Bioscience, Biotechnology, and Biochemistry

ISSN: 0916-8451 (Print) 1347-6947 (Online) Journal homepage: https://www.tandfonline.com/loi/tbbb20

Efficient Production of γ-Polyglutamic Acid by

Bacillus subtilis (natto) in Jar Fermenters

Yoshihiro Ogawa, Fumio Yamaguchi, Katsumi Yuasa & Yasutaka Tahara

To cite this article: Yoshihiro Ogawa, Fumio Yamaguchi, Katsumi Yuasa & Yasutaka Tahara
(1997) Efficient Production of γ-Polyglutamic Acid by Bacillus�subtilis (natto) in Jar Fermenters,
Bioscience, Biotechnology, and Biochemistry, 61:10, 1684-1687, DOI: 10.1271/bbb.61.1684

To link to this article: https://doi.org/10.1271/bbb.61.1684

Published online: 12 Jun 2014.

Submit your article to this journal

Article views: 1327

View related articles

Citing articles: 55 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=tbbb20
Biosci. Biotech. Biochem., 61 (10),1684-1687,1997

Efficient Production of y-Polyglutamic Acid by Bucillus subtilis (nutto) in Jar Fermenters

Yoshihiro OGAWA, t Fumio YAMAGUCHI, Katsumi YUASA, and Yasutaka TAHARA *
Research and Development Divisiol1, Kikkoman Corporation, 399 Noda, Noda-shi, Chiba 278, Japan
* Department of Applied Biological Chemistry, Faculty ofAgriculture, Shizuoka University, 836 Ohya, Shizuoka-shi, Shizuoka
422, Japan
Received March 26, 1997

The large scale fermentation of y-polyglutamic acid (y-PGA) by Bacillus subtilis (natto) was done
using a 30-liter jar fermenter. A stable cultivation without foaming could be done with addition of 3 0ft) NaCI
to the medium. The y-PGA productivity became higher with increasing speed of agitation and amounts of
glutamic acid added to the broth. Finally, we were able to obtain about 35 mg/ml of y-PGA under the
optimum conditions. The glutamic acid added to the medium was efficiently converted into y-PGA in the
stationary phase. To discover the role of L-glutamic acid added to the medium for y-PGA biosynthesis by
Bacillus suhtilis (natto), the radioactivity incorporated into y-PGA from 14C-L-glutamic acid was measured.
As a result, radioactive y-PGA was detected in the medium. Then, the glutamic acid in the medium was
transported into the cells and actually polymerized as the glutamic acid unit of y-PGA.

Key words: Bacillus subtilis (natto); y-polyglutamic acid; salvage pathway

y-Polyglutamic acid (y-PGA) is the main constituent of
the sticky material of natto, a traditional food in Japan,
j,e., cooked soybeans fermented with Bacillus subtilis
(natto).l 6) y-PGA is a water-soluble and biodegradable
substance which comprises D- and L-glutamic acid polym-
erized through y-glutamyl bonds. Its use in the fields of
food, cosmetics, or medicine has been proposed, and
biodegradable fibers 7) and hydrogels 8 ,9) have already been
made from 'Y-PGA.
y-PGA was discovered as a component of the capsules
from strain MR-l, and produced the most sticky material on natto fer-
mentation compared with other strains in our stock culture collection. 2 1)
The parent strain, MR-l, was isolated from the Miyagino strain (a stock
culture of Miyagino Natto Seisakujo) as a commercially useful strain as
a natto starter.
Culture conditions. B. subtilis (natto) strain MR-141 was cultured on
Nutrient Broth (Difco Laboratories) plates containing 1.5% agar for 7
days at 40°C to induce spore formation. After cultivation, 10 ml of sterilized
standard saline (0,9% NaCl) was added to each plate, and the spores were
suspended. One-ml portions (5 x 10 8 cell/ml) of the spore suspension were
divided into several micro tubes and stored in - 80 C until use for
G

of Bacillus amhracis 10 ) and Bacillus mesentericus,ll) and cultivation,

was isolated from a culture filtrate of Bacillus subtilis. 12.13)
y-PGA producing bacteria are divided into two types as to
the requirement of glutamic acid for y-PGA biosynthesis.
Glutamic acid-independent production of y-PGA using
B. subtilis SE,141 B. licheniformis A3S 15 ) and B. subtilis
TAM-416) have been reported as de novo y-PGA producing
bacteria. On the other hand, other bacteria such as B. subtilis
ATCC994S,13) B. anthracis,17) B. sllbtilis F-2-0118) and B.
subtilis IF0333S 19 ,20) were stimulated to produce 'Y-PGA
by the addition of L-glutamic acid to the medium. In these
bacteria, salvage I,-PGA production was suggested, but
there was no evidence that the glutamic acid units in y-PGA
were formed from the glutamic acid in the medium. Kunioka
and Goto reported that glutamic acid added to the medium
appeared to regulate the de novo y-PGA production from
citrate, and not to be incorporated into )!_PGA.20)
To produce 'Y-PGA on an industrial scale, we searched
for suitable conditions for y-PGA fermentation in a jar
fermenter. Moreover, to clarify the role of glutamic acid in
the medium, we examined the incorporation of 14C-glutamic
acid into y-PGA.
For cultivation in a 30-liter jar fermenter, 20 liters of MSG medium
comprising 6% maltose, 7% soy sauce, 3% sodium L-glutamate, 0,25%
K 2 HP0 4 , 0.05% MgS0 4 · 7H 2 0, and 3% NaCI or GSG medium con-
taining 6% glucose instead of maltose was used, As an antifoaming agent,
0.1 % silicone oil (SH200-100cs; Toray Dow Corning Silicone Co.) was
added. After sterilization, an appropriate volume of 20% NaOH sterilized
by autoclaving was added to the medium to adjust the pH to 8. The spore
solution (1 m!) was diluted with 10 ml of standard saline, and then added
to the medium to start the cultivation. The temperature and agitation were
maintained at 40°C and 400 rpm, respectively, for every cultivation time.
The aeration was maintained at 0.1 vvm until 18 h and then at I vvm after
18 h to prevent loss of the medium as foam. Fifty ml of the culture broth
was removed from the jar fermenter at appropriate times and stored at
- 20 u C until analysis.
y-PGA analysis. The amount of glutamic acid in the broth was measured
with an amino acid (Model L-8500; Hitachi Co.). Cells were
removed from the culture broth by centrifugation, and each supernatant
was filtered through a membrane filter (0,45 ,urn) to obtain a culture filtrate,
y-PGA in the culture filtrate was hydrolyzed with 6 N HCI at lOODC for 4
hours, and the amount of glutamic acid after hydrolysis was regarded as
the total glutamic acid, The amount of free glutamic acid derived from
sodium L-glutamate in the culture filtrate was measured without hydrolysis.
The amount of y-PGA was calculated as glutamic acid equivalent to the
difference between the total and free glutamic acid.

The measurement of glucose and maltose in the medium, The glucose

Materials and Methods remaining in the culture filtrate was measured using a Glucose Test
Bacterial strains. Among many y-PGA producing strains in our stock Wako (Wako Pure Chemical Industries, Ltd.). To calculate the maltose
culture collection, Bacillus subtilis (natto) strain MR-141 was used. This concentration, a 10,u1 sample was incubated with 10,ul (1,4 U) of
strain was selected as a mutant defective in ammonia production derived IX-glucosidase (AGH-211; Toyobo Co., Ltd.) at 37°C for 30min. After

t Corresponding author.
Abbreviation: y-PGA. y-polyglutamic acid.

NII-Electronic Library Service

 Production of y-Polyglutamic Acid
incubation, glucose produced through hydrolysis from maltose was
measured using a Glucose Test Wako.
l'+C-Glucose and 14C-L-glutamic acid incol1)()}'(ltioll into y-PGA. For a
small-scale culture in a 50-ml flask. 10 ml of GYG medium comprising
3°/c) glucose, 2.5% yeast extract, 3% sodium L-glutamate, 0.25% K 2 HP0 4 ,
0.05% MgS0 4 ' 7H 2 0. and 3%) NaCl was used. The spore solution (10 .ul)
was inoculated into the GYG medium containing 185 kBq of [U_ 14C]
D-glucose (New England Nuclear, U.S.A .. 37-185 MBq/.umoJ) or [U_ 14C]
L-glutamic acid (New England Nuclear, U.S.A., 10.95 MBq/.umol). The
cells were grown at 37 C at 120 rpm on a rotary shaker for 72 h. Fifty .ul
of the borth was removed from the flask at appropriate times and mixed
with 50.u1 of the SDS-sample buffer (2% SDS, 30o/() glycerol, and 0.25 M
Tris hydroxyaminomethane, pH 6.8), and then boiled for 2 min. Each
sample solution (20.uI) \vas put on a 4-15°;;) gradient acrylamide slab gel
(Daiichi Pure Chemicals, Ltd.), and then electrophoresis was perfonned
to separate the labeled y-PGA from free 14( -glucose or 14C-glutamic acid.
The y-PGA was detected by staining with alcian blue as described by
Yamaguchi et al.,22) and the radioactivity of y-PGA was detected by
autoradiography. The spots corresponding to y-PGA were cut from the
gel and the radioactivity was counted in a toluene scintillation solution.
Results
Effects of the NaCI concentration on y~PGA fermentation
At first, y-PGA production in a jar fermenter could not
be done, because the viscous broth foamed out of the jar
fermenter. To prevent foaming, several antifoaming agents
were examined, but the foaming of the viscous medium was
not suppressed by any antifoaming agent, such as oil or
silicone. The most effective treatment for preventing
foaming was the addition of NaCl to the medium.
Figure 1 shows the effects of the NaC] concentration on
y-PGA fermentation. With ] or 2% NaCl, y-PGA pro-
duction became maximum at 24 and 50 h, respectively, al-
though free glutamic acid remained in the medium. With
N aCI at these concentrations, the medium changed into a
mousse after 24 and 50 h, respectively, and then foamed
out. In the case of 3 or 5% NaCL foaming of the medium
was prevented and the y-PGA fermentation could be done
stably until cultivation was stopped. But the y-PGA
production with 5% NaCI was repressed after 48 h, and the
amount of y-PGA remained at 12 mg/ml. An effective
y-PGA fermentation was possible with 3°1<) NaCl. With this
concentrations, glutamic acid added to the medium seemed
to be efficiently converted into j)-PGA, because the final
amount of y-PGA was equal with the total glutamic acid
level, and free glutamic acid was not detected in the medium.
Efficient conversion was also observed even with in-
l%NaCI
,-.,. 30....--------,
E
~ 25
-.5 \".-"'-- --A. ~.
20 \",
'"C , _._ --ill
~ 15
Q
E 10
CI:l
~ 5 /,,'" ...--.
C!:l 0 I...../_ _ _ _ _ _ _ _-......--J
o ~ ~ n %0
Cultivation Time
Fig. 1. Effects of the NaCI Concentration on y-PGA Fermentation.
B. ~lIhlili.) (mllW) strain MR-141 WetS cultured in it 30-liter Jar fermenter containing 20 liters of MSG medium containing various amounts of NaCI. The culture conditions
and 1,-PGA analysis were described under MaterIals and Methods. Symbols: .A., total glutamic acid; . , free glutamic acid; . , y-PGA.
1685
creasing amounts of sodium L-glutamate in the medium
(data not shown). With every concentration of sodium L-
glutamate, the y-PGA production continued until the free
glutamic acid had been consumed. Since the amount of total
glutamic acid was maintained at a constant level during
these cultivations, it was deduced that strain MR-141 could
not use glutamic acid as a carbon source.
Effects of agitation on y-PGA fermentation
To obtain efficient production of y-PGA, the effects of
agitation on y-PGA fermentation were examined using
MSG medium containing 6% sodium L-glutamate (Fig. 2).
Under agitation at 300 or 350 rpm, the rates of y-PGA
production were low, and above 15 mg/ml of free glutamic
acid remained in the broth. In the case of 400 or 450 rpm
agitation, the rates of y-PGA production were prominently
higher than with 300 or 350 rpm. But, the agitation effect
seemed to be maximum at 400 rpm, since 450 rpm gave
similar y-PGA production.
y-PGA fermentation in a glucose or maltose medium
Figure 3 shows the course of y-PGA production in
medium containing either glucose or maltose as a carbon
source. In the GSG medium containing glucose as a carbon
source, the cell numbers quickly increased up to 18 hand
was maintained at a constant level after 24 h. The glucose
concentration in the medium quickly decreased in the
logarithmic phase, and was not detected after 30 h. The
pH of the broth also decreased while glucose remained.
However, when glucose had been consumed, the pH quickly
increased for several hours and then was maintained at from
7 to 8. The initial y-PGA production increased with cell
growth in the logarithmic phase. Moreover, after 48 h, when
glucose had been consumed, the amount of y-PGA also
gradually increased without cell growth. However, the
efficient conversion of free glutamic acid into ,},-PGA did
not occur within 90 h, because the rate of y-PGA produc-
tion was slightly decreased in the stationary phase. Thus,
the yield of y-PGA in the GSG medium remained at only
25mg/ml.
On the other hand, in the case of cultivation in the MSG
medium containing maltose as a carbon source, the cell
growth pattern resembled that in the GSG medium, but the
speed of assimilation of maltose was very low and maltose
remained until 90 h. Then, the pH was maintained at a low
2%NaCI 3%NaCI 5%NaCI
~ ~ n %0 ~ ~ n %0 24 48 72 96
(h)
NII-Electronic Library Service
1686 Y. OGAWA et al.

300rpm 350rpm 400rpm 450rpm

50~--------~ ~--------~ ------------~ r----------~

~ 40
b.O
-5
"'0 30
o
-< 20

~ O~~~~~~ ~~~~~~ ~~--~~--~ ~--~~----~

o 24 48 72 96 0 24 48 72 96 0 24 48 72 96 0 24 48 72 96
Cultivation Time (h)
Fig. 2. Effects of Agitation on 'Y-PGA Fermentation.
B. subtilis (natfo) stram MR·141 was cultured in a 30-liter jar fermenter containing 20 liters of a MSG medium containing 6% sodium L-glutamate. The agitation speed
waS changed from 300 to 450 rpm. The other culture conditions and symbols are the same as those in Fig. 1.

G S G Medium MS G Medium 3
700.....----------.
"'C
~600 4) .......
'c 500 +-'::15
;::;:'400
coo..
"-0
+' 0
'\j;300 0.0
;;;;:200 "-0 2
l00L-_ _ -------J 00
(,),...
.5: x
(/)
4)
.-<
+-'w
.- 0..
> I
:;:;~
(,)
co 0
o+'
.- c:
"'C.-
co
c::::
0
0 24 48 72
Cultivation Time (h)
Fig. 4. 14C-Glucose or 14C-L-Glutamic Acid Incorporation into 'Y-PGA.
B. subtilis (natta) strain MR-141 was cultured in a 50-ml flask with rotary shaking
using 10ml of the GYG medium described under Materials and Methods. The
radioactivity incorporated into y-PGA from derived 14C-glucose C.) or 14C_L_
glutamic acid (.) was measured.

production in a jar fermenter described above, the incor-

o~------~------~
o 24 48 72 96 0 24 48 72 96
Cultivation Time (h)
Fig. 3. Effects of the Carbon Source on 'Y-PGA Fermentation.
B. subtilis (l1atto) strain MR-141 was cultured in a 30-liter Jar fermenter containing
20 liters of MSG medium containing 5(% sodium L-glutamate or GSG medium
containing 6% glucose 1I1stead of maltose The other culture conditions and symbols
are the same as those in Fig. I.
value in the stationary phase, and the total glutamic acid
level did not change during the cultivation. The initial
y-PGA production also increased with cell growth in the
MSG medium. Moreover, the rate of y-PGA production in
the stationary phase was almost equal to the initial rate.
Finally, 35 mg/m! of y-PGA, the highest yield in our ex-
periment, was obtained.
14C-Glucose and 14C-L-glutamic acid incorporation into
y-PGA
From the effects of glutamic acid addition on y-PGA
poration of the glutamic acid into y-PGA was apparent. To
clarify the role ofL-glutamic acid and the pathway ofy-PGA
biosynthesis, the incorporation of 14C-glucose or 14C_L_
glutamic acid into y-PGA was examined.
Radioactivity was detected in the y-PGA produced in the
medium containing 14C-glucose or 14C-L-glutamic acid
(Fig. 4). The radioactivity of 14C-L-glutamic acid was
actually incorporated into y-PGA and increased con-
comitantly with y-PGA production. But, under culture
conditions like these in a 50-ml flask, the yield of y-PGA
was maintained at only 7 or 8 mg/ml after 48 h. Then, only
about 35% of the total radioactivity of 14C-L-glutamic
acid was detected in y-PGA. On the other hand, the
incorporation of radioactivity of 14C-glucose into y-PGA
was only about 6% and was maintained at this low level
for 72 h. On autoradiography, intermediate compounds of
low molecular weight were not detected even in the early
period of cultivation (Fig. 5). Moreover, other protein bands
labeled with 14C-glucose or 14C-L-glutamic acid were
scarcely detected.

NII-Electronic Library Service

 Production of y-Polyglutamic Acid
Cultivation Time (h)
o 5 8 11 14 17 20 24 28 32 36 40 44 48 54 60 66 72
,;,
Fig. 5. Autoradiogram of y-PGA Labeled with 14C-L-GJutamic Acid.
Culture broth (10 pi) of B. subtifis (natto) strain MR-141 was analyzed by SDS-
PAGE. The radioactivity ineorporated into y-PGA from l"C-L-glutamic acid was
detected by autoradiography.
Discussion
To produce y-PGA on an industrial scale, we searched
for suitable conditions for y-PGA fermentation in a jar
fermenter. We succeeded in obtaining very efficient pro-
duction of y-PGA (about 35 mg/ml). However, even under
our suitable conditions, supplemented air was not efficiently
dispersed in the broth due to its high viscosity, such as
above 1500 mPas. Thus, more effective y-PGA production
may be possible with extra oxygen.
From the results of 14C-glucose and 14C-glutamic acid
incorporation, it was deduced that y-PGA may be syn-
thesized via both de novo and salvage pathways in the
logarithmic stage with cell growth, but in the stationary
stage when glucose has been consumed, y-PGA may be
produced independently of cell growth and exclusively
synthesized from free glutamic acid in the medium via a
salvage pathway. 'We obtained the same results in the case
of strain NR-I isolated from the Naruse strain, which is a
strain commercially used as a natto starter (data not shown).
But, on a small scale cultivation such as in a 50-ml flask,
efficient y-PGA production like that in a jar fermenter was
not possible, because the dissolved oxygen concentration
may be limited.
K unioka and Goto reported that glutamic acid added to
the medium was not assimilated by B. subtilis IF03335,
and may be a regulator of the enzyme involved in y-PGA
biosynthesis. 20 ) Then, the glutamic acid unit in y-PGA was
assumed to be synthesized via a de novo pathway from citric
acid and ammonium sulfate. But, our cultivation results
suggested that effective conversion of glutamic acid added
to the medium into y-PGA was very important for producing
a large amount of '},-PGA (Fig. 3). Moreover, the radio-
activity of glutamic acid was actually incorporated into
y- PGA (Fig. 4). Therefore, in the case of B. subtilis (natto) ,
L-glutamic acid in the medium is likely to serve as a material
for y-PGA biosynthesis, rather than as a regulator. The
difference between B. subtilis (natto) and B. subtilis IF03335
suggests that the mechanism of y-PGA biosynthesis may be
different in these strains or under each sel of culture con-
ditions.
Ito et al. suggested an intermediate of y-PGA existed in
the medium of B. subtilis TAM-4, which synthesized y-PGA
1687
via a de novo pathway. 16) But, in our experiments, an
intermediate was not detected in the autoradiogram (Fig.
5). Therefore, in the case of B. subtilis (natto), y-PGA seems
~y-PGA
to be secreted into the medium at a high molecular mass.
Moreover, other proteins bands labeled with 14C-glutamic
acid were scarcely detected under the y-PGA bands. Thus,
glutamic acid was efficiently incorporated into y-PGA rather
than protein.
Tory suggested that polyglutamyl synthetase activity
detected in membrane particles required ATP for y-PGA
biosynthesis. 23) In our case, y-PGA biosynthesis appeared
to require energy for the polymerization reaction, because
high productivity was obtained by using maltose, which was
gradually used as an energy source during cultivation (Fig.
3), and by increasing the agitation speed to increase aerobic
metabolism (Fig. 2). Thus, we consider that the y-PGA
productivity in the stationary phase will be further increased
by the addition of suitable amounts of sugar and glutamic
acid as energy and material sources, respectively.
References
1) H. Fujii, Nippon Nogeikagaku Kaishi (in Japanese), 36, 1000~1O04
(J 962).
2) H. Fujii, Nippon Nogeikagaku Kaishi (in Japanese), 37, 346-350
(1963).
3) H. Fujii, Nippon Nogeikagaku Kaishi (in Japanese), 37, 407-411
(1963).
4) H. Fujii, Nippon Nogeikagaku Kaishi (in Japanese), 37, 474-477
(1963).
5) H. Fujii, Nippon Nogeikagaku Kaishi (in Japanese), 37, 615-618
(1963).
6) H. Fujii, Nippon Nogeikagaku Ka ish i (in Japanese), 37, 619-622
(1963).
7) K. Yahata, J. Sadanobu, and T. Endo, Polym. Prepr., Jpn., 41, 1077
(1992).
8) H. J. Choi, R. Yang, and M. Kunioka, J. Appl. Polym. Sci., 58,
807-814 (1995).
9) H. 1. Choi and M. Kunioka, Radiat. Phys. Chem., 46, 175-179
(1995).
10) G. Ivanovics and L. Erdos, Z. lmmunitiits/orsch., 90, 5-19 (1937).
11) G. Ivanovics and V. Bruckner, Z. lmmunitiit~rorsch., 90, 304-318
(1937).
12) M. Bovarnick, J. Bioi. Chem., 145,415-424 (1942).
13) c. B. Thorne. C. G. Gomez, H. E. Noyes, and R. D. Housewright,
J. Bacterio!" 68, 307-315 (1954).
14) S. Sawa, S. Murao. T. Murakawa, and S. Omata, Nippon
Nogeikagaku Kaishi (in Japanese), 45, 124-129 (1971).
15) C. Cheng, Y. Asada, and T. Aida, Agric. Bioi. Chem., 53, 2369-2375
(1989).
16) Y. Ito, T. Tanaka, T. Ohmachi, and Y. Asada, Biosci. Biotech.
Biochem., 60, 1239-1242 (1996).
17) C. B. Thorne, C. G. Gomez, G. R. Blind, and R. D. Housewright,
J. Bacterio!', 65, 472-478 (1953).
18) H. Kubota, T. Matsunobu, K. Uotani, H. Takebe, A. Satoh, T.
Tanaka, and M. Taniguchi, Biosci. Biotech. Biochem., 57,1212-1213
(1993).
19) A. Gato and M. Kunioka, Biosci. Biotech. Biochem., 56, 1031-1035
(1992).
20) M. Kuniaka and A. Goto, Appl. Microbial. Biotechnol., 40, 867-872
(1994).
21) F. Yamaguchi, Y. Ogawa, E. Yamazaki, and K. Yuasa, Japan Kokai
Tokkyo Koho, 8154616 (June 18, 1996).
22) F. Yamaguchi, Y. Ogawa, M. Kikuchi, K. Yuasa, and H. Motai,
Biosci. Biotech. Biochem., 60, 255-258 (1996).
23) F. A. Troy, J. Bioi. Chem., 248, 305-315 (1973).
NII-Electronic Library Service