Perceiving molecular evolution processes in Escherichia coliby comprehensive metabolite and gene expression profiling
© Vijayendran et al.; licensee BioMed Central Ltd. 2008
Received: 10 September 2007
Accepted: 10 April 2008
Published: 10 April 2008
Evolutionary changes that are due to different environmental conditions can be examined based on the various molecular aspects that constitute a cell, namely transcript, protein, or metabolite abundance. We analyzed changes in transcript and metabolite abundance in evolved and ancestor strains in three different evolutionary conditions - excess nutrient adaptation, prolonged stationary phase adaptation, and adaptation because of environmental shift - in two different strains of bacterium Escherichia coli K-12 (MG1655 and DH10B).
Metabolite profiling of 84 identified metabolites revealed that most of the metabolites involved in the tricarboxylic acid cycle and nucleotide metabolism were altered in both of the excess nutrient evolved lines. Gene expression profiling using whole genome microarray with 4,288 open reading frames revealed over-representation of the transport functional category in all evolved lines. Excess nutrient adapted lines were found to exhibit greater degrees of positive correlation, indicating parallelism between ancestor and evolved lines, when compared with prolonged stationary phase adapted lines. Gene-metabolite correlation network analysis revealed over-representation of membrane-associated functional categories. Proteome analysis revealed the major role played by outer membrane proteins in adaptive evolution. GltB, LamB and YaeT proteins in excess nutrient lines, and FepA, CirA, OmpC and OmpA in prolonged stationary phase lines were found to be differentially over-expressed.
In summary, we report the vital involvement of energy metabolism and membrane-associated functional categories in all of the evolutionary conditions examined in this study within the context of transcript, outer membrane protein, and metabolite levels. These initial data obtained may help to enhance our understanding of the evolutionary process from a systems biology perspective.
Most micro-organisms grow in environments that are not favorable for their growth. The level of nutrients available to them is rarely optimal. These microbes must adapt to environmental conditions that consist of excess, suboptimal (limiting) or fluctuating levels of nutrients, or famine. Evolution can be studied by observing its processes and consequences in the laboratory, specifically by culturing a micro-organism in varying nutrient environments [1–4]. Extensively studied microbial evolutionary processes include nutrient-limited adaptive evolution [5–7] and famine-induced prolonged stationary phase evolution [8–10]. During prolonged carbon starvation, micro-organisms can undergo rapid evolution, with mutants exhibiting a 'growth advantage in stationary phase' (GASP) phenotype . These mutants, harboring a selective advantage, out-compete their siblings and take over the culture through their progeny [11–13]. Adaptive evolution of micro-organisms is a process in which specific mutations result in phenotypic attributes that are responsible for fitness in a particular selective environment . Laboratory studies conducted under these evolutionary conditions can address fundamental questions regarding adaptation processes and selection pressures, thereby explaining modes of evolution.
In this study we used Escherichia coli K-12 strains (MG1655 and DH10B) subjected to the following processes: a serial passage system (excess nutrient adaptive evolution studies), constant batch culture (prolonged stationary phase evolution studies), and culture with nutrient alteration after adaptation to a particular nutrient (examining pleiotropic effects due to environmental shift). During adverse conditions, micro-organisms are known to exploit limited resources more quickly and are observed to assimilate various metabolites. Some of these residual metabolites comprise an alternative resource that the organism can metabolize . Continual assimilation of metabolites and the various compounds metabolized by the organism offer a specific niche that allows the organism to evolve with genetic capacity to utilize those assimilated metabolites . Hence, a detailed metabolite analysis of these evolved populations would enhance our understanding of these evolutionary processes. Along with data generated from transcriptomics approaches, metabolomics data will be vital in obtaining a global view of an organism at a particular time point, during which metabolite behavior closely reflects the actual cellular environment and the observed phenotype of that organism.
Strains and their evolved conditions
MG1655 grown in glucose (ancestor)
DH10B grown in glucose (ancestor)
MG1655 grown in galactose (ancestor)
DH10B grown in galactose (ancestor)
MG1655 adapted about 1,000 generations in glucose (evolved)
DH10B adapted about 1,000 generations in glucose (evolved)
MGAdp (glucose evolved strains) grown in galactose (evolved)
DHAdp (glucose evolved strains) grown in galactose (evolved)
MG1655 grown in prolonged stationary phase (37 days; evolved)
DH10B grown in prolonged stationary phase (37 days; evolved)
In this study we developed a picture of laboratory molecular evolutionary processes in two different strains by integrating multidimensional metabolome and gene expression data, in order to identify metabolites and genes that are vital to the evolutionary process.
The Adp line cultures (MGAdp and DHAdp) were maintained in prolonged exponential growth phase by daily passage into fresh medium for about 1,000 generations, undergoing many rounds of exponential phase growth. The Stat line cultures (MGStat and DHStat) were maintained in constant batch culture for 37 days, during which no nutrients were added after the initial inoculation and no cells were removed (unlike the preceding setup). For the AdpGal line cultures (MGAdpGal and DHAdpGal), Adp lines (glucose adapted) were grown in medium containing galactose as carbon source, thus creating an environmental shift for the cells with respect to the standard nutrient source. During this period of adaptation, both Adp lines (evolved) exhibited increased fitness in their growth, whereas Stat lines (evolved) exhibited growth behavior similar to that of their ancestors. The samples of MG, DH, MGGal, DHGal, MGAdp, DHAdp, MGAdpGal, DHAdpGal, MGStat, and DHStat lines grown in the respective carbon sources (Table 1) were harvested during the mid-exponential phase of growth for both metabolome and transcriptome analysis.
In the metabolome analysis, from about 200 peaks in each chromatogram about 100 metabolites were identified by gas chromatography-mass spectrometry. In the transcriptome analysis a whole genome microarray consisting of 4,288 open reading frames of Escherichia coli K-12 was used. To examine the multivariate measures of variability of the metabolite and gene expression profiles for the obtained data, and for clustering the biological samples, we applied principal components analysis (PCA). In order to identify parallel metabolite accumulation and gene expression, we applied pair-wise correlation plot analysis. To examine the extent of parallelism among the evolved lines, gene-metabolite correlation networks were constructed and their topologic properties were studied. By mapping the correlation networks to Gene Ontology (GO) functional annotations, the functional relevance of the networks was determined. Subsequently, the functional modules that were statistically significantly over-represented in respective evolution processes were identified.
Statistically significant metabolites involved in various evolved conditions
Total number of metabolites taken into account
Number of over-abundant metabolites (d i ≥ 1)
Number of less abundant metabolites (d i ≤ -1)
Total number of differentially abundant metabolites
Number of intersecting metabolites
Total number of intersecting metabolites
Gene expression profiling
Several studies have used gene expression profiling to study molecular evolution, but these studies were confined to a single type of evolutionary process and were focused on a single molecular aspect that characterizes a cell (transcript abundance) [17–20]. In our study we focused on three evolutionary conditions in two strains and two molecular aspects of a cell (transcript and metabolite abundance). This approach allowed us to integrate metabolome and transcriptome datasets to elucidate the process of adaptive evolution under laboratory conditions.
Statistically significant genes involved in various evolved conditions
Total number of genes taken into account
Number of over-expressed genes (d i ≥ 1)
Number of under-expressed genes (d i ≤ -1)
Total number of differentially expressed genes
Number of intersecting genes
Total number of intersecting genes
Extent of changes
Direction of the observed extent of changes
To examine the level of observed change among the strains, we calculated the pair-wise Pearson correlation coefficient (r; PCC) for all of the metabolites and significantly correlating genes. All genes having a threshold of r ≤ -0.9 or ≥ 0.9 and all metabolites were plotted on both axes of a matrix containing either all pair-wise metabolite or gene expression profile correlations. When these correlations (r) are color coded, this facilitates use of visual inspection to determine the degree of positive and negative correlation among the samples in question. The correlation map of Adp, AdpGal, and Stat line comparisons exhibited various degrees of negative correlation (Figure 5g-l). Among these, Stat line comparisons (MG/MGStat versus DH/DHStat) exhibited a high degree of negative correlation when compared with AdpGal and Adp line comparisons in both metabolite and gene expression correlation maps (Fig. 5i, l), suggesting elevated levels of variability due to selection among the Stat lines. The correlation map of the Adp line comparison (MG/MGAdp versus DH/DHAdp) revealed a lower degree of negative correlation than did the other line comparisons in both metabolite and gene expression correlation maps (Figure 5g, j), denoting a reduced level of variability caused by selection among the Adp lines.
Gene-metabolite correlation network analysis
All-against-all metabolite and gene expression profile comparisons for Adp, AdpGal, and Stat matrices were used to generate evolution-specific co-expression networks constructed using r (PCC). There was a significant, strong dependence between co-expression and functional relevance of the networks, attesting to the potential of co-expression network analysis (Figure 6a). In co-expression networks, nodes correspond to genes or metabolites, and edges link two genes or metabolites if they have a threshold correlation coefficient (r) at or above which genes or metabolites are considered to be changed differentially, exhibiting similar behavior. Correlation networks as such inherently contain corresponding large noise components, which were largely eliminated by setting the threshold of r at 0.9. The correlation networks based on the high threshold r of 0.9 reported here are less likely to contain noise while being sufficiently dense for analyses of topologic properties.
Evaluation of evolution-specific networks
With respect to a number of parameters describing their common topologic properties, all evolution-specific co-expression networks (Adp: 4,170 nodes and 23,086 edges; AdpGal: 4,136 nodes and 20,501 edges; and Stat: 4,166 nodes and 54,028 edges) were found to be similar except for the average degree (see Additional data file 8). The average degree (<k>) is the average number of edges per node . The Stat co-expression network exhibits higher <k> than do the Adp and AdpGal networks, which is consistent with its greater numbers of edges. The parameter <k> gives only a rough approximation of how dense the network is. The average clustering coefficient (<C>) is a measure of network density and characterizes the overall tendency of nodes to form clusters . For all of the evolution-specific coexpression networks, <C> was approximately constant and high (about 0.05) when compared with randomly generated networks of similar size, for which the observed <C> was quite low (about 0.0008). The average path length <l> is the average shortest path between all pairs of nodes . For all of the evolution-specific co-expression networks, the <l> was approximately constant and low (about 6.97; Figure 6e). When analyzing the networks' generic features, the clustering coefficients C(k) of all of the networks were more or less constant, implying that they did not exhibit a hierarchical structure (Figure 6b). The node degree (k) distribution of all of the networks appeared to have an exponential drop-off in the tail, following a power law (Figure 6c). Overall, these evaluations suggest that the global properties of these evolution-specific co-expression networks are indistinguishable.
Evolution-specific intersection networks
Strain-specific and evolution-specific networks were screened for the set of nodes N, for which there is a link (r ≥ 0.9) between two nodes a and b in both strains in the particular evolution type, in order to build evolution-specific intersection networks. By examining the intersection networks of both strains, we found that the path length distribution varied among networks. All intersection networks differed in <k>, which is consistent with their varying numbers of edges. The average clustering coefficient <C> was slightly higher in the Adp intersection network (<C> Adp intersection = 0.113, AdpGal intersection = 0.07, and Stat intersection = 0.089), demonstrating high network density and tendency of nodes to form clusters in the Adp intersection network (see Additional data file 8). The average path length <l> was almost equal in all cases, but its distribution in the Adp intersection network differed, indicating high network navigability (Figure 6f, g). Based on the observations of the global properties of the evolution-specific intersection networks, the Adp intersection network can be distinguished from other intersection networks, demonstrating its unique characteristics.
Parallelism and functional relevance of molecular evolution
The generated networks were examined for functional coherence by assigning GO functional annotations to the networks' entities, and the level of parallelism in the representation of these functional categories was elucidated. Parallel evolution is the independent development of similar traits in distinct but evolutionarily related lineages through similar selective factors on both lines . Parallel evolution of similar traits across both lines are used as an indicator that the change is adaptive . Previous studies in E. coli and Saccharomyces cerevisiae have demonstrated parallel changes in independently adapted lines of replicate populations by utilizing gene expression profiling [17, 19]. Here, we examined the parallelism of metabolite and gene expression levels among the evolved lines of different populations that exhibited similar growth behavior.
Parallelism in outer-membrane protein expression
In this study we examined the metabolome and transcriptome profiles of excess nutrient adaptive evolution, pleiotropic environmental shift changes, and prolonged stationary phase evolution in two strains of E. coli K-12. We found significant influence of genes involved in transport and membrane related functional categories in all evolutionary conditions evaluated in this study. In earlier studies, during prolonged nutrient limited chemostat culture of bacterial populations, it was reported that the populations tend toward mutational adaptation in transport systems in order to increase the efficiency with which they utilize limited nutrients [25–28]. For example, glucose limited chemostat evolved strains attained diverse mutations at several loci in LamB porin, which increased glucose permeability [27–29]. An earlier study of adaptation of Ralstonia in selective environments  resulted in morphologic changes in the outer cell envelope in all of the lineages examined.
In long-term stationary phase cultures, cells lose their integrity and release their cellular components into the medium as cells enter the death phase . For cell maintenance and growth, the surviving cells scavenge nutrient sources from the cellular debris (amino acids from proteins, carbohydrates from the cell wall, and lipids from cell membrane material and DNA) of their dead siblings . This nutrient scavenging process due to nutrient limitation enhances the availability of carbon sources by reconstruction of the OM composition (glycerophospholipids, lipopolysaccharides and proteins) and there by improving the permeability of the OM . The OM of E. coli consists of a lipid bilayer structure composed of an outer layer consisting of lipopolysaccharide and an inner layer consisting of phospholipids . The genes involved in the biosynthetic pathways of fatty acids (key building blocks for the phospholipid components of cell membranes) and lipids were over-expressed in Stat lines (see Additional data file 10). Other major components of the OM are proteins; these largely consist of porins, which co-exist with lipopolysaccharide . The OM of the cell is the first point of contact with the external environment, and therefore its cellular constituents may be the most sensitive to the external environment. Consistent with this hypothesis, OM proteins FepA, CirA, OmpC, and OmpA were differentially over-expressed in Stat lines (Figure 8), and the genes belonging to the membrane-associated GO functional categories were significantly over-represented in the corresponding evolutionary networks as well (Figure 7f). This demonstrates the reliability of the correlation network analysis, which was sufficiently robust to identify significant changes in the integrated metabolite and gene profiling dataset.
Mutation rates in stationary phase are known to be influenced by the genetic background of the strain . Initial isogenic long-term stationary phase cultures are highly dynamic and are known to yield different 'growth advantage in stationary phase' mutations due to significant genotypic diversity in these cultures . Consistent with this hypothesis, when we applied PCA (Figure 5c, f) and correlation plot analysis (Figure 5i, l), the metabolite and gene expression levels of Stat lines exhibited low degrees of parallelism when compared with their ancestor lines. Likewise, when GO functional annotations were mapped onto the Stat co-expression network, we found that none of the GO functional categories was significantly over-represented, denoting a low level of parallelism (Figure 7c). However, when applied to the Stat intersection co-expression network, membrane-associated GO functional categories were significantly over-represented (Figure 7f). These observations demonstrate the parallelism in membrane-associated categories in the Stat intersection co-expression network but not in the Stat co-expression network. It suggests the existence of parallelism in membrane-associated categories but not in similar membrane-associated genes in Stat lines. From this we can conclude that distinct but functionally related genes are involved in the parallelism in the Stat intersection co-expression network.
We analyzed two different strains under three different evolutionary conditions. Integration of metabolome and gene expression data within the context of evolution facilitated investigation of the path of evolution and their degree of parallelism. Classifying microarray data according to significantly over-represented GO functional categories showed that the transport related categories had the greater overall representation. Similarly, by mapping the GO annotation to the correlation networks, we found that the membrane associated functional categories were significantly over-represented. The OM of the cell is the first point of contact with the external environment, which acts as a barrier that is quite resistant to insult and acts as a channel for nutrient transport. Components of the OM may therefore be the cellular constituents that are most sensitive to the external environment. Analyses of the OM proteins of the ancestor and evolved strains revealed clear differential regulation of the OM proteins.
In summary, all of the evolutionary experiments reported in this study demonstrate the vital role played by the involvement of the membrane associated components in the evolutionary process. These studies show that adaptive evolution in excess nutrient conditions are appropriate for examining the extent of parallelism in the evolutionary process of the evolved populations, whereas the prolonged stationary phase conditions are useful in understanding the evolution of microbial diversity among evolved populations and the dynamic state of the evolved condition. Such studies will certainly advance our understanding of the process of evolution immensely and, along with constructed models , will be an ideal initial source of data for systems biology study of microbial evolution.
Materials and methods
Strain and culture conditions
Both the bacterial strains MG1655 and DH10B used in this study are derivatives of E. coli K-12. All of the experiments were conducted in 250 ml of M9 minimal medium supplemented with 4 g/l glucose or galactose in covered 1 l Erlenmeyer flasks at 37°C. Adaptation to excess nutrient experiments were carried out in the presence of 4 g/l glucose through serial passage at exponential phase for about 1,000 generations. The cells were grown overnight and were diluted by passage into fresh medium. Passage of each culture into fresh medium was conducted in a laminar flow station using standard sterile technique practices. Serial passage was conducted for 37 days at exponential phase for about 1,000 generations. For adaptation due to environmental shift experiments, the strains that were adapted to excess nutrient (glucose) condition for about 1,000 generations were grown in 4 g/l galactose. For prolonged stationary phase adaptation experiments, both the strains were incubated for 37 days in M9 minimal medium with 4 g/l glucose as initial source of carbon. The evolved populations were frozen using liquid nitrogen and stored in a freezer at -80°C.
Approximately equal numbers of cells (7 × 109) were taken from the exponential phase of growth for all of the experiments. Cells were disrupted using acid washed glass beads at maximum speed in a Ribolyser (Q-BIOgene, Heidelberg, Germany) at a setting of 6.5 m/second, twice for 45 seconds in the presence of 80% methanol. Subsequently, metabolites were derived using methoxylamine hydrochloride and N-methyl-N-(trimethylsilyl)trifluoroacetamide in the presence of ribitol as the internal standard. Sample volumes of 1 μl were analysed using a TraceGC gas chromatograph coupled to a PolarisQ ion trap mass spectrometer (Thermo Finnigan, Dreieich, Germany). Derived metabolites were evaporated at 250°C in splitless mode and separated on a 30 m × 0.25 mm Equity-5 column with 0.25 μm coating (Supelco, Bellefonte, California, USA). Metabolites were identified by comparison with purified standards, the NIST 2005 database (NIST) and the Golm Metabolome Database . Selected metabolite peak areas were automatically quantified using the processing setup implemented in the Xcalibur 1.4 software (Thermo Finnigan, Dreieich, Germany). The relative response ratios calculated from the peak areas were normalized by the internal standard ribitol and dry mass of the sample. For both the strains in all the biologic experiments, six replicates were used, which consisted of three independent biologic replicates and three technical replicates. The variation among the biological replicates was estimated to be relatively low (see Additional data file 11 [part a]).
Gene expression profiling
E. coli K12 V2 OciChip™ arrays containing 4,288 gene specific oligonucleotide probes representing the complete E. coli K-12 genome were utilized in this study (Ocimum Biosolutions, Hyderabad, India). Total RNA was isolated using RNeasy kit (Qiagen, Hilden, Germany), in accordance with the manufacturer's instructions. Reverse transcription, labeling, and scanning were performed as described previously . Hybridization was carried out in accordance with the manufacturer's instructions (Ocimum Biosolutions, Hyderabad, India).
Microarray data analysis
Mean signal and mean local background intensities were determined for each spot of the microarray images, by using the ImaGene 6.0 software for spot detection, image segmentation, and signal quantification (Biodiscovery, Los Angeles, California, USA). After subtraction of the local background intensities from the signal intensities, the average intensity in both channels was subsequently normalized using the LOWESS (locally weighted scatterplot smoothing) method using the GeneSight 4.0 software package (Biodiscovery, Los Angeles, California, USA). The normalized log2 ratios were used to represent the data graphically and to calculate Wilcoxon rank sum test P value using MapMan software , with functional classifications based on MultiFun and GO terms, a cell function assignment scheme, with slight modification [38, 39]. The SAM add-in to Microsoft Excel was used for comparisons of replicate array experiments . For both of the strains in all of the biologic experiments, three or more replicates were used, which consisted of three biologic replicates. The variation among the biologic replicates was estimated to be relatively low (see Additional data file 11 [part b]). The ArrayExpress repository  accession number for the microarray data is E-MEXP-1166, which consists of 29 hybridizations.
All of the networks reported in this study were constructed based on PCC r ≥ 0.9 measure (nodes that correspond to genes or metabolites with r ≥ 0.9 were linked by an edge). All-against-all metabolite and gene expression profile r values of evolution-specific matrices were used to generate evolution-specific co-expression network. Strain-specific and evolution-specific matrices were used to generate evolution-specific intersection co-expression network. Intersection co-expression networks are the network over the set of nodes N, where there is a link (r ≥ 0.9) between two nodes i and j if they are connected in both of the strains in the particular evolutionary condition in context. Topologic properties of the networks were analyzed using the Pajek program .
Network functional analysis
Network visualization and functional analysis was achieved using Cytoscape . Networks were screened for highly linked clusters of genes or metabolites using MCODE . Genes in the networks were functionally categorized using their GO biologic process annotation terms , and the over-represented GO terms were identified with BINGO . The hypergeometric test was used for this purpose, with the Benjamini and Hochberg false discovery rate correction (a false discovery rate-controlled P value cutoff of ≤ 0.05).
Outer membrane protein analysis
Approximately equal numbers of extracted cells (7 × 109) were disrupted by ultrasonication with 5 ml of 50 mmol/l Tris/HCl (pH 7.3), containing 0.7 mg of DNase I (Sigma, Taufkirchen, Germany) and 0.5 mmol/l protease inhibitor (Pefabloc SC; Centerchem, Inc., Norwalk, CT, USA). After the unbroken cells were removed by centrifugation, the supernatant was treated with ice-cold 0.1 mol/l sodium carbonate (pH 11). Eventually, the carbonate treated membranes were collected and subsequently analysed by SDS one-dimensional gel electrophoresis. Excised protein bands were subjected to tryptic digestion and mass spectra were obtained on a Ultraflex MALDI-TOF/TOF (Bruker Daltonics, Bremen, Germany). Peptide masses were searched against the E. coli database located on our local server using MASCOT search engine (Matrix Science Ltd., London, U.K) with a mass cutoff of 100 ppm.
Additional data files
The following additional data are available with the online version of this paper. Additional data file 1 is a table listing the identified metabolites of the ancestral and evolved strains by gas chromatography-mass spectrometry. Additional data file 2 is a table listing significantly altered metabolites in all of the evolved conditions. Additional data file 3 is a table listing significantly altered genes in all of the evolved conditions. Additional data file 4 is a table listing significant GO functional categories involved in all of the evolved conditions. Additional data file 5 is a figure showing the integration of transcriptome and metabolome data during the comparison of ancestral and evolved strains in excess nutrient adaptive evolution. Additional data file 6 is a figure showing the gene expression and metabolite abundance level in the pentose phosphate pathway in excess nutrient adapted strains. Additional data file 7 is a figure showing PCA analyses for both the ancestor and evolved lines of both the strains grown in two different media. Additional data file 8 is a table listing common topologic properties of all evolution co-expression networks. Additional data file 9 is a figure showing the gene expression and metabolite abundance level in histidine biosynthesis pathway in excess nutrient adapted strains. Additional data file 10 is a figure showing the integration of transcriptome and metabolome data during the comparison of ancestral and evolved strains in prolonged stationary phase evolution. Additional data file 11 is a figure showing metabolite abundance level and gene expression level among the biologic replicates.
nicotinamide adenine dinucleotide phosphate
principal components analysis
Pearson correlation coefficient
significance analysis of microarrays
We thank Steven E Finkel (University of Southern California), Rashmi Prasad (University of Bielefeld), and Rileen Sinha (Fritz Lipmann Institute) for helpful comments and critical reading of the manuscript. We should like to thank Manuela Meyer and Eberhard Wünsch for their technical assistance. The work was supported by a scholarship from the NRW International Graduate School in Bioinformatics and Genome Research.
- Elena SF, Lenski RE: Evolution experiments with microorganisms: the dynamics and genetic bases of adaptation. Nat Rev Genet. 2003, 4: 457-469. 10.1038/nrg1088.PubMedView ArticleGoogle Scholar
- Finkel SE: Long-term survival during stationary phase: evolution and the GASP phenotype. Nat Rev Microbiol. 2006, 4: 113-120. 10.1038/nrmicro1340.PubMedView ArticleGoogle Scholar
- Wright BE: Stress-directed adaptive mutations and evolution. Mol Microbiol. 2004, 52: 643-650. 10.1111/j.1365-2958.2004.04012.x.PubMedView ArticleGoogle Scholar
- Zinser ER, Kolter R: Escherichia coli evolution during stationary phase. Res Microbiol. 2004, 155: 328-336. 10.1016/j.resmic.2004.01.014.PubMedView ArticleGoogle Scholar
- Lenski RE, Travisano M: Dynamics of adaptation and diversification: a 10,000-generation experiment with bacterial populations. Proc Natl Acad Sci USA. 1994, 91: 6808-6814. 10.1073/pnas.91.15.6808.PubMedPubMed CentralView ArticleGoogle Scholar
- Sniegowski PD, Gerrish PJ, Lenski RE: Evolution of high mutation rates in experimental populations of E. coli. Nature. 1997, 387: 703-705. 10.1038/42701.PubMedView ArticleGoogle Scholar
- Papadopoulos D, Schneider D, Meier-Eiss J, Arber W, Lenski RE, Blot M: Genomic evolution during a 10,000-generation experiment with bacteria. Proc Natl Acad Sci USA. 1999, 96: 3807-3812. 10.1073/pnas.96.7.3807.PubMedPubMed CentralView ArticleGoogle Scholar
- Finkel SE, Kolter R: Evolution of microbial diversity during prolonged starvation. Proc Natl Acad Sci USA. 1999, 96: 4023-4027. 10.1073/pnas.96.7.4023.PubMedPubMed CentralView ArticleGoogle Scholar
- Loewe L, Textor V, Scherer S: High deleterious genomic mutation rate in stationary phase of Escherichia coli. Science. 2003, 302: 1558-1560. 10.1126/science.1087911.PubMedView ArticleGoogle Scholar
- Bjedov I, Tenaillon O, Gerard B, Souza V, Denamur E, Radman M, Taddei F, Matic I: Stress-induced mutagenesis in bacteria. Science. 2003, 300: 1404-1409. 10.1126/science.1082240.PubMedView ArticleGoogle Scholar
- Lombardo MJ, Aponyi I, Rosenberg SM: General stress response regulator RpoS in adaptive mutation and amplification in Escherichia coli. Genetics. 2004, 166: 669-680. 10.1534/genetics.166.2.669.PubMedPubMed CentralView ArticleGoogle Scholar
- Zinser ER, Kolter R: Prolonged stationary-phase incubation selects for lrp mutations in Escherichia coli K-12. J Bacteriol. 2000, 182: 4361-4365. 10.1128/JB.182.15.4361-4365.2000.PubMedPubMed CentralView ArticleGoogle Scholar
- Zinser ER, Kolter R: Mutations enhancing amino acid catabolism confer a growth advantage in stationary phase. J Bacteriol. 1999, 181: 5800-5807.PubMedPubMed CentralGoogle Scholar
- Raamsdonk LM, Teusink B, Broadhurst D, Zhang N, Hayes A, Walsh MC, Berden JA, Brindle KM, Kell DB, Rowland JJ, Westerhoff HV, van Dam K, Oliver SG: A functional genomics strategy that uses metabolome data to reveal the phenotype of silent mutations. Nat Biotechnol. 2001, 19: 45-50. 10.1038/83496.PubMedView ArticleGoogle Scholar
- Fernie AR, Trethewey RN, Krotzky AJ, Willmitzer L: Metabolite profiling: from diagnostics to systems biology. Nat Rev Mol Cell Biol. 2004, 5: 763-769. 10.1038/nrm1451.PubMedView ArticleGoogle Scholar
- Tusher VG, Tibshirani R, Chu G: Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci USA. 2001, 98: 5116-5121. 10.1073/pnas.091062498.PubMedPubMed CentralView ArticleGoogle Scholar
- Ferea TL, Botstein D, Brown PO, Rosenzweig RF: Systematic changes in gene expression patterns following adaptive evolution in yeast. Proc Natl Acad Sci USA. 1999, 96: 9721-9726. 10.1073/pnas.96.17.9721.PubMedPubMed CentralView ArticleGoogle Scholar
- Riehle MM, Bennett AF, Lenski RE, Long AD: Evolutionary changes in heat-inducible gene expression in lines of Escherichia coli adapted to high temperature. Physiol Genomics. 2003, 14: 47-58.PubMedView ArticleGoogle Scholar
- Cooper TF, Rozen DE, Lenski RE: Parallel changes in gene expression after 20,000 generations of evolution in Escherichia coli. Proc Natl Acad Sci USA. 2003, 100: 1072-1077. 10.1073/pnas.0334340100.PubMedPubMed CentralView ArticleGoogle Scholar
- Fong SS, Joyce AR, Palsson BO: Parallel adaptive evolution cultures of Escherichia coli lead to convergent growth phenotypes with different gene expression states. Genome Res. 2005, 15: 1365-1372. 10.1101/gr.3832305.PubMedPubMed CentralView ArticleGoogle Scholar
- Wolfe CJ, Kohane IS, Butte AJ: Systematic survey reveals general applicability of 'guilt-by-association' within gene coexpression networks. BMC Bioinformatics. 2005, 6: 227-10.1186/1471-2105-6-227.PubMedPubMed CentralView ArticleGoogle Scholar
- Barabasi AL, Oltvai ZN: Network biology: understanding the cell's functional organization. Nat Rev Genet. 2004, 5: 101-113. 10.1038/nrg1272.PubMedView ArticleGoogle Scholar
- Schluter D: Parallel evolution and inheritance of quantitative traits. Am Nat. 2004, 163: 809-822. 10.1086/383621.PubMedView ArticleGoogle Scholar
- Bull JJ, Badgett MR, Wichman HA, Huelsenbeck JP, Hillis DM, Gulati A, Ho C, Molineux IJ: Exceptional convergent evolution in a virus. Genetics. 1997, 147: 1497-1507.PubMedPubMed CentralGoogle Scholar
- Helling RB, Vargas CN, Adams J: Evolution of Escherichia coli during growth in a constant environment. Genetics. 1987, 116: 349-358.PubMedPubMed CentralGoogle Scholar
- Sonti RV, Roth JR: Role of gene duplications in the adaptation of Salmonella typhimurium to growth on limiting carbon sources. Genetics. 1989, 123: 19-28.PubMedPubMed CentralGoogle Scholar
- Notley-McRobb L, Ferenci T: Adaptive mgl-regulatory mutations and genetic diversity evolving in glucose-limited Escherichia coli populations. Environ Microbiol. 1999, 1: 33-43. 10.1046/j.1462-2920.1999.00002.x.PubMedView ArticleGoogle Scholar
- Notley-McRobb L, Ferenci T: The generation of multiple co-existing mal-regulatory mutations through polygenic evolution in glucose-limited populations of Escherichia coli. Environ Microbiol. 1999, 1: 45-52. 10.1046/j.1462-2920.1999.00003.x.PubMedView ArticleGoogle Scholar
- Notley-McRobb L, Ferenci T: Experimental analysis of molecular events during mutational periodic selections in bacterial evolution. Genetics. 2000, 156: 1493-1501.PubMedPubMed CentralGoogle Scholar
- Riley MS, Cooper VS, Lenski RE, Forney LJ, Marsh TL: Rapid phenotypic change and diversification of a soil bacterium during 1000 generations of experimental evolution. Microbiology. 2001, 147: 995-1006.PubMedView ArticleGoogle Scholar
- Liu X, Ferenci T: An analysis of multifactorial influences on the transcriptional control of ompF and ompC porin expression under nutrient limitation. Microbiology. 2001, 147: 2981-2989.PubMedView ArticleGoogle Scholar
- Nikaido H, Nakae T: The outer membrane of Gram-negative bacteria. Adv Microb Physiol. 1979, 20: 163-250.PubMedView ArticleGoogle Scholar
- Nikaido H: Molecular basis of bacterial outer membrane permeability revisited. Microbiol Mol Biol Rev. 2003, 67: 593-656. 10.1128/MMBR.67.4.593-656.2003.PubMedPubMed CentralView ArticleGoogle Scholar
- Covert MW, Knight EM, Reed JL, Herrgard MJ, Palsson BO: Integrating high-throughput and computational data elucidates bacterial networks. Nature. 2004, 429: 92-96. 10.1038/nature02456.PubMedView ArticleGoogle Scholar
- Kopka J, Schauer N, Krueger S, Birkemeyer C, Usadel B, Bergmuller E, Dormann P, Weckwerth W, Gibon Y, Stitt M, Willmitzer L, Fernie AR, Steinhauser D: GMD@CSB.DB: the Golm Metabolome Database. Bioinformatics. 2005, 21: 1635-1638. 10.1093/bioinformatics/bti236.PubMedView ArticleGoogle Scholar
- Ruberg S, Tian ZX, Krol E, Linke B, Meyer F, Wang Y, Puhler A, Weidner S, Becker A: Construction and validation of a Sinorhizobium meliloti whole genome DNA microarray: genome-wide profiling of osmoadaptive gene expression. J Biotechnol. 2003, 106: 255-268. 10.1016/j.jbiotec.2003.08.005.PubMedView ArticleGoogle Scholar
- Thimm O, Blasing O, Gibon Y, Nagel A, Meyer S, Kruger P, Selbig J, Muller LA, Rhee SY, Stitt M: MAPMAN: a user-driven tool to display genomics data sets onto diagrams of metabolic pathways and other biological processes. Plant J. 2004, 37: 914-939. 10.1111/j.1365-313X.2004.02016.x.PubMedView ArticleGoogle Scholar
- Serres MH, Riley M: MultiFun, a multifunctional classification scheme for Escherichia coli K-12 gene products. Microb Comp Genomics. 2000, 5: 205-222.PubMedView ArticleGoogle Scholar
- Serres MH, Goswami S, Riley M: GenProtEC: an updated and improved analysis of functions of Escherichia coli K-12 proteins. Nucleic Acids Res. 2004, D300-D302. 10.1093/nar/gkh087. 32 Database
- Parkinson H, Kapushesky M, Shojatalab M, Abeygunawardena N, Coulson R, Farne A, Holloway E, Kolesnykov N, Lilja P, Lukk M, Mani R, Rayner T, Sharma A, William E, Sarkans U, Brazma A: ArrayExpress: a public database of microarray experiments and gene expression profiles. Nucleic Acids Res. 2007, D747-D750. 10.1093/nar/gkl995. 35 Database
- Batagelj V, Mrvar A: PAJEK: program for large network analysis. Connections. 1998, 21: 47-57.Google Scholar
- Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B, Ideker T: Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003, 13: 2498-2504. 10.1101/gr.1239303.PubMedPubMed CentralView ArticleGoogle Scholar
- Bader GD, Hogue CW: An automated method for finding molecular complexes in large protein interaction networks. BMC Bioinformatics. 2003, 4: 2-10.1186/1471-2105-4-2.PubMedPubMed CentralView ArticleGoogle Scholar
- Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, Harris MA, Hill DP, Issel-Tarver L, Kasarskis A, Lewis S, Matese JC, Richardson JE, Ringwald M, Rubin GM, Sherlock G: Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet. 2000, 25: 25-29. 10.1038/75556.PubMedPubMed CentralView ArticleGoogle Scholar
- Maere S, Heymans K, Kuiper M: BiNGO: a Cytoscape plugin to assess overrepresentation of gene ontology categories in biological networks. Bioinformatics. 2005, 21: 3448-3449. 10.1093/bioinformatics/bti551.PubMedView ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.