Lactic acid production by Bacillus
coagulans—Kinetic studies and
optimization of culture medium for
batch and continuous fermentations
T. Payot, Z. Chemaly, and M. Fick
Laboratoire des Sciences du Ge´nie Chimique, CNRS, E.N.S.A.I.A. Vandoeuvre cedex, France
Bacillus coagulans is an atypical strain for lactic acid production; the thermophile character of this strain
(growth at 52°C) proves that it is particularly adapted for industrial production of lactate without sterile
conditions.
In the first step, continuous culture was performed to define the experimental domain of aeration and nitrogen
supplementation. The aerobic condition showed a positive influence on growth and a negative effect on lactate
production. At steady state for pH 6.4, the concentrations of biomass and lactic acid were 3.9 and 19.5 g l
21
,
respectively, without aeration and 4.6 and 11 g l
21
with aeration. The nitrogen source is essential for the
fermentation process. Pulses of different types of yeast extract (liquid and powder) were added into the fermentor
at steady state. After pulses, biomass concentration increased two and three times with powdered yeast extract
and liquid yeast extract, respectively. Liquid yeast extract was more efficient for growth than powdered yeast
extract probably due to degradation of vitamins in a spray dryer. Secondly, a factorial fractional experimental
design was performed to optimize batch fermentation. Temperature and pH control, the initial concentration of
sugar, and the nitrogen source were optimized. For the initial sucrose concentration of 60 g l
21
, productions of
biomass and lactic acid were 3.1 and 55 g l
21
, respectively. The maximal specific production rate of lactic acid
is high (6.1 60.3 g/l zh/g/l cell) in comparison with mesophilic lactic acid bacteria. © 1998 Elsevier Science
Inc.
Keywords: Lactic acid; Bacillus coagulans; continuous fermentation; kinetic studies
Introduction
Lactic acid is one of the most widely used organic acids in
the food industry and is a very common substrate for
chemical synthesis.
1
Recently, there has been an increased
interest in lactate because it can be used as raw material for
the production of biodegradable polymers with applications
in medical (high resorption thread), pharmaceutical (low
diffusion drug), and food industries (packaging).
2
Now,
production of lactic acid is achieved with a fermentation
process using whey permeate
3,4
or molasses
5,6
as carbon
source. Various mesophilic lactic bacterial strains
4
are used
to produce lactic acid such as Lactobacillus helveticus,
4
Lactobacillus delbrueckii,
7
and Lactobacillus plantarum;
8,9
however, these mesophilic strains are not adapted for
industrial production of lactate because high contamination
risks are subversive. Thermophilic strains are ideal for this
production because sterile conditions are not necessary
anymore. For this reason, a number of studies used Bacillus
coagulans as a lactic acid producer
10–12
at high tempera-
ture
10
(55°C). Sucrose and yeast extract were supplemented
with sulfate, phosphate, magnesium, manganese, and iron
salts.
11
In this paper, an atypical strain of B. coagulans was
studied in batch and continuous mode to prove the real
potentiality and the performance of this bacteria. Optimiza-
tion of temperature, pH regulation, nitrogen source concen-
tration, and salt supplementation was performed since these
factors are essential for lactic acid producers.
3,8,13–15
First,
continuous processes were performed to define the experi-
mental domain conditions such as influence of aeration,
nitrogen, and dilution rate. After that, an experimental
design was developed to find the influence of nutrients such
Address reprint requests to Dr. M. Fick, CRNS E.N.S.A.I.A., Lab des
Sciences du Genie Chimique, BP 172, 2 avenue la Foret de Haye, 54505
Vandoeuvre cedex, France.
Received 1 May 1997; revised 29 June 1998; accepted 16 July 1998
Enzyme and Microbial Technology 24:191–199, 1999
© 1998 Elsevier Science Inc. All rights reserved. 0141-0229/99/$–see front matter
655 Avenue of the Americas, New York, NY 10010 PII S0141-0229(98)00098-2
as yeast extract, sugar, salts, and to optimize physical
parameters (pH and temperature). Finally, advantages and
the inconvenience of B. coagulans TB/04 were discussed
and compared with another B. coagulans
11
and with Lac-
tobacillus rhamnosus studied in our laboratory.
Materials and methods
Strains and medium
B. coagulans TB/04 is a characterized strain isolated by natural
selection using continuous culture. In our experimental conditions,
this Bacillus was shown to be a homofermentative lactic acid
bacteria. The bioconversion capability was always superior to
90%. The bacterial was grown on sucrose.
Selection of bacterial strain
Four strains characterized as B. coagulans were compared (indus-
trial collection, Brussels Biotech, Brussels, Belgium). The most
productive strain was used for this work and named B. coagulans
TB/04.
Biomass analysis
Cell concentration was estimated by measurement of optical
density at 570 nm correlated to dry cell weights.
Sugar and organic acids analysis
Sugar and lactic concentrations were determined by offline high
performance liquid chromatography (Waters Associate, Milford,
MA). Sucrose, fructose, d-glucose and l-lactic acid concentrations
were determined enzymatically with a sucrose/d-glucose diagnos-
tic kit (Boehringer Mannheim 1113950, Mannheim, Germany);
d-glucose/d-fructose diagnostic kit (Boehringer Mannheim
139106); and a l-lactic acid diagnostic kit (Boehringer Mannheim
139084). The l-lactic acid was measured by a second enzymatic
method with a glucose-lactate analyzer YSI 2000 (Bioblock,
Illkirch, France). HPLC was used as a qualitative screening of
media composition and for estimation of by-products using a
UV-RI detector. Enzymatic techniques were used to determine the
concentration of metabolites.
Conductivity measurements
Samples and online measurements were performed with a conduc-
tivity probe WTW
16,17
(Wissenschaftlich-Technische Werksta¨tten
GmbH D-8120, Weilheim, Germany) LF 96 with a Tetracon 96
conductivity cell. This technique was used for the online estima-
tion of lactic acid concentration.
Preparation of bacterial extract
Fermentation broth was recovered after batch fermentation and
centrifugation (4,500 rpm for 20 min). Cells were washed three
times with saline water (0.9% NaCl) and disintegrated in a
bead-mill (5 min with 0.1 mm glass beads). Cell debris was
washed two times with water the homogenate was hydrolyzed with
6nH
2
SO
4
at 90°C for 2 h. The preparation was neutralized with
6nNH
3
and centrifuged at 4,000 rpm for 30 min. The supernatant
was concentrated by atomization (100°C and 150 ml h
21
).
Concentrate was used as bacterial extract.
Fermentation protocol
Sterile culture medium (200 ml) was inoculated with 15 ml
activation medium (cell concentration about3gl
21
) and incubated
for 12 h at 50°C in a 250-ml mini-reactor (Wheaton Instrument,
Millville, NJ). The agitation speed was set up at 200 rpm, and then
Figure 1 Continuous bioreactor
Table 1 Activation medium and culture medium for
B.
coagulans
Activation medium Culture medium
Sucrose 20 g l
21
100gl
21
Tryptone 16 g l
21
—
Yeast extract 10 g l
21
2gl
21
NaCl 5 g l
21
—
(NH
4
)
2
SO
4
— 1.7gl
21
(NH
4
)
2
HPO
4
— 1.0gl
21
pH 6.4 6.4
Temperature 50°C 50°C
Papers
192 Enzyme Microb. Technol., 1999, vol. 24, February 15/March
100 ml of this culture was aseptically transferred into 900 ml
culture medium in a 2-l fermentor (Setric 2M SGI, Toulouse,
France) with an Ingold pH probe.
Continuous fermentation
Medium feeding and the stage probe were activated with a
peristaltic pump (Ismatec and Maton Lesquin, Santa Clara, CA,
TH 50LF52) after a batch phase of 12–14 h. The activation and
culture media were defined for B. coagulans BB/ZVHB (Table 1).
Fermentations (1 l) were performed to study the most important
parameters for growth of and the lactic acid production by B.
coagulans (Figure 1).
At steady state, pulse and shift techniques were used to study
the influence of different types of nitrogen sources. Pulses were
performed with liquid yeast extract (Biokar ref. 112002, Beauvais,
France) and powder yeast extract (Biospringer ref. 103022, Mai-
sons-Alfort, France).
Batch fermentation and experimental design
The influence of different nutrients, pH, and temperature on the
batch production (Figure 2) of lactic acid by B. coagulans were
studied (Table 2). A factorial fractional experimental design 2
IV
9–5
was used and the maximal specific lactic acid production rate (g
lactic acid g
21
cells h
21
), maximal specific growth rate (h
21
),
maximal specific sugar consumption rate (g g
21
cells h
21
),
maximal concentration of biomass (g l
21
), end lactic concentration
(g l
21
), and lactic acid yield (%, lactate produced/initial sucrose
concentration) were determined.
Fermentation conditions. Lactic acid fermentation was per-
formed at 50 or 52°C. The broth was stirred at 300 rpm and the pH
was maintained at 6.2 or 6.4 by automatic addition of 2–8 nNH
3
.
Results
Continuous fermentation
Influence of dilution rate. A continuous fermentation was
performed to find the influence of dilution rate on growth
and lactic acid production at pH 6.4 and 50°C. Medium was
prepared with 100 g l
2
sucrose, liquid yeast extract (4 g l
2
),
and ammonium salts (NH
4
)
2
HPO
4
(1.0 g l
2
) and NH
4
)
2
SO
4
(1.7 g l
2
). This study was performed to estimate maximal
specific growth rate (m
max
) at steady state.
The continuous state was started after a batch phase of
18 h. Fresh medium was added into the reactor with a
peristaltic pump at a very low dilution rate (0.03 h
21
). At
steady state for each dilution rate, concentrations of bio-
mass, sugars, lactic acid, and other organic acids were
determined. The results showed a low consumption of
sucrose at all dilution rates (,25%), only a quarter of
sucrose was used by the cells, the global yield did not
exceed 20% (Table 3), and the lactate/sucrose yield de-
creased as the dilution rate increased. The highest biomass
was obtained at a 0.15 h
21
dilution rate while the maximum
lactate was achieved at 0.03 h
21
. Fructose was not detected
at low dilution rates (below 0.1 h
21
) but increased with the
dilution rate. The total concentration of other products
(acetate, pyruvate, ethanol, fumarate, formate, and citrate)
did not exceed2gl
21
at all dilution rates considered
Decreased sucrose consumption could be due to a high
concentration of sucrose or a deficiency of vitamins, pep-
tides, or salts. Since sucrose is hydrolyzed to glucose and
fructose, the fructose utilization pathway seems to be
limited
11
(fructose accumulation at high dilution rates). The
homofermentative character of the strain for our conditions
is confirmed by the amounts of acetate and ethanol pro-
duced (,2gl
21
). Finally, the maximal specific growth rate
was estimated around 0.3 h
21
.
Influence of pH regulation. The second continuous fer-
mentation was performed at 50°C to test pH influence on
biomass and lactic acid production. In order to reduce
inhibition pressure, a new continuous culture at high dilu-
tion rate (0.2 h
21
) was conducted between pH values of
6.0–7.0 with2gl
21
yeast extract and 60 g l
21
sucrose
(molasses) without ammonium salts. The results show that
optimal pH for lactic acid production is defined around 6.5
(Table 4). Glucose and fructose were not detected for pH
values between 6.2–6.8, but a very low concentration of
fructose (,2gl
21
) was detected for pH values between
6.0–7.0 (data not shown). Other by-products such as ace-
tate, pyruvate, ethanol, fumarate, formate, and citrate were
detected at very low concentrations (total ,2gl
21
60.5).
Influence of aeration. This parameter was studied under
the same continuous culture described for the previous
study. For two different pH values, the culture medium was
Figure 2 Batch bioreactor
Table 2 Parameters studied using a factorial fractional experi-
mental design
N° Parameter Level
(21) Level
(1)
1 Molasses 120 g l
21
a60gl
21
a
2 Powder yeast extract 0 g l
21
2gl
21
3 Liquid yeast extract 0 g l
21
2gl
21
dw
4 Bacterial extract 0 g l
21
2gl
21
dw
5 (NH
4
)
2
SO
4
0gl
21
2gl
21
6 (NH
4
)
2
HPO
4
0gl
21
1gl
21
7 pH 6.1 6.4
8 Temperature 50°C 52°C
9 Tween 80 0 ml l
21
1mll
21
Nine parameters studied for two levels (21) (1). Dry weight, dw
a
Sucrose equivalent concentration
Lactic acid production by
B. coagulans
: T. Payot
et al.
Enzyme Microb. Technol., 1999, vol. 24, February 15/March 193
aerated (0.06 m
3
h
21
). The biomass concentration increased
under aerobic conditions (Table 5) with a decrease in lactic
acid production.
Temperature effect. This continuous culture was identical
to that of previous studies (same medium composition and
dilution rate) with a pH regulation at 6.5. The temperature
was tested between 50–58°C. The optimal temperature for
growth and lactic acid production was found between
50–52°C. B. coagulans grew at 58°C. This shows its
thermophile character (Table 6).
Influence of yeast extract. A continuous culture was
performed with a low concentration of nitrogen source:
sucrose, 60 g l
21
; yeast extract,2gl
21
; without ammonium
salts, pH 6.5, and a temperature of 50°C. Two pulses of
yeast extract (powder and liquid yeast extract, respectively)
were successively added into the fermentor. A first pulse of
powder yeast extract was added at steady state. After
waiting for the steady state to be reached again, a second
pulse of liquid yeast extract was then added. Figure 3 shows
the influence of the two types of yeast extract on growth.
The single yeast extract concentration curve corresponds to
both types since they have the same dry weight. Yeast
extracts considerably increase biomass (two times and three
times, respectively, with powdered yeast extract and liquid
yeast extract). Liquid yeast extract is more efficient than the
powder at the same dry weight. Supplementation of the
medium with liquid yeast extract and powdered yeast
extract increased the lactic acid production from about 9 to
19gl
21
and 13.5 g l
21
, respectively (data not shown).
Batch fermentation
Eighteen batch fermentations were performed to study the
influence of nine parameters at two different levels of
experimental design (Tables 2 and 7). Experiment n°15 was
repeated (15A, B, and C) for statistical evaluation of this
study. These experiments were performed to evaluate the
influence of a high concentration of molasses (diluted 5
times 512°Bricks 5120gl
21
equivalent sucrose),
different nitrogen sources (yeast extract of liquid and
powdered types and bacterial extract), ammonium salts and
Table 3 Influence of dilution rate on the production of biomass, lactic acid, and other by-products produced by
B. coagulans
and
residual sugars left in the fermentor
Dilution rate
(h
21
)Biomass
(g l
21
)Lactic acid
(g l
21
)
Sucrose,
residual
(g l
21
)Glucose
(g l
21
)Fructose
(g l
21
)Yield
(%) By-products
a
(g l
21
)
0.03 0.7 20.1 77 0 0 93 1
0.05 0.9 19.5 78 0 0 92 1.1
0.1 0.9 19 79 0 0 93 0.9
0.125 1.1 18.5 79 0 0.2 90 0.9
0.15 1.1 18 79 0 0.5 89 1
0.175 1.05 17.6 79.5 0 0.6 90 1.2
0.185 1.02 16.4 80.5 0 0.7 88 1.3
0.2 1 14.9 81 0 0.9 87 1.4
0.225 0.95 13.7 83 0 1.7 88 1.3
0.25 0.92 12.5 83 0 2.4 86 1.4
0.275 0.88 12.3 83 0 2.9 84 1.6
0.3
b
0.5 10 82 0.9 4.8 83 1.9
a
By-products such as acetate, pyruvate, ethanol, formate, citrate, and fumarate
b
m
max
evaluation (washout dilution rate)
Table 4 Influence of pH on the growth of and lactic acid
production by
B. coagulans
TB/04 in continuous fermentation at
0.2 h
21
dilution rate and 50°C
pH regulation
(3 Nammonia) Biomass
(g l
21
)Lactic acid
(g l
21
)Sucrose residual
(g l
21
)
6.0 3.8 13 41.5
6.2 3.9 17.5 37.6
6.4 3.9 19.5 35.5
6.5 4 22.5 32.3
6.6 3.95 18.8 35.7
6.8 4 18.5 36.5
7.0 3.8 15.4 40.2
Table 5 Influence of aeration on the growth of and lactic acid
produced by
B. coagulans
TB/04
pH O
2
Biomass(g l
21
) Lactic acid(g l
21
)
6.0 23.8 13
6.0 14.5 9
6.4 23.9 19.5
6.4 14.6 11
Table 6 Temperature influence on the growth of and lactic acid
production by
B. coagulans
TB/04
Temperature
(°C) Biomass
(g l
21
)Lactic acid
(g l
21
)Sucrose
(g l
21
)
50 4.1 22 32.3
52 4.2 21.9 32.2
54 3.6 19 34.6
56 3.4 16.3 38.1
58 3 13 41.4
Papers
194 Enzyme Microb. Technol., 1999, vol. 24, February 15/March
Tween 80 (as surfactant to increase transfer of sugar and
lactic acid through membrane cells). The effect of the
bacterial extract was tested to evaluate the feasibility of
substituting yeast extract. High sugar concentration was
added to demonstrate sucrose inhibition. Further small
variations in pH and temperature were tested to determine
the process stability with a view to industrial scaleup.
The overall results obtained under the 15 different
experimental conditions were represented in Table 8.
Under condition n°15A, biomass (3.1 g l
21
) and lactic
acid (55 g l
21
) were obtained without accumulation of
fructose (Figures 4 and 5).
Analysis errors were estimated for lactic acid and su-
crose concentration at 5% (HPLC, enzymatic technique, and
conductivity) and 10% (HPLC), respectively. The total
concentration of the by-products did not exceed1gl
21
with
a very low accumulation of acetate and ethanol (Figure 6).
These results agree with the results obtained in chemostatic
studies: the same maximum specific growth rate (0.29 h
21
)
was obtained in both studies (Batch and Chemostat).
(Figure 7).
The mean kinetics of the three optimum experiments
(n°15A, B, and C) were calculated and summarized in Table
9. The results obtained with these three experiments show
that error does not exceed 10% for all the parameters
considered; moreover, the results of this culture could be
interpreted after global realization of experimental design.
Errors were determined with statistic studies given by
experimental design.
18
The effect of each parameter tested
is summarized in Table 10.
Discussion
B. coagulans TB/04 is an atypical strain for lactic acid
production. The optimum temperature for growth and lactic
acid production was around 52°C as with other thermophilic
Figure 3 Pulses of yeast extract. Effect on growth:
yeast extract theoretical concentration given by Eq. (1)
(■); concentration of biomass after the pulse of powder
yeast extract (E); and concentration of biomass after the
pulse of liquid yeast extract (h)
Y
5
Y
0(2
tD
)
Eq. (1)
where
Y
is the concentration of yeast extract;
Y
0
, the
initial concentration of yeast extract;
t
, the time after the
pulse, and
D
the dilution rate
Table 7 Factorial fractional experimental design
Sucrose Powder
Y.E. Liquid
Y.E. Bacterial
extract (NH
4
)
2
SO
4
(NH
4
)
2
HPO
4
pH Temperature Tween 80
121212121212121211
212121211 112121
321121211 1211 21
41 1212121211 1 1
521211211211 1 21
6121121211211 1
72111212111211
81 1121121212121
921212112111121
10 1 21211 1 21211 1
11 211211 1 211211
12 1 1 211211212121
13 212111 1 121211
14 1 2111212112121
15A 211112121211 21
15B 211112121211 21
15C 211112121211 21
16 1 1 1 1 1 1 1 1 1
Part of planning chosen for experiments, 1–16 experimental design; A–C, replicate experiments
Lactic acid production by
B. coagulans
: T. Payot
et al.
Enzyme Microb. Technol., 1999, vol. 24, February 15/March 195