Elsevier

Journal of Biotechnology

Volume 132, Issue 4, 1 December 2007, Pages 359-374
Journal of Biotechnology

Topology of the global regulatory network of carbon limitation in Escherichia coli

https://doi.org/10.1016/j.jbiotec.2007.08.029Get rights and content

Abstract

One fundamental shortcoming of biotechnological processes operating under carbon-limiting conditions is the high-energy demand (maintenance) of the cells. Although the function of the central carbon metabolism in supplying precursors and energy for biosynthesis has been thoroughly characterized, its regulation and dynamic behaviour during carbon-limited growth has not yet been revealed. The current work demonstrates a time series of metabolic flux distributions during fed-batch cultivation of Escherichia coli K-12 W3110 applying a constant feed rate. The fluxes in glycolysis, pentose phosphate pathway and biosynthesis fell significantly, whereas TCA cycle fluxes remained constant. The flux redistribution resulted in an enhanced energy generation in the TCA cycle and consequently, in a 20% lower biomass yield. The intracellular alarmones ppGpp and cAMP accumulated in large quantities after the onset of nutrient limitation, subsequently declining to basal levels. The network topology of the regulation of the central metabolic pathways was identified so that the observed metabolic and regulatory behaviour can be described. This provides novel aspects of global regulation of the metabolism by the cra, crp and relA/spoT modulons. The work constitutes an important step towards dynamic mathematical modelling of regulation and metabolism, which is needed for the rational optimization of biotechnological processes.

Introduction

In biotechnological processes, the cell growth often is controlled by substrate limitation. This strategy is particularly applied in fed-batch cultivations employing exponential and constant feeding profiles, which not least guarantees respectable high cell densities. Substrate-limited growth, however, results in an excessive energy consumption for the maintenance of cellular functions (Lengeler et al., 1999, Pirt, 1982) as well as in other disadvantageous stress-related effects. Enfors’ research group (Bylund et al., 1998, Larsson et al., 1996, Schweder et al., 1999, Teich et al., 1999, Xu et al., 1999a), as well as Hewitt et al., 2000, Hewitt et al., 1999, Hewitt and Nebe-Von-Caron (2001) and Lapin et al. (2006) found that in large-scale fed-batch processes the extracellular gradients of substrate concentrations have profound effects on the growth yield, product formation and viability of the cell population. Evidently, such findings are the result of the interaction of the individual cell with its abiotic environment, which determines the cell's regulatory response—and thus, its metabolic state. Consequently, the rational optimization of biotechnological processes (metabolic engineering) requires dynamic mathematical models comprising both regulation and metabolism. Ongoing research in several academic groups and industry is focused on modelling the central carbon metabolism (glycolysis, pentose phosphate pathway and TCA cycle) of Escherichia coli and its regulation during glucose-limited growth (Ellison et al., 2006, Heijnen et al., 2006). The first step in modelling is the identification of the model structure (topology), which relies on the knowledge from previous publications (bottom-up approach). Detailed information about the transcriptional regulation of the genes encoding metabolic enzymes in E. coli is available from databases (Keseler et al., 2005, Salgado et al., 2006). However, the regulators significantly affecting the central carbon metabolism during carbon limitation and consequently, the model topology of the global genetic regulatory network3 have not yet been identified. In the complementary top-down approach the model structure is deduced from the system's response to an external stimulus, applying proper systems-level experimental tools like proteomics, transcriptomics, metabolomics and metabolic flux analysis (MFA). However, the dynamic metabolic behaviour of the relevant pathways in response to carbon limitation still remains to be clarified in E. coli, which can be studied in fed-batch cultivations. When applying a constant feed rate, the substrate limitation continually increases and a succession of (physiological) quasi-steady states can be achieved (Dunn and Mor, 1975). This allows investigating the metabolic and regulatory response during the transition from exponential to carbon-limited growth.

Bacteria control metabolism and growth rate through global genetic regulatory systems (see the footnote 3), i.e. regulons and modulons (Lengeler et al., 1999, Neidhardt and Savageau, 1996). Prominent examples in E. coli are the catabolite repression (crp modulon) and the stringent response (relA/spoT modulon), two processes that are active under carbon-limiting conditions. During stringent response (reviewed in Braeken et al. (2006), Cashel et al. (1996) and Lengeler et al. (1999)), the limitation of nutrients leads to the intracellular accumulation of ppGpp (guanosine 3′,5′-bis(diphosphate)), which is supposed to bind to the RNA polymerase (Artsimovitch et al., 2004). The transcription of genes involved in the translation process – in particular of ribosomal RNA and ribosomal proteins – is negatively regulated by ppGpp. As a result, the protein biosynthesis rate declines, which in turn also leads to a reduction in growth rate (Cashel et al., 1996, Lengeler et al., 1999). During amino acid limitation, the synthesis of ppGpp or guanosine pentaphosphate (pppGpp), collectively referred to as (p)ppGpp, is mediated by RelA (GDP pyrophosphokinase/GTP pyrophosphokinase). Under amino acid-limiting conditions, the ribosome-bound RelA protein is stimulated by uncharged tRNAs at the A site of ribosomes (Wendrich et al., 2002). However, the accumulation of (p)ppGpp depends also on the dual activity of the SpoT protein as (p)ppGpp-hydrolase or (p)ppGpp-synthetase. Although it is known from a homologous protein of Streptococcus dysgalactiae subsp. equisimilis that the opposing activities of SpoT are reciprocally regulated (Hogg et al., 2004, Mechold et al., 2002), the regulation of the SpoT protein in E. coli is still hypothetical. The most important issue for understanding growth control is the signalling mechanism, which leads to accumulation of ppGpp under carbon-limiting conditions, an aspect that is still not entirely clarified.

Besides various effects on growth-related functions (Cashel et al., 1996), the alarmone ppGpp is known to be involved in the regulation of the sigma S factor concentration (σS; rpoS gene) on the transcriptional and posttranscriptional level (Hengge-Aronis, 2002). As an alternative subunit of RNA polymerase, σS is involved in the regulation of transcription in the general stress response in E. coli (also designated as ‘stationary phase response’). It is assumed that elevated levels of σS negatively regulate σD-dependent housekeeping genes, such as the TCA cycle genes (Patten et al., 2004). Moreover, ppGpp influences the competition between different stress-related sigma factors in the binding of the RNA polymerase core enzyme at the expense of the sigma factor σD (Jishage et al., 2002) and the RNA polymerase availability (Barker et al., 2001a, Barker et al., 2001b, Cashel et al., 1996, Jensen and Pedersen, 1990, Traxler et al., 2006). The crp modulon belongs to a group of global genetic regulatory systems, which can be subsumed under the term catabolite control. One basic feature of these systems is that the presence or absence of an extracellular carbon source is indicated by an intracellular metabolite (catabolite) that serves as a signal for derepression (catabolite activation) or deactivation (catabolite repression) of catabolic genes (Saier et al., 1996). The crp modulon includes catabolic operons for the utilization of various carbon sources and is regulated by the Crp–cAMP complex. The synthesis of the alarmone cAMP (cyclic 3′,5′-AMP) by the enzyme adenylate cyclase (CyaA) is stimulated by the phophorylated EIIAGlc protein, a component of the E. coli phosphoenolpyruvate:carbohydrate phosphotransferase system (PTS) (reviewed in Lengeler et al. (1999) and Postma et al. (1993)). It is assumed that a low glucose uptake rate by the PTS and a high ratio of phosphoenolpyruvate and pyruvate concentrations (cPEP/cPyr) lead to the phosphorylation of the EIIAGlc protein (Hogema et al., 1998). Consequently, limited glucose availability leads to the synthesis of cAMP and the transcriptional regulator complex Crp–cAMP is formed. Catabolite control is also exerted by the catabolite repressor/activator protein Cra (formerly designated FruR), which regulates numerous genes involved in the carbon and energy metabolism (the cra modulon) (reviewed in Ramseier (1996), Saier and Ramseier (1996) and Saier et al. (1996)). The regulator protein Cra is inactivated by the catabolites fructose 1-phosphate and fructose 1,6-bis(phosphate) (Saier and Ramseier, 1996).

The present study demonstrates the signal formation and dynamic metabolic responses of E. coli K-12 W3110 exposed to an increasing carbon limitation during fed-batch cultivation applying a constant feed rate. The observed decrease in the biomass yield is shown to result from a substantial carbon flow into the TCA cycle and from the subsequent oxidation of the carbon source. Simultaneously, most of the fluxes in the central carbon metabolism and in the biosynthesis pathways decreased significantly. This rearrangement is supposed to re-establish a balance between anabolism and catabolism after nutrient limitation. The genetic regulatory systems responsible for the illustrated metabolic responses are proposed and assembled to a global regulatory network. A new method for determining the qualitative time course of the intracellular cAMP concentration is presented. For the first time the profile of the intracellular cAMP level is shown in fed-batch cultivations of E. coli wild-type cells. Most importantly, the resetting of the cAMP signal could be demonstrated. The suggested network takes account of the observed signal resetting (cAMP and ppGpp) and of a probable stringent response-signalling pathway during carbon limitation. The provided network topology is novel inasmuch as it comprehensively explains the obtained systems-level data of the metabolic transition from exponential to carbon-limited growth typical of fed-batch processes.

Section snippets

Bacterial strain and fed-batch cultivation

The fed-batch cultivations were carried out with the bacterial strain E. coli K-12 W3110 (DSM 5911, German Collection of Microorganisms and Cell Cultures) in a 30-l bioreactor (Bioengineering AG, Wald, Switzerland). Minimal medium supplemented with glucose as the carbon source was used. The batch medium (batch volume VR,0 = 17 l) consisted of 8.8 g l−1 glucose·H2O, 2.0 g l−1 Na2SO4·10H2O, 2.68 g l−1 (NH4)2SO4, 1.0 g l−1 NH4Cl, 14.6 g l−1 K2HPO4, 4.02 g l−1 NaH2PO4·2H2O, 0.01 g l−1 thiamine HCl; 0.3 mM CaCl2·2H2

Fed-batch cultivation

Five independent fed-batch cultivation experiments applying a constant feed rate were carried out to study the response of the central carbon metabolism of E. coli concerning glucose limitation. The time profiles of the extracellular concentrations of glucose, acetate and biomass are presented in Fig. 1a. Acetate was produced during the consumption of glucose in the batch phase and was used up after 1 h of glucose-limited growth (Fig. 1a). In response to the limited availability of the carbon

Topology of the network regulating the central carbon metabolism

The work presented here demonstrates the formation and resetting of the intracellular signals of carbon limitation, cAMP and ppGpp, and the redistribution of metabolic fluxes during glucose-limited fed-batch growth. In order to explain the observed dynamic behaviour the topology of the global genetic regulatory network of the regulation of central carbon metabolism is identified. This involves the examination of global genetic regulatory systems (regulons and modulons) that are known to be

Acknowledgments

The authors would like to thank Insilico Biotechnology AG for providing the software tool Insilico Discovery. We would also like to thank Petra Schlack, Andreas Freund and Achim Hauck for excellent technical assistance in the analysis of nucleotides and during the cultivations; Heiko Schwarz for the support in the development of the analytical procedure for the determination of nucleotides; Volker Windeisen for the in vitro synthesis of ppGpp; Jochen Rebell, Institut für Organische Chemie

References (77)

  • T.M. Ramseier

    Cra and the control of carbon flux via metabolic pathways

    Res. Microbiol.

    (1996)
  • T.M. Ramseier et al.

    In vitro binding of the pleiotropic transcriptional regulatory protein, FruR, to the fru, pps, ace, pts and icd operons of Escherichia coli and Salmonella typhimurium

    J. Mol. Biol.

    (1993)
  • J.W. Schmid et al.

    Metabolic design based on a coupled gene expression-metabolic network model of tryptophan production in Escherichia coli

    Metab. Eng.

    (2004)
  • H. Senn et al.

    The growth of Escherichia coli in glucose-limited chemostat cultures: a re-examination of the kinetics

    Biochim. Biophys. Acta

    (1994)
  • T.M. Wendrich et al.

    Dissection of the mechanism for the stringent factor RelA

    Mol. Cells

    (2002)
  • G. An et al.

    Cloning the spoT gene of Escherichia coli: identification of the spoT gene product

    J. Bacteriol.

    (1979)
  • D. Bhatnagar et al.

    Cyclic AMP transport across membrane vesicles of ultra-violet light irradiated Escherichia coli

    J. Biosci.

    (1984)
  • F. Bylund et al.

    Substrate gradient formation in the large-scale bioreactor lowers cell yield and increases by-product formation

    Bioprocess Eng.

    (1998)
  • M.W. Calhoun et al.

    Energetic efficiency of Escherichia coli: effects of mutations in components of the aerobic respiratory chain

    J. Bacteriol.

    (1993)
  • M. Cashel et al.

    The stringent response

  • C. Chassagnole et al.

    Dynamic modeling of the central carbon metabolism of Escherichia coli

    Biotechnol. Bioeng.

    (2002)
  • A.J. Cozzone et al.

    Control of isocitrate dehydrogenase catalytic activity by protein phosphorylation in Escherichia coli

    J. Mol. Microbiol. Biotechnol.

    (2005)
  • M. Cserjan-Puschmann et al.

    Metabolic approaches for the optimisation of recombinant fermentation processes

    Appl. Microbiol. Biotechnol.

    (1999)
  • A. Death et al.

    Between feast and famine: endogenous inducer synthesis in the adaptation of Escherichia coli to growth with limiting carbohydrates

    J. Bacteriol.

    (1994)
  • I.J. Dunn et al.

    Variable-volume continuous cultivation

    Biotechnol. Bioeng.

    (1975)
  • M. Emmerling et al.

    Metabolic flux responses to pyruvate kinase knockout in Escherichia coli

    J. Bacteriol.

    (2002)
  • W. Epstein et al.

    Adenosine 3′:5′-cyclic monophosphate as mediator of catabolite repression in Escherichia coli

    Proc. Natl. Acad. Sci. U.S.A.

    (1975)
  • R.B. Gennis et al.

    Respiration

  • D.R. Gentry et al.

    Mutational analysis of the Escherichia coli spoT gene identifies distinct but overlapping regions involved in ppGpp synthesis and degradation

    Mol. Microbiol.

    (1996)
  • M.T. Hansen et al.

    Simple downshift and resulting lack of correlation between ppGpp pool size and ribonucleic acid accumulation

    J. Bacteriol.

    (1975)
  • R. Hengge-Aronis

    Signal transduction and regulatory mechanisms involved in control of the σS (RpoS) subunit of RNA polymerase

    Microbiol. Mol. Biol. Rev.

    (2002)
  • C.J. Hewitt et al.

    An industrial application of multiparameter flow cytometry: assessment of cell physiological state and its application to the study of microbial fermentations

    Cytometry

    (2001)
  • C.J. Hewitt et al.

    Studies related to the scale-up of high-cell-density E. coli fed-batch fermentations using multiparameter flow cytometry: effect of a changing microenvironment with respect to glucose and dissolved oxygen concentration

    Biotechnol. Bioeng.

    (2000)
  • C.J. Hewitt et al.

    Use of multi-staining flow cytometry to characterise the physiological state of Escherichia coli W3110 in high cell density fed-batch cultures

    Biotechnol. Bioeng.

    (1999)
  • B.M. Hogema et al.

    Inducer exclusion in Escherichia coli by non-PTS substrates: the role of the PEP to pyruvate ratio in determining the phosphorylation state of enzyme IIAGlc

    Mol. Microbiol.

    (1998)
  • J.L. Ingraham et al.

    Growth of the Bacterial Cell

    (1983)
  • Cited by (59)

    View all citing articles on Scopus
    1

    Present address: Institute of Biochemical Engineering, University of Stuttgart, Allmandring 31, D-70569 Stuttgart, Germany.

    2

    Present address: Fraunhofer Institute for Manufacturing Engineering and Automation, Department Cleanroom Manufacturing, Nobelstrasse 12, 70569 Stuttgart, Germany.

    View full text