Bacteria can elaborate complex patterns of development that are dictated by temporally ordered patterns of gene expression, typically under the control of a master regulatory pathway. For some processes, such as biofilm development, regulators that initiate the process have been identified but subsequent phenotypic changes such as stress tolerance do not seem to be under the control of these same regulators. A hallmark feature of biofilms is growth within a self-produced extracellular matrix. In this study we used metabolomics to compare Salmonella cells in rdar colony biofilms to isogenic csgD deletion mutants that do not produce an extracellular matrix. The two populations show distinct metabolite profiles. Even though CsgD controls only extracellular matrix production, metabolite signatures associated with cellular adaptations associated with stress tolerances were present in the wild type but not the mutant cells. To further explore these differences we examine the temporal gene expression of genes implicated in biofilm development and stress adaptations. In wild type cells, genes involved in a metabolic shift to gluconeogenesis and various stress-resistance pathways exhibited an ordered expression profile timed with multicellular development even though they are not CsgD regulated. In csgD mutant cells, the ordered expression was lost. We conclude that the induction of these pathways results from production of, and growth within, a self produced matrix rather than elaboration of a defined genetic program. These results predict that common physiological properties of biofilms are induced independently of regulatory pathways that initiate biofilm formation.
Citation: White AP, Weljie AM, Apel D, Zhang P, Shaykhutdinov R, Vogel HJ, et al. (2010) A Global Metabolic Shift Is Linked to Salmonella Multicellular Development. PLoS ONE 5(7): e11814. https://doi.org/10.1371/journal.pone.0011814
Editor: Alfredo Herrera-Estrella, Cinvestav, Mexico
Received: March 2, 2010; Accepted: June 22, 2010; Published: July 27, 2010
Copyright: © 2010 White 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: This work was supported by grants from the Canadian Institutes of Health Research (http://www.cihr-irsc.gc.ca) to MGS and HJV. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Bacteria frequently grow in multicellular communities that can exhibit complex phenotypes. How the cells organize and how these phenotypes are regulated is of fundamental importance in many areas of microbiology. Biofilms are examples of bacterial multicellular behavior. Biofilm is an umbrella term describing the growth of bacterial cells encased within an extracellular matrix usually in association with surfaces. Biofilms are thought to be the most prevalent form of bacterial life in nature and represent an evolutionarily conserved strategy for survival and persistence . In addition, they are implicated in >60% of human infectious diseases with tremendous health and economic impacts . Characteristics of biofilms include high cell densities, nutrient limitation, and matrix components that serve to link individual cells together . Cells within a biofilm also display remarkable stress tolerance including reduced susceptibility to antibiotics. Many factors contribute to this elevated resistance including regulatory mechanisms ,  and physical and chemical protection by protein, polysaccharide or nucleic acid polymers in the extracellular matrix . These polymers can also aid survival by nutrient trapping, buffering and water retention . The stress tolerance of biofilms is a common feature independent of how they are formed. However, it remains to be answered how biofilms grown under different conditions show common phenotypes without a coordinating regulatory pathway.
For Salmonella, the best characterized biofilm state is a coordinated multicellular behaviour termed the rdar morphotype , . The formation of rdar (red, dry and rough) colonies is marked by a shift from smooth to an aggregative morphology which results from the ordered production of extracellular matrix components . The matrix in rdar morphotype colonies is primarily comprised of curli fimbriae (also called thin aggregative fimbriae or Tafi) and several exopolysaccharides (EPS), including cellulose and an O-antigen capsule , , . These polymers are produced in response to starvation , triggered through activation of σS (RpoS), the sigma factor that regulates the general stress response , , and ultimately CsgD, the main transcriptional regulator of the rdar morphotype , . CsgD activation is controlled by a complex regulatory cascade leading to increased intracellular levels of bis (3′-5′) cyclic dimeric guanosine monophosphate (c-di-GMP) , , . CsgD controls aggregation by directly stimulating transcription of csgBAC, encoding the curli structural proteins, and adrA, encoding a diguanylate cyclase that activates cellulose production , . The rdar morphotype polymers provide a survival advantage through enhanced resistance to desiccation and disinfection , , , allowing cells to survive for up to several years . The rdar morphotype is hypothesized to represent a critical state in the transmission of Salmonella between hosts , .
Microarrays , , , mutagenesis ,  and proteomics , , along with many other approaches, have been successfully used to identify differences between planktonic and biofilm cells. While each approach has its own merit, there is still much to learn about biofilm-specific regulatory networks  and stress resistance mechanisms, which are predicted to be related to heterogeneity ,  and/or slow growth of cells . Here, we used a combined approach of metabolomics and transcriptional analysis to compare extracellular matrix-embedded, wild-type S. enterica serovar Typhimurium (hereafter referred to as S. Typhimurium) to a matrix-deficient csgD deletion mutant. We wanted to determine if cells that lack the machinery for polymer production (due to altered regulation) would accumulate precursors and share similar metabolism as wild-type cells or whether there was a specific metabolic adaptation linked to the aggregation process. The use of luciferase reporters in transcriptional analysis allowed for temporal resolution during this early event in biofilm formation. Metabolic differentiation and stress-resistance pathways were activated in wild-type cells as part of a global transcriptional upshift coinciding with the time of aggregation. The dynamic temporal program and lack of expression in csgD mutant cells suggests that many of the adaptations in wild-type cells occurred in response to the microenvironment generated by aggregation. We hypothesize that growth within the self-produced matrix regulates a core set of “biofilm” traits independent of the macro environment. This could be an important step in understanding the regulation and physiology of cells in bacterial biofilms.
Characterization of the small molecule metabolites produced by bacteria represents a non-biased approach to investigate cellular activity. For our experiments, metabolites were extracted from S. Typhimurium wild-type and csgD mutant colonies grown for two or five days on 1% tryptone medium (T agar). Under these conditions, wild-type cells form aggregative, rdar morphotype colonies, whereas csgD deletion mutant cells form smooth colonies that lack EPS production (Figure 1; , ). Despite the differences in colony diameter (Fig. 1), the starting CFU numbers were similar: Day 2 - 2.80±0.48×109 for wild-type and 1.77±0.27×109 for the csgD mutant (n = 10, P = 2.8×10−6, two-tailed Student's paired t-Test); and Day 5 - 3.23±0.44×109 for wild-type and 3.12±0.39×109 for the csgD mutant (n = 8, P = 0.61, two-tailed Student's paired t-Test).
1 µL of cells from overnight cultures of each strain (A) were grown at 28°C for 50 h on T agar (B), 100 h on T agar supplemented with 100 µg/mL Congo red (C), or 75 h on T agar supplemented with 200 µg/mL calcofluor (D). Colonies in (B) were stained for glycogen production (see Materials and Methods); a dark brown color is indicative of the presence of glycogen . Colonies in (D) were visualized under UV light; the white color is indicative of calcofluor binding . ΔcsgD and ΔpckA/ΔppsA strains are deficient for rdar morphotype formation (A, C), glycogen (B) and cellulose (D) production.
GC-MS and 1H NMR metabolite profiles were initially compared by unsupervised principal component analysis , which confirmed that there were significant differences between strains, and the time of growth, with no sample outliers (data not shown). A final supervised model of the spectra was generated using orthogonal partial least square discriminate analysis (Figure 2). The explained variance in metabolite data (R2) and predictive ability (Q2) were high for both GC-MS (R2 = 0.976, Q2 = 0.865) and 1H NMR (R2 = 0.912, Q2 = 0.741) models. The GC-MS spectra were clearly divided into four groups corresponding to each strain after 2 or 5 days of growth, whereas the 1H NMR profiles displayed more batch variation and the groupings were not as distinct (Figure 2).
GC-MS and 1H NMR profiles of metabolites extracted from wild-type or csgD mutant colonies grown on T agar for 2 days (open symbols) or 5 days (closed symbols) at 28°C were compared by orthogonal partial least square discriminate analysis (OPLS-DA). Score plots derived from the OPLS-DA models are shown. Each symbol represents one sample; for GC-MS, n = 9 for all groups except for wt at day 2 (n = 8), for 1H NMR, n = 8 for all four groups. The x-axis is the primary component, which represents all variance related to the csgD mutation. The y-axis is the first orthogonal component, which represents the variance related to the day of growth and is unrelated to mutation status.
Summary of metabolomic analysis
In total, 25 metabolites were detected at statistically different concentrations (Table 1). Many compounds detected at higher levels in wild-type colonies were the end products of gluconeogenesis, including glucose and its polymer, glycogen, as well as galactose, mannose, and glycerol-3-phosphate, an important gluconeogenesis intermediate . Trehalose, glycine-betaine (betaine) and glutamate, three of the major osmoprotectants used by S. Typhimurium , were found at higher concentrations in wild-type colonies. Additional osmoprotectants detected were carnitine and galactinol (Text S1). Other compounds more abundant in wild-type colonies were glutathione, nicotinamide adenine dinucleotide (NAD+), octanoic acid and pyroglutamate (Text S1). The major compounds detected at higher levels in csgD mutant colonies were the upper TCA cycle intermediates succinate, fumarate and malate, the polyamine compounds putrescine and cadaverine, and adenosine monophosphate (AMP), adenine and proline.
Plotting these metabolites onto a simplified S. Typhimurium metabolic map (Figure 3) indicated that gluconeogenesis was specifically activated in wild-type cells and/or repressed in csgD mutant cells. We hypothesized that a block in gluconeogenesis in csgD mutant cells was responsible for the accumulation of TCA cycle intermediates and polyamine compounds. The detection of higher levels of AMP in csgD mutant colonies was consistent with this hypothesis, since AMP is a potent inhibitor of the gluconeogenesis-specific enzyme fructose bisphosphatase (Fbp) . The presence of numerous osmoprotectants in the wild-type cells was unexpected because T agar is a low osmolarity medium; osmoprotectants normally only accumulate during growth under high osmolarity conditions .
Compounds shown were identified at statistically higher concentrations in wild-type colonies (red) or csgD mutant colonies (blue). The schematics for gluconeogenesis, the TCA cycle, and related pathways were adapted from the EcoCYC™ database (www.ecocyc.org). Genes encoding important enzymes are listed in italics; their expression was monitored using promoter luciferase fusions. Genes encoding enzymes that catalyze key reactions in gluconeogenesis are underlined.
Reporters for transcriptional analysis
Based on our working model of cellular states, promoter-luciferase transcriptional fusions were generated for genes at regulatory checkpoints in several key metabolic pathways (Figure 3). In addition, reporters were generated for genes a) identified as important in related expression studies (J.S. Happe, R.J. Martinuzzi, V. Kostenko, M.G. Surette, unpublished) or b) whose protein products were identified by proteomic analysis of wild-type rdar morphotype colonies (A.P. White, W. Kim, M.G. Surette, unpublished). Control reporters that contain synthetic promoters designed to measure σ70 and σS activity - sig70_7  and sig38H4 , respectively - were also included. In total, reporters were generated for 59 single gene or multiple gene operons (Table S1).
Luciferase expression was initially monitored in wild-type and csgD mutant strains grown on T agar as individual or mixed-strain colonies (data not shown). However, the narrow linear range of detection by the camera system made it difficult to resolve differences in expression. Furthermore, while development of the rdar colony is an ordered process, it does not occur uniformly throughout the colony ,  making temporal expression profiling in the colony complicated. To overcome these problems, we analyzed gene expression during growth of the strains in 1% tryptone liquid media. Wild-type cells grown under these conditions have a clear aggregation phenotype, coupled with increased gene expression , and the multicellular aggregates formed share many of the characteristics of cells in rdar morphotype colonies , . In contrast, csgD deletion mutant cells do not aggregate under these same growth conditions , .
Transcriptional profiling reveals a global metabolic shift coinciding with aggregation
Most of the reporters in S. Typhimurium wild-type cells displayed a distinct temporal pattern of activation with peak expression occurring at the time of aggregation (Figure 4A). Aggregation was predicted to begin at 25 h, based on an increase in σS activity and activation of essential rdar morphotype genes, including csgDEFG, csgBAC and adrA (Figure 5A). The coordinated activation of genes from many different functional categories (Table S1) is indicative of a global metabolic shift in wild-type cells. In contrast, the majority of operons analyzed, including the rdar morphotype genes (Figure 5A), had low expression in the csgD deletion mutant cultures (Figure 4B) and no correlation to the temporal pattern observed in wild-type cultures.
Each wild-type (A) or csgD mutant (B) reporter strain contains a plasmid-based promoter-luciferase (luxCDABE) fusion designed to measure gene expression by light production. For each reporter, shown is the ratio of lux activity at each time point divided by the maximum luminescence in the wt reporter strain. Blue and red indicate low and high expression, respectively. Gene (or operon) names are listed on the left of each panel; sig38H4 and sig70_7 are synthetic reporters designed to measure σS and σ70 activity, respectively. Genes that are essential for rdar morphotype formation are shown in red; mlrA encodes a transcriptional regulator required for csgDEFG expression . Arrows in (A) signify the beginning of the aggregation process at 25 h.
Each red (wild-type) or blue (csgD mutant) curve represents the raw, non-normalized gene expression value (light counts per second; CPS) in each strain as a function of time. Expression profiles are shown for operons encoding proteins essential for rdar morphotype aggregation (A), gluconeogenesis-specific enzymes (B), glycolytic or gluconeogenic bi-directional enzymes (C), upper tricarboxylic acid cycle enzymes (D), or proteins involved in osmoprotection (E). Gene (operon) names are listed on the upper left in each panel. σS activity is represented by expression measured from the sig38H4 reporter. Vertical lines in each graph in (A) represent the beginning of the aggregation process at 25 h; these lines are shown for all other reporters to highlight the coordinated timing of gene expression. For the majority of genes, promoter activity peaks at or near the time of aggregation in wild-type cells.
Global transcription rates were elevated at the time of aggregation. Wild-type cells had a two-fold increase in σ70 activity and nearly three-fold increase in σS activity relative to csgD mutant cells (Figure 4, Table S1). Since σ70 and σS compete for binding to the RNA polymerase holoenzyme and drive expression of genes required for vegetative growth and stress responses, respectively , , these results were indicative of physiological differences between wild-type and csgD mutant cultures.
Carbon flux into gluconeogenesis is increased in S. Typhimurium rdar morphotype cells
To monitor carbon flux, we analyzed the expression of genes encoding key enzymes in gluconeogenesis, glycolysis and TCA cycles (Figure 3). Four key gluconeogenesis-specific enzymes, malic enzyme (maeB), PEP synthase (ppsA), PEP carboxylase (pckA) and fructose bisphosphatase (fbp) , were all significantly up-regulated in wild-type cultures relative to csgD mutant cultures (Figure 4, Figure 5B). The largest change in gene expression was measured for pckA, which was elevated 45-fold (Table S1). GpmA, pgmI, gapA, fbaA, and fbaB genes, encoding enzymes that catalyze reversible steps in gluconeogenesis and glycolysis, were also induced in the wild-type strain at the time of aggregation (Figure 4A, Figure 5C). We hypothesized that elevated expression of these enzymes was necessary for increased carbon flux between PEP and fructose-1,6,-bisphosphate (Figure 3). Since upper TCA cycle intermediates are essential starting points for gluconeogenesis , we monitored the expression of succinate dehydrogenase (sdhCD), fumarate reductase (fumAC), and malate dehydrogenase (mdh); each of these enzymes were up-regulated in wild-type cultures coinciding with the time of aggregation (Figure 4A, Figure 5D). Lower expression of these genes in the csgD mutant cultures indicated that the elevated concentrations of TCA intermediates measured by metabolomics were not due to increased amounts of enzyme, but were more likely caused by inhibition of gluconeogenesis. Expression of aceBA, coding for enzymes in the glyoxylate shunt, and sucAB, coding for enzymes catalyzing conversion of 2-ketoglutarate to succinate for complete TCA cycling, were not different between wild-type and csgD mutant strains (Figure 4, Figure 5D, Table S1). For the majority of metabolic reporters analyzed, expression profiles were similar in both strains until the estimated time of aggregation at which point expression was induced in wild-type cells (Figure 5B, C, D). This confirmed that there was a metabolic shift linked to the aggregation process.
The production of sugars from gluconeogenesis should be an essential pathway for aggregation in S. Typhimurium, particularly when strains are grown on amino-acid based media, such as tryptone or LB . This was confirmed here since a ppsA/pckA mutant strain was unable to form rdar morphotype colonies and synthesize EPS or glycogen (Figure 1). However, strains carrying single deletions in ppsA or pckA were not impaired indicating that either arm of gluconeogenesis was sufficient to generate the precursor sugars required for polysaccharide production. Since gluconeogenesis is an energy consuming process that may be controlled by the adenylate energy charge , we also investigated the expression of adenylate kinase (Adk). Adk catalyzes the reversible conversion of ATP+AMP into two ADP molecules and is known to buffer ATP levels during periods of rapid ATP consumption . Adk expression was elevated in wild-type cultures relative to csgD mutant cultures (Figure 4; Table S1), suggesting that aggregating cells have an increased requirement for ATP. Furthermore, since the Adk reaction is the only route of de novo synthesis of ADP from AMP in Salmonella  these results could explain the increased AMP levels detected in csgD mutant colonies.
Enzymes for osmoprotectant synthesis and accumulation are up-regulated in S. Typhimurium rdar morphotype cells
OtsBA, coding for enzymes that catalyze trehalose biosynthesis , and kdpFABC, which encodes a high affinity potassium import system coupled to glutamate accumulation  were up-regulated in wild-type cultures at the time of aggregation (Figure 5E). Glycine betaine (betaine) or its precursor, choline, cannot be synthesized de novo by Salmonella  but can be transported into cells via the well-characterized proP and proU (proVWX) import systems . Expression of proP was induced in the wild-type cultures (Figure 4), but proVWX was not (Table S1). Proline can also act as an osmoprotectant and be transported through the proP and proU systems , however, we could not explain the higher levels of proline in csgD mutant colonies based on these results. yehZYXW, encoding a putative osmoprotectant import system , and osmE and osmY, osmotically-inducible genes encoding proteins of unknown function , were highly induced in wild-type cultures timed with aggregation (Figure 4, Figure 5E, Table S1). This suggested that during the aggregation process, cells in wild-type cultures were exposed to an environment of increased osmolarity.
Defences against reactive oxygen species (ROS) are induced in S. Typhimurium rdar morphotype cells
The tri-peptide glutathione (L-γ-glutamylcysteinylglycine; GSH) is a major reducing agent and acts as a detoxifying compound through non-enzymatic deactivation of ROS and the action of glutathione-S-transferase enzymes . Expression of gshA, coding for the enzyme catalyzing the first step in GSH synthesis, was not elevated in the wild-type background (Table S1). However, STM4267, encoding a glutathione-S-transferase, and yghA, encoding a putative glutathionylspermidine synthase, were up-regulated in wild-type cells coinciding with aggregation (Figure 4, Table S1). Glutathionylspermidine, a conjugate of GSH and spermidine, can also function as a detoxifying compound . Several oxidative stress-relieving enzymes, including cytosolic superoxide dismutase (SodA), a putative peroxidase (STM0402), and a NADH∶quinone oxidoreductase (WrbA)  were identified as abundant by proteomic analysis of wild-type rdar morphotype colonies (data not shown). Each of these genes, along with soxS from the soxRS superoxide response regulon , were expressed at higher levels in wild-type cultures (Figure 4, Table S1). Increased wraB expression in wild-type cells could explain the increased levels of oxidized NAD+ detected by metabolomics.
Expression levels of polyamine biosynthesis enzymes are similar in wild-type S. Typhimurium and csgD mutant strains
Polyamines have diverse roles within cells, including stabilization of phosphate charges on nucleic acids and other negatively charged polymers and scavenging of ROS . We analyzed expression of speA, speB, speC, and cadBA genes encoding four of the main decarboxylation enzymes for production of putrescine and cadaverine (Figure 3). speA, speB and speC had similar magnitudes of expression in wild-type and csgD mutant cultures, whereas cadBA was slightly elevated in wild-type cultures (Figure 4, Table S1). These results fit our hypothesis that polyamines accumulated in csgD mutant cells as a result of reduced carbon flux into gluconeogenesis and other biosynthetic pathways.
Levels of intracellular iron are limiting during growth in 1% tryptone
Iron limitation is known to activate csgD expression and formation of the rdar morphotype  and is also known to induce expression of different iron acquisition systems . EntCEBA, encoding enzymes for the biosynthesis of enterobactin, fhuA, encoding an outer membrane receptor for ferrichrome siderophores produced by fungi, and fhuF, which encodes a protein involved in the ferrioxamine B system , were induced in both wild-type and csgD mutant cultures during growth (Figure 4), indicating that iron was limiting during growth. However, the fhuA and entCEBA operons were induced higher in wild-type cultures (Table S1), suggesting that aggregation may also affect intracellular iron levels.
ABC transporters are up-regulated during growild-typeh
In Gram-negative bacteria, many transporters from the ATP-binding cassette (ABC) superfamily function as nutrient importers that utilize high-affinity periplasmic-binding proteins (PBP) to define their specificity . Several PBP, with specificities for carbohydrates, amino acids, peptides, or unknown substrates (Table S1), were identified as abundant by proteomic analysis of wild-type rdar morphotype colonies (data not shown). Expression of these operons, including lsrACDBFGE, encoding the transport and processing system for the AI-2 signalling molecule , were induced in both strain backgrounds at later time points during growth when nutrient limitation would occur (Figure 4, data not shown). Yrb, opp, lsr, glt and argT operons were up-regulated in wild-type cultures coinciding with aggregation (Figure 4; Table S1). The induction of diverse nutrient import systems may be necessary for cells to harvest all available nutrients in the current growth media or could represent an example of carbon source foraging , where cells expend energy to broaden their search for alternative energy sources.
Starvation in non-differentiating bacteria is known to induce a myriad of molecular changes to allow for more efficient nutrient scavenging and increased stress resistance . The results described indicate that the Salmonella rdar morphotype is a specialized multicellular physiology adapted to this survival response. This may be critical for Salmonella transmission by ensuring that enough cells survive to infect new hosts , , . Analyzing the metabolome and identification of the major metabolites by NMR and GC-MS revealed that rdar morphotype cells have a shift in central metabolism to gluconeogenesis and production of small molecules that aid in osmotic stress response. These changes were observed at the transcriptional level as part of a global temporal shift that was timed with aggregation.
S. Typhimurium rdar morphotype cells displayed increased carbon flux into gluconeogenesis at the onset of aggregation. In particular, PEP synthase and PEP carboxylase enzymes were required to synthesize sugars for production of EPS and glycogen. This result was undoubtedly influenced by growth on amino-acid based media, however, the significant up-regulation of gluconeogenesis in aggregation-positive wild-type cells compared to aggregation-negative csgD mutant cells was striking. This observation has implications for many types of natural biofilms and is likely not restricted to Salmonella. Polysaccharides are usually essential for aggregation to occur , , such as with alginate, Pel and Psl polysaccharides in Pseudomonas aeruginosa , and VPS in Vibrio cholerae . Glycogen is also important because it is known to enhance S. enterica survival  and was recently shown to play a critical role in transmission of V. cholerae . Under the conditions investigated, it is assumed that carbon flux is controlled by the catabolite repressor/activator (Cra) protein, which activates gluconeogenesis enzymes (ppsA, pckA, fbp) and represses sugar catabolism enzymes . In agreement with this hypothesis, the addition of glucose during growth leads to inhibition of the rdar morphotype (A.P. White and M.G. Surette, unpublished). Collectively, these results suggest that blocking gluconeogenesis may be an effective means to prevent or reduce biofilm formation in a wide variety of bacteria.
S. Typhimurium rdar morphotype cells displayed numerous stress-resistance adaptations that coincided with aggregation. Several osmoprotectants were detected at high levels in rdar morphotype colonies and transcriptional analysis verified that systems for osmoprotectant synthesis and transport were induced. Osmoprotectants are predicted to enhance desiccation survival by causing a reduction in water stress . We also observed that wild-type cells had an increased capacity for ROS defence, which would partially alleviate the damage caused to DNA, lipids and proteins known to occur during desiccation . Finally, the induction of nutrient acquisition systems as part of a carbon foraging or starvation response ,  would ensure swift revival of cells after long periods of “metabolic dormancy”. Our results agree with a recent study by Hinton and colleagues  who investigated S. Typhimurium biofilms using proteomic and microarray analysis. Similar stress-resistance adaptations have also been observed in other biofilm systems, including evidence for increased osmoprotection in E. coli  and ROS defence in P. aeruginosa . Each of these main stress adaptations are known to be controlled by σS , , , , and σS activity was measured to be almost three times higher in S. Typhimurium wild-type cells compared to csgD mutant cells. In E. coli, which shares most features of rdar morphotype regulation , CsgD was shown to have a stabilizing effect on σS protein levels  which could partially explain our findings. The only other metabolome comparison of biofilm and planktonic cells was performed with P. aeruginosa, and although these cell types had different metabolic profiles, individual metabolites were not identified .
One of the most intriguing questions arising from this study is how is the signal for Salmonella aggregation linked to metabolism and stress resistance? The transcriptional regulator CsgD is the most obvious candidate, acting in concert with σS , . However, analysis of the CsgD regulon in E. coli, did not reveal any gene targets linked to global carbon flux and relatively few that were directly linked to stress resistance . Based on these findings, we hypothesize that the primary role of CsgD is to control the aggregation process and that the majority of adaptations are the consequence of production of an extracellular matrix. Stress-inducing changes in the microenvironment of aggregated or biofilm cells have been observed before. In P. aeruginosa, the chelation of ions by extracellular DNA present in the biofilm caused activation of antibiotic and stress resistance pathways in the adjacent cells . It is possible that synthesis of an extracellular matrix by rdar morphotype cells causes an increase in the local osmolarity around aggregated cells  or mimics an increase in osmolarity by reducing the water activity, which, in turn, would expose cells to increased oxidative stress . The signal for S. Typhimurium cells to aggregate and the accompanying changes may be akin to a developmental process, such as sporulation in Myxococcus and Bacillus spp. , . Another possibility is that the rdar morphotype adaptations are genetically programmed changes that occur prior to experiencing the environmental stress (anticipatory regulation ).
c-di-GMP is a key regulatory molecule in the aggregation process. High intracellular levels have been linked to aggregation in numerous bacterial species, including Salmonella, E. coli, Pseudomonas spp. and Vibrio spp. , . In Salmonella and E. coli, there is a complex interplay between c-di-GMP, CsgD, σS, and other global regulators, such as CsrA (Text S1) , , . Although we didn't attempt to dissect this regulatory network, both curli production (via activation of csgD) and cellulose production (via activation of adrA) are indicators of high intracellular levels of c-di-GMP . While cellulose production through AdrA appears to be a very specific response to a specific c-di-GMP signalling pathway, a recent study challenges the role of some diguanylate cyclases in modulating cytoplasmic c-di-GMP pools in S. Typhimurium . There is evidence that a high concentration of c-di-GMP can regulate expression of soxS (ROS defence), fur (iron acquisition), and other global regulatory proteins in E. coli , however the physiological relevance of this study is questionable. In recent experiments performed with P. aeruginosa, Starkey et al.  found that the number of genes regulated in response of c-di-GMP was relatively small compared to the number of genes differentially regulated as a result of aggregation.
The elaborate temporal program associated with the Salmonella rdar morphotype is initiated through the aggregation regulator CsgD. The demands for exopolysaccharide production in turn cause changes in the expression of metabolic genes associated with gluconeogenesis. Finally the microenvironment that results from being embedded in a self-produced matrix results in the induction of numerous pathways associated with stress tolerance. Thus what appears to be a defined temporal program is not coordinated through a master regulatory pathway but is the result of the cell producing and responding to its own matrix (Fig. 6). Since growth in multicellular aggregates and biofilms is common among microorganisms, our findings may represent a general phenomenon that helps to explain some of the inherent resistant properties of biofilms.
Aggregation is initiated by the activation of the CsgD regulon; BapA is a large cell-surface protein involved in biofilm formation . The metabolic demand of polysaccharide production leads to induction of gluconeogenesis and the subsequent response to the self-produced matrix activates pathways that lead to general biofilm phenotypes. These later processes represent emergent behaviors and are not under control of a “biofilm specific’ regulatory cascade.
Materials and Methods
Bacterial strains, media and growth conditions
S. Typhimurium strain ATCC 14028 was used as the wild-type strain in this study. The isogenic csgD mutant strain (ΔcsgD), previously named ΔagfD , has a 612 bp deletion in csgD (encoding amino acids 6 to 210 in the mature protein). Strains were grown for 16–20 h at 37°C with agitation in Miller's Luria-Bertani broth (1.0% salt) or LB without salt (LBns), supplemented with 50 µg/ml Kanamycin (Kan), if necessary, before performing additional experiments. To obtain colonies, 1 µl aliquots of overnight cultures were spotted on T agar (1% tryptone, 1.5% Difco agar) and incubated at 28°C for up to 5 days. For bioluminescence assays, reporter strain cultures were diluted 1 in 600 in T broth supplemented with 50 µg/ml Kan to a final volume of 150 µl in 96-well clear-bottom black plates (9520 Costar; Corning Inc.). The culture in each well was overlaid with 50 µl mineral oil prior to starting the assays. Cultures were assayed for luminescence (0.1s) and absorbance (620 nm, 0.1s) every 30 min during growth at 28°C with agitation in a Wallac Victor2 (Perkin-Elmer Life Sciences, Boston, Mass.).
Construction of luciferase reporters
Promoter-containing DNA regions were PCR amplified from wild-type S. Typhimurium, purified (Qiagen Inc.), digested with XhoI and BamHI (Invitrogen Canada Inc.), and ligated using T4 DNA ligase (Invitrogen Canada Inc.) into pCS26-Pac (XhoI-BamHI) or pU220 (BamHI-XhoI) reporter vectors containing the luxCDABE operon from Photorhabdus luminescens . All primers used for reporter construction are listed in Table S2. Salmonella strains were transformed with plasmids via electroporation (Bio-Rad Laboratories Inc.).
The csgDEFG (agfDEFG), csgBAC (agfBAC), adrA and mlrA reporters have been previously described , . The promoter sequences in the sig38H4  and sig70_7  reporters are (ATAATTCCATGCGGTTTCGCTAAAATCATGTATACTTATTATCAATT) and (AATAATTCTTGATATTTATGCTTCCGGCTCGTATTTTACGTGCAATT), respectively; the −35 and −10 promoter regions are underlined. These reporters were selected from a library constructed with the above sequences with four degenerate positions in each promoter (K. Pabbaraju and M.G. Surette, unpublished). Light production as the result of transcription from these synthetic promoters reflects σS- or σ70-RNA polymerase activity.
Construction and characterization of S. Typhimurium deletion mutants
ΔppsA and ΔpckA mutant strains were created by deletion mutagenesis of wild-type S. Typhimurium using a chloramphenicol cassette as described . Chromosomal loci of the generated mutants were verified by PCR using a primer specific to the insert and a primer that annealed to sequence that flanked the disrupted loci (Table S2). To ensure the absence of secondary mutations, all generated deletions were moved into a clean wild-type background by P22 transduction . The pckA::cat mutant was cured of the chloramphenicol cassette as previously described . The unmarked ΔpckA mutant was transduced with the P22 lysate of ppsA::cat to generate a ΔpckA/ΔppsA double mutant. The mutants were phenotypically tested by examining their capacity to grow in M9 minimal media supplemented with 0.2% glucose, 0.2% glycerol, 0.4% acetate, 0.4% citrate, or 0.4% succinate as previously described .
Staining of colonies for glycogen production
An aqueous iodine solution (0.01 M I2, 0.03 M KI)  was initially tested for glycogen staining but did not stain glycogen intensely enough. Therefore, the iodine concentration was increased to 0.1 M and the solution was vortexed for 5 min prior to staining. 5 mL of solution was added to each plate and swirled around the entire plate area and left to stain for 5 min before taking pictures.
Extraction of metabolites from S. Typhimurium colonies
Wild-type or csgD mutant colonies were removed from T agar after 2 or 5 days growth and placed into 2 mL sterile vials containing 0.2 g of 0.1 mm Zirconia/Silica beads (BioSpec Products Inc., Bartlesville, OK, USA); two colonies were added to each vial. Immediately following the addition of 1 mL of ice-cold methanol, cells were homogenized for 2 min using a Mini-Beadbeater 8 (BioSpec Products Inc., Bartlesville, OK, USA). Beads and cell debris were sedimented by centrifugation (20,000g, 2 min), the supernatant was removed and filtered through a 0.22 µ Spin-X centrifuge tube filter (Costar, Corning Inc) by centrifugation (20,000g, 1 min). Samples were evaporated to dryness using a Centrivap concentrator (Labconco Corp., Kansas City, MO) and were stored at −80°C prior to NMR or GC-MS analysis. We chose to extract metabolites using ice-cold methanol because this method yielded the most comprehensive array of metabolites in E. coli when six commonly used procedures were compared . To determine the number of colony forming units (CFU) at the time of extraction, colonies removed from agar were resuspended in 0.5 mL of phosphate-buffered saline, homogenized in a tissue grinder for ∼20 s, serially diluted in triplicate, plated in duplicate in 5 µL drops onto LB agar and incubated at 28°C overnight.
Preparation of samples for 1H NMR analysis
Dried samples were resuspended in 600 µL of deionized water and filtered through pre-wetted NanoSep 3K filters (Pall, Ann Arbor, MI, USA) by centrifugation (20,000g, 60min) to remove any dissolved proteins. 130 µL of metabolite sample buffer (0.5 M sodium phosphate (monobasic)+2.5 mM 2,2-dimethyl-2-silapentane-5-sulfonate (DSS)) and 10 µL of 1M sodium azide was added to bring the volume of each sample to ∼650 µL. pH values ranged between 7.2 and 7.4 for all samples tested (data not shown), therefore samples were not pH-adjusted prior to analysis.
Preparation of samples for GC-MS analysis
For GC-MS, dried samples were resuspended in 60 µl of methoxyamine in anhydrous pyridine (20 mg/ml), transferred to a glass vial and incubated overnight at room temperature on a rotary shaker. 60 µl of N-methyl-N-trimethylsilyltrifluoroacetamide (MSTFA) and 6.0 µl of chlorotrimethylsilane (TMS-Cl) were added and the reaction was continued for one hour. A 100 µl aliquot of the reaction mixture was diluted with 900 µl of hexane prior to analysis.
1H NMR analysis
All experiments were performed on a Bruker Advance 600 MHz spectrometer (Bruker Daltonics) operating at 600.22 MHz and equipped with a 5-mm TXI probe at 298 K for solution-state analysis. All one-dimensional 1H NMR spectra were acquired using a standard Bruker noesypr1d pulse sequence in which the residual water peak was irradiated during the relaxation delay of 1.0 s and during the mixing time of 100 ms. A total of 256 scans were collected into 65,536 data points over a spectral width of 12,195 Hz, with a 5-s repetition time. A line broadening of 0.5 Hz was applied to the spectra prior to Fourier transformation, phasing and baseline correction. To confirm spectral assignments, a 1H,13C heteronuclear single quantum correlation (HSQC) and a 1H,1H total correlation (TOCSY) spectra were acquired. A standard echo/antiecho-TPPI gradient selection pulse sequence  was used for HSQC spectrum. The parameters comprised a J-coupling delay of 0.86 ms, time domain points of 2 k (F2) and 256 (F1), spectral width (1H) of 12 ppm, spectral width (13C) of 169 ppm, GARP 13C decoupling, 80 scans/increment, acquisition time of 0.14 s, and a relaxation delay of 1.6 s. A phase sensitive homonuclear Hartman-Hahn transfer using DIPSI2 sequence for mixing with water suppression using exitation sculping with gradients  was used for TOCSY spectrum with parameters comprised a TOCSY mixing time 0.12 s, time domain points of 2k (F2) and 400 (F1), spectral width (both 1H) of 12 ppm, 64 scans/increment, acquisition time of 0.14 s, and a relaxation delay of 1.0 s.
Metabolite identification and quantification from one-dimensional 1H NMR spectra was achieved using the Profiler module of Chenomx NMR Suite version 4.6 (Chenomx. Inc., Edmonton, Canada). Chenomx Profiler is linked to a database of metabolites whose unique NMR spectral signatures are encoded at various spectrophotometer frequencies, including 600 MHz. Two-dimensional 1H NMR was employed to confirm compound identities where necessary. Metabolites were quantified by comparison to the internal standard DSS, which also served as a chemical shift reference.
Experiments were performed on an Agilent 5975B inert XL gas chromatograph (6890N) and mass spectrometer (EI/CI) (Agilent Technologies Canada Inc., Mississauga, Ont). Individual metabolites were identified by comparison to the HSALLMASS compound database using the Agilent MSD Security ChemStation software.
One-dimensional 1H NMR spectra were imported into Chenomx NMR Suite version 4.6 (Chenomx) for spectral binning. All shifts related to the solvent (i.e., in the range of 4.5–5.0 ppm) and DSS were excluded, and the remaining spectral regions were divided into 0.04-ppm bins. GC-MS spectra were also divided into 0.04-ppm bins. GC-MS spectra were processed as peaks deconvoluted using the Automated Mass Spectral Deconvolution and Identification System (AMDIS, Version 2.64, NIST, US) and subsequently filtered using spectconnect (http://spectconnect.mit.edu/; PMID: 17263323). Chemometric analysis was performed using SIMCA-P version 11.5 (Umetrics) with unsupervised PCA (to look for outliers and other anomalous sources of variance) or orthogonal partial least square discriminate analysis (OPLS-DA). Variables were scaled to unit variance to ensure equal contributions to the models.
Comparison of promoter-luciferase reporter expression in wild-type S. Typhimurium and csgD deletion mutant strains.
(0.21 MB DOC)
PCR primers used for reporter construction or mutagenesis.
(0.21 MB DOC)
The authors wish to thank Dae-Kyun Ro for use of the Agilent GC-MS system, Wook Kim, Dave Schreimer and members of the Southern Alberta Mass Spectrometry Center for help with proteomic analysis, and Jennifer Happe for sequencing of reporter plasmids from the S. Typhimurium random promoter library. We also thank Ken Sanderson, Bill Kay, Shawn Lewenza, Heidi Mulcahy, Eoin O'Grady, Chris Sibley and members of the Surette laboratory for helpful discussions and critical reading of the manuscript.
Conceived and designed the experiments: APW MGS. Performed the experiments: APW AW DA PZ RS. Analyzed the data: APW AW DA PZ RS HJV MGS. Contributed reagents/materials/analysis tools: AW HJV. Wrote the paper: APW MGS.
- 1. Stoodley P, Sauer K, Davies DG, Costerton JW (2002) Biofilms as complex differentiated communities. Annu Rev Microbiol 56: 187–209.P. StoodleyK. SauerDG DaviesJW Costerton2002Biofilms as complex differentiated communities.Annu Rev Microbiol56187209
- 2. Fux CA, Costerton JW, Stewart PS, Stoodley P (2005) Survival strategies of infectious biofilms. Trends Microbiol 13: 34–40.CA FuxJW CostertonPS StewartP. Stoodley2005Survival strategies of infectious biofilms.Trends Microbiol133440
- 3. Kolter R, Greenberg EP (2006) The superficial life of microbes. Nature 441: 300–302.R. KolterEP Greenberg2006The superficial life of microbes.Nature441300302
- 4. Mah TF, Pitts B, Pellock B, Walker GC, Stewart PS, et al. (2003) A genetic basis for Pseudomonas aeruginosa biofilm antibiotic resistance. Nature 426: 306–310.TF MahB. PittsB. PellockGC WalkerPS Stewart2003A genetic basis for Pseudomonas aeruginosa biofilm antibiotic resistance.Nature426306310
- 5. Mulcahy H, Charron-Mazenod L, Lewenza S (2008) Extracellular DNA chelates cations and induces antibiotic resistance in Pseudomonas aeruginosa biofilms. PLoS Pathog 4: e1000213.H. MulcahyL. Charron-MazenodS. Lewenza2008Extracellular DNA chelates cations and induces antibiotic resistance in Pseudomonas aeruginosa biofilms.PLoS Pathog4e1000213
- 6. Sutherland I (2001) Biofilm exopolysaccharides: a strong and sticky framework. Microbiology 147: 3–9.I. Sutherland2001Biofilm exopolysaccharides: a strong and sticky framework.Microbiology14739
- 7. Collinson SK, Doig PC, Doran JL, Clouthier S, Trust TJ, et al. (1993) Thin, aggregative fimbriae mediate binding of Salmonella enteritidis to fibronectin. J Bacteriol 175: 12–18.SK CollinsonPC DoigJL DoranS. ClouthierTJ Trust1993Thin, aggregative fimbriae mediate binding of Salmonella enteritidis to fibronectin.J Bacteriol1751218
- 8. Romling U, Sierralta WD, Eriksson K, Normark S (1998) Multicellular and aggregative behaviour of Salmonella typhimurium strains is controlled by mutations in the agfD promoter. Mol Microbiol 28: 249–264.U. RomlingWD SierraltaK. ErikssonS. Normark1998Multicellular and aggregative behaviour of Salmonella typhimurium strains is controlled by mutations in the agfD promoter.Mol Microbiol28249264
- 9. White AP, Gibson DL, Kim W, Kay WW, Surette MG (2006) Thin aggregative fimbriae and cellulose enhance long-term survival and persistence of Salmonella. J Bacteriol 188: 3219–3227.AP WhiteDL GibsonW. KimWW KayMG Surette2006Thin aggregative fimbriae and cellulose enhance long-term survival and persistence of Salmonella.J Bacteriol18832193227
- 10. Gibson DL, White AP, Snyder SD, Martin S, Heiss C, et al. (2006) Salmonella produces an O-antigen capsule regulated by AgfD and important for environmental persistence. J Bacteriol 188: 7722–7730.DL GibsonAP WhiteSD SnyderS. MartinC. Heiss2006Salmonella produces an O-antigen capsule regulated by AgfD and important for environmental persistence.J Bacteriol18877227730
- 11. Solano C, Garcia B, Valle J, Berasain C, Ghigo JM, et al. (2002) Genetic analysis of Salmonella enteritidis biofilm formation: critical role of cellulose. Mol Microbiol 43: 793–808.C. SolanoB. GarciaJ. ValleC. BerasainJM Ghigo2002Genetic analysis of Salmonella enteritidis biofilm formation: critical role of cellulose.Mol Microbiol43793808
- 12. Zogaj X, Nimtz M, Rohde M, Bokranz W, Romling U (2001) The multicellular morphotypes of Salmonella typhimurium and Escherichia coli produce cellulose as the second component of the extracellular matrix. Mol Microbiol 39: 1452–1463.X. ZogajM. NimtzM. RohdeW. BokranzU. Romling2001The multicellular morphotypes of Salmonella typhimurium and Escherichia coli produce cellulose as the second component of the extracellular matrix.Mol Microbiol3914521463
- 13. Gerstel U, Romling U (2001) Oxygen tension and nutrient starvation are major signals that regulate agfD promoter activity and expression of the multicellular morphotype in Salmonella typhimurium. Environ Microbiol 3: 638–648.U. GerstelU. Romling2001Oxygen tension and nutrient starvation are major signals that regulate agfD promoter activity and expression of the multicellular morphotype in Salmonella typhimurium.Environ Microbiol3638648
- 14. Lacour S, Landini P (2004) SigmaS-dependent gene expression at the onset of stationary phase in Escherichia coli: function of sigmaS-dependent genes and identification of their promoter sequences. J Bacteriol 186: 7186–7195.S. LacourP. Landini2004SigmaS-dependent gene expression at the onset of stationary phase in Escherichia coli: function of sigmaS-dependent genes and identification of their promoter sequences.J Bacteriol18671867195
- 15. Weber H, Polen T, Heuveling J, Wendisch VF, Hengge R (2005) Genome-wide analysis of the general stress response network in Escherichia coli: sigmaS-dependent genes, promoters, and sigma factor selectivity. J Bacteriol 187: 1591–1603.H. WeberT. PolenJ. HeuvelingVF WendischR. Hengge2005Genome-wide analysis of the general stress response network in Escherichia coli: sigmaS-dependent genes, promoters, and sigma factor selectivity.J Bacteriol18715911603
- 16. Kader A, Simm R, Gerstel U, Morr M, Romling U (2006) Hierarchical involvement of various GGDEF domain proteins in rdar morphotype development of Salmonella enterica serovar Typhimurium. Mol Microbiol 60: 602–616.A. KaderR. SimmU. GerstelM. MorrU. Romling2006Hierarchical involvement of various GGDEF domain proteins in rdar morphotype development of Salmonella enterica serovar Typhimurium.Mol Microbiol60602616
- 17. Pesavento C, Becker G, Sommerfeldt N, Possling A, Tschowri N, et al. (2008) Inverse regulatory coordination of motility and curli-mediated adhesion in Escherichia coli. Genes Dev 22: 2434–2446.C. PesaventoG. BeckerN. SommerfeldtA. PosslingN. Tschowri2008Inverse regulatory coordination of motility and curli-mediated adhesion in Escherichia coli.Genes Dev2224342446
- 18. Simm R, Remminghorst U, Ahmad I, Zakikhany K, Romling U (2009) A role for the EAL-like protein STM1344 in regulation of CsgD expression and motility in Salmonella enterica serovar Typhimurium. J Bacteriol 191: 3928–3937.R. SimmU. RemminghorstI. AhmadK. ZakikhanyU. Romling2009A role for the EAL-like protein STM1344 in regulation of CsgD expression and motility in Salmonella enterica serovar Typhimurium.J Bacteriol19139283937
- 19. Anriany YA, Weiner RM, Johnson JA, De Rezende CE, Joseph SW (2001) Salmonella enterica serovar Typhimurium DT104 displays a rugose phenotype. Appl Environ Microbiol 67: 4048–4056.YA AnrianyRM WeinerJA JohnsonCE De RezendeSW Joseph2001Salmonella enterica serovar Typhimurium DT104 displays a rugose phenotype.Appl Environ Microbiol6740484056
- 20. Scher K, Romling U, Yaron S (2005) Effect of heat, acidification, and chlorination on Salmonella enterica serovar Typhimurium cells in a biofilm formed at the air-liquid interface. Appl Environ Microbiol 71: 1163–1168.K. ScherU. RomlingS. Yaron2005Effect of heat, acidification, and chlorination on Salmonella enterica serovar Typhimurium cells in a biofilm formed at the air-liquid interface.Appl Environ Microbiol7111631168
- 21. Apel D, White AP, Grassl GA, Finlay BB, Surette MG (2009) Long-term survival of Salmonella enterica serovar Typhimurium reveals an infectious state that is underrepresented on laboratory media containing bile salts. Appl Environ Microbiol 75: 4923–4925.D. ApelAP WhiteGA GrasslBB FinlayMG Surette2009Long-term survival of Salmonella enterica serovar Typhimurium reveals an infectious state that is underrepresented on laboratory media containing bile salts.Appl Environ Microbiol7549234925
- 22. White AP, Surette MG (2006) Comparative genetics of the rdar morphotype in Salmonella. J Bacteriol 188: 8395–8406.AP WhiteMG Surette2006Comparative genetics of the rdar morphotype in Salmonella.J Bacteriol18883958406
- 23. Beloin C, Valle J, Latour-Lambert P, Faure P, Kzreminski M, et al. (2004) Global impact of mature biofilm lifestyle on Escherichia coli K-12 gene expression. Mol Microbiol 51: 659–674.C. BeloinJ. ValleP. Latour-LambertP. FaureM. Kzreminski2004Global impact of mature biofilm lifestyle on Escherichia coli K-12 gene expression.Mol Microbiol51659674
- 24. Moorthy S, Watnick PI (2005) Identification of novel stage-specific genetic requirements through whole genome transcription profiling of Vibrio cholerae biofilm development. Mol Microbiol 57: 1623–1635.S. MoorthyPI Watnick2005Identification of novel stage-specific genetic requirements through whole genome transcription profiling of Vibrio cholerae biofilm development.Mol Microbiol5716231635
- 25. Whiteley M, Bangera MG, Bumgarner RE, Parsek MR, Teitzel GM, et al. (2001) Gene expression in Pseudomonas aeruginosa biofilms. Nature 413: 860–864.M. WhiteleyMG BangeraRE BumgarnerMR ParsekGM Teitzel2001Gene expression in Pseudomonas aeruginosa biofilms.Nature413860864
- 26. O'Toole GA, Kolter R (1998) Initiation of biofilm formation in Pseudomonas fluorescens WCS365 proceeds via multiple, convergent signalling pathways: a genetic analysis. Mol Microbiol 28: 449–461.GA O'TooleR. Kolter1998Initiation of biofilm formation in Pseudomonas fluorescens WCS365 proceeds via multiple, convergent signalling pathways: a genetic analysis.Mol Microbiol28449461
- 27. Prigent-Combaret C, Vidal O, Dorel C, Lejeune P (1999) Abiotic surface sensing and biofilm-dependent regulation of gene expression in Escherichia coli. J Bacteriol 181: 5993–6002.C. Prigent-CombaretO. VidalC. DorelP. Lejeune1999Abiotic surface sensing and biofilm-dependent regulation of gene expression in Escherichia coli.J Bacteriol18159936002
- 28. Sauer K, Camper AK, Ehrlich GD, Costerton JW, Davies DG (2002) Pseudomonas aeruginosa displays multiple phenotypes during development as a biofilm. J Bacteriol 184: 1140–1154.K. SauerAK CamperGD EhrlichJW CostertonDG Davies2002Pseudomonas aeruginosa displays multiple phenotypes during development as a biofilm.J Bacteriol18411401154
- 29. Tremoulet F, Duche O, Namane A, Martinie B, Labadie JC (2002) A proteomic study of Escherichia coli O157:H7 NCTC 12900 cultivated in biofilm or in planktonic growth mode. FEMS Microbiol Lett 215: 7–14.F. TremouletO. DucheA. NamaneB. MartinieJC Labadie2002A proteomic study of Escherichia coli O157:H7 NCTC 12900 cultivated in biofilm or in planktonic growth mode.FEMS Microbiol Lett215714
- 30. Ghigo JM (2003) Are there biofilm-specific physiological pathways beyond a reasonable doubt? Res Microbiol 154: 1–8.JM Ghigo2003Are there biofilm-specific physiological pathways beyond a reasonable doubt?Res Microbiol15418
- 31. Boles BR, Thoendel M, Singh PK (2004) Self-generated diversity produces “insurance effects” in biofilm communities. Proc Natl Acad Sci USA 101: 16630–16635.BR BolesM. ThoendelPK Singh2004Self-generated diversity produces “insurance effects” in biofilm communities.Proc Natl Acad Sci USA1011663016635
- 32. Stewart PS, Franklin MJ (2008) Physiological heterogeneity in biofilms. Nat Rev Microbiol 6: 199–210.PS StewartMJ Franklin2008Physiological heterogeneity in biofilms.Nat Rev Microbiol6199210
- 33. Lewis K (2007) Persister cells, dormancy and infectious disease. Nat Rev Microbiol 5: 48–56.K. Lewis2007Persister cells, dormancy and infectious disease.Nat Rev Microbiol54856
- 34. Trygg J, Holmes E, Lundstedt T (2007) Chemometrics in metabonomics. J Proteome Res 6: 469–479.J. TryggE. HolmesT. Lundstedt2007Chemometrics in metabonomics.J Proteome Res6469479
- 35. Fraenkel DG (1996) Glycolysis. In: Curtiss R III, Ingraham JL, Lin ECC, Low KB, Magasanik B, et al., editors. Escherichia coli and Salmonella, Cellular and Molecular Biology. 2nd ed. Washington D.C.: ASM Press. pp. 189–198.DG Fraenkel1996Glycolysis.R. Curtiss IIIJL IngrahamECC LinKB LowB. MagasanikEscherichia coli and Salmonella, Cellular and Molecular Biology. 2nd edWashington D.C.ASM Press189198
- 36. Lucht JM, Bremer E (1994) Adaptation of Escherichia coli to high osmolarity environments: osmoregulation of the high-affinity glycine betaine transport system proU. FEMS Microbiol Rev 14: 3–20.JM LuchtE. Bremer1994Adaptation of Escherichia coli to high osmolarity environments: osmoregulation of the high-affinity glycine betaine transport system proU.FEMS Microbiol Rev14320
- 37. Underwood AH, Newsholme EA (1965) Some Properties of Fructose 1,6-Diphosphatase of Rat Liver and Their Relation to the Control of Gluconeogenesis. Biochem J 95: 767–774.AH UnderwoodEA Newsholme1965Some Properties of Fructose 1,6-Diphosphatase of Rat Liver and Their Relation to the Control of Gluconeogenesis.Biochem J95767774
- 38. Stocki SL, Annett CB, Sibley CD, McLaws M, Checkley SL, et al. (2007) Persistence of Salmonella on egg conveyor belts is dependent on the belt type but not on the rdar morphotype. Poultry Sci 86: 2375–2383.SL StockiCB AnnettCD SibleyM. McLawsSL Checkley2007Persistence of Salmonella on egg conveyor belts is dependent on the belt type but not on the rdar morphotype.Poultry Sci8623752383
- 39. Grantcharova N, Peters V, Monteiro C, Zakikhany K, Romling U (2010) Bistable expression of CsgD in biofilm development of Salmonella enterica serovar typhimurium. J Bacteriol 192: 456–466.N. GrantcharovaV. PetersC. MonteiroK. ZakikhanyU. Romling2010Bistable expression of CsgD in biofilm development of Salmonella enterica serovar typhimurium.J Bacteriol192456466
- 40. White AP, Gibson DL, Grassl GA, Kay WW, Finlay BB, et al. (2008) Aggregation via the red, dry, and rough morphotype is not a virulence adaptation in Salmonella enterica serovar Typhimurium. Infect Immun 76: 1048–1058.AP WhiteDL GibsonGA GrasslWW KayBB Finlay2008Aggregation via the red, dry, and rough morphotype is not a virulence adaptation in Salmonella enterica serovar Typhimurium.Infect Immun7610481058
- 41. Sauer U, Eikmanns BJ (2005) The PEP-pyruvate-oxaloacetate node as the switch point for carbon flux distribution in bacteria. FEMS Microbiol Rev 29: 765–794.U. SauerBJ Eikmanns2005The PEP-pyruvate-oxaloacetate node as the switch point for carbon flux distribution in bacteria.FEMS Microbiol Rev29765794
- 42. Sezonov G, Joseleau-Petit D, D'Ari R (2007) Escherichia coli physiology in Luria-Bertani broth. J Bacteriol 189: 8746–8749.G. SezonovD. Joseleau-PetitR. D'Ari2007Escherichia coli physiology in Luria-Bertani broth.J Bacteriol18987468749
- 43. Chapman AG, Fall L, Atkinson DE (1971) Adenylate energy charge in Escherichia coli during growth and starvation. J Bacteriol 108: 1072–1086.AG ChapmanL. FallDE Atkinson1971Adenylate energy charge in Escherichia coli during growth and starvation.J Bacteriol10810721086
- 44. Gutierrez JA, Csonka LN (1995) Isolation and characterization of adenylate kinase (adk) mutations in Salmonella typhimurium which block the ability of glycine betaine to function as an osmoprotectant. J Bacteriol 177: 390–400.JA GutierrezLN Csonka1995Isolation and characterization of adenylate kinase (adk) mutations in Salmonella typhimurium which block the ability of glycine betaine to function as an osmoprotectant.J Bacteriol177390400
- 45. Checroun C, Gutierrez C (2004) Sigma(s)-dependent regulation of yehZYXW, which encodes a putative osmoprotectant ABC transporter of Escherichia coli. FEMS Microbiol Lett 236: 221–226.C. ChecrounC. Gutierrez2004Sigma(s)-dependent regulation of yehZYXW, which encodes a putative osmoprotectant ABC transporter of Escherichia coli.FEMS Microbiol Lett236221226
- 46. Carmel-Harel O, Storz G (2000) Roles of the glutathione- and thioredoxin-dependent reduction systems in the Escherichia coli and Saccharomyces cerevisiae responses to oxidative stress. Annu Rev Microbiol 54: 439–461.O. Carmel-HarelG. Storz2000Roles of the glutathione- and thioredoxin-dependent reduction systems in the Escherichia coli and Saccharomyces cerevisiae responses to oxidative stress.Annu Rev Microbiol54439461
- 47. Bollinger JM Jr, Kwon DS, Huisman GW, Kolter R, Walsh CT (1995) Glutathionylspermidine metabolism in Escherichia coli. Purification, cloning, overproduction, and characterization of a bifunctional glutathionylspermidine synthetase/amidase. J Biol Chem 270: 14031–14041.JM Bollinger JrDS KwonGW HuismanR. KolterCT Walsh1995Glutathionylspermidine metabolism in Escherichia coli. Purification, cloning, overproduction, and characterization of a bifunctional glutathionylspermidine synthetase/amidase.J Biol Chem2701403114041
- 48. Andrade SL, Patridge EV, Ferry JG, Einsle O (2007) Crystal structure of the NADH:quinone oxidoreductase WrbA from Escherichia coli. J Bacteriol 189: 9101–9107.SL AndradeEV PatridgeJG FerryO. Einsle2007Crystal structure of the NADH:quinone oxidoreductase WrbA from Escherichia coli.J Bacteriol18991019107
- 49. Storz G, Imlay JA (1999) Oxidative stress. Curr Opin Microbiol 2: 188–194.G. StorzJA Imlay1999Oxidative stress.Curr Opin Microbiol2188194
- 50. Shah P, Swiatlo E (2008) A multifaceted role for polyamines in bacterial pathogens. Mol Microbiol 68: 4–16.P. ShahE. Swiatlo2008A multifaceted role for polyamines in bacterial pathogens.Mol Microbiol68416
- 51. Masse E, Vanderpool CK, Gottesman S (2005) Effect of RyhB small RNA on global iron use in Escherichia coli. J Bacteriol 187: 6962–6971.E. MasseCK VanderpoolS. Gottesman2005Effect of RyhB small RNA on global iron use in Escherichia coli.J Bacteriol18769626971
- 52. Braun V, Hantke K, Koster W (1998) Bacterial iron transport: mechanisms, genetics and regulation. In: Sigel A, Sigel H, editors. Metal Ions in Biological Systems. New York: Marcel Dekker, Inc. pp. 67–145.V. BraunK. HantkeW. Koster1998Bacterial iron transport: mechanisms, genetics and regulation.A. SigelH. SigelMetal Ions in Biological SystemsNew YorkMarcel Dekker, Inc67145
- 53. Davidson AL, Maloney PC (2007) ABC transporters: how small machines do a big job. Trends Microbiol 15: 448–455.AL DavidsonPC Maloney2007ABC transporters: how small machines do a big job.Trends Microbiol15448455
- 54. Taga ME, Miller ST, Bassler BL (2003) Lsr-mediated transport and processing of AI-2 in Salmonella typhimurium. Mol Microbiol 50: 1411–1427.ME TagaST MillerBL Bassler2003Lsr-mediated transport and processing of AI-2 in Salmonella typhimurium.Mol Microbiol5014111427
- 55. Liu M, Durfee T, Cabrera JE, Zhao K, Jin DJ, et al. (2005) Global transcriptional programs reveal a carbon source foraging strategy by Escherichia coli. J Biol Chem 280: 15921–15927.M. LiuT. DurfeeJE CabreraK. ZhaoDJ Jin2005Global transcriptional programs reveal a carbon source foraging strategy by Escherichia coli.J Biol Chem2801592115927
- 56. Matin A, Auger EA, Blum PH, Schultz JE (1989) Genetic basis of starvation survival in nondifferentiating bacteria. Ann Rev Microbiol 43: 293–316.A. MatinEA AugerPH BlumJE Schultz1989Genetic basis of starvation survival in nondifferentiating bacteria.Ann Rev Microbiol43293316
- 57. Ryder C, Byrd M, Wozniak DJ (2007) Role of polysaccharides in Pseudomonas aeruginosa biofilm development. Curr Opin Microbiol 10: 644–648.C. RyderM. ByrdDJ Wozniak2007Role of polysaccharides in Pseudomonas aeruginosa biofilm development.Curr Opin Microbiol10644648
- 58. Yildiz FH, Schoolnik GK (1999) Vibrio cholerae O1 El Tor: identification of a gene cluster required for the rugose colony type, exopolysaccharide production, chlorine resistance, and biofilm formation. Proc Natl Acad Sci U S A 96: 4028–4033.FH YildizGK Schoolnik1999Vibrio cholerae O1 El Tor: identification of a gene cluster required for the rugose colony type, exopolysaccharide production, chlorine resistance, and biofilm formation.Proc Natl Acad Sci U S A9640284033
- 59. McMeechan A, Lovell MA, Cogan TA, Marston KL, Humphrey TJ, et al. (2005) Glycogen production by different Salmonella enterica serotypes: contribution of functional glgC to virulence, intestinal colonization and environmental survival. Microbiology 151: 3969–3977.A. McMeechanMA LovellTA CoganKL MarstonTJ Humphrey2005Glycogen production by different Salmonella enterica serotypes: contribution of functional glgC to virulence, intestinal colonization and environmental survival.Microbiology15139693977
- 60. Bourassa L, Camilli A (2009) Glycogen contributes to the environmental persistence and transmission of Vibrio cholerae. Mol Microbiol 72: 124–138.L. BourassaA. Camilli2009Glycogen contributes to the environmental persistence and transmission of Vibrio cholerae.Mol Microbiol72124138
- 61. Saier MH Jr, Ramseier TM (1996) The catabolite repressor/activator (Cra) protein of enteric bacteria. J Bacteriol 178: 3411–3417.MH Saier JrTM Ramseier1996The catabolite repressor/activator (Cra) protein of enteric bacteria.J Bacteriol17834113417
- 62. Potts M (1994) Desiccation tolerance of prokaryotes. Microbiol Rev 58: 755–805.M. Potts1994Desiccation tolerance of prokaryotes.Microbiol Rev58755805
- 63. Hamilton S, Bongaerts RJ, Mulholland F, Cochrane B, Porter J, et al. (2009) The transcriptional programme of Salmonella enterica serovar Typhimurium reveals a key role for tryptophan metabolism in biofilms. BMC Genomics 10: 599.S. HamiltonRJ BongaertsF. MulhollandB. CochraneJ. Porter2009The transcriptional programme of Salmonella enterica serovar Typhimurium reveals a key role for tryptophan metabolism in biofilms.BMC Genomics10599
- 64. Gualdi L, Tagliabue L, Landini P (2007) Biofilm formation-gene expression relay system in Escherichia coli: modulation of sigmaS-dependent gene expression by the CsgD regulatory protein via sigmaS protein stabilization. J Bacteriol 189: 8034–8043.L. GualdiL. TagliabueP. Landini2007Biofilm formation-gene expression relay system in Escherichia coli: modulation of sigmaS-dependent gene expression by the CsgD regulatory protein via sigmaS protein stabilization.J Bacteriol18980348043
- 65. Gjersing EL, Herberg JL, Horn J, Schaldach CM, Maxwell RS (2007) NMR metabolomics of planktonic and biofilm modes of growth in Pseudomonas aeruginosa. Anal Chem 79: 8037–8045.EL GjersingJL HerbergJ. HornCM SchaldachRS Maxwell2007NMR metabolomics of planktonic and biofilm modes of growth in Pseudomonas aeruginosa.Anal Chem7980378045
- 66. Brombacher E, Baratto A, Dorel C, Landini P (2006) Gene expression regulation by the Curli activator CsgD protein: modulation of cellulose biosynthesis and control of negative determinants for microbial adhesion. J Bacteriol 188: 2027–2037.E. BrombacherA. BarattoC. DorelP. Landini2006Gene expression regulation by the Curli activator CsgD protein: modulation of cellulose biosynthesis and control of negative determinants for microbial adhesion.J Bacteriol18820272037
- 67. Aguilar C, Vlamakis H, Losick R, Kolter R (2007) Thinking about Bacillus subtilis as a multicellular organism. Curr Opin Microbiol 10: 638–643.C. AguilarH. VlamakisR. LosickR. Kolter2007Thinking about Bacillus subtilis as a multicellular organism.Curr Opin Microbiol10638643
- 68. Errington J (2003) Regulation of endospore formation in Bacillus subtilis. Nat Rev Microbiol 1: 117–126.J. Errington2003Regulation of endospore formation in Bacillus subtilis.Nat Rev Microbiol1117126
- 69. Tagkopoulos I, Liu Y-C, Tavazoie S (2008) Predictive behaviour within microbial genetic networks. Science 320: 1313–1317.I. TagkopoulosY-C LiuS. Tavazoie2008Predictive behaviour within microbial genetic networks.Science32013131317
- 70. Hengge R (2009) Principles of c-di-GMP signalling in bacteria. Nat Rev Microbiol 7: 263–273.R. Hengge2009Principles of c-di-GMP signalling in bacteria.Nat Rev Microbiol7263273
- 71. Weber H, Pesavento C, Possling A, Tischendorf G, Hengge R (2006) Cyclic-di-GMP-mediated signalling within the sigma network of Escherichia coli. Mol Microbiol 62: 1014–1034.H. WeberC. PesaventoA. PosslingG. TischendorfR. Hengge2006Cyclic-di-GMP-mediated signalling within the sigma network of Escherichia coli.Mol Microbiol6210141034
- 72. Solano C, Garcia B, Latasa C, Toledo-Arana A, Zorraquino V, et al. (2009) Genetic reductionist approach for dissecting individual roles of GGDEF proteins within the c-di-GMP signaling network in Salmonella. Proc Natl Acad Sci U S A 106: 7997–8002.C. SolanoB. GarciaC. LatasaA. Toledo-AranaV. Zorraquino2009Genetic reductionist approach for dissecting individual roles of GGDEF proteins within the c-di-GMP signaling network in Salmonella.Proc Natl Acad Sci U S A10679978002
- 73. Mendez-Ortiz MM, Hyodo M, Hayakawa Y, Membrillo-Hernandez J (2006) Genome-wide transcriptional profile of Escherichia coli in response to high levels of the second messenger 3′,5′-cyclic diguanylic acid. J Biol Chem 281: 8090–8099.MM Mendez-OrtizM. HyodoY. HayakawaJ. Membrillo-Hernandez2006Genome-wide transcriptional profile of Escherichia coli in response to high levels of the second messenger 3′,5′-cyclic diguanylic acid.J Biol Chem28180908099
- 74. Starkey M, Hickman JH, Ma L, Zhang N, De Long S, et al. (2009) Pseudomonas aeruginosa rugose small-colony variants have adaptations that likely promote persistence in the cystic fibrosis lung. J Bacteriol 191: 3492–3503.M. StarkeyJH HickmanL. MaN. ZhangS. De Long2009Pseudomonas aeruginosa rugose small-colony variants have adaptations that likely promote persistence in the cystic fibrosis lung.J Bacteriol19134923503
- 75. Bjarnason J, Southward CM, Surette MG (2003) Genomic profiling of iron-responsive genes in Salmonella enterica serovar Typhimurium by high-throughput screening of a random promoter library. J Bacteriol 185: 4973–4982.J. BjarnasonCM SouthwardMG Surette2003Genomic profiling of iron-responsive genes in Salmonella enterica serovar Typhimurium by high-throughput screening of a random promoter library.J Bacteriol18549734982
- 76. 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.KA DatsenkoBL Wanner2000One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products.Proc Natl Acad Sci U S A9766406645
- 77. Maloy SR, Stewart VJ, Taylor RK (1996) Genetic analysis of pathogenic bacteria: a laboratory manual. Plainview, NY: Cold Spring Harbor Laboratory Press. pp. 466–468.SR MaloyVJ StewartRK Taylor1996Genetic analysis of pathogenic bacteria: a laboratory manualPlainview, NYCold Spring Harbor Laboratory Press466468
- 78. Mercado-Lubo R, Gauger EJ, Leatham MP, Conway T, Cohen PS (2008) A Salmonella enterica serovar typhimurium succinate dehydrogenase/fumarate reductase double mutant is avirulent and immunogenic in BALB/c mice. Infect Immun 76: 1128–1134.R. Mercado-LuboEJ GaugerMP LeathamT. ConwayPS Cohen2008A Salmonella enterica serovar typhimurium succinate dehydrogenase/fumarate reductase double mutant is avirulent and immunogenic in BALB/c mice.Infect Immun7611281134
- 79. Govons S, Vinopal R, Ingraham J, Preiss J (1969) Isolation of mutants of Escherichia coli B altered in their ability to synthesize glycogen. J Bacteriol 97: 970–972.S. GovonsR. VinopalJ. IngrahamJ. Preiss1969Isolation of mutants of Escherichia coli B altered in their ability to synthesize glycogen.J Bacteriol97970972
- 80. Maharjan RP, Ferenci T (2003) Global metabolite analysis: the influence of extraction methodology on metabolome profiles of Escherichia coli. Anal Biochem 313: 145–154.RP MaharjanT. Ferenci2003Global metabolite analysis: the influence of extraction methodology on metabolome profiles of Escherichia coli.Anal Biochem313145154
- 81. Schleucher J, Schwendinger M, Sattler M, Schmidt P, Schedletzky O, et al. (1994) A general enhancement scheme in heteronuclear multidimensional NMR employing pulsed field gradients. J Biomol NMR 4: 301–306.J. SchleucherM. SchwendingerM. SattlerP. SchmidtO. Schedletzky1994A general enhancement scheme in heteronuclear multidimensional NMR employing pulsed field gradients.J Biomol NMR4301306
- 82. Hwang TL, Shaka AJ (1995) Water suppression that works. Excitation sculping using arbitrary waveforms and pulsed field gradients. J Magn Reson Series A 112: 275–279.TL HwangAJ Shaka1995Water suppression that works. Excitation sculping using arbitrary waveforms and pulsed field gradients.J Magn Reson Series A112275279
- 83. Brown PK, Dozois CM, Nickerson CA, Zuppardo A, Terlonge J, et al. (2001) MlrA, a novel regulator of curli (AgF) and extracellular matrix synthesis by Escherichia coli and Salmonella enterica serovar Typhimurium. Mol Microbiol 41: 349–363.PK BrownCM DozoisCA NickersonA. ZuppardoJ. Terlonge2001MlrA, a novel regulator of curli (AgF) and extracellular matrix synthesis by Escherichia coli and Salmonella enterica serovar Typhimurium.Mol Microbiol41349363
- 84. Latasa C, Roux A, Toledo-Arana A, Ghigo J, Gamazo C, et al. (2005) BapA, a large secreted protein required for biofilm formation and host colonization of Salmonella enterica serovar Enteritidis. Mol Microbiol 58: 1522–1539.C. LatasaA. RouxA. Toledo-AranaJ. GhigoC. Gamazo2005BapA, a large secreted protein required for biofilm formation and host colonization of Salmonella enterica serovar Enteritidis.Mol Microbiol5815221539