Available online at www.sciencedirect.

com

Systems and synthetic metabolic engineering for amino acid

production – the heartbeat of industrial strain development
Judith Becker and Christoph Wittmann

With a world market of more than four million tons per year, we review recent advances in systems and synthetic
L-amino acids are among the most important products in metabolic engineering towards amino acids and related
industrial biotechnology. The recent years have seen a compounds. This will cover major achievements for the
tremendous progress in the development of tailor-made strains two major industrial production organisms – Corynebacter-
for such products, intensively driven from systems metabolic ium glutamicum and Escherichia coli.
engineering, which upgrades strain engineering into a concept
of optimization on a global scale. This concept seems Concepts and strategies for industrial strain
especially valuable for efficient amino acid production, engineering
demanding for a global modification of pathway fluxes – a The high economic interest has stimulated the develop-
challenge with regard to the high complexity of the underlying ment of over producing strains from early on. For many
metabolism, superimposed by various layers of metabolic and decades this has been carried out by iterative rounds of
transcriptional control. random mutagenesis and selection that provided mutants
with remarkable performance, that were however, limited
Address by growth defects, low stress tolerance or enhanced
Institute of Biochemical Engineering, Technische Universität nutrient demand owing to the inherent accumulation
Braunschweig, Germany
of detrimental mutations during strain development.
Corresponding author: Wittmann, Christoph Enabled by recombinant DNA technology, the focus of
(c.wittmann@tu-braunschweig.de) strain engineering recently shifted to rational approaches,
which provided a next level of production strains. For
existing industrial processes, it seems, however, not easy
Current Opinion in Biotechnology 2012, 23:718–726 to reach the high performance of the classical strains [2].
This review comes from a themed issue on Tissue, cell and Obviously, the typically local concepts of metabolic
pathway engineering engineering fail to achieve pathway engineering on a
Edited by Hal Alper and Wilfried Weber global scale with full consideration of the complex inter-
For a complete overview see the Issue and the Editorial
actions throughout the entire pathway network [3,4]. This
particularly holds for amino acid production, demanding
Available online 13th January 2012
for high yield, titer and productivity and thus a global
0958-1669/$ – see front matter, # 2012 Elsevier Ltd. All rights modification of pathway fluxes within a complex inter-
reserved.
connected network of reactions that is tightly constrained
http://dx.doi.org/10.1016/j.copbio.2011.12.025 by different layers of metabolic [5–7] and transcriptional
regulation [4,8,9]. The recent years have seen a tremen-
dous improvement of amino acid producers by systems
Introduction metabolic engineering, the upgrade of strain engineering
From the pioneering discovery of Corynebacterium gluta- to optimization on a global scale. This builds on the
micum as natural glutamate producer almost 60 years ago, powerful omics technologies of systems biology for in-
amino acid production has evolved into an industry with a depth profiling [10] and even model-based design of
rich portfolio of products and applications (Figure 1). This different cell factories of interest [11]. At present,
includes the dominating products L-glutamate and L- synthetic metabolic engineering is driving this even
lysine with annual market volumes of around 2.5 and further. The assembly of artificial pathways within cells
1.5 million tons, respectively, and a predicted market now allows to overcome natural boundaries set by the
growth of 6–8% per year. Almost all amino acids are strain specific genome repertoire and to extend traditional
exclusively derived by microbial fermentation, because amino acid pathways towards non-natural chemicals of
this provides the biologically active L-enantiomer, excep- huge commercial interest.
tions being methionine where also the D-enantiomer is
bio-available [1] and the non-chiral amino acid glycine. L-Lysine
The biosynthetic routes towards amino acids are hereby L-lysine production by C. glutamicum is one of the most
closely linked to the network of central metabolic path- impressive examples for the power of systems and syn-
ways via the supply for carbon building blocks, nitrogen, thetic metabolic engineering. The gram-positive soil
reducing power and energy – a truly complex biochemical bacterium C. glutamicum is the dominating organism for
network to be tackled by the metabolic engineer. Here, industrial production of about 1.5 million tons of this

Current Opinion in Biotechnology 2012, 23:718–726 www.sciencedirect.com

 Systems and synthetic metabolic engineering for amino acid production Becker and Wittmann 719

Figure 1

Product Production strain Application Market size [Mt/a] Titer [g/L] Yield [g/g]e
1 2 3 50 100 150 0.5 1.0

L-glutamate C. glutamicum

L-lysine C. glutamicum

L-methioninea C. glutamicum

L-threonine E. coli 0.23

L-phenylalanineb E. coli 0.08

L-tryptophanb C. glutamicum 0.05

L-arginine C. glutamicum 0.01

L-valine E. coli 0.005

L-alaninec C. glutamicum 0.005

L-serine C. glutamicum 0.003

Cadaverined C. glutamicum 0.1

Putrescined E. coli 0.1

Current Opinion in Biotechnology

Systems and synthetic metabolic engineering of E. coli and C. glutamicum for the production of amino acids and related compounds. Data comprise
major application of the different compounds (food and feed ; health and hygiene ; agriculture and technical application ; textiles, packaging
and housing and transportation and energy ), current market sizes and production performance (titer and yield) of strains obtained rationally for
the production of L-glutamate [40], L-lysine [11], L-methionine [1,50], L-threonine [61], L-phenylalanine [8,68,69], L-tryptophan [8,68,69], L-arginine
[45,68], L-valine [36,68], L-alanine [48,68], L-serine [43], cadaverine [54,55,56] and putrescine [59].
a
Strains were obtained by a combined approach of random mutagenesis and metabolic engineering.
b
Market size denotes production of the D/L-racemate.
c
The yield was corrected for the additional glucose demand required for biomass formation in the initial fermentation step, considering a biomass yield
of 0.5 g g 1 [70].
d
The rough estimate of market size in the segment of polyamides is based on the assumption that economically attractive production costs can be
achieved in bio-based production.
e
The yields given in dark blue represent experimentally achieved values by rationally engineered production strains. The yields in light blue represent
the theoretical maximum as deduced from in silico modeling for the production of L-glutamate [71], L-lysine [72], L-methionine [51], L-threonine [33], L-
phenylalanine [73], L-tryptophan [73], L-arginine [73], L-valine [74], L-alanine [73], L-serine [73] and cadaverine [52]. The theoretical maximum yield of
putrescine was assumed to be equimolar to the theoretical maximum yield of arginine.

amino acid for animal nutrition (Figure 1). The tin-opener strategies, however, allowed only limited improvement
to the first rational overproducers was a detailed under- and did not provide industrially competitive lysine produ-
standing of the contribution of individual reactions and cers, most probably because of their local character,
pathways to the carbon flux through the C. glutamicum cell, neglecting the complex network regulation in C. glutami-
created by sophisticated 13C flux studies [3,12–19]. This cum. An interesting study towards a more global concept of
allowed the identification and implementation of genetic strain engineering compared the genome sequence be-
targets beyond the terminal biosynthetic pathway itself in tween the wild type of C. glutamicum and a classical lysine
order to enhance the supply of NADPH and oxaloacetate, producer [30]. Out of the massive set of point mutations
required for lysine production. As example, the detailed present, selected beneficial modifications could be ident-
understanding of fluxes in C. glutamicum predicted the ified. These were then implemented into the non-produ-
gluconeogenetic fructose 1,6-bisphosphatase as truly novel cing wild type in subsequent optimization rounds and
target for enhanced NADPH supply [16] which could finally provided a strain with interesting properties, that
then be successfully implemented for improved pro- was a yield of 0.4 g g 1 and a final titer of 100 g L 1 [31].
duction [20,21]. Further prominent examples enhanced More recently, systems-wide global pathway engineering
lysine production by increasing the flux through the recently provided the best lysine producer known so far
NADPH supplying pentose phosphate pathway (PPP) [11]. The design of a genetic blueprint for the superior
using overexpression of individual enzymes [22], modifi- strain was based on global model-based prediction, inte-
cation of the kinetic properties of flux controlling enzymes grating in silico and in vivo fluxes to a priori predict a multi-
[22,23] or interruption of the competing glycolytic pathway target combination for the desired production properties
[24]. Complementary, also increased supply of oxaloace- (Figure 2). The only 12 predicted traits were then imple-
tate [25,26] and recently, also attenuation of the competing mented stepwise into the genome of the non-producing
TCA cycle flux was shown beneficial [27–29]. All these wild type, creating the stable, completely defined minimal

www.sciencedirect.com Current Opinion in Biotechnology 2012, 23:718–726

720 Tissue, cell and pathway engineering

Figure 2

Glucose

Psod
tkt tal zwf opcA pgl tkt-operon
Glucose 6-P Pentose 5-P
zwf pgl
tkt
Fructose 6-P tkt

fbp Peftu Erythrose 4-P

fbp
tal
Fructose 1,6-BP
Sedoheptulose 7-P

Dihydroxyacetone P Glyceraldehyde 3-P

Psod C 932 T
lysC lysC
G1A
L-Aspartatyl-P L-Aspartate
hom
Ppck
PhomT 176 C Phosphoenolpyruvate Δpck pck
hom

L-Threonine Aspartate semialdehyde

Acetyl-CoA
Pyruvate Pyruvate

Oxaloacetate
L-2,3-Dihydrodipicolinate
PsodC 1372 T
dapB Psod
dapB
pyc

pyc
L-Δ1-Tetrahydrodipicolinate Isocitrate
Malate
Picd A 1G
ddh icd icd
Pddh 2-Oxoglutarate
ddh ddh

meso-Diaminopimelate
PlysA lysA
lysA lysA

L-Lysine

L-Lysine ex
Current Opinion in Biotechnology

Systems-wide metabolic pathway engineering of the rational lysine hyper-producing C. glutamicum LYS-12 strain by genome-based implementation
of 12 genomic traits in the non-producing wild type C. glutamicum ATCC 13032 [11]. Targeted genes comprise dapB, encoding dihydrodipicolinate
reductase, ddh, ecoding diaminopimelate dehydrogenase, fbp, encoding fructose 1,6-bisphosphatase, hom, encoding homoserine dehydrogenase,
icd, encoding isocitrate dehydrogenase, lysA, encoding diaminopimelate decarboxylase, lysC, encoding aspartokinase, pck, encoding
phosphoenolpyruvate carboxykinase, pgl, encoding 6-phosphogluconolactonase, pyc, encoding pyruvate carboxylase, tal, encoding transaldolase,
tkt, encoding transketolase and zwf, encoding glucose 6-phosphate dehydrogenase. Details on the modifications are given in the gray boxes. Bold
arrows indicate reactions with increased metabolic fluxes whereby green reactions represent direct amplification targets. Gene deletion or attenuation
is shown in red. ‘X’ denotes gene deletion.

Current Opinion in Biotechnology 2012, 23:718–726 www.sciencedirect.com

 Systems and synthetic metabolic engineering for amino acid production Becker and Wittmann 721

mutation strain C. glutamicum LYS-12. Under industrial of 2.1 g L 1 h 1 and a yield of 0.3 g g 1 [36]. Shortly,
conditions on a molasses based medium, C. glutamicum global strain engineering comprised modification of the
LYS-12 exhibited remarkable production properties. This biosynthetic valine pathway by pathway deregulation and
includes a high titer of 120 g L 1, reached within 30 h of plasmid-based overexpression (Figure 3b). Carbon with-
cultivation. The carbon conversion yield of 0.55 g g 1 is drawal towards isoleucine was abolished by ilvA deletion.
already rather close to the theoretical optimum predicted This was complemented by global regulatory changes
for zero growth (Figure 1). The high vitality of the mini- introduced by plasmid-based over expression of the tran-
mally modified strain, with a specific glucose uptake rate as scriptional regulator Lrp and deletion of lacI to allow
high as that of the parent wild type, finally enabled a constitutive expression of the overexpressed genes under
productivity of 4.0 g L 1 h 1. With these properties, control of the trc or tac promoters. Finally, also the valine
LYS-12 is the first synthetic strain that competes with export was improved by over expression of the corre-
classical producers optimized for more than 50 years, sponding exporter YgaZH. Alternatively, also C. glutami-
showing the enormous potential of systems-wide pathway cum was modified by rational strategies [37–39]. This also
engineering. generated interesting valine overproducers, which so far
do not reach the performance of E. coli strains, but display
L-Threonine interesting candidates to be optimized further.
L-Threonine is currently produced by fermentation of
classical Escherichia coli strains [4,5]. Beyond early studies L-Glutamate
on rational engineering of E. coli by targeted over expres- The leading amino acid on the global market is the flavor
sion of the biosynthetic threonine gene cluster cloned enhancer monosodium glutamate (Figure 1). Rational
from a threonine producing mutant [32], a genetically engineering of producing C. glutamicum mainly focused
defined threonine producer was recently constructed by on the improved supply of the glutamate precursor 2-
systems metabolic engineering on basis of the wild type oxoglutarate. This involved targeted flux increase
E. coli W3110 [33]. Hereby, systems-wide profiling, through anaplerosis by over expression of pyruvate
integrating data from genome, transcriptome, proteome carboxylase [26] or by deletion of competing pyruvate
and in silico flux response analysis, provided key charac- kinase [40]. Beyond this, production was engineered by
teristics of threonine producers [34]. First steps involved reducing the undesired 2-oxoglutare consumption via
engineering of the threonine pathway by release from oxoglutarate dehydrogenase on the basis of antisense
feedback inhibition, removal of transcriptional attenu- RNA expression of odhA [41] or over expression of the
ation, and concomitant plasmid-based over expression ODHC-inhibitor odhI [42]. The yields and titers achieved
(Figure 3a). This was combined with deletion or down- by these targeted strategies are still far from that of
regulation of competing pathways towards lysine, meth- classical strains used in industrial production that
ionine, glycine, isoleucine, and acetate and the accumulate more than 100 g L 1 [2]. Also the yields of
abolishment of threonine re-uptake. To account for the the synthetic strains are still rather low. A systems meta-
increased demand for oxaloacetate in the threonine pro- bolic engineering strategy, as described above for other
ducer, the expression of anaplerotic PEP carboxylase was products, seems promising to bring this further.
fine-tuned by promoter exchange. Further benefit was
taken from deletion of the transcriptional regulator iclR Other L-amino acids
controlling expression of the glyoxylate shunt [33]. The Beyond the major products L-lysine, L-threonine, L-glu-
final strain produced 82 g L 1 threonine with a maximum tamate and L-valine metabolic engineering was continu-
productivity of 1.7 g L 1 h 1 and a carbon conversion ously applied to establish or improve also other amino
yield of 0.39 g g 1. Further modifications seem necessary acid production processes [2] as summarized in Figure 1.
to bring the production performance closer to that of Interesting production strains for L-serine [43], L-cysteine
currently applied classical strains and the theoretical [44], L-arginine [45,46], L-citrulline [45], L-ornithine [47]
optimum (Figure 1). Metabolic engineering was also and L-alanine [48,49] could be derived in recent years. For
applied to extend the natural substrate range of E. coli. the aromatic amino acids, L-tryptophan, L-tyrosine and L-
Implementation of b-fructofuranosidase mediated phenylalanine, targeted engineering was even crucial to
sucrose-assimilation and threonine production with a establish commercially competitive fermentation pro-
yield of 0.28 g g 1 [35]. cesses [8]. However, production performance is still far
away from what can be expected as theoretical optimum
L-Valine (Figure 1).
For the production of L-valine, strain engineering has
been carried out for both, C. glutamicum and E. coli. L- Targeted engineering also provided a basic L-methionine
valine biosynthesis branches off from the central metab- producer, where classical approaches so far failed [50].
olite pyruvate. A comprehensive approach recently cre- Though the production is still rather low, first systems-
ated a strain with impressive production properties, oriented studies demonstrated the feasibility of a fermen-
including a high titer of 61 g L 1 valine a productivity tative production process and employed in-silico pathway

www.sciencedirect.com Current Opinion in Biotechnology 2012, 23:718–726

722 Tissue, cell and pathway engineering

Figure 3

(a) Glucose

P
thrA thrB thrC

rhtC
Glucose 6-P Pentose 5-P
pBRThrABCR3

Ap
rhtA Fructose 6-P
rhtB rop

Glyceraldehyde 3-P
thrA lysC
C 1034 T C 1055 T

L-Aspartatyl-P L-Aspartate Acetate

ppc
Phosphoenolpyruvate
lysA ΔlysA P
ppc P
acs
L-Lysine Aspartate semialdehyde
Acetyl-CoA acs
thrA P Pyruvate
thrABC
metA ΔmetA
thrABC Oxaloacetate
L-Methionine L-Homoserine

thrB
Glyoxylate
L-Homoserine-P Isocitrate
ilvA ilvA Malate icIR
C 290 T
thrC
L-Isoleucine ΔicIR

tdh Δtdh
Succinate
Glycine L-Threonine tdcC
ΔtdcC

L-Threonine ex

(b) Glucose

Glucose 6-P Pentose 5-P

lacl
Fructose 6-P
Δlacl

Glyceraldehyde 3-P

Phosphoenolpyruvate ilvED
L-Isoleucine ilvBN ilvC

ilvBNmut pKBRilvBN mutCED

ilvA ΔilvA Pyruvate 2-Acetolactate
ilvIH ilvC Ap
L-Threonine Oxaloacetate
+
2,3-Dihydroxyisovalerate

ilvD

2-Ketoisovalerate
P
ygaZH lrp Lrp ilvE

pTrc184ygaZHIrp
- L-Valine
Cm
+ P
ygaZH ygaZH
ilvJ

L-Valine ex
Current Opinion in Biotechnology

Current Opinion in Biotechnology 2012, 23:718–726 www.sciencedirect.com

 Systems and synthetic metabolic engineering for amino acid production Becker and Wittmann 723

modeling and well-designed experiments to predict glutamicum [60]. In E. coli putrescine is a natural metab-
promising targets for further engineering such as the olite and part of the ornithine degrading pathway.
promising use of higher reduced sulfur sources for pro- Hereby, accumulation and secretion of putrescine was
duction [1,51]. initiated by over expression of the putrescine-forming
ornithine decarboxylase speC combined with deletion of
Related products and non-natural chemicals putrescine and ornithine consuming reactions [59].
The rising era of bio-based economy is currently driving
the engineering of cells for production of various chemi- Conclusion
cals beyond traditional products [2]. This also holds for Current advances in systems and synthetic metabolic
the industrial amino acid producers, C. glutamicum and E. engineering illustrate the power of these global concepts
coli. Prominent products with high industrial relevance for industrial strain development. For existing industrial
are diamines, meaning that they are important building processes, it seems, however, not easy to approach the
blocks for bio-based polyamides with excellent material high performance of the classical strains [2] or even the
properties [52]. Particularly, cadaverine (1,5-diaminopen- theoretical boundaries of optimum production (Figure 1).
tane) and putrescine (1,4-diaminobutane) appear very Probably except for the most advanced example of lysine
promising to replace current petrochemical routes to production [11], the production properties of rationally
the huge market of polyamides. Cadaverine is directly created amino acid producers created so far leave still
accessible via decarboxylation of lysine, suggesting lysine space for substantial improvement in the future
hyper-producing C. glutamicum as straightforward pro- (Figure 1). We can expect a next level of strains upon
duction organism. The key enzyme lysine decarboxylase integration of newly developed strategies into systems
is naturally not present in C. glutamicum, making heter- and synthetic metabolic engineering. This might involve
ologous expression a crucial modification to establish the use of reduced-genome microorganism that was
production in this organism [53]. Hereby, the introduc- recently shown to breed a superior threonine producing
tion of high-expression synthetic genes constructed on strain of E. coli [61]. Overall, the tremendously growing
basis of the ldcC gene from E. coli provided the most toolbox for metabolic engineers will further speed up
suited strategy to boost cadaverine production [54]. development times – and probably increase the heartbeat
Systems metabolic engineering of basic producers then of industrial strain development. In addition, also the
included amplification and deregulation of the biosyn- integration of new strategies from model-based design,
thetic pathway at various levels, improved supply of the synthetic biology or evolutionary engineering will con-
carbon precursor oxaloacetate by targeted modification of tribute to better production strains. This will enable the
the pyruvate node and downregulation of the competing crossing of boundaries, set by the natural gene repertoire
pathway towards threonine [54]. Production was further of specific species and drive the creation of novel syn-
optimized by deletion of a newly discovered pathway thetic pathways for the use of a broad range of raw
towards the undesired by-product N-acetyl-cadaverine materials [2,35,57,62–64] and the production of a huge
that allowed by-product free cadaverine production at a portfolio of non-natural compounds [2,54,65–67]. At one
yield of 0.17 g g 1 [55]. More recently, transcriptome- future time point, amino acids might display only a small
based identification of a putative cadaverine exporter subset of products obtained by C. glutamicum or E. coli.
enabled a next level of production strains [56]. Beyond They will, however, always remain as the cradle of this
this, the substrate range for cadaverine production was fascinating development.
extended towards xylose allowing sustainable production
from hemicellulose-based raw material [57]. In addition,
cadaverine production was also established for E. coli as Acknowledgements
production host that naturally comprises the entire bio- The authors gratefully acknowledge support by the German Federal
Ministry of Education and Research (BMBF) through the grant ‘Bio-based
synthetic pathway [58]. Similarly, also putrescine pro- Polyamides through Fermentation’ (No. 0315239A) within the initiative
duction was demonstrated for E. coli [59] and C. Bioindustry21.

( Figure 3 Legend ) Metabolic engineering strategy for the creation of a genetically defined threonine production strain (a) on basis of E. coli K-12 [33]
and a genetically defined valine producing strain (b) on the basis of E. coli W [36]. Genetic modifications are indicated in the gray boxes or on the
plasmid. Targeted genes comprise acs, encoding acetyl-CoA synthetase, iclR, encoding transcriptional regulator IclR, ilvA, encoding threonine
dehydratase, ilvBN, ecoding acetohydroxy acid synthase I, ilvC acetohydroxy acid isomeroreductase, ilvD, encoding dihydoxy acid dehydratase, ilvE,
encoding branched chain amino acid aminotransferase, ilvIH, encoding acetohydroxy acid synthase III, ilvJ, encoding valine transporter, lacI, encoding
transcriptional regulator LacI, lrp, encoding transcriptional regulator Lrp, lysA, encoding diaminopimelate decarboxylase, lysC, encoding aspartokinase
III, metA, encoding homoserine succinyltransferase, ppc, encoding phopsphoenolpyruvate carboxylase, rhtABC, encoding threonine exporters, tdh,
encoding threonine dehydrogenase, tdcC, encoding threonine transporter, thrA, encoding aspartokinase I, thrB, encoding homoserine kinase, thrC,
encoding threonine synthase and ygaZH, encoding valine exporter. Bold arrows indicate reactions with increased carbon flow whereby green
reactions represent direct amplification targets (by promoter exchange and/or by plasmid-based over expression). Gene deletion or attenuation is
shown in red. ‘X’ denotes gene deletion. Gray arrows indicates induction (+) or repression ( ). The abbreviation mut specifies the implementation of six
specific point mutations G59A, C60T, T62A, A63C, A64T, and G66C into the ilvN gene [74].

www.sciencedirect.com Current Opinion in Biotechnology 2012, 23:718–726

724 Tissue, cell and pathway engineering

References and recommended reading miniaturized scale. Biotechnol Bioeng 2004, 87:1-6.
Papers of particular interest, published within the period of review, 16. Kiefer P, Heinzle E, Zelder O, Wittmann C: Comparative
have been highlighted as:  metabolic flux analysis of lysine-producing Corynebacterium
glutamicum cultured on glucose or fructose. Appl Environ
 of special interest Microbiol 2004, 70:229-239.
 of outstanding interest Together with [20] this paper highlights the value of metabolic flux
analysis for knowledge-based strain engineering. Based on comparative
1. Bolten CJ, Schröder H, Dickschat J, Wittmann C: Towards flux studies of C. glutamicum on different substrates, fructose 1,6-
methionine overproduction in Corynebacterium glutamicum – bisphosphatase was predicted as truly novel target towards improved
methanethiol and dimethyldisulfide as reduced sulfur lysine production, allowing targeted improvement of lysine production by
sources. J Microbiol Biotechnol 2010, 20:1196-1203. flux rerouting.

2. Becker J, Wittmann C: Bio-based production of chemicals, 17. Marx A, de Graaf A, Wiechert W, Eggeling L, Sahm H:
materials and fuels – Corynebacterium glutamicum as Determination of the fluxes in the central metabolism of
versatile cell factory. Curr Opin Biotechnol 2011; DOI:10.1016/ Corynebacterium glutamicum by Nuclear Magnetic
j.copbio.2011.11.012, in press. Resonance Spectroscopy combined with metabolite
balancing. Biotechnol Bioeng 1996, 49:111-129.
3. Becker J, Klopprogge C, Wittmann C: Metabolic responses to
pyruvate kinase deletion in lysine producing Corynebacterium 18. Marx A, Striegel K, de Graaf A, Sahm H, Eggeling L: Response of
glutamicum. Microb Cell Fact 2008, 7:8. the central metabolism of Corynebacterium glutamicum to
different flux burdens. Biotechnol Bioeng 1997, 56:168-180.
4. Lee SY, Park JH: Integration of systems biology with
bioprocess engineering: L-threonine production by systems 19. Petersen S, de Graaf AA, Eggeling L, Möllney M, Wiechert W,
metabolic engineering of Escherichia coli. In Biosystems Sahm H: In vivo quantification of parallel and bidirectional
Engineering I. Edited by Wittmann C, Krull R.. In Adv Biochem Eng fluxes in the anaplerosis of Corynebacterium glutamicum. J
Biotechnol, vol 120. Edited by Scheper T Adv Biochem Eng Biol Chem 2000, 275:35932-35941.
Biotechnol. vol 120Springer; 2010:1-19. 20. Becker J, Heinzle E, Klopprogge C, Zelder O, Wittmann C:
5. Dong X, Quinn PJ, Wang X: Metabolic engineering of  Amplified expression of fructose 1,6-bisphosphatase in
Escherichia coli and Corynebacterium glutamicum for the Corynebacterium glutamicum increases in vivo flux through the
production of L-threonine. Biotechnol Adv 2010, 29:11-23. pentose phosphate pathway and lysine production on different
carbon sources. Appl Environ Microbiol 2005, 71:8587-8596.
6. Kalinowski J, Cremer J, Bachmann B, Eggeling L, Sahm H, Together with [16] this paper highlights the value of metabolic flux
Pühler A: Genetic and biochemical analysis of the analysis for knowledge-based strain engineering. Based on comparative
aspartokinase from Corynebacterium glutamicum. Mol flux studies of C. glutamicum on different substrates, fructose 1,6-
Microbiol 1991, 5:1197-1204. bisphosphatase was predicted as truly novel target towards improved
lysine production, allowing targeted improvement of lysine production by
7. Moritz B, Striegel K, De Graaf AA, Sahm H: Kinetic properties of flux rerouting.
the glucose-6-phosphate and 6-phosphogluconate
dehydrogenases from Corynebacterium glutamicum and their 21. Georgi T, Rittmann D, Wendisch VF: Lysine and glutamate
application for predicting pentose phosphate pathway flux in production by Corynebacterium glutamicum on glucose,
vivo. Eur J Biochem 2000, 267:3442-3452. fructose and sucrose: roles of malic enzyme and fructose-1,6-
bisphosphatase. Metab Eng. 2005, 7:291-301.
8. Ikeda M: Towards bacterial strains overproducing L-
tryptophan and other aromatics by metabolic engineering. 22. Becker J, Klopprogge C, Herold A, Zelder O, Bolten CJ, Wittmann C:
Appl Microbiol Biotechnol 2006, 69:615-626. Metabolic flux engineering of L-lysine production in
Corynebacterium glutamicum-over expression and
9. Krömer JO, Sorgenfrei O, Klopprogge K, Heinzle E, Wittmann C: modification of G6P dehydrogenase. J Biotechnol 2007,
In-depth profiling of lysine-producing Corynebacterium 132:99-109.
glutamicum by combined analysis of the transcriptome,
metabolome, and fluxome. J Bacteriol 2004, 186:1769-1784. 23. Ohnishi J, Katahira R, Mitsuhashi S, Kakita S, Ikeda M: A novel
gnd mutation leading to increased L-lysine production in
10. Kohlstedt M, Becker J, Wittmann C: Metabolic fluxes and Corynebacterium glutamicum. FEMS Microbiol Lett 2005,
beyond – systems biology understanding and engineering of 242:265-274.
microbial metabolism. Appl Microbiol Biotechnol 2010,
88:1065-1075. 24. Marx A, Hans S, Möckel B, Bathe B, de Graaf AA: Metabolic
phenotype of phosphoglucose isomerase mutants of
11. Becker J, Zelder O, Häfner S, Schröder H, Wittmann C: From zero Corynebacterium glutamicum. J Biotechnol 2003, 104:185-197.
 to hero – design-based systems metabolic engineering of
Corynebacterium glutamicum for L-lysine production. Metab 25. Petersen S, Mack C, de Graaf AA, Riedel C, Eikmanns BJ, Sahm H:
Eng 2011, 13:159-168. Metabolic consequences of altered phosphoenolpyruvate
This paper for the first time reports a synthetic lysine hyper-producing carboxykinase activity in Corynebacterium glutamicum reveal
strain of C. glutamicum that can compete with the classical industrial anaplerotic regulation mechanisms in vivo. Metab Eng 2001,
strains from 50 years of optimization and displays a benchmark in strain 3:344-361.
engineering for amino acid production. Integrating only 12 genomic traits,
predicted by computational design, the non-producing wild type was 26. Peters-Wendisch PG, Schiel B, Wendisch VF, Katsoulidis E,
successfully converted into an efficient cell factory for the important bulk Möckel B, Sahm H, Eikmanns BJ: Pyruvate carboxylase is a
product lysine. major bottleneck for glutamate and lysine production by
Corynebacterium glutamicum. J Mol Microbiol Biotechnol 2001,
12. Wittmann C: Analysis and engineering of metabolic pathway 3:295-300.
fluxes in Corynebacterium glutamicum. Adv Biochem Eng
Biotechnol 2010, 120:21-49. 27. Becker J, Klopprogge C, Schröder H, Wittmann C: Metabolic
engineering of the tricarboxylic acid cycle for improved lysine
13. Wittmann C, Heinzle E: Genealogy profiling through strain production by Corynebacterium glutamicum. Appl Environ
improvement by using metabolic network analysis: metabolic Microbiol 2009, 75:7866-7869.
flux genealogy of several generations of lysine-producing
Corynebacteria. Appl Environ Microbiol 2002, 68:5843-5859. 28. Becker J, Buschke N, Bücker R, Wittmann C: Systems level
engineering of Corynebacterium glutamicum –
14. Wittmann C, Kiefer P, Zelder O: Metabolic fluxes in reprogramming translational efficiency for superior
Corynebacterium glutamicum during lysine production with production. Eng Life Sci 2010, 10:430-438.
sucrose as carbon source. Appl Environ Microbiol 2004,
70:7277-7287. 29. Blombach B, Schreiner ME, Moch M, Oldiges M, Eikmanns BJ:
Effect of pyruvate dehydrogenase complex deficiency on L-
15. Wittmann C, Kim HM, Heinzle E: Metabolic network analysis of lysine production with Corynebacterium glutamicum. Appl
lysine producing Corynebacterium glutamicum at a Microbiol Biotechnol 2007, 76:615-623.

Current Opinion in Biotechnology 2012, 23:718–726 www.sciencedirect.com

 Systems and synthetic metabolic engineering for amino acid production Becker and Wittmann 725

30. Ohnishi J, Mitsuhashi S, Hayashi M, Ando S, Yokoi H, Ochiai K, 46. Xu M, Rao Z, Xu H, Lan C, Dou W, Zhang X, Xu H, Jin J, Xu Z:
Ikeda M: A novel methodology employing Corynebacterium Enhanced production of L-arginine by expression of
glutamicum genome information to generate a new L-lysine- Vitreoscilla hemoglobin using a novel expression system in
producing mutant. Appl Microbiol Biotechnol 2002, 58:217-223. Corynebacterium crenatum. Appl Biochem Biotechnol 2011,
163:707-719.
31. Ikeda M, Ohnishi J, Hayashi M, Mitsuhashi S: A genome-based
approach to create a minimally mutated Corynebacterium 47. Lee YJ, Cho JY: Genetic manipulation of a primary metabolic
glutamicum strain for efficient L-lysine production. J Ind pathway for L-ornithine production in Escherichia coli.
Microbiol Biotechnol 2006, 33:610-615. Biotechnol Lett 2006, 28:1849-1856.
32. Miwa K, Tsuchida T, Kurahashi O: Construction of L-threonine 48. Jojima T, Fujii M, Mori E, Inui M, Yukawa H: Engineering of sugar
overproducing strains of Escherichia coli K-12 using metabolism of Corynebacterium glutamicum for production of
recombinant DNA techniques. Agric Biol Chem 1983, amino acid L-alanine under oxygen deprivation. Appl Microbiol
47:2329-2334. Biotechnol 2010, 87:159-165.
33. Lee KH, Park JH, Kim TY, Kim HU, Lee SY: Systems metabolic 49. Zhang X, Jantama K, Moore JC, Shanmugam KT, Ingram LO:
 engineering of Escherichia coli for L-threonine production. Production of L-alanine by metabolically engineered
Mol Syst Biol 2007, 3:149. Escherichia coli. Appl Microbiol Biotechnol 2007, 77:355-366.
This paper impressively reports the creation of a genetically defined
threonine producing E. coli strain through systems metabolic engineering. 50. Park SD, Lee JY, Sim SY, Kim Y, Lee HS: Characteristics
of methionine production by an engineered Corynebacterium
34. Lee JH, Lee DE, Lee BU, Kim HS: Global analyses of glutamicum strain. Metab Eng 2007, 9:327-336.
transcriptomes and proteomes of a parent strain and an L-
threonine-overproducing mutant strain. J Bacteriol 2003, 51. Krömer JO, Wittmann C, Schröder H, Heinzle E: Metabolic
185:5442-5451. pathway analysis for rational design of L-methionine
production by Escherichia coli and Corynebacterium
35. Lee JW, Choi S, Park JH, Vickers CE, Nielsen LK, Lee SY: glutamicum. Metab Eng 2006, 8:353-369.
Development of sucrose-utilizing Escherichia coli K-12 strain
by cloning beta-fructofuranosidases and its application for L- 52. Kind S, Wittmann C: Bio-based production of the platform
threonine production. Appl Microbiol Biotechnol 2010, chemical 1,5-diaminopentane. Appl Microbiol Biotechnol 2011,
88:905-913. 91:1287-1296.

36. Park JH, Jang YS, Lee JW, Lee SY: Escherichia coli W as a new 53. Mimitsuka T, Sawai H, Hatsu M, Yamada K: Metabolic
 platform strain for the enhanced production of L-valine by engineering of Corynebacterium glutamicum for cadaverine
systems metabolic engineering. Biotechnol Bioeng 2011, fermentation. Biosci Biotechnol Biochem 2007,
108:1140-1147. 71:2130-2135.
This work uses the concept of systems metabolic engineering to create a
genetically defined valine producing strain on basis of the less frequently 54. Kind S, Jeong WK, Schröder H, Wittmann C: Systems-wide
used E. coli strain W. Engineering mainly focused on the terminal pathway  metabolic pathway engineering in Corynebacterium
and on the deletion of competing reactions. glutamicum for bio-based production of diaminopentane.
Metab Eng 2010, 12:341-351.
37. Radmacher E, Vaitsikova A, Burger U, Krumbach K, Sahm H,
Eggeling L: Linking central metabolism with increased 55. Kind S, Jeong WK, Schröder H, Zelder O, Wittmann C:
pathway flux: L-valine accumulation by Corynebacterium  Identification and elimination of the competing N-
glutamicum. Appl Environ Microbiol 2002, 68:2246-2250. acetyldiaminopentane pathway for improved production of
diaminopentane by Corynebacterium glutamicum. Appl
38. Holatko J, Elisakova V, Prouza M, Sobotka M, Nesvera J, Patek M: Environ Microbiol 2010, 76:5175-5180.
Metabolic engineering of the L-valine biosynthesis pathway in These two papers demonstrate the extension of amino acid production
Corynebacterium glutamicum using promoter activity pathways towards non-natural chemicals. Fermentation of C. gluta-
modulation. J Biotechnol 2009, 139:203-210. micum that was metabolically engineered for the production of the
industrial monomer cadaverine opens access to high-value polya-
39. Bartek T, Blombach B, Zönnchen E, Makus P, Lang S, mides. The synthetic cell factory was streamlined by systems-wide
Eikmanns BJ, Oldiges M: Importance of NADPH supply for pathway engineering.
improved L-valine formation in Corynebacterium glutamicum.
Biotechnol Prog 2010, 26:361-371. 56. Kind S, Kreye S, Wittmann C: Metabolic engineering of cellular
transport for overproduction of the platform chemical 1,5-
40. Sawada K, Zen-in S, Wada M, Yokota A: Metabolic changes in a diaminopentane in Corynebacterium glutamicum. Metab Eng
pyruvate kinase gene deletion mutant of Corynebacterium 2011, 13:617-627.
glutamicum ATCC 13032. Metab Eng 2010, 12:401-407.
57. Buschke N, Schröder H, Wittmann C: Metabolic engineering of
41. Kim J, Hirasawa T, Sato Y, Nagahisa K, Furusawa C, Shimizu H: Corynebacterium glutamicum for production of 1,5-
Effect of odhA overexpression and odhA antisense RNA diaminopentane from hemicellulose. Biotechnol J 2011,
expression on Tween-40-triggered glutamate production by 6:306-317.
Corynebacterium glutamicum. Appl Microbiol Biotechnol 2009,
81:1097-1106. 58. Qian ZG, Xia XX, Lee SY: Metabolic engineering of Escherichia
coli for the production of cadaverine: a five carbon diamine.
42. Kim J, Fukuda H, Hirasawa T, Nagahisa K, Nagai K, Wachi M, Biotechnol Bioeng 2011, 108:93-103.
Shimizu H: Requirement of de novo synthesis of the OdhI
protein in penicillin-induced glutamate production by 59. Qian ZG, Xia XX, Lee SY: Metabolic engineering of Escherichia
Corynebacterium glutamicum. Appl Microbiol Biotechnol 2009, coli for the production of putrescine: a four carbon diamine.
86:911-920. Biotechnol Bioeng 2009, 104:651-662.
43. Peters-Wendisch P, Stolz M, Etterich H, Kennerknecht N, Sahm H, 60. Schneider J, Wendisch VF: Putrescine production by
Eggeling L: Metabolic engineering of Corynebacterium engineered Corynebacterium glutamicum. Appl Microbiol
glutamicum for L-serine production. Appl Environ Microbiol Biotechnol 2010, 88:859-868.
2005, 71:7139-7144.
61. Lee JH, Sung BH, Kim MS, Blattner FR, Yoon BH, Kim JH, Kim SC:
44. Nakamori S, Kobayashi SI, Kobayashi C, Takagi H: Metabolic engineering of a reduced-genome strain of
Overproduction of L-cysteine and L-cystine by Escherichia Escherichia coli for L-threonine production. Microb Cell Fact
coli strains with a genetically altered serine acetyltransferase. 2009, 8:2.
Appl Environ Microbiol 1998, 64:1607-1611.
62. Kawaguchi H, Sasaki M, Vertes AA, Inui M, Yukawa H:
45. Ikeda M, Mitsuhashi S, Tanaka K, Hayashi M: Reengineering of a Engineering of an L-arabinose metabolic pathway in
Corynebacterium glutamicum L-arginine and L-citrulline Corynebacterium glutamicum. Appl Microbiol Biotechnol 2008,
producer. Appl Environ Microbiol 2009, 75:1635-1641. 77:1053-1062.

www.sciencedirect.com Current Opinion in Biotechnology 2012, 23:718–726

726 Tissue, cell and pathway engineering

63. Tsuchidate T, Tateno T, Okai N, Tanaka T, Ogino C, Kondo A: 69. Leuchtenberger W, Huthmacher K, Drauz K: Biotechnological
Glutamate production from beta-glucan using production of amino acids and derivatives: current
endoglucanase-secreting Corynebacterium glutamicum. Appl status and prospects. Appl Microbiol Biotechnol 2005,
Microbiol Biotechnol 2011, 90:895-901. 69:1-8.
64. Seibold G, Auchter M, Berens S, Kalinowski J, Eikmanns BJ: 70. Krömer JO, Bolten CJ, Heinzle E, Schröder H, Wittmann C:
Utilization of soluble starch by a recombinant Physiological response of Corynebacterium glutamicum to
Corynebacterium glutamicum strain: growth and lysine oxidative stress induced by deletion of the transcriptional
production. J Biotechnol 2006, 124:381-391. repressor McbR. Microbiology 2008, 154:3917-3930.
65. Smith KM, Cho KM, Liao JC: Engineering Corynebacterium 71. Chinen A, Kozlov YI, Hara Y, Izui H, Yasueda H: Innovative
glutamicum for isobutanol production. Appl Microbiol metabolic pathway design for efficient L-glutamate
Biotechnol 2010, 87:1045-1055. production by suppressing CO2 emission. J Biosci Bioeng 2007,
103:262-269.
66. Dekishima Y, Lan EI, Shen CR, Cho KM, Liao JC: Extending
carbon chain length of 1-butanol pathway for 1-hexanol 72. Melzer G, Esfandabadi ME, Franco-Lara E, Wittmann C: Flux
synthesis from glucose by engineered Escherichia coli. J Am Design: in silico design of cell factories based on correlation of
Chem Soc 2011, 133:11399-11401. pathway fluxes to desired properties. BMC Syst Biol 2009,
3:120.
67. Boghigian BA, Salas D, Ajikumar PK, Stephanopoulos G, Pfeifer
BA: Analysis of heterologous taxadiene production in K- and 73. Varma A, Boesch BW, Palsson BO: Biochemical production
B-derived Escherichia coli. Appl Microbiol Biotechnol 2011; capabilities of Escherichia coli. Biotechnol Bioeng 1993, 42:59-73.
DOI:10.1007/s00253-011-3528-4, in press.
74. Park JH, Kim TY, Lee KH, Lee SY: Fed-batch culture of
68. Ikeda M: Amino acid production processes. Adv Biochem Eng Escherichia coli for L-valine production based on in silico flux
Biotechnol 2003, 79:1-35. response analysis. Biotechnol Bioeng 2011, 108:934-946.

Current Opinion in Biotechnology 2012, 23:718–726 www.sciencedirect.com