Analysis of type 1 fimbriae expression in verotoxigenic Escherichia coli: a comparison between serotypes O157 and O26

Andrew J. Roe1, Carol Currie1, David G. E. Smith1 and David L. Gally1

Zoonotic and Animal Pathogens Research Laboratory, Medical Microbiology, Teviot Place, University of Edinburgh, Edinburgh EH8 9AG, UK1

Author for correspondence: David Gally. Tel: +44 131 651 1342. Fax: +44 131 650 6531. e-mail: d.gally{at}ed.ac.uk


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Previous research has shown that verotoxin-producing Escherichia coli (VTEC) O157 strains appear unable to express type 1 fimbriae although other serotypes such as O26 and O118 can. This study has investigated the molecular basis of this difference. The study confirmed the presence of a 16 bp deletion within the regulatory region of fimA (fim switch) in 63 VTEC O157 strains but not in other VTEC serotypes tested. The fim switch was shown to be detectable only in the phase off orientation in VTEC O157, but detection of the switch in the phase on orientation correlated with the degree of mannose-sensitive yeast agglutination in VTEC O26. Repair of the 16 bp deletion in the VTEC O157 fim switch region restored phase-variable expression of fimA in a permissive background. Non-O157 VTEC, especially O26 and O118, can be pathogenic in cattle; the role of type 1 fimbriae in this and colonization is discussed.

Keywords: Escherichia coli O157, VTEC, type 1 fimbriae, phase variation, adherence

Abbreviations: A/E, attaching and effacing; IHF, integration host factor; VTEC, verotoxigenic E. coli


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Ruminants, particularly cattle, act as the main reservoir for verotoxigenic Escherichia coli (VTEC) (Beutin, 1999 ; Borczyk et al., 1987 ; Chart, 1998 ; Mead & Griffin, 1998 ; Ørskov et al., 1987 ). E. coli O157 is the most common serotype associated with VTEC infections in the UK and the USA, other serotypes such as O26 and O118 being prevalent in other countries (Beutin, 1999 ; Mead & Griffin, 1998 ). Except in the case of neonatal cattle (DeanNystrom et al., 1997 ), VTEC O157 does not cause disease in the bovine host, but diarrhoea and haemorrhagic colitis have been associated with other VTEC serotypes, especially O26 (Pearson et al., 1999 ; Stordeur et al., 2000 ). The very fact that VTEC O157 does not cause disease in ruminants may have helped its successful spread to 20–30% of herds in England and Scotland as well as 40–70% of herds in the USA in recent years (Elder et al., 2000 ; Gansheroff & O’Brien, 2000 ; Synge, 1999 , 2000 ).

In humans, the serious sequelae of haemorrhagic colitis and haemolytic uraemic syndrome stem from the activity of the shiga-like toxins, and include gastrointestinal tract colonization and damage involving intimate attachment and attaching and effacing (A/E) lesion formation (Frankel et al., 1998 ; Kaper, 1998 ). Intimin is the key candidate initiating bacterial binding (DeVinney et al., 1999 ; Kenny, 1999 ) in the human host, possibly at the Peyer’s patches of the small intestine (Phillips & Frankel, 2000 ).

There is a great deal of interest in understanding the successful dissemination of VTEC O157 among beef and dairy cattle and the reasons why it does not cause colitis and diarrheoa in this host. A logical starting point for such investigations is the carriage and expression of established bacterial adhesins, although to date there is no clear evidence that E. coli O157 colonizes the bovine gastrointestinal tract by binding to the epithelium. E. coli O157 does not carry genes for many of the established adhesins that could play a role in colonizing the bovine host, such as K99 and F41 (our unpublished data), but it does carry the genes for type 1 fimbriae. Type 1 fimbriae are the most common adhesin produced by E. coli and mediate adherence to mannose-containing glycoproteins found on the surfaces of many eukaryotic cells. Type 1 fimbriae have been shown to play an important role in the colonization of E. coli in the human urinary tract (Connell et al., 1996 ; Donnenberg & Welch, 1996 ; Mulvey et al., 1998 ) and, from their prevalence, it is assumed that they are important in the faecal–oral cycle, for binding within the gastointestinal tract and/or for transfer between hosts (Orndorff & Bloch, 1990 ). For cattle, it has been established that type 1 fimbriae bind well to both immature and mature (keratinized) rumen epithelium in vitro (Galfi et al., 1998 ).

While the majority of VTEC O157 strains do not appear to be able to express type 1 fimbriae, other VTEC serotypes, notably O26 and O118, can express this adhesin (Enami et al., 1999 ). Initial adherence by factors such as type 1 fimbriae may make subsequent intimate adherence and A/E lesion formation more likely to occur (Mack et al., 1999 ). Expression of type 1 fimbriae is phase variable, being dependent on an invertible DNA element (fim switch) that contains the promoter for fimA, the gene for the main fimbrial subunit (Abraham et al., 1985 ; Klemm, 1984 ; Olsen & Klemm, 1994 ). In the off orientation, type 1 fimbriae cannot be produced. Inversion of the fim switch requires the activity of two recombinases, FimB and FimE, as well as numerous co-factors (Gally et al., 1996 ; Klemm, 1986 ; McClain et al., 1991 ).

The research presented follows on from the findings of Li et al. (1997) , who identified a 16 bp deletion in E. coli O157 within the regulatory switch region that controls expression of type 1 fimbriae. They designed a primer that bridged this deletion as an effective probe for E. coli O157 strains. We have confirmed the presence of this deletion in 63 human and cattle E. coli O157 strains and have demonstrated that the phase switch controlling type 1 fimbrial expression is detectable only in the off orientation in VTEC O157. Repair of the 16 bp deletion restores phase-variable expression of fimA.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Media and growth conditions.
The bacterial strains and plasmids used in this study are listed in Table 1. The growth media used in this work were L-broth (per litre, 5 g NaCl, 5 g yeast extract and 10 g tryptone; Oxoid), L-agar (1·5% agar) and 6% sucrose agar (Blomfield et al., 1991 ), which was used to select for recombinant bacteria. Sucrose agar consisted of L-agar with the addition of 6% sucrose in the place of NaCl. Indicator plates of minimal MOPS supplemented with X-Gal were used as described previously (Gally et al., 1994 ).


View this table:
[in this window]
[in a new window]
 
Table 1. Strains, plasmids and primers used in this study

 
Strains.
VTEC strains were obtained from various sources. ZAP1 (O157:H7, VT1- VT2+) was provided by Dr Fiona Thompson-Carter of the National Reference Laboratories, Aberdeen, and was isolated from a patient infected during the Redhouse Dairy (Lothian, Scotland) outbreak of 1994. The strains used in the fim-region screen (Fig. 1) were also provided by Dr Fiona Thompson-Carter and represent a cross-section of human-disease serotypes (O26, O111, O86, O128 and O157) and bovine isolates of different phage types (2, 14, 21/28) and PFGE patterns. ZAP26 (strain 413/89-1, O26:H- VTEC) was a bovine isolate provided by Dr Trinad Chakraborty (Djafari et al., 1997 ). MPRL4269 (ZAP77) and MPRL4270 (ZAP78) are both O26 human-disease-associated strains, shown to be VT1-/VT2- and VT1+/VT2-, respectively.



View larger version (91K):
[in this window]
[in a new window]
 
Fig. 1. (a) PCR amplification of the fim region of VTEC and non-VTEC strains. E. coli O157 strains can be seen to have a smaller product when analysed using primers C1 and C2 (described in Table 1). The lanes contain the following: 100 bp markers (bright band of 600 bp) (lane 1); E. coli K-12 MG1655 (lane 2); ZAP28, ZAP32, ZAP36, ZAP40 (lanes 3–6), all O157 strains of differing phage types; ZAP21 (lane 7, O26), ZAP78 (lane 8, O26), ZAP72 (lane 9, O111), ZAP73 (lane 10, O111), ZAP74 (lane 11, O128), ZAP76 (lane 12, O86); ZAP1, ZAP48, ZAP52, ZAP56 (lanes, 13–16, further phage types of O157); a blank control (lane 17); a repeat of the 100 bp ladder (lane 18). Further details of the strains are given in Table 1. (b) PCR analysis of the fim deletion region using O157-specific primers (see text and Table 1). The same strains were analysed as described for (a).

 
Agglutination assay.
A single colony of the strain to be tested was used to inoculate 5 ml LB broth, cultured at 37 °C without shaking, then subcultured twice for overnight static growth. Agglutination was assayed on glass slides by mixing 15 µl bacterial culture with an equal volume of bakers’ yeast suspension (10 mg ml-1). Mannose inhibition of agglutination was confirmed using 3% {alpha}-D-mannose in the yeast suspension.

Switch-orientation assay.
Primers fim5' and fim3' (Table 1) were used to amplify a 604 bp region that incorporates the invertible fimA promoter element. The resulting PCR product was digested with HinfI, and the products separated using a 6% polyacrylamide Tris/borate EDTA gel run at 100 V for 2 h. The products were visualized after staining with ethidium bromide using Flowgen gel-analysis software and hardware.

Cloning of fim switch regions into pDG19 and allelic exchange.
The fim switch region from the strains was amplified using primers C1 and C2 (Table 1). Primer C1 incorporates a SphI site and C2 incorporates a BamHI site at its 5'-terminus, allowing the amplified fragment to be ligated into pDG19 (Table 1). Allelic exchange of the cloned switch region into BGEC144 (Table 1) was performed as described previously (Blomfield et al., 1991 ; Leathart & Gally, 1998 ).

‘Repair’ of the O157 fim switch.
The 16 bp of K-12 sequence absent in E. coli O157 ZAP1 was introduced into the plasmid carrying the ZAP1 fim switch region (pAJR1; Table 1) by using specific primers. Two primers were designed (rep1 and rep2; Table 1), each of which had an 8 bp 5'-terminus that complemented the K-12 region absent from the sequence of ZAP1. In conjunction with primers C1 and C2 (Table 1), amplification using Pwo polymerase (Roche) produced two blunt-ended products of 317 bp and 424 bp. These products were cleaned using a Qiagen PCR clean-up kit and ligated using standard conditions; 1 µl of this ligation reaction was then used as a template for PCR using primers C1 and C2 (Table 1) to produce a fim switch region that was identical to that of ZAP1 (apart from the ‘repair’ of the 16 bp deletion).

Switching-frequency assay.
Strains were plated on minimal glucose X-Gal agar at 37 °C. Using a sterile cork-borer, six ‘off’ (white) colonies were excised from the plate, diluted in PBS and plated on minimal glucose X-Gal agar at 42 °C (300–1000 c.f.u.). Plates were counted for white and blue colonies, and the frequency of off-to-on inversion was calculated as described previously (Gally et al., 1993 ).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Detection of fim switch deletion in VTEC isolates
To confirm the presence of the 16 bp deletion in E. coli O157, a number of human-outbreak and cattle isolates were tested. The isolates covered a range of phage types and a variety of PFGE patterns (Fiona Thomson-Carter, Aberdeen E. coli Reference Laboratory). All of the human (27) and bovine strains (36) tested possessed the deletion, as demonstrated by a smaller PCR product generated across the fim switch (Fig. 1a). A selection of these were confirmed by use of the deletion-detection primer designed by Li et al. (1997) (Fig. 1b, Table 1). A small number of shiga-like-toxin-positive E. coli O26 (3), O111 (4), O86 (1) and O128 (1) isolates were tested and, as anticipated, were shown not to contain the deletion (Fig. 1a, b). These data confirm the specificity of this deletion to E. coli O157 and the potential use of the deletion PCR primer as a molecular probe (Li et al., 1997 ).

Genotypic and phenotypic analyses of type 1 fimbrial expression
To confirm that E. coli O157 does not produce type 1 fimbriae whereas E. coli O26 can, a simple slide agglutination test was performed on one human E. coli O157 outbreak strain and three E. coli O26 strains (two human and one bovine). The O157 strain was negative for agglutination, whereas the O26 strains varied in their ability to agglutinate, but all three were positive (Fig. 2, bottom panel).



View larger version (44K):
[in this window]
[in a new window]
 
Fig. 2. PCR and restriction digestion of the fim switch from VTEC isolates indicates the on–off status within the bacterial population. The product was run either undigested or after digestion with HinfI. Lanes: 1, 100 bp marker; 2, ZAP26 uncut; 3, ZAP26 cut; 4, ZAP77, uncut; 5, ZAP77, cut; 6, ZAP78 uncut; 7, ZAP78 cut; 8, ZAP1 uncut; 9, ZAP1 cut. The results of the yeast agglutination assay are summarized below the gel image.

 
To determine whether agglutination ability correlated with the phase state of the fim switch in each culture, a PCR and restriction digestion of the fim switch was carried out as described in Methods. As shown in Fig. 2, the higher the proportion of the bacterial population with the fim switch in the on orientation, the better the ability to agglutinate. Moreover, the O157 strain appeared to be locked in the off orientation (>99% off). Without phase switching from the off to the on phase orientation, E. coli O157 will be unable to express type 1 fimbriae.

Cloning and analysis of fim switch regions
After differences in the phase state of the fim switch between VTEC isolates had been established, the contribution of the fim switch region alone to this behaviour was determined. To do this, isogenic fimA–lacZYA strains containing the three O26 switch regions and one O157 switch region were created, allowing quantification of switching frequencies. This approach uses allelic exchange to replace the fim switch of E. coli K-12 MG1655 with the relevant VTEC switch region adjacent to a fimA–lacZYA transcriptional fusion (Leathart & Gally, 1998 ; see Methods). This allows switching frequencies to be measured with the same recombinase and host factors present. SR1 resulted from the exchange of the E. coli O157 switch region into BGEC144 (fimB+E-). The frequency of FimB-promoted off-to-on switching in E. coli K-12 containing its wild-type switch was measured as 3x10-3 per cell per generation (Fig. 3). However, SR1 gave no detectable switching (<1x10-5 per cell per generation; Fig. 3). SR2-4 resulted from exchange of the three E. coli O26 regions into BGEC144 and all three strains had frequencies similar to that of E. coli K-12 (2x10-3 to 3·5x10-3; Fig. 3). These data show that the E. coli O157 switch region is unable to invert in a permissive background, proving that sequence changes in this switch region prevent off-to-on inversion of the switch to allow expression of fimA. Interestingly, while there are clear differences in the degree of expression of type 1 fimbriae (and switch orientation) between the three O26 strains (Fig. 2), these differences were not transferred with the respective switch regions and cannot be attributed to sequence variation in these regions.



View larger version (11K):
[in this window]
[in a new window]
 
Fig. 3. Switching frequencies of strains containing the fim switch region from VTEC isolates (Table 1). SR1–4 contain the switches from ZAP1, MPRL4269, MPRL4270 and ZAP26, respectively exchanged into BGEC144. SR5 is the ZAP1 fim switch region containing a 13 bp repair exchanged into BGEC144, whereas SR6 contains the ZAP1 fim switch region with the full 16 bp repair exchanged into BGEC144 (Table 1). Switching frequencies were measured as described in Methods.

 
All four VTEC switch regions were sequenced (Fig. 4). While numerous single base-pair variations were evident, the previously characterized 16 bp deletion in E. coli O157 appeared to be the most likely change preventing fim switching.



View larger version (40K):
[in this window]
[in a new window]
 
Fig. 4. Sequences of fim switch regions for the VTEC isolates used in this study and for E. coli K-12 MG1655. The diagram shows the IHF-binding site located within the invertible element (a) (Blomfield et al., 1997 ) and the 16 bp deletion region (b) identified by Li et al. (1997) .

 
Repair of the 16 bp deletion from E. coli O157
To ascertain whether this deletion is responsible for the failure of the E. coli O157 switch region to undergo inversion from off to on, a repair was engineered by site-specific PCR and cloning methods, followed by allelic exchange of the repaired switch into BGEC144 (SR6). This construct was confirmed by sequencing (Fig. 4). During this construction, another clone was isolated that was found to contain only 13 of the 16 bp absent in the E. coli O157 sequence (Fig. 4). This partial repair was also exchanged into BGEC144 to create SR5. FimB off-to-on switching was measured in both SR5 and SR6 (Fig. 3). It was clear that repair of the full 16 bp restored a wild-type switching frequency (2·5x10-3). The incomplete repair was able to invert from off to on but at a 10-fold-reduced frequency (2·8x10-4). In summary, these data confirm that the 16 bp deletion in the E. coli O157 switch region prevents phase-variable expression of fimA, the main fimbrial subunit gene.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The data presented in this work confirm that the majority of VTEC O157 strains contain a 16 bp deletion in the fim switch that has so far not been detected in any other E. coli serotype. The fim switch from one E. coli O157 human-outbreak strain (ZAP1) was shown to remain in the off orientation despite culture in conditions that favour expression of type 1 fimbriae. In contrast, the three E. coli O26 strains examined showed variation in their ability to produce type 1 fimbriae, and this variation correlated well with the proportion of bacteria containing the fim switch in the on orientation. The switch regions from all four strains were cloned and exchanged into the E. coli K-12 chromosome to replace the wild-type switch and create fimA–lacZYA transcriptional fusions. The switch region from ZAP1 failed to promote expression of the fusion and the switch was still locked in the off orientation. This proves that sequence variation within the ZAP1 fim switch region prevents inversion and expression of fimA. All three O26 regions were able to produce phase-variable expression of fimA at rates equivalent (within twofold) to that of E. coli K-12.

The four switch regions were sequenced and the 16 bp deletion was identified as the major difference between the O26 and O157 strains. This deletion was repaired in the E. coli K-12 fusion background and this repaired strain was then able to phase-vary fimA expression at frequencies equivalent to both E. coli K-12 and the O26 switch replacement strains. While other changes within the fim region of E. coli O157 may prevent type 1 fimbrial expression, this single deletion within the regulatory region is sufficient to nullify expression and is the likely reason why E. coli O157 cannot produce type 1 fimbriae. The reason why the 16 bp deletion prevents phase variation is unclear. The regulation of type 1 fimbrial phase variation has been studied in detail, with most binding sites for required co-factors being identified within the fim switch (Blomfield et al., 1997 ; Gally et al., 1994 , 1996 ; Olsen et al., 1998 ). The 16 bp deletion lies at the edge of a known integration host factor (IHF)-binding site (Fig. 4) but does not involve the majority of bases considered important for IHF binding (Blomfield et al., 1997 ). However, it is possible that the lack of fimA expression is caused by a failure of IHF to bind and bend the DNA at this site. Alternatively, the required DNA conformation for inversion may be strict and may be affected by the 16 bp deletion. A consistent distance of 314±1 bp between the 9 bp inverted repeats was observed in urinary tract E. coli isolates (Leathart & Gally, 1998 ). For recombination the inverted repeats need to: (1) be brought together by DNA bending proteins (IHF and Lrp) and (2) be in phase to allow appropriate recombinase binding and activity. The 16 bp deletion could interfere with either or both of these requirements. In the course of repairing the 16 bp deletion, a 3 bp deletion was created at the same site (Fig. 4); the strain containing this deletion was able to phase vary but at a 10-fold reduced rate compared to the full repair and E. coli K-12. This 3 bp deletion is not within the IHF-binding site and supports the premise that the 16 bp deletion prevents recombination by interfering with strict spatial requirements.

Clear differences were observed in the degree of expression of type 1 fimbriae (and switch orientation) between the three O26 strains tested. However, all three imported switch regions had similar off-to-on switching frequencies. Potentially, other genetic factors in the relevant host strain are responsible for the differences in expression levels. This could include differences in recombinase expression or activity. However, it is possible that, as shown for urinary tract isolates, regulators of other adhesins or virulence determinants may cross-talk with type 1 fimbriae to alter expression levels (Xia et al., 2000 ). Expression of type III secretion organelles and intimate attachment may not be compatible with the presence of fimbrial adhesions on the bacterial surface. The potential for coordinate expression between type 1 fimbriae and type III secretion in enterohaemorrhagic E. coli (EHEC) serotypes such as O26 remains to be investigated. Certainly, O157 EHEC infections do not seem to need type 1 fimbriae but, as with rabbit diarrhoeagenic E. coli (RDEC), O26 strains may be more likely to form A/E lesions once the bacterium is brought into close proximity to the host-cell epithelium by a fimbrial adhesin. Our preliminary data have demonstated that the level of type 1 expression does correlate with ability of E. coli to bind to bovine colonic epithelium. Therefore, expression of type 1 fimbriae by E. coli O26/O118 could enhance colonization in the bovine colon as well as that in the rumen (Galfi et al., 1998 ). Obviously, host factors such as competitive flora and mucus secretion may also modulate this activity. Other factors that could explain VTEC O26/O118 virulence in the bovine host (in comparison to VTEC O157), including the level and activity of secreted proteins, are currently being investigated.


   ACKNOWLEDGEMENTS
 
This research was supported by a Ministry of Agriculture Fisheries and Food Research Fellowship to D.G. and D.G.E.S.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Abraham, J. M., Freitag, C. S., Clements, J. R. & Eisenstein, B. I.(1985). An invertible element of DNA controls phase variation of type 1 fimbriae of Escherichia coli. Proc Natl Acad Sci USA 82, 5724-5727.[Abstract]

Beutin, L.(1999). Escherichia coli O157 and other types of verocytotoxigenic E. coli (VTEC) isolated from humans, animals and food in Germany. In Escherichia coli O157 in Farm Animals , pp. 121-146. Edited by C. S. Stewart & H. J. Flint. Wallingford, UK:CABI Publishing.

Blomfield, I. C., McClain, M. S. & Eisenstein, B. I.(1991). Type 1 fimbriae mutants of Escherichia coli K-12 – characterization of recognized afimbriate strains and construction of new fim deletion mutants. Mol Microbiol 5, 1439-1445.[Medline]

Blomfield, I. C., Kulasekara, D. H. & Eisenstein, B. I.(1997). Integration host factor stimulates both FimB- and FimE-mediated site-specific DNA inversion that controls phase variation of type 1 fimbriae expression in Escherichia coli. Mol Microbiol 23, 705-717.[Medline]

Borczyk, A. A., Karmali, M. A., Lior, H. & Duncan, L. M. C. (1987). Bovine reservoir for verotoxin-producing Escherichia coli O157:H7. Lancet 1, 98.

Chart, H.(1998). Are all infections with Escherichia coli O157 associated with cattle? Lancet 352, 1005.[Medline]

Connell, H., Agace, W., Klemm, P., Schembri, M., Marild, S. & Svanborg, C.(1996). Type 1 fimbrial expression enhances Escherichia coli virulence for the urinary tract. Proc Natl Acad Sci USA 93, 9827-9832.[Abstract/Free Full Text]

DeanNystrom, E. A., Bosworth, B. T. & Moon, H. W.(1997). Pathogenesis of O157:H7 Escherichia coli infection in neonatal calves. Adv Exp Med Biol 412, 47-51.[Medline]

DeVinney, R., Stein, M., Reinscheid, D., Abe, A., Ruschkowski, S. & 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, 2389-2398.[Abstract/Free Full Text]

Djafari, S., Ebel, F., Deibel, C., Kramer, S., Hudel, M. & Chakraborty, T.(1997). Characterization of an exported protease from Shiga toxin-producing Escherichia coli. Mol Microbiol 25, 771-784.[Medline]

Donnenberg, M. S. & Welch, R. A. (1996). Virulence Determinants of Uropathogenic Escherichia coli. Washington, DC: American Society for Microbiology.

Elder, R. O., Keen, J. E., Siragusa, G. R., Barkocy-Gallagher, G. A., Koohmaraie, M. & Laegreid, W. W.(2000). Correlation of enterohemorrhagic Escherichia coli O157 prevalence in faeces, hides, and carcasses of beef cattle during processing. Proc Natl Acad Sci USA 97, 2999-3003.[Abstract/Free Full Text]

Enami, M., Nakasone, N., Honma, Y., Kakinohana, S., Kudaka, J. & Iwanaga, M.(1999). Expression of type I pili is abolished in verotoxin-producing Escherichia coli O157. FEMS Microbiol Lett 179, 467-472.[Medline]

Frankel, G., Phillips, A. D., Rosenshine, I., Dougan, G., Kaper, J. B. & Knutton, S.(1998). Enteropathogenic and enterohaemorrhagic Escherichia coli: more subversive elements. Mol Microbiol 30, 911-921.[Medline]

Galfi, P., Neogrady, S., Semjen, G., Bardocz, S. & Pusztai, A.(1998). Attachment of different Escherichia coli strains to cultured rumen epithelial cells. Vet Microbiol 61, 191-197.[Medline]

Gally, D. L., Bogan, J. A., Eisenstein, B. I. & Blomfield, I. C.(1993). Environmental regulation of the fim switch controlling type-1 fimbrial phase variation in Escherichia coli K-12 – effects of temperature and media. J Bacteriol 175, 6186-6193.[Abstract]

Gally, D. L., Rucker, T. J. & Blomfield, I. C.(1994). The leucine-responsive regulatory protein binds to the fim switch to control phase variation of type 1 fimbrial expression in Escherichia coli K-12. J Bacteriol 176, 5665-5672.[Abstract]

Gally, D. L., Leathart, J. & Blomfield, I. C.(1996). Interaction of fimB and fimE with the fim switch that controls the phase variation of type 1 fimbriae in Escherichia coli K-12. Mol Microbiol 21, 725-738.[Medline]

Gansheroff, L. J. & O’Brien, A. D.(2000). Escherichia coli O157:H7 in beef cattle presented for slaughter in the US: higher prevalence rates than previously estimated. Proc Natl Acad Sci USA 97, 2959-2961.[Free Full Text]

Guyer, M. S., Reed, R. R., Steitz, J. A. & Low, K. B.(1981). Identification of a sex-factor-affinity site in E. coli as a gamma delta. Cold Spring Harbor Symp Quant Biol 45, 135-140.[Medline]

Kaper, J. B.(1998). Enterohemorrhagic Escherichia coli. Curr Opin Microbiol 1, 103-108.[Medline]

Kenny, B.(1999). Phosphorylation of tyrosine 474 of the enteropathogenic Escherichia coli (EPEC) Tir receptor molecule is essential for actin nucleating activity and is preceded by additional host modifications. Mol Microbiol 31, 1229-1241.[Medline]

Klemm, P.(1984). The fimA gene encoding the type-1 fimbrial subunit of Escherichia coli – nucleotide-sequence and primary structure of the protein. Eur J Biochem 143, 395-399.[Abstract]

Klemm, P.(1986). 2 regulatory fim genes, fimB and fimE, control the phase variation of type 1 fimbriae in Escherichia coli. EMBO J 5, 1389-1393.[Abstract]

Leathart, J. B. S. & Gally, D. L.(1998). Regulation of type 1 fimbrial expression in uropathogenic Escherichia coli: heterogeneity of expression through sequence changes in the fim switch region. Mol Microbiol 28, 371-381.[Medline]

Li, B. G., Koch, W. H. & Cebula, T. A.(1997). Detection and characterization of the fimA gene of Escherichia coli O157:H7. Mol Cell Probes 11, 397-406.[Medline]

McClain, M. S., Blomfield, I. C. & Eisenstein, B. I.(1991). Roles of fimB and fimE in site-specific DNA inversion associated with phase variation of type 1 fimbriae in Escherichia coli. J Bacteriol 173, 5308-5314.[Medline]

Mack, D. R., Michail, S., Wei, S., McDougall, L. & Hollingsworth, M. A.(1999). Probiotics inhibit enteropathogenic E. coli adherence in vitro by inducing intestinal mucin gene expression. Am J Physiol–Gastrointest Liver Physiol 39, G941-G950.

Mead, P. S. & Griffin, P. M.(1998). Escherichia coli O157:H7. Lancet 352, 1207-1212.[Medline]

Mulvey, M. A., LopezBoado, Y. S., Wilson, C. L., Roth, R., Parks, W. C., Heuser, J. & Hultgren, S. J.(1998). Induction and evasion of host defenses by type 1-piliated uropathogenic Escherichia coli. Science 282, 1494-1497.[Abstract/Free Full Text]

Olsen, P. B. & Klemm, P.(1994). Localization of promoters in the fim gene cluster and the effect of H-NS on the transcription of fimB and fimE. FEMS Microbiol Lett 116, 95-100.[Medline]

Olsen, P. B., Schembri, M. A., Gally, D. L. & Klemm, P.(1998). Differential temperature modulation by H-NS of the fimB and fimE recombinase genes which control the orientation of the type 1 fimbrial phase switch. FEMS Microbiol Lett 162, 17-23.[Medline]

Orndorff, P. E. & Bloch, C. A.(1990). The role of type 1 pili in the pathogenesis of Escherichia coli infections – a short review and some new ideas. Microb Pathog 9, 75-79.[Medline]

Ørskov, F., Ørskov, I. & Villar, J. A. (1987). Cattle as a reservoir of verotoxin-producing Escherichia coli O157:H7. Lancet 2, 276.[Medline]

Pearson, G. R., Bazeley, K. J., Jones, J. R., Gunning, R. F., Green, M. J., Cookson, A. & Woodward, M. J.(1999). Attaching and effacing lesions in the large intestine of an eight-month-old heifer associated with Escherichia coli O26 infection in a group of animals with dysentery. Vet Rec 145, 370-373.[Medline]

Phillips, A. & Frankel, G.(2000). Intimin-mediated tissue specificity in enteropathogenic Escherichia coli interaction with woman intestinal organ culture. J Infect Dis 181, 1496-1500.[Medline]

Stordeur, P., China, B., Charlier, G., Roels, S. & Mainil, J.(2000). Clinical signs, reproduction of attaching/effacing lesions, and enterocyte invasion after oral inoculation of an O118 enterohaemorrhagic Escherichia coli in neonatal calves. Microbes Infect 2, 17-24.[Medline]

Synge, B. A.(1999). Animal studies in Scotland. In Escherichia coli O157 in Farm Animals , pp. 91-98. Edited by C. S. Stewart & H. J. Flint. Wallingford, UK:CABI Publishing.

Synge, B. A.(2000). Verocytotoxin-producing Escherichia coli: a veterinary view. J Appl Microbiol 88, 31S-37S.

Xia, Y., Gally, D., Forsman-Semb, K. & Uhlin, B. E.(2000). Regulatory cross-talk between adhesin operons in Escherichia coli: inhibition of type 1 fimbriae expression by the PapB protein. EMBO J 19, 1450-1457.[Abstract/Free Full Text]

Received 9 August 2000; revised 20 September 2000; accepted 25 September 2000.