emerging corynebacterium glutamicum systems biology corynebacterium... · emerging corynebacterium...

Journal of Biotechnology 124 (2006) 74–92

Emerging Corynebacterium glutamicum systems biology

Volker F. Wendisch a,∗, Michael Bott a, Jorn Kalinowski c,Marco Oldiges b, Wolfgang Wiechert d

a Institute of Biotechnology 1, Research Center Juelich, D-52425 Juelich, Germanyb Institute of Biotechnology 2, Research Center Juelich, Germanyc Institute for Genome Research, University of Bielefeld, Germany

d Department of Simulation, Institute of Systems Engineering, University of Siegen, Germany

Received 8 July 2005; received in revised form 12 October 2005; accepted 1 December 2005

Abstract

Corynebacterium glutamicum is widely used for the biotechnological production of amino acids. Amino acid producingstrains have been improved classically by mutagenesis and screening as well as in a rational manner using recombinant DNAtechnology. Metabolic flux analysis may be viewed as the first systems approach to C. glutamicum physiology since it com-bines isotope labeling data with metabolic network models of the biosynthetic and central metabolic pathways. However, onlythe complete genome sequence of C. glutamicum and post-genomics methods such as transcriptomics and proteomics haveallowed characterizing metabolic and regulatory properties of this bacterium on a truly global level. Besides transcriptomicsand proteomics, metabolomics and modeling approaches have now been established. Systems biology, which uses systematicgcfo©

KT

1

a

f

0

enomic, proteomic and metabolomic technologies with the final aim of constructing comprehensive and predictive models ofomplex biological systems, is emerging for C. glutamicum. We will present current developments that advanced our insight intoundamental biology of C. glutamicum and that in the future will enable novel biotechnological applications for the improvementf amino acid production.

2005 Elsevier B.V. All rights reserved.

eywords: Systems biology; White biotechnology; Corynebacterium glutamicum; Amino acid production; Strain development; Genomics;ranscriptomics; DNA microarray; Proteomics; Metabolic flux analysis; Metabolite profiling; Metabolomics; Mathematical modeling

. Introduction

Many fields of biology including biotechnologyre experiencing a shift towards systems-level charac-

∗ Corresponding author. Tel.: +49 2461 615169;ax: +49 2461 612710.

E-mail address: [email protected] (V.F. Wendisch).

terizations. Systems-level characterizations are estab-lished e.g. in developmental biology, ecology orimmunology, but in molecular biology they onlybecame possible by the success of the genome sequenc-ing efforts: the sequencing of whole genomes ratherthan single genes or operons allowed for the firsttime unraveling the complete genetic information ofan organism. On the other hand, the elucidation of

168-1656/$ – see front matter © 2005 Elsevier B.V. All rights reserved.doi:10.1016/j.jbiotec.2005.12.002

V.F. Wendisch et al. / Journal of Biotechnology 124 (2006) 74–92 75

Fig. 1. Integration of C. glutamicum systems biology approaches with examples of their unique contributions to the characterization of metabolicand regulatory network topologies.

biochemical pathways, their regulatory principles suchas feedback-inhibition and the metabolic networkanalysis under nonequilibrium conditions are well-established in biochemistry and physiology. Today, theprospect of systems biology based on molecular enti-ties excites researchers from all disciplines of biology.

What is systems biology? Currently, no generallyaccepted definition of systems biology exists. However,there is general agreement that systems biology (Fig. 1)determines the complete repertoire of genes (genome),mRNAs (transcriptome), proteins (proteome), metabo-lites (metabolome), fluxes of an organism or cell aswell as their interactions across all levels based onglobal analysis techniques and integrates these data intopredictive models (Aderem, 2005; Bork and Serrano,2005; Hood et al., 2004; Kirschner, 2005; Liu, 2005;Westerhoff and Palsson, 2004). Systematic perturba-tions by genetic means or by exposition to environ-mental stimuli are performed to assess the robustnessof a cellular system and its modularity as well as todrive model refinement. Ultimately, emergent systemsproperties, i.e. those that cannot be extrapolated fromthe sum of the properties of all individual elements,have to be unraveled.

How do we approach systems biology conceptu-ally and experimentally? Physics has been invoked as

an example of how to build on the interdependenceand iterative refinement of experimentation, theory andtechnology (Liu, 2005). Theoretical models based onall available experimental data predicted unforeseensystems properties that could only be verified experi-mentally after advancing technology. Based on anoma-lies of a model of the solar system the existence of amissing planet had been predicted in the 18th centurybefore Neptune had been discovered, however, sys-tems analysis needs to mature further, both in physicsand in biology, to be successfully applied to very com-plex systems such as the weather or a living cell (Endyand Brent, 2001). Current systems biology approachesare mostly concentrated on collecting and integratingexperimental data and building first models, while thegap between simulation and prediction of unforeseensystems properties to be assessed experimentally stillneeds to be bridged.

Why Corynebacterium glutamicum systems biol-ogy? Systems biology of the bacterium C. glutamicummight be useful more generally for developing and test-ing concepts and in validating and sharpening exper-imental systems biology tools as e.g. this relativelysimple, single-celled microorganism lacks extensivecompartmentalization, thus reducing spatial complex-ity (Bork and Serrano, 2005). While in molecular

76 V.F. Wendisch et al. / Journal of Biotechnology 124 (2006) 74–92

biology tools and concepts had mainly been developedin viruses and bacteria before having been transferredto higher eukaryotes (Brock, 1999), systems biologyapproaches have been initiated in bacteria and highereukaryotes in parallel (Ideker et al., 2001) with the pos-sibility of mutual benefit.

Systems biology is of fundamental significance andat the same time holds the promise to lead to practicalinnovations in biotechnology, medicine and engineer-ing. Mostly the impact of systems biology on medicinee.g. on the drug discovery process is discussed (Hood etal., 2004; Hood and Perlmutter, 2004). Here, we focuson the prospects of systems biology approaches to thedevelopment of white biotechnology and in particularon the bacterium C. glutamicum. Since its discoveryas l-glutamate producing bacterium in 1957, C. glu-tamicum has become the work-horse of large-scaleproduction of amino acids by fermentation processes.The annual market growth for amino acids is esti-mated to be about 10% and currently the productionof amino acids in fermentative processes with C. glu-tamicum amounts to 1,500,000 t of l-glutamate peryear, 550,000 t of l-lysine per year while amounts of l-glutamine (2000 t per year) and branched-chain aminoacids are lower (Hermann, 2003). Due to its importancein amino acid production, C. glutamicum has been sub-ject of physiological, biochemical and genetic studiesfor four decades and a number of well-defined fermen-tation processes based on this bacterium are in place. Inaddition, C. glutamicum is classified as a ‘generally rec-oiddtCmo

2

omips

These two genome sequences allowed the evaluationof the sequencing quality by sequence comparison andresequencing of all differences found before releasingthe second sequence (acc. no. BX927147). None of thepoint mutations that were detected inside of 61 codingregions was found to be a sequencing error (data notshown) revealing that both genome sequences are ofoutstanding quality.

However, both genome sequences differ by the pres-ence or absence of insertion element copies and a puta-tive prophage (Kalinowski, 2005) showing that thesegenetic elements can change the genome within shorttime. In addition, the ATCC 13032 strain differs frommany C. glutamicum strains as it does not possess an S-layer (Hansmeier et al., 2004). Another C. glutamicumstrain, named “strain R”, displayed even larger differ-ences and was found to contain a number of additionalgenetic islands (Suzuki et al., 2005).

A complete genome sequence is brought to lifeby annotation, supported by powerful bioinformat-ics tools, either for predicting genes, operons andother signals in DNA, or finding homologous proteinsin databases and predicting their cellular localizationas well as their function. Especially, gene predic-tion is critical since a complete gene set is essentialfor the post-genomic methods transcriptome and pro-teome analysis. Gene prediction tools should be runin combinations using intrinsic and extrinsic methods(McHardy et al., 2004) and also in this case the remain-ing “intergenic” regions should be carefully inspectedflogs

ssiftcCjid1g

gnized as safe’ (GRAS) organism. Therefore, it lendstself as an ideal host not only for the established pro-uction of amino acids, but also for the production ofiverse compounds ranging from organic acids to pro-eins. To fully exploit the biotechnological potential of. glutamicum and for its development as a platformicroorganism for white biotechnology systems biol-

gy approaches may prove helpful and even essential.

. Genomics

The determination of the whole genome sequencef the C. glutamicum wild-type strain ATCC 13032arked an important breakthrough. Due to its outstand-

ng industrial interest a number of competing genomerojects led to the publication of two complete genomeequences (Ikeda et al., 2003; Kalinowski et al., 2003).

or small genes which often escape prediction. Table 1ists a collection of genes that were annotated basedn strong bioinformatics evidence in the BX927147enome sequence, but not in the NC003450 genomeequence.

The comparison of genome sequences of relatedpecies is of advantage for prediction of codingequences since mutations accumulate less frequentlyn coding regions than in non-coding regions. There-ore, the C. glutamicum genome can be compared withhat of the closely related amino acid producer, C. effi-iens (Nishio et al., 2003), and those of the pathogens. diphtheriae (Cerdeno-Tarraga et al., 2003) and C.

eikeium (Tauch et al., in press). Especially, the compar-son to C. efficiens is of high value, since both genomesiffer in the G + C content of their DNAs by around0%, although the encoded protein sequences and theene order are very highly conserved (Nakamura et


Table 1C. glutamicum genes annotated in BX927147 but not in BA000036/NC003450 and their coding evidence

Locus tag Genename

Gene product Length(aa)

Blastp hit toa; evalue Tblastn vs.C. efficiensb

Remark

cg0449 Hypothetical protein 210 CE0390; 5.0e−13cg0494 Extremely conserved hypothetical

protein33 CE0433; 1.3e−13

cg0757 Conserved hypothetical protein 67 DIP1807; 4.3e−15cg0759 prpD2 Propionate catabolism protein PrpD 504 NCgl0627, NCgl0628 Frameshift in the

NC003450 genomecg0869 Hypothetical protein predicted by

Glimmer/Critica127 CE0775; 6.1e−37

cg0877 rsrA Putative anti-sigma factor 89 DIP0710, 2e−15 Cdcg1000 Conserved hypothetical protein 66 CE0952, 3.3e−19cg1070 Conserved hypothetical protein 63 CE1009, 2.3e−28cg1172 Putative plasmid maintenance killer

protein93 PP1586, 1e−24

cg1241 Hypothetical protein predicted byGlimmer/Critica

87 CE1147, 1.4e−09

cg1542 Putative membrane protein 82 CE1479, 4.7e−13cg1704 Bacterial regulatory proteins; ArsR

family119 CE0874, 3.6e−34

cg1741 Putative membrane protein 84 CE1667, 8.5e−23cg1943 Hypothetical protein predicted by

Glimmer/Critica123 CE0505, 2.3e−41

cg1972 Putative translation elongation factor(GTPase)

111 COG4108

cg2029 Hypothetical protein predicted byGlimmer

253 COG5340

cg2325 Conserved hypothetical protein 43 COG0456 2.6e−18cg2525 Putative membrane protein 46 8.2e−20cg2526 Putative secreted or membrane

protein68 CE2202, 3.6e−18

cg2555 Conserved hypothetical protein 37 1.3e−12cg2791 rpmJ Ribosomal protein L36 40 CE2426, 2.4e−17cg3008 porA Porin 45 Lichtinger et al. (1998)cg3009 porH Porin 57 CE2560, 6e−20 Hunten et al. (2005)cg3077 Putative membrane protein 93 COG5158cg3117 cysX Ferredoxin-like protein 82 CE2643, 5e−43cg3257 Conserved hypothetical protein 46 2.2e−12cg3273 Hypothetical protein predicted by

Glimmer98 COG1917

cg3430 Conserved hypothetical protein 99 CE2945, 9.2e−46cg3433 Hypothetical protein predicted by

Glimmer/Critica73 CE0052, 1.8e−17

a Similarity to proteins from C. glutamicum NC003450 (NCgl), C. efficiens (CE), C. diphtheriae (DIP), and Pseudomonas putida (PP) or tothe Clusters of Orthologous Groups of proteins (COG).

b Similarity to the translated nucleotide sequence of the C. efficiens YS-314 genome (BA000035).

al., 2003). This evolutionary pressure for mutation tohigher G + C content very efficiently uncovered the truegene starts as well as putative regulator binding sitesand even promoters.

Using a combination of bioinformatics tools, cross-genome comparisons revealed corynebacterial gene

families as exemplified for the genes for 127 differentDNA-binding transcriptional regulators in the genomesequence of C. glutamicum which could be groupedinto 25 different families (Brune et al., 2005). It turnedout that only 28 regulators were conserved between allfour corynebacterial genomes, apparently representing


essential regulators that are controlling cell division,stress response, metal homeostasis as well as carbohy-drate and sulfur metabolism.

Although prediction of gene functions and of reg-ulatory signals is of outstanding importance whenanalysing thousands of genes within a genome, predic-tions often are only indicative of a gene’s affiliation to adistinct gene family or of the cellular localization of theencoded protein. In addition, as these annotations areassociated with a probability score which often givesno clear prediction, functional analysis of predictedgenes by experimentation may become essential. Intro-ducing a defined deletion into the chromosome whichremoves the gene of interest is often necessary to pre-cisely identify its function, especially when multiplecandidates for a given function must be assessed. Onthe other hand, physiological characterization of strainsoverexpressing a predicted gene or biochemical charac-terization of purified proteins may be useful. For thesepurposes, C. glutamicum is well suited since efficientgene replacement and overexpression techniques arewell developed (Kirchner and Tauch, 2003).

By the combination of predictions in silico withfunctional analyses by gene deletion or overexpressionand phenotypic characterizations, complex metabolicand regulatory pathways can be reconstructed, therebyproviding a basis for flux and metabolome analy-ses. Reconstruction of the methionine biosynthesis(Ruckert et al., 2003), the prediction and functionalanalysis of transaminase genes (McHardy et al., 2003),tfteea

ctdTOtitayt

approach was extended further by introducing addi-tional mutations (Georgi et al., in press; Ohnishi et al.,2005).

The ultimate goal of these bioinformatic and func-tional analyses is a fully and deeply annotated genomesequence. Due to the speed with which mutations canbe generated and with which high-throughput genomicand post-genomic analyses can be performed, this goalmight be in reach within the not so far future. A fullyannotated genome will then no longer be a simple“inventory” but might serve as “blueprint” for SystemsBiology approaches as well as for rational strain devel-opment in industrial production processes.

3. Transcriptomics

DNA microarray technology for C. glutamicum hasbeen established and has been applied to transcriptomeanalysis (reviewed in Wendisch (2003)), to genomecomparisons (Stansen et al., in press) and in medicaldiagnostics (Roth et al., 2004). Many studies publishedin 2004 and 2005 employed transcriptome analysisfor the characterization of diverse aspects of C. glu-tamicum physiology: regulation of sulfur (Rey et al.,2005), nitrogen (Silberbach et al., 2005), phosphorus(Rittmann et al., in press; Wendisch and Bott, 2005)and carbon metabolism (Gerstmeir et al., 2004; Netzeret al., 2004a; Stansen et al., in press), glutamate pro-duction triggered by ethambutol addition (RadmacherepcerslaaTi(

oslon

he analysis of six putative trehalose mycolyltrans-erase genes (Brand et al., 2003) as well as that ofhe three pathways of trehalose biosynthesis (Wolft al., 2003) and those of sulfonate and sulfonatester degradation are examples of such combinednalyses.

Another example is the use of metabolic models inonjunction with genome comparisons between a wild-ype and a producer strain originally established by ran-om mutagenesis and selection for higher production.he prototype of such experiments was described byhnishi et al. (2002), who identified mutations within

hree genes known to be involved in l-lysine productionn a classically derived producer strain. By introducinghese mutations into the C. glutamicum wild-type strainstrain was obtained which displayed a high l-lysineield while retaining a high vitality and thereby leadingo an increased productivity (Ohnishi et al., 2002). This

t al., 2005) or by a temperature-shift (Stansen et al., inress), lysine production (Kromer et al., 2004), serineatabolism (Netzer et al., 2004b), magnesium (Pallerlat al., 2005) and iron availability (Krug et al., 2005),egulation by the two-component signal transductionystem MtrA-MtrB (Moker et al., 2004), by the regu-ator of Clp protease gene expression ClgR (Engels etl., 2005; Engels et al., 2004) and by the regulator ofconitase gene expression AcnR (Krug et al., 2005).ranscriptome analysis of a spontaneous C. glutam-

cum mutant allowed mapping the mutated gene locusNetzer et al., 2004a).

While transcriptome analyses helped to advanceur knowledge about C. glutamicum in each of thesetudies, the characterizations of stimulons and regu-ons are most important from a systems level pointf view and are a cornerstone to define the regulatoryetwork topology (Wendisch, 2005). While a stimulon


comprises the group of genes that change expressionin response to a stimulus, e.g. nitrogen or phosphatestarvation (Ishige et al., 2003; Silberbach et al., 2005),a regulon comprises the group of genes controlled by atranscriptional regulator, e.g. by AcnR or McbR (Kruget al., 2005; Rey et al., 2005).

C. glutamicum responds to ammonium availabilityinvolving genetic regulation by the transcriptional reg-ulator AmtR (Burkovski, 2003). To identify the ammo-nium starvation stimulon it was important to distin-guish nitrogen-dependent from growth-rate dependenteffects as growth ceases when ammonium becomesscarce. This could be achieved by comparing twoammonium-limited continuous cultures differing bythe dilution rate with an ammonium-sufficient batchculture (Silberbach et al., 2005). The identification ofthe ammonium-starvation stimulon allowed unravelingthe physiological strategy of C. glutamicum to respondto the limiting availability of a nitrogen source. LikeE. coli (Zimmer et al., 2000), C. glutamicum scav-anges nitrogen from alternative sources when starvedfor ammonium as deduced from the increased expres-sion of genes for transport and metabolism of aminoacids, urea, etc. (Silberbach et al., 2005). In contrast,altered expression of e.g. genes for ribosomal pro-teins was a consequence of reduced growth rather thannitrogen limitation per se (Silberbach et al., 2005).Continuous cultures are useful to determine long-termadaptation to an environmental stimulus, but the kinet-ics of a stimulus response can only be determined insss2iiipi1hopWttsa

To characterize a regulon, a combination of tran-criptome analyses and bioinformatics with biochemi-cal and genetic experiments is often used. By globalgene expression analysis, McbR of C. glutamicumwas shown to control expression of 86 genes. Forty-five of these genes are organized in 22 operons thatshare a conserved motif in their promoter regions andcode for proteins involved in all aspects of transportand metabolism of the macroelement sulfur, i.e. (S-adenosyl)methionine and cysteine biosynthesis, sulfatereduction, uptake and utilization of sulfur-containingcompounds (Rey et al., 2005). Band-shift experimentsrevealed that S-adenosylhomocysteine, a product of S-adenosylmethionine-dependent transmethylation reac-tions, prevented the binding of McbR to the conservedsequence motif, thus providing a link between sulfurand C1 metabolism. As the McbR regulon also includestwo putative regulators and as at least a subset ofthe McbR regulon might be additionally controlled byother regulators, the hierarchy and network topology ofsulfur-dependent regulation remains to be elucidated.

The TetR-type transcriptional regulator AcnR wasshown to repress aconitase gene (acn) expression in C.glutamicum and DNA microarray analyses indicatedthat acn is its primary target gene (Krug et al., 2005).AcnR is active as a homodimer and binds to the acnpromoter in the region from −11 to −28 relative tothe transcription start. This AcnR binding motif andthe organization of its gene into an operon togetherwith acn is conserved in corynebacteria and mycobac-tscaotiiAiaabybtds

hift experiments combined with time-resolved tran-criptome analyses as demonstrated for the phosphatetarvation response of C. glutamicum (Ishige et al.,003; Wendisch and Bott, 2005). While in C. glutam-cum, but not in E. coli, expression of ushA encod-ng a secreted 5′-nucleotidase/UDP-sugar hydrolase isnduced by phosphate starvation (Rittmann et al., inress), the phosphate starvation stimulon of C. glutam-cum is in general similar to that of E. coli (Wanner,996) with the characteristic induction of operons forigh affinity uptake of inorganic phosphate and ofrganophosphate as well as of genes for secreted phos-hatases (Ishige et al., 2003; Rittmann et al., in press;endisch and Bott, 2005). Taken together, to charac-

erize stimulons as well as the dynamics of a responseo a stimulus it will prove useful to combine steady-tate analysis of continuous cultures with time-resolvednalysis of pulse experiments.

eria (Krug et al., 2005). As the activities of citrateynthase and isocitrate dehydrogenase are relativelyonstant on various carbon sources while aconitasectivities were 2.5–4-fold higher on propionate, citrate,r acetate than on glucose (Krug et al., 2005), it appearshat aconitase is a major control point and AcnR anmportant regulator of tricarboxylic acid cycle activityn C. glutamicum. Future studies will reveal whethercnR control is restricted to acn or whether its regulon

s more complex. To this end, so-called ChIP-to-chipnalyses (Buck and Lieb, 2004) will be helpful as theyllow identifying those regions of the genome that areound by a transcriptional regulator. Thus, these anal-ses identify the genes and operons directly regulatedy a transcriptional regulator and they facilitate consis-ent model building as direct regulatory effects can beistinguished from indirect effects propagated as con-equence of primary regulatory event(s).


4. Proteomics

Proteins fulfill many different functions. They arethe major catalysts of metabolism and key to regulatoryprocesses. Due to the fact that proteins are composed of20 different amino acids, their physicochemical prop-erties are highly diverse. Thus, their comprehensiveidentification and characterization is much more dif-ficult than that of nucleic acids.

The method currently used most frequently to estab-lish the protein composition of cells is two-dimensionalpolyacrylamide gel electrophoresis (2D-PAGE) andprotein identification by mass spectrometry. 2D-PAGEseparates proteins according to their isoelectric point(isoelectric focusing) and their apparent molecularmass (sodium dodecylsulfate-polyacrylamide gel elec-trophoresis). Protocols for 2D-PAGE of C. glutam-icum proteins have been established in recent years

(for review see Schaffer and Burkovski (2005)) andused to separate the proteins of different cell fractions(Fig. 2), i.e. soluble fraction, membrane fraction andsecreted proteins (Hermann et al., 1998, 2000, 2001). Inthese studies, 42 proteins were identified by N-terminalmicrosequencing, Western blotting with specific anti-sera or mass spectrometry. A first high-resolution ref-erence map for cytoplasmic and membrane-associatedproteins of glucose-grown C. glutamicum was estab-lished by using narrow immobilized pH gradients andtryptic peptide mass fingerprinting (Schaffer et al.,2001). In this work, 152 different proteins of the cyto-plasmic fraction were identified as well as 13 proteinsof the membrane fraction. Recently, the cytosolic, cellsurface and extracellular proteomes of C. efficiens cellsgrown Luria-Broth were analysed by 2D-PAGE and164, 49 and 89 proteins of the three fractions wereidentified by peptide mass fingerprinting, respectively.

F rding tU

ig. 2. Fractionation of C. glutamicum cells into cellular entities acco
C, ultracentrifugation; ASB-14, amidosulfobetaine 14; AIEC, anion excha
o Hermann et al. (2001) and Schlusener et al. (2005). Abbreviations:
nge chromatography.


For comparison, the corresponding fractions of C. glu-tamicum grown in Luria-Broth were also analysed andcompared to that of C. efficiens, revealing only minordifferences between the two species (Hansmeier et al.,in press). The latter study led to the identification of119 additional C. glutamicum proteins.

A major limitation of current 2D-PAGE is the failureto separate integral membrane proteins. For examplein the study by Schaffer et al. (2001), only proteinswith maximally one transmembrane helix were iden-tified in the membrane fraction. Recently, an alterna-tive method was established (Fig. 2) to separate andidentify intrinsic membrane proteins of C. glutamicum(Schlusener et al., 2005). It involves the initial sepa-ration of amidosulfobetaine-14-solubilized membraneproteins by anion exchange chromatography (AIEC)and subsequent separation of the individual AIEC frac-tions by SDS-PAGE followed by the identification ofindividual protein bands by peptide mass fingerprint-ing after digestion with trypsin and CNBr cleavage(Schlusener et al., 2005). Of the 170 different iden-tified proteins, 50 were membrane-integral with up to13 transmembrane helices. This number corresponds to7.5% of the predicted membrane proteome. Thus, thismethod appears to be suitable for the qualitative andpossibly also semiquantitative comparison of mem-brane proteomes.

Phosphorylation of proteins is a central regulatorymechanism both in eukaryotes and prokaryotes. In con-trast to phosphohistidine and phosphoaspartate, thertepcbCepgia(facnw

mass fingerprinting, 41 of the detected proteins wereidentified, most of which act in central metabolic path-ways such as glycolysis or the citric acid cycle. Thispoints to an important role of protein phosphorylationon serine and threonine residues (about 20 proteinsreacted with phosphothreonine-specific antibodies) inC. glutamicum’s regulatory processes. However, theexact residues phosphorylated in the different proteinsand the effects of phosphorylation on protein activityremain to be elucidated.

Regulation within the cell may occur on many dif-ferent levels, i.e. transcription, mRNA stability, trans-lation, protein stability, or protein activity. The firsttwo of these can be analysed at the RNA level withDNA microarrays, whereas the others require analy-sis at the protein level. 2D-PAGE was used to studythe effects of nitrogen limitation (Nolden et al., 2001;Silberbach et al., 2005), to identify proteins specifi-cally induced during growth on propionate (Claes etal., 2002), to compare the proteome of glucose- andacetate-grown cells (Gerstmeir et al., 2003), to anal-yse the influence of externally added valine (Lange etal., 2003), to determine proteins induced by heat shock(Barreiro et al., 2005), or to characterize mutants lack-ing regulatory subunits of the Clp protease, which ledto the identification of a novel transcriptional regula-tor ClgR controlling expression of clpP1P2, clpC andseveral additional genes (Engels et al., 2005; Engels etal., 2004).

Recent studies indicate that many cellular processesatinmtappvscHu(i

o

esidues involved in two-component signal transduc-ion by sensor kinases and response regulators (Mokert al., 2004), phosphoserine, phosphothreonine andhosphotyrosine residues are quite stable and proteinsarrying these phosphorylated residues can be analysedy 2D-PAGE. In a recent study, phosphorus-containing. glutamicum proteins were analysed using two differ-nt approaches. By in vivo radiolabeling using [33P]-hosphoric acid followed by 2D-PAGE and autoradio-raphy, about 60 labelled proteins were detected. Bymmunostaining of 2D gels with antibodies directedgainst phosphoserine, about 90 proteins were detectedBendt et al., 2003). The latter method is specificor phosphorylated amino acids, whereas the formerlso detects proteins modified by other phosphorus-ontaining groups, e.g. covalently linked flavine ade-ine dinucleotide. Thirty-one spots were detected bothith radiolabeling and immunostaining. By peptide

re carried out by multiprotein complexes. The iden-ification and analysis of their components providesnsight into how the protein repertoire of a cell is orga-ized into functional units. The currently preferredethod to identify interactions between individual pro-

eins on large scale are two-hybrid systems (Fieldsnd Song, 1989), whereas the composition of multi-rotein complexes is elucidated e.g. by tandem affinityurification (Gavin et al., 2002; Ho et al., 2002) of inivo tagged proteins. In C. glutamicum, no large scaletudies on protein–protein interaction and the identifi-ation of protein complexes have yet been performed.owever, in vivo tagging and affinity purification wassed to identify a cytochrome-bc1-aa3-supercomplexNiebisch and Bott, 2003), confirming that this methods applicable to C. glutamicum.

There is no doubt that proteins and thus all kindf studies on proteins are of central importance for a


holistic understanding of the cell. As is evident fromabove, proteome analysis of C. glutamicum is in anadvanced state, although protein identification has notyet been performed exhaustively for 2D gels or SDSgels of membrane proteins. Some key issues that haveto be solved in the future in order to obtain convincingmathematical models of metabolism and its regulationare quantification of intracellular protein concentra-tions and modifications and a far better knowledge ofregulation at the level of enzyme activity. The contribu-tion of in-depth biochemical and biophysical character-izations of individual proteins and enzymes to systemsbiology often is underestimated. However, as large-scale methods for thorough biochemical and biophysi-cal enzyme analyses are not yet available, the analysisof individual proteins is still an indispensable elementof systems biology. Moreover, it appears necessary tofocus at first instance on individual subsystems of thecell (e.g., glycolysis, citric acid cycle or biosynthesispathways) and to obtain reliable data for these, sincemodels can only be as good as the data they are basedon. Subsequently, the subsystems can be combined andintertwined to a complete picture of the cell.

5. Metabolomics and fluxomics

Metabolites function as substrates and productsof proteins as the major biocatalysts in the cell, butalso have important regulatory functions. Fructose-1,6-bitfrt2

crpuoslppm

tein translation to posttranslational modification (e.g.phosphorylation), by product and substrate metaboliteconcentrations and by allosteric control (inhibition oractivation of enzymes by other metabolites). Althoughthe in vitro enzyme properties can be extracted from adatabase (Schomburg et al., 2002) they do not correctlyreflect the enzyme properties in vivo in the presence ofthe whole metabolic network.

Metabolomics is currently gaining more and moreinterest because the metabolome is downstream of geneand protein regulation and thus adds valuable informa-tion not accessible by other methods, like the in vivostatus of allosteric control (Allen et al., 2003; Fiehnet al., 2000a). Metabolomics approaches have beendeveloped relatively late as low amounts of metabo-lites are present within the cells (3–5% dry cell weight(Neidhardt and Umbarger, 1996)) and their high degreeof chemical diversity complicates sample preparationand analytical procedures making it almost impossibleto separate and quantify all metabolites using a singleexperimental technique as is possible for RNA (DNA-chip) and proteins (2D-gel electrophoresis).

The metabolomics approach has been widelyapplied in plants (Fiehn et al., 2000a,b) but wasalso developed for microbial systems as E. coli(Chassagnole et al., 2002; Oldiges et al., 2004; Schaeferet al., 1999), S. cerevisiae. (Theobald et al., 1997;Visser et al., 2004) and B. subtilis (Soga et al.,2003). In addition to established GC–MS techniques(Schauer et al., 2005; Strelkov et al., 2004) morerpoaimat(asobnimcb

isphosphate, AMP or ATP e.g. modify enzyme activ-ties in C. glutamicum (reviewed in Eikmanns (2005))o allow the cell to adapt to environmental changes veryast. In addition, several metabolites function in geneticegulation by modulating the activities of e.g. transcrip-ional regulators like C. glutamicum McbR (Rey et al.,005).

Metabolomics deals with the quantification of intra-ellular metabolite concentrations while fluxomicsefers to the reaction velocities between metaboliteools in the biochemical network. Although both val-es (pool and flux) seem to be connected to eachther, no direct quantitative flux data can be derivedimply from the concentration of the reacting metabo-ites because the flux is also determined by the kineticroperties and amount of the corresponding enzymeresent in the cell. Hence, the in vivo flux may be deter-ined by genetic regulation from transcription to pro-

obust and sensitive analytical devices for LC–MS cou-ling became available and accelerate the applicationf metabolomics to diagnostics, functional genomicsnd systems biology (Fernie et al., 2004). Because ofts high impact as an industrial amino acid producingicroorganism C. glutamicum surprisingly played onlyminor role in this area so far, although some rela-

ive concentration data of the C. glutamicum wild typeStrelkov et al., 2004) and first quantitative data of anmino acid producing C. glutamicum strain were pre-ented (Magnus et al., 2005). From a systemic pointf view, quantitative metabolite data are advantageous,ecause kinetic models of the underlying biochemicaletwork can be derived from them, although the exper-mental effort to achieve absolute quantitative data of

etabolites at cytosolic level is much higher even inomplex biological sample matrices and a large num-er of metabolites.


According to Fiehn et al. (2000a), metabolome anal-ysis is multifunctional making use of different analyt-ical approaches depending on the topic of the studyand some examples of applications to C. glutamicumhave been described. Metabolome analysis can thus besubdivided into (i) target analysis, aiming at measur-ing distinct substrates or products e.g. to distinguishdifferences after genetic modification, (ii) metabolicprofiling (Magnus et al., 2005), focusing on a completepathway or linked pathways and (iii) metabolomics(Strelkov et al., 2004), striving for an unbiased, semi-quantitative overview of whole-cell metabolic patterns.For a more rapid analysis, metabolic fingerprinting(Fiehn, 2001) can be used, which reduces the ana-lytical effort to the analysis of those metabolites withbiochemical relevance. Quite recently, another variantof high-throughput classification was presented, calledmetabolic footprinting (Allen et al., 2003), which dif-fers from the aforementioned methods by analysingextracellular metabolites instead of intracellularpools.

In order to setup kinetic metabolic models, themetabolic profiling approach is preferred, becausequantitative metabolite data for a set of pathways(e.g. glycolysis, pentose-phosphate-pathway, TCA-cycle, amino acid pathway) are provided. For the func-tional genomics approach metabolomics or metabolicfingerprinting is well suited, because data of relativeconcentration changes for a high number of compoundscan contribute to the identification of the function ofaF

msmtf2dabpte

ia

may be explained by mass action law kinetics fol-lowed by the allosteric processes, controlling the activ-ity of enzymes. Changes on the genetic or the pro-teomic level are slower and the enzyme concentra-tions can be assumed to be constant for a few minutesafter exposition to a stimulus. Keeping this in mind,a stimulus-response experiment (Oldiges and Takors,2005) using a substrate step increase (glucose pulse),will show no effect for the genetic or proteomic prop-erties of metabolism within an observation window of1 or 2 min. Hence, the recorded metabolome responsewill only contain information about the enzyme kinet-ics and at the same time minimizing the superim-posed effects of gene and protein regulation. Thesestimulus-response experiments can serve as a datasource to setup dynamic kinetic models (Wiechert,2002) and to identify and quantify kinetic enzymeparameters/effectors in order to describe the biochemi-cal network on a systemic level. Current developmentsaim at the use of 13C-labelled glucose as pulse sub-strate instead of naturally labelled glucose (Mashegoet al., 2004). This will have the advantage that thestimulus shift (step increase) from glucose limitationto glucose saturation will be replaced by a stimulusshift from 13C limitation to 13C saturation, circumvent-ing the physiological switch from glucose limitation tosaturation.

The use of 13C-labelled substrates (e.g. glucose) isa general technique for the determination of the flux-ome, i.e. all intracellular metabolic fluxes within thesadmtldti1

tawtmD((

gene product in mutant strains (Fernie et al., 2004;iehn et al. 2000a; Raamsdonk et al., 2001).

Unlike transcriptome and proteome analysis,etabolome analysis has to be very fast (sub-second

cale) to monitor physiological changes in the environ-ent of the cell. Hence, quenching of metabolic activi-

ies for metabolomics studies has to be even faster thanor transcriptome and proteome analysis (Lange et al.,001). Several automated techniques allowing repro-ucible high-frequency sampling (Fig. 3) (Buziol etl., 2002; Schaefer et al., 1999; Visser et al., 2004) haveeen applied mainly for E. coli and S. cerevisiae, butreliminary results for an l-valine producing C. glu-amicum strain have been presented recently (Magnust al., 2005).

Roels (1983) ranked the relaxation times follow-ng a particular stimulus for several cellular responsesccording to their time scale. The fastest responses

toichiometric biochemical network. Metabolic fluxnalysis was the first approach to C. glutamicum strainevelopment with a systems level focus since it requiresetabolic network models of the biosynthetic and cen-

ral carbon metabolic pathways. The investigation of-lysine production by C. glutamicum was a majorriving force for the development of this 13C labellingechnology, although it is not limited to a certain biolog-cal species. After several hours of feeding a mixture of3C-labelled and unlabelled glucose to a growing cul-ure, the 13C labelling information has been distributednd equilibrated in the stoichiometric metabolic net-ork and accumulated in the amino acids of the pro-

eome. After protein hydrolysis, 13C information iseasured via GC–MS (Christensen and Nielsen, 1999;auner and Sauer, 2000) or more detailed using NMR

Marx et al., 1996). The 13C metabolic flux analysis13C MFA) uses the determined extracellular reaction


Fig. 3. Experimental setup for stimulus-response experiments used by (Schaefer et al., 1999). Automated rapid sampling unit in the lowerforeground.

rates and the 13C amino acid data to calculate the intra-cellular fluxes in the network (Wiechert, 2001a). Thismethod has been intensively used for flux measure-ments to improve l-lysine production with C. glutam-icum and revealed the importance of NADPH supply(Marx et al., 1996, 1999, 2003, 1997; Wittmann andHeinzle, 2002), of different anaplerotic pathways anda metabolic futile cycle in anaplerosis (Petersen et al.,2000, 2001; Peters-Wendisch et al., 1996; Wendischet al., 2000). A series of multiple flux determinationsduring a fed-batch process for l-lysine production withC. glutamicum was described using the sensor–reactorconcept (Drysch et al., 2003; El Massaoudi et al., 2003).In combination with a rapid sampling device (samplingfrequency < 1 Hz) such concept will provide sample

material for intracellular metabolome and transient 13Cmetabolome analysis using mass spectrometric meth-ods. This combination of metabolome and 13C fluxomeanalysis is no longer limited to growth-coupled sys-tems and will presumably allow faster access to theequilibrated 13C labelling information in the metabolicpathways of C. glutamicum under various growth andproduction conditions.

Quantitative metabolome and fluxome data are thebasis for a quantitative analysis of the metabolic net-work properties in C. glutamicum with mathematicalmodelling tools. In combination with semi-quantitativemetabolomics data to support functional genomics theycontribute to the systems-based investigation of C. glu-tamicum.


6. Mathematical modeling

Due to its integrative nature, modeling has become akey activity of any research project in systems biology.In general, mathematical modeling intends to representthe structure and behavior of a real system by means ofmathematical concepts. For example structural graphs,signal flow diagrams, equation systems, ordinary andpartial differential equations, logical rules, or proba-bilistic relations have been used in the past to describebiological systems (Wiechert, 2002).

Mathematics is a universal discipline, and thusmodel based methods developed for one specific organ-ism can be carried over to a similar system if the under-lying structures and data are modified appropriately.This means that tools developed for prokaryotic organ-isms like E. coli or B. subtilis will probably also applyto C. glutamicum. Likewise, methods from eukaryoticsingle cells, like S. cerevisiae, have good chances tobe adapted as long as cellular compartmentation andthe details of genetic regulation do not play a centralrole. On the other hand methods for plant and animalcells typically become more and more different fromprokaryotic models because additionally intracellulardiffusion and intra/intercellular signaling are becom-ing important.

It should be distinguished between a mathematicalmodel as an abstract system representation, mathemat-ical methods applicable to a certain class of modelsand the concrete implementations of these methodsahsmvtfitl

ttt2l(m

2001) is now well developed. Designed 13C label-ing experiments (Mollney et al., 1999) could evenresolve entangled fluxes like the different anapleroticas well as reverse fluxes (Petersen et al., 2000,2001) catalysed by PEP carboxylase (Eikmanns et al.,1989), PEP carboxykinase (Riedel et al., 2001) andpyruvate carboxylase (Peters-Wendisch et al., 1998,1997). Genetic (Peters-Wendisch et al., 2001; Riedelet al., 2001) and a series of flux analyses of differ-ent Corynebacterium strains (Petersen et al., 2000,2001) revealed a futile cycle having a strong influ-ence on lysine production. Current developments in13C MFA aim at an application to strain compari-son (Shirai et al., 2005), instationary process con-ditions (Drysch et al., 2004; Wiechert and Noh,2005), miniaturized reactors (Wittmann et al., 2004)and high throughput methods for metabolic screen-ing (Fischer et al., 2004; Wittmann and Heinzle,2002).

Flux balance analysis (FBA) (Edwards et al., 2002)and metabolic network analysis (MNA) (Schilling etal., 1999) are also based on stoichiometry and areconcerned with the exploration of the space of feasi-ble fluxes that is given by the stoichiometric relationsand reaction irreversibility assumptions. MNA tries toobtain a general picture of the metabolic flexibility ofan organism to adapt to different external conditionsby computing the elementary flux modes (Stelling etal., 2002) as well as to give information concerningthe importance of single reactions (Klamt and Gilles,2iratFlflmaVmbt

Ftt

s model based software tools. Mathematical toolsave been developed for graphical and textual modelpecification, stationary and instationary simulation,athematical systems analysis, experimental model

alidation, model based prediction and in silico sys-ems optimization (Jong, 2002; Wiechert, 2002). Theollowing examples show where C. glutamicum hasn recent years been the subject of theoretical inves-igations and tool development on different abstractionevels.

Metabolic flux analysis (MFA) aiming at the quan-itative determination of metabolic fluxes certainly ishe method that most profited from systemic ques-ions on the Corynebacterium metabolism (de Graaf,000; Vallino and Stephanopoulos, 1993). The under-ying mathematics for the classical stoichiometric MFAHeijden et al., 1994) and for the more powerful but alsoore complicated 13C MFA (Wiechert and Wurzel,

004). Although a genome-wide model of C. glutam-cum is currently not available, it will certainly beanked in between the very flexible E. coli (Reed etl., 2003) and the rather rigidly constrained B. sub-ilis (Fischer and Sauer, 2005; Stelling et al., 2004).BA is focused on how the metabolic flux map must

ook like to optimize one single, freely chosen targetux. Unfortunately, the theoretically computed opti-um for lysine production by C. glutamicum is still far

way from the real productivity (Stephanopoulos andallino, 1991). Obviously, more constraints like ther-odynamic restrictions (Mavrovouniotis, 1993) must

e introduced to obtain better estimates of the produc-ivity limits.

All stoichiometry based methods (MFA, MNA,BA) are limited because they are assuming a sta-

ionary metabolic state whereas the regulation ofhe pathways is completely left out. However, more


complex mechanistic models for stationary and insta-tionary states based on enzyme kinetic relations(Heinrich and Schuster, 1996) allow a detailed anal-ysis of cellular regulation, e.g. single branch pointsof the Corynebacterium metabolism (Vallino andStephanopoulos, 1994a,b). Sensitivity based methodslike metabolic control theory are widely used estab-lished tools (Kell and Westerhoff, 1986). However,mechanistic models must rely on many biologicalassumptions and consequently, their prediction mustalways be taken with care (Wiechert, 2004). As a com-promise between stoichiometric and regulatory mod-els a multi scale model of the anaplerotic reactionsof Corynebacterium in the context of global stoi-chiometry (Petersen et al., 2003) has been developedand validated with a series of stationary experiments.A Bayesian approach for parameter estimation andrigorous error estimation of this model (von Lieresand Wiechert, 2004) taking all sources of errors intoaccount showed that the model can predict the effectof several genetic modifications in the anaplerosis witha reasonable precision (Petersen et al., 2003; Riedel etal., 2001; Peters-Wendisch et al., 1998).

The investigation of metabolism under highlydynamic conditions has been driven forward by thedevelopment of stimulus-response experiments and, inparticular, rapid sampling techniques (see above andFig. 3). The goal of stimulus-response experimentsusually is to verify hypotheses on pathway regulation,to obtain more insight into the regulatory structureoeitteCateb

otttat

field, mainly because the required level of model detailis not clear (Jong, 2002). It is still under discussion towhat extent gene transcription, RNA degradation, genetranslation, protein phosphorylation or protein turnovermust be taken into account. Also, the available mea-surements related to genetic and proteomic processesare at best semi quantitative which strongly hampersmodel validation.

As a consequence, phenomenological and statisti-cal approaches correlating transcriptome, proteome,fluxome and metabolome data currently dominate thefield (Bro and Nielsen, 2004). DNA microarray datawas interpreted in the context of the metabolic net-work structure (Oh and Liao, 2000), used to reconstructgene regulatory signals by using network componentanalysis (Liao et al., 2003), to obtain further con-straints on metabolic fluxes (Akesson et al., 2004) orwas mapped to physiological states (Stephanopouloset al., 2002). For C. glutamicum, combinations of tran-scriptome with proteome analyses have been used toidentify targets to improve valine production (Langeet al., 2003) and to characterize nitrogen regulation(Silberbach et al., 2005), acetate regulation (Gerstmeiret al., 2003) or regulation by the Clp protease (Engelset al., 2005; Engels et al., 2004). While the phaseshift from growth to lysine production was studiedby combined transcriptome, metabolome and fluxomeanalyses (Kromer et al., 2004), serine utilizing andpantothenate producing C. glutamicum strains wereanalysed using combined transcriptome and fluxomea

7

rsrmotatoCmh

f single pathways, to find hitherto unknown enzymeffectors, or to determine enzyme kinetic parametersn vivo. High performance computational methods hado be developed and many different model variants hado be considered in the modeling process (Haunschildt al., 2005). Rapid sampling of a l-valine producing. glutamicum strain (Magnus et al., 2005) revealeds a preliminary result of model based data evaluationhat the acetolactate synthase (IlvBN) and the l-valinexport show a high degree of flux control in the l-valineiosynthesis pathway.

Clearly, there cannot be a systemic understandingf a microorganism or even parts of it without takinghe genetic and protein regulation into account. One ofhe goals of C. glutamicum systems biology must behe quantitative prediction of the influence of geneticlterations on the metabolic fluxes. However, the quan-itative modeling of genetic regulation is still an open

nalyses (Huser et al., 2005; Netzer et al., 2004b).

. Conclusion

Although systems biology approaches have onlyecently been established for C. glutamicum, impres-ive progress has since been made especially withespect to fundamental insight into principles ofetabolic regulation. Current and future efforts focus

n the integration of quantitative data from genomics,ranscriptomics, proteomics, metabolomics and fluxnalysis to build and evaluate metabolic and regula-ory models of C. glutamicum. Eventually, the potentialf these approaches for the rational improvement of. glutamicum strains for amino acid production and,ore broadly, for white biotechnology will fully be

arnessed.


Acknowledgements

The support of the Bundesministerium furBildung und Forschung (BMBF) through grant031U113D/031U213D is gratefully acknowledged.The project was partially funded by the DeutscheForschungsgemeinschaft (DFG) through grants(TA241/2-1, TA 241/2-2).

References

Aderem, A., 2005. Systems biology: its practice and challenges. Cell121, 511–513.

Akesson, M., Forster, J., Nielsen, J., 2004. Integration of gene expres-sion data into genome-scale metabolic data. Metab. Eng. 6,285–293.

Allen, J., Davey, H.M., Broadhurst, D., Heald, J.K., Rowland, J.J.,Oliver, S.G., Kell, D.B., 2003. High-throughput classification ofyeast mutants for functional genomics using metabolic footprint-ing. Nat. Biotechnol. 21, 692–696.

Barreiro, C., Gonzalez-Lavado, E., Brand, S., Tauch, A., Martin, J.F.,2005. Heat shock proteome analysis of wild-type Corynebac-terium glutamicum ATCC 13032 and a spontaneous mutant lack-ing GroEL1, a dispensable chaperone. J. Bacteriol. 187, 884–889.

Bendt, A.K., Burkovski, A., Schaffer, S., Bott, M., Farwick, M.,Hermann, T.C., 2003. Towards a phosphoproteome map ofCorynebacterium glutamicum. Proteomics 3, 1637–1646.

Bork, P., Serrano, L., 2005. Towards cellular systems in 4D. Cell121, 507–509.

Brand, S., Niehaus, K., Puhler, A., Kalinowski, J., 2003. Identifi-cation and functional analysis of six mycolyltransferase genesof Corynebacterium glutamicum ATCC 13032: the genes cop1,

coupled rapid stopped-flow sampling technique for measure-ments of metabolite dynamics on a subsecond time scale.Biotechnol. Bioeng. 80, 632–636.

Cerdeno-Tarraga, A.M., Efstratiou, A., Dover, L.G., Holden, M.T.,Pallen, M., Bentley, S.D., Besra, G.S., Churcher, C., James, K.D.,De Zoysa, A., Chillingworth, T., Cronin, A., Dowd, L., Feltwell,T., Hamlin, N., Holroyd, S., Jagels, K., Moule, S., Quail, M.A.,Rabbinowitsch, E., Rutherford, K.M., Thomson, N.R., Unwin,L., Whitehead, S., Barrell, B.G., Parkhill, J., 2003. The completegenome sequence and analysis of Corynebacterium diphtheriaeNCTC13129. Nucl. Acids Res. 31, 6516–6523.

Chassagnole, C., Noisommit-Rizzi, N., Schmid, J.W., Mauch, K.,Reuss, M., 2002. Dynamic modeling of the central carbonmetabolism of Escherichia coli. Biotechnol. Bioeng. 79, 53–73.

Christensen, B., Nielsen, J., 1999. Isotopomer analysis usingGC–MS. Metab. Eng. 1, 282–290.

Claes, W.A., Puhler, A., Kalinowski, J., 2002. Identification of twoprpDBC gene clusters in Corynebacterium glutamicum and theirinvolvement in propionate degradation via the 2-methylcitratecycle. J. Bacteriol. 184, 2728–2739.

Dauner, M., Sauer, U., 2000. GC–MS analysis of amino acids rapidlyprovides rich information for isotopomer balancing. Biotechnol.Progr. 16, 642–649.

de Graaf, A.A., 2000. Metabolic flux analysis of Corynebacteriumglutamicum. In: Schugerl, K., Bellgardt, K.H. (Eds.), BioreactionEngineering—Modeling and Control. Springer, Chapter 4.2.

Drysch, A., El Massaoudi, M., Mack, C., Takors, R., de Graaf, A.A.,Sahm, H., 2003. Production process monitoring by serial map-ping of microbial carbon flux distributions using a novel SensorReactor approach. II. (13)C-labeling-based metabolic flux anal-ysis and l-lysine production. Metab. Eng. 5, 96–107.

Drysch, A., El Massaoudi, M., Wiechert, W., de Graaf, A.A., Takors,R., 2004. Serial flux mapping of Corynebacterium glutamicumduring fed-batch l-lysine production using the sensor reactorapproach. Biotechnol. Bioeng. 85, 497–505.

E

E

E

E

E

E

dwards, J., Covert, M., Palsson, B.O., 2002. Metabolic modellingof microbes: the flux-balance approach. Environ. Microbiol. 4,133–140.

ikmanns, B., 2005. Central metabolism: tricarboxylic acid cycle andanaplerotic reactions. In: Eggeling, L., Bott, M. (Eds.), Handbookon Corynebacterium glutamicum. CRC Press, Boca Raton, USA,pp. 241–276.

ikmanns, B.J., Follettie, M.T., Griot, M.U., Sinskey, A.J., 1989. Thephosphoenolpyruvate carboxylase gene of Corynebacterium glu-tamicum: molecular cloning, nucleotide sequence, and expres-sion. Mol. Gen. Genet. 218, 330–339.

l Massaoudi, M., Spelthahn, J., Drysch, A., de Graaf, A., Takors,R., 2003. Production process monitoring by serial mappingof microbial carbon flux distributions using a novel sensorreactor approach. I. Sensor reactor system. Metab. Eng. 5,86–95.

ndy, D., Brent, R., 2001. Modelling cellular behaviour. Nature 409,391–395.

ngels, S., Ludwig, C., Schweitzer, J.E., Mack, C., Bott, M., Schaffer,S., 2005. The transcriptional activator ClgR controls transcriptionof genes involved in proteolysis and DNA repair in Corynebac-terium glutamicum. Mol. Microbiol. 57, 576–591.

cmt1, and cmt2 can replace each other in the synthesis of tre-halose dicorynomycolate, a component of the mycolic acid layerof the cell envelope. Arch. Microbiol. 10, 10.

Bro, C., Nielsen, J., 2004. Impact of ‘ome’ analysis in inversemetabolic engineering. Metab. Eng. 6, 204–211.

Brock, T.D., 1999. Milestones in Microbiology. ASM Press, Wash-ington, D.C.

Brune, I., Brinkrolf, K., Kalinowski, J., Puehler, A., Tauch,A., 2005. The individual and common repertoire of DNA-binding transcriptional regulators of Corynebacterium glutam-icum, Corynebacterium efficiens, Corynebacterium diphtheriaeand Corynebacterium jeikeium deduced from the completegenome sequences. BMC Genom. 6, 86.

Buck, M.J., Lieb, J.D., 2004. ChIP-chip: considerations for thedesign, analysis, and application of genome-wide chromatinimmunoprecipitation experiments. Genomics 83, 349–360.

Burkovski, A., 2003. Ammonium assimilation and nitrogen controlin Corynebacterium glutamicum and its relatives: an example fornew regulatory mechanisms in actinomycetes. FEMS Microbiol.Rev. 27, 617–628.

Buziol, S., Bashir, I., Baumeister, A., Claassen, W., Noisommit-Rizzi, N., Mailinger, W., Reuss, M., 2002. New bioreactor-


Engels, S., Schweitzer, J.E., Ludwig, C., Bott, M., Schaffer, S.,2004. clpC and clpP1P2 gene expression in Corynebacteriumglutamicum is controlled by a regulatory network involv-ing the transcriptional regulators ClgR and HspR as wellas the ECF sigma factor sigmaH. Mol. Microbiol. 52, 285–302.

Fernie, A.R., Trethewey, R.N., Krotzky, A.J., Willmitzer, L., 2004.Innovation—metabolite profiling: from diagnostics to systemsbiology. Nat. Rev. Mol. Cell Biol. 5, 763–769.

Fiehn, O., 2001. Combining genomics, metabolome analysis, andbiochemical modelling to understand metabolic networks. Com-par. Funct. Genom. 2, 155–168.

Fiehn, O., Kopka, J., Dormann, P., Altmann, T., Trethewey, R.N.,Willmitzer, L., 2000a. Metabolite profiling for plant functionalgenomics. Nat. Biotechnol. 18, 1157–1161.

Fiehn, O., Kopka, J., Trethewey, R.N., Willmitzer, L., 2000b. Iden-tification of uncommon plant metabolites based on calcula-tion of elemental compositions using gas chromatography andquadrupole mass spectrometry. Anal. Chem. 72, 3573–3580.

Fields, S., Song, O.K., 1989. A novel genetic system to detectprotein–protein interactions. Nature 340, 245–246.

Fischer, E., Sauer, U., 2005. Large-scale in vivo fluxes reveal rigid-ity and suboptimal performance of B. subtilis metabolism. Nat.Genet. 37, 636–640.

Fischer, E., Zamboni, N., Sauer, U., 2004. High-throughputmetabolic flux analysis based on gas chromatography–mass spec-trometry derived 13C constraints. Anal. Biochem. 325, 308–316.

Gavin, A.C., Bosche, M., Krause, R., Grandi, P., Marzioch, M.,Bauer, A., Schultz, J., Rick, J.M., Michon, A.M., Cruciat, C.M.,Remor, M., Hofert, C., Schelder, M., Brajenovic, M., Ruffner, H.,Merino, A., Klein, K., Hudak, M., Dickson, D., Rudi, T., Gnau,V., Bauch, A., Bastuck, S., Huhse, B., Leutwein, C., Heurtier,M.A., Copley, R.R., Edelmann, A., Querfurth, E., Rybin, V.,Drewes, G., Raida, M., Bouwmeester, T., Bork, P., Seraphin, B.,

G

G

G

H

H

of the biotechnologically important soil bacterium Corynebac-terium efficiens YS-314 in comparison to those of Corynebac-terium glutamicum ATCC 13032. Proteomics 5.

Haunschild, M., Freisleben, B., Takors, R., Wiechert, W., 2005.Investigating the dynamic behaviour of biochemical networksusing model families. Bioinformatics 21, 1617–1625.

Heijden, R.T.J.M.V.D., Heijnen, J.J., Hellinga, C., Romein, B., Luy-ben, K.C.A.M., 1994. Linear constraint relations in biochemicalreaction systems. I. Classification of the calculability and the bal-anceability of conversion rates. Biotechnol. Bioeng. 43, 3–10.

Heinrich, R., Schuster, S., 1996. The Regulation of Cellular Systems.Kluwer Academic Publishers.

Hermann, T., 2003. Industrial production of amino acids by coryne-form bacteria. J. Biotechnol. 104, 155–172.

Hermann, T., Finkemeier, M., Pfefferle, W., Wersch, G., Kramer, R.,Burkovski, A., 2000. Two-dimensional electrophoretic analysisof Corynebacterium glutamicum membrane fraction and surfaceproteins. Electrophoresis 21, 654–659.

Hermann, T., Pfefferle, W., Baumann, C., Busker, E., Schaffer, S.,Bott, M., Sahm, H., Dusch, N., Kalinowski, J., Puhler, A., Bendt,A.K., Kramer, R., Burkovski, A., 2001. Proteome analysis ofCorynebacterium glutamicum. Electrophoresis 22, 1712–1723.

Hermann, T., Wersch, G., Uhlemann, E.M., Schmid, R., Burkovski,A., 1998. Mapping and identification of Corynebacterium glu-tamicum proteins by two-dimensional gel electrophoresis andmicrosequencing. Electrophoresis 19, 3217–3221.

Ho, Y., Gruhler, A., Heilbut, A., Bader, G.D., Moore, L., Adams,S.L., Millar, A., Taylor, P., Bennett, K., Boutilier, K., Yang, L.,Wolting, C., Donaldson, I., Schandorff, S., Shewnarane, J., Vo,M., Taggart, J., Goudreault, M., Muskat, B., Alfarano, C., Dewar,D., Lin, Z., Michalickova, K., Willems, A.R., Sassi, H., Nielsen,P.A., Rasmussen, K.J., Andersen, J.R., Johansen, L.E., Hansen,L.H., Jespersen, H., Podtelejnikov, A., Nielsen, E., Crawford, J.,Poulsen, V., Sorensen, B.D., Matthiesen, J., Hendrickson, R.C.,Gleeson, F., Pawson, T., Moran, M.F., Durocher, D., Mann, M.,

H

H

H

H

I

Kuster, B., Neubauer, G., Superti-Furga, G., 2002. Functionalorganization of the yeast proteome by systematic analysis of pro-tein complexes. Nature 415, 141–147.

eorgi, T., Rittmann, D., Wendisch, V.F., in press. Lysine and glu-tamate production by Corynebacterium glutamicum on glucose,fructose and sucrose: roles of malic enzyme and fructose-1,6-bisphosphatase. Metab. Eng. 7, 291–301.

erstmeir, R., Cramer, A., Dangel, P., Schaffer, S., Eikmanns, B.J.,2004. RamB, a novel transcriptional regulator of genes involvedin acetate metabolism of Corynebacterium glutamicum. J. Bac-teriol. 186, 2798–2809.

erstmeir, R., Wendisch, V.F., Schnicke, S., Ruan, H., Farwick, M.,Reinscheid, D., Eikmanns, B.J., 2003. Acetate metabolism and itsregulation in Corynebacterium glutamicum. J. Biotechnol. 104,99–122.

ansmeier, N., Bartels, F.W., Ros, R., Anselmetti, D., Tauch,A., Puhler, A., Kalinowski, J., 2004. Classification of hyper-variable Corynebacterium glutamicum surface-layer proteins bysequence analyses and atomic force microscopy. J. Biotechnol.112, 177–193.

ansmeier, N., Chao, T., Puhler, A., Tauch, A., Kalinowski, J., inpress. The cytosolic, cell surface and extracellular proteomes

Hogue, C.W., Figeys, D., Tyers, M., 2002. Systematic identifica-tion of protein complexes in Saccharomyces cerevisiae by massspectrometry. Nature 415, 180–183.

ood, L., Heath, J.R., Phelps, M.E., Lin, B., 2004. Systems biol-ogy and new technologies enable predictive and preventativemedicine. Science 306, 640–643.

ood, L., Perlmutter, R.M., 2004. The impact of systems approacheson biological problems in drug discovery. Nat. Biotechnol. 22,1215–1217.

unten, P., Schiffler, B., Lottspeich, F., Benz, R., 2005. PorH, a newchannel-forming protein present in the cell wall of Corynebac-terium efficiens and Corynebacterium callunae. Microbiology151, 2429–2438.

user, A.T., Chassagnole, C., Lindley, N.D., Merkamm, M., Guyon-varch, A., Elisakova, V., Patek, M., Kalinowski, J., Brune, I., Puh-ler, A., Tauch, A., 2005. Rational design of a Corynebacteriumglutamicum pantothenate production strain and its characteriza-tion by metabolic flux analysis and genome-wide transcriptionalprofiling. Appl. Environ. Microbiol. 71, 3255–3268.

deker, T., Galitski, T., Hood, L., 2001. A new approach to decod-ing life: systems biology. Annu. Rev. Genom. Hum. Genet. 2,343–372.


Ikeda, H., Ishikawa, J., Hanamoto, A., Shinose, M., Kikuchi, H.,Shiba, T., Sakaki, Y., Hattori, M., Omura, S., 2003. Completegenome sequence and comparative analysis of the industrialmicroorganism Streptomyces avermitilis. Nat. Biotechnol. 21,526–531.

Ishige, T., Krause, M., Bott, M., Wendisch, V.F., Sahm, H., 2003.The phosphate starvation stimulon of Corynebacterium glutam-icum determined by DNA microarray analyses. J. Bacteriol. 185,4519–4529.

Jong, H.D., 2002. Modeling and simulation of genetic regulatorysystems: a literature review. J. Comput. Biol. 9, 67–103.

Kalinowski, J., 2005. The genomes of amino acid-producingcorynebacteria. In: Eggeling, L., Bott, M. (Eds.), Handbook onCorynebacterium glutamicum. CRC Press, Boca Raton, USA,pp. 37–56.

Kalinowski, J., Bathe, B., Bartels, D., Bischoff, N., Bott, M.,Burkovski, A., Dusch, N., Eggeling, L., Eikmanns, B.J., Gaigalat,L., Goesmann, A., Hartmann, M., Huthmacher, K., Kramer, R.,Linke, B., McHardy, A.C., Meyer, F., Mockel, B., Pfefferle,W., Puhler, A., Rey, D.A., Ruckert, C., Rupp, O., Sahm, H.,Wendisch, V.F., Wiegrabe, I., Tauch, A., 2003. The completeCorynebacterium glutamicum ATCC 13032 genome sequenceand its impact on the production of L-aspartate-derived aminoacids and vitamins. J. Biotechnol. 104, 5–25.

Kell, D.B., Westerhoff, H.V., 1986. Metabolic control theory: its rolein microbiology and biotechnology. FEMS Microbiol. Rev. 39,305–320.

Kirchner, O., Tauch, A., 2003. Tools for genetic engineering in theamino acid-producing bacterium Corynebacterium glutamicum.J. Biotechnol. 104, 287–299.

Kirschner, M.W., 2005. The meaning of systems biology. Cell 121,503–504.

Klamt, S., Gilles, E.D., 2004. Minimal cut sets in biochemical reac-tion networks. Bioinformatics 20, 226–234.

Kromer, J.O., Sorgenfrei, O., Klopprogge, K., Heinzle, E., Wittmann,

K

L

L

L

L

formed by a low molecular mass polypeptide. Biochemistry 37,15024–15032.

Liu, E.T., 2005. Systems biology, integrative biology, predictive biol-ogy. Cell 121, 505–506.

Magnus, J.B., Hollwedel, D., Oldiges, M., Wandrey, C., Takors, R.,2005. Modelling and analysis of the metabolic network dynam-ics in a Corynebacterium glutamicum l-valine production strain.In: BioPerspectives 2005 mit 23. DECHEMA-Jahrestagung derBiotechnologen. Wiesbaden, Germany.

Marx, A., de Graaf, A.A., Wiechert, W., Eggeling, L., Sahm, H.,1996. Determination of the fluxes in the central metabolism ofCorynebacterium glutamicum by nuclear magnetic resonancespetroscopy combined with metabolite balancing. Biotechnol.Bioeng. 49, 111–129.

Marx, A., Eikmanns, B.J., Sahm, H., de Graaf, A.A., Eggeling, L.,1999. Response of the central metabolism in Corynebacteriumglutamicum to the use of an NADH-dependent glutamate dehy-drogenase. Metab. Eng. 1, 35–48.

Marx, A., Hans, S., Mockel, B., Bathe, B., de Graaf, A.A., 2003.Metabolic phenotype of phosphoglucose isomerase mutants ofCorynebacterium glutamicum. J. Biotechnol. 104, 185–197.

Marx, A., Striegel, K., de Graaf, A.A., Sahm, H., Eggeling, L., 1997.Response of the central metabolism of Corynebacterium glutam-icum to different flux burdens. Biotechnol. Bioeng. 56, 168–180.

Mashego, M.R., Wu, L., Van Dam, J.C., Ras, C., Vinke, J.L., VanWinden, W.A., Van Gulik, W.M., Heijnen, J.J., 2004. MIRACLE:mass isotopomer ratio analysis of U-C-13-labeled extracts. A newmethod for accurate quantification of changes in concentrationsof intracellular metabolites. Biotechnol. Bioeng. 85, 620–628.

Mavrovouniotis, M.L., 1993. Identification of localized and dis-tributed bottlenecks in metabolic pathways. In: ISMB-93, Pro-ceedings of the First International Conference on IntelligentSystems for Molecular Biology, pp. 275–283.

McHardy, A.C., Goesmann, A., Puhler, A., Meyer, F., 2004. Devel-opment of joint application strategies for two microbial gene

M

M

M

N

N

C., 2004. In-depth profiling of lysine-producing Corynebac-terium glutamicum by combined analysis of the transcrip-tome, metabolome, and fluxome. J. Bacteriol. 186, 1769–1784.

rug, A., Wendisch, V.F., Bott, M., 2005. Identification of AcnR,a TetR-type repressor of the aconitase gene acn in Corynebac-terium glutamicum. J. Biol. Chem. 280, 585–595.

ange, C., Rittmann, D., Wendisch, V.F., Bott, M., Sahm, H., 2003.Global expression profiling and physiological characterization ofCorynebacterium glutamicum grown in the presence of l-valine.Appl. Environ. Microbiol. 69, 2521–2532.

ange, H.C., Eman, M., van Zuijlen, G., Visser, D., van Dam, J.C.,Frank, J., de Mattos, M.J.T., Heijnen, J.J., 2001. Improved rapidsampling for in vivo kinetics of intracellular metabolites in Sac-charomyces cerevisiae. Biotechnol. Bioeng. 75, 406–415.

iao, J.C., Boscolo, R., Yang, Y.-L., Tran, L.M., Sabatti, C., Roy-chowdhury, V.R., 2003. Network component analysis: recon-struction of regulatory signals in biological systems. PNAS 100,15522–15527.

ichtinger, T., Burkovski, A., Niederweis, M., Kramer, R., Benz,R., 1998. Biochemical and biophysical characterization of thecell wall porin of Corynebacterium glutamicum: the channel is

finders. Bioinformatics 20, 1622–1631.cHardy, A.C., Tauch, A., Ruckert, C., Puhler, A., Kalinowski,

J., 2003. Genome-based analysis of biosynthetic aminotrans-ferase genes of Corynebacterium glutamicum. J. Biotechnol. 104,229–240.

oker, N., Brocker, M., Schaffer, S., Kramer, R., Morbach, S.,Bott, M., 2004. Deletion of the genes encoding the MtrA-MtrBtwo-component system of Corynebacterium glutamicum has astrong influence on cell morphology, antibiotics susceptibilityand expression of genes involved in osmoprotection. Mol. Micro-biol. 54, 420–438.

ollney, M., Wiechert, W., Kownatzki, D., de Graaf, A.A., 1999.Bidirectional reaction steps in metabolic networks. IV. optimaldesign of isotopomer labeling experiments. Biotechnol. Bioeng.66, 86–101.

akamura, Y., Nishio, Y., Ikeo, K., Gojobori, T., 2003. The genomestability in Corynebacterium species due to lack of the recombi-national repair system. Gene 317, 149–155.

eidhardt, F.C., Umbarger, H.E., 1996. Chemical Composition ofEscherichia coli. In: Neidhardt, F.C. (Ed.), Escherichia coliand Salmonella typhimurium—Cellular and Molecular Biology.ASM Press.


Netzer, R., Krause, M., Rittmann, D., Peters-Wendisch, P.G.,Eggeling, L., Wendisch, V.F., Sahm, H., 2004a. Roles of pyru-vate kinase and malic enzyme in Corynebacterium glutamicumfor growth on carbon sources requiring gluconeogenesis. Arch.Microbiol. 182, 354–363.

Netzer, R., Peters-Wendisch, P., Eggeling, L., Sahm, H., 2004b.Cometabolism of a nongrowth substrate: l-serine utilization byCorynebacterium glutamicum. Appl. Environ. Microbiol. 70,7148–7155.

Niebisch, A., Bott, M., 2003. Purification of a cytochrome bc1-aa3

supercomplex with quinol oxidase activity from Corynebac-terium glutamicum. Identification of a fourth subunity ofcytochrome aa3 oxidase and mutational analysis of dihemecytochrome c1. J. Biol. Chem. 278, 4339–4346.

Nishio, Y., Nakamura, Y., Kawarabayasi, Y., Usuda, Y., Kimura,E., Sugimoto, S., Matsui, K., Yamagishi, A., Kikuchi, H., Ikeo,K., Gojobori, T., 2003. Comparative complete genome sequenceanalysis of the amino acid replacements responsible for thethermostability of Corynebacterium efficiens. Genome Res. 13,1572–1579.

Nolden, L., Ngouoto-Nkili, C.E., Bendt, A.K., Kramer, R.,Burkovski, A., 2001. Sensing nitrogen limitation in Corynebac-terium glutamicum: the role of glnK and glnD. Mol. Microbiol.42, 1281–1295.

Oh, M.K., Liao, J.C., 2000. Gene expression profiling by DNAmicroarrays and metabolic fluxes in Escherichia coli. Biotechnol.Progr. 16, 278–286.

Ohnishi, J., Katahira, R., Mitsuhashi, S., Kakita, S., Ikeda, M., 2005.A novel gnd mutation leading to increased l-lysine productionin Corynebacterium glutamicum. FEMS Microbiol. Lett. 242,265–274.

Ohnishi, J., Mitsuhashi, S., Hayashi, M., Ando, S., Yokoi, H.,Ochiai, K., Ikeda, M., 2002. A novel methodology employingCorynebacterium glutamicum genome information to generatea new l-lysine-producing mutant. Appl. Microbiol. Biotechnol.

O

O

P

P

P

Petersen, S., Mack, C., De Graaf, A.A., Riedel, C., Eikmanns, B.J.,Sahm, H., 2001. Metabolic consequences of altered phospho-enolpyruvatecarboxykinase activity in Corynebacterium glutam-icum reveal anaplerotic regulation mechanisms in vivo. Metab.Eng. 3, 344–361.

Peters-Wendisch, P.G., Kreutzer, C., Kalinowski, J., Patek, M., Sahm,H., Eikmanns, B.J., 1998. Pyruvate carboxylase from Corynebac-terium glutamicum: characterization, expression and inactivationof the pyc gene. Microbiology 144, 915–927.

Peters-Wendisch, P.G., Schiel, B., Wendisch, V.F., Katsoulidis, E.,Mockel, B., Sahm, H., Eikmanns, B.J., 2001. Pyruvate carboxy-lase is a major bottleneck for glutamate and lysine production byCorynebacterium glutamicum. J. Mol. Microbiol. Biotechnol. 3,295–300.

Peters-Wendisch, P.G., Wendisch, V.F., de Graaf, A.A., Eikmanns,B.J., Sahm, H., 1996. C3-carboxylation as an anaplerotic reactionin phosphoenolpyruvate carboxylase-deficient Corynebacteriumglutamicum. Arch. Microbiol. 165, 387–396.

Peters-Wendisch, P.G., Wendisch, V.F., Paul, S., Eikmanns, B.J.,Sahm, H., 1997. Pyruvate carboxylase as an anaplerotic enzymein Corynebacterium glutamicum. Microbiology 143, 1095–1103.

Raamsdonk, L.M., Teusink, B., Broadhurst, D., Zhang, N., Hayes,A., Walsh, M.C., Berden, J.A., Brindle, K.M., Kell, D.B., Row-land, J.J., Westerhoff, H.V., van Dam, K., Oliver, S.G., 2001. Afunctional genomics strategy that uses metabolome data to revealthe phenotype of silent mutations. Nat. Biotechnol. 19, 45–50.

Radmacher, E., Stansen, K.C., Besra, G.S., Alderwick, L.J.,Maughan, W.N., Hollweg, G., Sahm, H., Wendisch, V.F.,Eggeling, L., 2005. Ethambutol, a cell wall inhibitor of Mycobac-terium tuberculosis, elicits l-glutamate efflux of Corynebac-terium glutamicum. Microbiology 151, 1359–1368.

Reed, J.L., Vo, T.D., Schilling, C.H., Palsson, B.O., 2003. Anexpanded genome-scale model of Escherichia coli K-12 (iJR904GSM/GPR). Genome Biol. 4, R54.

R

R

R

R

R

58, 217–223.ldiges, M., Kunze, M., Degenring, D., Sprenger, G.A., Takors, R.,

2004. Stimulation, monitoring, and analysis of pathway dynam-ics by metabolic profiling in the aromatic amino acid pathway.Biotechnol. Progr. 20, 1623–1633.

ldiges, M., Takors, R., 2005. Applying metabolic profiling tech-niques for stimulus-response experiments: chances and pitfalls.Technol. Transf. Biotechnol.: From Lab to Industry to Produc-tion, 173–196.

allerla, S.R., Knebel, S., Polen, T., Klauth, P., Hollender, J.,Wendisch, V.F., Schoberth, S.M., 2005. Formation of volutingranules in Corynebacterium glutamicum. FEMS Microbiol.Lett. 243, 133–140.

etersen, S., de Graaf, A.A., Eggeling, L., Mollney, M., Wiechert, W.,Sahm, H., 2000. In Vivo Quantification of Parallel and Bidirec-tional Fluxes in the Anaplerosis of Corynebacterium glutamicum.J. Biol. Chem. 275, 35932–35941.

etersen, S., Lieres, E., Graaf, A.A.D., Sahm, H., Wiechert, W.,2003. A multi-scale approach for the predictive modeling ofmetabolic regulation. In: Kholodenko, B.N., Westerhoff, H.V.(Eds.), Metabolic Engineering in the Post Genomic Era. HorizonScientific Press (Chapter10).

ey, D.A., Nentwich, S.S., Koch, D.J., Ruckert, C., Puhler, A., Tauch,A., Kalinowski, J., 2005. The McbR repressor modulated bythe effector substance S-adenosylhomocysteine controls directlythe transcription of a regulon involved in sulphur metabolism ofCorynebacterium glutamicum ATCC 13032. Mol. Microbiol. 56,871–887.

iedel, C., Rittmann, D., Dangel, P., Mockel, B., Petersen, S., Sahm,H., Eikmanns, B.J., 2001. Characterization of the phospho-enolpyruvate carboxykinase gene from Corynebacterium glu-tamicum and significance of the enzyme for growth and aminoacid production. J. Mol. Microbiol. Biotechnol. 3, 573–583.

ittmann, D., Sorger-Herrmann, U., Wendisch, V.F., in press.The phosphate starvation inducible gene ushA encodes a 5′-nucleotidase required for growth of Corynebacterium glutam-icum on nucleotides as phosphorus source. Appl. Environ. Micro-biol. 71, 4339–4344.

oels, J.A., 1983. Energetics and Kinetics in Biotechnology. ElsevierBiomedical Press.

oth, S.B., Jalava, J., Ruuskanen, O., Ruohola, A., Nikkari, S., 2004.Use of an oligonucleotide array for laboratory diagnosis of bac-teria responsible for acute upper respiratory infections. J. Clin.Microbiol. 42, 4268–4274.


Ruckert, C., Puhler, A., Kalinowski, J., 2003. Genome-wide analysisof the L-methionine biosynthetic pathway in Corynebacteriumglutamicum by targeted gene deletion and homologous comple-mentation. J. Biotechnol. 104, 213–228.

Schaefer, U., Boos, W., Takors, R., Weuster-Botz, D., 1999. Auto-mated sampling device for monitoring intracellular metabolitedynamics. Anal. Biochem. 270, 88–96.

Schaffer, S., Burkovski, A., 2005. Proteomics. In: Eggeling, L., Bott,M. (Eds.), Handbook of Corynebacterium glutamicum. CRCPress, pp. 99–118.

Schaffer, S., Weil, B., Nguyen, V.D., Dongmann, G., Gunther, K.,Nickolaus, M., Hermann, T., Bott, M., 2001. A high-resolutionreference map for cytoplasmic and membrane-associated pro-teins of Corynebacterium glutamicum. Electrophoresis 22,4404–4422.

Schauer, N., Steinhauser, D., Strelkov, S., Schomburg, D., Allison,G., Moritz, T., Lundgren, K., Roessner-Tunali, U., Forbes, M.G.,Willmitzer, L., Fernie, A.R., Kopka, J., 2005. GC–MS librariesfor the rapid identification of metabolites in complex biologicalsamples. FEBS Lett. 579, 1332–1337.

Schilling, C.H., Edwards, J.S., Palsson, B.O., 1999. Towardmetabolic phenomics: analysis of genomic data using flux bal-ances. Biotechnol. Progr. 15, 288–295.

Schlusener, D., Fischer, F., Kruip, J., Rogner, M., Poetsch, A., 2005.Mapping the membrane proteome of Corynebacterium glutam-icum. Proteomics.

Schomburg, I., Chang, A., Schomburg, D., 2002. BRENDA, enzymedata and metabolic information. Nucl. Acids Res. 30, 47–49.

Shirai, T., Nakato, A., Izutani, N., Nagahisa, K., Shioya, S., Kimura,E., Kawarabayasi, Y., Yamagishi, A., Gojobori, T., Shimizu, H.,2005. Comparative study of flux redistribution of metabolic path-way in glutamate production by two coryneform bacteria. Metab.Eng. 7, 59–69.

Silberbach, M., Schafer, M., Huser, A.T., Kalinowski, J., Puhler,

S

S

S

S

S

Stephanopoulos, G., Vallino, J.J., 1991. Network rigidity andmetabolic engineering in metabolite overproduction. Science252, 1675–1681.

Strelkov, S., von Elstermann, M., Schomburg, D., 2004. Compre-hensive analysis of metabolites in Corynebacterium glutamicumby gas chromatography/mass spectrometry. Biol. Chem. 385,853–861.

Suzuki, N., Okayama, S., Nonaka, H., Tsuge, Y., Inui, M., Yukawa,H., 2005. Large-scale engineering of the Corynebacterium glu-tamicum genome. Appl. Environ. Microbiol. 71, 3369–3372.

Tauch, A., Kaiser, O., Hain, T., Goesmann, A., Weisshaar, B., Alber-meier, A., Bekel, T., Bischoff, N., Brune, I., Chakraborty, T.,Kalinowski, J., Meyer, F., Rupp, O., Schneiker, S., Viehoever,P., Puhler, A., in press. Complete genome sequence and analy-sis of the multiresistant nosocomial pathogen Corynebacteriumjeikeium k411, a lipid-requiring bacterium of the human skinflora. J. Bacteriol. 187, 4671–4782.

Theobald, U., Mailinger, W., Baltes, M., Rizzi, M., Reuss, M., 1997.In vivo analysis of metabolic dynamics in Saccharomyces cere-visiae. 1 Experimental observations. Biotechnol. Bioeng. 55,305–316.

Vallino, J.J., Stephanopoulos, G., 1993. Metabolic flux distributionsin Corynebacterium glutamicum during growth and lysine over-production. Biotechnol. Bioeng. 41, 633–646.

Vallino, J.J., Stephanopoulos, G., 1994a. Carbon flux distributions atthe glucose 6-phosphate branch point in Corynebacterium glu-tamicum during lysine overproduction. Biotechnol. Progr. 10,327–334.

Vallino, J.J., Stephanopoulos, G., 1994b. Carbon flux distributionsat the pyruvate branch point in Corynebacterium glutamicumduring lysine overproduction. Biotechnol. Progr. 10, 320–326.

Visser, D., van Zuylen, G.A., van Dam, J.C., Eman, M.R., Proll, A.,Ras, C., Wu, L., van Gulik, W.M., Heijnen, J.J., 2004. Analysisof in vivo kinetics of glycolysis in aerobic Saccharomyces cere-visiae by application of glucose and ethanol pulses. Biotechnol.

v

W

W

W

W

A., Kramer, R., Burkovski, A., 2005. Adaptation of Corynebac-terium glutamicum to ammonium limitation: a global analysisusing transcriptome and proteome techniques. Appl. Environ.Microbiol. 71, 2391–2402.

oga, T., Ohashi, Y., Ueno, Y., Naraoka, H., Tomita, M., Nish-ioka, T., 2003. Quantitative metabolome analysis using capillaryelectrophoresis mass spectrometry. J. Proteom. Res. 2, 488–494.

tansen, K.C., Uy, D., Delaunay, S., Eggeling, L., Goergen, J.-L., Wendisch, V.F., in press. Characterization of a lactate uti-lization operon of Corynebacterium glutamicum induced dur-ing temperature-triggered glutamate production. Appl. Environ.Microbiol. 71, 5920–5928.

telling, J., Klamt, S., Bettenbrock, K., Schuster, S., Gilles, E.D.,2002. Metabolic network structure determines key aspects offunctionality and regulation. Nature 420, 190–193.

telling, J., Sauer, U., Szallasi, Z., Doyle, F.J., Doyle, H., 2004.Robustness of cellular functions. Cell 118, 675–685.

tephanopoulos, G., Hwang, D., Schmitt, W.A., Misra, J.,Stephanopoulos, G., 2002. Mapping physiological statesfrom microarray expression measurements. Bioinformatics 18,1054–1063.

Bioeng. 88, 157–167.on Lieres, E., Wiechert, W., 2004. Bayes statistics and markov

chain monte carlo simulation: an alternative method for param-eter identification and error estimation. In: Proceedings of theFifth EUROSIM Congress on Modeling and Simulation, ESIEEParis, Marne la Vallee, France.

anner, B.L., 1996. Phosphorus assimilation and control of the phos-phate regulon. In: Neidhardt, F.C., Curtiss III, R., Ingraham, J.L.,Lin, E.C.C., Low, K.B., Magasanik, B., Reznikoff, W.S., Riley,M., Schaechter, M., Umbarger, H.E. (Eds.), Escherichia coli andSalmonella: Cellular and Molecular Biology, 2nd ed. ASM Press,Washington, D.C., pp. 1357–1381.

endisch, V.F., 2003. Genome-wide expression analysis inCorynebacterium glutamicum using DNA microarrays. J.Biotechnol. 104, 273–285.

endisch, V.F., 2005. Towards improving production of fine chem-icals by systems biology: amino acid production by Corynebac-terium glutamicum. Chim. Oggi 23, 49–52.

endisch, V.F., Bott, M., 2005. Phosphorus metabolism ofCorynebacterium glutamicum. In: Eggeling, L., Bott, M. (Eds.),Handbook on Corynebacterium glutamicum. CRC Press, BocaRaton, USA, pp. 377–396.


Wendisch, V.F., de Graaf, A.A., Sahm, H., Eikmanns, B.J., 2000.Quantitative determination of metabolic fluxes during couti-lization of two carbon sources: comparative analyses withCorynebacterium glutamicum during growth on acetate and/orglucose. J. Bacteriol. 182, 3088–3096.

Westerhoff, H.V., Palsson, B.O., 2004. The evolution of molecu-lar biology into systems biology. Nat. Biotechnol. 22, 1249–1252.

Wiechert, W., 2001a. 13C metabolic flux analysis. Metab. Eng. 3,195–206.

Wiechert, W., 2002. Modeling and simulation: tools for metabolicengineering. J. Biotechnol. 94, 37–63.

Wiechert, W., 2004. Validation of metabolic models: concepts, tools,and problems. In: Kholodenko, B.N., Westerhoff, H.V. (Eds.),Metabolic Engineering in the Post Genomic Era. Horizon Bio-science (Chapter 11).

Wiechert, W., Noh, K., 2005. From stationary to instationarymetabolic flux analysis. Adv. Biochem. Eng. Biotechnol. 92,145–172.

Wiechert, W., Wurzel, M., 2001. Metabolic isotopomer labelingsystems. Part I. Global dynamic behaviour. Math. Biosci. 169,173–205.

Wittmann, C., Heinzle, E., 2002. Genealogy profiling through strainimprovement by using metabolic network analysis: metabolicflux genealogy of several generations of lysine-producingcorynebacteria. Appl. Environ. Microbiol. 68, 5843–5859.

Wittmann, C., Kim, H.M., Heinzle, E., 2004. Metabolic networkanalysis of lysine producing Corynebacterium glutamicum at aminiaturized scale. Biotechnol. Bioeng. 87, 1–6.

Wolf, A., Kramer, R., Morbach, S., 2003. Three pathways for tre-halose metabolism in Corynebacterium glutamicum ATCC13032and their significance in response to osmotic stress. Mol. Micro-biol. 49, 1119–1134.

Zimmer, D.P., Soupene, E., Lee, H.L., Wendisch, V.F., Khodursky,A.B., Peter, B.J., Bender, R.A., Kustu, S., 2000. Nitrogen regula-tory protein C-controlled genes of Escherichia coli: scavengingas a defense against nitrogen limitation. Proc. Natl. Acad. Sci.U.S.A. 97, 14674–14679.

emerging corynebacterium glutamicum systems biology corynebacterium... · emerging corynebacterium...

Documents