ArticlePDF Available

Unstructured Models for Lactic Acid Fermentation–A Review

Authors:
  • Ferhat Abbas University of Setif-1, Setif, Algeria

Abstract

To describe a microbial process, two kinds of models can be developed, structured and unstructured models. Contrary to structured models, which take into account some basic aspects of cell structure, their function and composition, no physiological characteri-zation of cells is considered in unstructured models, which only consider total cellular concentration. However, in spite of their simplicity, unstructured models have proven to accurately describe lactic acid fermentation in a wide range of experimental conditions and media. A partial link between cell growth and production, namely the Luedeking and Piret model, is mostly considered by the authors. Culture pH is the main parameter to be considered for model development. Acidic pH leads to inhibitory concentrations of undis-sociated lactic acid, the main inhibitory component, which causes cessation of growth and then production. On the other hand, pH control at optimal value for LAB growth allows to overcome product inhibition (by the total lactic acid produced or its undissociated part); hence nutritional limitations have to be considered for model development. Nitrogen is mainly involved in cessation of growth, owing to the fastidious nutritional requirements of LAB, while lactic acid production ceased when carbon was exhausted from the me-dium. The lack of substrate inhibition when usual concentrations of carbon substrate are used should be noted.
ISSN 1330-9862 review
(FTB-2353)
Unstructured Models for Lactic Acid Fermentation
A Review
Abdallah Bouguettoucha
1
, Béatrice Balannec
2,3
and Abdeltif Amrane
2,3
*
1
Laboratory of Chemical Engineering Process, Faculty of Technology, Ferhat Abbas University,
DZ-19000 Setif, Algeria
2
National School of Chemistry of Rennes, University of Rennes 1, CNRS, UMR 6226,
Avenue du Général Leclerc, CS 50837, FR-35708 Rennes Cedex 7, France
3
European University of Brittany, 5 Boulevard Laënnec, FR-35000 Rennes, France
Received: September 28, 2009
Accepted: December 22, 2009
Summary
To describe a microbial process, two kinds of models can be developed, structured
and unstructur ed models. Contrary to structured models, which take into account some
basic aspects of cell structure, their function and composition, no physiological characteri-
zation of cells is considered in unstructured models, which only consider total cellular
concentration. However, in spite of their simplicity, unstructured models have proven to
accurately describe lactic acid fermentation in a wide range of experimental conditions
and media. A partial link between cell growth and production, namely the Luedeking and
Piret model, is mostly considered by the authors. Culture pH is the main parameter to be
considered for model development. Acidic pH leads to inhibitory concentrations of undis-
sociated lactic acid, the main inhibitory component, which causes cessation of growth and
then production. On the other hand, pH control at optimal value for LAB growth allows
to overcome product inhibition (by the total lactic acid produced or its undissociated part);
hence nutritional limitations have to be considered for model development. Nitrogen is
mainly involved in cessation of growth, owing to the fastidious nutritional requirements
of LAB, while lactic acid production ceased when carbon was exhausted from the me
-
dium. The lack of substrate inhibition when usual concentrations of carbon substrate are
used should be noted.
Key words: lactic acid bacteria, growth inhibition, unstructured models, nutritional limita
-
tions
Introduction
Kinetic models enable bioengineers to design and con
-
trol microbial processes. Mathematical models, together
with carefully designed experiments, allow the improve
-
ment of the evaluation and knowledge concerning sys
-
tem behaviour (1). To describe a microbial process, two
kinds of models can be developed, structured and un
-
structured models. Structured models take into account
some basic aspects of cell structure, their function and
composition, and have been reported to accurately de
-
scribe lactic acid fermentation (1,2), but they can seem
complex. Only total cellular concentration is considered
in unstructured models, and hence they do not involve
any physiological characterization of the cells. However,
they have proven to accurately describe lactic acid fer
-
mentation in a wide range of experimental conditions
and media (3–8). An exhaustive review of the available
3
A. BOUGUETTOUCHA et al.: Models for Lactic Acid Fermentation, Food Technol. Biotechnol. 49 (1) 3–12 (2011)
*Corresponding author; Phone: ++33 2 2323 8155; Fax: ++33 2 2323 8120; E-mail: abdeltif.amrane@univ-rennes1.fr
unstructured models is therefor e presented thereafter. The
paper is divided into two main parts: models for growth
kinetics are reviewed in the first part, followed by the
available models for production kinetics in the second
part.
Growth Kinetics
Some growth phases can be characterized by the ex
-
amination of growth time-courses: the lag phase, the ex
-
ponential growth phase, the deceleration growth phase,
the stationary phase and the phase of exponential decay.
The complexity of this biological phenomenon requires
the use of nonlinear mathematical models to identify
growth parameters. These models are described there
-
after.
Bacterial growth can be described as follows:
/1/
where X is the concentration of biomass, m is the specific
growth rate and t is time.
Some authors (9–14) also took into account cell death
in their growth kinetics model and considered an expo
-
nential decay for the decline phase:
/2/
with k
d
as the specific death rate.
Carbon limitation
The limitation by the carbon substrate (lack of inhi-
bition by the product) is often described by the Monod
model (15):
/3/
where S and k
S
are the substrate concentration and sub
-
strate saturation constant, respectively.
Some authors added to the Monod model a term for
inhibition by the carbon substrate (16), namely they con
-
sidered the Haldane equation:
/4/
where k
i
is the substrate inhibition constant.
An exponential substrate inhibition model (17)was
also tested:
/5/
However, according to Altiok et al. (16), a substrate
inhibition term did not appear to be relevant to describe
their experimental data. Regarding the carbon source
(whey), substrate concentration was too low to be inhi
-
bitory, taking into consideration the high productivities
reported for conventional batch cultures even at high
initial glucose or lactose concentrations (above 100 g/L)
(18–20).
Bâati et al. (21) proposed the following relation based
on the Monod model (Eq. 3) and governing cell multi
-
plication at low temperatures:
/6/
where q
p
is the specific production rate, T is temperature
and T
max
a maximum temperature beyond which there
is no more growth; the terms a (yield of biomass on lac
-
tate), b (maximal maintenance) and c (constant of affi
-
nity) are constants; and k
a
is the substrate catabolic con
-
stant of affinity of the non-proliferating cells.
Growth kinetics on multiple substrates was also de
-
scribed. Nancib (13) examined the effect of both glucose
(G) and fructose (F) concentrations during batch cultures
of Lactobacillus casei ssp. rhamnosus on date juice for
lactic acid production and took into account cell death
in their model:
/7/
where G and F are the glucose and fructose concentra
-
tions, respectively.
Bajpai-Dikshit et al.(22) proposed the following mo
-
del which involved the intracellular enzyme level to de
-
scribe the growth of Lactobacillus rhamnosus on multiple
substrates:
/8/
where m
max,i
is the maximum specific growth rate on the
substrate S
i
,
e
e
i
max,i
is the specific relative growth enzyme
levels inside the cell and k
Si
is the substrate saturation
constant of the substrate S
i
. The net specific growth rate
m on a medium containing two substrates in terms of
individual growth rates can be defined as:
/9/
where m
i
is the specific growth on the substrate i, a
1
and
a
2
are the control coefficients corresponding to the gene
-
tic and metabolic regulations inside the cell, respectively.
According to these authors, the developed kinetic model
can be used for the design and operation of batch and
continuous reactors, such as fed-batch and chemostat re
-
actors.
However, a carbon substrate limitation model (Mo
-
nod) is only record ed in case of a high nitrogen supple
-
mentation of culture media (23,24); while in the general
case, cessation of growth can be attributed to the defi
-
ciency in peptide sources (25,26) or in growth factors
(27,28).
Product inhibition
Inhibition by the total lactic acid formed
Luedeking and Piret (3) proposed a linear relation
between the inhibitory end-product and the specific
growth rate:
/10/
This relation where d is a constant and P is the lactic
acid concentration matched the results recorded for Lac
-
tobacillus delbrueckii grown on glucose (3). Belhocine (29)
4
A. BOUGUETTOUCHA et al.: Models for Lactic Acid Fermentation, Food Technol. Biotechnol. 49 (1) 3–12 (2011)
d
·
dt
=
X
Xm
d
d
··
dt
=-
X
XkXm
max
S
=
+
mm
S
Sk
max
2
Si
/
=
++
mm
S
Sk S k
max
Si
exp
S
k
æö
ç÷
=-
+
èø
mm
S
kS
()
max
max
p
a
a·q
-
=-
+
+-
m
TT
S
b
kS
cT T
maxG dG G maxF dF F
GF
d
d
GF
tG F
éùéù
=-+-
êúêú
++
ëûëû
mm
X
kX kX
kk
i
max,i i
max,i
i
Si i
e
S
e
kS
æö
ç÷
ç÷
èø
=
+
m
m
i1122
=+mamam
max
·=-mm dP
showed that the results obtained in continuous culture
of Lactobacillus helveticus grown on lactose depend on the
proportional inhibition (Eq. 10) rather than on substrate
limitation (Eq. 3) and/or a non-competitive inhibition by
the product, which were considered by many authors as
discussed below.
Several authors held into account in their growth
model an inhibition by the formed product. If this pro
-
duction is non-competitive, the specific growth rate be
-
comes (30–33):
/11/
where P and k
P
are the product concentration and pro
-
duct inhibition constant, respectively.
Owingtothelowk
S
values (some tens of mg/L),
which appear negligible in comparison with the experi
-
mental residual lactose concentration (1 or 2 g/L), a sub
-
strate limitation model (Eq. 3) may lead to the following
simplified expression:
/12/
while the specific growth rate remains constant only
during the exponential growth phase. Tayeb et al. (34),
as well as Ajbar and Fakeeha (35) considered therefore
only the product inhibition, without considering the
carbon substrate limitation:
/13/
Rogers et al.(4) attempted to describe experimental
data on batch culture using Streptococcus cremoris HP
1
var-
ious growth models, including the Kendall model (36):
/14/
the Monod model (Eq. 3) (15); a non-competitive inhibi
-
tion, with (Eq. 11) or without (Eq. 13) substrate limita
-
tion (34,37), or a modified non-competitive inhibition:
/15/
as well as the Edwards model (38), which was only a
modified non-competitive inhibition with an additional
term to account for a possible substrate inhibition:
/16/
According to Rogers et al. (4), if compared to Eq. 11,
Eqs. 15 and 16 did not improve growth fitting, so the
simplest non-competitive inhibition (Eq. 11) should be
preferred. These authors reported that lactose limitation
and lactic acid inhibition had a significant effect on the
growth, while initial substrate concentration and cell mor
-
tality did not affect growth significantly.
Ben Youssef et al. (12) modified the non-competitive
inhibition by adding a term to account for the effect of a
critical lactic acid concentration P
c
:
/17/
Some authors (39,40) added product inhibition term
to the Monod relation:
/18/
where P
inh
is the product concentration above which bac
-
teria do not grow.
Amongst the several models tested by Burgos-Rubio
et al. (41), a simplified expression of the above relation
was proposed:
/19/
The above expression was also considered by Mer
-
cier et al. (42) and Moldes et al. (43), who considered the
maximum lactic acid concentration, P
max
, instead of the
inhibitory concentration, P
inh
.
Kumar Dutta et al. (7), as well as Kwon et al. (20),
modified the above expression by considering the addi
-
tion of a toxic power for the product n:
/20/
Boonmee et al. (8) considered substrate limitation
and inhibition, as well as product inhibition, and hence
proposed the following model:
/21/
where P
m
and P
i
are the maximum inhibitory lactate con-
centration and the threshold level of lactate before an
inhibitory effect, respectively.
Nandasana and Kumar (14) modified the above mo-
del by considering an exponential decay for product in-
hibition:
/22/
It should be observed that lactose can only be inhi
-
bitory at high concentration levels (above 100 g/L) (18–20),
which is obviously not the case with whey, the usual
substrate used for lactic acid production. Consequently,
Pinelli et al. (33) did not consider the substrate inhibition
term in their model:
/23/
In agreement with Gonçalves et al. (44), who con
-
sidered both substrate and product inhibitions, Åker
-
berg et al. (45) added a product inhibition term to the
substrate inhibition relation of Briggs-Haldane (46):
/24/
With the following expressions, the pH dependence of
the parameters is described:
/25/
/26/
5
A. BOUGUETTOUCHA et al.: Models for Lactic Acid Fermentation, Food Technol. Biotechnol. 49 (1) 3–12 (2011)
P
max
SP
=
++
mm
Sk
SkPk
max max
S
+
mm m
S
Sk
P
max
P
=
+
mm
k
Pk
1
2
1
æö
ç÷
=-
èø
X
k
k
m
P
13
SP
æöæö
ç÷ç÷
=-
++
èøèø
m
Sk
kk
kSkP
Pi
1
SP i
S
æöæöæö
ç÷ç÷ç÷
=
+++
èøèøèø
m
Sk k
k
kkPkS
P
max
SP c
1
k
kP
æöæö
æö
ç÷ç÷
ç÷
=-
++
èø
èøèø
mm
SP
kS P
max
Sinh
1
æö
ç÷
=-
+
èø
mm
SP
kS P
max
inh
1
æö
ç÷
=-
èø
mm
P
P
n
max
Smax
1
S
kS
æö
ç÷
=-
+
èø
mm
P
P
ii
max
Si mi
1
æöæöæö
-
ç÷ç÷ç÷
=+
++ -
èøèøèø
mm
Sk PP
kSkS PP
i
max
Si P
exp
P
k
æöæö
æö
ç÷ç÷
ç÷
=-
++
èø
èøèø
mm
Sk
kSkS
max
SP
exp
k
æö
æö
ç÷
ç÷
=-
+
èø
èø
mm
SP
kS
()
n
max pH
2
Si
=-
++
mm
S
KP
kSSk
()()
[] []
m
max
12
1/H H
++
=
++
m
m
kk
mm
()()
[] []
pm
pH
p1 p2
1/H H
++
=
++
K
K
kk
where K
pH
is the parameter representing the pH depen
-
dence of product inhibition, and m
m
, k
m
, K
pm
and k
p
are
kinetic parameters that describe the effect of the pH on
m
max
and K
pH
.
A product inhibition term was also added to the
Briggs-Haldane relation by Ajbar and Fakeeha (35):
/27/
Ishizaki and Ohta (47) examined the fermentation of
L-lactate in batch culture of Streptococcus sp. IO-1 at vari
-
ous carbon substrate concentrations. They reported un
-
competitive inhibition, and hence proposed the follow
-
ing relationship:
/28/
where K
m
and k
p
are the Michaelis constant and the lac
-
tate inhibition constant for cell growth, respectively.
Some authors used a complex model to validate their
experimental results; Biazar et al. (48) tried to solve the
equation related to the growth kinetics of Lactobacillus
helveticus using an Adomian decomposition method (49,
50):
/29/
where K
iS
is the substrate concentration at which the sub-
strate inhibition factor is:
e
=
(/ )
.
SK
iS
n
1
0 368
; and K
iP
is the
lactic acid inhibition concentration at which the product
inhibition factor is:
e
=
(/ )
.
PK
iP
n
2
0 368
.
Peeva and Peev (51) considered only the inhibitory
effect by the product and hence proposed the following
model:
/30/
/31/
where k
d
is the cell death rate, k
p
is a coefficient for
product inhibition and P
max
is the theoretical lactic acid
concentration obtained after total substrate consumption.
Growth inhibition by lactic acid is only observed in
experiments carried out at acidic pH (52,53)orinthe
absence of pH control (54), therefore pH control is need
-
ed at its optimal value for lactic acid production (5.9)
(55,56) during culturing to overcome this inhibition. More
-
over, the lack of product inhibition during culturing at
usual carbon substrate concentrations, like the lactose con
-
tent of whey, has been clearly demonstrated (57). This
constitutes one of the main drawbacks of the above models.
Inhibition by undissociated lactic acid
It is now recognised that the main inhibitory compo
-
nent is the undissociated form of lactic acid. Inhibition
by weak organic acids is related to the solubility of the
undissociated form within the cytoplasmic membrane
and the insolubility of the ionised acid form (54,58); the
result is an acidification of the cytoplasm and the col
-
lapse of the motive force, causing an inhibition of nu
-
trient transport (59,60). It should be observed that in case
of pH control at its optimal value for lactic acid pro
-
duction (5.9), the final free lactic acid concentration during
culture on whey (approx. 0.3 g/L) is below the inhibi
-
tory threshold (54), leading to the absence of inhibitory
effect (61,62).
Yeh et al. (63) assumed a non-competitive inhibition:
/32/
with [HL] as the undissociated lactic acid concentration.
Some authors (9,11,64) added to the Monod model a
term to account for the inhibition by the lactate ion L
and an exponential decay to account for the inhibitory
effect of the undissociated lactic acid HL:
/33/
where L
max
and HL
max
are the dissociated and undisso
-
ciated lactic acid inhibition constants, respectively.
Leroy and De Vuyst (65) also added to the Monod
model an inhibitory term involving the concentration of
undissociated lactic acid with a toxic power n:
/34/
where g
N
is the remaining self-inhibition coefficient as
-
cribed to the limited availability of nutrients.
A drawback of the above models is their develop-
ment to describe cultures carried out at pH controlled at
the optimal value, namely close to 6, leading to a final
free lactic acid concentration below the inhibitory thresh-
old (54). However, such models can be useful to de-
scribe cultures carried out at acidic pH or without pH
control. With this aim, Amrane and Couriol (66) noted
that the specific growth rate decreased when undisso-
ciated lactic acid concentrations increase, and consequently
proposed the following logistic equation to describe cul
-
tures carried out without pH control:
/35/
where m
0
and [HL]
C
are the constants.
Vereecken and Van Impe (67) also proposed an ex
-
ponential decay, which involved the undissociated lactic
acid concentration and pH (or hydrogen ion concentra
-
tion [H
+
]:
/36/
/37/
where [HL]
min
is the minimum inhibitory concentration
of undissociated lactic acid, while a and b are constants.
To describe Lactobacillus plantarum growth in cucum
-
ber juice (vegetable fermentation), Passos et al.(10,68) took
into account NaCl and undissociated acetic acid concen
-
trations, [HA], added to the juice, in addition to a carbon
substrate limitation (Monod model), an inhibitory pH (hy
-
drogen ion) effect and an undissociated lactic acid con
-
centration effect:
/38/
6
A. BOUGUETTOUCHA et al.: Models for Lactic Acid Fermentation, Food Technol. Biotechnol. 49 (1) 3–12 (2011)
n
max
2
max
Si
1-
S
æö
ç÷
=
++
èø
mm
SP
P
kSk
()
max
mp
1
S
Pk S
=
++
mm
K
1
2
n
n
max d
SiSiP
exp exp
P
k
K
æö
æö
ç÷
ç÷
=---
+
èø
èø
mm
SS
kS K
()
max p d
1=--mm kP k
a
max
6.13· 0.056=-a P
[]
HL
max
SHL
HL
æö
ç÷
=
++
èø
mm
SK
Sk K
[]
[]
[]
max
S
max
max
1exp
æö
æö
ç÷
ç÷
=--
ç÷
ç÷
éù
+
ëû
èø
èø
mm
SL HL
kS L
HL
[]
[]
n
max N
S
max
1
æö
ç÷
=-
ç÷
+
èø
mm g
SHL
kS HL
[]
[]
max 0
C
exp
æö
ç÷
=--
èø
mm m
HL
HL
()
()
[][]
max min
exp - -=mm kHLHL
m
[]
2
b
a
H
+
=+k
m
[]
[]
[]
2.6
2.0
0
max
11
0.056 69
H
HL
H
+
+
æö
éù
æö
ëû
æö
ç÷
ç÷
=--
ç÷
ç÷
éù
ø
èø
ëû
èø
mm
S
S
/39/
pH inhibitory effect
In addition to the undissociated lactic acid concentra
-
tion (54), pH also plays a significant role in the inhibi
-
tory effect (69). Moreover, according to Fu and Mathews
(70), the inhibitory effect of acidic product resulted mainly
from the action of the proton ions, and hence pH can be
used as a basic parameter in kinetic models based on
the Monod model with m
max
and k
S
functions of the pH:
/40/
where the optimal parameter sets are correlated with pH
through the following empirical equations (4£pH£7) (70):
/41/
and /42/
Nitrogen limitation
The above models involved only the carbon substrate
limitation, mainly through the Monod model, as well as
product inhibition, thr ough the total concentration of the
produced acid or its undissociated form, or the conse-
quence of this production, a pH decrease. However, it is
not the general feature recorded during lactic acid fer-
mentation. Growth inhibition by lactic acid is observed
in experiments carried out at acidic pH (52,53)orinthe
absence of pH control (54), which is why pH control is
needed at its optimal value for lactic acid production (5.9)
(55,56) to overcome this inhibition. Moreover, the com
-
plex substrates containing peptidic nitrogen and growth
factors (71–73) added to the culture media are major con
-
tributors to the production cost of the final product (74,
75), and hence nitrogen limitations are usually observed
instead of carbon limitation of growth.
Amrane and Prigent (6,76) proposed a logistic func
-
tion (Eq. 13) to describe experimental data:
/43/
where c and d are constants.
Growth time-course was accurately fitted by means
of the above model; however, it was not completely
satisfactory from a cognitive point of view. Indeed, all
growth parameters did not have an obvious biological
meaning (6).
Consequently, the Verlhust model (77–79), which was
successfully applied to describe LAB growth (52,53,61,
62,80–84) may be preferred to the above model, since this
logistic expression involves only growth parameters:
/44/
where X
max
is the maximum biomass concentration.
The term
1
X
X
max
was taken in a global way for
an increasing lack of nutrients, namely for nitrogen limi
-
tation (23,24,78).
According to Lan et al. (85), growth kinetics can be
satisfactorily described by a modified Verlhust model,
which indirectly takes into account the inhibitory effect
of the product through an exponent m (86):
/45/
where k is an empirical constant related to the maximum
specific growth rate.
In addition to nutritional limitations through the Verl
-
hust model, Altiok and et al. (16) also considered an in
-
hibition by the produced lactic acid:
/46/
where f and h are parameters related to the 'toxic power'
for biomass and the inhibitory product, respectively. These
authors showed that the inhibitory effects on both bio
-
mass and product increased with the increase of h and f
toxic power values.
However, since the main inhibitor of growth is the
undissociated form of lactic acid (54,60), Bouguettoucha
et al. (62) replaced the total lactic acid concentration by
its undissociated form in the inhibition term:
/47/
where [HL]
inh
is the undissociated lactic acid inhibitory
threshold value, 8.5 g/L (69).
Owing to the fastidious nutritional requirements of
lactic acid bacteria (especially those concerning nitrogen)
(87–89), it appears difficult to include these limitations
in a growth model, and the available literature lacks in
models involving nutritional limitations in 'a direct way'.
Leh and Charles (5) tried to solve this difficulty, since
they considered carbon and nitrogen substrate limita
-
tions in their model. To account for both limitations, the
following modification of the Monod relation (Eq. 3) was
considered:
/48/
In this relation, pr and k
pr
are the concentration and
the saturation constant of 'usable proteins', respectively.
The difficulties encountered in the use of this model come
from the definition of the 'usable proteins'.
If the carbon substrate saturation constant k
S
is ne
-
glected during growth when compared to the carbon sub
-
strate concentration S (90), the above equation (Eq. 48)
can be simplified, leading to the Monod equation modi
-
fied to account for a nitrogen substrate limitation:
/49/
7
A. BOUGUETTOUCHA et al.: Models for Lactic Acid Fermentation, Food Technol. Biotechnol. 49 (1) 3–12 (2011)
[]
[]
[]
[]
[]
[]
NaCl NaCl
NaCl
1.7
0
1.5
0.35 1 1 ·
5.8 150
1.6
·1 1
4.47 11.8
æö
æö
=+ -
ç÷
ç÷
ø
èø
æö
æö
ç÷
+-
ç÷
ø
èø
m
HA HA
HA
()
pH
max
S
()
pH
=
+
mm
S
kS
()
()
()
2
max
0.265
0.523exp 0.16 pH 5.0
0.614 pH 4.0
=---
-
m
()
()
()
2
S
106.4
0.605exp 0.85 pH 5.0
0.65 pH 4.0
=-+
+-
k
()
max
max
1
·exp
1
=
+
-
mm
m
c d·t
c
max
max
1
X
X
æö
ç÷
=-
èø
mm
m
max
1
æö
ç÷
=-
èø
m
X
k
X
hf
max
max max
11
XP
XP
æöæö
ç÷ç÷
=- -
èøèø
mm
[]
[]
max
max
inh
11
XHL
XHL
æö
æö
ç÷
ç÷
=- -
ç÷
èø
èø
mm
max
pr pr
SS
1
·
=
+++
mm
kk
kk
pr S S pr
max
pr
·=
+
mm
pr
pr k
By adding the product inhibition term to Eq. 48, the
specific growth rate becomes:
/50/
The above model can only be 'usable' if a clear defi
-
nition and a relevant method for the determination of
the really 'usable nitrogen' by bacteria is given (91), which
is not at all obvious.
Schepers et al. (92) assumed that the specific growth
rate was a function of the carbon and nitrogen substrates,
the product and the pH, and proposed the following com
-
plex relation:
/51/
The above equation is the combination of four parts:
the first one (A) characterizes a strong interaction be-
tween pH and whey permeate, WP, for maximum spe-
cific growth rate deduced from multifactor kinetic ana-
lysis, leading to its involvement in the growth model
through coded factors, pH
c
and WP
c
(93); Monod sub-
strate limitations were assumed for carbon and nitrogen
substrate effects (part B); exponential inhibition by the
undissociated lactic acid and logistic inhibition by lactate
were taken into account through the third part (C); and
the last part (D) corresponds to the pH effect (combi
-
nation of Gaussian pH effect).
Production Kinetics
The model of Luedeking and Piret (3,94)isthemost
widely used concerning the kinetics of production.
Amongst others, Kumar Dutta et al. (7), Boonmee et al.
(8), Altiok et al. (16), Tayeb et al. (34), Keller and Ger
-
hardt (39), Burgos-Rubio et al. (41), Åkerberg et al. (45),
Biazar et al. (48), Roy et al. (52), Wang et al.(64), Vázquez
and Murado (83,84), Bibal et al. (95), Ye et al. (96), Ha et
al. (97) are authors who have shown that lactic acid pro
-
duction is partially associated with growth and then pro
-
posed the following relation:
/52/
In this relation, q
p
is the specific productivity rate, A
and B are coefficients for growth- and non-growth-asso
-
ciated production, respectively.
Total link between growth and production has been
considered by some authors (13,98), leading to the follow
-
ing simplified Luedeking and Piret expression:
q
p
=A·m /53/
However, this particular case is only recorded for
media supplemented with high concentrations of nitro
-
gen.
During lactic acid production by L. casei, Peeva and
Peev (51) found that acid production was mainly non
-
-growth associated, and hence proposed the following
relationship:
/54/
where b is the biomass productivity coefficient.
Amrane and Prigent (6,76) noted that the beginning
of the production is accurately described by Luedeking
and Piret relation, namely for significant values of the
specific growth rate. However, almost half of the lactic
acid is produced during the deceleration and the sta
-
tionary growth phases, whereas the specific growth rate
tends towards the zero value. This part of production is
not satisfactorily described by the Luedeking and Piret
relation, which cannot account for the decrease of the
specific production rate at low specific growth rates. The
Luedeking and Piret model was therefore modified by
introducing an additive term:
/55/
where H is a constant.
Rogers et al. (4) tested two substrate-dependent mo-
dels, in addition to the Luedeking-Piret model, and ob-
tained the following relation:
/56/
which was improved by Jørgensen and Nikolajsen (99):
/57/
where C is a constant.
Rogers et al. (4) also tested a substrate limitation mo
-
del which described more accurately their experiment
with S. cremoris:
/58/
The above substrate limitation model was also con
-
sidered by Berry et al. (11), as well as by Ben Youssef et
al. (12), who replaced the substrate saturation constant
k
S
by the affinity constant of the resting cells for glucose
k
S
rc
, which is the function of k
S
.
In addition to substrate limitation, Boonmee et al. (8)
took into account substrate inhibition, as well as pos
-
sible limitation and inhibition by the lactate in their
model:
/59/
where P
m
and P
i
are maximum and inhibitory threshold
of lactic acid concentrations for lactic acid production,
respectively.
Similarly to the growth (Eq. 22), Nandasana and
Kumar (14) modified the model by Boonmee et al. (8)by
considering an exponential decay for product inhibition:
/60/
8
A. BOUGUETTOUCHA et al.: Models for Lactic Acid Fermentation, Food Technol. Biotechnol. 49 (1) 3–12 (2011)
max
2
pr
pr
·
()·
1
=
+
+
mm
pr
P
pr k
k
[]
()
()
()
[] []
--
opt
max c c
Sz
A
B
HL
pH pKa
LL
pKa pH
C
pH pH
n
2
D
·pH ·WP · ·
exp
110
··
1exp
110
·e
14243
1442 4 43
14444444244444443
144 244 3
-
-
-
æö
ç÷
=+
ç÷
++
èø
æö
æö
ç÷
ç÷
-
ç÷
ç÷
+
èø
ç÷
ç÷
æö
æö
ç÷
ç÷
ç÷
+-
ç÷
ç÷
ç÷
+
èø
èø
èø
æö
ç÷
ç÷
ç÷
èø
mm b
SZ
Sk Zk
P
k
P
kk
s
P
d
d
==+m
P
qA·B
Xt
()
p
1
p
=-bqkP
a
()
p
1exp ·
éù
=+- -
ëû
mmqA·B H
p
··=+mqA BS
p
··=+-mqA BCS
p
S
·
æö
ç÷
=+
+
èø
m
S
qA B
kS
·
i
i
ppmax
Si mi
·1AB
æöæöæö
-
ç÷ç÷ç÷
=+ +
++ -
èøèøèø
m
PP
Sk
qq
kSkS PP
i
ppmax
Si P
·exp
Sk P
AB
kSkS k
æöæö
æö
ç÷ç÷
ç÷
=+ -
++
èø
èøèø
mqq
The model was found to provide good predictions
of experimental lactic acid production data.
In addition to a substrate limitation, Bâati et al. (21)
also considered an exponential decay for product inhi
-
bition:
/61/
where q
pmax
is the maximum specific lactic acid produc
-
tion rate, which was given by the following expression:
/62/
where K
b
and K
c
are constants.
Similarly to the relation proposed for growth (Eq.
19), Moldes et al.(43) proposed a logistic relation for the
production rate:
/63/
This relation can be helpful to describe experimental
data, but its biological meaning is not at all obvious,
owing to the absence of involvement of the biomass.
Monteagudo et al.(40) also considered a logistic
term for the lactic acid inhibition, which was added to
the Luedeking-Piret relation:
/64/
where P'
max
is the concentration greater than P
max
, above
which bacteria do not produce lactic acid.
Since the undissociated form of lactic acid is the main
growth inhibitor (54,60), Balannec et al. (82) considered
the undissociated form of the product instead of its total
amount. The inhibitory term was added to the non-
-growth-associated part of the production, to account for
cessation of production in case of culture without pH
controloratacidicpH:
/65/
On the other hand, during culturing at pH controlled
at 5.9, the exhaustion of the carbon substrate caused ces
-
sation of p roduction; a corrective term was therefore in
-
troduced to account for this behaviour (100):
/66/
The parameter S
lim
, which corresponds to the limit
-
ing lactose concentration (3 g/L), deduced from several
runs on whey supplemented with various yeast extract
concentrations (24,91), has recently been introduced in
the above relation in place of the residual lactose con
-
centration S
res
:
/67/
To avoid the use of two expressions for production
rate (Eqs. 61 and 63), depending on culture conditions,
both above expressions were merged, leading to a uni
-
que expression taking into account both effects, a nutri
-
tional limitation effect and an inhibitory effect:
/68/
Conclusion
Nutritional limitations (carbon and nitrogen) and pro
-
duct inhibition are mainly considered to account for cessa
-
tion of growth. However, analysis of lactic acid bacteria
culture shows that pH is the main factor to be consi
-
dered for model development. The undissociated form
of lactic acid is the main inhibitor, whose concentration
increased at acidic pH, and thus in addition to the pH
effect caused the cessation of growth. Some authors take
into account the inhibitory effect of the undissociated
lactic acid in their growth model, and hence also involve
the pH through the Henderson–Hasselbach equation. To
overcome inhibitory effects, pH is usually controlled at
its optimal value (close to 6). Under these conditions and
the usual culture conditions, the lack of lactic acid in
-
hibition, as well as carbon substrate inhibition, is clearly
demonstrated. Therefore, nutritional limitations cause the
cessation of growth. Complex substrates containing pep
-
tidic nitrogen and growth factors are generally added to
culture media, owing to the fastidious nutritional require
-
ments of lactic acid bacteria, and hence nitrogen limi
-
tation is usually observed instead of carbon limitation of
growth. However, there is a lack of models involving nu
-
tritional limitations in 'a direct way' in the available lite
-
rature, owing to the difficulty to characterize the 'usable
nitrogen'. However, some models are available, involv-
ing nitrogen limitations in 'a direct way' or indirectly
through the Verlhust expression, for instance.
The Luedeking and Piret model, involving a partial
link between growth and production, is the most widely
used to describe production kinetics. As for the growth,
pH is the main parameter to be considered for model
development. At acidic pH, cessation of production re-
sulted from both inhibitory effects of the undissociated
part of lactic acid and pH, and has been considered by
some authors. On the other hand, the control of pH at
its optimal culture value leads to cessation of produc
-
tion due to carbon exhaustion from the medium, since
LAB are unable to use the carbon components released
by autolysis of dead cells; the Monod model is mainly
considered to account for this behaviour.
References
1. C.J. Gadjil, K.V. Venkatesh, Structured model for batch cul
-
ture growth of Lactobacillus bulgaricus, J. Chem. Tech. Bio
-
technol. 68 (1997) 89–93.
2. J. Nielsen, K. Nikolajsen, J. Villadsen, Structured model
-
ling of a microbial system: I. A theoretical study of lactic
acid fermentation, Biotechnol. Bioeng. 38 (1991) 1–10.
3. R. Luedeking, E.L. Piret, Transient and steady states in con
-
tinuous fermentation: Theory and experiments, J. Biochem.
Microbiol. Technol. Eng. 1 (1959) 431–459.
4. P.L. Rogers, L. Bramall, I.J. McDonald, Kinetic analysis of
batch and continuous culture of Streptococcus cremoris HP
1
,
Can. J. Microbiol. 24 (1978) 372–380.
5. M.B. Leh, M. Charles, The effect of whey protein hydroly
-
zate average molecular weight on the lactic acid fermen
-
tation, J. Ind. Microbiol. Biotechnol. 4 (1989) 77–80.
6. A. Amrane, Y. Prigent, Lactic acid production from lactose
in batch culture: Analysis of the data with the help of a
mathematical model; Relevance for nitrogen source and pre
-
9
A. BOUGUETTOUCHA et al.: Models for Lactic Acid Fermentation, Food Technol. Biotechnol. 49 (1) 3–12 (2011)
PPmax P
S
exp(- · )P=
+
S
qq k
Sk
)
(
2
Pmax b c
·KTK=-q
'
0
max
d
1
dt
æö
ç÷
=-
èø
PP
PP
P
()
p
'
max
·1
æö
ç÷
=+ -
ç÷
èø
m
P
qAB
P
[]
[]
p
inh
·1
HL
AB
HL
æö
ç÷
=+-
ç÷
èø
q
m
·
res
p
1
S
AB
S
æö
=+-
ç÷
èø
q
m
·
lim
p
1
S
AB
S
æö
=+-
ç÷
èø
q
m
[]
[]
lim
p
inh
11
SHL
SHL
æö
æö
ç÷
=+- -
ç÷
ç÷
èø
èø
q
m
culture assessment, Appl. Microbiol. Biotechnol. 40 (1994)
644–649.
7. S. Kumar Dutta, A. Mukherjee, P. Chakraborty, Effect of
product inhibition on lactic acid fermentation: Simulation
and modelling, Appl. Microbiol. Biotechnol. 46 (1996) 410–
413.
8. M. Boonmee, N. Leksawasdi, W. Bridge, P.L. Rogers, Batch
and continuous culture of Lactococus lactis NZ133: Experi
-
mental data and model development, Biochem. Eng. J. 14
(2003) 127–135.
9. V.M. Yabannavar, D.I.C. Wang, Analysis of mass transfer
for immobilized cells in an extractive lactic acid fermenta
-
tion, Biotechnol. Bioeng. 37 (1991) 544–550.
10. F.V. Passos, H.P. Fleming, D.F. Ollis, R.M. Felder, R.F. Mc
Feeters, Kinetics and modeling of lactic acid production by
Lactobacillus plantarum, Appl. Environ. Microbiol. 60 (1994)
2627–2636.
11. A.R. Berry, C.M.M. Franco, Wei Zhang, A.P.J. Mildelberg,
Growth and lactic acid production in batch culture of Lac
-
tobacillus rhamnosus in a defined medium, Biotechnol. Lett.
21 (1999) 163–167.
12. C. Ben Youssef, G. Goma, A. Olmos-Dichara, Kinetic mo
-
delling of Lactobacillus casei ssp. rhamnosus growth and lac
-
tic acid production in batch cultures under various medi
-
um conditions, Biotechnol. Lett. 27 (2005) 1785–1789.
13. A. Nancib, Lactic acid production by Lactobacillus casei
subsp. rhamnosus growing on date juice: Batch, fed-batch
and continuous culture kinetics and optimization, PhD The
-
sis, University of Nancy, Nancy, France (2007) (in French).
14. A.D. Nandasana, S. Kumar, Kinetic modeling of lactic acid
production from molasses using Enterococus faecalis RKY1,
Biochem. Eng. J. 38 (2008) 277–284.
15. J. Monod: Research of the Growth of Bacterial Cultures, Her-
mann, Paris, France (1942) (in French).
16. D. Altiok, F. Tokatli, S. Harsa, Kinetic modelling of lactic
acid production from whey by Lactobacillus casei (NRRL
B-441), J. Chem. Tech. Biotechnol. 81 (2006) 1190–1197.
17. S. Aiba, M. Shoda, M. Nagatani, Kinetics of product inhi-
bition in alcohol fermentation, Biotechnol. Bioeng. 10 (1969)
845–864.
18. M.A. Mehaia, M. Cheryan, Production of lactic acid from
sweet whey permeate concentrates, Process Biochem. 22 (1987)
185–188.
19. M. Hujanen, Y.Y. Linko, Optimization of
L(+)-lactic acid pro
-
duction employing statistical experimental design, Biotech
-
nol. Tech. 8 (1994) 325–330.
20. S. Kwon, I.K. Yoo, W.G. Lee, H.N. Chang, Y.K. Chang, High
-
-rate continuous production of lactic acid by Lactobacillus
rhamnosus in a two-stage membrane cell-recycle bioreactor,
Biotechnol. Bioeng. 73 (2001) 25–34.
21. L. Bâati, G. Roux, B. Dahhou, J.L. Uribelarrea, Unstruc
-
tured modelling growth of Lactobacillus acidophilus as a func
-
tion of the temperature, Math. Comput. Simulat. 65 (2004)
137–145.
22. J. Bajpai-Dikshit, A.K. Suresh, K.V. Venkatesh, An optimal
model for representating the kinetics of growth and pro
-
duct formation by Lactobacillus rhamnosus on multiple sub
-
strates, J. Biosci. Bioeng. 96 (2003) 481–486.
23. A. Amrane, Y. Prigent, Growth and lactic acid production
coupling for Lactobacillus helveticus cultivated on supple
-
mented whey: Influence of peptidic nitrogen deficiency, J.
Biotechnol. 55 (1997) 1–8.
24. A. Amrane, Y. Prigent, Influence of yeast extract concen
-
tration on batch cultures of Lactobacillus helveticus: Growth
and production coupling, World J. Microbiol. Biotechnol. 14
(1998) 529–534.
25. K.W. Turner, T.D. Thomas, Uncoupling of growth and acid
production in lactic streptococci, New Zeal. J. Dairy Sci. 10
(1975) 162–167.
26. F. Mozzi, G.S. de Giori, G. Oliver, G.F. de Valdez, Effect of
culture pH on the growth characteristics and polysaccha
-
ride production by Lactobacillus casei, Milchwissenschaft, 49
(1994) 667–670.
27. N.C. Major, A.T. Bull, Lactic acid productivity of a conti
-
nuous culture of Lactobacillus delbrueckii, Biotechnol. Lett. 7
(1985) 401–405.
28. A. Aeschlimann, U. von Stockar, The production of lactic
acid from whey permeate by Lactobacillus helveticus, Bio
-
technol. Lett. 11 (1989) 195–200.
29. D. Belhocine, Lactose valorisation by lactic acid fermen
-
tation, PhD Thesis, University of Rennes 1, Rennes, France
(1987) (in French).
30. G.E. Powell, Growth kinetics of non-competitive product
inhibition in batch cultures, J. Chem. Tech. Biotechnol. 34
(1984) 33–38.
31. H. Ohara, K. Hiyama, T. Yoshida, Non-competitive pro
-
duct inhibition in lactic acid fermentation from glucose,
Appl. Microbiol. Biotechnol. 36 (1992) 773–776.
32. R. Cachon, C. Diviès, Modeling of growth and lactate fer
-
mentation by Lactococcus lactis subsp. lactis biovar. diacety
-
lactis in batch culture, Appl. Microbiol. Biotechnol. 40 (1993)
28–33.
33. D. Pinelli, A.R. González-Vara, D. Matteuzzi, F. Magelli,
Assessment of kinetic models for the production of
L- and
D-lactic acid isomers by Lactobacillus casei DMS 20011 and
Lactobacillus coryniformis DMS 20004 in continuous fermen
-
tation, J. Ferment. Bioeng. 83 (1997) 209–212.
34. J. Tayeb, C. Bouillanne, J.M. Desmazeaud, Computerized
control of growth with temperature in a mixed culture of
lactic acid bacteria, J. Ferment. Technol. 62 (1984) 461–470.
35. A. Ajbar, A.H. Fakeeha, Static and dynamic behaviour of a
class of unstructured models of continuous bioreactors with
growth associated product, Bioprocess Biosyst. Eng. 25 (2002)
21–27.
36. D.G. Kendall, Stochastic processes and population growth,
J. Roy. Stat. Soc. B, 11 (1949) 230–282.
37. N.D. Ierusalimsky: Bottlenecks in Metabolism as Growth
Rate Controlling Factors. In: Microbial Physiology and Conti
-
nuous Culture, E.O. Pawell, C.G.T. Evans, R.E. Strange, D.W.
Tempest (Eds.), HMSO, London, UK (1967).
38. V.H. Edwards, The influence of high substrate concentra
-
tions on microbial kinetics, Biotechnol. Bioeng. 12 (1970) 679–
712.
39. A.K. Keller, P. Gerhardt, Continuous lactic acid fermenta
-
tion of whey to produce a ruminant feed supplement high
in crude protein, Biotechnol. Bioeng. 17 (1975) 997–1018.
40. J.M. Monteagudo, L. Rodriguez, J. Rincón, J. Fuertes, Ki
-
netics of lactic acid fermentation by Lactobacillus delbrueckii
on beet molasses, J. Chem. Tech. Biotechnol. 68 (1997) 271–
276.
41. C.N. Burgos-Rubio, M.R. Okos, P.C. Wankat, Kinetic study
of the conversion of different substrates to lactic acid using
Lactobacillus bulgaricus, Biotechnol. Progr. 16 (2000) 305–314.
42. P. Mercier, L. Yerushalmi, D. Rouleau, D. Dochain, Kine
-
tics of lactic acid fermentation on glucose and corn by Lac
-
tobacillus amylophilus, J. Chem. Tech. Biotechnol. 55 (1992) 111–
121.
43. A.B. Moldes, J.L. Alonso, J.C. Parajó, Cogeneration of cello
-
biose and glucose from pretreated wood and bioconver
-
sion to lactic acid: A kinetic study, J. Biosci. Bioeng. 87 (1999)
787–792.
44. L.M.D. Gonçalves, A.M.R.B. Xavier, J.S. Almeida, M.J.T. Car
-
rondo, Concomitant substrate and product inhibition ki
-
netics in lactic acid production, Enzyme Microb. Technol. 13
(1991) 314–319.
45. C. Åkerberg, K. Hofvendahl, G. Zacchi, B. Hahn-Hägerdal,
Modelling the influence of pH, temperature, glucose and
10
A. BOUGUETTOUCHA et al.: Models for Lactic Acid Fermentation, Food Technol. Biotechnol. 49 (1) 3–12 (2011)
lactic acid concentrations on the kinetics of lactic acid pro
-
duction by Lactococcus lactis ssp. lactis ATCC 19435 in whole
-
-wheat flour, Appl. Microbiol. Biotechnol. 49 (1998) 682–690.
46. J.F. Andrews, A mathematical model for continuous cul
-
ture of microorganisms utilizing inhibitory substrates, Bio
-
technol. Bioeng. 10 (1968) 707–723.
47. A. Ishizaki, T. Ohta, Batch culture kinetics of
L-lactate fer
-
mentation employing Streptococcus IO-1, J. Ferment. Bioeng.
67 (1989) 46–51.
48. J. Biazar, M. Tango, E. Babolian, R. Islam, Solution of the
kinetic modelling of lactic acid fermentation using Ado
-
mian decomposition method, Appl. Math. Comput. 144 (2003)
433–439.
49. G. Adomian: Stochastic Systems, Academic Press, New York,
NY, USA (1983).
50. G. Adomian: Nonlinear Stochastic Operator Equations, Aca
-
demic Press, New York, NY, USA (1986).
51. L. Peeva, G. Peev, A new method for pH stabilization of
the lactoacidic fermentation, Enzyme Microb. Technol. 21 (1997)
176–181.
52. D. Roy, A. Leduy, J. Goulet, Kinetics of growth and lactic
acid production from whey permeate by Lactobacillus hel
-
veticus, Can. J. Chem. Eng. 65 (1987) 597–603.
53. S. Norton, C. Lacroix, J.C. Vuillemard, Kinetic study of con
-
tinuous whey permeate fermentation by immobilised Lac
-
tobacillus helveticus for lactic acid production, Enzyme Mi
-
crob. Technol. 6 (1994) 457–466.
54. G. Gätje, G. Gottschalk, Limitation of growth and lactic acid
production in batch and continuous cultures of Lactobacil
-
lus helveticus, Appl. Microbiol. Biotechnol. 34 (1991) 446–449.
55. T.P. Hanson, G.T. Tsao, Kinetic studies of the lactic acid
fermentation in batch and continuous cultures, Biotech. Bio-
eng. 14 (1970) 233–252.
56. K.V. Venkatesh, M.R. Okos, P.C. Wankat, Kinetic model of
growth and lactic acid production from lactose by Lacto-
bacillus bulgaricus, Process Biochem. 28 (1993) 231–241.
57. A. Amrane, Y. Prigent, Influence of an initial addition of
lactic acid on growth, acid production and their coupling
for batch cultures of Lactobacillus helveticus, Bioprocess Bio-
syst. Eng. 19 (1998) 307–312.
58. L.C. McDonald, H.P. Fleming, H.M. Hassan, Acid tolerance
of Leuconostoc mesenteroides and Lactobacillus plantarum, Appl.
Environ. Microbiol. 56 (1990) 2120–2124.
59. G.R. Bender, R.E. Marquis, Membrane ATPases and acid
tolerance of Actinomyces viscos and Lactobacillus casei, Appl.
Environ. Microbiol. 53 (1987) 2124–2128.
60. E.R. Kashket, Bioenergetics of lactic acid bacteria: Cytoplas
-
mic pH and osmotolerance, FEMS Microbiol. Rev. 46 (1987)
233–244.
61. A. Bouguettoucha, B. Balannec, S. Nacef, A. Amrane, A
generalised unstructured model for batch cultures of Lac
-
tobacillus helveticus, Enzyme Microb. Technol. 41 (2007) 377–
382.
62. A. Bouguettoucha, B. Balannec, A. Amrane, Unstructured
generalized models for the analysis of the inhibitory and
the nutritional limitation effects on Lactobacillus helveticus
growth Models validation, Biochem. Eng. J. 39 (2008) 566–
574.
63. P.L.H. Yeh, R.K. Bajpai, E.L. Iannotti, An improved kinetic
model for lactic acid fermentation, J. Ferment. Bioeng. 71
(1991) 75–77.
64. H. Wang, M. Seki, S. Furusaki, Mathematical model for ana
-
lysis of mass transfer for immobilized cells in lactic acid
fermentation, Biotechnol. Progr. 11 (1995) 558–564.
65. F. Leroy, L. De Vuyst, Growth of the bacteriocin-producing
Lactobacillus sakei strain CTC 494 in MRS broth is strongly
reduced due to nutrient exhaustion: A nutrient depletion
model for the growth of lactic acid bacteria, Appl. Environ.
Microbiol. 67 (2001) 4407–4413.
66. A. Amrane, C. Couriol, Unstructured model for seed cul
-
tures without pH control of Lactobacillus helveticus growing
on supplemented whey permeate, J. Chem. Tech. Biotechnol.
77 (2002) 950–957.
67. K.M. Vereecken, J.F. Van Impe, Analysis and practical im
-
plementation of a model for combined growth and meta
-
bolite production of lactic acid bacteria, Int. J. Food Micro
-
biol. 73 (2002) 239–250.
68. F.V. Passos, H.P. Fleming, H.M. Hassan, R.F. McFeeters,
Effect of malic acid on the growth kinetics of Lactobacillus
plantarum, Appl. Microbiol. Biotechnol. 63 (2003) 207–211.
69. A. Amrane, Y. Prigent, Differentiation of pH and free lactic
acid effects on the various growth and production phases
of Lactobacillus helveticus, J. Chem. Tech. Biotechnol. 74 (1999)
33–40.
70. W. Fu, A.P. Mathews, Lactic acid production from lactose
by Lactobacillus plantarum: Kinetic model and effects of pH,
substrate and oxygen, Biochem. Eng. J. 3 (1999) 163–170.
71. C. Cornelius, T. Erpicum, P. Jacques, P. Thonart, Compa
-
rison of fermentation industrial components such as corn
steep and yeast extract for lactic acid bacteria production,
Med. Fac. Landbouw. Univ. Gent, 61 (1996) 1461–1463.
72. M. Hujanen, Y.Y. Linko, Effect of temperature and various
nitrogen sources on
L(+)-lactic acid production by Lacto
-
bacillus casei, Appl. Microbiol. Biotechnol. 45 (1996) 307–313.
73. I. Mierau, E.R.S. Kunji, G. Venema, J. Kok, Casein and pep
-
tide degradation in lactic acid bacteria, Biotech. Gen. Eng.
Rev. 14 (1997) 279–301.
74. S. Tejayadi, M. Cheryan, Lactic acid from cheese whey per-
meate. Productivity and economics of a continuous mem-
brane bioreactor, Appl. Microbiol. Biotechnol. 43 (1995) 242–
248.
75. I.K. Yoo, H.N. Chang, E.G. Lee, Y.K. Chang, S.H. Moon,
Effect of B vitamin supplementation on lactic acid produc-
tion by Lactobacillus casei, J. Ferment. Bioeng. 84 (1997) 172–
175.
76. A. Amrane, Y. Prigent, Mathematical model for lactic acid
production from lactose in batch culture: Model develop-
ment and simulation, J. Chem. Tech. Biotechnol. 60 (1994) 241–
246.
77. R.A. Moraine, P. Rogovin, Kinetics of polysaccharide B-1459
fermentation, Biotechnol. Bioeng. 8 (1996) 511–524.
78. C. Diaz, P. Lelong, P. Dieu, C. Feuillerat, M. Salome, On
-
-line analysis and modelling of microbial growth using a
hybrid system approach, Process Biochem. 34 (1999) 39–47.
79. A. Pandey, C.R. Soccol, D. Mitchell, New developments in
solid state fermentation: I. Bioprocesses and products, Pro
-
cess Biochem. 35 (2000) 1153–1169.
80. M. Özilgen: Kinetics of Food Processes Involving Pure or
Mixed Cultures of Lactic Acid Bacteria. In: Lactic Acid Bac
-
teria: Current Advances in Metabolism, Genetics and Applica
-
tions, NAT O ASI Series H, Vol. 98, T.F. Bozogh lu, B. Ray (Eds.),
Springer-Verlag, Berlin, Germany (1996).
81. A. Amrane, Analysis of the kinetics of growth and lactic
acid production for Lactobacillus helveticus growing on sup
-
plemented whey permeate, J. Chem. Tech. Biotechnol. 80
(2005) 345–352.
82. B. Balannec, A. Bouguettoucha, A. Amrane, Unstructured
model for batch cultures without pH control of Lactobacil
-
lus helveticus Inhibitory effect of the undissociated lactic
acid, Biochem. Eng. J. 35 (2007) 289–294.
83. J.A. Vázquez, M.A. Murado, Mathematical tools for objec
-
tive comparison of microbial culture: Application to evalu
-
ation of 15 peptones for lactic acid bacteria productions,
Biochem. Eng. J. 39 (2008) 276–287.
84. J.A. Vázquez, M.A. Murado, Unstructured mathematical mo
-
del for biomass, lactic acid and bacteriocin production by
lactic acid bacteria in batch fermentation, J. Chem. Tech.
Biotechnol. 83 (2008) 91–96.
11
A. BOUGUETTOUCHA et al.: Models for Lactic Acid Fermentation, Food Technol. Biotechnol. 49 (1) 3–12 (2011)
85. C.Q. Lan, G. Oddone, D.A. Mills, D.E. Block, Kinetics of
Lactobacillus lactis growth and methabolite formation under
aerobic and anaeorobic conditions in the presence or ab
-
sence of hemin, Biotechnol. Bioeng. 95 (2006) 1070–1080.
86. W. Messens, J. Verluyten, F. Leroy, L. De Vuyst, Modelling
growth and bacteriocin production by Lactobacillus curva
-
tus LTH 1174 in response to temperature and pH values
used for European sausage fermentation processes, Int. J.
Food Microbiol. 81 (2003) 41–52.
87. M. Desmazeaud, Lactic acid bacteria nutrition: State of the
art, Lait, 63 (1983) 267–316.
88. G.G. Pritchard, T. Coolbear, The physiology and biochem
-
istry of the proteolytic system in lactic acid bacteria, FEMS
Microbiol. Rev. 12 (1993) 179–206.
89. S. Torriani, M. Vescovo, G. Scolari, An overview on Lacto
-
bacillus helveticus, Ann. Microbiol. Enzymol. 44 (1994) 163–
191.
90. M.B. Leh, The effect of whey protein hydrolyzates on the
lactic acid fermentation, PhD Thesis, University of Lehigh,
Bethlehem, PA, USA (1987).
91. A. Amrane, Y. Prigent, Analysis of growth and production
coupling for batch cultures of Lactobacillus helveticus with
the help of an unstructured model, Process Biochem. 34 (1999)
1–10.
92. A.W. Schepers, J. Thibault, C. Lacroix, Lactobacillus helveti
-
cus growth and lactic acid production during pH-control-
led batch cultures in whey permeate/yeast extract medi
-
um. Part I: Multiple factor kinetic analysis, Enzyme Microb.
Technol. 30 (2002) 176–186.
93. A.W. Schepers, J. Thibult, C. Lacroix, Lactobacillus helveti
-
cus growth and lactic acid production during pH-control
-
led batch cultures in whey permeate/yeast extract medi
-
um. Part II: Kinetic modeling and model validation, Enzyme
Microb. Technol. 30 (2002) 187–194.
94. R. Luedeking, E.L. Piret, A kinetic study of the lactic acid
fermentation. Batch process at controlled pH, J. Biochem.
Microbiol. Technol. Eng. 1 (1959) 393–412.
95. B. Bibal, C. Kapp, G. Goma, A. Pareilleux, Continuous cul
-
ture of Streptococcus cremoris on lactose using various me
-
dium conditions, Appl. Microbiol. Biotechnol. 32 (1989) 155–159.
96. K. Ye, S. Jin, K. Shimizu, Performance improvement of lac
-
tic acid fermentation by multistage extractive fermenta
-
tion, J. Ferment. Bioeng. 81 (1996) 240–246.
97. M.Y. Ha, S.W. Kim, Y.W. Lee, M.J. Kim, S.J. Kim, Kinetics
analysis of growth and lactic acid production in pH-con
-
trolled batch cultures of Lactobacillus casei KH-1 using yeast
extract/corn steep liquor/glucose medium, J. Biosci. Bio
-
eng. 96 (2003) 134–140.
98. R.Y. Peng, T.C.K. Yang, H.E. Wang, Y.C. Lin, C. Cheng, Mo
-
delling of lactic acid fermentation An improvement of
Luedeking's model, J. Chin. Agr. Chem. Soc. 35 (1997) 485–
494.
99. M.H. Jørgensen, K. Nikolajsen, Mathematic model for lac
-
tic acid formation with Streptococcus cremoris from glucose,
Appl. Microbiol. Biotechnol. 25 (1987) 313–316.
100. A. Amrane, Batch cultures of supplemented whey perme
-
ate using Lactobacillus helveticus: Unstructured model for
biomass formation, substrate consumption and lactic acid
production, Enzyme Microb. Technol. 28 (2001) 827–834.
12
A. BOUGUETTOUCHA et al.: Models for Lactic Acid Fermentation, Food Technol. Biotechnol. 49 (1) 3–12 (2011)
... The dishes were inoculated with agar disks (d = 10 mm) of a fully grown culture, incubated at 25 • C, and the diameter of the growing colonies was measured daily. The resulting data were used for the modeling of the growth kinetics by applying the logistic curve model (Equation (1)) and the reverse autocatalytic growth model (Equation (2)) [26,27]. ...
Article
Full-text available
In the present study, an unknown fungal strain was isolated from the fruiting body of a local Dryad’s Saddle mushroom (Polyporaceae). The molecular identification of the isolate was performed by amplification of the ITS1-5.8S-ITS2 region and the strain was identified with 100.00% confidence as Epicoccum nigrum. The morphological characteristics, including the distinctive colony pigmentation, conidiophore structure, and conidial shape, were determined to ensure comprehensive characterization of the fungus. The modeling of the kinetics of the growth process was conducted with the applying the logistic curve model and the reverse autocatalytic growth model, and the concentrations of the compounds in the nutrient medium required for the E. nigrum development were established. Controlled submerged cultivation was carried out for cultural liquid obtaining, which was further used for the evaluation of the biological activities. The untreated cultural liquid demonstrated antimicrobial activity against Sclerotinia sclerotiorum where the minimal inhibitory concentration was 1.25 mg/mL. Antimicrobial activity was also detected toward Botrytis cinerea (2.5 mg/mL) and Aspergillus flavus (2.5 mg/mL). The direct utilization of crude cultural liquid for phytopathogenic control is a sustainable approach that will provide the opportunity for the development of an environmentally friendly manufacturing process.
... Model describes biomass growth through Monod kinetics, whereas the production of other products is assumed to be dependent upon the concentration of the limiting component and the concentration of biomass as of a second order reaction. However, in an actual scenario, the product formation could be growth linked or non-growth linked or combination of both, which could also depend upon the type of the reactant molecules [48,49]. ...
Article
Full-text available
A variety of mathematical models have been developed to simulate the biochemical and physico-chemical aspects of the anaerobic digestion (AD) process to treat organic wastes and generate biogas. However, all these models, including the most widely accepted and implemented Anaerobic Digestion Model No.1, remain incapable of adequately representing the material balance of AD and are therefore inherently incapable of material conservation. The absence of robust mass conservation constrains reliable estimates of any kinetic parameters being estimated by regression of empirical data. To address this issue, the present work involved the development of a “framework” for a mass-conserving atomistic mathematical model which is capable of mass conservation, with a relative error in the range of machine precision value and an atom balance with a relative error of ±0.02% whilst obeying the Henry’s law and electroneutrality principle. Implementing the model in an Excel spreadsheet, the study calibrated the model using the empirical data derived from batch studies. Although the model shows high fidelity as assessed via inspection, considering several constraints including the drawbacks of the model and implementation platform, the study also provides a non-exhaustive list of limitations and further scope for development.
... Different models have been used in the modeling of lactic acid production from various substrates. Some of them are summarized in the works of Zacharof & Lovitt [13], Bouguettoucha et al. [14], and Gordeev et al. [15]. ...
Preprint
Full-text available
Lactic acid production from lactose by Lactiplantibacillus plantarum AC 11S was studied. The influence of initial substrate concentration was investigated, and optimal pH and temperature were determined. An unstructured mathematical model was developed comprising equations for bacterial growth, substrate consumption, and product formation. The model was solved with different terms for specific growth rates considering substrate and/or product inhibition. The best bacterial growth and lactic acid production were achieved at pH = 6.5 and 30 oC. Production of lactic acid was mainly growth-associated and at substrate concentration over 15 g/L a considerable product inhibition was observed. The parameters of different models were determined and compared. While the modified Gompertz equation gave the best fit for biomass growth, a variant of the logistic equation including a term for product inhibition described in the best way the entire set of equations comprising bacterial growth, substrate consumption, and product formation at different initial substrate concentrations.
... KS represents substrate concentration at which the growth rate is equal to half of the maximum growth rate. The Monod equation describes the increase in cell concentration with time and considers limiting substrate concentration [13][14][15]. It is widely accepted that lactic acid fermentation is limited by the substrate and inhibited by the product. ...
Article
Full-text available
Lactic acid is a versatile chemical with wide application in many industries. It can be produced by the fermentation of different sugars by various lactobacilli and investigations on lactic acid production from different substrates and by different strains are still in progress. The present study aimed to study lactic acid production from lactose by Lactiplantibacillus plantarum AC 11S and to choose a mathematical model describing in the best way the experimental data obtained. The influence of initial substrate concentration was investigated, and optimal pH and temperature were determined. An unstructured mathematical model was developed comprising equations for bacterial growth, substrate consumption, and product formation. The model was solved with different terms for specific growth rates considering substrate and/or product inhibition. The best bacterial growth and lactic acid production were achieved at pH = 6.5 and 30 °C. Production of lactic acid was mainly growth-associated, and at initial substrate concentration over 15 g/L, a considerable product inhibition was observed. The parameters of different models were determined and compared. The modified Gompertz equation gave the best fit when solving only the equation for biomass growth at different initial substrate concentrations. Solving the entire set of differential equations for bacterial growth, substrate consumption, and product formation, the best results were obtained when using a variant of the logistic equation for biomass growth. This variant included a term for product inhibition and described in the best way all experimental data. Solving the model for different biomass concentrations showed that an increase in biomass led to a shorter lag phase and the stationary phase was reached faster. The results obtained, optimum conditions and the kinetic model, are good bases for studying pH-controlled fermentation, as well as a continuous process.
... The most appropriate models describing the dependence of the growth and the biosynthesis of metabolic products by substrate concentration are those of Moser and Andrews. The major disadvantage of the mentioned above models is the absence of a clear biological sense and the impossibility of describing the process kinetics caused by the constantly increasing biomass concentration and the effect of the limited volume of the medium on the growth and metabolite synthesis (Bouguettoucha et al., 2011;Feng et al., 2016). Also, they don't provide information about at what development stage the metabolites are produced, which is important when the process will be up-scaled. ...
Article
Full-text available
The aim of this study was the evaluation of the basic bioprocess parameters by applying mathematical modelling of the growth kinetics of Ganoderma resinaceum GA1M isolated from Bulgaria as well as assessment of the antioxidant and antimicrobial activity of the obtained biomass. The models of Monod, Moser, Andrews and Verhulst in combination with the Luedeking–Piret were applied for kinetics modelling of the cultivation process. The in vitro antioxidant activity was evaluated by applying four different methods. The disk diffusion method and broth microdilution methods were employed to determine the antimicrobial activity of the mycelium biomass extracts. Mathematical modeling of the kinetics of the cultivation process was performed and basic bioprocess parameters were determined. The obtained experimental data were in agreement with the applied mathematical models. The water extract of the mycelium biomass showed the highest antioxidant potential and the highest total phenolic content. The methylene chloride, methanol and ethanol extracts of the biomass demonstrated antimicrobial activity.
... (2)) to include the effects of inhibitory acids and mass transfer through immobilization matrix. Acid inhibition can be well represented by a non-competitive inhibition function (Bouguettoucha et al., 2011;Cachon and Diviès, 1993;Powell, 1984). An effectiveness factor (η) represents a ratio between the rates of the immobilized and free cell systems (Calvelo and Cunningham, 1970). ...
... The rate of biomass formation can be explained by Eq. (1) where X is the biomass concentration (g L −1 ), is the specific growth rate (h −1 ) and r x is the volumetric biomass productivity rate (g L −1 h −1 ). Due to the precise predictability of unstructured and non-segregated models reported for the metabolism of LAB [20][21][22], the Modified Logistic model (MLM), Gompertz model (GM), Richards model (RM), and Schnute model (SM) were used to describe the kinetics of P. pentosaceus as represented in Table 1. ...
Article
Full-text available
Pediocin (PE) is an antimicrobial peptide possessing a plethora of applications in food preservation. In this study, an economical and sustainable method for the production of PE by Pediococcus pentosaceus CRA51 was achieved by utilizing low-cost renewable feedstocks and industrial by-products viz., Palmyra palm jaggery (PJ), Cassava fibrous waste hydrolysate (CFWH) and Whey protein hydrolysate (WPH)) as substrates. WPH served as an elite nitrogen source yielding a 1.75-fold increment in the PE titer than the conventional glucose-based medium. In addition, the kinetics of growth, concomitant production kinetics of lactic acid and PE by P. pentosaceus CRA51 were investigated by employing different substrate-independent mathematical models such as Modified Logistic, Gompertz, Richards, and Schnute. Based on the overall model fitting parameters and statistical significance obtained on various carbon and nitrogen substrates, Richards model exhibited the best fit with the predictions close to the experimental data. The insights derived from kinetic studies and medium-based techno-economic analysis demonstrates that the utilization of PJ and WPH as alternate carbon and nitrogen source would be an economically viable strategy for the enhanced production of PE. It is envisaged that the maximum specific growth rate (μm)μm){\mu }_{m}) and biomass yield (YX/S)YX/S){Y}_{X/S}) obtained on PJ (0.346 h⁻¹, 0.257 g g⁻¹) and WPH-based media (0.204 h⁻¹, 0.141 g g⁻¹) can be leveraged for the optimization of high cell density fermentation, and further for the large-scale economic production of PE by P. pentosaceus CRA51. Graphical Abstract
... Two most commonly used modeling methods of cellular metabolism are differential equation-based models [22,23] and flux balance analysis (FBA)-based models [24]. Admittedly, differential equation-based models can be easily connected with process control, but most of them treat the whole metabolic network as a "black box" [25,26]. On the other hand, fine-grained "white-box" differential equation-based models [23] are typically costly to construct due to limited data availability [19,27] and complex enzyme kinetic mechanisms [28]. ...
Article
Full-text available
In microorganisms, different from primary metabolism for cellular growth, secondary metabolism is for ecological interactions and stress responses and an important source of natural products widely used in various areas such as pharmaceutics and food additives. With advancements of sequencing technologies and bioinformatics tools, a large number of biosynthetic gene clusters of secondary metabolites have been discovered from microbial genomes. However, due to challenges from the difficulty of genome-scale pathway reconstruction and the limitation of conventional flux balance analysis (FBA) on secondary metabolism, the quantitative modeling of secondary metabolism is poorly established, in contrast to that of primary metabolism. This review first discusses current efforts on the reconstruction of secondary metabolic pathways in genome-scale metabolic models (GSMMs), as well as related FBA-based modeling techniques. Additionally, potential extensions of FBA are suggested to improve the prediction accuracy of secondary metabolite production. As this review posits, biosynthetic pathway reconstruction for various secondary metabolites will become automated and a modeling framework capturing secondary metabolism onset will enhance the predictive power. Expectedly, an improved FBA-based modeling workflow will facilitate quantitative study of secondary metabolism and in silico design of engineering strategies for natural product production.
Article
Sweet whey permeate powder (WPP, 85% lactose) was reconstituted, supplemented with yeast extract and used for the production of lactic acid by Lactobacillus bulgaricus. With the pH maintained at 5. 6 and the temperature at 45 degree C, complete lactose utilization and good yields of lactic acid were obtained at 5, 10, and 15% WPP in a batch system. Higher substrate concentrations resulted in incomplete lactose utilization and growth inhibition. Under comparable conditions in the CSTR membrane recycle bioreactor (complete utilization of 134 g 1** minus **1 lactose and product concentration of 110-117 g 1** minus **1 lactate), the optimum productivity was 27-84 g 1** minus **1h** minus **1 at cell concentration of 30-63 g 1** minus **1. This is 6-18 times better than the batch reactor.
Article
Batch and continuous fermentation kinetics of Lactobacillus bulgaricas were examined in detail. The batch kinetic model includes the effect of pH, lactate and substrate. The effect of pH on the enzyme β-galactosidase is reflected in the value of μmax, the Monod's growth constant. Empirical expressions to estimate model parameters based on pH were established. These were further used to estimate the cell mass and lactic acid concentration for the fermentation proceeded with no pH control. At high pH lactic acid was mostly formed by growth associated mechanisms and at low pH it was due to the cell maintenance. The relationship between ATP concentration and cell mass in a batch fermentation could be approximated by a Leudeking—Piret like relationship. The continuous fermentation was studied with varying pH and initial substrate conditions. A maximum productivity of 6·2 g/litre per hour was obtained at pH 5·6 for the CSTR system, which was three times the productivity obtained for a batch system at the same pH. The CSTR system had a high productivity over a very small range of retention times. Three stages were found to be sufficient to utilize completely the lactose, but at the cost of decreasing the productivity to 4·8 g/litre per hour.
Article
For Lactobacillus helveticus growing on supplemented whey permeate, some simple criteria have been proposed for assessing carbon and nitrogen source limitations:•irrespective of supplementation, carbon limitation coincided with the end of lactic acid production.•carbon limitation in absence of nitrogen deficiency led to an abrupt cessation of growth.•nitrogen deficiency in the absence of carbon limitation resulted in an increase of the overall product on biomass yield YP/X during the second half of the cultivation.At pH near to the optimum (pH=6), even at the end of reaction, the final concentration of free lactic acid was below the inhibitory threshold (0.45–0.60 g l−1). In these conditions the shift from total to partial growth-associated production was tightly controlled by the concentration of available peptidic nitrogen.
Article
A kinetic model was developed that describes growth and lactic acid production rates of Lactobacillus helveticus during pH-controlled batch cultures in whey permeate-yeast extract medium as a function of four variables: sugar, nitrogen substrate and lactic acid concentrations, and pH. Since available nitrogen substrate could not be measured directly, it was related to medium composition through a response surface model of maximum biomass concentration in nitrogen-limited growth cultures. The effect of pH on growth was modeled with a Gaussian equation instead of the traditional second order polynomial term. Product inhibition was modeled with a logistic term for lactate and an exponential term for undissociated lactic acid. The kinetic model accurately described the experimental data with few exceptions.
Article
An on-line estimator based on the biomass concentration was developed for detecting and quantifying the growth phases encountered in batch cultures. The bioprocess is considered to be a hybrid system. Its evolution is characterized by different successive phases, each of them being described by a specific model of the growth dynamics. For each new biomass concentration value, the estimator selects the most appropriate model in terms of growth simulation from time of inoculation and estimates the corresponding kinetics parameters.This estimator generates a smooth growth rate trend, thus facilitating its on-line monitoring. The detection of the transitions between phases favours the discovery of possible factors affecting the growth. Finally, the duration of the respective phases, along with the kinetic parameters of the model selected at the last measurement summarize microbial growth and allow comparison between different batches experiments.A culture of recombinant Escherichia coli in a 2.5 l bioreactor, with its biomass concentration assessed every 20 min. is described as an example. The transitions between growth phases are well detected and the simulations generated by the successive estimated models accurately reflect the experimental data.
Article
Growth and production coupling for L. helveticus growing at constant pH on a medium largely supplemented with yeast extract and peptons has been examined as a function of initial lactic acid addition: these conditions allowed to avoid side effects of cultivation pH and nitrogen source deficiency on coupling. Growth of L. helveticus strain milano was reduced by half for a free lactic acid concentration of 0.34 g.l-1; both growth and production were fully inhibited for a critical end-product concentration of 83 g.l-1, despite the fact that lactose was still available. The extent of autolysis 8 h after growth ceased was practically independent of initial lactic acid concentration: the same conclusion held for the part played by a growth-associated mechanism in total acid production. During the slowing down phase of growth, production was non-growth-associated: production rate remained close to its maximal value, while growth rate declined.