• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of iaiPermissionsJournals.ASM.orgJournalIAI ArticleJournal InfoAuthorsReviewers
Infect Immun. Nov 2008; 76(11): 5062–5071.
Published online Sep 15, 2008. doi:  10.1128/IAI.00654-08
PMCID: PMC2573367

Contribution of the Ler- and H-NS-Regulated Long Polar Fimbriae of Escherichia coli O157:H7 during Binding to Tissue-Cultured Cells[down-pointing small open triangle]


The expression of the long polar fimbriae (LPF) of enterohemorrhagic Escherichia coli (EHEC) O157:H7 is controlled by a tightly regulated process, and, therefore, the role of these fimbriae during binding to epithelial cells has been difficult to establish. We recently found that histone-like nucleoid-structuring protein (H-NS) binds to the regulatory sequence of the E. coli O157:H7 lpf1 operon and “silences” its transcription, while Ler inhibits the action of the H-NS protein and allows lpf1 to be expressed. In the present study, we determined how the deregulated expression of LPF affects binding of EHEC O157:H7 to tissue-cultured cells, correlating the adherence phenotype with lpf1 expression. We tested the adherence properties of EHEC hns mutant and found that this strain adhered 2.8-fold better than the wild type. In contrast, the EHEC ler mutant adhered 2.1-fold less than the wild type. The EHEC hns ler mutant constitutively expressed the lpf genes, and, therefore, we observed that the double mutant adhered 5.6-fold times better than the wild type. Disruption of lpfA in the EHEC hns and hns ler mutants or the addition of anti-LpfA serum caused a reduction in adhesion, demonstrating that the increased adherence was due to the expression of LPF. Immunogold-labeling electron microscopy showed that LPF is present on the surface of EHEC lpfA+ strains. Furthermore, we showed that EHEC expressing LPF agglutinates red blood cells from different species and that the agglutination was blocked by the addition of anti-LpfA serum. Overall, our data confirmed that expression of LPF is a tightly regulated process and, for the first time, demonstrated that these fimbriae are associated with adherence and hemagglutination phenotypes in EHEC O157:H7.

Enterohemorrhagic Escherichia coli (EHEC) O157:H7, a serious food-borne pathogen which causes diarrhea that is often bloody, accompanied by severe abdominal cramps, can result in a life-threatening condition known as the hemolytic-uremic syndrome (reviewed in reference 20). The organism can be found living in the intestines of healthy cattle, but eating contaminated meat, especially ground beef, has resulted in multiple outbreaks worldwide (12). Recent E. coli O157:H7 outbreaks have drawn attention to food-borne illnesses, and though CDC reports indicated that the overall number of cases is declining in the United States, consumption of produce, particularly leafy vegetables and spinach, is becoming increasingly associated with human infections. Furthermore, it was recently demonstrated that strains associated with these outbreaks represent an emergent subpopulation that has acquired critical factors that contribute to more severe disease (4, 5, 17, 34). This alternate source of infection has set up new challenges for the scientific community to try identifying novel determinants and regulatory mechanisms implicated in the colonization, survival, and/or pathogenic processes.

During the infectious process, EHEC adheres to the intestinal epithelium, where it produces Shiga toxins responsible for the hemorrhagic symptoms. Adhesion of E. coli O157:H7 to enterocytes induces the formation of the attaching and effacing (A/E) lesion (reviewed in reference 42). The A/E phenotype is mainly conferred by the locus of enterocyte effacement (LEE), a pathogenicity island containing genes coding for structural components of a type III secretion apparatus, translocator and secreted effector proteins, an adhesin (intimin), and the intimin receptor, Tir (reviewed in reference 39). Previous studies of EHEC and enteropathogenic E. coli (EPEC) have shown that the expression of LEE-encoded virulence factors is regulated by a complex assortment of environmental cues and a variety of regulatory elements encoded inside and outside the LEE (reviewed in reference 18). Various regulators outside the LEE have been characterized in A/E-producing E. coli strains, including the histone-like nucleoid-structuring protein (H-NS), the integrated host factor, two nucleoproteins, Fis and Hha, and the quorum-sensing regulators QseA and QseD (2, 11, 24, 28, 31, 32, 35). H-NS plays an important role in the transcriptional “silencing” of genes located in the different operons of the LEE, and it is responsive to multiple environmental signals and regulatory proteins (18). A/E-producing E. coli strains, including EPEC and EHEC, also encode Ler, an additional H-NS-like protein (19). The Ler protein is particularly interesting, as it is encoded within the LEE and induces the expression of the genes in the pathogenicity island by counteracting the H-NS-mediated silencing (3, 19). In addition, it has been reported that Ler regulates the expression of proteins encoded outside the LEE and which are not essential for A/E lesion formation, including EspC in EPEC or a “long fine” fimbria in EHEC (7).

In comparison to the well-established regulatory mechanisms controlling expression of LEE-encoded genes, very little is known about the mechanisms regulating E. coli O157:H7 colonization factors, including those controlling fimbria expression. EHEC O157:H7 contains two nonidentical lpf loci homologous to the long polar fimbriae (LPF) of Salmonella enterica serovar Typhimurium (reviewed in reference 42). Expression of the E. coli O157:H7 lpf operon 1 (lpf1) in E. coli K-12 has been linked to increased adherence to tissue-cultured cells and has been associated with the appearance of peritrichous long fimbriae (37). E. coli O157:H7 strains harboring mutations in one or both of the lpf loci have diminished colonization abilities in swine and sheep animal models (14) and also displayed an altered human intestinal tissue tropism (10). Additionally, expression of lpf1 is known to be regulated in response to growth phase, temperature, and pH, and in vivo data, using a lamb model, support the role of LPF as a colonization factor associated with persistence in the intestine of infected animals (41). Recently we clarified the connection between regulatory proteins and expression of the lpf1 loci in response to environmental cues (40). We demonstrated that H-NS protein functions as a transcriptional silencer, while the LEE-encoded Ler functions as an antisilencer of LPF expression, by interacting directly with the lpf promoter region. In the present study, we investigated whether the deregulated expression of LPF in the different regulatory mutants affects binding of EHEC O157:H7 to tissue-cultured cells, correlating the adherence phenotype with lpf1 expression.


Bacterial strains, plasmids, media, and growth conditions.

Bacterial strains and plasmids used in this study are listed in Table Table1.1. Strains were routinely grown in Luria-Bertani (LB) broth or Dulbecco's minimal Eagle's medium (DMEM [Gibco/Invitrogen]) at 37°C. The pH of DMEM was adjusted to 6.5 as previously described (40). Antibiotics were added to media at the following concentrations: kanamycin, 50 μg/ml; ampicillin, 100 μg/ml; chloramphenicol, 30 μg/ml; and tetracycline, 12.5 μg/ml.

Bacterial strains and plasmids used in this study

Recombinant DNA techniques.

Standard methods were used to perform plasmid purification, PCR, ligation, restriction digests, transformation, conjugation, and gel electrophoresis (29).

Construction of EHEC O157:H7 lpfA single, double, and triple mutants.

EHEC strains defective in LPF expression were constructed in the chromosome of strains EDL933, AC425, and SDP01 (Table (Table1)1) by marker exchange as follows. The suicide vector pLPF::cat (lpfA::cat inserted into pCVD442 [37]) was introduced into EDL933, AC425, and SDP01 by conjugation using the donor strain SM10 (λ Pir) (33). Colonies resistant to sucrose, chloramphenicol, and the respective antibiotic markers already present in the different EHEC strains were tested for ampicillin sensitivity. The presence of the cat cassette within the chromosomal lpfA gene of AGT203, TMS002, and TMS003 was confirmed by PCR as described before (37) and by phenotypic assays (unable to express the LPF but still able to form A/E lesions [data not shown]). These new mutants, referred as TMS002 (hns lpfA) and TMS003 (hns ler lpfA), were utilized in our adhesion, hemagglutination (HA) assays, and in immunogold electron microscopy experiments (described below), to show that LPF is expressed on the surface of EHEC O157:H7 and that these fimbriae are associated with adherence to tissue-cultured cells and/or agglutination of erythrocytes.

Bacterial adhesion to cultured epithelial cells.

Adherence of bacterial strains to cultured HeLa epithelial cells was evaluated as previously described (37). DMEM with 10% (vol/vol) heat-inactivated fetal bovine serum, 2 mM l-glutamine, penicillin (100,000 IU/liter), and streptomycin (100 mg/liter) was used to grow HeLa cells prior to bacterial infection. Bacteria were grown statically in LB broth overnight at 37°C and inoculated at a multiplicity of infection of approximately 10:1 onto semiconfluent cultured epithelial cell monolayers grown on 24-well microtiter plates. Before use, the cells were washed with sterile phosphate-buffered saline (PBS [pH 7.4]) and replenished with DMEM alone or supplemented with 1% d-mannose. Bacteria and cells were incubated for 3 h at 37°C and 5% CO2, and then cells were washed five times with PBS. The infected monolayers were fixed and stained with Giemsa's solution for microscopic evaluation. To quantify E. coli adherence, the bacteria were recovered with 0.1% Triton X-100 in PBS buffer and plated on Luria agar plates containing the proper antibiotic. When required, the bacterial suspensions were combined with preimmune or anti-LpfA serum (1:5,000; LpfA antiserum from S. enterica serovar Typhimurium was kindly provided by A. J. Bäumler, University of California, Davis) for 30 min at room temperature prior to infecting the HeLa cells. Inhibition of LPF binding by the different sera was determined after 3 h postinfection. Adherence data are expressed as the percentage of adherent bacteria (or CFU per ml) recovered from triplicate wells after subtraction of bacteria attached to plastic wells, and the means were gathered from two independent experiments. Statistical differences are expressed as the P value determined by a paired Student's t test.

Electron microscopy.

Strains were grown overnight at 37°C under static conditions in DMEM, and then bacteria were centrifuged at 3,000 × g. The bacteria were resuspended in PBS (pH 7.4) and allowed to adhere to Formvar-carbon-coated copper grids (200 mesh; Electron Microscopy Sciences) as previously described (38). For immunogold labeling of LPF fimbriae, bacteria were reacted with anti-LpfA serum and 15-nm gold-labeled anti-rabbit immunoglobulin G (AuroProbe GAR G15, RPN422; Amersham Biosciences, United Kingdom) and negatively stained with 2% potassium-phosphotungstic acid (pH 6.8). Specimens were examined in a Phillips 201 electron microscope.

HA assay.

HA activity was measured on a 96-well round-bottom plate, adapting a method described by Evans et al. (9). Bacteria were grown overnight statically in DMEM (pH 6.5) at 37°C and then were harvested and suspended in PBS (pH 7.4) to a concentration of 1 × 108 CFU/ml. Type A human blood (Immunocor, Inc., Norcross, GA) was diluted to a 10% concentration with PBS to test for HA, and 10% supplemented with 1% mannose in PBS to test for mannose-resistant HA. The same procedure was used for freshly drawn mouse, rat, guinea pig, and sheep blood obtained from the University of Texas Medical Branch Animal Research Center. Bacterial cells were mixed with the appropriate species of blood (ca. 0.1 ml). After observation for about 5 min at room temperature, the 96-well plate was placed on the surface of ice, incubated for 2 h, and monitored every 15 min. The results were recorded as follows. A reaction of ++++ was instantaneous and complete, involving all of the erythrocytes (even sheet of erythrocytes across the well). Lesser degrees of HA were recorded as +++, ++, and +, whereas reactions containing a small erythrocyte pellet at the bottom of the well were considered negative (−). When required, the bacterial suspensions were combined with anti-LpfA serum (1:2,500 or 1:5,000) for 15 min at room temperature prior to the addition of the different species of blood. The inhibition of LPF binding to erythrocytes was monitored every 15 min, and the assay was performed on ice for 1 h.

Heat-extracted proteins and Western blot analysis.

Bacterial samples were grown statically in DMEM (pH 6.5) overnight at 37°C, and cell concentrations were standardized based on optical density at 600 nm readings. Cells (4 × 109) were collected and then harvested by centrifugation at 3,000 × g for 10 min, resuspended in 160 μl of PBS (pH 7.4), and incubated at 60°C for 30 min. The supernatant of centrifuged samples was mixed with 40 μl of 5× sodium dodecyl sulfate-sample buffer (16) and boiled for 5 min. Heat-extracted proteins were separated onto a 12% sodium dodecyl sulfate-polyacrylamide gel and transferred onto polyvinylidene difluoride Immobilon-P transfer membrane (Millipore, Bedford, MA). Membranes were probed with the polyclonal anti-LpfA antibody and blot developed with an ECL-Plus kit (Amersham, Piscataway, NJ).


Ler- and H-NS-regulated LPF-mediated adherence to tissue-cultured cells.

When we initiated our studies characterizing the lpf operons in EHEC O157:H7, a major obstacle that we encountered was the difficulty of visualizing and purifying fimbriae from E. coli O157:H7 because of inconsistent expression of LPF. We hypothesized at the time that this was due to unknown, stringent regulatory mechanisms controlling expression of LPF. Therefore, we undertook the task to establish the connection between regulatory mechanisms acting on the lpf1 loci in response to environmental cues. Transcriptional analysis of the lpf1 promoter fused to a reporter revealed that expression of lpf was enhanced ~1.7-fold in the hns mutant, and a significant reduction in lpfA1p::lacZ enzymatic activity was observed in a ler mutant strain (40). We also found that the levels of enzymatic activity in the Δler Δhns double mutant became constitutive, suggesting that EHEC O157:H7 possess other regulatory factors that may be controlling expression of lpf in the absence of Ler and H-NS proteins (40). Because we identified the optimal conditions controlling expression of LPF, the participation of these fimbriae in the adherence phenotype of EHEC O157:H7 was next established.

We have previously demonstrated that differences in the adherence phenotype of different EHEC O157:H7 lpfA mutant strains to tissue-cultured cells, after a 3- or 6-h infection, were minimal compared to their respective wild-type strains (10, 37). Similar to other lpfA single-mutant strains, only modest reduction in the adherence of the AGT203 (lpfA) strain was observed compared with wild-type strain EDL933 (Fig. (Fig.1),1), indicative of the role played by other factors in adhesion to HeLa cells under the experimental conditions tested. Because we found that Ler and H-NS controlled expression of LPF, we then tested the adherence properties of the AC425 (hns) mutant strain. This strain adhered 2.2-fold better than the wild type, when mannose was present in the assay. (The standard adhesion assay utilized mannose to block unspecific binding.) Although the difference was not statistically different, the result suggested that a gene or genes encoding adhesion factors are expressed in the absence of the negative regulator (Fig. (Fig.1).1). In contrast, the CB49 (ler) mutant adhered 4.3-fold less than the wild type, a result which could be predicted because the LEE-encoded genes are poorly expressed in the absence of the positive regulator Ler. Because we observed that the SDP01 (hns ler) double mutant constitutively expressed the lpf genes, we next asked whether this deregulation had an effect in adhesion. The hns ler double mutant (which we predicted should not express LEE-encoded genes) adhered 7.2-fold times better than the wild type, which strongly suggested that this mutant hyperexpressed an adhesin, likely the LPF (Fig. (Fig.11).

FIG. 1.
Adhesion of E. coli O157:H7 wild-type and mutant strains to tissue-cultured cells. HeLa cells were incubated for 3 h with EHEC strains EDL933 (wild type), AGT203 (lpfA), AC425 (hns), CB49 (ler), SDP01 (hns ler), TMS002 (hns lpfA), and TMS003 (hns ler ...

We next determined whether mannose was having an effect on the adherence of our mutant strains. Interestingly, we observed an overall increase in the adherence to HeLa cells of the wild type as well as the mutant strains. In the absence of this sugar, the wild type and the lpfA (AGT203) mutant adhered at similar levels (Fig. (Fig.1).1). In contrast, the hns (AC425) mutant adhered 2.8-fold better than the wild type and the ler (CB49) mutant adhered 2.1-fold less than the parent strain (Fig. (Fig.1),1), supporting our previous findings in the presence of mannose. Finally, the ler hns (SDP01) mutant adhered 5.6-fold better than the wild type; however, the number of mutant bacteria recovered after infection was significantly higher than that of the organisms obtained in the adhesion assay supplemented with mannose. Overall, the data suggested that mannose has an overall effect on adhesion but does not seem to direct block LPF-mediated adherence. To confirm that the increased adherence in the hns single and hns ler double mutants was due to the lpf genes, we constructed lpfA double and triple mutants. In standard adherence assays, we showed that the TMS002 (hns lpfA) double mutant had a 2.5-fold reduction in adhesion (3.1-fold in the absence of mannose) compared to the hns mutant (Fig. (Fig.1).1). Additionally, strain TMS003 (hns ler lpfA triple mutant) demonstrated that the hyperadherence observed in the hns ler double mutant was reduced 18.6-fold (11.8-fold in the absence of mannose) and confirmed that lpf gene expression is regulated by H-NS, Ler, and other uncharacterized regulatory factors and that LPF is responsible for the adherence phenotype observed in the different mutant strains.

Qualitative analysis of the adherent organisms supported our quantitative results because we did not observe unspecific binding of the bacterial strains to plastic wells or to the coverslips used in the Giemsa staining (Fig. (Fig.2).2). As we have previously reported, E. coli O157:H7 does not form distinct microcolonies, and instead, the wild-type bacteria attached to the cells in small aggregates were distributed on top of the HeLa cells (Fig. (Fig.2A).2A). The lpfA mutant adhered similarly to the wild-type strain (Fig. (Fig.2B).2B). The ler mutant also adhered to the HeLa cells, but, in contrast, very few aggregates were observed (Fig. (Fig.2C).2C). The hns and the ler hns mutant strains adhered better than any of the strains tested, and larger aggregates were observed, many of them not in direct contact with the HeLa cells (Fig. 2D and E). Finally, the mutation of the lpfA gene in the hns and ler hns strains (TMS002 and TMS003, respectively) confirmed that the increased adherence and formation of larger aggregates were associated with the expression of LPF (Fig. 2F and G). Adherence to plastic wells without cells was minimum, no aggregates were observed, and similar numbers of bacteria were recovered in the wild type and the isogenic mutants (~2.1 × 103 to 3.8 × 103 bacteria per well). Also, autoaggregation assays were performed to determine whether this was the reason of the increase adherence/aggregation phenotypes, but the results indicated that the wild type and the mutants aggregated at the same rate (data not shown). The reason for the apparent aggregation is unknown at this moment, but a link between these virulence-associated regulators (H-NS and Ler) and the lpf operon was established and strongly suggested that this fimbrial adhesin might be playing an important and still unidentified role in pathogenesis of EHEC or other LPF-expressing E. coli strains.

FIG. 2.
Qualitative adherence assays. Shown are Giemsa-stained HeLa cells infected after 3 h of incubation with EHEC strains (A) EDL933, (B) AGT203 (lpfA), (C) CB49 (ler), (D) AC425 (hns), (E) SDP01 (hns ler), (F) TMS002 (hns lpfA), and (G) TMS003 (hns ler lpfA ...

Finally, we investigated whether the antiserum raised against the LpfA protein was able to block adherence of the EHEC strains to HeLa cells. For all of the strains tested, adherence to HeLa cells was reduced at the concentrations of anti-LpfA serum selected (1:5,000) (Table (Table2).2). It is important to note that the antiserum effect was statistically significant in the strains AC425 and SDP01 that had increased expression of LPF (P < 0.05) when comparing infections with and without the sera. Surprisingly, the anti-LpfA serum had a significant effect on the adherence of the TMS003 strain (P < 0.05) but not on any of the other strains tested. To confirm that the blocking of adherence was a specific effect due to antibodies against LpfA, an additional experiment was performed to determine whether rabbit preimmune serum also inhibited adherence of our EHEC strains. As indicated in Table Table2,2, the rabbit serum had no effect in blocking adherence of the EHEC strains at the same concentration used with the anti-LpfA serum. Together, these data suggested that LPF-mediated adherence could be blocked by the corresponding sera.

Adherence of EHEC strain EDL933 and its mutants to HeLa cells in the presence of anti-LpfA serum

Detection of the H-NS- and Ler-regulated, surface-localized LPF.

Based on the data presented above, we then determined if LPF fimbriae were localized on the surface of E. coli O157:H7 strains. The ultrastructural studies were performed with strain EDL933 and the hns, ler hns, or ler hns lpfA mutant strains, using immunogold labeling electron microscopy techniques and anti-LpfA antibodies. Very few gold particles were seen decorating EHEC EDL933 grown under LPF-inducing conditions, confirming our previous results indicating that LPF is barely localized on the surface of the wild-type strain under optimal laboratory conditions (Fig. (Fig.3A)3A) (37). In contrast, the hns (AC425) mutant expressed LPF, as exhibited by the presence of gold particles at the surface of the bacteria (Fig. (Fig.3B).3B). The result that confirmed our transcriptional data and adhesion phenotype was performed with the ler hns (SDP01) mutant because this strain was predicted to constitutively expressed LPF. Our immunogold pictures confirmed that LPF was localized on the surface of SDP01, as observed by the presence of several gold particles decorating these fimbriae (Fig. (Fig.3C).3C). Remarkably, the constitutively expressed LPF was located at the poles of the bacteria and two distinct structures were observed: long, individually expressed fimbriae and the formation of bundles at one of the poles of the organism (Fig. (Fig.3C).3C). As negative control and to confirm our observations, we tested strain TMS003 (ler hns lpfA), and the results indicated that the triple mutant was not labeled with anti-LpfA antibodies (Fig. (Fig.3D3D).

FIG. 3.
LPF expression was visualized by immunogold-labeling electron microscopy and Western blot analysis. EHEC strains (A) EDL933 (A), AC425 (hns) (B), SDP01 (hns ler) (C), and TMS003 (hns ler lpfA) (D) grown under optimal conditions of LPF expression were ...

Finally, we compared heat-extracted protein profiles of the E. coli O157:H7 parent strain with the different mutant strains, grown in DMEM under inducing conditions. A Western blot of the heat-extracted proteins and anti-LpfA serum revealed a band of ~15 kDa in the samples derived from the AC425 (hns) and SDP01 (ler hns) mutant strains, suggesting that these strains expressed LpfA. In contrast, this protein was absent from the heat extracts of the wild-type E. coli O157:H7 strain. The band was also absent from the AGT203 (lpfA), CB49 (ler), TMS002 (hns lpfA), and TMS003 (ler hns lpfA), under the conditions tested (data not shown). Overall, the results strongly suggest that LPF is present on the surface of the hns and the ler hns mutant strains and confirmed that lpf expression is tightly regulated and associated with the adherence phenotype.

Expression of LPF increases HA of E. coli O157:H7 mutant strains.

Type 1 fimbriae are expressed by most E. coli strains and are characterized by their ability to bind to mannose derivatives; such binding can be inhibited by the addition of exogenous α-d-mannose, a phenotype termed mannose-sensitive HA. We have shown that E. coli O157:H7 LPF1 proteins are related to the type 1 fimbriae of E. coli and S. enterica serovar Typhimurium. To determine whether the mutant strains expressing LPF displayed HA or mannose-sensitive HA phenotypes, the bacteria were grown under LPF-inducing conditions and mixed with erythrocytes derived from different animal species or of human origin. As shown in Table Table3,3, the wild-type E. coli O157:H7 strain EDL933 did not agglutinate human, sheep, rat, mouse, or guinea pig erythrocytes. Similarly, AGT203 (lpfA) and CB49 (ler) mutant strains did not agglutinate any of the cells tested, except for strain CB49, which hemagglutinated a 25% sheep erythrocyte suspension. In contrast, the AC425 (hns) and SDP01 (ler hns) hemagglutinated rat, mouse, and guinea pig red blood cells, and this HA was mannose resistant (Table (Table3).3). To determine whether the HA observed in these mutants was due to the expression of LPF, we tested the TMS002 (hns lpfA) and TMS003 (ler hns lpfA) strains. To our surprise, these two mutant strains agglutinated all of the red blood cells tested and the HA was mannose resistant. Finally, to determine whether the HA of rat and mouse erythrocytes observed with the AC425 and SDP01 strains was due to the expression of LPF, the HA assay was performed in the presence of anti-LpfA serum. As shown in Table Table4,4, the antibody partially inhibited the HA phenotype. This piece of data indicated that LPF contributes to the HA phenotype and that in the absence in the regulatory mutants of LPF, a putative hemagglutinin is expressed.

HA assay results
Inhibition of HA with anti-LpfA serum


In this study, we performed further characterization of a chromosomal locus containing genes encoding the long polar fimbriae 1 of E. coli O157:H7. We analyzed the roles of several regulators in the expression of lpf1 and found that H-NS functions as a silencer, while Ler functions as an antisilencer of this operon (40). Furthermore, our data are the first to unequivocally show that LPF is present on the surface of the hns and ler hns mutant strains and confirm that lpf expression is tightly regulated and associated with the adherence phenotype. Our findings also supported the role of these surface-expressed fimbriae in the agglutination of red blood cells. However, we still do not know whether additional hemagglutinins are expressed in the absence of the regulator H-NS and the LPF fimbriae. It is clear that, up to this date, not much was known about the regulation of putative or novel colonization factors of E. coli O157:H7 different from LEE-encoded intimin. Consequently, our study is important because it expands the knowledge about the function of two well-established virulence regulators, Ler and H-NS, and helps us to understand how they coordinate and control the expression of genes outside the LEE pathogenicity island, which participate in adherence and, potentially, intestinal colonization, as in the case of LPF.

Initial observations while trying to establish the role of Ler as a global regulator of adherence were by Elliott et al. (7), who reported that, in addition to the effects of Ler on LEE-located genes, Ler regulated the expression of proteins encoded elsewhere in the genome. Their data suggested that Ler was regulating fimbrial expression and adherence phenotypes because a mutation of the ler gene was associated with enhanced adherence to tissue culture monolayers, altered adherence patterns, and expression of long “fine” fimbriae (7). In contrast, our present data disproved these observations because our results showed that Ler acts as a positive regulator of the lpf operon, playing a similar regulatory role to that observed on LEE-encoded genes and that a mutation in the ler gene produced a reduction in the adherence phenotype. We also did not observe the presence of fimbriae in the ler mutant under any of the conditions tested. Our present results are further validated by work reported by Ogierman et al. (21), who showed that Ler increases the level of intimin in some EHEC O157:H strains, but the adherence to tissue-cultured cells was shown to be independent of intimin. Accordingly, their data suggested that Ler is regulating the expression of an alternative adherence factor responsible for the phenotype in EHEC O157:H. No further studies have been published describing the alternative adherence factors of those EHEC O157:H strains, but it is plausible to suggest that LPF1, located in the genome of every single E. coli O157 strain that we have analyzed so far (data not shown), might be responsible for the intimin-independent adhesion in those EHEC O157:H strains. Furthermore, we proposed that E. coli O157:H7, under optimal intestinal environmental conditions, induces the expression of LPF, LEE-encoded proteins, and probably some other non-LEE-encoded proteins, to cause the full pathogenic phenotype. However, the deregulation of LPF only occurs under specific and well-defined environmental conditions that might be coordinated by or dependent on the expression of LEE-encoded proteins.

In the case of H-NS, its role as a regulator of adherence factors (other than LEE-encoded genes) in pathogenic E. coli has not been well documented. Scott et al. (30) performed a transposon mutagenesis in a Shiga toxin-producing E. coli (STEC) 091:H21 strain B2F1, which does not carry the LEE genes, to identify genes involved in binding to human colonic epithelial T84 cells. They found that several of the mutants displayed reduced adherence to T84 cells, and some of those mutants mapped to the hns gene. Their overall analysis demonstrated that H-NS regulated the expression of several genes in the LEE-negative B2F1 STEC strain; however, they did not identify any H-NS-regulated adherence factor. At the moment, we do not know whether the B2F1 STEC strain carries the lpf1 loci, but we could proposed that the reason the hns mutation resulted in reduced adherence was due to the absence of the lpf1 genes. If this is true, it might indicate that those STEC strains that are LEE negative, but possess the lpf1 loci, have an advantage over those lpf-negative strains, because they can use these fimbriae to colonize specific intestinal regions. Another H-NS-regulated surface structure identified in E. coli strains is the curli fibers (22). Curli are surface organelles that mediate binding to soluble matrix proteins, and the genes encoding these fibers are highly conserved between S. enterica serovar Typhimurium and E. coli (27). In the case of E. coli O157:H7 strains, the role of curli fibers has been associated with the ability of some curli-positive strains to bind tissue-cultured epithelial cells as a prerequisite before bacterial invasion (15, 43). In our case, the prototype strain selected for our studies does not produce curli fibers under the conditions tested and, furthermore, does not invade tissue-cultured cells under any condition (data not shown); therefore, the potential contribution of curli to the hyperadherence phenotype observed with our E. coli O157:H7 hns mutants can be disregarded.

A quite surprising result was the fact that the E. coli O157:H7 hns ler double mutant adhered better than the hns single mutant. Our previous study demonstrated that the lpf operon becomes constitutively expressed in the absence of the H-NS and Ler regulators, which suggested that E. coli O157:H7 possesses other regulatory factors controlling expression of lpf (40). Our present study suggests that in the hns mutant strain, Ler activated, but also modulated, expression of the lpf1 operon, while in the hns ler double mutant, the modulation effect was lost and the fimbria is deregulated and constitutively expressed. However, we cannot discount the participation of other adhesins in the overall binding to HeLa cells. Recent studies testing the effect of overproduction of Ler on global transcription levels, indicated that the Ler protein activated several virulence regulatory systems, including multiple fimbrial adhesins, in a strict fashion in response to environmental signals that are closely correlated to the LEE-encoded genes (1).

One pending issue of our previous publication that is, in part, answered by our functional studies described here has to do with the unusual characteristic of the E. coli O157:H7 lpf1 operon, which contains two open reading frames (lpfC and lpfC′) predicted to encode the putative outer membrane components of the fimbriae (37). The DNA sequence comparison with other related fimbrial outer membrane proteins indicated that lpfC is disrupted in E. coli O157:H7 and, due to the difficulty of expressing the fimbriae in vitro, we were unable to rule out, at the time, that the LPF subunits were not translocated and assembled on the surface of the bacterial cell, due to the lack of a functional outer membrane component. Nevertheless, the evidence presented here clearly indicated that the lack of expression of the fimbriae was due not to a nonfunctional operon, but instead, to a tight regulation mediated by H-NS and Ler. One additional characteristic associated with the regulation of the lpf1 operon is that the genes encoding these fimbrial proteins are present in multiple A/E lesion-forming bacteria and S. enterica serovar Typhimurium. Still, some of the lpf operons already characterized, specifically those in enteropathogenic E. coli O127:H6 and Citrobacter rodentium, which are complete and intact, are apparently not expressed (36). The lack of lpf expression in these pathogens could be attributed to tight regulation or to the absence of specific promoter elements which are only present in E. coli O157:H7; such evidence supporting this hypothesis has already been presented in our prior publication (40). However, the differences in expression could also be due to other regulatory mechanisms that are only found in some pathogens and that are used to repress the expression of the lpf genes under most of the conditions tested, and, perhaps, the repression of these lpf genes is only relieved under specific conditions which occur during an infection stage in humans, in an animal reservoir, or during a free phase in the environment.

The role of LPF in the HA phenotype of EHEC O157:H7 strains needs further investigation. It was clear from our early studies that it was not possible to detect LPF on wild-type EHEC O157:H7 isolate EDL933 under optimal laboratory conditions (37), and therefore, strain EDL933 did not show LPF-mediated HA of any of the red blood cells tested. When deregulation of LPF was obtained in the hns and the hns ler mutants, MRHA of erythrocytes from rodent species was obtained, indicating that the LPF fimbrial subunits interact with these cells and that the HA phenotype is different from the one observed in type 1 fimbria-expressing E. coli strains. Excitingly, disruption of the lpfA gene in the hns and the hns ler EDL933 mutants, which was anticipated to eliminate the LPF-mediated HA phenotype, produced an increase in the ability of the mutant strains to agglutinate rodent red blood cells as well as those from sheep and humans. Two previous studies indicated that expression of multiple horizontally acquired genes, scattered throughout the E. coli chromosome, is controlled by Ler and H-NS, and several of those genes encode proteins predicted to be associated with adhesion and/or agglutination (1, 23). Although the identity of the novel hemagglutinin(s) expressed by the TMS002 (hns lpfA) and TMS003 (hns ler lpfA) mutant strains is currently unknown, our data suggest that in the absence of LPF, the E. coli O157:H7 strain is capable of expressing alternative adhesins; however, type 1 fimbriae might not be one of them, because it has been demonstrated that E. coli O157:H7 or O157:H strains do not express type 1 fimbriae due to a 16-bp deletion in the invertible element, producing an “off” phenotype (8, 13, 26). The idea of expression of other adhesins is supported by our in vivo studies, where the absence of the lpf genes causes a different pattern of colonization in animal models of infection (14, 41), or in our ex vivo studies, where the mutant strain displayed a different tissue tropism in human intestinal biopsies (10).

An important number of regulators have been shown to modulate expression of adherence-associated genes in E. coli O157:H7 (reviewed in reference 18), although our studies are the first that clearly link two of those regulators, Ler and H-NS, with the control of LPF and demonstrate that a tight regulation is associated with the LPF-mediated adherence and HA phenotypes. Whether additional factors are connected to the regulatory mechanism of the lpf operon and whether, in the absence of LPF, additional adhesins/agglutinins are expressed are issues that have to be further clarified to understand the complexity of the E. coli O157:H7 colonization phenotypes.


We thank Stacy Diaz for critical reading of the manuscript.

The laboratory of A.G.T. was supported in part by institutional funds from the UTMB John Sealy Memorial Endowment Fund for Biomedical Research. M.M.P.A-H. received a scholarship from VIEP, BUAP, and from the Integral System for the Development of Consolidated and in Consolidation Academic Groups Project, Mexico (no. BUAP CA-89-2006-15-03), to perform a scientific visit to UTMB.


Editor: A. J. Bäumler


[down-pointing small open triangle]Published ahead of print on 15 September 2008.


1. Abe, H., A. Miyahara, T. Oshima, K. Tashiro, Y. Ogura, S. Kuhara, N. Ogasawara, T. Hayashi, and T. Tobe. 2008. Global regulation by horizontally transferred regulators establishes the pathogenicity of Escherichia coli. DNA Res. 1525-38. [PMC free article] [PubMed]
2. Beltrametti, F., A. U. Kresse, and C. A. Guzmán. 1999. Transcriptional regulation of the esp genes of enterohemorrhagic Escherichia coli. J. Bacteriol. 1813409-3418. [PMC free article] [PubMed]
3. Bustamante, V. H., F. J. Santana, E. Calva, and J. L. Puente. 2001. Transcriptional regulation of type III secretion genes in enteropathogenic Escherichia coli: Ler antagonizes H-NS-dependent repression. Mol. Microbiol. 39664-678. [PubMed]
4. CDC. 2006. Ongoing multistate outbreak of Escherichia coli serotype O157:H7 infections associated with consumption of fresh spinach—United States, September 2006. MMWR Morb. Mortal. Wkly. Rep. 551045-1046. [PubMed]
5. CDC. 2006. Preliminary FoodNet data on the incidence of infection with pathogens transmitted commonly through food—10 states, United States, 2005. MMWR Morb. Mortal. Wkly. Rep. 55392-395. [PubMed]
6. Donnenberg, M. S., and J. B. Kaper. 1991. Construction of an eae deletion mutant of enteropathogenic Escherichia coli by using a positive-selection suicide vector. Infect. Immun. 594310-4317. [PMC free article] [PubMed]
7. Elliott, S. J., V. Sperandio, J. A. Girón, S. Shin, J. L. Mellies, L. Wainwright, S. W. Hutcheson, T. K. McDaniel, and J. B. Kaper. 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. 686115-6126. [PMC free article] [PubMed]
8. Enami, M., N. Nakasone, Y. Honma, S. Kakinohana, J. Kudaka, and M. Iwanaga. 1999. Expression of type I pili is abolished in verotoxin-producing Escherichia coli O157. FEMS Microbiol. Lett. 179467-472. [PubMed]
9. Evans, D. J., Jr., D. G. Evans, and H. L. DuPont. 1979. Hemagglutination patterns of enterotoxigenic and enteropathogenic Escherichia coli determined with human, bovine, chicken, and guinea pig erythrocytes in the presence and absence of mannose. Infect. Immun. 23336-346. [PMC free article] [PubMed]
10. Fitzhenry, R., S. Dahan, A. G. Torres, Y. Chong, R. Heuschkel, S. Murch, M. Thomson, J. B. Kaper, G. Frankel, and A. D. Phillips. 2006. Long polar fimbriae and tissue tropism in Escherichia coli O157:H7. Int. J. Med. Microbiol. 296547-552. [PubMed]
11. Goldberg, M. D., M. Johnson, J. C. Hinton, and P. H. Williams. 2001. Role of the nucleoid-associated protein Fis in the regulation of virulence properties of enteropathogenic Escherichia coli. Mol. Microbiol. 41549-559. [PubMed]
12. Hussein, H. S., and L. M. Bollinger. 2005. Prevalence of Shiga toxin-producing Escherichia coli in beef cattle. J. Food Prot. 682224-2241. [PubMed]
13. Iida, K.-I., Y. Mizunoe, S. N. Wai, and S.-I. Yoshida. 2001. Type 1 fimbriation and its phase switching in diarrheagenic Escherichia coli strains. Clin. Diagn. Lab. Immunol. 8489-495. [PMC free article] [PubMed]
14. Jordan, D. M., N. Cornick, A. G. Torres, E. A. Dean-Nystrom, J. B. Kaper, and H. W. Moon. 2004. Long polar fimbriae contribute to colonization by Escherichia coli O157:H7 in vivo. Infect. Immun. 726168-6171. [PMC free article] [PubMed]
15. Kim, S. H., and Y. H. Kim. 2004. Escherichia coli O157:H7 adherence to HEp-2 cells is implicated with curli expression and outer membrane integrity. J. Vet. Sci. 5119-124. [PubMed]
16. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227680-685. [PubMed]
17. Manning, S. D., A. S. Motiwala, A. C. Springman, W. Qi, D. W. Lacher, L. M. Ouellette, J. M. Mladonicky, P. Somsel, J. T. Rudrik, S. E. Dietrich, W. Zhang, B. Swaminathan, D. Alland, and T. S. Whittam. 2008. Variation in virulence among clades of Escherichia coli O157:H7 associated with disease outbreaks. Proc. Natl. Acad. Sci. USA 1054868-4873. [PMC free article] [PubMed]
18. Mellies, J. L., A. M. S. Barron, and A. M. Carmona. 2007. Enteropathogenic and enterohemorrhagic Escherichia coli virulence gene regulation. Infect. Immun. 754199-4210. [PMC free article] [PubMed]
19. Mellies, J. L., S. J. Elliott, V. Sperandio, M. S. Donnenberg, and J. B. Kaper. 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. 33296-306. [PubMed]
20. Nataro, J. P., and J. B. Kaper. 1998. Diarrheagenic Escherichia coli. Clin. Microbiol. Rev. 11142-201. [PMC free article] [PubMed]
21. Ogierman, M. A., A. W. Paton, and J. C. Paton. 2000. Up-regulation of both intimin and eae-independent adherence of Shiga toxigenic Escherichia coli O157 by ler and phenotypic impact of a naturally occurring ler mutation. Infect. Immun. 685344-5353. [PMC free article] [PubMed]
22. Olsén, A., A. Arnqvist, M. Hammar, and S. Normark. 1993. Environmental regulation of curli production in Escherichia coli. Infect. Agents Dis. 2272-274. [PubMed]
23. Oshima, T., S. Ishikawa, K. Kurokawa, H. Aiba, and N. Ogasawara. 2006. Escherichia coli histone-like protein H-NS preferentially binds to horizontally acquired DNA in association with RNA polymerase. DNA Res. 13141-153. [PubMed]
24. Reading, N. C., and V. Sperandio. 2006. Quorum sensing: the many languages of bacteria. FEMS Microbiol. Lett. 2541-11. [PubMed]
25. Riley, L. W., R. S. Remis, S. D. Helgerson, H. B. McGee, J. G. Wells, B. R. Davis, R. J. Hebert, E. S. Olcott, L. M. Johnson, N. T. Hargrett, P. A. Blake, and M. L. Cohen. 1983. Hemorrhagic colitis associated with a rare Escherichia coli serotype. N. Engl. J. Med. 308681-685. [PubMed]
26. Roe, A. J., C. Currie, D. G. Smith, and D. L. Gally. 2001. Analysis of type 1 fimbriae expression in verotoxigenic Escherichia coli: a comparison between serotypes O157 and O26. Microbiology 147145-152. [PubMed]
27. Römling, U., Z. Bian, M. Hammar, W. D. Sierralta, and S. Normark. 1998. Curli fibers are highly conserved between Salmonella typhimurium and Escherichia coli with respect to operon structure and regulation. J. Bacteriol. 180722-731. [PMC free article] [PubMed]
28. Russell, R. M., F. C. Sharp, D. A. Rasko, and V. Sperandio. 2007. QseA and GrlR/GrlA regulation of the locus of enterocyte effacement genes in enterohemorrhagic Escherichia coli. J. Bacteriol. 1895387-5392. [PMC free article] [PubMed]
29. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
30. Scott, M. E., A. R. Melton-Celsa, and A. D. O'Brien. 2003. Mutations in hns reduce the adherence of Shiga toxin-producing E. coli 091:H21 strain B2F1 to human colonic epithelial cells and increase the production of hemolysin. Microb. Pathog. 34155-159. [PubMed]
31. Sharma, V. K., and R. L. Zuerner. 2004. Role of hha and ler in transcriptional regulation of the esp operon of enterohemorrhagic Escherichia coli O157:H7. J. Bacteriol. 1867290-7301. [PMC free article] [PubMed]
32. Sharp, F. C., and V. Sperandio. 2007. QseA directly activates transcription of LEE1 in enterohemorrhagic Escherichia coli. Infect. Immun. 752432-2440. [PMC free article] [PubMed]
33. Simon, R., U. Priefer, and A. Pühler. 1983. A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in Gram negative bacteria. Bio/Technology 1784-791.
34. Sivapalasingam, S., C. R. Friedman, L. Cohen, and R. V. Tauxe. 2004. Fresh produce: a growing cause of outbreaks of foodborne illness in the United States, 1973 through 1997. J. Food Prot. 672342-2353. [PubMed]
35. Sperandio, V., C. C. Li, and J. B. Kaper. 2002. Quorum-sensing Escherichia coli regulator A: a regulator of the LysR family involved in the regulation of the locus of enterocyte effacement pathogenicity island in enterohemorrhagic E. coli. Infect. Immun. 703085-3093. [PMC free article] [PubMed]
36. Tatsuno, I., R. Mundy, G. Frankel, Y. Chong, A. D. Phillips, A. G. Torres, and J. B. Kaper. 2006. The lpf gene cluster for long polar fimbriae is not involved in adherence of enteropathogenic Escherichia coli or virulence of Citrobacter rodentium. Infect. Immun. 74265-272. [PMC free article] [PubMed]
37. Torres, A. G., J. A. Giron, N. T. Perna, V. Burland, F. R. Blattner, F. Avelino-Flores, and J. B. Kaper. 2002. Identification and characterization of lpfABCCDE, a fimbrial operon of enterohemorrhagic Escherichia coli O157:H7. Infect. Immun. 705416-5427. [PMC free article] [PubMed]
38. Torres, A. G., K. J. Kanack, C. B. Tutt, V. Popov, and J. B. Kaper. 2004. Characterization of the second long polar (LP) fimbriae of Escherichia coli O157:H7 and distribution of LP fimbriae in other pathogenic E. coli strains. FEMS Microbiol. Lett. 238333-344. [PubMed]
39. Torres, A. G., and J. B. Kaper. 2001. PAIs of intestinal E. coli, p. 31-48. In J. Hacker and J. B. Kaper (ed.), Pathogenicity islands (PAIs) and the evolution of pathogenic microbes, vol. 1. Springer-Verlag, Berlin, Germany.
40. Torres, A. G., G. N. López-Sánchez, L. Milflores-Flores, S. D. Patel, M. Rojas-López, C. F. Martínez de la Peña, M. M. P. Arenas-Hernández, and Y. Martínez-Laguna. 2007. Ler and H-NS, regulators controlling expression of the long polar fimbriae of Escherichia coli O157:H7. J. Bacteriol. 1895916-5928. [PMC free article] [PubMed]
41. Torres, A. G., L. Milflores-Flores, J. G. Garcia-Gallegos, S. D. Patel, A. Best, R. M. La Ragione, Y. Martinez-Laguna, and M. J. Woodward. 2007. Environmental regulation and colonization attributes of the long polar fimbriae of Escherichia coli O157:H7. Int. J. Med. Microbiol. 297177-185. [PubMed]
42. Torres, A. G., X. Zhou, and J. B. Kaper. 2005. Adherence of diarrheagenic Escherichia coli strains to epithelial cells. Infect. Immun. 7318-29. [PMC free article] [PubMed]
43. Uhlich, G. A., J. E. Keen, and R. O. Elder. 2002. Variations in the csgD promoter of Escherichia coli O157:H7 associated with increased virulence in mice and increased invasion of HEp-2 cells. Infect. Immun. 70395-399. [PMC free article] [PubMed]
44. Woodall, L. A., P. W. Russell, S. L. Harris, and P. E. Orndorff. 1993. Rapid, synchronous, and stable induction of type 1 piliation in Escherichia coli by using a chromosomal lacUV5 promoter. J. Bacteriol. 1752770-2778. [PMC free article] [PubMed]

Articles from Infection and Immunity are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...