Enterohaemorrhagic Escherichia coli (EHEC) are responsible for outbreaks of food- and water-borne illness. The bovine gastrointestinal tract (GIT) is thought to be the principle reservoir of EHEC. Knowledge of the nutrients essential for EHEC growth and survival in the bovine intestine may help in developing strategies to limit their shedding in bovine faeces thus reducing the risk of human illnesses. To identify specific metabolic pathways induced in the animal GIT, the transcriptome profiles of EHEC O157:H7 EDL933 during incubation in bovine small intestine contents (BSIC) and minimal medium supplemented with glucose were compared. The transcriptome analysis revealed that genes responsible for the assimilation of ethanolamine, urea, agmatine and amino acids (Asp, Thr, Gly, Ser and Trp) were strongly up-regulated suggesting that these compounds are the main nitrogen sources for EHEC in BSIC. A central role for the gluconeogenesis pathway and assimilation of gluconeogenic substrates was also pinpointed in EHEC incubated in BSIC. Our results suggested that three amino acids (Asp, Ser and Trp), glycerol, glycerol 3-phosphate, L-lactate and C4-dicarboxylates are important carbon sources for EHEC in BSIC. The ability to use gluconeogenic substrates as nitrogen sources (amino acids) and/or carbon sources (amino acids, glycerol and lactate) may provide a growth advantage to the bacteria in intestinal fluids. Accordingly, aspartate (2.4 mM), serine (1.9 mM), glycerol (5.8 mM) and lactate (3.6 mM) were present in BSIC and may represent the main gluconeogenic substrates potentially used by EHEC. A double mutant of E. coli EDL933 defective for phosphoenolpyruvate synthase (PpsA) and phosphoenolpyruvate carboxykinase (PckA), unable to utilize tricarboxylic acid (TCA) intermediates was constructed. Growth competition experiments between EHEC EDL933 and the isogenic mutant strain in BSIC clearly showed a significant competitive growth advantage of the wild-type strain further illustrating the importance of the gluconeogenesis pathway in maintaining EHEC in the bovine GIT.
Citation: Bertin Y, Deval C, de la Foye A, Masson L, Gannon V, Harel J, et al. (2014) The Gluconeogenesis Pathway Is Involved in Maintenance of Enterohaemorrhagic Escherichia coli O157:H7 in Bovine Intestinal Content. PLoS ONE 9(6): e98367. https://doi.org/10.1371/journal.pone.0098367
Editor: Paul Hoskisson, University of Strathclyde, United Kingdom
Received: February 18, 2014; Accepted: May 1, 2014; Published: June 2, 2014
Copyright: © 2014 Bertin et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Funding provided by EU project ProsafeBeef within the 6th Framework Programme (Food-CT-2006-32241). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors confirm that i) Mickaël Desvaux is an editor for the journal PLOS ONE and ii) this does not alter the authors' adherence to PLOS ONE editorial policies and criteria.
Enterohaemorrhagic Escherichia coli (EHEC) are Shiga toxin-producing E. coli (STEC) responsible for human gastrointestinal illnesses, including bloody diarrhea . These disorders may be complicated by renal dysfunction, including the life-threatening haemolytic-uraemic syndrome (HUS), responsible for acute renal failure in children . Most outbreaks and sporadic cases of bloody diarrhea and HUS have been attributed to EHEC serotype O157:H7. STEC strains have been isolated from the intestine of various healthy domestic and wild animals, but ruminants, mainly cattle, are the principal reservoir , . While EHEC strains colonize the bovine gastrointestinal tract, cattle are asymptomatic because they do not express the globotriaosylceramide-3 (Gb3) receptor on their vascular endothelium. Binding to this receptor is thought to be necessary for the pathophysiological effects associated with Shiga toxin in the human host . EHEC strains are transmitted from cattle to humans by means of unpasteurized milk, undercooked meat, fruit, vegetables or water. Hides have been specifically identified as the principal source of EHEC contamination during slaughter .
In-depth knowledge of EHEC physiology and metabolism during residence in the bovine intestinal environment is critical to understand how it is shed in animals and to identify nutritional strategies to limit its shedding. According to Freter's nutrient niche theory, intestinal colonization by a given bacterial species requires the consumption of one or a small number of growth-limiting nutrients more efficiently than all of the other competitors in this ecosystem . In the mammalian intestine, nutrients are released from ingested foods and epithelial or bacterial cell debris. The lumen contains a wide variety of nutrients, at various concentrations and many of these are actively absorbed by the host's intestine. In the rumen, growth of E. coli O157:H7 is limited or may even be suppressed by the presence of the resident microbiota and by strictly anaerobic conditions , , . However, E. coli O157:H7 may survive passage through the acid barrier of the abomasum, and enter the lumen of the bovine small intestine which likely constitutes a more favourable environment for EHEC growth .
To survive and compete, heterotrophic microorganisms have developed the ability to tune their metabolism according to nutrient composition and availability in a given environment , . Although glucose is the preferred carbon and energy source for many bacteria, pathogenic isolates are capable of surviving by using a variety of carbon and nitrogen substrates, including other carbohydrates, lipids, glycolipids and amino acids. The capacity of EHEC to adapt to nutrient availability and environmental conditions is key to their residence time and survival in the intestine. However, our knowledge of the nutrients preferentially used by EHEC in the bovine intestine and the metabolic pathways required for persistence and growth is still limited. For example, Snider et al. demonstrated that fucose is a critical carbon source for the survival of EHEC in the bovine rectum . Also, in vivo colonization experiments showed that the genes agaB and dctA, coding for the specific transport of N-acetyl-glucosamine (GlcNAc) and C4 dicarboxylic acids respectively, influence EHEC colonization of the bovine gut . More recently, we showed that EHEC could catabolize the monosaccharides, released into the bovine small intestine from the mucus layer covering the enterocytes, simultaneously . Furthermore, using a co-culture model we also demonstrated that mucus-derived carbohydrates (GlcNAc, mannose, N-acetyl neuraminic acid and galactose) confer a competitive growth advantage to EHEC in the bovine small intestine . Similarly, ethanolamine released from phospholipids included in animal, plant and microbial cell membranes is an important nitrogen source that favours EHEC persistence in the bovine small intestine .
For their successful survival and multiplication, bacteria must exert specific control over genes required for adaptation to and growth within specific environments. As the transcription of genes encoding catabolic and anabolic reactions can be dynamically altered by bacteria according to nutrient availability, specific metabolic patterns can be deduced from the gene expression profiles , , , . In this study, a whole genome DNA microarray was used for the first time to compare whole transcriptomic profiles of EHEC incubated in bovine small intestine content (BSIC) and minimal medium. The objective of this study was to identify metabolic pathways activated by EHEC in BSIC in order to improve our understanding of the metabolic mechanisms controlling EHEC growth and survival in the bovine gut. Our results highlight the importance of gluconeogenic substrates as carbon and nitrogen sources for EHEC.
Materials and Methods
Animals were slaughtered in accordance with the guidelines of the local Ethics Committee and current INRA ethical guidelines for animal welfare (Permit number: 63345001). BSIC samples were collected after the slaughter of animals required for experiments specifically approved by the “Comité d’éthique en matière d’experimentation animale en Auvergne (Permit number: CE22-08) in the experimental slaughterhouse of the “Unité Mixte de Recherche sur les Herbivores”, INRA, Saint-Genes-Champanelle, France.
Bacterial strains and growth media
EHEC O157:H7 EDL933 was isolated from contaminated hamburger meat . The spontaneous nalidixic acid-resistant mutant EHEC EDL933 NalR previously described  was used in this study. The BSIC samples were collected as previously described , , . Briefly, the jejunum and the ileum were removed as a single piece from three beef cattle and the total luminal contents were collected in O2-free N2-saturated sterile flasks. The BSIC samples were then pooled, rapidly filtered through four layers of cheesecloth and immediately frozen at -80°C until use. Under these conditions of preparation and storage, the microbiota included in BSIC samples remains stable during two months . To obtain sterile BSIC, the intestinal contents were centrifuged twice at 2,000×g for 20 min and the resulting supernatants were filtered through a 0.22-μm nylon filter.
E. coli EDL933 was cultured in sterile BSIC or M9 minimal medium supplemented with 0.4% glucose, MgSO4 (1 mM), CaCl2 (0.1 mM) and trace metals (M9-Glc). Broth cultures were started from a single colony in Luria-Bertani (LB) broth and grown for 8 hours at 37°C with aeration. The bacterial cells were then diluted 50-fold in sterile BSIC or M9-Glc and cultured overnight without aeration (to minimize oxygen availability) at 39°C (bovine temperature). The next day, sterile BSIC or M9-Glc was inoculated by overnight BSIC or M9-Glc broth cultures (1/100 dilution), respectively (15 mL tubes containing 14 mL of BSIC or M9-Glc were used to limit oxygen availability). The cultures were then incubated at 39°C without shaking. The growth rate of E. coli EDL933 was very similar in either BSIC or M9-Glc media (μmax 1.25 h−1), although the lag phase and growth yield were 2 and 1.5 fold higher in M9-Glc medium, respectively.
DNA microarray experiments
1) RNA extraction and cDNA labeling.
Transcriptome analysis was performed from RNA collected when E. coli EDL933 reached the stationary growth phase (4.25 h and 7.5 h in BSIC and M9-Glc respectively). Bacterial suspensions were then centrifuged at 10,000×g for 15 min. The supernatants were stored at -20°C for further investigation and the bacterial pellets were rapidly resuspended in two volumes of RNAprotect Bacteria reagent (Qiagen) to stabilize the RNA. The suspensions were then centrifuged at 10,000×g for 15 min, and total RNA was purified from the bacterial pellet using the NucleoSpin RNA II kit (Macherey-Nagel). Contaminating DNA was removed using an RNAse-free DNAse I column as described by the manufacturer (Macherey-Nagel). RNA was quantified using a NanoDrop ND-1000 spectrophotometer and RNA integrity was electrophoretically verified by ethidium bromide staining. RNA samples were then precipitated by adding 1/10 volume of 3 M sodium acetate (pH 5.2) and 2.5 volumes of cold 100% ethanol before storage at −80°C until further use.
Total RNA samples (8 µg) were reverse transcribed to Cy3- or Cy5-labeled cDNA using the BD-Atlas PowerScript Fluorescent labeling kit (Invitrogen) and Cy-dye Post-Labelling Reactive Dye Pack (Amersham) following the manufacturers' instructions. Unincorporated Cy-dye was removed using the QIAquick PCR purification kit (Qiagen) and dye incorporation was quantified using a NanoDrop spectrophotometer.
2) Hybridization and scanning.
Corning Ultra-Gap II slides (Corning, Acton, MA) were spotted with the MWG E. coli O157:H7 array set (Ocimum) at the National Microbiology Laboratory (Canadian Science Centre for Human and Animal Health, Public Health Agency of Canada, Winnipeg). As previously described, the MWG array consists of 6,167 50-mer oligonucleotides covering the genomes of E. coli K-12 (MG1655) and E. coli O157:H7 strains Sakai (RIMD 0509952) and EDL933 (ATCC 700927) . Description of the array is available at the Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/projects/geo/) (platform accession number GPL6178). Two arrays were spotted per slide.
Prior to hybridization, the slides were incubated for 50 min at 42°C in prehybridization buffer (5×SSC, 0.1% SDS, 0.1% bovine serum albumin), washed twice at room temperature in distilled water for 10 min and dried by centrifugation for 5 min at 800×g. Cy3- and Cy5-labeled cDNA (200 pmol each) were then mixed, dried, resuspended in 50 µL hybridization buffer (5×SSC, 0.1% SDS, 50% deionized formamide, 0.2 g L−1 salmon sperm DNA) and denatured for 5 min at 95°C before cooling for 5 min at room temperature. Labelled cDNA was then applied to the microarray slide, covered with Hybri-Slip cover slips (Molecular Probes), placed in hybridization chambers (Corning) and hybridized overnight at 42°C. Following hybridization, the slides were washed twice in 0.1× SSC, 0.1% SDS (42°C, 10 min) and twice in 0.1× SSC (room temperature, 10 min). After drying by centrifugation (5 min at 800×g), the hybridized slides were scanned with a SYS-SN-ARRAY Agilent Microarray Scanner (Agilent, Santa Clara, CA, USA) at 10 µm resolution.
3) Data analysis.
The signal and background intensity values in both channels for each printed spot were obtained using GenePix Pro 6.0 software (Axon Instruments, USA). Spots with a reference signal lower than background plus two standard deviations were excluded. A total of four arrays were performed using biological replicates (RNA samples were collected from four cultures in BSIC and M9-Glc respectively on separate days). Two microarrays were hybridized with cDNA from BSIC and M9-Glc labelled with cy3 and cy5 respectively and other two hybridizations with cDNA samples conversely labelled. Empty and flagged spots were excluded, and intensities were transformed using a base-2 logarithm. Data normalisation and differential analyses were done using R software. A background correction based on a convolution model which was found to stabilize the variance of the log-ratios (normexp + offset) was processed using limma R package (http://www.bioconductor.org/packages/2.6/bioc/html/limma.html) . The mean spot intensity-dependent dye effect was corrected by performing a global Lowess normalization. The block effect was corrected via subtraction of the median by blocks on each slide following the Anapuce library method . In order to achieve consistency of log-ratio distributions between arrays, the log-ratios of each array were divided by their median-absolute-deviation, a robust measure of dispersion, and multiplied by the mean median-absolute-deviation computed on the four arrays . Technical replicates were then averaged within arrays. Statistical analysis was carried out using the anapuce R package. Differential analysis was done with the Varmixt method which relies on gene expression variances mixture modelling. This method leads to a more powerful test than a t-test assuming a specific variance for each gene, and to a more realistic assumption than assuming a common variance for all genes . The False Discovery Rate was controlled by adjusting p-values using the Benjamini Hochberg method . In this report, only genes showing a >2-fold up-regulation or down-regulation and a Benjamini-Hochberg False Discovery Rate-adjusted P value of <0.05 were considered to be differentially regulated.
In silico analysis
The list of genes was submitted to Database for Annotation, Visualization and Integrated Discovery (DAVID) (resource v6.7) (http://david.abcc.ncifcrf.gov/summary.jsp) to access a relational database of functional annotation , . The genes were categorized in Gene Ontology (GO) categories (“Biological Process” or “Compartment cellular”) using the chart feature offered by DAVID (http://www.geneontology.org/GO.nodes.html). The Gene Functional Classification tool in DAVID builds clusters of genes with significantly similar ontologies as tested against whole E. coli genomes. For each gene list clustered in the different GO terms, DAVID calculated an “Expression Analysis Systematic Explorer” (EASE) score (also called enrichment score) corresponding to a modified Fisher Exact P-value (ranged from 0 to 1). Fisher Exact P-value = 0 represents perfect enrichment. P-value smaller than 0.05 considered as strongly enriched in the annotation categories was used as standard cut-off level. BLAST (http://www.ncbi.nlm.nih.gov), xBASE (http:// xbase.bham.ac.uk), Kyoto Encyclopedia of Genes and Genomes (KEGG) (www.genome.jp/kegg) and EcoCyc (http://ecocyc.org) servers were also used in this study.
Quantitative PCR (qPCR)
Quantitative PCR experiments were performed to validate the microarray results. One microgram of each RNA sample was reverse transcribed using the SuperScript II Reverse Transcriptase kit (Invitrogen) with 3 µg of random primer and 100 units of SuperScript II Rnase H. Quantitative PCR runs were carried out using the Mastercycler ep realplex apparatus (Eppendorf) with 20 ng of cDNA, 0.5 µM of each primer, 3 mM of MgCl2, 10 µL of SYBR Premix Ex Taq mix (Takara Bio Inc.) in a final volume of 20 µL. Amplification conditions were as follows: 95°C for 15 s, 55°C for 15 s, and 72°C for 20 s. The tufA mRNA was used for normalization of mRNA quantification. The relative mRNA quantification was performed using primers designed to specifically amplify fragments of 90 to 200 bp (Table S1). Control samples lacking the reverse transcriptase were included to assess DNA contamination and triplicate samples were amplified in each case. Results were calculated using the comparative cycle threshold method.
The sequential construction of the double mutant EDL933ΔppsAΔpckA was performed by using a one-step PCR-based method . The genes ppsA (Z2731) and pckA (Z4758) were replaced by the genes conferring resistance to chloramphenicol and kanamycin respectively. Primers used to construct the double mutant (ppsA-Cm-F: CGCAGAAATGTGTTTCTCAAACCGTTCATTTATCACAAAAGGATTGTTCGGTGTAGGCTGGAGCTGCTTC, ppsA-Cm-R: TCTTCGGGGATCACATAAACCCGGCGACAAAACGCCGCCGGGGATTTATTCATATGAATATCCTCCTTAGT, pckA-Km-F: CAAAAAGACTTTACTATTCAGGCAATACATATTGGCTAAGGAGCAGTGAAAGCCACGTTGTGTCTCAAAATC and pckA-Km-R: CGTTTTGCTTTCTATAAGATACTGGATAGATATTCTCCAGCTTCAAATCATTAGAAAAACTCATCGAGCA) were designed according to the E. coli O157:H7 EDL933 genome sequence. Gene knockouts were confirmed by PCR analysis and DNA sequencing. The wild-type and mutant strains showed similar growth curves when incubated in M9 minimal medium supplemented with glucose (20 mM) for 24 h at 37°C. In contrast to the wild-type strain, the mutant EDL933ΔppsAΔpckA was unable to grow in M9 medium supplemented with sodium succinate (20 mM) as the sole carbon source.
Bacterial competition experiments
Competition experiments between E. coli EDL933 NalR and the double mutant EDL933ΔppsAΔpckA were performed in BSIC samples containing live endogenous microbiota. Precultures of each E. coli strain, inoculated from a single colony, were incubated in LB broth with the appropriate antibiotic for 8 hours at 37°C with aeration. The precultures were then diluted 50-fold in LB broth and grown overnight at 39°C without shaking. The next day, a BSIC sample was inoculated with approximately 5×103 bacteria mL−1 of each of the two strains tested and then incubated at 39°C without shaking. At each time point, the co-culture was 10-fold serially diluted in phosphate buffer (PBS) at pH 7.2 and plated on Sorbitol MacConkey (SMAC) agar plates containing nalidixic acid, kanamycin or chloramphenicol (50 µg mL−1 each). The plates were then incubated overnight at 37°C and the colony forming units (CFU) were counted. Each experiment was replicated at least three times. The presented values are the log10 mean number of CFU mL−1 ± standard error. Statistical analysis was done using a Student's t test for paired samples (two-tailed).
The concentrations of succinate, fumarate, lactate and malate were quantified by NMR as previously described . Briefly, 50 mM TSP-d4 was added to 0.5 mL of samples and analyzed from one-dimensional 1H NMR spectra. Peak areas were integrated, and the metabolite concentration was calculated relative to TSP-d4. Glycerol concentration was measured using an HPLC apparatus as previously described . Briefly, bacterial supernatants (1 mL) were deproteinized with 125 µL of 0.3 M BaOH and 125 µL of 5% ZnSO4. After centrifugation, the supernatants were filtered through a 0.22 µm nylon filter and injected in an HPLC apparatus (Agilent 1100 series) fitted with two columns (Rezex ROA 300 × 7.8 nm, Phenomenex) mounted in series in an oven (50 °C) with a refractometer as a detector. The mobile phase consisted of sulphuric acid in ultrapure water pumped at 0.7 mL min−1 and 54 bars. Amino acids were quantified by ion-exchange chromatography with post-column ninhydrin detection (Hitachi L8900). The bacterial supernatants were deproteinized as previously described  and Norleucine (2.9 mM) was added as the internal standard. Samples were incubated on ice for 15 min and centrifuged at 10000 × g for 15 min at 4°C.
Results and Discussion
Growth conditions used for performing microarray experiments
Microarray technology was used to identify global gene expression changes in EHEC EDL933 grown in BSIC, relative to the same strain grown in minimal M9 medium supplemented with glucose as the sole carbon source (M9-Glc). As previously reported , , the growth conditions were designed to mimic the physiological conditions of the bovine gut (see the Materials and Methods section). Transcriptome profiling was performed with RNA samples collected when the bacteria reached the stationary growth phase in BSIC and M9-Glc, respectively. This was done to simulate the growth conditions of the bacterium in its natural environment which are very similar to stationary growth phase conditions during laboratory cultivation . The two bacterial cultures were grown under identical conditions of aeration, pH (≈ 7.2–7.3), temperature and growth phase, and differed only in culture media composition. Indeed, the composition of M9-Glc is completely defined whereas the composition of BSIC is unknown (the small intestine probably contained numerous compounds at low concentration). The aim of the study was to identify metabolic pathways preferentially used by EHEC to assimilate compounds present in BISC and absent in M9-Glc.
Transcriptomic analysis of EHEC EDL933 growing in BSIC
Statistical analysis identified 658 genes which were transcribed at significantly different levels in EHEC EDL933 grown in BSIC compared to M9-Glc (≥ 2 fold change; p value<0.05). Of these, the transcription of 364 and 294 genes were up- and down-regulated, respectively, by the bacteria grown in BSIC compared with those grown in M9-Glc. To validate the microarray results independently, the expression of 34 differentially regulated genes (representing approximately 5% of genes with significantly altered expression) was also measured by qPCR (Table S1). Linear regression calculations showed a significant correlation between qPCR and microarray data (r2 = 0.8038) (Fig. S1).
To facilitate subsequent analysis, we first classified the genes into functional groups (Fig. 1). In addition, the differentially expressed genes were classified in different “Gene Ontology” (GO) categories with enrichment scores calculated for each group of genes (see the experimental procedure section and Tables S2 and S3). GO categories and enrichment scores discussed in this study are listed in Table 1. The relative changes in gene expression levels ranged from a 40.6-fold increase for yidY (encoding the MdtL multidrug efflux system) to a 25.9-fold decrease for bioF (encoding 8-amino-7-oxononanoate synthase) during incubation of E. coli EDL933 in BSIC compared with incubation in M9-Glc. Differentially expressed genes were members of nearly all functional categories (Fig. 1). However, a substantial bias in frequency was observed for genes within the “purines, pyrimidines, nucleosides and nucleotide metabolism”, “transcription” and “protein synthesis” functional categories (Fig. 1). Indeed, 77 of the 364 up-regulated genes (21%) in EDL933 incubated in BSIC were found within these three functional groups compared with only 4 of the 294 down-regulated genes (1.4%). Furthermore, genes up-regulated in BSIC were grouped into 11 and 50 catabolic and biosynthetic processes GO categories, respectively (Table S2), whereas genes down-regulated in BSIC were only grouped into 6 and 27 catabolic and biosynthetic processes GO categories, respectively (Table S3). These results are consistent with the notion that E. coli EDL933 may need to express more genes encoding catabolic enzymes in order to metabolize the greater variety of substrates available in BSIC compared with M9-Glc.
Genes up- and down-regulated in EHEC EDL933 incubated in BSIC compared to M9-Glc are shown in grey and white respectively.
Since the aim of this report was to increase our understanding of the nutritional basis of EHEC survival in the bovine intestine, we focused on genes up-regulated in EHEC EDL933 incubated in BSIC encoding the transport and metabolism of carbon and nitrogen sources.
Expression of genes involved in the metabolism of carbon sources
A high enrichment score was obtained for the genes categorized in the tricarboxylic acid (TCA) cycle GO category when E. coli EDL933 was incubated in BSIC in comparison with M9-Glc (Table 1). The genes associated with the TCA cycle are listed in Table 2. The Embden-Meyerof-Parnas (EMP), Entner-Doudoroff (ED) and Pentose-phosphate (PP) pathways are essential for E. coli central metabolism and provide the main routes of carbon flux. However, transcription of genes encoding enzymes required for the different steps of the EMP, ED and PP pathways was not significantly altered in EDL933 incubated in BSIC (GEO database: accession number GSE49468). The TCA cycle that can be activated by both glycolytic and gluconeogenic substrates, is essential for respiration-mediated ATP synthesis and the generation of precursors for many other biosynthetic pathways. In this study, the genes classified in the gluconeogenesis (GNG) GO category were significantly up-regulated by EHEC EDL933 in BISC (Table 1). GNG is essentially a reversal of glycolysis but certain GNG steps are irreversible and thus performed by specific enzymes. During incubation in BSIC, EDL933 up-regulated genes involved in irreversible GNG reactions. As shown in Table 2 and Fig. 2, transcription of the genes encoding the irreversible key GNG enzymes, phosphoenolpyruvate (PEP) synthase (ppsA), fructose-1,6-bisphosphatase (fbp) and PEP carboxykinase (pckA) was induced. In contrast, transcription of the genes coding for the enzymes catalyzing irreversible glycolysis steps (pfkA, pfkB and pykF) was not significantly altered (GEO database: accession number GSE49468).
Genes with transcription up-regulated (red) and not altered (black) in EHEC EDL933 incubated in BSIC compared with M9-Glc. Genes encoding enzymes involved in irreversible GNG reactions are shown in blue. G3P: glycerol-3-phosphate; G3P-ROH: glycerophosphodiester; DHAP: dihydroxyacetone phosphate; PEP: phosphoenopyruvate; Pi: inorganic phosphate; UQ: ubiquinone; UQH2: ubiquinol.
In addition to the observed GNG gene up-regulation, EHEC EDL933 induced pathways involved in assimilation of GNG substrates such as glycerol, glycerol-3P, L-lactate and C4-dicarboxylate.
1) Glycerol and glycerol 3-phosphate.
A significant enrichment score was obtained for the genes classified in the “glycerol metabolic process” GO category (Table 1). The genes glpF and glpT were up-regulated in EHEC EDL933 during incubation in BSIC (Table 2). These genes encode proteins responsible for the specific transport of glycerol and glycerol 3-phosphate (G3P), respectively, across the inner bacterial membrane (Fig. 2). In addition, the transcription of glpK, glpD and glpQ which encode glycerol kinase, G3P dehydrogenase and phosphodiesterase, respectively, was also induced (Table 2). Under aerobic conditions, GlpK converts glycerol to G3P and then GlpD reduces G3P into dihydroxyacetone phosphate. The latter compound is converted into glyceraldehyde-3P entering the GNG pathway (Fig. 2). GlpQ is required for degrading glycerophosphodiesters into G3P and the corresponding alcohol. In fact, the degradation of glycerol and G3P are closely linked (Fig. 2) and G3P dehydrogenase appears to be an essential membrane enzyme, functioning at the central junction of respiration, glycolysis / GNG and phospholipid biosynthesis.
2) C4-dicarboxylates and L lactate.
The gene dctA which encode a transporter for C4-dicarboxylates was up-regulated in EHEC EDL933 during incubation in BSIC (Table 2). C4-dicarboxylates (succinate, fumarate and malate) are gluconeogenic substrates that E. coli can use as carbon and energy sources , , . These compounds can directly enter the TCA cycle to participate in the cyclic flow of carbon. Interestingly, in vivo colonization experiments showed that dctA is required for EHEC colonization of the bovine gut .
The genes lldR and lldD responsible for aerobic L-lactate metabolism were also up-regulated in EHEC EDL933 incubated in BSIC (Table 2). L-lactate dehydrogenase (encoded by lldD) is a peripheral membrane protein that catalyzes the oxidation of L-lactate to pyruvate through the respiratory electron transport chain in vivo (Fig. 2) and allows E. coli to grow in a medium containing L-lactate as the sole carbon source . In addition, the transcription of aldA and fucA encoding aldehyde dehydrogenase and L-fuculose-1-phosphate aldolase, respectively, was induced in EHEC EDL933 incubated in BSIC. These genes are required for the conversion of L-fuculose-phosphate and L-lactaldehyde to L-lactate (Fig. 2; Table 2).
Expression of genes involved in the metabolism of nitrogen sources
In this report, genes in the “amine transport” and “amine catabolic process” GO categories were significantly up-regulated in EHEC EDL933 during incubation in BSIC (Table 1). Induction of these genes is likely indicative of nitrogen mobilization in EHEC to support the biosynthesis of compounds essential for bacterial growth such as peptides, proteins and nucleosides. In-depth microarray results analysis showed that the genes encoding the transport and / or catabolism of ethanolamine, urea and agmatine were up-regulated during incubation of EDL933 in BSIC.
Fourteen genes in the eut operon, which encodes the catabolism of ethanolamine (EA), were up-regulated in EHEC EDL933 incubated in BSIC (Table 3). In particular, the transcription of eutACT encoding ethanolamine ammonia-lyase (enzyme converting ethanolamine to acetaldehyde and free ammonia) was induced (Fig. 3). The increased expression of eut transcripts during incubation of EDL933 in BSIC observed in this report was previously reported based on qPCR results . Indeed, we have shown that free EA: i) constitutes an important nitrogen source for EHEC in BSIC and ii) is responsible for the induction of the eut gene cluster . EA is part of phosphatidylethanolamine, the most abundant phospholipid in bacterial, animal and plant cells membranes and is present in BSIC samples . Consequently, EA constitutes a constantly renewed source of nitrogen in the mammalian intestine and the ability to metabolize this compound likely favours EHEC persistence in this ecological niche .
Genes up- and down-regulated in EHEC EDL933 incubated in BSIC compared to M9-Glc are shown in red and green respectively. The release of ammonia is indicated in blue. THF: tetrahydrofolate.
2) Urea and agmatine.
The genes ureABD were up-regulated in EHEC incubated in BSIC (Table 3). The ure genes encode the enzyme urease that catalyzes the hydrolysis of urea into one carbon dioxide and two ammonium molecules (Fig. 3). Interestingly, the presence of the ure gene cluster appears to confer a competitive colonization advantage to STEC strains in the mouse intestine . The authors suggested that urease provides the bacterial cell with an easily assimilated source of nitrogen to gain a metabolic advantage over other endogenous microbiota . Despite the presence of the ure gene cluster, EHEC EDL933 displays urease-negative phenotype in vitro when cultured on Christensen urea agar plates , . However, ure genes are regulated by Fur (ferric uptake regulator) or an unidentified trans-acting factor and it has been postulate that the pathogen is capable of producing functional urease in the intestine given the correct physiological conditions . The in vivo production of urease by EHEC has not been documented, and the conditions necessary to induce these ure genes in the bovine intestine are unknown.
The gene speB encoding an agmatine ureohydrolase that catalyzes the release of putrescine and urea from agmatine was also up-regulated during incubation in BSIC (Table 3), whereas the transcription of adiC encoding the transport of agmatine out of the cell was repressed (9.26-fold decrease in expression; P: 0.0000104) (Fig. 3). The lumen of the mammalian gut contains a high amount of agmatine which is released from i) the gut microflora, ii) desquamated gastrointestinal epithelial cells and iii) ingested food . This compound could be used by EHEC to generate ammonia in the bacterial cytoplasm and act as a source of nitrogen. Furthermore, in E. coli, arginine decarboxylase catalyzes agmatine formation by the decarboxylation of arginine as the first step of polyamine biosynthesis (that play an important role in the stabilization of DNA and the modulation of genes translation) .
Further studies are needed to investigate urease production by EHEC in the animal digestive tract, as well as agmatine metabolism by this pathogen.
Expression of genes involved in the metabolism of amino acids
A significant enrichment score was calculated for the genes categorized in the “amino acid transport” GO category (Table 1). The genes encoding the transporters for aspartate (dctA), serine (sdaC, tdcC), tryptophan (tnaB), lysine (lysP) and phenylalanine (pheP) into the bacterial cells were up-regulated in EHEC EDL933 incubated in BSIC (Table 4, Fig. 2 and 3).
Gluconeogenic amino acids are both carbon and nitrogen sources for many bacteria: ammonia is first release due to deamination and second, deaminated amino acids yield α-keto acids that can enter either directly or indirectly into the TCA cycle. The transcription of the genes encoding the degradation of gluconeogenic amino acids such as serine, tryptophan and aspartate was induced during incubation of EHEC EDL933 in BSIC. The genes encoding serine deaminase (sdaA) and tryptophanase (tnaA) converting serine and tryptophan to pyruvate, respectively, and aspartate ammonia-lyase (aspA) required to degrade aspartate to fumarate, were also up-regulated (Table 4, Fig. 2 and 3). In addition, the transcription of ansB, which encodes asparaginase II, an enzyme that catalyzes the conversion of asparagine to aspartate, was induced (Table 4, Fig. 2 and 3).
The genes gcvHPRT coding for the glycine cleavage system which catalyzes the release of ammonia from glycine, were up-regulated (Table 4; Fig. 3) while the transcription of glyA, encoding serine hydroxymethyltransferase required to convert 5,10-methylene-THF to serine, was not altered. This suggested that glycine could only be used as nitrogen source by EHEC during incubation in BSIC.
Genes that code for the specific transport of glutamine across the bacterial membrane (glnHPQ) were up-regulated (Table 4; Fig. 3), whereas ybaS, which encodes glutaminase (converting glutamine to glutamate and ammonia), was down-regulated in EHEC EDL933 incubated in BSIC (6.9-fold decrease; P: 0.00003) (Fig. 3). In contrast, the EHEC strain induced the transcription of carA encoding carbamoyl-phosphate synthetase as well as pyrB and pyrI encoding aspartate carbamoyltransferase required to direct the flux of glutamine to pyrimidine synthesis (Fig. 3).
As shown above, genes coding for the transport of lysine and phenylalanine were up-regulated in EHEC EDL933, but the significance of this is unclear as i) phenylalanine cannot be used as a carbon or nitrogen source by E. coli  and ii) lysine is a ketogenic amino acid ultimately degraded to carbon dioxide in the TCA cycle. However, the induction of genes encoding specific transporters suggests that phenylalanine and lysine could be imported into the bacterial cytoplasm to serve as building blocks during protein synthesis.
Role of gluconeogenesis during the growth of EHEC EDL933 in BSIC
From our transcriptomic analyses, a central role for the GNG pathway was pinpointed in E. coli EDL933 incubated in BSIC suggesting that the utilization of gluconeogenic substrates provides a growth advantage for the bacteria. It is well documented that PEP synthase (PpsA) (converting pyruvate to PEP) and PEP carboxykinase (PckA) (converting oxaloacetate to PEP) are two key enzymes in the GNG pathway , , . To perform growth competition experiments, we constructed the double mutant EDL933ΔppsAΔpckA to completely block the conversion of gluconeogenic substrates and TCA cycle intermediates to PEP. A similar double mutant has been previously used to study the role of the GNG pathway in the virulence of Salmonella enterica in mice and the colonization of the mouse intestine by EHEC , . The wild-type and mutant strains used in our study showed similar growth curves when incubated in M9 minimal medium supplemented with glucose (20 mM) as the sole carbon source. However, in contrast to the wild-type strain, EDL933ΔppsAΔpckA was unable to grow in M9 medium supplemented with sodium succinate (20 mM) as the sole carbon source.
Growth competition assays are usually performed to compare the growth pattern of a wild-type strain and its isogenic mutant co-incubated in biological fluids or liquid growth medium , , , . A mutant strain that does not compete efficiently for nutrients fails to reach the same population density as the parent strain, whereas similar growth curves indicate that both strains are able to use limiting nutrients equally well or do not compete for the same limiting nutrient. The E. coli strains EDL933 and EDL933ΔppsAΔpckA were co-incubated in BSIC samples containing live endogenous microbiota under growth conditions that mimic the physiological conditions in BSIC (see the experimental procedure section). As shown in Fig. 4, similar growth curves were observed during the first 4 h of co-incubation, suggesting that nutrients present in BSIC were used by the two strains with equal efficiency. However, after 5 h of co-incubation, a significant growth defect was observed for EDL933ΔppsAΔpckA compared with the wild type strain. These results demonstrated that activation of the GNG pathway confers a competitive growth advantage to EHEC EDL933 in BSIC and suggested that assimilation of gluconeogenic substrates is required for maximal growth of EHEC in the bovine small intestine.
The BSIC samples were inoculated with a 1:1 mixture of the two strains. Bars represent the SEM of three independent experiments. ***, denotes statistical significance, P <0.01 as determined by the Student t test for paired samples.
To analyze potential assimilation of gluconeogenic substrates by EHEC EDL933, BSIC was then analyzed by NMR to measure succinate, fumarate, and lactate levels and by HPLC for glycerol and amino acid quantification. A total of 45 mM of free amino acids was detected, including aspartate (2.4 mM) and serine (1.9 mM). Although the transcription of the genes encoding proteins responsible for the transport and the catabolism of tryptophan was induced in EDL933 cultured in BSIC, tryptophan was below the limit of detection (15 µM). Glycerol (5.8 mM) and lactate (3.6 mM) were also detected in BSIC while succinate and fumarate were at levels under the limit of detection. Taken together, the results showed that aspartate, serine, glycerol and lactate are released into the bovine intestine and may constitute important gluconeogenic substrates that can be used by EHEC.
Since environmental conditions direct the expression of suites of genes necessary for optimal bacterial growth, our data indicate that the utilization of gluconeogenic substrates may play an important role in the colonization of the bovine intestine by EHEC. Gluconeogenic substrates appear to be nitrogen sources (amino acids) and carbon sources (amino acids, glycerol, L-lactate). Accordingly, it is well documented that the GNG pathway contributes to bacterial fitness in vivo and is required by pathogenic Enterobacteriaceae during host infection , , , . In particular, utilization of gluconeogenic substrates is required for a maximal colonization of the mouse intestine by EHEC .
Significant levels of gluconeogenic substrates are likely present in the bovine intestinal environment. Lactate-producing bacteria such as Lactobacillus, Streptococcus or Succinivibrio sp. are part of the bovine intestinal microbiota and produce lactate (from carbohydrate fermentation) that can accumulate in the bovine intestine . Aspartate and serine are abundant amino acids present in the mucus of the bovine small intestine  and may constitute direct sources of nutrients from the intestinal epithelium. Serine is also part of phosphatidylserine (a phospholipid present in the leaflet of prokaryote and eukaryote cell membranes) and may be released into BSIC from intestinal epithelium and cell debris from the endogenous microbiota. Glycerol is part of phospholipids (phosphatidylethanolamine, phosphatidylserine and phosphatidylcholine) present in cell membranes and, like serine, is released into the BSIC during cell renewal. It is interesting to note that glycerol is used as a dietary supplement for cattle as i) a substitute for starch-based ingredients (such as corn), ii) energy supplementation, and iii) a preventive aid for metabolic problems (such as ketosis) , . Glycerol is inexpensive and is currently abundant because it is a by-product of biodiesel fuel production. Consequently there is a renewed interest in incorporating this feedstock into ruminant diets. Interestingly, only 80% of the glycerol added to cattle diets is metabolized anaerobically by the ruminal microbiota , suggesting that part of the dietary glycerol can reach the small intestine.
The genome-wide expression profiling described in this report highlights the capacity of EHEC to adapt to the bovine digestive environment and opens numerous new avenues for future investigations. In particular, defining the precise role of different gluconeogenic substrates during the growth of EHEC in the bovine intestine should enhance our understanding of their role in the physiology of these pathogens, and may assist us in discovering new nutritional or ecological strategies (for example the use of probiotics) to limit EHEC carriage by ruminants.
Comparison of fold-changes of genes expression obtained by microarray and q-PCR. Thirty four genes with significantly altered expression in EHEC EDL933 cultured in BSIC compared to M9-Glc (both up and down-regulated) were selected from microarray data. The gene list is shown in Table S1. The expression ratios obtained by q-PCR and log2-transformed fold changes were plotted against one another. Linear regression calculations showed a significant correlation between qPCR and microarray data (r2 = 0.8038).
Sequence of primers used in relative mRNA quantification.
Classification of genes up-regulated in EHEC EDL933 incubated in BSIC compared with cells incubated in M9-Glc. The genes were classified in different “Gene Ontology” (GO) categories with enrichment scores calculated for each genes group.
Classification of genes down-regulated in EHEC EDL933 incubated in BSIC compared with cells incubated in M9-Glc. The genes were classified in different “Gene Ontology” (GO) categories with enrichment scores calculated for each genes group.
We thank Alexandra Durand for excellent technical assistance and David Duchez for HPLC analysis.
Conceived and designed the experiments: YB CD ADLF JH CM EF. Performed the experiments: YB CD ADLF. Analyzed the data: YB CD ADLF LM VG JH CM MD EF. Contributed reagents/materials/analysis tools: LM VG JH ADLF. Wrote the paper: YB CD ADLF LM VG JH CM MD EF.
- 1. Kaper JB, Nataro JP, Mobley HL (2004) Pathogenic Escherichia coli. Nat Rev Microbiol 2: 123–140.
- 2. Tarr PI, Gordon CA, Chandler WL (2005) Shiga-toxin-producing Escherichia coli and haemolytic uraemic syndrome. Lancet 365: 1073–1086.
- 3. Caprioli A, Morabito S, Brugere H, Oswald E (2005) Enterohaemorrhagic Escherichia coli: emerging issues on virulence and modes of transmission. Vet Res 36: 289–311.
- 4. Karmali MA, Gannon V, Sargeant JM (2010) Verocytotoxin-producing Escherichia coli (VTEC). Vet Microbiol 140: 360–370.
- 5. Pruimboom-Brees IM, Morgan TW, Ackermann MR, Nystrom ED, Samuel JE, et al. (2000) Cattle lack vascular receptors for Escherichia coli O157:H7 Shiga toxins. Proc Natl Acad Sci U S A 97: 10325–10329.
- 6. McEvoy JM, Doherty AM, Sheridan JJ, Thomson-Carter FM, Garvey P, et al. (2003) The prevalence and spread of Escherichia coli O157:H7 at a commercial beef abattoir. J Appl Microbiol 95: 256–266.
- 7. Freter R, Brickner H, Botney M, Cleven D, Aranki A (1983) Mechanisms that control bacterial populations in continuous-flow culture models of mouse large intestinal flora. Infect Immun 39: 676–685.
- 8. Chaucheyras-Durand F, Faqir F, Ameilbonne A, Rozand C, Martin C (2010) Fates of acid-resistant and non-acid-resistant Shiga toxin-producing Escherichia coli strains in ruminant digestive contents in the absence and presence of probiotics. Appl Environ Microbiol 76: 640–647.
- 9. Chaucheyras-Durand F, Madic J, Doudin F, Martin C (2006) Biotic and abiotic factors influencing in vitro growth of Escherichia coli O157:H7 in ruminant digestive contents. Appl Environ Microbiol 72: 4136–4142.
- 10. de Vaux A, Morrison M, Hutkins RW (2002) Displacement of Escherichia coli O157:H7 from rumen medium containing prebiotic sugars. Appl Environ Microbiol 68: 519–524.
- 11. Bertin Y, Girardeau JP, Chaucheyras-Durand F, Lyan B, Pujos-Guillot E, et al. (2011) Enterohaemorrhagic Escherichia coli gains a competitive advantage by using ethanolamine as a nitrogen source in the bovine intestinal content. Environ Microbiol 13: 365–377.
- 12. Baev MV, Baev D, Radek AJ, Campbell JW (2006) Growth of Escherichia coli MG1655 on LB medium: monitoring utilization of amino acids, peptides, and nucleotides with transcriptional microarrays. Appl Microbiol Biotechnol 71: 317–322.
- 13. Ferenci T (2001) Hungry bacteria—definition and properties of a nutritional state. Environ Microbiol 3: 605–611.
- 14. Snider TA, Fabich AJ, Conway T, Clinkenbeard KD (2009) E. coli O157:H7 catabolism of intestinal mucin-derived carbohydrates and colonization. Vet Microbiol 136: 150–154.
- 15. Dziva F, van Diemen PM, Stevens MP, Smith AJ, Wallis TS (2004) Identification of Escherichia coli O157: H7 genes influencing colonization of the bovine gastrointestinal tract using signature-tagged mutagenesis. Microbiology-Sgm 150: 3631–3645.
- 16. Bertin Y, Chaucheyras-Durand F, Robbe-Masselot C, Durand A, de la Foye A, et al. (2013) Carbohydrate utilization by enterohaemorrhagic Escherichia coli O157:H7 in bovine intestinal content. Environ Microbiol 15: 610–622.
- 17. Chang DE, Smalley DJ, Tucker DL, Leatham MP, Norris WE, et al. (2004) Carbon nutrition of Escherichia coli in the mouse intestine. Proc Natl Acad Sci U S A 101: 7427–7432.
- 18. Jandu N, Ho NK, Donato KA, Karmali MA, Mascarenhas M, et al. (2009) Enterohemorrhagic Escherichia coli O157:H7 gene expression profiling in response to growth in the presence of host epithelia. PLoS ONE 4: e4889.
- 19. Bergholz TM, Vanaja SK, Whittam TS (2009) Gene expression induced in Escherichia coli O157:H7 upon exposure to model apple juice. Appl Environ Microbiol 75: 3542–3553.
- 20. Riley LW, Remis RS, Helgerson SD, Mcgee HB, Wells JG, et al. (1983) Hemorrhagic Colitis Associated with a Rare Escherichia-Coli Serotype. N. Engl. J. Med. 308: 681–685.
- 21. Chagnot C, Zorgani MA, Astruc T, Desvaux M (2013) Proteinaceous determinants of surface colonization in bacteria: bacterial adhesion and biofilm formation from a protein secretion perspective. Front Microbiol 4: 303.
- 22. Chagnot C, Agus A, Renier S, Peyrin F, Talon R, et al. (2013) In vitro colonization of the muscle extracellular matrix components by Escherichia coli O157:H7: the influence of growth medium, temperature and pH on initial adhesion and induction of biofilm formation by collagens I and III. PLoS ONE 8: e59386.
- 23. Poirier K, Faucher SP, Beland M, Brousseau R, Gannon V, et al. (2008) Escherichia coli O157:H7 survives within human macrophages: global gene expression profile and involvement of the Shiga toxins. Infect Immun 76: 4814–4822.
- 24. Ritchie ME, Silver J, Oshlack A, Holmes M, Diyagama D, et al. (2007) A comparison of background correction methods for two-colour microarrays. Bioinformatics 23: 2700–2707.
- 25. Yang YH, Dudoit S, Luu P, Lin DM, Peng V, et al. (2002) Normalization for cDNA microarray data: a robust composite method addressing single and multiple slide systematic variation. Nucleic Acids Res 30: e15.
- 26. Smyth GK, Speed T (2003) Normalization of cDNA microarray data. Methods 31: 265–273.
- 27. Delmar P, Robin S, Daudin JJ (2005) VarMixt: efficient variance modelling for the differential analysis of replicated gene expression data. Bioinformatics 21: 502–508.
- 28. Benjamini Y, Hochberg Y (1995) Controlling the False Discovery Rate - a Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society Series B-Methodological 57: 289–300.
- 29. Huang da W, Sherman BT, Lempicki RA (2009) Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc 4: 44–57.
- 30. Huang da W, Sherman BT, Lempicki RA (2009) Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res 37: 1–13.
- 31. Datsenko KA, Wanner BL (2000) One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci U S A 97: 6640–6645.
- 32. Matulova M, Nouaille R, Capek P, Pean M, Delort AM, et al. (2008) NMR study of cellulose and wheat straw degradation by Ruminococcus albus 20. FEBS J 275: 3503–3511.
- 33. Christophe G, Guiavarch E, Creuly C, Dussap CG (2009) Growth monitoring of Fibrobacter succinogenes by pressure measurement. Bioprocess Biosyst Eng 32: 123–128.
- 34. Mondino A, Bongiovanni G, Fumero S, Rossi L (1972) An improved method of plasma deproteination with sulphosalicylic acid for determining amino acids and related compounds. J Chromatogr 74: 255–263.
- 35. Bacun-Druzina V, Butorac A, Mrvcic J, Landeka T, Dragicevic TL, et al. (2011) Bacterial Stationary-Phase Evolution. Food Technol. Biotechnol 49: 13–23.
- 36. Engel P, Kramer R, Unden G (1992) Anaerobic fumarate transport in Escherichia coli by an fnr-dependent dicarboxylate uptake system which is different from the aerobic dicarboxylate uptake system. J Bacteriol 174: 5533–5539.
- 37. Abo-Amer AE, Munn J, Jackson K, Aktas M, Golby P, et al. (2004) DNA interaction and phosphotransfer of the C4-dicarboxylate-responsive DcuS-DcuR two-component regulatory system from Escherichia coli. J Bacteriol 186: 1879–1889.
- 38. Janausch IG, Zientz E, Tran QH, Kroger A, Unden G (2002) C4-dicarboxylate carriers and sensors in bacteria. Biochim Biophys Acta 1553: 39–56.
- 39. Dong JM, Taylor JS, Latour DJ, Iuchi S, Lin EC (1993) Three overlapping lct genes involved in L-lactate utilization by Escherichia coli. J Bacteriol 175: 6671–6678.
- 40. Steyert SR, Kaper JB (2012) Contribution of urease to colonization by Shiga toxin-producing Escherichia coli. Infect Immun 80: 2589–2600.
- 41. Heimer SR, Welch RA, Perna NT, Posfai G, Evans PS, et al. (2002) Urease of enterohemorrhagic Escherichia coli: evidence for regulation by fur and a trans-acting factor. Infect Immun 70: 1027–1031.
- 42. Friedrich AW, Kock R, Bielaszewska M, Zhang W, Karch H, et al. (2005) Distribution of the urease gene cluster among and urease activities of enterohemorrhagic Escherichia coli O157 isolates from humans. J Clin Microbiol 43: 546–550.
- 43. Molderings GJ, Heinen A, Menzel S, Lubbecke F, Homann J, et al. (2003) Gastrointestinal uptake of agmatine: distribution in tissues and organs and pathophysiologic relevance. Ann N Y Acad Sci 1009: 44–51.
- 44. Panagiotidis CA, Blackburn S, Low KB, Canellakis ES (1987) Biosynthesis of polyamines in ornithine decarboxylase, arginine decarboxylase, and agmatine ureohydrolase deletion mutants of Escherichia coli strain K-12. Proc Natl Acad Sci U S A 84: 4423–4427.
- 45. Olukoya DK (1986) Nutritional variation in Escherichia coli. J Gen Microbiol 132: 3231–3234.
- 46. Tchawa Yimga M, Leatham MP, Allen JH, Laux DC, Conway T, et al. (2006) Role of gluconeogenesis and the tricarboxylic acid cycle in the virulence of Salmonella enterica serovar Typhimurium in BALB/c mice. Infect Immun 74: 1130–1140.
- 47. Miranda RL, Conway T, Leatham MP, Chang DE, Norris WE, et al. (2004) Glycolytic and gluconeogenic growth of Escherichia coli O157:H7 (EDL933) and E. coli K-12 (MG1655) in the mouse intestine. Infect Immun 72: 1666–1676.
- 48. Chao YP, Patnaik R, Roof WD, Young RF, Liao JC (1993) Control of gluconeogenic growth by pps and pck in Escherichia coli. J Bacteriol 175: 6939–6944.
- 49. Farrell MJ, Finkel SE (2003) The growth advantage in stationary-phase phenotype conferred by rpoS mutations is dependent on the pH and nutrient environment. J Bacteriol 185: 7044–7052.
- 50. Palchevskiy V, Finkel SE (2006) Escherichia coli competence gene homologs are essential for competitive fitness and the use of DNA as a nutrient. J Bacteriol 188: 3902–3910.
- 51. Pradhan S, Baidya AK, Ghosh A, Paul K, Chowdhury R (2010) The El Tor biotype of Vibrio cholerae exhibits a growth advantage in the stationary phase in mixed cultures with the classical biotype. J Bacteriol 192: 955–963.
- 52. Alteri CJ, Smith SN, Mobley HL (2009) Fitness of Escherichia coli during urinary tract infection requires gluconeogenesis and the TCA cycle. PLoS Pathog 5: e1000448.
- 53. Dougherty MJ, Boyd JM, Downs DM (2006) Inhibition of fructose-1,6-bisphosphatase by aminoimidazole carboxamide ribotide prevents growth of Salmonella enterica purH mutants on glycerol. J Biol Chem 281: 33892–33899.
- 54. Ljungh A, Wadstrom T (2006) Lactic acid bacteria as probiotics. Curr Issues Intest Microbiol 7: 73–89.
- 55. Montagne L, Toullec R, Lalles JP (2000) Calf intestinal mucin: Isolation, partial characterization, and measurement in ileal digesta with an enzyme-linked immunosorbent assay. J Dairy Sci 83: 507–517.
- 56. AbuGhazaleh AA, Abo El-Nor S, Ibrahim SA (2011) The effect of replacing corn with glycerol on ruminal bacteria in continuous culture fermenters. J Anim Physiol Anim Nutr (Berl) 95: 313–319.
- 57. DeFrain JM, Hippen AR, Kalscheur KF, Jardon PW (2004) Feeding glycerol to transition dairy cows: effects on blood metabolites and lactation performance. J Dairy Sci 87: 4195–4206.
- 58. Trabue S, Scoggin K, Tjandrakusuma S, Rasmussen MA, Reilly PJ (2007) Ruminal fermentation of propylene glycol and glycerol. J Agric Food Chem 55: 7043–7051.