SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Enterohaemorrhagic Escherichia coli (EHEC) are highly infectious pathogens capable of causing severe diarrhoeal illnesses. As a critical step during their colonization, EHEC adhere intimately to intestinal epithelial cells and generate F-actin ‘pedestal’ structures that elevate them above surrounding cell surfaces. Intimate adhesion and pedestal formation result from delivery of the EHEC type III secretion system (TTSS) effector proteins Tir and EspFU into the host cell and expression of the bacterial outer membrane adhesin, intimin. To investigate a role for DNA methylation during the regulation of adhesion and pedestal formation in EHEC, we deleted the dam (DNA adenine methyltransferase) gene from EHEC O157:H7 and demonstrate that this mutation results in increased interactions with cultured host cells. EHECΔdam exhibits dramatically elevated levels of adherence and pedestal formation when compared with wild-type EHEC, and expresses significantly higher protein levels of intimin, Tir and EspFU. Analyses of GFP fusions, Northern blotting, reverse transcription polymerase chain reaction, and microarray experiments indicate that the abundance of Tir in the dam mutant is not due to increased transcription levels, raising the possibility that Dam methylation can indirectly control protein expression by a post-transcriptional mechanism. In contrast to other dam-deficient pathogens, EHECΔdam is capable of robust intestinal colonization of experimentally infected animals.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Enterohaemorrhagic Escherichia coli (EHEC) are among the leading causes of food- and water-borne illnesses affecting humans in the US, Europe and Japan (for review, see Donnenberg and Whittam, 2001; Kaper et al., 2004; Spears et al., 2006). These bacteria are highly infectious and ingestion of only 100–200 organisms is sufficient to trigger debilitating diarrhoeal disease. Although most EHEC infections resolve spontaneously after 5–10 days of abdominal cramping and bloody diarrhoea, approximately 2–7% of cases progress to the potentially–fatal haemolytic uremic syndrome due, in part, to the production of cytotoxic Shiga toxins which are capable of promoting kidney failure.

As a critical step during their colonization and pathogenesis, EHEC strains produce ‘attaching and effacing’ (AE) lesions on the intestinal epithelial cell surface (for review, see Dean et al., 2005; Garmendia et al., 2005; Hayward et al., 2006). These lesions are marked by destruction of brush border microvilli and intimate bacterial adhesion to enterocytes. Localized cellular accumulation of actin filaments (F-actin) beneath these adherent bacteria generates ‘pedestals’ that can elevate them above the surrounding cell surfaces.

Both intimate adhesion and actin pedestal formation result from the transfer of E. coli secreted proteins (Esps) into the host cell where they interact with mammalian signalling molecules that control actin assembly. Transfer occurs via the bacterial type III secretion system (TTSS), which is encoded by a chromosomal pathogenicity island called the locus of enterocyte effacement (LEE) (for review, see Dean et al., 2005; Garmendia et al., 2005; Hayward et al., 2006). One of the Esps encoded within the LEE is Tir (translocated intimin receptor) (Kenny et al., 1997; Deibel et al., 1998; DeVinney et al., 1999), which is delivered, with the help of its chaperone CesT, into the plasma membrane of target cells where it adopts a hairpin-loop conformation and serves as a receptor for intimin, an adhesin encoded by the LEE eae gene (Jerse et al., 1990). Intimin binds to the central extracellular domain of Tir to trigger clustering of its N- and C-terminal cytoplasmic regions and subsequent actin assembly beneath the plasma membrane (Rosenshine et al., 1996). A non-LEE-encoded effector, EspFU (also called TccP), associates with Tir and is essential for F-actin assembly because it promotes recruitment of the host actin-nucleating machinery to sites of bacterial adherence (Campellone et al., 2004; Garmendia et al., 2004).

Regulation of effector translocation by the EHEC TTSS, and thus intimate adhesion and actin pedestal formation, are highly co-ordinated processes subjected to extensive transcriptional regulation (reviewed in Sperandio et al., 2003; Spears et al., 2006). The LEE activator and repressor proteins, GrlA and GrlR, respectively, control expression of the critical LEE-encoded regulator, Ler (Mellies et al., 1999; Deng et al., 2004). Ler is capable of activating the majority of operons found within the LEE (Friedberg et al., 1999; Elliott et al., 2000), but may do so in only subpopulations of bacteria (Roe et al., 2004). Expression of Ler, and therefore other LEE genes, are significantly influenced by non-LEE-encoded factors that respond to external stimuli, including global regulatory systems involving quorum sensing and the DNA organizing protein, H-NS (Sperandio et al., 1999; Kanamaru et al., 2000; Bustamante et al., 2001).

An additional mechanism by which EHEC virulence gene expression could be controlled is via DNA methylation. In non-pathogenic E. coli K-12 strains, the dam gene encodes a DNA adenine methyltransferase (Dam) with multiple functions during many cellular processes including chromosome replication, DNA mismatch repair, stimulation of the SOS response, and transcriptional regulation (Lobner-Olesen et al., 2005; Wion and Casadesus, 2006). Indeed, loss of Dam in E. coli K-12 alters the level of expression of many genes compared with wild type (Oshima et al., 2002; Lobner-Olesen et al., 2003; Robbins-Manke et al., 2005).

Dam has also been shown to play necessary roles in the pathogenesis of a number of bacterial species. For example, in Salmonella enterica Serovar Typhiumurium, Dam can control the expression of multiple virulence factors, including some TTSS effectors (the Sips) (Garcia-del Portillo et al., 1999; Heithoff et al., 1999). Thus, a S. enterica Serovar Typhiumurium dam mutant is severely defective at cell invasion, avirulent in a murine infection model, and can serve as an effective live vaccine (Garcia-del Portillo et al., 1999; Heithoff et al., 1999). Another enteric pathogen, Shigella flexneri exhibits partial defects in intracellular growth and cell-to-cell spread following dam inactivation (Honma et al., 2004). The dam gene is essential for the viability of Vibrio cholerae and some strains of Yersinia pseudotuberculosis, where overexpression of Dam leads to dysregulated secretion of type III effectors (the Yops) and attenuation of virulence (Julio et al., 2001). Lastly, Dam directs phase variation in uropathogenic E. coli, because methylation of regulatory promoter sequences controls the binding of different proteins that modulate transcription of the gene encoding the adhesive Pap pilus (Hernday et al., 2003).

The most common disease-causing EHEC serotype is E. coli O157:H7 and the genome of strain EDL933 has been completely sequenced. Not surprisingly, it contains a dam gene that is 99% identical to dam from E. coli K-12 (Perna et al., 2001). However, unlike K-12, which only possesses the archetypical dam gene, EDL933 harbours several prophages that are predicted to encode other putative Dam-like methyltransferases of unknown function. We asked if the chromosomal prototype dam gene is responsible for Dam methylation in EHEC, and because intimate adhesion and pedestal formation are critical for pathogenesis and are controlled by intimin, the TTSS, Tir and EspFU, whether Dam methylation affects host cell interactions by regulating the expression of these virulence factors. We show that deletion of the prototype dam gene of EHEC O157:H7 results in a loss of Dam methylation and a dramatic increase in adherence and actin pedestal formation on cultured human cells compared with wild-type EHEC. Increases in adherence and pedestal formation in vitro correlate with elevated protein levels of intimin, Tir and EspFU. Finally, consistent with its capacity for vigorous interaction with mammalian cells in vitro and unlike dam mutants of several other enteric pathogens, the dam mutant of EHEC O157:H7 is capable of robust colonization of the intestines of infected animals.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Deletion of the archetypical dam gene of EHEC O157:H7 results in a mutant that exhibits DNA methylation patterns, growth characteristics and mutator phenotypes similar to those of an E. coli K-12 dam mutant

To determine whether the prototype dam gene in EHEC encoded a functional methyltransferase that was largely responsible for Dam methylation in this bacterium, Lambda Red-mediated recombination (Murphy and Campellone, 2003) was used to construct EHECΔdam, a dam-deleted mutant of TUV93-0 (which is referred to as ‘wild-type EHEC’ hereafter), a derivative of EHEC EDL933 that is incapable of producing Shiga toxin. To assess the DNA methylation state of wild-type EHEC and EHECΔdam, these strains were transformed with pBR322, grown in Luria–Bertani (LB) media, and their DNA subjected to restriction enzyme analyses using Sau3A, DpnI and DpnII. Sau3A digests DNA independent of its methylation state, whereas DpnI and DpnII digest methylated and unmethylated DNA respectively. In Fig. 1A, DNA fragments migrating faster than 4.4 kb are those of pBR322 and those larger than this size are derived from chromosomal DNA. As expected, treatment of DNA from wild-type EHEC with Sau3A and DpnI resulted in identical DNA digestion patterns, whereas DpnII was without effect (Fig. 1A, lanes 1–3). Conversely, the Sau3A and DpnII digestion patterns were the same for EHECΔdam, but the DNA derived from this strain was resistant to DpnI digestion (Fig. 1A, lanes 4–6). Thus, the archetypical Dam protein is required for DNA methylation in EHEC under these common culturing conditions.

image

Figure 1. An EHEC O157:H7 dam mutant displays minor growth defects and an elevated mutation frequency. A. Bacterial DNA from EHEC/pBR322 and EHECΔdam/pBR322 was treated with Sau3A to digest methylated and non-methylated DNA (lanes 1,4), DpnI to digest methylated DNA (lanes 2,5) or DpnII to digest non-methylated DNA (lanes 3,6), and resolved on a 1% agarose gel. A DNA marker is shown to the left. B. EHEC cultures grown to saturation in LB (EHEC and EHECΔdam) or LB plus ampicillin (EHECΔdam/pBR322 and EHECΔdam/pBRdam) were diluted ∼1000-fold into fresh medium and grown with agitation at 37°C. OD600 values were measured at the indicated times. C. EHEC cultures grown as in B were titred on LB agar plates. D. EHEC cultures grown in LB or LB + ampicillin were fixed onto glass coverslips, stained with fluorescent anti-O157 antibodies, and examined microscopically. Examples of elongated EHEC bacilli are indicated with arrows. E. Equivalent levels of protein extracts from wild-type EHEC and EHECΔdam (see OmpA blotting in Fig. 5A) were separated by SDS-PAGE and blotted with an antibody to RecA. F. EHEC cultures grown to saturation were spread on LB agar plates to determine bacterial titres and on LB plus rifampicin plates to identify spontaneously resistant mutants. Data are the means ± SD of seven experiments.

Download figure to PowerPoint

In E. coli K-12, dam mutants display subtle growth defects, exhibit aberrant cellular morphology, have increased expression levels of SOS genes, and possess elevated mutation frequencies (Marinus, 1996; Lobner-Olesen et al., 2005; Wion and Casadesus, 2006). As observed for E. coli K-12, the optical densities of wild-type EHEC and EHECΔdam broth cultures were indistinguishable over time (Fig. 1B). These patterns were unaffected by the presence of a complementing Dam plasmid, pBRdam or a pBR322 vector control (Fig. 1B). However, when growth was measured by viable cell counts (colony-forming units), the dam mutant showed lower values than wild type, a phenotype that could be complemented by the presence of plasmid-encoded dam, but not pBR322 (Fig. 1C). This apparent discrepancy between dam-proficient and dam-deficient strains is explained by an increased tendency of EHECΔdam to form elongated bacilli (Fig. 1D) and/or interconnected filaments as has been observed for E. coli K-12 (Marinus, 1996). Such changes in bacterial morphology likely result from induction of the SOS regulon (Marinus, 1996). Consistent with SOS induction, immunoblotting for the RecA protein, a known component of the SOS response, indicated that it is expressed at higher levels in EHECΔdam than in wild-type EHEC (Fig. 1E). Additional microarray analyses indicated that transcription of recN, a member of the SOS regulon, was increased 2.9-fold in EHECΔdam relative to wild-type EHEC (KM, ALO and MM, manuscript in preparation). As expected, the elongated morphology of EHECΔdam could be suppressed by the dam plasmid (Fig. 1D), indicating that Dam expression can rescue the cell division defect.

Analogous to E. coli K-12, EHECΔdam also displayed an increase in spontaneous mutation rate, as determined by a more than 12-fold elevation in frequency of resistance to the RNA polymerase inhibitor, rifampicin (Fig. 1F). This mutator phenotype was also capable of being suppressed by the pBRdam plasmid (Fig. 1F). Thus, the collective growth-related phenotypes associated with deletion of dam in EHEC are comparable to those that result from inactivation of the dam gene in E. coli K-12.

An EHEC dam mutant adheres to cultured mammalian cells more rapidly than wild-type EHEC

To examine whether Dam regulates the ability of EHEC to interact with mammalian host cells, we aimed to next measure the efficiencies with which dam-proficient and dam-deficient EHEC strains could adhere to cultured epithelial cells. However, it was first necessary to determine whether EHEC dam mutants could replicate normally in conditions that have been established for optimal infection of host cells in vitro (serum-containing mammalian cell culture medium in an atmosphere containing 5% CO2). Therefore, EHEC strains grown to saturation in LB media were diluted into this infection medium and cultivated in a manner identical to that typically used during infection of cultured mammalian cells. In fact, EHECΔdam and EHECΔdam/pBR322 were capable of multiplying in this cell culture medium, but at rates that were slightly slower than wild-type EHEC or EHECΔdam/pBRdam+ when measured by optical density after 4–5 h of growth (Fig. 2A) and when comparing viable counts after 3–5 h of growth (Fig. 2B). Nevertheless, these levels were sufficient to allow an investigation of the proficiency of cell binding.

image

Figure 2. An EHEC O157:H7 dam mutant adheres to cultured mammalian cells more rapidly than wild-type EHEC. A. EHEC cultures grown to saturation in LB or LB plus ampicillin were diluted ∼200-fold into antibiotic-free cell culture medium and grown under conditions typically used during infection of mammalian cells. OD600 values were measured at the indicated times. B. EHEC cultures grown as in A were tittered on LB agar plates. C. EHEC strains grown in LB or LB plus ampicillin were diluted into antibiotic-free cell culture medium and used to infect HeLa cells. Infected samples were treated with anti-O157 antibodies to identify bacteria and phalloidin to visualize mammalian cells. Numbers of cell-associated bacteria were measured at the depicted time points. Data are the means ± SD of three experiments.

Download figure to PowerPoint

To compare the efficiency of binding by dam-proficient and dam-deficient strains, cultured HeLa cells were infected with EHEC for various times, fixed, and processed for microscopic examination. Bacteria were identified by staining with anti-O157 antibodies and the number of adherent bacteria per mammalian cell was quantified by immunoflourescence microscopy. Notably, EHECΔdam exhibited two- to threefold higher levels of cell binding than wild-type EHEC after 1 h, 2 h and 3 h of infection (Fig. 2C). This elevated cell binding by EHECΔdam was reduced to wild-type levels at the 1 h and 2 h time points by the complementing plasmid, pBRdam, but not by the pBR322 vector alone (Fig. 2C). Adherence by Dam-deficient bacteria was indistinguishable from that of Dam-proficient bacteria at 4 h post infection (Fig. 2C). These analyses indicate that Dam negatively regulates the initial association between EHEC and its host cells, because strains lacking this methyltransferase have enhanced cell-binding properties at early time points during infection in vitro.

An EHEC dam mutant generates actin pedestals more frequently than wild-type EHEC

To next examine whether Dam regulates the ability of EHEC to trigger actin pedestal formation within cultured cells, EHEC strains were grown in LB and used to infect HeLa cells for various times. Infected cells were treated with phalloidin to stain F-actin, and the number of actin pedestals per mammalian cell was quantified microscopically. Remarkably, after 3 h of infection, EHECΔdam generated greater than 60-fold more pedestals than wild-type EHEC (Fig. 3A). Similarly, after 4 h and 5 h of infection, EHECΔdam formed approximately 20- and 13-fold more pedestals than wild type (Fig. 3A). Unexpectedly, the pBR322 vector caused a significant suppression of actin pedestal formation by EHECΔdam (Fig. 3A). Nevertheless, the presence of the dam gene within pBR322 reduced pedestal formation even further, to levels similar to wild type (Fig. 3A), indicating that dam provided in trans can complement EHECΔdam for pedestal formation.

image

Figure 3. An EHEC O157:H7 dam mutant generates actin pedestals more frequently than wild-type EHEC. A. EHEC strains grown in LB or LB plus ampicillin were diluted into antibiotic-free cell culture medium and used to infect HeLa cells. Infected samples were treated with anti-O157 antibodies to identify bacteria and phalloidin to visualize F-actin. Frequencies of actin pedestal formation were measured at the depicted times. Data are the means ± SD of three experiments. B. EHEC strains were grown under pre-inducing conditions, diluted into antibiotic-free cell culture medium, and used to infect HeLa cells for 5 h. Infected samples were treated with anti-O157 antibodies to identify bacteria and phalloidin to visualize F-actin. C. Pedestal formation frequencies were measured for the infections described in B. Data are the mean of two experiments. Similar results were observed in at least four other experiments.

Download figure to PowerPoint

While these results demonstrate that EHECΔdam grown in LB can subsequently generate actin pedestals on mammalian cells in a rapid and efficient manner, cultivation in serum-free cell culture medium, instead of LB, has been previously shown to ‘pre-induce’ the ability of EHEC to signal to host cells (Beltrametti et al., 1999). The abilities of wild type and dam-deficient EHEC to stimulate actin assembly were therefore also compared after growth under these conditions. Similar to results obtained when bacteria were precultured in LB, EHECΔdam and EHECΔdam/pBR322 (Fig. 3B, bottom panels) triggered actin polymerization more frequently than wild-type EHEC and EHECΔdam/pBRdam+ (Fig. 3B, top panels) after 5 h of infection. Quantification of pedestal formation by wild-type EHEC and EHECΔdam revealed that, as expected, the overall frequency of pedestal formation by both strains was elevated dramatically when grown under pre-inducing conditions relative to bacteria grown in LB (compare Fig. 3C with 3A, right panel). Nevertheless, dam-deficient EHEC was still nearly five times more effective at generating pedestals than its parental strain (Fig. 3C).

To determine, at a higher resolution, whether these actin pedestals formed by EHECΔdam were morphologically similar to AE lesions normally generated by wild-type EHEC, infected HeLa cells were examined by electron microscopy. Scanning electron micrographs indicated that bound EHEC and EHECΔdam were both associated with filopodia on the surfaces of HeLa cells (Fig. 4A and B, left panels). The presence of these finger-like projections is likely a consequence of the translocated effector protein Map (Kenny, 2002). In addition, transmission electron microscopy demonstrated that wild-type EHEC and EHECΔdam adhered intimately to HeLa cells, as bacteria appeared very tightly opposed to the plasma membrane (Fig. 4A and B, right panels). Consistent with an accumulation of F-actin beneath sites of bacterial attachment, these cellular regions often contained an enrichment of electron-dense material (Fig. 4A and B, right panels). Thus, there are no discernible morphological differences between the pedestals generated by wild type and dam-deficient EHEC.

image

Figure 4. Pedestals generated by an EHEC O157:H7 dam mutant are morphologically indistinguishable from pedestals formed by wild-type EHEC. HeLa cells were infected with (A) wild-type EHEC or (B) EHECΔdam under conditions described in Fig. 3B and subjected to scanning electron microscopy (left panels) or transmission electron microscopy (middle and right panels).

Download figure to PowerPoint

Dam-deficient EHEC O157:H7 strains express elevated levels of proteins required for intimate adherence and actin pedestal formation

In E. coli K-12, Dam deficiency correlates with alterations in levels of gene expression which can either increase or decrease (Marinus, 1987; Lobner-Olesen et al., 2005; Wion and Casadesus, 2006). To establish a molecular basis for the rapid kinetics and elevated frequencies of adherence and actin pedestal formation by EHECΔdam in vitro, bacteria grown in LB were collected and examined by immunoblotting for several known or putative virulence factors: two surface adhesins (intimin and OmpA), a component of the type III translocon (EspD), and the two effectors known to be directly required for pedestal formation (Tir and EspFU). The levels of OmpA, a protein which appears to be involved in non-intimate adherence but not pedestal formation (Torres and Kaper, 2003) were equivalent in both dam-proficient and dam-deficient EHEC strains (Fig. 5A), suggesting that this adhesin is not involved in the increased cell binding by EHECΔdam. In contrast, each of the other EHEC proteins, which participate in different processes leading to intimate adherence and pedestal formation, were upregulated in EHECΔdam compared with wild-type EHEC (Fig. 5A). Intimin, a LEE-encoded outer membrane protein involved in initial and intimate adherence (Garmendia et al., 2005), was found in significantly greater quantities in dam-deficient strains (lanes 2–3) than in dam-proficient strains (lanes 1,4). EspD, a LEE-derived protein required for forming a pore in the host membrane to promote delivery of other effector proteins (Dean et al., 2005), was expressed at slightly higher levels in EHECΔdam (lane 2) than wild-type EHEC (lane 1). Tir, the LEE-encoded receptor for intimin (Kenny et al., 1997; Deibel et al., 1998), was also overexpressed in dam-deficient strains (lanes 2–3) compared with dam-proficient strains (lanes 1,4). Lastly, EspFU, an effector that links Tir to the actin assembly machinery and is encoded on a pathogenicity island distinct from the LEE (Campellone et al., 2004; Garmendia et al., 2004), was produced at higher levels in EHECΔdam (lane 6) than in wild type (lane 5). Overexpression of intimin, EspD, Tir and EspFU by EHECΔdam, relative to wild-type EHEC, was also observed when these bacteria were grown under pre-inducing conditions (Fig. 5B). Collectively, these results suggest that in the absence of Dam methylation, EHEC exhibits an upregulation of both LEE-encoded and non-LEE-derived proteins which stimulate intimate adherence and actin assembly.

image

Figure 5. Dam-deficient EHEC O157:H7 strains express elevated levels of proteins required for intimate adherence and actin pedestal formation. A. EHEC cultures grown in LB or LB plus ampicillin were collected by centrifugation and lysed in SDS-PAGE sample buffer. Protein samples derived from 0.1 OD units of each culture were separated by SDS-PAGE and examined by immunoblotting with antibodies to intimin, EspD, Tir, OmpA and myc-tagged EspFU. B. EHEC cultures grown under pre-inducing conditions were processed as in A.

Download figure to PowerPoint

Overexpression of Tir is not associated with discernibly increased levels of transcription

To determine whether high levels of Tir protein correlate with increased levels of tir transcription, dam-proficient and dam-deficient bacteria were transformed with a tir-gfp plasmid. This fusion consists of the tir promoter region, the Shine-Dalgarno sequence and the first few tir codons fused to the coding sequence of GFP (Roe et al., 2004). As a control, an EHEC strain harbouring a plasmid encoding a GFP-fusion to the promoter for the ribosomal subunit gene rpsM was also analysed. Upon growth of the strains in LB, the GFP fluorescence over time of dam-deficient strains harbouring rpsM::gfp, or tir::gfp were both somewhat lower than the corresponding wild-type strains (Fig. 6A), raising the possibility that transcription is globally reduced in the absence of Dam methylation. To assess the frequency of GFP expression by individual bacteria in these populations, culture samples were also examined for fluorescence microscopically. Each set of bacteria harbouring the rpsM::gfp reporter exhibited a relatively homogeneous expression pattern (Fig. 6B, top panels). In contrast, just a fraction of wild type and EHECΔdam bacteria fluoresced brightly (Fig. 6B, bottom panels), similar to previous observations that tir expression only occurs in a proportion of cultured bacteria (Roe et al., 2004). Overall, these results indicate that in EHECΔdam, tir is not differentially transcribed relative to a ribosomal protein, suggesting that the striking increase in the level of Tir protein may not be due to a corresponding increase in transcription.

image

Figure 6. Overexpression of tir is not due to increased levels of transcription. A. EHEC strains harbouring reporter plasmids containing the GFP gene downstream of the rpsM or tir promoters were grown in LB containing the appropriate antibiotics. Culture densities were estimated by measuring OD600, while GFP expression levels were quantified fluorometrically. Data are the means of triplicate experiments. B. EHEC samples derived from experiments described in A were fixed onto glass coverslips and examined microscopically for GFP fluorescence. C. RNA samples from EHEC cultures grown in LB were subjected to Northern analyses using DNA probes for tir or pgi. Arrows indicate positions of transcripts at the depicted sizes.

Download figure to PowerPoint

To assess transcription from the endogenous tir promoter located on the chromosome, rather than a plasmid-encoded copy, bacterial RNA was collected from wild-type EHEC and EHECΔdam and examined by Northern blotting using probes directed against tir and a control gene, pgi (phosphoglucose isomerase) (Fig. 6C). While the strongest tir hybridization signal was observed at 1.6 kb for each strain, additional bands consistent with a polycistronic message encoding tir, cesT (the Tir chaparone) and eae (intimin) were also observed. Indeed, the sizes of the approximately 5.2 kb, 2.2 kb and 1.6 kb RNA species correspond with the predicted tir-cesT-eae, tir-cesT and tir transcripts, respectively, although the 5.2 kb and 2.2 kb species were not characterized further. None of the tir-hybridizing transcripts appeared more abundant in the dam mutant compared with wild type, whereas the control pgi transcript from the same preparation of RNA appeared slightly more abundant in the mutant background, indicating that the low tir-specific signal in the dam mutant was not due to general degradation of mRNA in that preparation.

To assess the abundance of tir transcripts, along with those of other LEE genes like espD, eae and cesT when wild type and dam-deficient EHEC strains were grown under the same conditions used for measuring actin pedestal formation, transcription was measured globally using a DNA microarray. Consistent with GFP-fusion and Northern blotting data, these experiments revealed no significant differences (less than twofold) between wild-type EHEC and EHECΔdam in transcript levels of tir, espD, eae or cesT (KM, ALO and MGM, manuscript in preparation). The relative abundance of tir transcript in EHECΔdam versus wild type was 1.5:1. To confirm that tir transcription was not discernibly increased in EHECΔdam, RNA isolated from wild-type EHEC and EHECΔdam was also examined by quantitative reverse transcription polymerase chain reaction (RT-PCR). These experiments indicated that tir transcript levels were not significantly elevated relative to the control gene gapA (glyceraldehyde-3-phosphate dehydrogenase). The abundance of tir transcript in EHECΔdam versus wild type was 1.7:1. Overall, the observations that substantial increases in tir transcription could not be detected in the dam mutant in any of the above assays suggests that the ability of EHECΔdam to overexpress Tir, and possibly other factors like intimin, is not likely due to an increase in transcription.

EHECΔdam is capable of colonizing the intestines of gnotobiotic piglets

Dam-deficient mutants of some enteric pathogens, like S. enterica Serovar Typhiumurium, are avirulent in animal infection models and can serve as effective live vaccines (Low et al., 2001; Casadesus and Low, 2006). To assess the pathogenicity of the EHEC dam mutant, an established gnotobiotic piglet infection model in which wild-type EHEC normally colonizes the large intestine was utilized (Tzipori et al., 1992; Donohue-Rolfe et al., 2000). Bacteria were administered orally, the animals were sacrificed after 72 h, and tissues from the small and large intestine were prepared for histology and scored for colonization as described previously (Tzipori et al., 1992; Donohue-Rolfe et al., 2000). While wild-type EHEC colonized the entire piglet intestine, a control strain lacking intimin (EHECΔeae) did not colonize at all (Table 1). With the exception the proximal portions of the small intestine, EHECΔdam was observed to colonize the entire intestine, and was closely associated with the epithelium from the ileum through the spiral colon. Indeed, in the caecum, a primary site of wild-type EHEC colonization, the dam mutant colonized in a robust manner indistinguishable from wild type. Thus, unlike some other enteric pathogens, and in spite of its previously observed defects in growth in vitro (Fig. 1), EHECΔdam was capable of extensive intestinal colonization.

Table 1.  An EHECΔdam mutant colonizes the intestines of gnotobiotic piglets.
EHEC strainSmall intestineLarge intestine
DuodenumJejunumIleum
ProximalDistalCaecumSpiral colon
  1. Gnotobiotic piglets were infected orally with 5 × 109 EHEC, EHECΔdam or EHECΔeae. Animals were sacrificed 72 h after infection and the indicated tissues were removed, fixed, stained histologically and scored for colonization on a scale of 0–4. A score of 0 indicates the absence of bacteria, 1 denotes free bacteria in the lumen, 2 shows the appearance of bacteria associated with enterocytes, 3 and 4 equal greater degrees of enterocyte association (adapted from Tzipori et al., 1992). Values represent averaged colonization scores from at least three piglets.

  2. WT, wild type.

WT122333
Δdam001232
Δeae000000

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

In this study, we first identified the archetypical dam gene as the primary source of adenine methylation in EHEC, because inactivation of this gene resulted in a complete absence of detectable adenine methylation. Yet, when placed under control of an inducible promoter, the EHEC VT2-Sa prophage-derived dam homologue could be expressed and methylate GATC sequences (Radlinska and Bujnicki, 2001), indicating that such prophage-derived Dam proteins may be capable of methylating DNA. However, our results suggest that prophage-encoded EHEC dam genes are not expressed during standard bacterial culturing conditions, and thus we could not establish a biological function for any dam genes other than the archetype.

The alteration of SOS gene expression and increases in filamentation and spontaneous mutagenesis observed for the dam-deficient EHEC strains are entirely analogous to phenotypes of dam-deficient E. coli K-12 (Fig. 1), suggesting, not surprisingly, that adenine methylation controls similar cellular functions in these related bacteria. However, given that adenine methylation is a requirement for virulence in many enteric pathogens, an unexpected result was that the EHEC dam mutant demonstrated increased adherence to cultured mammalian cells compared with wild-type EHEC (Fig. 2). Higher levels of adherence were manifested most prominently in the first few hours after encountering cultured cells. This phenomenon likely reflects differences in gene expression that either precede and/or occur rapidly after these bacteria are introduced into mammalian cell culture medium. Enhanced levels of intimin, EspD, Tir and EspFU are readily apparent in cultured EHECΔdam prior to infection (Fig. 5), demonstrating that these bacteria are already equipped for efficient infection. After 4 h of infection, however, no differences in adherence were observed between the wild type and dam-deficient strains. Because several adhesive mechanisms are induced by pathogenic E. coli after incubation with mammalian cells (Giron et al., 2002), expression of such putative pathways at later time points could explain the restriction of measurable differences in binding between the mutant and wild-type strains to early time points.

The relative increase in mammalian cell interaction by the EHECΔdam strain was also reflected by a striking difference in actin pedestal formation. When grown under ‘non-inducing’ conditions (in LB), the dam-deficient mutant generated pedestals as much as 60-fold more efficiently than wild type, and even under ‘inducing’ growth conditions that are known to dramatically increase pedestal formation by wild-type EHEC, the mutant was almost fivefold better at pedestal formation (Fig. 3). The higher levels of EspD, Tir, EspFU and intimin that we observed under both of these conditions likely contribute to this enhanced signalling capacity of EHECΔdam. Surprisingly, reporter fusion, Northern blotting, microarray and RT-PCR analysis of tir revealed no significant difference in the transcription levels of EHECΔdam versus wild-type bacteria, suggesting that alteration of transcription initiation frequency or message stability are not the primary cause of the increased production of this protein. At least two mechanisms of post-transcriptional regulation mediated by Dam methylation have been described for other bacteria (Bell and Cupples, 2001; Camacho and Casadesus, 2005), and further investigation of the regulation of espD, tir, eae, espFU and other genes that promote host cell interaction will be required to confirm and characterize the post-transcriptional nature of regulation and its potential mechanisms.

Among enteric pathogens harbouring TTSSs, such as Shigella, Yersinia and Salmonella, improper regulation of gene expression following disruption of Dam methyltransferase activity results in a decrease in levels of entry into cultured target cells by these intracellular pathogens, and a concomitant modest to very dramatic decreases in virulence, raising the possibility that such strains might serve as good vaccine candidates or that the Dam protein may be a useful therapeutic target (Low et al., 2001; Casadesus and Low, 2006). In contrast, our data indicate that in spite of some in vitro growth defects, dam inactivation in EHEC results in enhanced interactions with cultured cells, likely due in part to the constitutively high expression levels of intimin, EspD, Tir and EspFU. Consistent with this, EHECΔdam is also capable of extensive colonization of gnotobiotic piglets at levels equal to or only slightly less than wild-type EHEC (Table 1). The finding that epithelial colonization of the small bowel and spiral colon by EHECΔdam was slightly less extensive than by wild type (in contrast to its enhanced relative level of interaction with cultured cells) is likely a result of a dramatic increase in the levels of intimin, Tir and EspFU, and type III translocation by wild-type EHEC upon infection of the mammalian host (MB and JL, manuscript in preparation). Regardless, the finding that EHECΔdam interacts extensively with mammalian cells in vitro and appears to retain significant virulence in vivo indicates that therapeutic strategies that target Dam methyltransferase in EHEC, or that utilize EHEC dam mutants as attenuated vaccine strains, are not particularly promising. On the other hand, investigation into how Dam methyltransferase influences virulence gene regulation, perhaps through post-transcriptional means, may shed light on regulatory networks that are integral to the pathogenesis of EHEC and other bacteria.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Strains, plasmids and DNA manipulations

Most EHEC strains used in this study were derived from TUV93-0, a Shiga toxin-deficient form of prototype EDL933. TUV93-0 (Campellone et al., 2002) was generated in the following manner. Plasmid pTUV9 (Donohue-Rolfe et al., 2000), a derivative of suicide vector pGP704 that contains the genes for levansucrase (sacB) and a 1433 bp fragment representing a 500 bp deletion of stx-1, was introduced into strain EDL933 by plate mating with E. coli pTUV9/SM10 (lambda pir) (Miller and Mekalanos, 1988). EDL933 exconjugants were grown on LB plates containing sucrose and sucrose-resistant colonies were screened by ELISA for Stx1 and Stx2 production. Toxin-negative colonies were further analysed by PCR and by Southern blot analysis using EcoRI digested DNA from stx-1 (933J) and stx-2 (933W) converting phages as probes. At some time during this strain construction, the stx2 gene was also inactivated. EHECΔdam (KM67) was created by Lambda Red-mediated recombination in TUV93-0. The upstream region of TUV93-0 dam was generated by PCR using primers ATCATCGAGCTCCTGGTCGTTCTCGTCAATCTTCTG and TGCCGATCAACGTCTCATGCGGCCGCAGCGCGATTTTTCTTCATGCTGAC, while the dam downstream region was generated with primers AATGGCAGAAATTCGAAAGCGGCCGCACCCGCGAAAAAATAATTCTCAAGG and ATCATCGCATGCAATCACCACGCTGAAGTATTTGGC. PCR products were digested either with SacI plus NotI or NotI plus SphI. Both fragments were ligated together in a mixture containing the SacI-SphI backbone of pUC19 to generate the Δdam-containing plasmid pKM210. This plasmid contains an in-frame fusion of dam consisting of the first six dam codons, a 9 bp sequence containing a NotI site, and the last four dam codons. The fusion is flanked by 1.3 kb of sequence upstream and 1.2 kb of sequence downstream of dam. A derivative of pKM210, pKM212 (Δdam::kan), was generated by ligating a kanamycin drug resistance cassette into the NotI site. Plasmid digests containing this dam substitution were electroporated into TUV93-0 containing the Red-producing plasmid pTP223 as described previously (Murphy and Campellone, 2003). Primers used to test for the presence/absence of the dam or dam::kan gene by PCR were TCGGTGTTTCTCAACACC and CGCTTGTTGTTCAAGCGT. A chloramphenicol resistant EHECΔdam derivative (used in Fig. 5, right panels) was generated by Lambda Red-mediated recombination by electroporating TUV93-0 with a PCR product containing the cat gene flanked by 50 nucleotides of dam-specific sequence. Wild-type EDL933 and a Δdam mutant of EDL933, KM69 (KM, ALO and MM, manuscript in preparation), were used in microarray experiments. The construction of the EHECΔeae strain has been described elsewhere (Murphy and Campellone, 2003). The Dam complementation plasmid, pBRdam (pMQ148) is a derivative of pBR322, contains the dam gene on a BamHI fragment, and expresses low levels of Dam protein (roughly twofold that of wild type; not enough to induce mutagenesis) (Arraj et al., 1990). The EspFU-myc plasmid and reporter plasmids for measuring transcription of rpsM, and tir have been described previously (Campellone et al., 2004; Roe et al., 2004). EHEC genomic DNA was prepared using the MasterPure DNA Purification kit (Epicentre Biotechnologies) and digested with restriction enzymes (New England Biolabs) by standard methods.

Bacterial and mammalian cell culture

For routine passage, all EHEC strains were grown in LB media (plus appropriate antibiotics) at 37°C. Prior to infections, EHEC were usually grown to saturation in LB (plus 50 ug ml−1 ampicillin to maintain pBR322-based plasmids, 25 ug ml−1 kanamycin to maintain pEspFU-myc, or 15 ug ml−1 chloramphenicol to maintain GFP-reporter plasmids), diluted 500-fold and grown for an additional 15 h. To pre-induce EHEC (e.g. in Figs 3B and C, 4 and 5B), bacteria were diluted 500-fold into DMEM + 100 mM HEPES pH 7.4 and grown in 5% CO2 without agitation for 15 h. To measure EHEC growth rates, bacteria were diluted to 2 × 106 ml−1 in LB (plus appropriate antibiotics) or 107 ml−1 in DMEM + 3% FBS + 25 mM HEPES pH 7.4, and culture optical densities (OD600) and colony-forming units measured hourly. Representative experiments are shown in Figs 1 and 2. To measure spontaneous mutation rates, EHEC cultures grown in LB broth (plus appropriate antibiotics) were spread on LB agar plates or LB agar plates containing 100 ug ml−1 rifampicin and the frequency of RifR mutants per ml of culture was calculated. For RT-PCR and microarray experiments, bacteria grown in DMEM + 100 mM HEPES were diluted 200-fold into DMEM + 3% FBS + 25 mM HEPES pH 7.4 and grown in 5% CO2 without agitation for 5 h, conditions used for optimal infection of cultured cells. HeLa cell cultures were maintained in DMEM + 10% FBS at 37°C in 5% CO2. To measure the interactions between mammalian cells and EHEC, HeLa cells grown to 50–90% confluency on 12 mm glass coverslips were infected with 107 bacteria in DMEM + 3% FBS + 25 mM HEPES pH 7.4 for up to 5 h as described previously (Campellone et al., 2002).

Immunofluorescence microscopy

Bacteria and HeLa cells were fixed onto glass coverslips by treatment with PBS + 2.5% paraformaldehyde for 20–30 min as described previously (Campellone et al., 2002). Prior to staining, mammalian cell membranes were permeabilized with 0.1% TritonX-100. Bacteria were detected by treatment with rabbit anti0157 antibodies (diluted 1:500 in PBS + 1%BSA) for 30 min prior to washing and addition of Alexa488 goat anti-rabbit antibodies (1:200). F-actin was visualized by staining with Alexa568-phalloidin (1:100).

Immunoblotting

To prepare EHEC lysates, bacteria cultured in LB were centrifuged, washed once with PBS, and resuspended in SDS-PAGE sample buffer (Campellone et al., 2002). Protein samples were boiled for 10 min, centrifuged, and sample amounts corresponding to 0.1 OD units were separated by 10% SDS-PAGE prior to transferring to PVDF membranes. Membranes were blocked for 30 min in PBS + 5% milk (PBSM) before treatment with rabbit anti-RecA (diluted 1:10 000 in PBSM; a gift from Kendall Knight), sheep anti-intimin (1:1000; a gift from Alison O'Brien), mouse anti-EspD and anti-TirM (1:500; gifts from Trinad Chakraborty), rabbit anti-OmpA (1:10 000; a gift from Carol Kumamoto) or mouse antic-myc (1:300; Sigma) for 2–3 h, similar to previous experiments (Roe et al., 2003; Campellone et al., 2004; Campellone and Leong, 2005). Following washes, membranes were treated with secondary antibodies and developed. Similar results were observed in at least three sets of immunoblots.

Quantification of EHEC adherence and actin pedestal formation

The cell-binding frequencies of EHEC strains were quantified by measuring the number of bacteria (identified by anti-O157-staining) associated with randomly chosen HeLa cells. One hundred and fifty cells were examined after 1 h, 2 h or 3 h of infection and 50 cells were examined after 4 h of infection in each of three separate experiments. Pedestal formation frequencies were quantified by measuring the number of localized F-actin pedestals (identified by intense phalloidin-staining) on randomly chosen HeLa cells. One hundred and fifty cells were examined after 3 h of infection and 50 cells were examined after 4 h or 5 h of infection in each of three separate experiments.

Electron microscopy

Samples of infected HeLa cells in buffered formalin were processed as described previously for electron microscopy (Tzipori et al., 1989).

Transcriptional analyses

Enterohaemorrhagic Escherichia coli strains harbouring the appropriate GFP reporter plasmids (pAJR75 and pAJR145; Roe et al., 2004) were grown to saturation in LB and diluted into fresh LB to a starting OD600 of 0.05. Aliquots were removed at subsequent time intervals and fluorescence levels determined using a Fluostar 96-well plate reader. OD600 readings were taken in parallel, and all samples with an OD higher than 1.0 were diluted to accurately measure cell density. Cultures appeared to still be in exponential phase growth at OD 3.5 when measured in this way. The promoter-less plasmid pAJR70 transformed into EHEC strains acted as a control for fluorescence produced independent of promoter activity. For single cell analyses of GFP expression, 20 μl aliquots were dried on glass slides, washed three times with PBS, and cover slips applied using DAKO fluorescent mounting medium. Northern blotting was performed on EHEC cultures grown to an OD600 of 3.5 in LB, similar to experiments described previously (Roe et al., 2003; Zhang et al., 2004). Oligonucleotide sequences used to create the PCR-generated tir probe were ATGCCTATTGGTAATCTTGGTC and TTAAGAGTATCGAGCGGACC. For microarray experiments, RNA was isolated from EDL933 and its Δdam derivative, KM69, using the MasterPure RNA Purification kit (Epicentre Biotechnologies), processed as recommended in the Affymetrix Expression Analysis Technical Manual (http://www.affymetrix.com), and hybridized to GeneChip®E. coli Genome 2.0 Arrays. These arrays did not have espFU sequences, so expression from this locus could not be monitored. Raw data were processed using the Microarray Suite version 5.0 software (Affymetrix) and exported to Microsoft Excel (Microsoft Corporation). The data files can be accessed at http://users.umassmed.edu/martin.marinus/arrays/. For RT-PCR experiments, RNA isolated using the MasterPure RNA Purification kit was used as substrate for first strand cDNA synthesis with Superscript III RT (Invitrogen) as recommended by the supplier. The sequence of the RT primers were CTGGTGGGTTATTCGAAGTATTC for tir and GAGATGTGAGCGATCAGGTCC for gapA. PCR primers were CTAAACTTTGGTTGGCGTTGG and TCGCAGTTTCAGTTGCACTTG for tir and CGCTGAACGTGATCCGGCTAA and ACCAGCGGTGATGTGTTTACG for gapA. PCR was performed with Quantitect SYBR Green (Qiagen) in an Opticon 2 Continuous Fluorescence Detector (MJ Research) for 40 cycles according to the manufacturer's instructions. Relative gene expression values were determined using the 2–ΔΔCT method (Livak and Schmittgen, 2001). Each experiment was carried out in triplicate and included controls to detect contaminating DNA.

Animal infections

Gnotobiotic piglet infections were performed as described previously (Tzipori et al., 1992). Briefly, animals were derived by cesarean section and maintained in microbiological isolation. Twenty-four h after cesarean delivery, animals were given a single oral challenge of 5 × 109 colony-forming units of EHEC (24 piglets), EHECΔdam (10 piglets) or EHECΔeae (three piglets). Animals were monitored for up to 72 h, sacrificed, and formalin-fixed tissues from the small and large intestine were prepared for histology by haematoxylin and eosin staining and scored for colonization. To confirm the absence of cross-contamination between isolators, gut contents were cultured for bacterial growth and individual colonies tested for antibiotic resistance. In all cases, bacteria harvested from infected piglets displayed the appropriate antibiotic resistance profile (data not shown).

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We thank Ken Knight, Alison O'Brien, Trinad Chakraborty and Carol Kumamoto for providing antibodies and Alla Segova for instruction in RT-PCR. This work was supported by Grants NIH R01-A46454 to J.M.L., the DEFRA VTRI Fellowship to A.J.R. and D.L.G., and the Danish Natural Sciences Research Council to A.L.O.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • Arraj, J.A., Wu, T.-H., and Marinus, M.G. (1990) Expression of a DNA methylation (dam) gene in Escherichia coli K-12. Curr Microbiol 20: 133136.
  • Bell, D.C., and Cupples, C.G. (2001) Very-short-patch repair in Escherichia coli requires the dam adenine methyltransferase. J Bacteriol 183: 36313635.
  • Beltrametti, F., Kresse, A.U., and Guzman, C.A. (1999) Transcriptional regulation of the esp genes of enterohemorrhagic Escherichia coli. J Bacteriol 181: 34093418.
  • Bustamante, V.H., Santana, F.J., Calva, E., and Puente, J.L. (2001) Transcriptional regulation of type III secretion genes in enteropathogenic Escherichia coli: Ler antagonizes H-NS-dependent repression. Mol Microbiol 39: 664678.
  • Camacho, E.M., and Casadesus, J. (2005) Regulation of traJ transcription in the Salmonella virulence plasmid by strand-specific DNA adenine hemimethylation. Mol Microbiol 57: 17001718.
  • Campellone, K.G., and Leong, J.M. (2005) Nck-independent actin assembly is mediated by two phosphorylated tyrosines within enteropathogenic Escherichia coli Tir. Mol Microbiol 56: 416432.
  • Campellone, K.G., Giese, A., Tipper, D.J., and Leong, J.M. (2002) A tyrosine-phosphorylated 12-amino-acid sequence of enteropathogenic Escherichia coli Tir binds the host adaptor protein Nck and is required for Nck localization to actin pedestals. Mol Microbiol 43: 12271241.
  • Campellone, K.G., Robbins, D., and Leong, J.M. (2004) EspFU is a translocated EHEC effector that interacts with Tir and N-WASP and promotes Nck-independent actin assembly. Dev Cell 7: 217228.
  • Casadesus, J., and Low, D. (2006) Epigenetic gene regulation in the bacterial world. Microbiol Mol Biol Rev 70: 830856.
  • Dean, P., Maresca, M., and Kenny, B. (2005) EPEC's weapons of mass subversion. Curr Opin Microbiol 8: 2834.
  • Deibel, C., Kramer, S., Chakraborty, T., and Ebel, F. (1998) EspE, a novel secreted protein of attaching and effacing bacteria, is directly translocated into infected host cells, where it appears as a tyrosine-phosphorylated 90 kDa protein. Mol Microbiol 28: 463474.
  • Deng, W., Puente, J.L., Gruenheid, S., Li, Y., Vallance, B.A., Vazquez, A., et al. (2004) Dissecting virulence: systematic and functional analyses of a pathogenicity island. Proc Natl Acad Sci USA 101: 35973602.
  • DeVinney, R., Stein, M., Reinscheid, D., Abe, A., Ruschkowski, S., and Finlay, B.B. (1999) Enterohemorrhagic Escherichia coli O157:H7 produces Tir, which is translocated to the host cell membrane but is not tyrosine phosphorylated. Infect Immun 67: 23892398.
  • Donnenberg, M.S., and Whittam, T.S. (2001) Pathogenesis and evolution of virulence in enteropathogenic and enterohemorrhagic Escherichia coli. J Clin Invest 107: 539548.
  • Donohue-Rolfe, A., Kondova, I., Oswald, S., Hutto, D., and Tzipori, S. (2000) Escherichia coli O157:H7 strains that express Shiga toxin (Stx) 2 alone are more neurotropic for gnotobiotic piglets than are isotypes producing only Stx1 or both Stx1 and Stx2. J Infect Dis 181: 18251829.
  • Elliott, S.J., Sperandio, V., Giron, J.A., Shin, S., Mellies, J.L., Wainwright, L., et al. (2000) The locus of enterocyte effacement (LEE)-encoded regulator controls expression of both LEE- and non-LEE-encoded virulence factors in enteropathogenic and enterohemorrhagic Escherichia coli. Infect Immun 68: 61156126.
  • Friedberg, D., Umanski, T., Fang, Y., and Rosenshine, I. (1999) Hierarchy in the expression of the locus of enterocyte effacement genes of enteropathogenic Escherichia coli. Mol Microbiol 34: 941952.
  • Garcia-del Portillo, F., Pucciarelli, M.G., and Casadesus, J. (1999) DNA adenine methyltransferase mutants of Salmonella enterica Serovar Typhiumurium show defects in protein secretion, cell invasion, and M cell cytotoxicity. Proc Natl Acad Sci USA 96: 1157811583.
  • Garmendia, J., Phillips, A.D., Carlier, M.F., Chong, Y., Schuller, S., Marches, O., et al. (2004) TccP is an enterohaemorrhagic Escherichia coli O157:H7 type III effector protein that couples Tir to the actin-cytoskeleton. Cell Microbiol 6: 11671183.
  • Garmendia, J., Frankel, G., and Crepin, V.F. (2005) Enteropathogenic and enterohemorrhagic Escherichia coli infections: translocation, translocation, translocation. Infect Immun 73: 25732585.
  • Giron, J.A., Torres, A.G., Freer, E., and Kaper, J.B. (2002) The flagella of enteropathogenic Escherichia coli mediate adherence to epithelial cells. Mol Microbiol 44: 361379.
  • Hayward, R.D., Leong, J.M., Koronakis, V., and Campellone, K.G. (2006) Exploiting pathogenic Escherichia coli to model transmembrane receptor signalling. Nat Rev Microbiol 4: 358370.
  • Heithoff, D.M., Sinsheimer, R.L., Low, D.A., and Mahan, M.J. (1999) An essential role for DNA adenine methylation in bacterial virulence. Science 284: 967970.
  • Hernday, A.D., Braaten, B.A., and Low, D.A. (2003) The mechanism by which DNA adenine methyltransferase and PapI activate the pap epigenetic switch. Mol Cell 12: 947957.
  • Honma, Y., Fernandez, R.E., and Maurelli, A.T. (2004) A DNA adenine methyltransferase mutant of Shigella flexneri shows no significant attenuation of virulence. Microbiology 150: 10731078.
  • Jerse, A.E., Yu, J., Tall, B.D., and Kaper, J.B. (1990) A genetic locus of enteropathogenic Escherichia coli necessary for the production of attaching and effacing lesions on tissue culture cells. Proc Natl Acad Sci USA 87: 78397843.
  • Julio, S.M., Heithoff, D.M., Provenzano, D., Klose, K.E., Sinsheimer, R.L., Low, D.A., and Mahan, M.J. (2001) DNA adenine methyltransferase is essential for viability and plays a role in the pathogenesis of Yersinia pseudotuberculosis and Vibrio cholerae. Infect Immun 69: 76107615.
  • Kanamaru, K., Kanamaru, K., Tatsuno, I., Tobe, T., and Sasakawa, C. (2000) SdiA, an Escherichia coli homologue of quorum-sensing regulators, controls the expression of virulence factors in enterohaemorrhagic Escherichia coli O157:H7. Mol Microbiol 38: 805816.
  • Kaper, J.B., Nataro, J.P., and Mobley, H.L. (2004) Pathogenic Escherichia coli. Nat Rev Microbiol 2: 123140.
  • Kenny, B. (2002) Mechanism of action of EPEC type III effector molecules. Int J Med Microbiol 291: 469477.
  • Kenny, B., DeVinney, R., Stein, M., Reinscheid, D.J., Frey, E.A., and Finlay, B.B. (1997) Enteropathogenic E. coli (EPEC) transfers its receptor for intimate adherence into mammalian cells. Cell 91: 511520.
  • Livak, K.J., and Schmittgen, T.D. (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2–ΔΔCT method. Methods 25: 402408.
  • Lobner-Olesen, A., Marinus, M.G., and Hansen, F.G. (2003) Role of SeqA and Dam in Escherichia coli gene expression: a global/microarray analysis. Proc Natl Acad Sci USA 100: 46724677.
  • Lobner-Olesen, A., Skovgaard, O., and Marinus, M.G. (2005) Dam methylation: coordinating cellular processes. Curr Opin Microbiol 8: 154160.
  • Low, D.A., Weyand, N.J., and Mahan, M.J. (2001) Roles of DNA adenine methylation in regulating bacterial gene expression and virulence. Infect Immun 69: 71977204.
  • Marinus, M.G. (1987) DNA methylation in Escherichia coli. Annu Rev Genet 21: 113131.
  • Marinus, M.G. (1996) Methylation of DNA. In Escherichia Coli and Salmonella: Cellular and Molecular Biology. Neidhardt, F.C., Curtiss, R., Ingraham, J.L., Lin, E.C.C., Low, K.B., Magasanik, B., et al. (eds). Washington, DC: American Society for Microbiology Press, pp. 782791.
  • Mellies, J.L., Elliott, S.J., Sperandio, V., Donnenberg, M.S., and Kaper, J.B. (1999) The Per regulon of enteropathogenic Escherichia coli: identification of a regulatory cascade and a novel transcriptional activator, the locus of enterocyte effacement (LEE)-encoded regulator (Ler). Mol Microbiol 33: 296306.
  • Miller, V.L., and Mekalanos, J.J. (1988) A novel suicide vector and its use in construction of insertion mutations: osmoregulation of outer membrane proteins and virulence determinants in Vibrio cholerae requires toxR. J Bacteriol 170: 25752583.
  • Murphy, K.C., and Campellone, K.G. (2003) Lambda Red-mediated recombinogenic engineering of enterohemorrhagic and enteropathogenic E. coli. BMC Mol Biol 4: 11.
  • Oshima, T., Wada, C., Kawagoe, Y., Ara, T., Maeda, M., Masuda, Y., et al. (2002) Genome-wide analysis of deoxyadenosine methyltransferase-mediated control of gene expression in Escherichia coli. Mol Microbiol 45: 673695.
  • Perna, N.T., Plunkett, G., III, Burland, V., Mau, B., Glasner, J.D., Rose, D.J., et al. (2001) Genome sequence of enterohaemorrhagic Escherichia coli O157:H7. Nature 409: 529533.
  • Radlinska, M., and Bujnicki, J.M. (2001) Cloning of enterohemorrhagic Escherichia coli phage VT-2 dam methyltransferase. Acta Microbiol Pol 50: 161167.
  • Robbins-Manke, J.L., Zdraveski, Z.Z., Marinus, M., and Essigmann, J.M. (2005) Analysis of global gene expression and double-strand-break formation in DNA adenine methyltransferase- and mismatch repair-deficient Escherichia coli. J Bacteriol 187: 70277037.
  • Roe, A.J., Hoey, D.E., and Gally, D.L. (2003) Regulation, secretion and activity of type III-secreted proteins of enterohaemorrhagic Escherichia coli O157. Biochem Soc Trans 31: 98103.
  • Roe, A.J., Naylor, S.W., Spears, K.J., Yull, H.M., Dransfield, T.A., Oxford, M., et al. (2004) Co-ordinate single-cell expression of LEE4- and LEE5-encoded proteins of Escherichia coli O157:H7. Mol Microbiol 54: 337352.
  • Rosenshine, I., Ruschkowski, S., Stein, M., Reinscheid, D.J., Mills, S.D., and Finlay, B.B. (1996) A pathogenic bacterium triggers epithelial signals to form a functional bacterial receptor that mediates actin pseudopod formation. EMBO J 15: 26132624.
  • Spears, K.J., Roe, A.J., and Gally, D.L. (2006) A comparison of enteropathogenic and enterohaemorrhagic Escherichia coli pathogenesis. FEMS Microbiol Lett 255: 187202.
  • Sperandio, V., Mellies, J.L., Nguyen, W., Shin, S., and Kaper, J.B. (1999) Quorum sensing controls expression of the type III secretion gene transcription and protein secretion in enterohemorrhagic and enteropathogenic Escherichia coli. Proc Natl Acad Sci USA 96: 1519615201.
  • Sperandio, V., Torres, A.G., Jarvis, B., Nataro, J.P., and Kaper, J.B. (2003) Bacteria-host communication: the language of hormones. Proc Natl Acad Sci USA 100: 89518956.
  • Torres, A.G., and Kaper, J.B. (2003) Multiple elements controlling adherence of enterohemorrhagic Escherichia coli O157:H7 to HeLa cells. Infect Immun 71: 49854995.
  • Tzipori, S., Gibson, R., and Montanaro, J. (1989) Nature and distribution of mucosal lesions associated with enteropathogenic and enterohemorrhagic Escherichia coli in piglets and the role of plasmid-mediated factors. Infect Immun 57: 11421150.
  • Tzipori, S., Montanaro, J., Robins-Browne, R.M., Vial, P., Gibson, R., and Levine, M.M. (1992) Studies with enteroaggregative Escherichia coli in the gnotobiotic piglet gastroenteritis model. Infect Immun 60: 53025306.
  • Wion, D., and Casadesus, J. (2006) N6-methyl-adenine: an epigenetic signal for DNA–protein interactions. Nat Rev Microbiol 4: 183192.
  • Zhang, L., Chaudhuri, R.R., Constantinidou, C., Hobman, J.L., Patel, M.D., Jones, A.C., et al. (2004) Regulators encoded in the Escherichia coli type III secretion system 2 gene cluster influence expression of genes within the locus for enterocyte effacement in enterohemorrhagic E. coli O157:H7. Infect Immun 72: 72827293.