and then lowered to 3 during the production phase. A high level of kojic
acid production (31.00 g/l) was achieved in fermentation where the pH was not controlled
throughout the cultivation or not controlled during the growth phase but kept at 3 during the pro-
duction phase. This result indicates that the pH control strategy is important; with the variation
in culture pH during the growth phase being most significant and critical to kojic acid production.
[Key words: kojic acid, fed-batch culture, sago starch, pH control, Aspergillus flavus]
Kojic acid (5-hydroxy-2-hydroxymethyl-@-pyrone) con-
tinues to attract attention because of its economic potential
in the fields of medicine, food science, cosmetics and agri-
culture (14). Kojic acid can be produced by aerobic fer-
mentation of Aspergillus spp. using various sources of car-
bon such as glucose, sucrose, acetate, ethanol, arabinose
and xylose (5, 6). However, glucose is the best carbon
source and acts as a precursor for kojic acid synthesis (7).
During the fermentation, kojic acid is formed directly from
glucose through a multi-step reaction without cleavage of
the carbon chain (8, 9).
Polysaccharides are a poor source of carbon for kojic acid
production. No Kojic acid was produced when starch was
used as a carbon source by A. oryzae (10) and very little
when maltose was used (11). Recently, high levels of pro-
duction of kojic acid using gelatinized sago starch as sub-
strate were achieved in shake flask cultures of A. flavus
which is capable of secreting amylolytic enzymes during its
growth (12). However, the production by this fungus in a
pilot scale fermentor (50 l) using gelatinized sago starch as a
substrate was greatly reduced due to non-optimal conditions
of aeration. Efficient oxygen transfer is difficult to achieve
in fermentations using high concentrations of starch due to a
high solution viscosity and imperfect mixing. A High aera-
tion rate (>7080% saturation) was required during the
growth phase to produce mycelium that contained enzymic
activities for the conversion of glucose to kojic acid during
the production phase by A. flavus (13).
Another important aspect of kojic acid fermentation is the
culture pH. Substantially high levels of kojic acid were pro-
duced by A. flavus in submerged fermentations at a pH of
between 6 and 7 (11) and also at pH3 (14). The highest
level of production by A. parasiticus was obtained at two
optimal pH values, 4.5 and 6.2 (15). Wilson (16) reported
that more kojic acid was obtained at a pH of between 2 and
3 and small changes to the culture pH could greatly reduce
the production. This means that the fermentation process is
* Corresponding author. e-mail: shioya@bio.eng.osaka-u.ac.jp
phone/fax: +81-(0)6-6879-7444
ROSFARIZAN ET AL. J. BIOSCI. BIOENG., 100
very sensitive to changes in the culture pH. For fermenta-
tion using starches as a carbon source, the pH is important
not only for the growth of the fungus and kojic acid synthe-
sis but also for the secretion of amylolytic enzymes and hy-
drolysis of starch to glucose. To improve kojic acid produc-
tion using starches, a two-phase pH control strategy is re-
quired. The culture pH for enhancement of the secretion of
amylolytic enzymes during the growth phase may be differ-
ent to the optimal pH for kojic acid synthesis during the pro-
duction phase. However, information on this is not yet re-
ported.
In this paper, a fermentation technique which can be used
to enhance kojic acid production by A. flavus Link 44-1 us-
ing gelatinized sago starch in an 8-l fermentor is proposed.
The effect of two different feeding modes and pH control
strategies during fed-batch fermentation on the dissolved
oxygen tension (DOT) level and activity of amylolytic en-
zymes which are involved in kojic acid production using
sago starch as a substrate is discussed.
MATERIALS AND METHODS
Microorganism and medium The fungus, A. flavus Link
44-1 was used for kojic acid production. After cultivation of this
strain, no aflatoxin was found but a safety check should be per-
formed before every application. Solid medium containing (g/l):
sucrose, 140; KH
2
PO
4
, 1; MgSO
4
7H
2
O, 0.25; NH
4
NO
3
, 2.5; tech-
nical agar, 15 was used for the preparation of slants for spore
production. The optimized medium for kojic acid production
described by Ariff et al. (13) was used in all experiments. The
medium consisted of (g/l): yeast extract, 5; KH
2
PO
4
, 1 and
MgSO
4
7H
2
O, 0.5 and different concentrations of gelatinized sago
starch were added according to the needs of each experiment. The
sago starch used in this study was supplied by Nee Seng Le Co.,
Sarawak (Malaysia). Gelatinized starch was prepared by heating
the starch slurry to just above 70C, which is the gelatinization
temperature of sago starch (17). Prior to gelatinization, the pH of
the starch slurry was adjusted to 3 using concentrated HCl.
Fermentor and fermentations All fermentations were per-
formed using an 8-l stirred tank fermentor (MBF-800 ME; Eyela,
Tokyo) equipped with temperature, pH and dissolved oxygen con-
trollers. Three six-bladed turbine impellers (diameter, d=40 mm)
were used for agitation and the impeller speed (N) was fixed at
600 rpm (impeller tip speed=FNd=1.26 m/s) for all fermentations.
A polarographic dissolved oxygen probe (Ingold, Switzerland) was
used to measure DOT levels and a steam-sterilizable glass pH elec-
trode (Ingold) was used to monitor the culture pH.
For batch fermentation, 1 ml of a spore suspension (approxi-
mately 310
8
spores/ml) was introduced into 3 l of medium con-
taining 100 g/l of gelatinized starch. Two fed-batch fermentations
with different feeding modes were carried out. Both fermentations
were initiated with 2 l of initial batch culture using a medium con-
taining 60 g/l of gelatinized sago starch. In the first type of fed-
batch fermentation, 1 l of 140 g/l gelatinized sago starch was added
only once at 2 d of fermentation. In the second type, 200 ml of
140 g/l gelatinized sago starch was added intermittently at 2-d
intervals for 10 d (i.e., 200 ml of gelatinized starch was added 5
times during the fermentation).
The aeration control strategy for optimum kojic acid production
as suggested by Ariff et al. (13) was employed in all fermentations.
With this strategy, DOT was controlled at 80% and 30% of satura-
tion during growth and production phases, respectively, using glu-
cose as a carbon source. The DOT level in the culture broth was
controlled via a sequential cascade control of the rate of airflow.
The maximum and minimum set points of permitted airflow were
5 l/min and 1 l/min, respectively. Initially, all three methods of fer-
mentation (batch and two fed-batch cultures) were carried out
without pH control (strategy A). Three control strategies were ap-
plied to intermittent fed-batch fermentation; (i) pH was controlled
at 3 throughout the fermentation (strategy B), (ii) pH was con-
trolled at 4 during the growth phase and then switched to 3 during
the production phase (strategy C), and (iii) pH was not controlled
during the growth phase (initial pH3) but was kept at 3 during the
production phase (strategy D). The pH control system consisted of
a standard PI controller and dead band control with two pumps for
acid and alkali. The culture pH was controlled by adding either
1.0 N NaOH or 1.0 N HCl. The pH control system could ade-
quately control the culture pH to within 0.1. The temperature
within the fermentor was maintained at 30C.
Analytical methods During the fermentation, 5 ml samples
were collected at time intervals for analysis. The samples were
centrifuged at 10, 000 rpm for 10 min. The supernatants were used
for chemical determinations while the pellets were washed with
acetate buffer (0.1 M, pH5.0) 3 times. In order to remove the
starchy materials attached to the fungal mycelia, the pellets were
resuspended in a 1% v/v =-amylase (Termamyl 120L) solution ob-
tained from NOVO, Malaysia. The mixture was kept in a water
bath at 100C for 15 min to hydrolyse the insoluble starch to solu-
ble sugars. The suspended mycelia free from starchy materials
were then filtered using a preweighed microfiber filter paper and
dried in an oven at 95C for measurements of dry weight. The con-
centration of kojic acid in the culture was determined using a colo-
rimetric method as proposed by Bentley (18). The glucose concen-
tration was determined with a glucose analyzer (Model 2700; YSI
Bioanalytical Products, Spring, USA). =-Amylase activity was
assayed as described by Ariff and Webb (19). The effect on the ac-
tivity of =-amylase was determined at a pH ranging from 2.5 to
8.0. A 0.05 M citrate phosphate buffer was used to prepare solu-
tions 2.5 to 5 while for a higher pH (6.0 to 8.0), 0.05 M phosphate
buffer was used.
RESULTS AND DISCUSSION
Batch fermentation A typical time course of batch
kojic acid fermentation by A. flavus using gelatinized sago
starch as a carbon source is shown in Fig. 1. The growth of
the fungus was rapid during the initial stages of the fermen-
tation and reached a stationary phase after about 5 d. During
active growth, starch was hydrolyzed to glucose by amylo-
lytic enzymes secreted by the fungus and the glucose con-
centration in the culture increased. Substantial levels of
activity by both amylolytic enzymes (=-amylase and gluco-
amylase) were detected with a maximum reached after
about 5 d of fermentation. As the fermentation progressed,
a decrease in the activity of amylolytic enzymes was ob-
served possibly due to a reduction in culture pH that deacti-
vated the enzymes.
The accumulation of glucose in the culture reached a
maximum of 16.0 g/l after about 4 d of fermentation. Kojic
acid started to be produced when growth reached a station-
ary phase (5 d) in which there was no significant increase
in mycelial dry weight. In the fermentation using the same
fungal strain and pure glucose as a carbon source, produc-
tion started after 2 d (13). This means that the onset of pro-
duction was about 3 d earlier than when starch was used as a
carbon source. The delay in the onset of kojic acid produc-
KOJIC ACID PRODUCTION FROM SAGO STARCH VOL. 94, 2002 101
tion from starch may be due to the time required to hydro-
lyze starch to glucose prior to use in the synthesis.
Rapid kojic acid production occurred during the non-
growing phase where rapid consumption of glucose was
also observed, indicating that glucose was used for kojic
acid synthesis. During the fermentation, the glucose that
accumulated in the culture was converted to kojic acid
through the action of cell-bound enzymes (9). The cell-
bound enzyme system consisted of glucose-6-phosphate
dehydrogenase, hexokinase and gluconate dehydrogenase,
which are involved in the direct synthesis of kojic acid from
glucose (11). It is well known that glucose acts as a pre-
cursor in kojic acid synthesis (20). Kojic acid production
ceased when all the glucose in the culture was depleted.
Kojic acid reached a maximum concentration (4.51 g/l)
after 10 d of fermentation. The yield of kojic acid based on
the starch consumed and overall productivity was 0.045
g/g and 0.45 g/l/d, respectively (Table 1). The reduced kojic
acid production in batch fermentation using gelatinized sago
starch as a carbon source may be due to non-optimal aera-
tion conditions. Figure 1 also shows that the DOT level de-
creased to 0% saturation after about 2-d fermentation and
remained there. This indicates that the control system failed
to maintain the DOT at the required value (i.e., 80% satura-
tion during active growth phase). The high medium viscos-
ity and imperfect mixing during the initial stages of the fer-
mentation led to a low volumetric oxygen transfer rate and
hence, reduced the DOT level.
Fed-batch fermentation Figure 2 shows a typical
time course of fed-batch fermentation of kojic acid, in which
a large volume of highly concentrated starch was added
once after 2 d. The growth profile and maximum cell con-
centration were similar to those for batch fermentation. A
slightly higher level of =-amylase activity than that ob-
tained in batch fermentation was observed. However, glu-
cose production was increased more than two-fold, suggest-
ing that the hydrolysis of starch by the secreted amylolytic
enzymes was more efficient than for batch fermentation. It
is interesting that the DOT level dropped to 0% saturation
after about 4 d of fermentation. This means that the required
level during the growth phase was slightly improved com-
pared to batch fermentation but still failed to maintain high
levels (7080% saturation) during the growth phase which
is required for optimal kojic acid fermentation (13). Kojic
acid was rapidly produced during the non-growing phase
and reached a maximum concentration of 16.43 g/l when
the glucose became depleted. Compared to batch fermen-
tation, the amount of kojic acid produced was increased
about 3.6-fold. However, the yield (0.164 g kojic acid/g
starch) and overall productivity (0.97 g/l/d) was only about
2 and 3 times higher than for batch fermentation, respec-
tively (Table 1).
Further improvement in the production of kojic acid was
achieved by feeding a small volume of concentrated starch
intermittently to the active fungal culture at 2 d intervals
(Fig. 3). In this fermentation run, the DOT level was main-
tained at about 4050% saturation during the growth phase
and dropped to 0% saturation after 6 d. Very high levels of
glucose accumulated after 4 d and then there was a marked
reduction due to consumption for kojic acid synthesis.
FIG. 1. A time course of batch kojic acid fermentation using sago
starch as a carbon source. Symbols: solid circle, cell concentration;
solid lozenge, glucose; solid square, kojic acid; open circle, =-amy-
lase; open square, glucoamylase; broken line, DOT; dotted line, pH.
TABLE 1. The performance of batch and fed-batch fermentation for kojic acid production by A. flavus
with different pH control strategies using gelatinized sago starch as a carbon source
Parameters Batch
Fed-batch
Addition of
starch at 48 h
Addition of starch at 48 h intervals
pH control strategy A
a
A
a
A
a
B
a
C
a
D
a
X
m
(g/l) 16.30 16.00 16.68 16.23 16.52 16.74
P
m
(g/l) 4.51 16.43 31.07 7.26 0.98 31.23
t (d) 10 17 15 15 15 15
P (g/l/d) 0.45 0.97 2.07 0.61 0.065 2.08
Y
K/S
(g/g) 0.045 0.164 0.357 0.0834 0.0113 0.359
X
m
, Maximum cell concentration obtained during fermentation; P
m
, maximum kojic acid concentration; t, time taken to reach maximum concen-
tration of kojic acid; P, productivity; Y
K/S
, yield of kojic acid based on starch consumed.
a
Different pH control strategies: (A) without pH control (initial pH3), (B) pH was controlled at 3 throughout the fermentation, (C) pH was con-
trolled at 4 during the growth phase and then 3 during the production phase, (D) pH was not controlled during the growth phase but was kept at 3
during the production phase.
ROSFARIZAN ET AL. J. BIOSCI. BIOENG., 102
Rapid production started after 4 d and reached a maximum
concentration of 31.07 g/l after 15 d of fermentation. Obvi-
ously, the high levels of DOT during the growth phase en-
hanced the production of mycelia with a great ability to syn-
thesize kojic acid in which the mycelia effectively trans-
formed glucose into kojic acid during the production phase.
The results of this study also showed that different levels
of DOT greatly influenced the secretion of amylolytic en-
zymes, especially =-amylase (Figs. 13). High DOT levels
during the growth phase enhanced =-amylase production
and resulted in high levels of glucose in the culture. The low
glucose accumulation during the growth phase in batch fer-
mentation caused a very low level of kojic acid production
compared to fed-batch fermentation. It has been reported
that =-amylase and glucoamylase production by Aspergillus
spp. was greatly influenced by the aeration conditions (DOT
and shear rate) (21).
The yield and overall productivity of kojic acid produc-
tion by A. flavus using gelatinized sago starch were com-
parable to those obtained using pure glucose as a carbon
source (13, 14). The level of production obtained in this
study was also comparable to that reported in the literature
(10, 22). The yield and overall productivity of batch kojic
acid fermentation by A. oryzae was 0.26 g kojic acid/g
glucose and 3.43 g/l/d, respectively (10). A slightly higher
yield (0.43 g kojic acid/g glucose) was obtained in repeated
batch fermentation by using pure glucose as a substrate in
immobilized cells of A. oryzae (22). One of the aim of our
report is to utilize an inexpensive carbon source such as
gelatinized sago starch and reduce production costs. In ad-
dition, the kojic acid produced by fermentation using starch
has a higher solubility and cytotoxicity than that produced
using glucose (Hwei, A. N. M., B. Sc. thesis, Universiti
Putra Malaysia, Selangor, 1996).
Effect of culture pH The effect of controlling the
culture pH during intermittent fed-batch fermentation on the
growth of A. flavus, secretion of amylolytic enzymes and
production of kojic acid is shown in Fig. 4 and the perfor-
mances of each strategy are given in Table 1. The profile of
DOT was similar for all fermentations, indicating that the
difference in growth, amylolytic enzymes and kojic acid
was influenced not by the aeration conditions but by the
culture pH. In fermentation without pH control (pH control
strategy A), the process was started with an initial pH of 3
and the culture pH during the growth phase was increased
to around 3.5 after 2 d and then decreased to 3 during the
production phase as can be seen in Fig. 4c. Based on the
pH and amylolytic activity profile in fermentation without
pH control, fermentations with a two-phase control strategy
in which the culture pH was controlled differently during
growth and production, were proposed. Under control strat-
egy C, the culture pH was kept at 4 during the growth phase
and switched to 3 during the production phase. For com-
parison, fermentation where the culture pH was controlled
at 3 throughout the fermentation (control strategy B) was
also carried out. A slightly lower rate of growth and maxi-
mum cell concentration was attained with strategy B. Sig-
nificantly less =-amylase activity and glucose was obtained
with strategy B than strategy A, suggesting that the growth
of the fungus and secretion of =-amylase were greatly in-
hibited at low pH. The low level of kojic acid production
obtained with strategy B was due to the small amount of
glucose accumulated in the growth phase that could be used
for the conversion to kojic acid. Although a rather high level
FIG. 2. A time course of fed-batch kojic acid fermentation using
sago starch as a carbon source added after 2 d of fermentation to an 8-l
fermentor. Symbols: solid circles, cell concentration; solid lozenges,
glucose; solid squares, kojic acid; open circles, =-amylase; open
squares, glucoamylase; broken line, DOT; dotted line, pH. An arrow
indicates starch addition.
FIG. 3. A time course of fed-batch kojic acid fermentation using
sago starch as a carbon source added at 2-d intervals. Symbols: solid
circles, cell concentration; solid lozenges, glucose; solid squares, kojic
acid; open circles, =-amylase; open squares, glucoamylase; broken
line, DOT; dotted line, pH. Arrows indicate starch addition.
KOJIC ACID PRODUCTION FROM SAGO STARCH VOL. 94, 2002 103
of activity of =-amylase and glucose accumulation was ob-
tained with strategy C, the glucose that accumulated was not
consumed during the production phase and very little kojic
acid was produced. The low kojic acid production for fer-
mentation strategy C could be due to the culture pH during
the growth phase not being suitable for the production of
mycelia with the ability to synthesize kojic acid. This is be-
cause a high level of production (31.23 g/l) was achieved
when the culture pH during the growth phase was not con-
trolled but the pH was kept at 3 during the production phase
(pH control strategy D, Fig. 4c) the same as for strategy C.
Kojic acid production using strategy D was almost the
same as for strategy A. In addition, amylolytic enzymic ac-
tivities, growth and glucose profiles were almost the same
as those in strategy A. This result indicates that the culture
pH during the growth phase was critical to kojic acid pro-
duction. It can be suggested that during growth the culture
pH should be controlled precisely at around 3.3 to 3.7 as
shown in Fig. 4c, which might be critical for mycelia pro-
duction by enzymes necessary for direct synthesis of kojic
acid.
Figure 5 shows the effect of pH on the activity of the =-
amylase of A. flavus. The highest level of activity was ob-
served at pH4. A slight reduction in activity was observed
FIG. 4. Effect of pH control strategy on the performance of intermittent fed-batch fermentation of kojic acid. (a) Cell concentration (large
symbol) and =-amylase (small symbol), (b) DOT (line) and glucoamylase (symbol), (c) kojic acid (symbol) and pH (line), (d) glucose (symbol).
Symbols: solid lozenge and solid line; without pH control (strategy A), open circle and broken line; pH was kept at 3 throughout the fermentation
(strategy B), solid circle and dotted line; pH was kept at 4 during the growth phase and then switched to 3 during the production phase (strategy C),
open lozenge and one dotted chain line; pH was not controlled during the growth phase but was kept at 3 during the production phase (strategy D).
FIG. 5. Effect of pH on the activity of A. flavus =-amylase.
ROSFARIZAN ET AL. J. BIOSCI. BIOENG., 104
at a pH higher than 6 and lower than 3.5, indicating that the
=-amylase of A. flavus was stable under acidic conditions.
This means that the culture pH for all the fermentations
investigated did not significantly influence the activity of =-
amylase present in the culture medium. The pH-dependency
of glucoamylase activity for this strain is also not significant
between pH2 to 4 (data not shown). In other words, the
effect of culture pH on kojic acid production did not depend
on the pH-dependency of =-amylase or glucoamylase activ-
ities. The results of this study indicated that the influence of
culture pH during the growth phase was much more sensi-
tive than that during the production phase. However, the in-
fluence of culture pH may also depend on the type of carbon
or nitrogen source used. In fermentation using glucose as a
carbon source and peptone as a nitrogen source, the optimal
pH for high kojic acid production was between 2 and 3 (10).
To further investigate the significance of culture pH dur-
ing the growth or production phase using different carbon
sources, the performances of fermentations using glucose
and starch under different pH control strategies were pre-
sented (Table 2). In fermentation using glucose as a carbon
source, high levels of kojic acid production (3548 g/l) were
achieved in fermentations in which the culture pH was not
controlled during the growth phase (initial pH3); and either
controlled or not during the production phase. Higher kojic
acid production rates during production phase were also
achieved in fermentations, in which the culture pH was not
controlled during the growth phase. Similar results were ob-
tained in fermentations using sago starch as a carbon source.
During fermentation, glucose is directly converted to kojic
acid by a multistep enzyme reaction without any cleavage
into small fragments (7, 9), however it has also been re-
ported (23) that other organic acids such as oxalic, succinic,
lactic, acetic and lactic acid were produced during fermenta-
tion using glucose as a carbon source by the same strain of
A. flavus.
Since kojic acid fermentation by A. flavus using sago
starch as a carbon source was very sensitive to culture pH,
further work on the effect of pH on the secretion of enzymes
related to the direct production of kojic acid such as glucose
dehydrogenase and gluconate dehydrogenase should be car-
ried out. The effect of very small changes in pH between 3
and 4 (e.g., 3.0, 3.2, 3.4, 3.6, 3.8 and 4.0) during the growth
phase on kojic acid production as well as related enzymes
excreted and/or accumulated inside the cell should be in-
vestigated. Using the optimal pH during the growth phase,
the effect of different pH values (e.g., pH2 to 3) during the
production phase on kojic acid production should also be
investigated.
ACKNOWLEDGMENTS
One of authors, Rosfarizan Mohamad, is indebted to JSPS Core
University Program for funding the research under Long Term In-
vitation Program in International Center for Biotechnology, Osaka
University, Japan.
REFERENCES
1. Kayahara, H., Shibata, N., Tadasa, K., Maeda, H., Kotani,
T., and Ichimoto, I.: Amino acids and peptide derivatives of
kojic acid and their antifungal properties. Agric. Biol. Chem.,
54, 24412442 (1990).
2. Le Blanch, D. T. and Akers, H. A.: Maltol and ethyl maltol
from larch tree to successful food additives. Food Technol.,
26, 7887 (1989).
3. Gomes, A. J., Lunardi C. N., Gonzales, S., and Tedesco,
A. C.: The antioxidant action of Polypodium leucotomos ex-
tract and kojic acid: reactions with reactive oxygen species.
Braz. J. Med. Biol. Res., 34, 14871494 (2001).
4. Smith, C. N. and Lindsay, C. D.: Kojic acid reduces the
cytotoxic effects of sulfur mustard on cultures containing
human melanoma cells in vitro. J. Appl. Toxicol., 21, 435
440 (2001).
5. Beelik, A.: Kojic acid. Adv. Carbohydr. Chem., 11, 145183
(1956).
6. Presscot, S. C. and Dunn, C. G.: Industrial microbiology,
3rd ed., p. 609618. Mc Graw-Hill Book, New York (1959).
7. Arnstein, H. R. V. and Bentley, R.: The biosynthesis of
kojic acid 1. Production from [1-14] and [3: 4-14C] glucose
and [2-14C]-1: 3-dihydroxyacetone. Biochem. J., 54, 493
TABLE 2. Comparison of the performance in kojic acid fermentation by A. flavus using different pH control strategies
during growth and production phases with glucose and starch as carbon sources
Carbon source
pH control strategy
P
m
(g/l)
Kojic acid
production rate
a
(g/l/d)
Y
K/G
or
Y
K/S
Y
X/G
or
Y
X/S
Growth phase Production phase
Glucose (10%) 2.20 2.20 7.58 0.529 0.126 0.00228
2.50 2.50 33.24 2.348 0.645 0.360
3.00 3.00 30.12 1.994 0.518 0.540
3.50 3.50 23.23 1.355 0.218 0.419
4.00 4.00 17.72 1.267 0.361 0.362
3.00
in
# 35.40 6.600 0.446 0.409
3.00
in
3.00 48.79 5.340 0.685 0.287
3.00
in
2.75 40.51 3.430 0.690 0.345
Starch (8.7%) 3.00 3.00 7.26 0.745 0.0834 0.187
3.00
in
# 31.07 2.070 0.357 0.192
3.00
in
3.00 31.23 2.295 0.359 0.192
4.00 3.00 0.98 0.080 0.0113 0.190
P
m
, Maximum kojic acid concentration; Y
K/G
, yield of kojic acid based on glucose consumed during the production phase; Y
X/ G
, yield of cell
based on glucose consumed during the growth phase; Y
K/S
, yield of kojic acid based on starch consumed; Y
X/S
, yield of cell based on starch con-
sumed; 3.00
in
, initial pH3 (without control during the growth phase); #, pH was not controlled during the production phase.
a
Kojic acid production rate during the production phase.
KOJIC ACID PRODUCTION FROM SAGO STARCH VOL. 94, 2002 105
508 (1953).
8. Kitada, M. and Fukimbara, T.: Studies on kojic acid fer-
mentation (VII). The mechanism of the conversion of glucose
to kojic acid. J. Ferment. Technol., 49, 847851 (1971).
9. Bajpai, P., Agrawala, P. K., and Vishwanathan, L.: En-
zymes relevant to kojic acid biosynthesis in Aspergillus fla-
vus. J. Gen. Microbiol., 127, 131136 (1981).
10. Kitada, M., Ueyama, H., and Fukimbara, T.: Studies on
kojic acid fermentation (I). Cultural condition in submerged
culture. J. Ferment. Technol., 45, 11011107 (1967).
11. Basappa, S. C., Sreenivasamurthy, V., and Parpia, H. A. B.:
Aflatoxin and kojic acid production by resting cells of As-
pergillus flavus Link. J. Gen. Microbiol., 61, 8186 (1970).
12. Rosfarizan, M., Ariff, A. B., Hassan, M. A., and Karim,
M. I. A.: Kojic acid production by Aspergillus flavus using
gelatinised and hydrolysed sago starch as carbon sources.
Folia Microbiol., 43, 459464 (1998).
13. Ariff, A. B., Salleh, M. S., Ghani, B., Hassan, M. A., Rusul,
G., and Karim, M. I. A.: Aeration and yeast extract require-
ments for kojic acid production by Aspergillus flavus Link.
Enz. Microb. Technol., 19, 545550 (1996).
14. Ariff, A. B., Rosfarizan, M., Herng, L. S., Madihah, M. S.,
and Karim, M. I.: Kinetics and modelling of kojic acid pro-
duction by Aspergillus flavus Link in batch fermentation and
resuspended mycelial system. World J. Microbiol. Biotech-
nol., 13, 195201 (1997).
15. Lin, M. T., Mahajan, J. R., Dianese, J. C., and Takatsu, A.:
High production of kojic acid crytals by Aspergillus para-
siticus UNBF A12 in liquid medium. Appl. Environ. Micro-
biol., 32, 298299 (1976).
16. Wilson, B. J.: Miscellaneous Aspergillus toxins, p. 235250.
In Ciegler, A. (ed.), Microbes toxins, fungal toxins, VI. Aca-
demic Press, New York (1971).
17. Haska, N. and Ohta, Y.: Mechanism of hydrolysis of the
treated sago starch granules by raw starch digesting amylase
from Penicillium brunncum. Starch/Starke, 44, 2228 (1992).
18. Bentley, R.: Preparation and analysis of kojic acid. Methods
Enzymol., 3, 238241 (1957).
19. Ariff, A. B. and Webb, C.: The influence of different fer-
mentor configurations and modes of fermentor operation on
glucoamylase production by Aspergillus awamori. Asia Pac.
J. Mol. Biol. Biotechnol., 4, 183195 (1996).
20. Megalla, S. E., Nassar, A. Y., and Gohar, M. A. S.: The
role of cooper (I) nicotinic acid complex on kojic acid biosyn-
thesis by Aspergillus flavus. J. Basic Microbiol., 27, 2933
(1987).
21. Rousset, S. and Schlich, P.: Amylase production in sub-
merged culture using principal component analysis. J. Fer-
ment. Bioeng., 68, 339343 (1989).
22. Kwak, M. Y. and Rhee, J. S.: Control mycelial growth for
kojic acid production using ca-alginate immobilized fungal
cells. Appl. Microbiol. Biotechnol., 36, 578583 (1992).
23. Rosfarizan, M., Ariff, A. B., Hassan, M. A., and Karim,
M. I. A.: Influence of pH on kojic acid fermentation by As-
pergillus flavus. Pakistan J. Biol. Sci., 3, 977982 (2000).