TMP 5 C54

Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

REGULAR ISSUE

Mienda et al.
_______________________________________________________________________________________________________

IIOAB-India Mienda et al. IIOABJ; Vol. 5; Issue 2; 2014: 1623
16



























w
w
w
.
i
i
o
a
b
.
o
r
g







































































































































w
w
w
.
i
i
o
a
b
.
w
e
b
s
.
c
o
m

S
Y
S
T
E
M
S

B
I
O
L
O
G
Y



IN SILICO METABOLIC ENGINEERING INTERVENTIONS OF ESCHERICHIA
COLI FOR ENHANCED ETHANOL PRODUCTION, BASED ON GENE
KNOCKOUT SIMULATION
Bashir Sajo Mienda
1,2
*, Mohd Shahir Shamsir
1
, Idris Shehu
3
, Abdulkarim Ali Deba
4
, Ibrahim A
Galadima
5

1
Bioinformatics Research Group (BIRG), Biosciences and Health Sciences Department, Faculty of Biosciences and
Medical Engineering, Universiti Teknologi Malaysia, Skudai 81310 Johor Bahru, MALAYSIA
2
Department of Biological Sciences, Adamawa State University, P.M.B 25, Mubi, Adamawa State, NIGERIA
3
Department of Microbiology, Kaduna State University, P.M.B 2339, Kaduna, Kaduna State, NIGERIA
4
Science Education Programme, Abubakar Tafawa Balewa University, P.M.B 0248 Bauchi, Bauchi State, NIGERIA
5
Department of Biological Sciences, Federal University Kashere, P.M.B 0182, Gombe State, NIGERIA

ABSTRACT















*Corresponding author: Email: [email protected]

[I] INTRODUCTION

Ethanol is one of the leading biofuels which is widely pursued
in industry [1]. The amount of world ethanol production for
transport fuel is continuously increasing and it was forecasted
that the global use of biofuel including bioethanol is expected to
nearly double in 2017 [2]. Accordingly, there have recently
been several reports on the use of Escherichia coli systems
metabolic engineering for biofuels production [2, 3]. Therefore,
systems metabolic engineering, combining advanced
computational tools and synthetic biology can provide novel
solutions and strategies to further re-engineer E. coli metabolic
pathways for enhanced ethanol productions.

Genome scale metabolic reconstructions are of utmost
importance as productive tools for their capacity to integrate
genomic information to predict desired phenotype and to
simulate whole-cell physiology in an interconnected system [4].
Several genome scale metabolic models of Escherichia coli
have been published recently [5], [6] with ability to reasonably
predict accurate growth rates, metabolite excretion rates and
growth phenotypes on a number of substrate and genetic
condition that is consistent with experimental observation [5],
[7]

The basic and applied uses of genome-scale metabolic
reconstruction of E. coli have been primarily focused on six (6)
applications: (1) metabolic engineering, (2) model driven
discovery, (3) prediction of cellular phenotypes, (4) analyses of
biological network properties, (5) studies of evolutionary
processes and (6) models of interspecies interactions [8].
However, the understanding of these applications coupled with
computational predictive power will expedite forward the E.
coli systems metabolic engineering for the production of ethanol
and other compound of interest.

Systems metabolic engineering, combining computational tools and synthetic microbiology is gaining
more attention in enhancing ethanol production in Escherichia coli. However, investigation on metabolic
engineering of E. coli for increased ethanol production based on genes knockout simulation using
minimization of metabolic adjustment (MOMA) under the OptFlux software platform still remained largely
unexplored. Here, we show that in silico genes knockout encoding for the competing pathways enzymes
like pyruvate formate lyase (pflA/b0902) and lactate dehydrogenase (ldhA/b1380) under anaerobic
conditions in E. coli BMKO eliminated lactate production, reduced carbon flux towards acetate, and
enhance ethanol flux. The corresponding genetic perturbations led to a small improvement in ethanol flux
from 100% to 100.18% on glucose and 100% to 100.08% on gluconate and it is capable of substantial
increase. We hypothesize that the ability of E. coli under anaerobic condition to use other alternative
pathways to produce acetyl-CoA, acetate and ethanol is substrate dependent. Our results demonstrates
that the OptFlux framework can prospectively and effectively predict metabolic engineering interventions
using E. coli genome-scale model, based on gene knockout simulation. This would lead not only to better
understanding of E. coli systems metabolic engineering, but also could guide future experimental work
for strain improvement.



Received on: 5
th
-Mar-2014
Revised on: 25
th
-Mar-2014
Accepted on: 31
st
Mar-2014
Published on: 15
th
June-2014

Escherichia coli, ethanol,
metabolic engineering,
prediction, OptFlux and
gene knockout simulation
KEY WORDS

ISSN: 0976-3104
RESEARCH ARTICLE OPEN ACCESS

REGULAR ISSUE
Mienda et al.
_______________________________________________________________________________________________________________________


IIOAB-India Mienda et al. IIOABJ; Vol. 5; Issue 2; 2014: 1623
17



























w
w
w
.
i
i
o
a
b
.
o
r
g







































































































































w
w
w
.
i
i
o
a
b
.
w
e
b
s
.
c
o
m

S
Y
S
T
E
M
S

B
I
O
L
O
G
Y


The advent of genome-scale metabolic models of Escherichia
coli has concurrently stimulated the development of several
computational tools/software to study E. coli systems in silico.
We previously reported that computational breakthroughs and
synthetic microbiology can be synergistically combined to
improve strain performance for increase ethanol production [9].
In particular, an open source user friendly computational tool
for metabolic engineering applications called OptFlux [10], has
been developed and implemented with a genome scale
metabolic model of E. coli to predict the phenotype simulation
of both wild-type and mutant organism, using the method of
Flux Balance Analysis (FBA), Minimization of Metabolic
Adjustment (MOMA) or Regulatory on/off Minimization
(ROOM) of Metabolic flux changes [10]. This software
platform can also be used to computationally predict gene
deletions strategies using either MOMA or ROOM simulation
for over production of target compound of interest in E. coli
stoichiometric model.

However, few studies have reported on the use of computational
methods such as OptKnock for studying various gene deletions
to increase the production of lactic acid using the E. coli
stoichiometric model constructed elsewhere [11]. But there is
little information available on the use of the OptFlux software
platform with E. coli genome scale model iAF1260 [5] as a
reference metabolic engineering application tool to predict post
perturbation effects of competing pathway genes as novel
strategies to increase ethanol production in E. coli. Although
other workers [1], [12] have reported the experimental deletion
of certain competing pathway gene(s) / enzymes such as lactate
dehydrogenase A (ldhA), pyruvate formate lyase A (pflA) and
pyruvate dehydrogenase (pdh) to increase ethanol production in
E. coli. Their approach is unique as they imported ethanol
production pathway (PET operon) from Zymomonas mobilis
which becomes integrated in to the pyruvate formate lyase
(pflB) locus.

Even though a lot of work has been reported on ethanol
production in metabolically engineered E. coli strains [2], very
little work has been reported on increasing ethanol production
in E. coli based on gene knockout simulation using the OptFlux
software platform. In this study, we investigate whether ethanol
flux from glucose and gluconate can be improved, based on in
silico genes knockout simulation of pyruvate dissimilation
pathways (pflA /b0902, ldhA /1380 and frdBC /b4152, b4153)
in E. coli stoichiometric model, and how the OptFlux software
platform can prospectively and effectively predict genetic
perturbations in relations to strain performance and ethanol
production.

[II] MATERIALS AND METHODS

2.1. Model

The work described herein uses the metabolic reconstruction of
Escherichia coli iAF1260 [5]. This model has been functionally tested
and validated against experimental data to be predictive for
computations of growth rates, metabolite excretion rates, and growth
phenotypes on a number of substrate and genetic conditions [5], [7]

2.2. Flux balance analysis

Flux balance analysis (FBA) and mininmization of Metabolic Adjustment
(MOMA) were implemented using the Java programming , within the
framework of the OptFlux open source platform (http://www.optflux.org)
[10]. This provides free user-friendly tools for the ME community aiming
to be the reference platform in the field. All simulation of mutant strain
and wild type were performed using the OptFlux v3.06.

Substrates (glucose and gluconate) uptake rates for the solitary carbon
substrates in each simulation were constrained to a maximum uptake
rate of10 mmol/gDW/h. For anaerobic simulations the oxygen uptake
rate was set to be 0.0 mmol/gDW/h. This value was chosen based on
slightly close experimental observations of anaerobic growth of E. coli
[23, 24, 25].

2.3. Gene knockout using the OptFlux software
platform

Flux balance analysis predict metabolic flux distributions at steady state
by using linear programming where as minimization of metabolic
adjustment (MOMA) employs quadratic programming to identify point in
flux space, which is more or less closest to the wild-type point and
consistent with the gene knockout constraint [26]. MOMA was
implemented under the OptFlux software platform as described in their
original documentation.

OptFlux simulations using MOMA were run to completion for single and
double knockout strains. The knockout gene(s) were pflA/b0902,
ldhA/b1380 and frdBC/b4152, b4153. The single knockout strains are
designated as BMKO1 (pflA or ldhA) and the double knock out strains
are designated as BMKO2 (frdBC or pflA ldhA) see Table-1. The
strains were further described as BM101 (pflA), BM102 (ldhA),
BM103 (frdBC) and BM104 (pflA ldhA). The gene(s) knockout were
performed using glucose and gluconate as substrates to increase
ethanol production in Escherichia coli genome stoichiometric model
Table: 1. E. coli In silico models used in this study
E. coli strains Description Source
E. coli BMKO Wild-type Model (Feist et al., 2007; Le Novre et al.,
2006)
BM101 BMKO1 pflA This study
BM102 BMKO1 ldhA This study
BM103 BMKO2 frdBC This study
BM104 BMKO2 pflA ldhA This study
REGULAR ISSUE
Mienda et al.
_______________________________________________________________________________________________________________________


IIOAB-India Mienda et al. IIOABJ; Vol. 5; Issue 2; 2014: 1623
18



























w
w
w
.
i
i
o
a
b
.
o
r
g







































































































































w
w
w
.
i
i
o
a
b
.
w
e
b
s
.
c
o
m

S
Y
S
T
E
M
S

B
I
O
L
O
G
Y


[III] RESULTS

3.1. Glucose as the substrate

When glucose was used as the substrate or carbon source, all the
strains are able to ferment glucose and produce a reasonable
amount of ethanol as the main fermentation product with the
exception of strain BM103 which lost its ability to grow under
anaerobic condition as a result of knocking out frdBC (b4152,
b4153). As shown in fig 1, the deletion of pflA (Strain BM101)
led to slightly lower growth rates. The pflA (BM101) single
deletion strain maintained about 98% of the wild-type growth
rate under anaerobic condition. On the other hand, the growth
rate of ldhA (BM102) single deletion strain is about 55.58% of
the wild-type growth rate, which is the lowest growth rate
reported in this study. The double knockout strain BM104
(pflA ldhA) led to slightly lower growth rates [Figure1 and
Table2] as in strain BM101 (98% of the wild-type). There is
ultimately no growth in strain BM103 as a result of the deletion
of the frdBC genes.

Table: 2. E. coli strain design properties on glucose under the OptFlux software platform

Strains Knockout
genes
Growth rates
(1/h)
% Growth
rate
Ethanol
(mmol/gDW/h)
Acetate
(mmol/gDW/h)

% ethanol
BMKO
(WT)
- 0.18814504 100 8.495536 8.61685 100
BM101 pflA 0.18466564 98.15 8.51065 0.00 100.18
BM102 ldhA 0.10457363 55.58 8.48512 8.61453 99.88
BM103 frdBC 0.00 0.00 0.00 0.00 0.00
BM104 pflA ldhA 0.18466564 98.15 8.51065 8.61147 99.88

Maximum uptake rates for glucose were set to be 10 mmol/gDW/h and the corresponding Oxygen uptake rate was 0.0 mmol/gDW/h for anaerobic
simulation

A reasonable amount of ethanol was formed as main
fermentation products of E. coli BMKO, and all the in silico
mutant models constructed in this study with the exception of
mutant strain BM103. The mutant strains BM101 (pflA) and
BM104 (pflA ldhA) had a slightly higher ethanol flux
(100.18% or 0.18% higher) than the wild-type [Figure1 and
Table2]. Other mutants have similar ethanol flux rate as the
wild-type. There was no detectable lactate flux produced by the
wild-type but a slight lactate was produced by BM101 and
BM104 (data not shown). Other by-products such as acetate and
formate were produced along with ethanol by the wild-type
BMKO. No acetate production was observed after deletion of
pflA (strain BM101). The mutant strains BM102 and BM104
produced slightly less acetate than the wild-type BMKO.














Fig: 1. Growth rates of E. coli BMKO and mutants on glucose and gluconate. Error bars indicate standard deviation on the
replicates

REGULAR ISSUE
Mienda et al.
_______________________________________________________________________________________________________________________


IIOAB-India Mienda et al. IIOABJ; Vol. 5; Issue 2; 2014: 1623
19



























w
w
w
.
i
i
o
a
b
.
o
r
g







































































































































w
w
w
.
i
i
o
a
b
.
w
e
b
s
.
c
o
m

S
Y
S
T
E
M
S

B
I
O
L
O
G
Y


3.2. Gluconate as the substrate

Gluconate was chosen as the solitary carbon source for E. coli
BMKO to produce ethanol. When gluconate was used, all of the
mutants produced ethanol flux in combination with acetate,
succinate and formate. The ethanol flux of the wild-type (E. coli
BMKO) is slightly higher than the acetate. Overall, gluconate
was utilized faster by the strain than glucose with 90% higher
growth rate than when glucose is used as the substrate. Similar
trend for growth rates was maintained for all mutants as
depicted in Figure 1 and Table 3. Single gene knockout
strain BM101 (pflA) led to substantially similar acetate
production as the wild-type [Figure 3 and Table 3] with a
slight increase in acetate production. However, no lactate was
produced. The ethanol flux in strain BM101 is lower [Table 3]
than that of the wild-type and exhibited no lactate production.



Fig: 2. Ethanol productivity rates of E. coli BMKO and mutants on glucose and gluconate. Error bars indicate standard
deviation on the replicates

Knocking out ldhA (strain BM102) led to increase ethanol flux
(100.08%) that is 0.08% high than the wild-type (BMKO) as
shown in Figure 2 and Table 3 and only a slightly less
acetate (0.045% less) is produced relative to the wild-type
[Figure 3 and Table 3). Double knocked out strain BM104
(pflA ldhA) does not efficiently direct more carbon flow
towards ethanol production [Figure2 and Table 3]. The
mutant strain BM102 (ldhA) obtained a slightly higher ethanol
flux on gluconate [Table 3] than that of the wild-type (BMKO)
with much less acetate production. No detectable flux for lactate
is realized.



Fig: 3. Acetate productivity rates of E. coli BMKO and mutants on glucose and gluconate. Error bars indicate standard
deviation on the replicates


REGULAR ISSUE
Mienda et al.
_______________________________________________________________________________________________________________________


IIOAB-India Mienda et al. IIOABJ; Vol. 5; Issue 2; 2014: 1623
20



























w
w
w
.
i
i
o
a
b
.
o
r
g







































































































































w
w
w
.
i
i
o
a
b
.
w
e
b
s
.
c
o
m

S
Y
S
T
E
M
S

B
I
O
L
O
G
Y



Table: 3. E. coli strain design properties on gluconate under the OptFlux software platform

Strains Knockout
genes
Growth rates
(1/h)
% Growth
rate
Ethanol
(mmol/gDW/h)
Acetate
(mmol/gDW/h)

% ethanol
BMKO
(WT)
- 0.35828092 100 14.50847 14.24288 100
BM101 pflA 0.22370522 62.44 14.46933 14.25845 99.73
BM102 ldhA 0.29466613 83.58 14.52007 14.23636 100.08
BM103 frdBC 0.00 0.00 0.00 0.00 0.00
BM104 pflA ldhA 0.29136069 81.32 14.46933 14.25845 99.73

Maximum uptake rates for gluconate were set to be 10 mmol/gDW/h and the corresponding Oxygen uptake rate was 0.0 mmol/gDW/h for anaerobic
simulation.

[IV] DISCUSSION

Eschericha coli systems metabolic engineering is a powerful
approach that can be used to generate renewable compounds
such as biofuels for growing human population. Systems
biology and in silico analysis have the potential to accelerate the
re-engineering of new strains of E. coli to increase its
production potential through model driven analysis [7]. This
work present progress towards this goal through a systematic
flux balance analysis (FBA) of the production potential for
ethanol from E. coli using in silico gene knockout simulation
strategies under the OptFlux software platform. Gluconate and
glucose are considered to be the two main solitary carbon
sources.

Eschericha coli carries out mixed acid fermentation under
anaerobic conditions and produces acetate, formate, succinate,
lactate and ethanol as fermentation products. The deletion of
mixed acid fermentation reactions / genes such as frdBC and
ldhA have been previously reported [13-15] to increase the
intracellular pool of NADH in E. coli. It was also previously
[15] established that knocking NADH consuming reactions
under anaerobic conditions led to completely no growth in E.
coli, this might be the reason why in silico knocking out of
frdBC in this study led to completely no growth in strain
BM103[Figure1]. The mutant strain BM103 (frdBC) lost its
ability to grow anaerobically due to lack of presence of an
important NADH-consuming pathway (frdBC / b4152, b4153)
as an electron sink [Figure-4]. Such a strain is un able to
recycle NADH, thereby creating a driving force for reaction that
consume NADH [1517]. Three (3) different in silico gene
knockout strategies for production of ethanol in E. coli genome
scale model [5] were carried out using the OptFlux software
platform [10]. This framework uses minimization of metabolic
adjustment to predict phenotype growth rates after introduction
of genetic perturbation (gene knockout) under anaerobic
conditions (see materials and methods).

There are four routes for pyruvate dissimilation in E. coli
KO11: pfl, ldhA pdh and recombinant PET operon from
Zymomonas mobilis encoding for pyruvate decarboxylase (pdc)
and alcohol dehydrogenase II (adhII) genes [1]. We reason that
in silico gene knockout simulation of pyruvate dissimilating
enzymes and other competing pathways (encoded by the
ldhA/b1380, frdBC/b4152,b4153 and pflA/b0902 genes
(without the use of PET operon from Z. Mobilis) in E. coli
genome-scale model could play essential roles in ethanol
production.

Under anaerobic conditions, when glucose was used as carbon
source, 2 ATP and 2NADH [Figure4] are generated during
glycolysis [1]. The required additional ATP per glucose can be
realized by dissimilation of pyruvate through PFL pathway
[Figure4] which has been demonstrated in E. coli, and the
increase pool of NADH is used to reduce metabolic
intermediates and create fermentative products [1, 18]. When a
more oxidized substrate like gluconate is used, only 1 ATP and
1 NADH are generated through the Entner-Doudoroff pathway
[Figure4]. The consequence of this oxidized substrate
necessitate the cell to produced more oxidized products such as
acetate, lactate, formate succinate and ethanol as an alternative
means to balance the accumulating NADH [1, 19]. It was
theoretically established that 1 mole of gluconate is required to
produced 0.5 moles of acetate, 1. 5 moles of ethanol and 1.5
ATP whereas 1 mole of glucose is required to produced 2 mole
of ethanol and 2 ATP [1, 12]. On the other hand the growth rate
of the mutant strain BM101 (pflA) is about 98% of the wild-
type (BMKO) under anaerobic condition, while that of the
mutant strain BM102 (ldhA) is about 55.5% of the wild-type
REGULAR ISSUE
Mienda et al.
_______________________________________________________________________________________________________________________


IIOAB-India Mienda et al. IIOABJ; Vol. 5; Issue 2; 2014: 1623
21



























w
w
w
.
i
i
o
a
b
.
o
r
g







































































































































w
w
w
.
i
i
o
a
b
.
w
e
b
s
.
c
o
m

S
Y
S
T
E
M
S

B
I
O
L
O
G
Y


(considered as marginal growth) see Figure1and Table2.
This seems to clearly indicate the significance of ldhA in the
growth rate of E. coli mutant strain on glucose under anaerobic
conditions. It could utilized only a little glucose under anaerobic
conditions because of the lack of major anaerobic respiratory
pathways[20]
Glucose
2 pyruvate
frdBC
ldhA
X
X
X
pflA
pflB
Lactate
Succinate
Formate
2 Acetyl-CoA
pta
Acetyl-P
ack
Acetate
ATP
Acetaldehyde
acdH
Ethanol
Gluconate
2 NADH
2 NAD
+
Embden-Meyerhof
Pathaway
2 ATP
2 NADH
2 Pyruvate
Entner-Doudoroff
pathway
1 ATP
1 NADH
2 Pyruvate
2 NAD
+
2 NADH
PDH complex
NADH
NADH
adhE


Fig: 4. Pathways involved in fermentative utilization of glucose and gluconate by E. coli BMKO and its constructed mutant
strains. The pathways along with the deleted competing fermentation pathways are shown. The enzymes in red represent the
pathways that were blocked via gene knockout. The knockout genes encode for lactate dehydrogenase (ldhA), Pyruvate formate
lyase (pflA) and fumarate reductase (frdBC)

When glucose was used as a solitary carbon source, ethanol is
primarily produced by the wild-type model (BMKO) as well as
small amount of acetate [Figure2, -3 and Table2). The
knocking out of pflA in strain BM101 led to no acetate
production when glucose is the substrate, whereas strain BM101
on gluconate produced acetate in combination with ethanol
fermentative products [Figure-3]. Acetate production is
believed to have serious effects on the production of ethanol and
other value added compounds when E. coli is used as chassis
host. This is because more carbon flux is directed toward the
production of acetate rather than ethanol, succinate or lactate in
E. coli central metabolism. The inability of strain BM101 to
produced acetate in silico using glucose as substrate is in
agreement with a work previously reported elsewhere [1],
where deletion of pflA in E. coli KO 11 totally eliminated the
acetate production. Therefore the acetate produced by wild-type
BMKO seems to be primarily from PFL pathway, and in the
absence of pflA, the other alternative way for the cell to
produce acetyl-CoA is via the PDH pathway [1], and/or
activation of pflB [Figure-4]. E. coli pyruvate formate lyase B
REGULAR ISSUE
Mienda et al.
_______________________________________________________________________________________________________________________


IIOAB-India Mienda et al. IIOABJ; Vol. 5; Issue 2; 2014: 1623
22



























w
w
w
.
i
i
o
a
b
.
o
r
g







































































































































w
w
w
.
i
i
o
a
b
.
w
e
b
s
.
c
o
m

S
Y
S
T
E
M
S

B
I
O
L
O
G
Y


(pflB), is a central enzyme of anaerobic metabolism, it catalyzes
the co-enzyme A-dependent non-oxidative cleavage of pyruvate
to acetyl-CoA and formate in anaerobically growing cells, its
activity is induced under anaerobic conditions [21, 22].

The decrease and/or absence of acetate production could
increase ethanol production in E. coli, as more carbon flux will
be directed toward ethanol production or other NADH
consuming reactions (succinate and lactate). The flux to ethanol
in mutant strain BM101 (pflA) and BM104 (pflAladhA)
was slightly higher than the wild-type [Table 2, 3 and Figure
2]. We can hypothesize that mutant strain BM101 can produce
certain amount of ethanol in combination with acetate as
fermentation products [Figure2, 3 and Table 3] when
gluconate was used as substrate, while on the hand the same
strain BM101, produced ethanol and other fermentation
products but without formation of acetate when glucose was
used as the substrate [Table 2]. This might be as a result of the
fact that gluconate is more oxidized than glucose and in
addition, the cell might have used the alternative pathways to
produce acetyl-CoA and acetate through the PDH pathway
and/or pflB is induced to produce acetate under anaerobic
condition with gluconate as susbtrate. Pyruvate formate lyase
(pflB), has been previously reported [20, 21] to be induced
under anaerobic condition, but yet no acetate was produced in
mutant strain BM101 (pflA) when glucose was the substrate
under the same condition. We can hypothesize that anaerobicity
is not the only factor that is critical to induction of pflB activity,
this is because strain BM101 (pflA) produced acetate under
anaerobic condition with gluconate as substrate. The activity of
pflB seems to be substrate dependent, stressing the need for
careful selection of substrate to reduce acetate formation and
increase ethanol production in E. coli.

When gluconate was used as substrate, mutant strain BM101
(pflA) lead to substantially similar acetate production as the
wild-type [Figure 3 and Table 3] with a slight increase in
acetate production. The ethanol flux was only 0.08% higher
than that of the wild-type on gluconate. Surprisingly, no lactate
was produced. This in silico result is contrary to the findings
reported elsewhere [1], where the mutant strain deficient of pflA
exhibited an increase in lactate production, although their
Mutant strain had a foreign gene designated as PET operon on
its pflB locus from Z. Molbilis.

Double knockout strain BM104 (pflA/ldhA) does not seem
to efficiently direct carbon flow towards ethanol production
[Figure2 and Table 2, -3]. This result is contrary to the work
reported previously [1], where double knockout strain (pflA
ldhA) demonstrated the highest ethanol production on both
glucose and gluconate substrates respectively. Normally when
pflA is deleted, the only way for the cell to produce acetyl-CoA
and ATP is via PDH pathway and/or activation of pflB under
anaerobic conditions [21, 22], in such circumstances excess
NADH will be generated and thus, can only be re-oxidized to
NAD
+
through the native ethanol pathway or exogenous PET
operon from Z. Mobilis [1]. The fact that the double knockout
strain BM104 of this study is deficient of exogenous PET
operon from Z. Mobilis, could be the reason why it cannot
produce higher ethanol on both glucose and gluconate
substrates. The nutant strain BM102 (ldhA) obtained a slightly
higher ethanol flux (0.08% higher) on gluconate from the wild
type (BMKO) with much less acetate production. Since the
acetate production was slightly lower than the wild-type
(BMKO), we can hypothesize that flux through PFL is
responsible for the increase acetate production. It was
previously reported that PFL appears to be the preferred route
of acetyl-CoA generation over the PDH pathway when the ldhA
was knockout [1]

[IV] CONCLUTION

Systems metabolic engineering, combining computational tools
and synthetic microbiology is gaining more attention in
enhancing ethanol production in E. coli. Overall, significant
ethanol flux seen in computationally predicted strain of E. coli
model constructed in this study based on gene knockout
simulation using the OptFlux software platform proves to be an
important approach for strain improvement. While engineered
strain of E. coli already exist that produce ethanol [1, 2, 12] and
some of the strain constructed here are similar to previously
published work [1, 2]. The mutant strains constructed in this
study indicates that pflA might be responsible for acetate
production in E. coli when glucose was used under anaerobic
condition, while the production of acetate when gluconate was
used under the same condition might be as a result of activation
of pflB and/or pyruvate dehydrogenase complex. We
hypothesize and reason that the ability of E. coli under
anaerobic condition to use other alternative pathways to produce
acetyl-CoA, acetate and ethanol is substrate dependent. We
finally showed that the OptFlux software platform can
prospectively and effectively predict metabolic engineering
interventions using E. coli genome-scale model, based on gene
knockout simulation, and it could guide future experimental
work rather than relying solely on the use of the conventional
experimental trial by error approach.

CONFLICT OF INTEREST

Author declares no conflict of interest.

ACKNOWLEDGEMENTS

We would like to thank Pablo Carbonell, Sara Careia and Paulo Maia, for
providing insight on the use of the OptFlux software platform.

FINANCIAL DISCLOSURE

This work is not supported by any fund.



REGULAR ISSUE
Mienda et al.
_______________________________________________________________________________________________________________________


IIOAB-India Mienda et al. IIOABJ; Vol. 5; Issue 2; 2014: 1623
23



























w
w
w
.
i
i
o
a
b
.
o
r
g







































































































































w
w
w
.
i
i
o
a
b
.
w
e
b
s
.
c
o
m

S
Y
S
T
E
M
S

B
I
O
L
O
G
Y



REFERENCES

[1] Hildebrand A, Schlacta T, Warmack R, Kasuga T, Fan Z. [2013]
Engineering Escherichia coli for improved ethanol production
from gluconate. J Biotechnol 168:101106.
[2] Jang Y, Myoung J, Choi S, Jun Y, Young D, et al. [2012]
Engineering of microorganisms for the production of biofuels
and perspectives based on systems metabolic engineering
approaches. Biotechnol Adv 30(5): 9891000.
[3] Ho J, Uk H, In D, Yup and S. [2013] Production of bulk
chemicals via novel metabolic pathways in microorganisms.
Biotechnol Adv 31(6):925935.
[4] Fong SS, Burgard AP, Herring CD, EM et al. [2005] In silico
design and adaptive evolution of Escherichia coli for production
of lactic acid. Biotechnol Bioeng 91(5): 643648.
[5] Feist AM, Henry CS, Reed, et al. [2007] A genome-scale
metabolic reconstruction for Escherichia coli K-12 MG1655
that accounts for 1260 ORFs and thermodynamic information.
Mol Syst Biol 3(121):121.
[6] Orth JD, Conrad TM, Na J, et al. [2011] A comprehensive
genome-scale reconstruction of Escherichia coli metabolism.
Mol Syst Biol 7(535):535.
[7] Feist AM, Zielinski, JD Orth, et al. [2010] Model-driven
evaluation of the production potential for growth-coupled
products of Escherichia coli. Metab Eng 12(3):173186.
[8] McCloskey D, Palsson B, Feist AM. [2013] Basic and applied
uses of genome-scale metabolic network econstructions of
Escherichia coli. Mol Syst Biol 9 (661): 661.
[9] Mienda BS, Shamsir MS. [2013] Thermotolerant micro-
organisms in Consolidated Bioprocessing for ethanol
production: A review. Res Biotechnol 4(4): 16.
[10] Rocha I, Maia P, Evangelista P, et al. [2010] OptFlux: an open-
source software platform for in silico metabolic engineering.
BMC Syst Biol 4(45).
[11] Reed JL, Vo TD, Schilling CH, Palsson BO. [2003] An
expanded genome-scale model of Escherichia coli K-12 (iJR904
GSM/GPR). Genome Biol 4( 9):54.
[12] Fan Z, Wu W, Hildebrand A, Kasuga T, Zhang R, Xiong X.
[2012] A novel biochemical route for fuels and chemicals
production from cellulosic biomass. PLoS One 7(2):31693.
[13] Atsumi S, Cann AF, Connor MR, Shen CR, et al. [2008]
Metabolic engineering of Escherichia coli for 1-butanol
production. Metab Eng 10(6):305311.
[14] Atsumi S, Liao JC. [2008] Metabolic engineering for advanced
biofuels production from Escherichia coli. Curr Opin
Biotechnol 19(5): 414419.
[15] Shen CR, Lan EI, Dekishima Y, et al. [2011] Driving forces
enable high-titer anaerobic 1-butanol synthesis in Escherichia
coli. Appl Environ Microbiol 77(9): 29052915.
[16] Kim Y, Ingram LO, and Shanmugam. [2007] Construction of an
Escherichia coli K-12 mutant for homoethanologenic
fermentation of glucose or xylose without foreign genes. Appl
Environ Microbiol 73(6):176671.
[17] Yomano LP, York, Zhou S, Shanmugam KT, LO Ingram.
[2008] Re-engineering Escherichia coli for ethanol production.
Biotechnol Lett 30(12): 20972103.
[18] Lawford HG and Rousseau JD. [1995] Comparative Energetics
of Glucose and Xylose Metabolism in Ethanologenic
Recombinant Escherichia coil. Appl Biochem Biotechnol 51-52:
179-195.
[19] Snchez AM, Bennett GN, San KY. [2005] Effect of different
levels of NADH availability on metabolic fluxes of Escherichia
coli chemostat cultures in defined medium. J Biotechnol 117(4):
395405.
[20] Lee, Lee D, Kim TY, Kim BH. Lee J, Lee SY. [2005]
Metabolic Engineering of Escherichia coli for Enhanced
Production of Succinic Acid , Based on Genome Comparison
and In Silico Gene Knockout Simulation. Appl Environ
Microbiol 71(12):78807887.
[21] Hasona A, Kim Y, Healy FG, Ingram LO, Shanmugam KT.
[2004] Pyruvate Formate Lyase and Acetate Kinase Are
Essential for Anaerobic Growth of Escherichia coli on Xylose. J
Bacteriol 186(22):75937600.
[22] Yang J, Naik SG.cOrtillo, et al. [2009, The iron-sulfur cluster
of pyruvate formate-lyase activating enzyme in whole cells:
cluster interconversion and a valence-localized [4Fe-4S]2+
state. Biochemistry 48(39): 92349241.
[23] Varma FG, Boesch BW, Palsson BO. [1993] Stoichiometric
interpretation of Escherichia coli glucose catabolism under
various oxygenation rates. Appl Environ Microbiol 59(8): 2465
2473.
[24] Edwards , RU Ibarra, Palsson BO. [2001] In silico predictions
of Escherichia coli metabolic capabilities are consistent with
experimental data Nat Biotechnol 19(2):125130.
[25] Fischer E, Zamboni N, Sauer U. [2004] High-throughput
metabolic flux analysis based on gas chromatographymass
spectrometry derived 13C constraints. Anal Biochem
325(2):308316.
[26] Vitkup D and Church GM.[2002] Analysis of optimality in
natural and perturbed metabolic networks. Proc Natl Acad Sci
USA 99, 1511215117.

You might also like