CreA influences the metabolic fluxes of Aspergillus nidulans during growth on glucose and xylose

Helga David, Astrid Mørkeberg Krogh, Christophe Roca{dagger}, Mats Åkesson{ddagger} and Jens Nielsen

Center for Microbial Biotechnology, BioCentrum-DTU, Building 223, Technical University of Denmark, DK-2800 Kgs, Lyngby, Denmark

Correspondence
Jens Nielsen
jn{at}biocentrum.dtu.dk


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The physiological phenotype of Aspergillus nidulans was investigated for different genetic and environmental conditions of glucose repression through the quantification of in vivo fluxes in the central carbon metabolism using 13C-metabolic-flux analysis. The particular focus was the role of the carbon repressor CreA, which is the major regulatory protein mediating carbon repression in many fungal species, in the primary metabolism of A. nidulans. Batch cultivations were performed with a reference strain and a deletion mutant strain (creA{Delta}4) using [1-13C]glucose as carbon source. The mutant strain was also grown on a mixture of [1-13C]glucose and unlabelled xylose. Fractional enrichment data were measured by gas chromatography-mass spectrometry. A model describing the central metabolism of A. nidulans was used in combination with fractional enrichment data, and measurements of extracellular rates and biomass composition for the estimation of the in vivo metabolic fluxes. The creA-mutant strain showed a lower maximum specific growth rate than the reference strain when grown on glucose (0·11 and 0·25 h–1, respectively), whereas the specific growth rate of the mutant strain grown on the glucose/xylose mixture was identical to that on glucose (0·11 h–1). Different patterns and increased levels of extracellular polyols were observed both upon deletion of the creA gene and upon addition of xylose to the growth medium of the mutant strain. Concerning metabolic fluxes, the major change observed in the flux distribution of A. nidulans upon deletion of the creA gene was a 20 % decrease in the flux through the oxidative part of the pentose-phosphate pathway. Addition of xylose to the growth medium of the mutant resulted in an increase of about 40 % in the activity of the oxidative part of the pentose-phosphate pathway, as well as decreases in the fluxes through the Embden–Meyerhof–Parnas pathway and the tricarboxylic acid cycle (in the range of 20–30 %). The derepression of key pathways leads to alterations in the demands for cofactors, thereby imposing changes in the central metabolism due to the coupling of the many different reactions via the redox and energy metabolism of the cells.


Abbreviations: EMP, Embden–Meyerhof–Parnas; PP, pentose phosphate; SFL, summed fractional labelling

{dagger}Present address: Centro de Recursos Microbiológicos, Biotechnology Unit, Faculty of Sciences and Technology, New University of Lisbon, 2829-516 Caparica, Portugal.

{ddagger}Present address: Novo Nordisk A/S, BioProcess Laboratories, Bagsværd, Denmark.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Aspergilli are a group of filamentous fungi that play an important role as cell factories for the production of organic acids, for example citric acid by Aspergillus niger (Kubicek & Röhr, 1986), of enzymes, for example {alpha}-amylase by Aspergillus oryzae (Carlsen & Nielsen, 2001), and of polyketides, for example statins by Aspergillus terreus (Manzoni & Rollini, 2002). Besides their industrial importance, aspergilli can act as pathogens for humans and animals, being responsible for diseases that range from allergy-type illness to life-threatening generalized infections, most commonly caused by Aspergillus fumigatus and Aspergillus flavus, but also Aspergillus nidulans and others (http://www.Aspergillus.man.ac.uk). The industrial and medical relevance of these filamentous fungi has provided the rationale for conducting large-scale genome research on particular species, and several genome-sequencing projects have been completed recently (A. fumigatus, A. niger, A. nidulans) or are nearing completion (A. flavus, A. oryzae, Aspergillus parasiticus, Aspergillus terreus) (GOLD, Genomes Online Database, http://www.genomesonline.org/). In fact, the information from whole genome sequences may increase the significance of these filamentous fungi in both fundamental and applied research, as genomic information opens the way for application of novel technologies like genome-wide transcription analysis and proteome analysis. In particular, it may lead to the mapping of the mechanisms underlying their pathogenicity, as well as assist in the design of metabolic engineering strategies to produce strains with improved properties (Nielsen, 2001).

Aspergilli have larger genomes and more genes than the eukaryotic model organism Saccharomyces cerevisiae, which endow them with more extensive metabolic capabilities for sustaining their saprophytic way of life, in which they are constantly confronted with changing nutrient conditions. Carbon repression is a regulatory mechanism involved in the adaptation of these organisms to this lifestyle. The effect of this phenomenon is that the synthesis of enzymes and permeases participating in the catabolism of alternative carbon sources is repressed when a more favourable source (e.g. glucose) is present in the growth medium (Ruijter & Visser, 1997). In spite of being an important mechanism for the survival of these organisms, carbon repression represents a major problem in large-scale production, where it is common to use complex sugar mixtures as substrate, since this mechanism prevents the co-consumption of substrates.

Carbon repression in aspergilli has been the focus of several studies, in particular for the model organism A. nidulans (for reviews see Arst & Bailey, 1977; Kelly & Hynes, 1982; Ruijter & Visser, 1997; Scazzocchio et al., 1995). The major regulatory gene controlling carbon repression in A. nidulans, A. niger and A. oryzae is creA (Agger et al., 2002; Ruijter & Visser, 1997; Shroff et al., 1997). The mechanism of repression by the CreA protein is understood in great detail, but little is known about the signal transduction pathway for carbon repression in aspergilli (Arst & Bailey, 1977; Felenbok et al., 2001; Ruijter & Visser, 1997; Scazzocchio et al., 1995). CreA is a DNA-binding protein that has two zinc fingers, which bind to specific short sequences in the promoter of target systems, thereby preventing the transcription of target genes (Ruijter & Visser, 1997). Studies aimed at investigating the physiological consequences of mutations in the creA gene have been undertaken by several research groups. It has been reported that the deletion of the creA gene has drastic effects on the morphology of A. nidulans under both carbon repressing and derepressing conditions (Dowzer & Kelly, 1991; Shroff et al., 1997). Furthermore, a creA-mutant strain and a reference strain of A. nidulans have been compared with respect to enzyme activities, metabolite concentrations and polyol pools, and the differences found pointed to alterations in the fluxes through the central pathways upon deletion of the creA gene (van der Veen et al., 1995).

In this work, we focused on the role of CreA in the physiological phenotype of A. nidulans through the quantification of the fluxes in the central carbon metabolism for different conditions of glucose repression. The fluxes were quantified through metabolic-flux analysis based on stationary carbon-isotope-labelling experiments using fractional enrichment data (for reviews on 13C metabolic-flux analysis see Christensen & Nielsen, 2000b; Szyperski, 1998; Wiechert & de Graaf, 1996; Wiechert, 2001). Carbon-labelling experiments were performed with a reference strain and a derepressed mutant strain (creA{Delta}4), in batch mode, using labelled glucose as substrate. In addition, the mutant cells were grown on a mixture of labelled glucose and unlabelled xylose. The latter is also a strongly repressing carbon source (Ruijter & Visser, 1997), although to a smaller extent than glucose. The study of the catabolism of xylose in the presence of glucose is relevant because both sugars are major components of lignocellulose (bulk of plant material), which represents an abundant and renewable carbon source, currently used in important industrial production processes (e.g. bioethanol) and with potential for utilization in the production of bulk chemicals. A metabolic model including all relevant biochemical conversions and transport processes, as well as the fate of the carbon atoms in the central carbon metabolism of A. nidulans, was derived from a metabolic reconstruction of the central carbon metabolism of A. niger (David et al., 2003). Fractional enrichment data and measurements of extracellular rates, as well as information on the biomass composition of the fungus, were then combined with the metabolic model for identification of the network topology and estimation of the in vivo fluxes in the reference and mutant strains at the different growth conditions studied.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Strains.
The two strains used in this study were the reference strain A. nidulans A187 (pabaA1 ya2) (obtained from the Fungal Genetics Stock Center, Kansas City, Kansas 66160-7420, USA) and the mutant strain A. nidulans creA{Delta}4 (pabaA1 ya2 argB2) (kindly donated by Joan Kelly, Department of Genetics, University of Adelaide, Adelaide 5005, Australia), which results from partial deletion of the creA gene (Shroff et al., 1997).

Growth medium.
A defined medium containing trace metal elements and vitamins was used in all cultivations. The medium used had the following composition: 15 g (NH4)2SO4 l–1, 3 g KH2PO4 l–1, 2 g MgSO4.7H2O l–1, 2 g NaCl l–1, 0·2 g CaCl2 l–1, 1 ml trace element solution l–1, 5 mg p-aminobenzoic acid (PABA) l–1 and 1 ml antifoam (sb2121) l–1. Trace element solution composition (per litre): 14·3 g ZnSO4.H2O, 13·8 g FeSO4.7H2O, 0·5 g NiCl2.6H2O and 2·5 g CuSO4.5H2O. The carbon sources used were either glucose (5 g l–1) or a mixture of glucose (2·5 g l–1) and xylose (2·5 g l–1). In the labelling experiments, [1-13C]glucose was introduced into the bioreactor, whereas the xylose used was unlabelled. The labelling patterns of the substrates were chosen so that the methodology adopted (metabolic-flux analysis based on GC-MS) enabled the discrimination of the fluxes in the central metabolic pathways.

Pre-cultivations.
Spores of both strains were propagated by inoculation on agar plates with the following medium composition (per litre): 10 g D-glucose, 2 g peptone, 1 g yeast extract, 1 g casamino acids, 50 ml salt solution, 1 ml trace element solution, 1 ml vitamin solution and 16 g agar. Salt solution composition (per litre): 10·4 g KCl, 10·4 g MgSO4.7H2O and 30·4 KH2PO4. Trace element solution composition (per 100 ml): 1·43 g ZnSO4.H2O, 1·38 g FeSO4.7H2O, 0·25 g CuSO4.5H2O and 0·05 g NiCl2.6H2O. Vitamin solution composition (per 100 ml): 100 mg each of the vitamins biotin, pyridoxin, thiamine, riboflavin, p-aminobenzoic acid (PABA) and nicotinic acid (http://www.fgsc.net/fgn48/Kaminskyj.htm). The spores were cultivated at 37 °C for 5–7 days and harvested by adding 10 ml 1 % Tween 80 in sterile water. The batch cultivations were inoculated with spores at a concentration of 105–106 spores ml–1.

Batch cultivations.
Batch cultivations were carried out in two types of bioreactors. To determine the physiological characteristics of the strains, cultivations were performed in well-controlled baffled 1 l in-house-manufactured bioreactors with a working volume of 0·8 l. The bioreactors were equipped with two disc-turbine impellers rotating at 100 r.p.m. at the beginning of the cultivation. The pH was kept constant at 3·0 by automatic addition of 2 M NaOH and the temperature was maintained at 30 °C. Air was used for sparging the bioreactor at a constant flow rate of 0·05 VVM (volume of gas per volume of liquid per minute). These conditions were maintained until the spores started to germinate, in order to prevent their agglomeration and thereby encourage filamentous growth. After germination started, the pH was increased to 6·0, the aeration to 1 VVM and the stirring speed to 900 r.p.m. The carbon-labelling experiments were carried out in 400 ml bioreactors specially designed for the purpose with a working volume of 300 ml. The batch cultivations in these bioreactors were performed in the same way as in the 1 l bioreactors.

Cell mass determination.
Cell dry weight was determined using nitrocellulose filters (pore size 0·45 µm, Gelman Sciences). The filters were pre-dried in a microwave oven at 150 W for 10 min and subsequently weighed. A known volume of cell culture was filtered and the residue was washed with distilled water and dried on the filter for 15 min in a microwave oven at 150 W. The filter was weighed again and the cell mass concentration was calculated.

Analysis of extracellular metabolites.
Fermentation samples were immediately filtered through SPE cartridge Chromafix 400-C18 columns (Macherey-Nagel) and stored at –20 °C until analysis. Glucose, xylose, glycerol and ethanol concentrations were determined by HPLC analysis using an Aminex HPX-87H column (Bio-Rad). The column was kept at 65 °C and eluted at 0·6 ml min–1 with 5 mM H2SO4. The compounds were detected refractometrically (Waters 410 Differential Refractometer). Xylitol, arabitol, erythritol, mannitol and galactitol concentrations were determined by high pH anion exchange chromatography (HPAEC) analysis using a CarboPac MA1 column (Dionex). The column was left at room temperature and eluted at 0·4 ml min–1 with high quality carbonate-free 612 mM NaOH (J. T. Baker). The eluent was degassed intensively with helium during 5 min before use. Detection of the polyols was performed by pulsed amperometric detection.

Analysis of total protein.
The concentration of solubilized protein was measured by the Bio-Rad Protein Assay, which is based on the method of Bradford (1976), using BSA as standard.

Analysis of fractional 13C enrichments.
Mycelia were harvested in the exponential growth phase, hydrolysed, derivatized and analysed by GC-MS for determination of labelling patterns of intracellular metabolites (glucose 6-phosphate and proteinogenic amino acids), according to Christensen & Nielsen (2000a). Summed fractional labellings (SFLs) were determined as described elsewhere (Christensen & Nielsen, 2000a). The consistency of the labelling measurements was evaluated through comparison of the SFLs of specific fragments obtained via different derivatization methods or different fragments of the same molecule resulting from a particular derivatization method. In general, the data were consistent, but inconsistencies were found for some fragments due to peak overlapping, tailing or broadening. In these cases, data were neglected in the analysis and flux calculations.

Computational methods.
Quantification of metabolic fluxes was accomplished by combining isotope and metabolite balancing. The mathematical framework adopted was the one described by Wiechert & de Graaf (1997) and Christensen & Nielsen (2000a), in which isotope balancing is carried out on a carbon atom basis, being based on fractional enrichments (fraction of labelled carbon atoms in a given position of a metabolite). In addition to a metabolic model (see Supplementary Table, available with the online version of this paper), the method makes use of measurements of extracellular rates (see Table 1) and fractional enrichment data of intracellular metabolites (see Table 2), as well as labelling patterns of substrates and biomass composition (see Table 3). For the sake of comparing different experiments, SFL data were normalized with respect to the amount of label entering the cells as substrate and the normalized values are presented in Table 2. Experimental standard deviations were calculated and are shown along with SFL data; however, in the flux quantification, a standard deviation of at least 1 % 13C-labelling was considered, whereas a standard deviation of 10 % was set for the fluxes. The metabolic fluxes were determined by an iterative process, in which an optimization method was used for minimizing the total deviation between measured and computed fractional enrichments and fluxes.


View this table:
[in this window]
[in a new window]
 
Table 1. Physiological parameters for batch cultivations of the reference strain and the creA-mutant strain growing on glucose or a mixture of glucose and xylose

 

View this table:
[in this window]
[in a new window]
 
Table 2. Measured summed fractional labellings (SFLs) for the different cultivations

Data are presented as SFL (SD). The values shown for SFLs refer to normalized data with respect to the amount of label entering in the form of substrate (104·5 % when the substrate is 100 % [1-13C]glucose, and 79·0 % when the substrate is the mixture of [1-13C]glucose and unlabelled xylose). For the calculations, a SD lower than 1 was considered to be 1, according to the method applied for measuring the labelling patterns (Christensen & Nielsen, 2000a).

 

View this table:
[in this window]
[in a new window]
 
Table 3. Cellular composition considered for determination of drain fluxes into biomass in the metabolic model of A. nidulans

Data are presented as content (normalized content).

 

   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Physiological parameters
The reference strain and the creA-mutant strain were compared in aerobic batch cultivations using glucose as the sole carbon source (5 g l–1). Moreover, the creA deletion strain was grown on a mixture of glucose (2·5 g l–1) and xylose (2·5 g l–1). For all cultivations, polyols were the only metabolites found to be excreted into the extracellular medium, namely mannitol, glycerol, erythritol and arabitol. In addition, xylitol was detected when a mixture of glucose and xylose was used as substrate. Physiological parameters for growth and metabolite production for the different cultivations performed are shown in Table 1. Fig. 1 depicts the concentration profiles of substrate and biomass for the batch cultivations carried out in the 1 l reactors with the reference strain and the creA-mutant strain.



View larger version (13K):
[in this window]
[in a new window]
 
Fig. 1. Concentration profiles for the batch cultivations in the 1 l reactors of the reference strain grown solely on glucose (a) and the creA-mutant strain grown solely on glucose (b). Open symbols, glucose; closed symbols, biomass.

 
Qualitative analysis of labelling patterns
Carbon-labelling experiments were carried out using [1-13C]glucose as substrate for both the reference strain and the mutant strain in order to quantitatively determine the effect of deleting the creA gene on the flux distribution in the central carbon metabolism of A. nidulans. Moreover, a mixture of [1-13C]glucose and unlabelled xylose was used as substrate in a cultivation of the creA-mutant strain to investigate the catabolism of a less favourable carbon source (xylose) in the presence of glucose.

The flux distribution in the central metabolism determines the labelling patterns of central metabolites and hence one may gain an insight into the active pathways and the topology of the metabolic network through the examination of summed fractional labelling (SFL) data. Table 2 summarizes the labelling patterns of several compounds measured by GC-MS for the different cultivations performed. These compounds correspond to fragments of building blocks (e.g. glucose 6-phosphate and proteinogenic amino acids) and the labelling patterns of the corresponding biosynthetic precursors may be deduced from the measured data by assuming biosynthetic routes for these building blocks and mapping their carbon skeletons to those of the precursors (see Table 2). Furthermore, with the aid of a model describing the carbon transitions involved in the conversion of substrates into precursor metabolites, it is possible to quantify the fluxes in the primary metabolism (Gombert et al., 2001).

Metabolic model
The metabolic model used for flux quantification was a non-compartmentalized model based on the inspection of the labelling data and on a metabolic reconstruction of the central carbon metabolism of A. niger (David et al., 2003). The model included information on the stoichiometry and carbon transitions of reactions participating in the major central pathways, namely the Embden–Meyerhof–Parnas (EMP) pathway, the pentose phosphate (PP) pathway, the tricarboxylic acid (TCA) cycle and anaplerotic reactions, as well as the uptake of sugars (glucose and xylose), and anabolism of polyols and of specific amino acids. In total, 43 reactions and 19 intracellular (or rather, balanced) metabolites were included in the metabolic model (see Supplementary Table, available with the online version of this paper).


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Carbon-labelling experiments
Carbon-labelling experiments should preferably be performed in continuous mode as metabolic stationarity is a necessary condition for metabolic-flux analysis (Wiechert et al., 2001). However, in the exponential phase of batch cultivations, the growth is balanced and hence it can be considered that the internal metabolites are in pseudo-steady state, and the use of batch cultures has recently been employed with success for analysis of various yeast (Raghevendran et al., 2004), Escherichia coli (Fischer & Sauer, 2003) and Bacillus subtilis mutants (Zamboni & Sauer, 2003). A potential pitfall associated with the analysis of different mutants in batch cultures is that the mutants may exhibit different maximum specific growth rates, and during unlimited growth in a batch culture the metabolic response is therefore likely to be a function not only of the mutation but also of the variation in the specific growth rate. However, because the basic objective of metabolic profiling is to map all the effects of a specific mutation (or a variation in medium composition), both direct and indirect, the use of batch cultures is well suited. In addition, for the analysis of filamentous fungi, batch cultures are preferred to continuous cultures, which are inherently difficult to conduct with these organisms. In particular with A. nidulans, continuous cultures are complicated to operate due to the occurrence of wall growth, sporulation and pellet formation. These problems may also occur in batch mode, but they can be overcome partially by harvesting the cells for subsequent analysis at an early stage of growth. In the light of these observations, we chose an experimental strategy where metabolic-flux analysis was based on batch experiments.

Conditions of glucose repression
It has been recently reported that CreA-mediated glucose repression in A. nidulans is a function of the specific growth rate (Ilyés et al., 2004). The study involved the measurement of the activity of {beta}-galactosidase, which is repressed by glucose, at different dilution rates in glucose-limited chemostat cultures of both a reference and a creA-deleted mutant. It was found that for specific growth rates of 0·068 h–1 or higher, glucose exerts repression on the synthesis of key enzymes, whereas for a specific growth rate of 0·015 h–1, glucose derepression occurs under the growth conditions considered. In our study, the cells were analysed in the exponential phase while growing at their maximal specific growth rates (0·11 h–1 for reference and 0·25 h–1 for mutant), which are well above the critical, derepressing growth rate, and hence repressing conditions were assured for the study.

Physiological parameters and metabolism of polyols
As can be seen in Table 1, although the mutant strain grew much slower on glucose than the reference strain, deletion of the creA gene did not have any significant effect on the biomass yield on glucose, since the specific glucose uptake rate of the mutant was lower than the one of the reference strain (1·4 vs 2·9 mmol (g dry wt)–1 h–1, respectively). On the other hand, larger amounts of polyols in the extracellular medium of the mutant cells reflected higher yields of polyols on glucose. Furthermore, the specific growth rate and the biomass yield of the mutant strain were the same when the cells were grown on glucose or on a mixture of glucose and xylose, but a higher yield of polyols was observed for growth on the sugar mixture (on a mass basis as well as on a carbon basis). When the mutant was grown on the mixture of glucose and xylose, it was observed that both sugars were taken up from the medium simultaneously, but glucose was consumed at a specific rate that was threefold that of xylose. Earlier it was reported that, when a reference strain was grown on a mixture of glucose and xylose, there was a sequential utilization of the two sugars (the xylose was only catabolized after the glucose had been depleted) (Prathumpai et al., 2004), and for that reason the cultivation of the reference strain on the sugar mixture was not pertinent to our study.

The patterns of polyol secretion during growth on glucose differed between the reference strain and the creA-mutant strain. The major polyol excreted by the reference strain was glycerol, followed by erythritol and mannitol, and only very small amounts of arabitol were detected in the extracellular medium (Table 1). Conversely, the mutant strain excreted predominantly erythritol and, to a smaller extent, glycerol and arabitol. Low levels of mannitol were produced by the mutant strain. Deletion of the creA gene resulted in an increase (of 62 %, on a carbon basis) in the total yield of polyols secreted into the extracellular medium during growth on glucose. The yields of arabitol and erythritol on glucose underwent significant increases (21-fold and 11-fold, respectively) and the mannitol yield increased 60 %, whereas glycerol secretion dropped about 80 %.

The increase in polyol secretion in the creA strain was consistent with a previous study by van der Veen et al. (1995). However, they reported that mannitol was the major polyol excreted by both the reference and a strain analogous to our creA strain (creAd-30). Moreover, they found a larger increase (about sixfold) in the level of extracellular polyols, including glycerol, upon mutation of the creA gene. These discrepancies might be explained by differences in the cultivation conditions (such as pH and oxygenation rate). In this work, experiments were carried out in well-controlled bioreactors, where the pH was kept constant through the addition of base, whereas in the above-mentioned study the strains were grown in shake-flasks, without pH control. Furthermore, in our study all samples were taken in the exponential phase, whereas van der Veen et al. (1995) took samples after a fixed time of cultivation of both strains (corresponding to different growth phases due to different specific growth rates).

When the creA mutant was grown on a mixture of glucose and xylose, the main polyol excreted was erythritol, followed by arabitol, mannitol and glycerol. Small amounts of xylitol were also detected in the extracellular medium. Comparison of the mutant strain on glucose and on the mixture of glucose and xylose revealed that on the latter there was an increase in the secretion of polyols (of about 60 %, on a carbon basis), which was more significant for mannitol (fivefold) and arabitol (threefold), whereas the yields of erythritol and glycerol were affected to a lesser extent (increases of 27 % and 10 %, respectively).

Polyols are synthesized and accumulate inside the cell, and eventually are secreted into the extracellular medium if not re-utilized by the cell. Reconsumption of polyols requires that the enzymes participating in their catabolic pathways are active, and typically these enzymes are subject to two forms of regulation, namely specific induction and carbon repression (Arst & Bailey, 1977). In particular, the enzymes involved in catabolism of glycerol (glycerol kinase and FAD-dependent glycerol-3-phosphate dehydrogenase) are induced by glycerol and dihydroxyacetone, but are also under carbon repression control mediated by the creA gene. Therefore, under repressing conditions, as in the case of the reference cells, it is expected that the intracellular pool of glycerol is not reconsumed to the same extent as in the mutant cells, and that accumulation of this polyol inside the cell leads ultimately to its secretion into the extracellular medium. This observation is in accordance with the decrease verified in the amount of glycerol excreted upon deletion of the creA gene. The major pathway for biosynthesis of glycerol in A. nidulans involves a dihydroxyacetone-phosphate phosphatase and a constitutive NADP+-dependent glycerol dehydrogenase. Under derepressing conditions, a second NADP+-dependent glycerol dehydrogenase is active, which has broader substrate specificity, contributing also to the formation of mannitol and erythritol. The enzymes participating in the catabolic pathway of mannitol (NADP+-dependent mannitol dehydrogenase and hexokinase or fructokinase) are not subject to significant carbon repression (Singh et al., 1988). In these conditions, it is expected that, when the creA gene is deleted, mannitol accumulates inside the cells owing to the activity of the second NADP+-dependent glycerol dehydrogenase, and eventually this polyol is secreted, resulting in a higher extracellular pool in the cultivation medium of the mutant cells.

Qualitative analysis of labelling patterns
The structure of the metabolic network, as well as the activity of some reactions and pathways, was partially deduced from direct inspection of the labelling data (shown in Table 2) as follows.

Glucose 6-phosphate branchpoint.
Glucose 6-phosphate (G6P) represents an important branchpoint for flux distribution, since it is the common intermediate of the Embden–Meyerhof–Parnas (EMP) pathway, the pentose phosphate (PP) pathway and the biosynthesis of cell wall components. When [1-13C]glucose is used as substrate, the label of glucose 6-phosphate (which corresponds to the SFL of the fragment Glc331) provides information on the flux through the PP pathway. In fact, in the oxidative part of this pathway there is loss of label when the first carbon of glucose 6-phosphate is converted into CO2, which causes a higher dilution of the label in the first atom of the hexose phosphate, due to the high reversibility of the reaction converting glucose 6-phosphate into fructose 6-phosphate. Thereby, in principle, a low SFL of glucose-6 phosphate indicates a high flux through the oxidative part of the PP pathway, and vice versa. However, the flux in the oxidative part of the PP pathway may be underestimated owing to the high reversibility of the non-oxidative part of this pathway, which opens the possibility for all positions in glucose 6-phosphate to become labelled. This leads to a decrease in the amount of label lost in the form of CO2 and subsequently to an increase in the SFL of glucose 6-phosphate. Comparing the SFL of the fragment Glc331 of the reference strain (86·96 % after normalization considering the input of label, see Table 2) with the one of the mutant strain (93·33 %), grown on 100 % [1-13C]glucose, it can be deduced that the flux through the oxidative part of the PP pathway decreased upon deletion of the creA gene. When the mutant was grown on a mixture of [1-13C]glucose and unlabelled xylose, the SFL of the fragment Glc331 (86·61 %) was lower than for growth on glucose alone, which indicates that this pathway was more active when the pentose was present in the medium. The changes in the activity of the oxidative branch of the PP pathway were also manifested in changes in the SFLs of the fragments Ser132 [which corresponds to the SFL of G3P(2,3)] and Ala99 or Ala116 [which correspond to the SFL of PYR(2,3)], which are inversely correlated to the changes in the activity of this pathway. In fact, as can be observed in Table 2, deletion of the creA gene led to an increase in the SFLs of the fragments Ser132 (26·94 % for the mutant vs 24·64 % for the reference), Ala99 (28·77 vs 25·67 %) and Ala116 (30·53 vs 27·24 %). On the other hand, addition of xylose to the growth medium of the creA-mutant cells led to a decrease in the SFLs of these fragments (Ser132, 22·61 % for growth on the sugar mixture vs 26·94 % for growth on glucose; Ala99, 21·43 vs 28·77 %; and Ala116, 23·42 vs 30·53 %).

Pyruvate branchpoint.
In A. nidulans, pyruvate (PYR) may be derived from phosphoenolpyruvate (PEP) or from malate in the reactions catalysed by the enzymes pyruvate kinase and malic enzyme, respectively (Hondmann & Visser, 1994). In the former reaction, the three carbon atoms of phosphoenolpyruvate give rise to the three carbon atoms of the pyruvate molecule, without interchange of positions; in the latter reaction, the three carbon atoms of pyruvate are derived from the first three carbon atoms of malate, whereas the fourth carbon atom of malate is released in the form of CO2. In turn, the carbon atoms of malate correspond to the ones in oxaloacetate (OAA) (without change of positions) due to interconversion of these two metabolites in the reaction catalysed by malate dehydrogenase (Hondmann & Visser, 1994). For the cultivations using [1-13C]glucose as substrate, the SFLs of the fragment Phe143 [corresponding to PEP(1,2)] were lower than the SFLs of the fragment Val143 [corresponding to PYR(1,2)], whereas the SFLs of the fragment Thr175 [corresponding to OAA(1,2)] were higher (see Table 2). These observations indicate that both of the above-mentioned reactions were active under the conditions considered, and hence both pyruvate kinase and malic enzyme were included in the model.

Oxaloacetate branchpoint.
During growth on glucose, oxaloacetate may originate from pyruvate in the anaplerotic reaction catalysed by the cytosolic enzyme pyruvate carboxylase (Osmani & Scrutton, 1983), as well as from malate in the reaction catalysed by malate dehydrogenase (both cytosolic and mitochondrial), as described previously. Assuming that pyruvate is only derived from phosphoenolpyruvate and that oxaloacetate is formed in the reaction catalysed by pyruvate carboxylase, in principle, it is expected that the second carbon atom of both pyruvate and oxaloacetate (fragment Asp115) is naturally labelled (1·1 %), if [1-13C] glucose is used as substrate. On the other hand, if oxaloacetate is formed in the reaction catalysed by malate dehydrogenase, the second carbon atom of this compound derives either from the third or the fourth carbon atom of 2-oxoglutarate (owing to scrambling due to the symmetry of the intermediates of TCA cycle succinate and fumarate), which in turn derives from oxaloacetate and acetyl coenzyme A. Thus, if the TCA cycle is active, the second carbon atom of oxaloacetate has a higher fractional labelling than the natural labelling. Similarly, a higher label in the second carbon atom of cytosolic oxaloacetate is to be expected if it is formed from citrate in the reaction catalysed by ATP:citrate lyase (Adams et al., 1997).

From the data presented in Table 2, it can be seen that the SFL of fragment Asp115 was higher for the creA mutant than for the reference strain (15·28 % and 11·36 %, respectively), during growth on glucose, which might indicate that the TCA cycle was more active in the former cells. When xylose was added to the growth medium, the flux through the TCA cycle decreased, which was suggested by a lower label in the second carbon atom of oxaloacetate (SFL of fragment Asp115 equal to 12·08 % for growth on the mixture).

Metabolic model
The metabolic model used for quantification of the fluxes is shown in Supplementary Table (available with the online version of this paper). More elaborate models were considered, but no significant improvement was achieved concerning the total deviation between experimental and computed results.

To describe growth, biomass production was regarded as a drain of biosynthetic precursors required to produce cellular components. The demands on each of these metabolites were estimated based on the biomass composition. In order to account for the effect of the specific growth rate on biomass composition, the cellular compositions considered for the reference and mutant strains of A. nidulans were based on the contents of the main biomass components of A. oryzae determined by Pedersen et al. (1999), for two different specific growth rates (see Table 3). It has been reported that deletion of the creA gene has a significant effect on the morphology of A. nidulans, which may be associated with a change in the cellular composition of the fungus. In all cultivations, dispersed mycelial growth was observed; however, the creA-mutant cells exhibited hyperbranching and increased hyphae diameter compared to the reference cells (data not shown). Moreover, it was observed that the cell wall content and its composition did not have an influence on the morphology, and it was hypothesized that the differences in morphology were most likely attributable to differences in the contents of polyols and lipids (data not shown). Therefore, simulations were performed considering different cellular compositions for the mutant strain, namely concerning polyols and lipids, in order to evaluate the sensitivity of the flux distribution to perturbations in biomass composition. Changes of 50 % in the contents of polyols and lipids did not lead to significant alterations in the fluxes, which differed by at most 9 % from the one based on the cell composition of A. oryzae. Hence, it was concluded that possible variations in the biomass composition have a small effect on the overall differences in the metabolic fluxes between the mutant and the reference strains.

In addition, as deletion of the creA gene is likely to lead to higher production of proteins, the effect of changing the protein content of biomass on the flux distribution was also assessed. The simulations showed that a twofold increase of the protein content would cause only minor changes on the fluxes in the central pathways (maximum of 8 %). The extracellular protein pool (unknown composition) of the creA strain was measured and it represented less than 4 % of the intracellular protein pool (as observed for the reference strain) and therefore its potential influence was regarded as minor.

Quantification of metabolic fluxes
Once all relevant metabolic pathways of the central carbon metabolism of A. nidulans were identified, quantification of metabolic fluxes was accomplished by employing the framework of metabolite and isotope balancing, as described above in Computational methods.

The net flux distributions computed for the reference cells and the creA-mutant cells grown in batch cultivation using glucose as the sole carbon source are shown in Fig. 2. The values presented are normalized with respect to the uptake of carbon, for the sake of comparison of the flux distributions. Thereby, the effect of the specific growth rate is eliminated in absolute terms, but the difference between the relative flux distributions may still be a function of the difference in specific growth rates. The major changes observed in the central carbon metabolism of the mutant compared with the reference strain corresponded to the activity of the oxidative part of the PP pathway, which decreased by about 20 %, while the activities of the EMP pathway and the TCA cycle underwent an increase of about 10 %.



View larger version (22K):
[in this window]
[in a new window]
 
Fig. 2. Comparison of the metabolic net flux distributions computed for the reference strain (upper values) and the creA-mutant strain (lower values, italic) of A. nidulans grown on glucose. All values shown refer to the net forward flux, i.e. the forward flux minus the backward flux. Reactions for which there are both forward and backward fluxes (and thus exchange of labelling) are indicated by *; the values shown for the lumped reaction involving the interconversion of pyruvate and oxaloacetate are net flux (forward flux-backward flux). All fluxes shown are normalized with respect to the carbon uptake rate [600 mmol (g dry wt)–1 h–1]. Abbreviations defined in Supplementary Table (available with the online version of this paper).

 
Flux maps describing the growth of creA-mutant cells on different carbon sources, namely on glucose and on the mixture of glucose and xylose (see Fig. 3) were compared. It was observed that when xylose was added to the medium there was an increase (of about 40 %) in the flux through the oxidative part of the PP pathway, as well as a drop in the fluxes through the EMP pathway and the TCA cycle (in the range of 20–30 %, in both cases).



View larger version (22K):
[in this window]
[in a new window]
 
Fig. 3. Comparison of the metabolic net flux distributions computed for the creA-mutant strain of A. nidulans grown on glucose (upper values) and on a mixture of glucose and xylose (lower values, italic). All values shown refer to the net forward flux, i.e. the forward flux minus the backward flux. Reactions for which there are both forward and backward fluxes (and thus exchange of labelling) are indicated by *; the values shown for the lumped reaction involving the interconversion of pyruvate and oxaloacetate are net flux (forward flux-backward flux). All fluxes shown are normalized with respect to the carbon uptake rate [600 mmol (g dry wt)–1 h–1]. Abbreviations defined in Supplementary Table (available with the online version of this paper).

 
It is difficult to draw solid conclusions concerning the fluxes through the anaplerotic reaction catalysed by pyruvate carboxylase and the reaction catalysed by malic enzyme, as these reactions were lumped together with the reaction catalysed by malate dehydrogenase in the model (resulting in a reversible reaction that involved the interconversion of pyruvate and oxaloacetate). However, through the approach adopted herein, it was possible to quantify both the forward and the backward fluxes in reversible reactions and thus gain some insight into the activity of the enzymes involved in this lumped reaction. If the forward flux through this lumped reaction is greater than the backward flux, it can be speculated that changes in the forward flux have an effect essentially on the activity of pyruvate carboxylase (and hence consumption of ATP), whereas changes in the backward flux affect mainly the activities of malic enzyme and malate dehydrogenase (and the conversion of NADH to NADPH). Therefore, the drop of about 10 % observed in the forward flux through this lumped reaction, upon deletion of the creA gene, may be an indication of decreased activity of the enzyme pyruvate carboxylase in the mutant strain. On the other hand, the backward flux through the lumped reaction remained approximately the same, which may indicate that the fluxes through malic enzyme and malate dehydrogenase did not change significantly. When xylose was added to the growth medium of the mutant strain, the forward flux through the lumped reaction (and the activity of pyruvate carboxylase) was not affected, whereas the backward flux (and the combined activities of malic enzyme and malate dehydrogenase) was subject to a decrease of 30 %.

Metabolic changes
The model proposed for the central carbon metabolism of A. nidulans and used for flux quantification comprised fueling reactions involved in the main catabolic pathways for the degradation of hexoses and pentoses. The major purposes of these fueling reactions are the generation of Gibbs free energy (in the form of ATP) and reducing power (in the form of NADPH), as well as the supply of biosynthetic precursors, which are required for biosynthesis of biomass. NADH is also produced in the fueling reactions, and is subsequently converted into ATP by oxidative phosphorylation or regenerated by substrate level phosphorylation (fermentation). Consequently, changes in the fluxes of these fueling reactions are manifested in alterations in the production of cofactors. The modifications observed in the flux distributions in the primary metabolism of A. nidulans were interpreted in terms of cofactor balancing.

Effect of a genetic change on the flux distribution: deletion of the creA gene.
Concerning the changes in cofactor generation when the mutant cells were compared to the reference cells for growth on glucose there seemed to be a decrease in the supply of NADPH, as well as an increase in the generation of NADH and ATP. In fact, the drop in the flux through the oxidative part of the PP pathway corresponds to a decrease in the supply of NADPH, while the increase in the activity of the EMP pathway results in the generation of additional ATP and NADH. The increased activity of the TCA cycle may have different effects on the recycling of cofactors depending on the cofactor (NADH or NADPH) used by isocitrate dehydrogenase in the conversion of isocitrate into 2-oxoglutarate. It has been reported that isocitrate dehydrogenase is subject to carbon repression in A. nidulans (Kelly, 1994). The flux through malic enzyme, which represents another source of NADPH, did not seem to be affected by the deletion of the creA gene, as mentioned above. Earlier reports state that the synthesis of malic enzyme appears to be relatively weakly regulated by carbon repression, and that induction is the major regulatory mechanism (Kelly & Hynes, 1981). Expression studies performed with reference cells of A. nidulans showed that the genes encoding some enzymes participating in the TCA cycle, namely aconitase and malate dehydrogenase, were down-regulated following the addition of glucose to a growth medium containing ethanol, a derepressing carbon source (Sims et al., 2004).

The increase in the activities of the EMP pathway and the TCA cycle may be explained in part in the light of the higher demand for ATP of the mutant strain. In fact, deletion of the creA gene leads to the derepression of genes involved in the catabolic pathways of carbon sources other than glucose, which may result in the emergence of futile cycles. For example, it has been reported that certain genes encoding enzymes unique to gluconeogenesis, such as fructose 1,6-bisphosphatase (acuG) and phosphoenolpyruvate carboxykinase (acuF), which are active in the catabolism of acetate or ethanol, are repressed by glucose to prevent the simultaneous activity of glycolysis and gluconeogenesis, and hence futile cycling (Sims et al., 2004). Other cycles, such as the mannitol cycle, whose net result is the generation of NADPH at the expense of NADH and ATP, may also arise in the creA-mutant strain, which allow the cells to meet their extra requirements for NADPH due to, for instance, a higher production of enzymes involved in the catabolic pathways of alternative carbon sources.

On the other hand, the changes observed in the relative flux distributions upon deletion of the creA gene are also the outcome of a decreased maximal specific growth rate (and thus decreased demand for NADPH for biosynthetic purposes).

Effect of an environmental change on the flux distribution: addition of xylose to the medium.
Similarly, the changes in the central metabolism observed when xylose was added to the growth medium of the creA-mutant cells were accompanied by changes in the production of cofactors, namely there seemed to be an increase in the supply of NADPH, as well as a decrease in the generation of NADH and ATP. The catabolism of xylose includes the NADPH-dependent reduction of the pentose to xylitol and the NAD+-dependent oxidation of the polyol to xylulose, which is subsequently phosphorylated to xylulose 5-phosphate, an intermediate of the PP pathway. Hence, when xylose was used as carbon source, the demand for NADPH increased and additional NADH was generated, compared to the catabolism of glucose. These changes may partially explain the ones observed in the production of cofactors in the fueling reactions. Indeed, when cells were grown on the mixture of glucose and xylose, the extra requirement for NADPH seemed to be met by an increased flux through the PP pathway, while the lower activity of the EMP pathway and the TCA cycle might be related to the extra generation of NADH in the catabolism of xylose.

Conclusions
In conclusion, we have performed the phenotypic characterization of A. nidulans for different (genetic and environmental) conditions of CreA-mediated glucose repression through in vivo metabolic-flux analysis. This enabled the global mapping of how the carbon repressor CreA influences the central metabolism of A. nidulans. The derepression of key pathways (e.g. the catabolism of glycerol, which leads to the onset of futile cycles, or the catabolism of xylose, which leads to alterations in the demands for cofactors) imposes changes in the central metabolism due to the coupling of the many different reactions, via the redox and energy metabolism of the cells.


   ACKNOWLEDGEMENTS
 
The authors would like to thank Thomas Grotkjær for software support and Gerald Hofmann for fruitful discussions. H. D. was funded through a research fellowship (SFRH/BD/3110/2000) of the III Community Support Framework financed by the European Social Fund and by a Portuguese National Fund from the Ministry of Science and Technology. Moreover, M. Å. acknowledges support from the Alf Åkerman foundation, Sweden. This work was also financed by the Danish Biotechnology Instrument Center, by the EUROFUNG consortium (QTRK3-1999-00729), by Vinnova, Sweden and by Erhvervsfremmestyrelsen, Denmark, via the Øresund Center Contract ETIF.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Adams, I. P., Dack, S., Dickinson, F. M., Midgley, M. & Ratledge, L. (1997). ATP:citrate lyase from Aspergillus nidulans. Biochem Soc Trans 25, S670.[Medline]

Agger, T., Petersen, J. B., O'Connor, S. M., Murphy, R. L., Kelly, J. M. & Nielsen, J. (2002). Physiological characterisation of recombinant Aspergillus nidulans strains with different creA genotypes expressing A. oryzae alpha-amylase. J Biotechnol 92, 279–285.[CrossRef][Medline]

Arst, H. N., Jr & Bailey, C. R. (1977). The regulation of carbon metabolism in Aspergillus nidulans. In Genetics and Physiology of Aspergillus nidulans, pp. 131–146. Edited by J. E. Smith & J. A. Pateman. London: Academic Press.

Bradford, M. M. (1976). A rapid and sensitive for the quantitation of microgram quantitites of protein utilizing the principle of protein-dye binding. Anal Biochem 72, 248–254.[CrossRef][Medline]

Carlsen, M. & Nielsen, J. (2001). Influence of carbon source on alpha-amylase production by Aspergillus oryzae. Appl Microbiol Biotechnol 57, 346–349.[Medline]

Christensen, B. & Nielsen, J. (2000a). Metabolic network analysis. A powerful tool in metabolic engineering. Adv Biochem Eng Biotechnol 66, 209–231.[Medline]

Christensen, B. & Nielsen, J. (2000b). Metabolic network analysis of Penicillium chrysogenum using 13C-labeled glucose. Biotechnol Bioeng 68, 652–659.[CrossRef][Medline]

David, H., Åkesson, M. & Nielsen, J. (2003). Reconstruction of the central carbon metabolism of Aspergillus niger. Eur J Biochem 270, 4243–4253.[Abstract/Free Full Text]

Dowzer, C. E. & Kelly, J. M. (1991). Analysis of the creA gene, a regulator of carbon catabolite repression in Aspergillus nidulans. Mol Cell Biol 11, 5701–5709.[Medline]

Felenbok, B., Flipphi, M. & Nikolaev, I. (2001). Ethanol catabolism in Aspergillus nidulans: a model system for studying gene regulation. Prog Nucleic Acid Res Mol Biol 69, 149–204.[Medline]

Fischer, E. & Sauer, U. (2003). Metabolic flux profiling of Escherichia coli mutants in central carbon metabolism using GC-MS. Eur J Biochem 270, 880–891.[Abstract/Free Full Text]

Gombert, A. K., Moreira dos, S. M., Christensen, B. & Nielsen, J. (2001). Network identification and flux quantification in the central metabolism of Saccharomyces cerevisiae under different conditions of glucose repression. J Bacteriol 183, 1441–1451.[Abstract/Free Full Text]

Hondmann, D. H. & Visser, J. (1994). Carbon metabolism. In Aspergillus: 50 Years On, pp. 61–139. Edited by S. D. Martinelli & J. R. Kinghorn. Amsterdam: Elsevier.

Ilyés, H., Fekete, E., Karaffa, L., Fekete, E., Sandor, E., Szentirmai, A. & Kubicek, C. P. (2004). CreA-mediated carbon catabolite repression of beta-galactosidase formation in Aspergillus nidulans is growth rate dependent. FEMS Microbiol Lett 235, 147–151.[CrossRef][Medline]

Kelly, J. M. (1994). Carbon catabolite repression. In Aspergillus: 50 Years On, pp. 355–367. Edited by S. D. Martinelli & J. R. Kinghorn. Amsterdam: Elsevier.

Kelly, J. M. & Hynes, M. J. (1981). The regulation of phosphoenolpyruvate carboxykinase and the NADP-linked malic enzyme in Aspergillus nidulans. J Gen Microbiol 123, 371–375.[Medline]

Kelly, J. M. & Hynes, M. J. (1982). The regulation of NADP-linked isocitrate dehydrogenase in Aspergillus nidulans. J Gen Microbiol 128, 23–28.[Medline]

Kubicek, C. & Röhr, M. (1986). Citric acid fermentation. Crit Rev Biotechnol 3, 331–373.

Manzoni, M. & Rollini, M. (2002). Biosynthesis and biotechnological production of statins by filamentous fungi and application of these cholesterol-lowering drugs. Appl Microbiol Biotechnol 58, 555–564.[CrossRef][Medline]

Nielsen, J. (2001). Metabolic engineering. Appl Microbiol Biotechnol 55, 263–283.[CrossRef][Medline]

Osmani, S. A. & Scrutton, M. C. (1983). The sub-cellular localisation of pyruvate carboxylase and of some other enzymes in Aspergillus nidulans. Eur J Biochem 133, 551–560.[Abstract]

Pedersen, H., Carlsen, M. & Nielsen, J. (1999). Identification of enzymes and quantification of metabolic fluxes in the wild type and in a recombinant Aspergillus oryzae strain. Appl Environ Microbiol 65, 11–19.[Abstract/Free Full Text]

Prathumpai, W., McIntyre, M. & Nielsen, J. (2004). The effect of CreA in glucose and xylose catabolism in Aspergillus nidulans. Appl Microbiol Biotechnol 63, 748–753.[CrossRef][Medline]

Raghevendran, V., Gombert, A. K., Christensen, B., Kotter, P. & Nielsen, J. (2004). Phenotypic characterization of glucose repression mutants of Saccharomyces cerevisiae using experiments with 13C-labelled glucose. Yeast 21, 769–779.[CrossRef][Medline]

Ruijter, G. & Visser, J. (1997). Carbon repression in aspergilli. FEMS Microbiol Lett 151, 103–114.[CrossRef][Medline]

Scazzocchio, C., Gavrias, V., Cubero, B., Panozzo, C., Mathieu, M. & Felenbok, B. (1995). Carbon catabolite repression in Aspergillus nidulans: a review. Can J Bot 73, S160–S166.

Shroff, R. A., O'Connor, S. M., Hynes, M. J., Lockington, R. A. & Kelly, J. M. (1997). Null alleles of creA, the regulator of carbon catabolite repression in Aspergillus nidulans. Fungal Genet Biol 22, 28–38.[CrossRef][Medline]

Sims, A. H., Robson, G. D., Hoyle, D. C., Oliver, S. G., Turener, G., Prade, R. A., Russell, H. H., Dunn-Coleman, N. & Gent, M. E. (2004). Use of expressed sequence tag analysis and cDNA microarrays of the filamentous fungus Aspergillus nidulans. Fungal Genet Biol 41, 199–212.[CrossRef][Medline]

Singh, M., Scrutton, N. S. & Scrutton, M. C. (1988). NADPH generation in Aspergillus nidulans: is the mannitol cycle involved? J Gen Microbiol 134, 643–654.[Medline]

Szyperski, T. (1998). 13C-NMR, MS and metabolic flux balancing in biotechnology research. Q Rev Biophys 31, 41–106.[CrossRef][Medline]

van der Veen, P., Ruijter, G. J. & Visser, J. (1995). An extreme creA mutation in Aspergillus nidulans has severe effects on D-glucose utilization. Microbiology 141, 2301–2306.[Medline]

Wiechert, W. (2001). 13C metabolic flux analysis. Metab Eng 3, 195–206.[CrossRef][Medline]

Wiechert, W. & de Graaf, A. A. (1996). In vivo stationary flux analysis by 13C labeling experiments. Adv Biochem Eng Biotechnol 54, 109–154.[Medline]

Wiechert, W. & de Graaf, A. A. (1997). Bidirectional reaction steps in metabolic networks: I. Modeling and simulation of carbon isotope labeling experiments. Biotechnol Bioeng 55, 101–117.[CrossRef]

Wiechert, W., Mollney, M., Petersen, S. & de Graaf, A. A. (2001). A universal framework for 13C metabolic flux analysis. Metab Eng 3, 265–283.[CrossRef][Medline]

Zamboni, N. & Sauer, U. (2003). Knockout of the high-coupling cytochrome aa3 oxidase reduces TCA cycle fluxes in Bacillus subtilis. FEMS Microbiol Lett 226, 121–126.[CrossRef][Medline]

Received 24 November 2004; revised 2 March 2005; accepted 7 April 2005.



This Article
Abstract
Full Text (PDF)
Supplementary table
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by David, H.
Articles by Nielsen, J.
Articles citing this Article
PubMed
PubMed Citation
Articles by David, H.
Articles by Nielsen, J.
Agricola
Articles by David, H.
Articles by Nielsen, J.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2005 Society for General Microbiology.