Journal of Applied Microbiology ISSN 1364-5072

ORIGINAL ARTICLE

Mn2+ and Mg2+ synergistically enhanced lactic acid

production by Lactobacillus rhamnosus FTDC 8313 via
affecting different stages of the hexose monophosphate
pathway
L.-C. Lew, S.-B. Choi, P.-L. Tan and M.-T. Liong
School of Industrial Technology, Universiti Sains Malaysia, Penang, Malaysia

Keywords
biotechnology, Lactobacillus, modelling,
probiotics.
Correspondence
Min-Tze Liong, School of Industrial Technology, Universiti Sains Malaysia, 11800 USM,
Penang, Malaysia.
E-mail: mintze.liong@usm.my
2013/1207: received 17 June 2013, revised
10 October 2013 and accepted 15 November
2013
doi:10.1111/jam.12399

Abstract
Aims: The study aimed to evaluate the effects of Mn2+ and Mg2+ on lactic
acid production using response surface methodology and to further study their
effects on interactions between the enzymes and substrates along the hexose
monophosphate pathway using a molecular modelling approach.
Methods and Results: A rotatable central composite design matrix for lactic
acid production was generated with two independent factors namely,
manganese sulfate and magnesium sulfate. The second-order regression model
indicated that the quadratic model was significant (P < 0
05), suggesting that
the model accurately represented the data in the experimental region. Threedimensional response surface showed that lactic acid production was high
along the region where the ratio of MnSO4 to MgSO4 was almost 1 : 1,
justifying the need for both Mg2+ and Mn2+ to be present simultaneously in
stimulating the production of lactic acid. Molecular docking simulation was
performed on a total of 13 essential enzymes involved in the hexose
monophosphate pathway for the production of lactic acid with four different
conditions namely in the presence of Mg2+, Mn2+, both Mg2+ and Mn2+ and
in the absence of metal ions. Results showed that the presence of both Mg2+
and Mn2+ within the binding site improved the binding affinity for substrates
in five enzymes namely, glucose-6-phosphate dehydrogenase, phosphogluconate
dehydrogenase, glyceraldehyde-3-phosphate dehydrogenase, phosphopyruvate
hydratase and pyruvate kinase.
Conclusions: Using response surface methodology and molecular modelling
approach, we illustrated that Mg2+ and Mn2+ synergistically enhanced lactic
acid production by Lactobacillus rhamnosus FTDC 8313 via affecting different
stages of the hexose monophosphate pathway.
Significance and Impacts of the Study: Mg2+ and Mn2+ synergistically
improved lactic acid production of Lact. rhamnosus via improved binding
affinity of the enzymesubstrate along the hexose monophosphate pathway,
instead of purely affecting growth as previously understood.

Introduction
Lactic acid bacteria (LAB) are characterized as Gram-positive, nonspore forming, facultative anaerobic bacteria that
produce lactic acid as the major fermentation product either
homo- or heterofermentatively. The homofermentative
644

pathway results in the transformation of glucose to pyruvate

through the EmbdenMeyerhofParnas pathway, yielding
2 mol of lactic acid from 1 mol of glucose. Meanwhile, in
heterofermentation, 1 mol of glucose produces 1 mol each
of lactate, carbon dioxide and either acetic or ethanol, via
the hexose monophosphate pathway. The ability of LAB to

Journal of Applied Microbiology 116, 644--653 2013 The Society for Applied Microbiology

L.-C. Lew et al.

produce lactic acid as the main product from a range of carbon sources has been a key contribution to industries typically as preservative, acidulant and flavouring for food
products. Lactic acid has also been widely used for many
years in textile and pharmaceutical industries, as well in cosmetic and skin care products. In addition, the demand for
lactic acid has also increased considerably over the years
attributed to its role as monomer in the production of biodegradable plastics (Wee et al. 2006). According to a recent
forecast by Global Industry Analysts Inc. (GIA), the world
market for lactic acid will reach 367
3 thousand metric tons
by 2017. Although lactic acid can be produced via chemical
synthesis, microbial fermentation is commonly preferred
attributed to a higher optical purity, a characteristic important for biodegradable polymers. Chemical synthesis from
petrochemical resources often produced racemic of DL-lactic acid, while an optically pure L(+) or D()-lactic acid
can be produced from fermentation of renewable sources
with the appropriate microbial strains (Hofvendahl and
Hahn-Hagerdal 2000). In addition, growing environmental
concerns arising from the pollution caused by the petrochemical industry and soaring international oil prices further strengthen the preference of microbial fermentation
over chemical synthesis.
Due to the extensive industrial use and enormous economical value, production of lactic acid from LAB has
been intensively studied throughout the years and optimization of lactic acid production has become the ultimate objective in most of the studies conducted. The
effects of different fermentation medium such as beet
molasses (Kotzamanidis et al. 2002), vine-trimming waste
(Bustos et al. 2004), soybean hydrolysate (Kwon et al.
2000), lactose and concentrated cheese whey (Schepers
et al. 2002) on lactic acid production has been investigated. Other means of increasing lactic acid production
includes, optimization of fermentation systems by utilizing the cell-recycle system, together with repeated batch
and continuous processes (Oh et al. 2003); utilization of
genome shuffling approach (Yu et al. 2008); development
mutant strain with high lactic acid production (Bhatt and
Srivastava 2012); utilization metabolic engineering
approach (Kyla-Nikkila et al. 2000); cell immobilization
(Senthuran et al. 1999); and development of continuous
electrodialysis fermentation system for higher production
of lactic acid (Min-tian et al. 2005). The effects of culture
temperature, nitrogen sources, pH, oxygen and growthstimulating elements such as B-vitamins and amino acids
on lactic acid production by LAB have also been reported
(Hujanen and Linko 1996; Fu and Mathews 1999). Lactic
acid production by Lactobacillus rhamnosus have been
demonstrated to be predominantly growth-associated
(Berry et al. 1999), while our previous study demonstrated that growth enhancement via the addition of

Mn2+ and Mg2+ enhanced lactic acid production

divalent metal ions led to enhanced production of lactic

acid (Lew et al. 2012). Despite various data supporting
the growth promoting effects of divalent metal ions,
information on the exact targets of divalent metal ions
during synthesis of lactic acid by LAB remains scarce.
Thus, to further understand the influence of divalent
metal ions on the production of lactic acid, a molecular
modelling approach was adopted with the aim of observing and studying the interactions between the enzymes
and substrates involved along the hexose monophosphate
pathway in the presence of divalent metal ions.
Materials and methods
Bacteria and media preparation
Lactobacillus rhamnosus FTDC 8313 was obtained from
the Culture Collection Centre of Bioprocess Technology
Division, School of Industrial Technology, Universiti
Sains Malaysia (Penang, Malaysia). The strain was activated in sterile de Man, Rogosa and Sharpe (MRS) broth
(Hi-Media, Mumbai, India) containing 0
2 mmol l1
MnSO4 and 0
8 mmol l1 MgSO4 for three consecutive
times using 10% (v/v) inoculum and incubated at 37C
for 24 h ahead of use. The stock cultures were stored at
20C in 40% (v/v) sterile glycerol.
Determination of lactic acid
Determination of lactic acid was performed according to
Lew et al. (2012). Sterile reconstituted skimmed milk (8%;
w/v) was supplemented with manganese sulfate, MnSO4
(Sigma-Aldrich, Steinheim, Germany) and magnesium sulfate, MgSO4 (Sigma-Aldrich) as according to Table 1. The
medium was then inoculated with 10% (v/v) inoculum and
incubated at 37C for 12 h. A 100 ll of 15
8 mol l1 HNO3
was added to 1
5 ml of sample to digest the protein. The fermentation broth was then centrifuged at 10 000 g for
15 min at 25C, filtered through a 0
2 lm cellulose acetate
syringe filter and stored at 20C prior to analyses. A highperformance liquid chromatography (HPLC) equipped with
UV/Vis detector (Jasco 875-UV, Tokyo, Japan) set at
220 nm was used to determine the concentration of lactic
acids. An Aminex HPX-87H column (300 9 7
8 mm; BioRad Laboratories, Richmond, CA, USA) maintained at 65C
was used with a degassed mobile phase of 0
004 mol l1
H2SO4 at a flow rate of 0
6 ml min1.
Response surface methodology
Response surface methodology was applied with two
independent factors namely, manganese sulfate (X1) and
magnesium sulfate (X2), to generate a central composite

Journal of Applied Microbiology 116, 644--653 2013 The Society for Applied Microbiology

645

L.-C. Lew et al.

Mn2+ and Mg2+ enhanced lactic acid production

Table 1 Matrix of the central composite design in coded levels for

the factors and response, for the production of lactic acid by Lactobacillus rhamnosus FTDC 8313 in the presence of MgSO4 and MnSO4
Standard
run

Block*

MnSO4
(X1)

MgSO4
(X2)

Response
(Y),

1
2
3
4
5
6
7
8
9
10
11
12
13
14

1
1
1
1
1
1
1
2
2
2
2
2
2
2

0
0
1
1
1
0
1
0
0
0
0
a
0
a

0
0
1
1
1
0
1
0
0
a
a
0
0
0

0
355
0
358
0
365
0
336
0
346
0
352
0
341
0
349
0
356
0
349
0
350
0
338
0
356
0
344
















0
001
0
001
0
002
0
001
0
001
0
003
0
002
0
002
0
001
0
003
0
002
0
001
0
001
0
002

*1 = batch 1; 2 = batch 2.
X1 = concentration of manganese sulfate added into the fermentation broth (1
7756
508 mmol l1; a = 0
7697
514 mmol l1);
X2 = concentration of magnesium sulfate added into the fermentation broth (1
2175
274 mmol l1; a = 0
3656
127 mmol l1).
Y = production of lactic acid (mg ml1) by Lact. rhamnosus FTDC
8313.
All results are means  standard deviation from three separate runs;
n = 3.

design (CCD) matrix with an alpha value of 1
414. The

treatment combinations were allocated into two blocks,
with the first and second block representing the first day
and second day of experiment, respectively. The first
block contained the full-factorial runs accompanied by
three centre runs while the second block contained the
axial runs accompanied by three centre runs. All experimental points are presented as the mean values of a triplicate determination. Screening and selection of factors
and their varying concentrations were determined as previously described (Lew et al. 2012).
The modelling and statistical analyses were carried out
using DESIGN EXPERT 5 (software version 5.07; Stat-Ease
Corp., Minneapolis, MN, USA). An experiment for validation purposes was carried out to confirm the legitimacy
and reproducibility of the model. The concentration of
lactic acid was assessed using a random point, and the
actual result was compared with the predicted value by
the model.
Molecular docking
A total of 13 essential enzymes involved in the hexose
monophosphate pathway for the production of lactic acid
were selected from Protein Data Bank (PDB) for the
646

computational study (Table 4). Molecular docking simulation was performed on these enzymes with four different conditions namely in the presence of magnesium
(Mg2+), manganese (Mn2+), both Mg2+ and Mn2+ and
without any metal ions. Different divalent metal ions
were docked using AUTODOCK 4.2 (Morris et al. 2009). A
total of 100 runs with 250 population size, with Lamarkian genetic searching algorithm and root mean square
tolerance of 1
0 
A were set as the docking input parameter. The lowest free energy of binding (FEB) of each conformation in the most populated cluster was selected.
Analysis and visualization of the docking results was performed using VMD (Humphrey et al. 1996) and LIGPLOT
(Wallace et al. 1995).
Statistical analysis
Data analysis was performed using SPSS Inc. software
(version 20.0; Chicago, IL, USA). Two-way analysis of
variance (ANOVA) was used to evaluate the significant differences between sample means, with significance level at
a = 0
05. Mean comparisons were assessed by Tukeys
test, and all data presented were mean values obtained
from three separate runs. P-values were stipulated to
indicate the general inclination of the factors studied
on the response variables with the respective statistical
significance.
Results
Lactic acid response surface
A rotatable CCD with an alpha value of 1
414, with a
fixed middle point of X1 (4
142 mmol l1 MnSO4) and
X2 (3
205 mmol l1 MgSO4), was used to generate the
design matrix and response (production of lactic acid;
Table 1). By fitting the experimental data with least
squares method, a simulated second-order expression was
obtained as follows:
Y 0
35  0
00069X1  0
0032X2  0
0062X12
 0
002X22 0
0085X1 X2

where Y is the predicted response of lactic acid, while X1

and X2 are the coded values of MnSO4 and MgSO4,
respectively.
The adequacy and fitness of the model were evaluated
using analysis of variance (ANOVA), and the data obtained
is presented in Table 2. The regression analyses indicated
that the quadratic model was significant (P < 0
05) suggesting that the model accurately represented the data in
the experimental region. The insignificant P-value
(0
1370) of lack-of-fit indicated a reasonable fit of the
model as an approximation to the true response. The

Journal of Applied Microbiology 116, 644--653 2013 The Society for Applied Microbiology

L.-C. Lew et al.

Mn2+ and Mg2+ enhanced lactic acid production

Table 2 Analysis of variance (ANOVA) of the second-order model (Y)*, for the production of lactic acid by Lactobacillus rhamnosus FTDC 8313 in
the presence of MgSO4 and MnSO4
Source

Sum of squares

Regression
Linear
Quadratic
Model
Residual
Lack-of-fit
Pure error
Correlation total

8
45
5
92
6
77
1
77
1
27
5
07
8
63

9
9
9
9
9
9
9

105
104
104
104
104
105
104

df

Mean square

2
3
5
7
3
4
13

4
23
1
97
1
35
2
54
4
24
1
27

9
9
9
9
9
9

105
104
104
105
105
105

F-value

P-value

0
55
7
77
5
33

0
5940
0
0125
0
0245

3
35

0
1370

*Y 0
35  0
00069X1  0
0032X2  0
0062X12  0
002X22 0
0085X1 X2 [Y = production of lactic acid (mg ml1) by Lact. rhamnosus FTDC
8313; X1 = concentration of manganese sulfate added into the fermentation broth (1
7756
508 mmol l1; a = 0
7697
514 mmol l1);
X2 = concentration of magnesium sulfate added into the fermentation broth (1
2175
274 mmol l1; a = 0
3656
127 mmol l1)].
R2 = 0
7920.
Significant at an a level of 0
05.

statistical analyses with coefficient estimates and the significance of the lactic acid response model are presented
in Table 3. Considering that a quadratic model was used,
analyses of coefficient estimates were performed on quadratic effects. Our data indicated that MnSO4 produced a
significant (P = 0
0121) quadratic effects on lactic acid
production, while MgSO4 did not. Despite this, the interaction of MnSO4 and MgSO4 produced a significant
effect on the production of lactic acid by Lact. rhamnosus
FTDC 8313 (P = 0
0119), indicating that both factors
exerted a synergistic effect and should be present together
to achieve a high production of lactic acid.
A three-dimensional (3-D) response surface for the production of lactic acid (Fig. 1) generated based on the second-order equation (Eqn 1) illustrated that an optimum
region could be observed. This region occurred in the
presence of 4
73 mmol ml1 MnSO4 and 4
42 mmol ml1
MgSO4, in tandem with our previous findings of an optimum growth region (10
59 log10 CFU ml1; Lew et al.
2012), justifying that the production of lactic acid was
growth-associated. However, from the 3-D response surface of lactic acid, it is also observed that lactic acid production was high at along the region where the ratio of
MnSO4 to MgSO4 was almost 1 : 1. When one of the divalent metal ions was at a high concentration and the other
at a low concentration, lactic acid production decreased.
Validation experiment was performed to further ascertain the predictions and the reliability of the regression
model. Lactic acid concentration obtained in the presence
of 4
73 mmol ml1 MnSO4 and 4
42 mmol ml1 MgSO4
were compared with the predicted value from the model.
Our results showed that the lactic acid concentration
obtained from actual experimentation was 0
356 mg ml1,
producing an error of 0
30% as compared to the predicted
value. The small error indicated that the prediction generated from the model was reliable and valid.

Table 3 Analysis of the coefficient estimates of the second-order

model (Y)* for the production of lactic acid by Lactobacillus rhamnosus FTDC 8313 in the presence of MgSO4 and MnSO4
Variable

Coefficient estimate

Standard error

t-value

P-value

Intercept
X1
X2
X12
X22
X1 X2

c0
c1
c2
c11
c22
c12

0
0021
0
0018
0
0018
0
0019
0
0019
0
0025

0
39
1
78
3
36
1
07
3
37

0
7103
0
1178
0
0121
0
3213
0
0119

=
=
=
=
=
=

0
35
0
00069
0
0032
0
0062
0
0020
0
0085

*Y c0  c1 X1  c2 X2  c11 X12  c22 X22 c12 X1 X2 [Y = production of

lactic acid (mg ml1) by Lact. rhamnosus FTDC 8313].
X1 = concentration of manganese sulfate added into the fermentation broth (1
7756
508 mmol l1; a = 0
7697
514 mmol l1);
X2 = concentration of magnesium sulfate added into the fermentation broth (1
2175
274 mmol l1; a = 0
3656
127 mmol l1).
Significant at an a level of 0
05.

Molecular docking
The hexose monophosphate pathway for the production of
lactic acid comprises of 13 enzymes (Fig. 2) and all the
enzymes studied had individual binding sites. Molecular
docking of Mn2+ and Mg2+ was performed within the substrate binding site prior to enzymesubstrate docking, for
enzymes that were void of divalent metal ions in their cocrystal structures. The presence of both Mg2+ and Mn2+
synergistically reduced the FEB for substrates, in five
enzymes along the hexose monophosphate pathway
(Table 4; Fig. 3); glucose-6-phosphate dehydrogenase,
phosphogluconate dehydrogenase, glyceraldehyde-3-phosphate dehydrogenase, phosphopyruvate hydratase and
pyruvate kinase. In addition to FEB, four of the five
enzymes (glucose-6-phosphate dehydrogenase, phosphogluconate dehydrogenase, glyceraldehyde-3-phosphate
dehydrogenase and pyruvate kinase) showed similar

Journal of Applied Microbiology 116, 644--653 2013 The Society for Applied Microbiology

647

L.-C. Lew et al.

Mn2+ and Mg2+ enhanced lactic acid production

0358911
0352968

Lactic acid (mg ml1)

0347024
0341081
0335138

5274
6508

4260

5325

3245
MgSO4 (mmol l1)

4141
2958
1217 1775
MnSO (mmol l1)

2231

Figure 1 Response surface of lactic acid (mg ml1) production by

Lactobacillus rhamnosus FTDC 8313 upon incubation for 12 h at
37C in 8% (w/v) in reconstituted skimmed milk from the effects of
manganese sulfate and magnesium sulfate.

binding conformation with their respective substrates,

where Mn2+ was located closely to the phosphates moiety
(<5 
A from the substrate) while Mg2+ was located at 9 
A
from the substrates (Table 5).
In the presence of both Mg2+ and Mn2+ within the
binding site, the FEB of beta-D-glucose-6-phosphate
towards glucose-6-phosphate dehydrogenase decreased
significantly from 4
79 kcal mol1 (without ion) to
8
59 kcal mol1, and this also far surpassed the effects
of individual divalent metal ions (Table 4). Mn2+ was
located 1
62 
A from b-D-glucose-6-phosphate (Table 5),
indicating a higher possibility to coordinate formation of
bonds with the phosphate moiety containing a 2 partial
charge (Lieberman et al. 2007). Considering that the
phosphate moiety of b-D-glucose-6-phosphate plays an
essential role in the enzymatic activity of glucose-6-phosphate dehydrogenase, the presence of Mn2+ may have
played a prominent role during catalysis, where Mn2+
could be chelated by the phosphorus moiety of b-D-glucose-6-phosphate (Fig. 3a). With a distance of 9 
A from
the phosphorus moiety, Mg2+ could not have a direct
hydrogen bonding or chelating effect with the substrate.
However, a single hydrogen bond was found between
Tyr415 with Mg2+ indicating that Mg2+ was utilized to
balance the polarity in Tyr residues and may have
increased the efficiency of the catalytic mechanism in glucose-6-phosphate dehydrogenase, despite the absence of
any direct interaction with the substrate.
A similar synergistic effect was observed in the interactions of phosphogluconate dehydrogenase, glyceralde648

hyde-3-phosphate dehydrogenase and pyruvate kinase

with 6-phopho-D-gluconate (Fig. 3b), glyceraldehyde-3phosphate (Fig. 3c) and phosphoenolpyruvate (Fig. 3e),
respectively. The binding affinity of these three enzymes
in the presence of Mg2+ and Mn2+ increased over threefold as compared to the absence of divalent metal ions
(Table 4). The FEB reflected a similar trend where the
binding affinities for enzymes towards their respective
substrates increased over fivefold in the presence of both
Mg2+ and Mn2+ as compared to the absence of divalent
metal ions. Phosphorus moiety from all the substrates
(6-phopho-D-gluconate, glyceraldehyde-3-phosphate and
phosphoenolpyruvate) were more favourably bound to
Mn2+ than Mg2+ with the distance of 1
65, 1
70 and
1
8 
A with atom O1P of the substrate in the presence of
A (Table 5). We
Mn2+ while Mg2+ was distanced over 16 
postulate a higher tendency of Mn2+ in forming coordinate bond with the substrate as compared to Mg2+.
Phosphopyruvate hydratase also showed better binding
towards its substrate with a threefold increase in FEB in
the presence of both divalent metal ions, as compared to
the absence of divalent metal ions. The binding orientation for phosphopyruvate hydratase (Fig. 3d) was also
different compared with the previously described four
enzymes, where Mg2+ was located nearer to the substrate,
and the substrate was sandwiched between both Mg2+
and Mn2+. Phosphopyruvate hydratase is recognized as a
metalloenzyme in the glycolytic pathway, utilizing two
Mg2+ ions in catalysis during conversion of 2-phosphoglycerate to phosphoenolpyruvate and produces a single
water molecule (Poyner et al. 1996; Pancholi 2001). In
our current simulation, a single Mn2+ replaced one Mg2+.
However, no significant differences in FEB values were
observed in the presence of a single Mg2+ or Mn2+
(Table 4). Mg2+ bounded with the carboxyl group of
2-phosphoglycerate (Fig. 3d) via coordination bonding,
stabilizing the negative charge on the deprotonated oxygen (carboxyl group) while increasing the acidity of the
alpha hydrogen (carboxyl group). In addition, Mn2+ was
closely positioned with the phosphorus moiety of 2-phosphoglycerate, forming coordination bond. The formation
of coordination bonds between both divalent metal ions
had synergistically promoted the heterolysis properties of
phosphopyruvate hydratase, leading to increased docking
binding affinity.
Discussion
In our previous study (Lew et al. 2012), we optimized
the growth of Lact. rhamnosus FTDC 8313 via the addition of Mn2+ and Mg2+. In addition, we also observed
that the concentration of total acids detected in the spent
broth reflected a similar response surface pattern as that

Journal of Applied Microbiology 116, 644--653 2013 The Society for Applied Microbiology

L.-C. Lew et al.

Mn2+ and Mg2+ enhanced lactic acid production

-D-Glucose-6P
Glucose-6-phosphate
isomerase

Mn2+

-D-Glucose-6P

Mg2+

Glucose-6-phosphate
dehydrogenase
6-P-D-Glucono-1,5-lactone

The presence of Mn2+ and Mg2+

significantly affected the binding
of -D-Glucose-6P to
Glucose -6-phosphate
dehydrogenase

Divalent metal ions

( Mn2+, Mg2+)

6-phosphogluconolactonase

6-P-D-Gluconate

The presence of Mn2+ and Mg2+

significantly affected the binding
of 6-P-D-Gluconate to
Phosphogluconate
dehydrogenase

Mn2+
Phosphogluconate
dehydrogenase
D-Ribulose-5-P
Ribulose phosphate-3epimirase

Mg2+

Divalent metal ions

( Mn2+, Mg2+)

D-Xylulose-5-P
Xylulose-5-phophate
phophosketolase

The presence of Mn2+ and Mg2+

significantly affected the
binding of Glyceraldehyde-3-P
to Glyceraldehyde-3phosphate dehydrogenase

Glyceraldehyde-3-P

Mg2+

Mn2+
Glyceraldehyde-3phosphate
dehydrogenase
1,3-P-Glycerate

Divalent metal ions

( Mn2+, Mg2+)

Phosphoglycerate kinase
3-P-Glycerate
Phosphoglycerate mutase

2-P-Glycerate

The presence of Mn2+ and Mg2+

significantly affected the
binding of 2-P-Glycerate
to phosphopyruvate
hydratase

Mg2+
Mn2+
Phosphopyruvate
hydratase

Divalent metal ions

( Mn2+, Mg2+)

PEP

Mn2+

Pyruvate kinase

Mg2+
Divalent metal ions
( Mn2+, Mg2+)

Pyruvate
D-lactate
dehydrogenase
D-Lactate

L-lactate dehydrogenase

The presence of Mn2+ and Mg2+ significantly affected the

binding of PEP to Pyruvate kinase

L-Lactate

Figure 2 Main enzymes of the hexose phosphate pathway for the production of lactic acid and the specific targets of Mg2+ and Mn2+ (present
synergistically) on enzymesubstrate interactions.

Journal of Applied Microbiology 116, 644--653 2013 The Society for Applied Microbiology

649

L.-C. Lew et al.

Mn2+ and Mg2+ enhanced lactic acid production

Table 4 Free energy of binding (FEB) of enzymes involved in the hexose monophosphate pathway upon docking against their natural substrates
in the presence and absence of divalent metal ions
FEB (kcal mol1)
No.

Enzyme

PDB ID

Without ion

Mg2+

Mn2+

Mg2+ and Mn2+

1
2
3
4
5
6
7
8
9
10
11
12
13

Glucose-6-phosphate isomerase
Glucose-6-phosphate dehydrogenase
6-phosphogluconolactonase
Phosphogluconate dehydrogenase
Ribulose phosphate-3-epimerase
Xylulose-5-phosphate phosphoketolase
Glyceraldehyde-3-phosphate dehydrogenase
Phosphoglycerate kinase
Phosphoglycerate mutase
Phosphopyruvate hydratase
Pyruvate kinase
D-lactate dehydrogenase
L-lactate dehydrogenase

3FF1
1DPG
3OC6
2IYO
2FLI
3AHE
3LC2
1VPE
1EJJ
3QN3
3TOT
1J49
3D4P

6
95
4
79
6
98
4
79
5
36
5
28
4
36
9
77
10
09
3
59
4
46
2
40
5
10

7
75
4
74
6
66
11
60
7
93
11
84
4
07
17
79
17
70
7
68
13
55
4
23
5
22

7
38
4
52
6
58
8
93
7
70
20
40
4
25
17
57
17
71
7
79
13
55
19
63
5
20

6
68
8
59*
7
09
12
05*
8
21
12
39
22
45*
18
06
17
76
10
73*
28
81*
6
38
6
39

PDB, Protein Data Bank.

*Significant reduction of FEB, from the synergism effects of Mg2+ and Mn2+ as compared to the absence of ions and presence of individual ions.

of growth. Our strain of Lact. rhamnosus FTDC 8313 is a

heterolactic fermenter with detectable acetic and lactic
acids via HPLC (Lew et al. 2013); lactic acid is a growthassociated predominant metabolite of the two acids. We
have previously postulated that Mn2+ and Mg2+ enhanced
growth leading to enhanced production of lactic acid.
However, we also hypothesize that Mn2+ and Mg2+ targeted different sites along the hexose monophosphate
pathway. Thus in this study, using the same parameters
obtained from our previous study on optimization of
growth, we would like to justify such a hypothesis. In this
study, it is observed that lactic acid production was high
along the region where the ratio of MnSO4 to MgSO4
was almost 1 : 1. When one of the divalent metal ions
was at a high concentration and the other at a low concentration, lactic acid production decreased, justifying the
need for both Mg2+ and Mn2+ to be present simultaneously in stimulating the production of lactic acid.
A molecular modelling approach was adopted to further
understand the influence of Mn2+ and Mg2+ on the production of lactic acid and to determine possible specific
targets of the divalent metal ions on enzymessubstrates
along the hexose monophosphate pathway. The hexose
monophosphate pathway for the production of lactic
acid comprises of 13 enzymes (Fig. 2). The crystallized
structures of all enzymes were selected based on criteria
such as origin, apo or holo structure, existence of divalent
metal ions and the types of structures (wild or mutant)
from the PDB (www.pdb.org). In addition, an assumption
was also made where enzymes with a same Enzyme
Commission (EC) number will adopt the same functional

650

domain despite originating from different organisms.

The functional domain is also known as binding site, which
will be the core region affecting the binding affinity of
enzymesubstrate. Thus, all the 13 enzymes studied had
individual binding sites, and molecular docking simulation
was performed utilizing a grid box with the size of
50 9 50 9 50 and grid spacing of 0
375 
A within the
binding site.
Mn2+ and Mg2+ have the same partial charges and are
known to play significant roles in electrostatic stabilization and enzyme activation (Andreini et al. 2008).
However, our current data reflected that both Mn2+ and
Mg2+ functioned differently in affecting the substrate
binding affinity among the five enzymes evaluated. Mn2+
has been reported to mainly affect redox processes due to
its nature as a transition element in enzymatic catalytic
mechanisms while Mg2+ could not (Silva and Williams
2001). Our present data suggested that Mn2+ did not act
as a redox centre, but formed coordination bond with
the electron lone pair from both oxygen atoms of the
substrate phosphorus moiety. This process stabilized the
substrate due to increased nucleophilicity, leading to
improved binding affinity towards the respective
enzymes.
In general, Mg2+ is one of the most abundant divalent metal ions in most organisms (Andreini et al.
2008), greatly needed to balance the polarity of
enzymes involved in catalytic properties (Sissi and Palumbo 2009), and is vital for the activation of various
important biological metabolic pathways (Sanwal 1970;

Ozer
et al. 2001; Romani and Maguire 2002; Andreini

Journal of Applied Microbiology 116, 644--653 2013 The Society for Applied Microbiology

L.-C. Lew et al.

Mn2+ and Mg2+ enhanced lactic acid production

(a)

(b)

Lys148
Asp235

Lys184

Mn2+
269

Mg2+

162

302
Tyr415

914

Ans188
Asn102

His240

Glu270

Trp267

268

His178

165

289
317

Beta-D-glucose
-6-phosphate

339

269
165 Mn2+

167

Mg2+

263

Lys266

6-phospho-D-gluconate
Glu216

(c)

(d)
Gln51

Asp192
Thr239

160

Mg2+

269

Asp183

211
188

His154
His191

2-phosphoglycerate
Mn2+
390
160

Mn2+

Lys332

170

144

Asp308

181

Glyceraldehyde
-3-phosphate

Glu162

Mg2+

199

Asp307
Lys193

Glu163

Asp239

(e)

IIe361

Thr353

Ans369
179

166

Mn2+

240

164

Mg2+

Thr348
Lys341

Phosphoenolpyruvate

Figure 3 Binding interactions of selected enzymes (a) glucose-6-phosphate dehydrogenase, (b) phosphogluconate dehydrogenase (c) glyceraldehyde-3-dehydrogenase (d) phosphopyruvate hydratase and (e) pyruvate kinase with their respective substrates in the presence of both Mn2+ and
Mg2+.

et al. 2008; Sissi and Palumbo 2009). However, our

current data illustrated that Mg2+ was located at distances surpassing the cut-offs of any important interactions, yet the presence of Mg2+ was significant in five
enzymes along the hexose monophosphate pathway.
This led us to hypothesize that Mg2+ played a different
role, in that of maintaining polarity within the binding
pocket and thus producing a stable binding environment for catalytic activities and favourable for the
actions of Mn2+.
In conclusion, using response surface methodology,
we have demonstrated that the interaction of Mn2+
and Mg2+ produced a significant effect on the produc-

tion of lactic acid by Lact. rhamnosus FTDC 8313, indicating that both factors exerted a synergistic effect and
should be present together to achieve a high production of lactic acid. Meanwhile, the 3-D response surface
indicated that when one of the divalent metal ions was
at a high concentration and the other at a low concentration, lactic acid production decreased, justifying the
need for both Mg2+ and Mn2+ to be present simultaneously in stimulating the production of lactic acid.
Using molecular modelling approach, we illustrated that
Mn2+ and Mg2+ targeted different sites along the
hexose monophosphate pathway, leading to improved
binding affinity for substrates in five enzymes namely,

Journal of Applied Microbiology 116, 644--653 2013 The Society for Applied Microbiology

651

L.-C. Lew et al.

Mn2+ and Mg2+ enhanced lactic acid production

Table 5 Distance of Mn2+ and Mg2+ (

A) with the phosphorus moiety of substrates
Distance between
interacting moiety with
divalent metal ion (
A)
Enzyme

Substrate

Atom

Mn2+

Mg2+

Glucose-6-phosphate dehydrogenase

Beta-D-glucose-6-phosphate

Phosphogluconate dehydrogenase

6-Phospho-D-gluconate

Glyceraldehyde-3-phosphate dehydrogenase

Glyceraldehyde-3-phosphate

Phosphopyruvate hydratase

2-phospho-glycerate

Pyruvate kinase

Phosphoenolpyruvate

O1P
O2P
O3P
O1P
O2P
O3P
O1P
O2P
O3P
O1P
O2P
O3P
O1P
O2P
O3P

3
64
1
62
2
69
1
65
4
16
3
90
1
70
1
88
3
50
3
90
4
03
1
60
3
38
1
64
1
79

9
14
11
26
10
04
17
44
19
01
16
67
19
87
19
48
18
51
4
08
1
81
4
09
16
29
15
64
17
45

glucose-6-phosphate
dehydrogenase,
phosphogluconate dehydrogenase, glyceraldehyde-3-phosphate dehydrogenase, phosphopyruvate hydratase and pyruvate
kinase.
Acknowledgement
This work was financially supported by the Science Fund
Grant (305/PTEKIND/613222) provided by the Malaysian
Ministry of Science, Technology and Innovation
(MOSTI), the FRGS grant (203/PTEKIND/6711239) provided by the Malaysian Ministry of Higher Education
(MOHE), the Research University grant (1001/PTEKIND/
815056) and USM Fellowship provided by Universiti
Sains Malaysia.
Conflict of interest
The authors declare that there are no conflict of interest.
References
Andreini, C., Bertini, I., Cavallaro, G., Holliday, G.L. and
Thornton, J.M. (2008) Metal ions in biological catalysis:
from enzyme databases to general principles. J Biol Inorg
Chem 13, 12051218.
Berry, A., Franco, C.M., Zhang, W. and Middelberg, A.J.
(1999) Growth and lactic acid production in batch culture
of Lactobacillus rhamnosus in a defined medium.
Biotechnol Lett 21, 163167.

652

Bhatt, S.M. and Srivastava, S.K. (2012) High yield of lactic

acid production by mutant strain of L. delbrueckii U12-1
and parameter optimization by Taguchi methodology.
Ann Biol Res 3, 25792592.
Bustos, G., Moldes, A.B., Cruz, J.M. and Domnguez, J.M.
(2004) Production of fermentable media from vinetrimming wastes and bioconversion into lactic acid by
Lactobacillus pentosus. J Sci Food Agric 84, 21052112.
Fu, W. and Mathews, A.P. (1999) Lactic acid production from
lactose by Lactobacillus plantarum: kinetic model and
effects of pH, substrate, and oxygen. Biochem Eng J 3,
163170.
Hofvendahl, K. and Hahn-Hagerdal, B. (2000) Factors
affecting the fermentative lactic acid production from
renewable resources. Enzyme Microb Technol 26, 87107.
Hujanen, M. and Linko, Y.Y. (1996) Effect of temperature and
various nitrogen sources on L(+)-lactic acid production
by Lactobacillus casei. Appl Microbiol Biotechnol 45,
307313.
Humphrey, W., Dalke, A. and Schulten, K. (1996) VMD:
visual molecular dynamics. J Mol Graph 14, 3338, 2738.
Kotzamanidis, C., Roukas, T. and Skaracis, G. (2002)
Optimization of lactic acid production from beet molasses
by Lactobacillus delbrueckii NCIMB 8130. World J
Microbiol Biotechnol 18, 441448.
Kwon, S., Lee, P.C., Lee, E.G., Keun Chang, Y. and Chang, N.
(2000) Production of lactic acid by Lactobacillus
rhamnosus with vitamin-supplemented soybean
hydrolysate. Enzyme Microb Technol 26, 209215.
Kyla-Nikkila, K., Hujanen, M., Leisola, M. and Palva, A.
(2000) Metabolic engineering of Lactobacillus helveticus

Journal of Applied Microbiology 116, 644--653 2013 The Society for Applied Microbiology

L.-C. Lew et al.

CNRZ32 for production of pure L-(+)-lactic acid. Appl

Environ Microbiol 66, 38353841.
Lew, L.C., Liong, M.T. and Gan, C.Y. (2012) Growth
optimization of Lactobacillus rhamnosus FTDC 8313 and
the production of putative dermal bioactives in the
presence of manganese and magnesium ions. J Appl
Microbiol 114, 526535.
Lew, L.C., Gan, C.Y. and Liong, M.T. (2013) Dermal
bioactives from lactobacilli and bifidobacteria. Ann
Microbiol 63, 10471055.
Lieberman, M., Marks, A.D., Smith, C.M. and Marks, D.B.
(2007) Marks Essential Medical Biochemistry. Philadelphia,
PA: Lippincott Williams & Wilkins.
Min-tian, G., Koide, M., Gotou, R., Takanashi, H., Hirata, M.
and Hano, T. (2005) Development of a continuous
electrodialysis fermentation system for production of lactic
acid by Lactobacillus rhamnosus. Process Biochem 40,
10331036.
Morris, G.M., Huey, R., Lindstrom, W., Sanner, M.F.,
Belew, R.K., Goodsell, D.S. and Olson, A.J. (2009)
Autodock4 and AutoDockTools4: automated docking
with selective receptor flexibility. J Comput Chem 16,
27852791.
Oh, H., Wee, Y.J., Yun, J.S. and Ryu, H.W. (2003) Lactic acid
production through cell-recycle repeated-batch bioreactor.
Appl Biochem Biotechnol 108, 603613.
zer, N., Aksoy, Y. and Og
us, I.H. (2001) Kinetic properties
O
of human placental glucose-6-phosphate dehydrogenase.
Int J Biochem Cell Biol 33, 221226.
Pancholi, V. (2001) Multifunctional alpha-enolase: its role in
diseases. Cell Mol Life Sci 58, 902920.
Poyner, R.R., Laughlin, L.T., Sowa, G.A. and Reed, G.H.
(1996) Toward identification of acid/base catalysts in the
active site of enolase: comparison of the properties of

Mn2+ and Mg2+ enhanced lactic acid production

K345A, E168Q, and E211Q variants. Biochemistry 35,

16921699.
Romani, A.M. and Maguire, M.E. (2002) Hormonal regulation
of Mg2+ transport and homeostasis in eukaryotic cells.
Biometals 15, 271283.
Sanwal, B.D. (1970) Regulatory mechanisms involving
nicotinamide adenine nucleotides as allosteric effectors. 3.
Control of glucose 6-phosphate dehydrogenase. J Biol
Chem 245, 16261631.
Schepers, A.W., Thibault, J. and Lacroix, C. (2002)
Lactobacillus helveticus growth and lactic acid production
during pH-controlled batch cultures in whey permeate/
yeast extract medium. Part I. multiple factor kinetic
analysis. Enzyme Microb Technol 30, 176186.
Senthuran, A., Senthuran, V., Hatti-Kaul, R. and Mattiasson,
B. (1999) Lactic acid production by immobilized
Lactobacillus casei in recycle batch reactor: a step towards
optimization. J Biotechnol 73, 6170.
Silva, J.J.R.F.D. and Williams, R.J.P. (2001) The Biological
Chemistry of the Elements: The Inorganic Chemistry of Life.
Oxford, New York: Oxford University Press.
Sissi, C. and Palumbo, M. (2009) Effects of magnesium and
related divalent metal ions in topoisomerase structure and
function. Nucleic Acids Res 37, 702711.
Wallace, A.C., Laskowski, R.A. and Thornton, J.M. (1995)
LIGPLOT: a program to generate schematic diagrams of
protein-ligand interactions. Protein Eng 8, 127134.
Wee, Y.J., Kim, J.N. and Ryu, H.W. (2006) Biotechnological
production of lactic acid and its recent applications. Food
Technol Biotechnol 44, 163172.
Yu, L., Pei, X., Lei, T., Wang, Y. and Feng, Y. (2008) Genome
shuffling enhanced L-lactic acid production by improving
glucose tolerance of Lactobacillus rhamnosus. J Biotechnol
134, 154159.

Journal of Applied Microbiology 116, 644--653 2013 The Society for Applied Microbiology

653