*For correspondence. E-mail firstname.lastname@example.org; Tel. (+61) 38344 8275; Fax (+61) 38344 8276.
Characterization and evidence of mobilization of the LEE pathogenicity island of rabbit-specific strains of enteropathogenic Escherichia coli
Article first published online: 18 JUN 2002
Volume 44, Issue 6, pages 1533–1550, June 2002
How to Cite
Tauschek, M., Strugnell, R. A. and Robins-Browne, R. M. (2002), Characterization and evidence of mobilization of the LEE pathogenicity island of rabbit-specific strains of enteropathogenic Escherichia coli. Molecular Microbiology, 44: 1533–1550. doi: 10.1046/j.1365-2958.2002.02968.x
- Issue published online: 18 JUN 2002
- Article first published online: 18 JUN 2002
- Top of page
- Experimental procedures
- Supplementary material
- Supporting Information
We have characterized the LEE pathogenicity islands (PAIs) of two rabbit-specific strains of enteropathogenic E. coli (REPEC), 83/39 (serotype O15:H-) and 84/110-1 (O103:H2), and have compared them to homologous loci from the human enteropathogenic and enterohaemorrhagic E. coli strains, E2348/69 and EDL933, and another REPEC strain, RDEC-1. All five PAIs contain a 34 kb core region that is highly conserved in gene order and nucleotide sequence. However, the LEE of 83/39 is significantly larger (59 540 basepairs) than those of the human strains, which are less than 44 kb, and has inserted into pheU tRNA. The regions flanking the 34 kb core of 83/39 contain homologues of two putative virulence determinants, efa1/lifA and senA. The LEE of 84/110-1 is approximately 85 kb and is located at pheV tRNA. Its core is almost identical to those of 83/39 and RDEC-1, apart from a larger espF gene, but its flanking regions contain trcA, a putative virulence determinant of EPEC. All three REPEC LEE PAIs contain a gene for an integrase, Int-phe. The LEE PAI of 84/110-1 is also flanked by short direct repeats (representing the 3′-end of pheV tRNA), suggesting that it may be unstable. To investigate this possibility, we constructed a LEE::sacB derivative of 84/110-1 and showed that the PAI was capable of spontaneous deletion. We also showed that Int-phe can mediate site-specific integration of foreign DNA at the pheU tRNA locus of E. coli DH1. Together these results indicate possible mechanisms of mobilization and integration of the LEE PAI.
- Top of page
- Experimental procedures
- Supplementary material
- Supporting Information
Enteropathogenic and enterohaemorrhagic strains of Escherichia coli are important human and veterinary pathogens (Okerman, 1987; Robins-Browne, 1987). Enteropathogenic E. coli (EPEC) causes diarrhoea in children and is a major cause of infant death in developing countries (Levine, 1987; Robins-Browne, 1987). Enterohaemorrhagic strains of E. coli (EHEC) are associated with haemorrhagic colitis and the haemolytic–uraemic syndrome, mainly in children in industrialized countries (Karmali et al., 1985; Karmali, 1989). These pathogenic varieties of E. coli are closely related to each other, and there is evidence that the most prevalent clone of EHEC, serotype O157:H7, is derived from EPEC (Feng et al., 1998). Moreover, both EPEC and EHEC appear to have evolved over time from non-pathogenic strains of E. coli by accumulating a variety of virulence factors, such as sets of virulence-related genes contained within plasmids, pathogenicity islands (PAIs), and bacteriophages (Kaper et al., 1999; Hacker and Kaper, 2000).
EPEC is distinguished from most other varieties of pathogenic E. coli by its ability to cause distinctive attaching-effacing (A/E) lesions in epithelial cells. These lesions are characterized by intimate bacterial adherence via pedestals containing polymerized actin to the membrane of epithelial cells lining the intestinal tract (Moon et al., 1983; Tzipori et al., 1989). Similar lesions are produced by most clones of EHEC, which differ from EPEC by virtue of their ability to produce phage-encoded Shiga toxins (Levine, 1987; Tzipori et al., 1989).
EPEC are host-specific and some strains are natural pathogens of animals, including laboratory animals such as rabbits. Rabbit-specific enteropathogenic E. coli (REPEC) cause severe diarrhoea in infant rabbits and produce characteristic A/E lesions in the intestinal mucosa (Robins-Browne et al., 1994a). Because REPEC show the same clinicopathological features, age and tissue specificity as the varieties of EPEC that infect humans, rabbits infected with REPEC provide an excellent model of human infections with EPEC (Robins-Browne et al., 1996), as they permit investigation of the pathogenic properties of these bacteria in a way that is not possible in humans.
In the prototypic human EPEC strain, E2348/69 (serotype O127:H6), the genes required for the production of A/E lesions are contained within a 35.6 kb PAI known as the locus of enterocyte effacement (LEE) (McDaniel and Kaper, 1997; Elliott et al., 1998). The LEE PAIs of the prototypic human EPEC and EHEC strains, E2348/69 and EDL933 (serotype O157:H7), respectively, have been completely sequenced, and that of the REPEC strain, RDEC-1 (serotype O15:H-), has been partially sequenced (Elliott et al., 1998; Perna et al., 1998; Zhu et al., 2001). The majority of genes within these PAIs are organized into five polycistronic operons: LEE1, LEE2, LEE3, tir and LEE4 (Elliott et al., 1998; Mellies et al., 1999). These genes encode (i) components of a type III secretion system (Jarvis et al., 1995); (ii) proteins secreted by this system, termed E. coli secreted proteins (Esps) (Frankel et al., 1998); (iii) a 94 kDa outer membrane protein adhesin known as intimin, which is the product of the eae gene (Jerse et al., 1990); (iv) the translocated intimin receptor, Tir, which is a type III secreted protein that becomes incorporated in the host cell membrane and acts as a receptor for intimin (Kenny et al., 1997); (v) CesT, a Tir chaperone (Abe et al., 1999; Elliott et al., 1999a); and (vi) Ler, a member of the H-NS family of transcriptional activators, which positively regulates transcription of most genes within the LEE as well as some outside LEE (Elliott et al., 2000). Genes encoding the type III secretion apparatus are highly conserved between EPEC, EHEC and REPEC, whereas intimin and the Esps that interact directly with the host cells are more variable (Frankel et al., 1998).
The LEE PAI of EHEC strain EDL933 is larger than that of E2348/69. It is 43.4 kb in size and contains a 7.5 kb putative prophage segment (designated 933 L) with homology to the P4 family of prophages (Perna et al., 1998). The LEE PAI of REPEC strain RDEC-1 contains a core region of 40 open reading frames (ORFs), all of which are shared with the LEE PAIs of the prototypic human EPEC and EHEC strains. However, the LEE PAI of RDEC-1 is larger than those of E2348/69 and EDL933 and contains the carboxyl end of a putative virulence determinant, efa1/lifA, but the exact size of the PAI is unknown as the integration site has not been reported (Zhu et al., 2001). Although there is a high degree of homology at the nucleotide level (89.3% identity) between the shared genes of the three sequenced LEE PAIs (Zhu et al., 2001), transfer of the intact LEE PAI from E. coli E2348/69 or the corresponding genes from the LEE PAI of RDEC-1 to E. coli K-12 confer the A/E phenotype on E. coli K-12 in vitro (McDaniel and Kaper, 1997; Karaolis et al., 1997), whereas this does not occur when the LEE PAI of EHEC strain EDL933 is introduced into the same recipient (Elliott et al., 1999b).
In general, PAIs are frequently linked to tRNA loci. The LEE PAIs of both E2348/69 and EDL933 have inserted into the selenocystyl (selC) tRNA locus located at min 82 on the chromosome of E. coli K-12 (McDaniel et al., 1995; Elliott et al., 1998; Perna et al., 1998). The LEE PAIs of EPEC of other serotypes, including O26:H11, O111:H2, O119:H6; O128:H2 and EHEC of serotypes O26:H11; O111:H-; O111:H8 have inserted into the pheU tRNA locus at min 94 of the E. coli K-12 chromosome (Wieler et al., 1997; Sperandio et al., 1998). Recently, the pheV tRNA locus at min 67 of the E. coli K-12 chromosome was also reported as the insertion site for the LEE PAI of an EHEC strain of serotype O103:H2 (Jores et al., 2001). The different locations of the insertion sites of the LEE in various strains of EPEC and EHEC suggest that LEE PAIs may have been acquired by different progenitors of EPEC and EHEC at various times (Wieler et al., 1997; Reid et al., 2000; Donnenberg and Whittam, 2001), or that this PAI can be mobilized and re-integrated within individual strains of E. coli.
Another feature of PAIs in general is that they are often flanked by small directly repeated (DR) DNA sequences resembling attachment (att) sites for phage integrases. Typical PAIs also carry mobility genes such as phage-related integrases, transposases or parts of insertion sequence (IS) elements, and are often unstable (Hacker et al., 1997). However, in contrast to many other PAIs, the LEE PAIs that have been characterized to date are not flanked by DRs and appear to be stable. The LEE PAI of EHEC EDL933 is flanked by a putative prophage containing a gene encoding an integrase that is a member of the family of site-specific recombinases. However, the location of the putative att sites suggests that the prophage inserted after chromosomal integration of the LEE (Perna et al., 1998). The LEE PAI of EPEC E2348/69 lacks obvious phage sequences and contains at one end remnants of IS600 and IS660 elements of Shigella subsp. (Donnenberg et al., 1997). The flanking regions of the RDEC-1 LEE PAI are unknown, but one end contains part of an IS2 element of E. coli K-12 (Zhu et al., 2001). In contrast to PAIs from uropathogenic E. coli (UPEC), Yersinia subsp., and Helicobacter pylori (Blum et al., 1994; Censini et al., 1996; Swenson et al., 1996; Buchrieser et al., 1998), mobilization or deletion of the LEE PAI has never been observed, and as with PAIs in general, the mechanism of integration is unknown.
In this study, we report the characterization of the LEE PAIs of REPEC strains 83/39 (serotype O15:H-) and 84/110-1 (serotype O103:H2). We describe the sequences of the REPEC LEE PAIs and compare them with those of the REPEC strain, RDEC-1, and the prototypic human EPEC and EHEC strains. We also report the identification of novel putative virulence determinants encoded on the LEE PAIs of these REPEC strains, and the results of studies to investigate the possible mechanism of integration and deletion of the LEE PAI of the REPEC strain, 84/110-1.
- Top of page
- Experimental procedures
- Supplementary material
- Supporting Information
Identification, sequence analysis and genetic organization of the LEE PAI of REPEC strain 83/39
Two overlapping inserts from cosmids p83/39(2) and p83/39(4) hybridized with the LEE A, C or D probes (McDaniel et al., 1995) (Fig. 1A). The sequence of the right end of the insert from p83/39(4) was identical to the E. coli K-12 sequence at min 94 of the chromosome (GenBank accession no. AE000486), indicating that p83/39(4) contained the right junction of the 83/39 LEE PAI. Further sequencing revealed that the integration site of the 83/39 LEE PAI is the pheU tRNA locus. Sequencing of the left end of the insert from cosmid p83/39(2) revealed the presence of an IS2 element of E. coli K-12. To confirm the pheU tRNA locus integration site and to obtain the left flanking region of the 83/39 LEE PAI, primers (see Experimental procedures) were designed to amplify the left junction. P54 from the cadC region, which is adjacent to the pheU tRNA locus in E. coli K-12, and P15 which is located within LEE (Fig. 1A) were used in a polymerase chain reaction (PCR) with genomic DNA from 83/39 as template. Sequence analysis of the 3.0 kb amplified fragment revealed that the left junction of the 83/39 LEE PAI is flanked by an IS3 element of E. coli K-12 (100% homology) and that there is a 255 basepair (bp) deletion in the homologous cadC locus adjacent to the IS3 element, compared with the sequence of E. coli K-12. The LEE PAI of 83/39 spanning the left and right junctions was then completely sequenced (GenBank accession no. AF453441). It comprised 59 540 bp including the IS3 element (Fig. 1A), with a G + C content of 40.5%, which is lower than the E. coli K-12 average of 50.8%, but is similar to those of the LEE PAIs of EPEC E2348/69 (38.4%), EHEC EDL933 (40.9%), and the SmaI fragment of RDEC-1 (41.3%) (Blattner et al., 1997; Elliott et al., 1998; Perna et al., 1998; Zhu et al., 2001). Analysis of the sequence revealed that the nucleotide identity between the LEE PAI of 83/39 and the 37,889-bp SmaI fragment of the RDEC-1 LEE PAI (GenBank accession no. AF200363) is >99.9% over 37 906 bp, and that the 34 nucleotide differences are either degenerate or present in non-coding regions.
The LEE PAI of 83/39 contains an internal 34 kb core region (nucleotides 3273–37309) comprising 40 ORFS, all of which correspond to genes located on the LEE PAIs of E2348/69, EDL933, and RDEC-1, and are in the same order and orientation. The 34 kb core regions of the LEE PAIs of 83/39 and RDEC-1 are nearly identical, in that they lack the enteric repetitive intergenic consensus (ERIC) element and orf3 (due to the absence of an ATG or alternative start codon), which are present in E2348/69 and EDL933. The regions flanking the 34 kb core contain 14 ORFs (Table 1). Interestingly, the right flanking region contains homologues of two putative virulence determinants of A/E strains of E. coli, namely (i) efa1/lifA, the largest gene in E. coli, which is involved with adherence of EHEC to epithelial cells (Nicholls et al., 2000)and with repression of host interleukins by EPEC (Klapproth et al., 2000); and (ii) senA, which encodes the ShET-2 enterotoxin of enteroinvasive E. coli and Shigella subsp. (Nataro et al., 1995). Adjacent to the pheU tRNA locus is an ORF that encodes a protein with >38% identity to integrases from a number of bacteriophages including P4 (GenBank accession no. CAA29379) and phi-R73 (GenBank accession no. A42465). This predicted protein also shows 40% identity to the integrase encoded by the putative prophage adjacent to the 34 kb core region of the LEE PAI of EHEC strain EDL933.
|ORF or IS||Locationa||Size (aa)||Source, similar proteinb||% Amino acid identity|
|IS3||598–897||99||K-12 transposase for IS3 (SWISSPROT P77681)||100 (99/99)|
|IS3||894–1760||288||K-12 transposase for IS3 (SWISSPROT P05822)||100 288/288|
|trunc IS2||2035–2400||121||K-12 transposase for IS2 (Swiss Prot P19776)||100 121/121|
|A||37673–38068||131||RDEC-1 putative transposase (GenBank AF200363)||100 131/131|
|E2348/69 putative transposase (GenBank AJ133705)||98 129/131|
|B||38240–47911||3223||EPEC LifA (GenBank AJ133705)||99 (3191/3223)|
|EHEC Efa1 (GenBank AF159462)||98 (3188/3223)|
|C||48566–49411||281||EHEC putative transposase (GenBank AE005528)||100 281/281|
|D||49789–50463||224||EHEC hypothetical protein (GenBank AP002563)||99 223/224|
|E||50512–51501||329||EHEC hypothetical protein (GenBank AP002563)||99 (327/329)|
|F||52109–53758||549||EIEC, Shigella spp. SenA, ShET2 enterotoxin (GenBank CAA90938)||38 209/547|
|G||53966–54193||75||EHEC hypothetical protein (GenBank AP002563)||98 (74/75)|
|H||54546–55286||246||EHEC putative transposases GenBank AP002563)||96 (244/246)|
|I||55421–57040||539||Mesorhizobium loti transposase (GenBank AP003007)||33 161/482|
|J||57037–58608||523||EDL933 LEE PAI L0015 (GenBank AF071034)||99 (504/509)|
|K||58725–59990||421||E. coli P4 integrase (SWISSPROT P08320)||54 222/411|
As REPEC strains, 83/39 and RDEC-1, are of the same serogroup (O15:H-), and because the integration site of the RDEC-1 LEE PAI had not been determined, we used PCR to determine if the LEE PAI of RDEC-1 is also inserted into the pheU tRNA locus. Using genomic DNA of RDEC-1 as template, the left junction was amplified with primers P54 and P15, and the right junction was amplified with primers P119 and P32. Sequence analysis of the ends of the 3.0 kb amplified fragment and the complete 1.7 kb fragment (GenBank accession no. AF454555) revealed that the LEE PAI of RDEC-1, like that of 83/39, is integrated into the pheU tRNA locus.
Functional analysis of the LEE PAI of E. coli strain 83/39
To determine if the LEE PAI of E. coli 83/39 contains all of the genes needed to express the A/E phenotype, E. coli HB101(p83/39) was tested for its ability to induce focal actin accumulation in HEp-2 cells in a 6 h fluorescent-actin staining (FAS) assay. Cosmid p83/39(2) contains the 34 kb core region of the LEE PAI of E. coli 83/39, apart from the last 176 bp of espF (Fig. 1A). E. coli HB101 containing this cosmid showed a positive FAS result, indicated by distinct fluorescence at the sites of bacterial attachment to HEp-2 cells (Fig. 2). No fluorescence associated with actin accumulation was seen in HEp-2 cells incubated with HB101 alone or with HB101 containing pHC79, the cosmid cloning vector. These results indicate that the cloned 34 kb core region of 83/39 contains all of the genes required to induce focal rearrangements similar to those induced by EPEC, and that intact espF is not necessary for expression of the A/E phenotype.
Identification, sequence analysis and genetic organization of the LEE PAI of E. coli 84/110-1
Two overlapping inserts from cosmids p84/110(7) and p84/110(2) hybridized with the LEE A, C or D probes (Fig. 1B). The left end of the insert from cosmid p84/110(7) contained ORFs identical to gspL and gspM from E. coli K-12 (GenBank accession no. AE000379) but, in 84/110-1, the gspM homologue was followed by a segment comprising 23 bp of the 3′-end of the pheV tRNA coding region, with a single internal basepair deletion (3′-TGGTGCC_GGACTCGGAATCGAA-5′). Immediately adjacent to this region was novel sequence belonging to the 84/110-1 LEE PAI. This result suggested that the 84/110-1 LEE PAI was integrated into the pheV tRNA locus located at min 67 of the chromosome of E. coli K-12. Sequencing of the right end of the insert from p84/110(2) revealed novel sequence and indicated that the overlapping inserts did not contain the complete LEE PAI of 84/110-1. To confirm that the pheV tRNA locus is the integration site of this PAI and to determine if it also contained a putative integrase gene, primers were designed that would amplify the right junction. P86 from yqgA, which is adjacent to the pheV tRNA locus in E. coli K-12, and primer P64, which is present in the putative integrase gene identified in the 83/39 LEE PAI, were used in a PCR reaction with genomic DNA from 84/110-1 as template. Sequence analysis of the 1.1 kb amplified fragment confirmed that the integration site of the LEE PAI of 84/110-1 is the pheV tRNA locus, and that a putative integrase gene, identical to that identified in 83/39, lies adjacent to this locus. These results indicated that the LEE PAI of 84/110-1 is flanked by short DRs, representing the 3′-end of the coding region of the pheV tRNA locus. This is the first time that a LEE PAI has been identified with this feature, although it is regularly found in other PAIs.
To determine the size of the 84/110-1 LEE PAI, we needed an internal fragment of the PAI that spanned the region from the right end of the insert from p84/110(2) to the pheV tRNA locus. This internal fragment was not represented in our cosmid library and we were unable to amplify this region using PCR. Accordingly, a double insertion mutant of E. coli 84/110-1, E. coli MT10, was constructed, which contained I-SceI restriction sites that flanked the unknown internal fragment (see Experimental procedures). MT10 was digested with I-SceI after which pulsed-field gel electrophoresis (PFGE) was used to separate the DNA fragments. Southern hybridization was performed using probes specific for eae (probe eae2) and the putative integrase gene (probe int-phe), which should be on same the fragment if the I-SceI restriction sites had been inserted correctly. The fragment which hybridized with these probes was approximately 50 kb in size (Fig. 3), making the entire LEE PAI of 84/110-1 approximately 85 kb in size, which is the largest LEE PAI described so far (Fig. 1B).
Sample sequencing of the inserts from cosmids p84/110(7) and p84/110(2) indicated that the LEE PAI of E. coli 84/110-1 also contained the 34 kb core region consisting of 40 ORFs, and like the LEE PAIs of 83/39 and RDEC-1, it is missing orf3 and the ERIC element. The sequenced sections of the core region of 84/110-1 (approximately 28 kb) exhibited >99.9% nucleotide identity to the same region of 83/39 and RDEC-1, except for espF where there is a 141 bp internal addition in 84/110-1 (or deletion in 83/39 and RDEC-1) (see Supplementary material), which encodes a proline-rich repeated sequence. Thus the EspF of 84/110-1 has 207 amino acid (aa) residues and three proline-rich repeats, whereas that of 83/39 and RDEC-1 comprises only 160 aa with two such repeats.
The left region of the PAI of strain 84/110-1 abutting the 34 kb core contains 10 ORFs, the key features of which are summarized in Table 2. This region is very similar to the left region flanking the LEE core of 83/39. In 84/110-1, immediately left of the 34 kb core are 645 bp of non-coding sequence, followed by a complete IS3 element, then ORFs similar to the putative prophage genes of the EDL933 LEE PAI, and finally E. coli K-12 sequence (including the 23 bp repeat) located at min 67 of the K-12 chromosome. In the case of 83/39, immediately adjacent to the left core region is the same 645 bp of non-coding sequence followed by an additional 128 bp of unique non-coding LEE sequence, then a truncated IS2 element, followed by 119 bp of one of the putative prophage genes of the LEE of 84/110-1, then an IS3 element, and finally E. coli K-12 sequence at min 94 that is missing 255 bp of cadC as indicated above.
|ORF or IS||Locationa||Size (aa)||Source, similar proteinb||% Amino acid identity|
|A||195–1037c||280||EDL933 LEE PAI L0010 (GenBank AF071034)||63 (71/112)|
|B||1128–1370c||80||EDL933 LEE PAI L0009 (GenBank AF071034)||87 (57/65)|
|C||1337–1771c||144||EDL933 LEE PAI L0008 (GenBank AF071034)||84 122/145|
|D||1821–2198c||125||EDL933 LEE PAI L0007 (GenBank AF071034)||93 116/124|
|E||2288–2764c||158||K-12 putative structural protein (GenBank AE000292)||89 109/122|
|F||2706–2819c||37||EHEC hypothetical protein (GenBank AE005277)||72 27/37|
|G||2819–3040c||73||K-12 hypothetical protein (GenBank AE000292)||95 (70/73)|
|H||3476–3751c||91||K-12 hypothetical protein (GenBank D90740)||98 (90/91)|
|IS3||3312–3611c||99||K-12 transposase for IS3 (SWISSPROT P77681)||100 (99/99)|
|IS3||3608–4474c||288||K-12 transposase for IS3 (SWISSPROT P05822)||100 288/288|
|I||1960–2154d||64||STEC ST43 (EMBL CAC81881)||98 (63/64)|
|J||2230–2814d||194||EPEC TrcA (DDBJ BAA36747)||53 104/193|
|K||3012–3650d||212||EPEC ORF2 (DDBJ BAA36748)||58 118/203|
|L||3941–5305d||454||EHEC hypothetical protein (DDBJ 34991)||69 (314–453)|
|Z||594–1858e||421||E. coli P4 integrase (SWISSPROT P08320)||54 222/411|
The region to the right of the 34 kb core of 84/110-1 that has been sequenced so far contains at least five ORFs, including a homologue of the putative virulence determinant, trcA, which encodes a protein involved in the formation of microcolonies by EPEC on cultured epithelial cells (Tobe et al., 1999). Approximately 34 kb of this region has not been sequenced, but we have determined that in contrast to 83/39 it does not contain efa1/lifA or senA.
Confirmation of the integration sites of the REPEC LEEs
To confirm the integration sites of the REPEC LEEs that were determined by sequence analysis, Southern hybridization was performed on I-CeuI digests of genomic DNA of REPEC 83/39, RDEC-1, and 84/110-1 using a probe specific for the conserved N-terminal region of the eae gene (probe eae1), which is a marker for LEE. I-CeuI cuts DNA of E. coli K-12 at each of the seven 23S ribosomal RNA (rrn) loci (Liu et al., 1993), thus allowing construction of genomic maps. The results of this analysis showed that the LEE PAIs and the predicted integration sites were located on the same fragments (Fig. 4), thus confirming the data derived from sequence analysis. Interestingly, the results also revealed that chromosomal rearrangements involving the fragments corresponding to fragments C and E of E. coli K-12 have evidently taken place in E. coli 83/39 (Fig. 4).
PCR analysis of spontaneous deletion of the LEE PAI from E. coli 84/110-1
Sequencing of the LEE PAI of 84/110-1 revealed that the right junction of the PAI contains an intact pheV tRNA locus, adjacent to a putative phage-related integrase gene, and that the left end is flanked by a segment comprising 23 bp of the 3′-end of the coding region of pheV tRNA with a single base deletion as described above. Because PAIs that are flanked by DRs are often unstable, we determined if the LEE PAI of 84/110-1 is capable of excision. To this end we introduced a counter-selectable marker, sacB, and a kanamycin-resistance gene into a non-coding region downstream of the last ORF (espF) of the 34 kb core region, generating E. coli MT11 (see Experimental procedures). sacB encodes the enzyme levansucrase, which is lethal to Gram-negative bacteria when they are grown in the presence of sucrose. E. coli MT11 was cultured on LB agar containing 10% sucrose, so that only derivatives of MT11 which had lost the sacB gene (and potentially the LEE PAI with it) would be viable. All sucrose-resistant derivatives of E. coli MT11 were susceptible to kanamycin and failed to hybridize with the eae gene probes indicating that an excision event had occurred. To determine the specificity of the excision event, 30 kanamycin-susceptible, eae derivatives of E. coli MT11 were analysed by colony PCR. Analysis of these mutants showed that two types were obtained. In the first type (27% of the total), the entire LEE PAI had excised permitting amplification of a 1.05 kb fragment with primers P74 and P86 (Fig. 5). Sequence analysis of this fragment revealed the pheV tRNA locus with a single internal base deletion at the 3′-end of the coding region followed by the gspM homologue, that is, the right and left junctions which had originally flanked the entire LEE PAI. In the second type of mutant (73% of the total), the left and right junction regions of the PAI were intact, as a 0.9 kb fragment was amplified with primers P74 and P126 and a 1.1 kb fragment amplified with primers P64 and P86. However, PCR analysis using primers that span the 34 kb core region and the IS3 element (see Experimental procedures) revealed that the complete 34 kb core region and the adjacent IS3 element had deleted. These results demonstrate that the LEE PAI of 84/110-1 is unstable and capable of spontaneous deletion. The combined deletion frequency was estimated to be approximately 6.6 × 10−6, which is similar to deletion frequencies of other PAIs (Rajakumar et al., 1997; Middendorf et al., 2001; Turner et al., 2001).
To investigate if the LEE PAI that was lost from the first type of mutant had circularized and was capable of independent replication, we attempted to amplify across the closed junction region of the excised PAI using outward facing primers on the LEE, P126 and P64 (Fig. 5). Plasmid and genomic DNA prepared from the PAI deletants was used as template in PCR reactions, in addition to colony PCR, but no PCR product was obtained. We also failed to amplify the eae gene in the PAI deletants using primers eaeF and eaeR, thus confirming that the 84/110-1 LEE PAI was not present as an extrachromo-somal element.
Characterization of the putative integrase of the LEE PAIs of REPEC
The integration sites of the LEE PAIs of all three REPEC strains characterized to date are the 3′-ends of the phenylalanine tRNA loci, pheU and pheV. Adjacent to these loci in all three REPEC strains is an identical ORF, designated int-phe, which encodes a predicted protein of 421 amino acids. Int-phe is a member of a family of site-specific recombinases, which contains the highly conserved RHRHY motif that is necessary for integrase function (Grainge and Jayaram, 1999). A clustal alignment of the amino acid sequence of Int-phe with the most similar integrases and other integrases for which function has been demonstrated (see Supplementary material) was used to construct a phylogenetic tree, which revealed a close relationship between the integrase and its site of integration (Fig. 6). This finding suggests that it is the type of integrase and not the virulence factors encoded on the PAI that determines the site of integration.
Site-specific recombinases, like the integrases of bacteriophages P4 and phi-R73, mediate recombination between a phage attachment site, attP (POP′), and a bacterial chromosomal attachment site, attB (BOB′), in which ‘O’ is the common core of the attachment site where strand exchange occurs (Landy, 1989; Sadowski, 1993; Yang and Mizuuchi, 1997). Integration results in formation of attL (BOP′), the left end junction, and attR (POB′), the right end junction, which contain sequences derived from the integrated phage and the bacterial chromosome.
To determine if the int-phe gene product that associated with the LEE PAIs was functional and sufficient to promote site-specific recombination between attP and attB, we investigated the product of the int-phe gene for integrase activity in a recombination deficient strain of E. coli, DH1. In the case of the LEE PAIs of REPEC, we hypothesized that the attB (BOB′) site was either of the two phe tRNA loci, and that the ‘O’ common core was the last 23 nucleotides of the phe tRNAs. To this end, we cloned the int-phe gene plus a portion of attR (POB′) (Fig. 7), which contains the ‘PO’ elements of attP (POP′) on a temperature-sensitive suicide plasmid, pMT27, that encodes resistance to kanamycin. After transformation into E. coli DH1 and selection for kanamycin-resistant transformants, we performed colony PCR to determine if the vector had integrated into either phe tRNA locus, our predicted attB site. The results showed that on each occasion pMT27 integrated into the DH1 chromosome, the site of integration was pheU tRNA (Fig. 8), with an integration frequency of approximately 1.4 × 10−6. The fragments amplified with primer pairs P64–P32 (0.9 kb) and P114–P32 (1.3 kb) were sequenced. Analysis of these sequences revealed that integration was precise and the sequence at the integration site was identical to the junction region of the 83/39 LEE PAI, namely, an intact pheU tRNA locus adjacent to the int-phe gene. Integration of pMT27 was not detected at the pheV tRNA locus.
Although we used E. coli DH1, a recombination deficient strain, as the recipient for pMT27, making homo-logous recombination a most unlikely explanation for the observed integration, we repeated the experiment using plasmid pMT29, which contained an insertionally inactivated int-phe gene. In these experiments, all bacterial colonies that grew on LB agar containing kanamycin after the 42°C incubation contained free plasmid that had not been cured from the bacterial cells and the phe tRNA loci were intact. A similarly low background level of DH1 cells, containing pMT27 which had not been cured, was detected in DH1 transformed with pMT27. The inability of pMT29 to integrate into the bacterial chromosome of DH1 indicated that the putative integrase, Int-phe, and not homologous recombination, was responsible for the integration.
- Top of page
- Experimental procedures
- Supplementary material
- Supporting Information
Comparison of the LEE PAIs of REPEC strains 83/39 and 84/110-1 with the homologous loci from EPEC E2348/69, EHEC EDL933 and REPEC RDEC-1 revealed that the 34 kb core region of these PAIs are highly conserved in linear gene order, nucleotide, and predicted protein sequences (Fig. 9). The core region of 83/39 is also functionally related to that of E2348/69 in terms of its ability to confer the capacity for fluorescent actin staining on a laboratory strain of E. coli (E. coli HB101), suggesting that the LEE PAI of 83/39 is sufficient for the production of A/E lesions. However, there are also major differences between the REPEC PAIs and those of EPEC E2348/69 and EHEC EDL933. In particular, the REPEC LEE PAIs are larger, contain other putative virulence determinants in regions flanking the 34 kb core, and have inserted into different sites of the chromosome. There are also minor differences in that the REPEC core regions do not contain an ERIC element or orf3, and the size of espF varies between these strains. These differences could be used to differentiate different LEE-positive strains of E. coli in epidemiological studies.
The G + C content of the fully sequenced LEE core regions are similar (approximately 39%) and considerably lower than the average for E. coli K-12 (50.8%), suggesting that the LEE was acquired from an unrelated organism via horizontal gene transfer. The highly conserved core region suggests that it had a common origin, but it is not clear how or when it was acquired by the putative ancestral strain(s) of EPEC and EHEC (Hacker and Kaper, 2000).
The LEE PAIs of REPEC 83/39 and RDEC-1 are closely related. Both of these PAIs have inserted into pheU tRNA leaving the coding region intact, and the 38 kb SmaI fragment of RDEC-1 is nearly identical to the corresponding region in 83/39. Although these strains are of the same serotype, O15:H-, and appear to contain near identical LEE PAIs, there are clear differences between them. For example, RDEC-1 possesses AF/R1 fimbriae (Wolf and Boedeker, 1990), whereas 83/39 possesses Ral fimbrial that resemble the K88 colonization fimbriae of ETEC (Adams et al., 1997). There are also differences in their chromosome, which in 83/39 appears to have undergone rearrangements compared with RDEC-1 and E. coli K-12. Rearrangements of this magnitude may result from homologous recombination between rRNA operons (Liu and Sanderson, 1998) or from large chromosomal insertions that are accompanied by inversions or translocations that restore the symmetry between the replication origin and terminus (Bergthorsson and Ochman, 1998).
At 59.5 kb, the LEE PAI of 83/39 is considerably larger than the corresponding PAIs of EPEC E2348/69 and EHEC EDL933. The regions flanking the 34 kb core contain 14 ORFs, two of which are homologues of genes which encode putative virulence determinants of E. coli, namely, efa1/lifA and senA (Nataro et al., 1995; Nicholls et al., 2000; Klapproth et al., 2000). Interestingly, efa1 is found in most other A/E strains of E. coli, including E2348/69, but generally is not co-located with the LEE PAI (Nicholls et al., 2000). PCR has revealed that EPEC E2348/69 and EHEC EDL933 also carry a homologue of senA and that it is also not co-located with LEE (unpublished data). The finding that these putative virulence determinants comprise part of the LEE PAI of 83/39 suggests that they may have been acquired at the same time as the 34 kb core. If the same were true of other A/E clones of E. coli, it would explain the widespread occurrence of these genes in EPEC and EHEC, even though they are not required for the production of A/E lesions.
The LEE PAI of REPEC 84/110-1 has inserted into the pheV tRNA locus and is around 85 kb in size, making it the largest LEE PAI to be identified so far. We determined the sequence of over 28 kb of the 84/110-1 core region, which appears to be nearly identical to the corresponding region of 83/39 (data not shown), apart from espF, which has either undergone a deletion (in the case of 83/39) or an addition in 84/110-1. Because the left region flanking the core of 83/39 is also present in the LEE of 84/110-1 and as the LEE cores of 83/39 and 84/110-1 appear to be identical, the LEE PAI of 83/39 may have originally possessed the same left flanking region as that of the 84/110-1 LEE PAI, but has subsequently undergone deletions, as part of the homing process and selection for stability (Hacker and Kaper, 2000).
Recently, the LEE PAI of a bovine EHEC strain of the same serotype as 84/110-1, O103:H2, was reported to have integrated into pheV tRNA (Jores et al., 2001). Comparison of the regions flanking the LEE core of these strains (GenBank accession no. AJ303141, AJ303142) revealed that the left flanking region of the 84/110-1 LEE core is identical to the corresponding region in EHEC O103:H2, including the 23 bp imperfect DR, except that the IS3 element in the LEE PAI of 84/110-1 has been replaced by a homologue of an IS629 element (which was originally identified in Shigella sonnei (Matsutani et al., 1987)), and by 4018 bp of sequence similar to those of the she PAI of Shigella flexneri (which has also inserted into pheV tRNA (Al Hasani et al., 2001)), and the putative prophage genes of the EDL933 LEE PAI. The right flanking sequences are also identical from the non-coding region between escF and espF to that last ORF that has been sequenced of the 84/110-1 LEE PAI, except for the espF gene. EspF of 84/110-1 has 207 aa and three proline-rich repeats, whereas EspF of EHEC O103:H2 comprises 294 aa with five such repeats. Interestingly, although the regions flanking the LEE cores of 84/110-1 and EHEC O103:H2 are very similar, these LEEs do not appear to be clonal, as the 84/110-1 LEE contains a gene for β intimin whereas that of EHEC O103:H2 LEE encodes ɛ intimin (Jores et al., 2001), and there are the aforementioned differences in EspF. This may indicate that the flanking regions were once part of an integrative element, such as a prophage, which integrated into the pheV tRNA locus before the different LEE cores were acquired, possibly via transposition, as indicated by the IS elements.
Several PAIs contain homologues of genes for the integrases of bacteriophages, which may have played a role in the original mobilization of the PAI (Hacker and Kaper, 2000). The 933L prophage within the LEE of EHEC EDL933 contains a gene that encodes a P4-like integrase that is a member of the family of site-specific recombinases. Given the absence of these genes from the LEE PAI of EPEC E2348/69, it is thought that the 933 L prophage inserted into the LEE PAI after the integration of the PAI into the chromosome (Perna et al., 1998). The LEE PAIs of 83/39, RDEC-1 and 84/110-all contain the int-phe gene that encodes an integrase that is another site-specific recombinase. PAI IIJ96 of uropathogenic E. coli strain J96, which contains the virulence determinants hlyII, prs and cnf1, has inserted into the pheU tRNA locus, is flanked by 135 bp imperfect DRs, and can spontaneously delete from the chromosome (Swenson et al., 1996). To determine if this PAI contained the int-phe gene, we sequenced the junction region adjacent to the pheU tRNA locus and identified the int-phe gene (GenBank accession no. AF453829). So far, six PAI elements of enterobacteria have been identified, which are co-located with either of the two identical phenylalanine tRNA loci, pheU and pheV, and all contain the int-phe gene. These are the three LEE PAIs of REPEC, the PAI IIJ96 of uropathogenic E. coli, the afa-8 gene cluster of E. coli, and the she PAI of S. flexneri (Al Hasani et al., 2001; Lalioui and Le Bouguenec, 2001). On the basis of this observation we hypothesized that the Int-phe integrase may play a role in the mobilization of these PAIs, and that it could function like an integrase of temperate bacteriophages, mediating RecA-independent recombination between a phage attachment site (attP/POP′) and a bacterial chromosomal site (attB/BOB′) (Landy, 1989; Yang and Mizuuchi, 1997). To investigate this possibility, we constructed a plasmid that contained the int-phe gene plus attR (POB′), which contains the ‘PO’ elements of attP (POP′), and is present in all the REPEC LEE PAIs that were investigated. Our results showed that the Int-phe integrase of 84/110-1 can mediate site-specific integration at the pheU tRNA locus. Integration occurred at a relatively low frequency but this could be due to use of an incomplete attP site. Similar work using the int-HPI gene of Yersinia pestis KUMA and a Π-dependent suicide vector showed that the composition of the attP site influenced the frequency of the recombination event (Rakin et al., 2001).
Surprisingly, integration was not observed at the identical pheV tRNA locus. In E. coli K-12, the phenylalanine tRNA coding regions and 18 bp upstream of the coding regions are identical, but further upstream and immediately downstream of the coding regions the sequences are different. After an integrative element inserts into a phe tRNA locus, the downstream regions become identical and consist of 114 bp of K-12-like sequence that is normally found downstream of the pheU tRNA. For this reason, the Int-phe integrase may preferentially mediate integration into the pheU tRNA locus, with pheV tRNA as an alternative but less preferred target. The high pathogenicity island (HPI) of Y. pestis is associated only with the asnT tRNA locus, but the HPI of Yersinia pseudotuberculosis shows no preference and can insert into any of the three asn tRNA loci in the chromosome (Buchrieser et al., 1998).
Integrases of the family of site-specific recombinases can also mediate excision (Yang and Mizuuchi, 1997). The she PAI of S. flexneri and the PAI IIJ96 of uropathogenic E. coli contain the int-phe gene, are flanked by 23 bp imperfect DRs (the 3′-end of the phe tRNA coding region), and are capable of spontaneous deletion (Swenson et al., 1996; Al Hasani et al., 2001). Because the LEE PAI of 84/110-1 also contains the int-phe gene and is flanked by the same DRs, we wanted to determine if the 84/110-1 LEE PAI was also capable of spontaneous excision. To this end, we constructed a LEE::sacB derivative of 84/110-1 to select for possible deletion/excision mutants of 84/110-1. Our results showed that the LEE PAI of 84/110-1 is capable of spontaneous deletion. Interestingly, two types of deletants were observed. In the first type, the complete PAI had excised, presumably by recombination between the 23 bp DRs that flank the 84/110-1 LEE PAI and mediated by Int-phe. In the second type, the integrase and the DRs do not appear to be involved as the junction regions of the PAI were intact, but the 34 kb core and the adjacent IS3 element had excised. The precise region that has deleted from the PAI is still to be determined as the region flanking the right side of the 34 kb core has not been completely sequenced, but as transposition of an IS3 element proceeds in a non-replicative manner (Sekine et al., 1996), it is likely that the IS3 element may have played a role in the deletion process.
This work may provide new insights into the evolution of the LEE pathogenicity island and indicate a possible mechanism of transfer. The functional phage-like integrase, the att sites or DRs, and the targeting of tRNA loci as integration sites suggests that the LEE PAI may have been acquired from a bacteriophage via horizontal transfer. If such a bacteriophage was responsible it could carry a variety of virulence determinants apart from the LEE, as there are PAIs, such as PAI IIJ96 of uropathogenic E. coli and the she PAI of S. flexneri that contain the same integrase gene and have integrated into the same tRNA loci as the LEE PAIs of 83/39 and 84/110-1, respectively, but carry different virulence determinants. The REPEC LEE PAIs are large, ranging from 60 to 85 kb, yet the size of the P4 genome, which encodes an homologous integrase gene, is only 11.6 kb (Inman et al., 1971). The amount of DNA that can be introduced into a phage is limited by the size of the capsid, but can extend beyond 100 kb (Ochman et al., 2000). Perhaps a larger bacteriophage with a similar integrase to that of P4 was responsible for the horizontal transfer of the PAIs to E. coli. This hypothetical phage may have introduced an expanded LEE PAI, perhaps similar to that of 83/39 which contains the efa1/lifA and senA genes that are present in most A/E strains of E. coli. Another possibility is that a P4-like bacteriophage originally integrated into the tRNA locus, thus creating a favourable environment for the integration of other mobile elements such as IS elements, transposases and other integrases which were associated with a variety of different virulence determinants. Our observation, using DNA hybridization and sample sequencing (data not shown), that almost all of the DNA sequence between the right end of the LEE core region and int-phe in 84/110-1 and 83/39 is different, provides some support for the latter hypothesis. Prophage DNA is non-essential so the integration of mobile elements into this DNA would be non-deleterious to the survival of the bacterium and may even enhance survival by inactivating the lytic cycle of a temperate bacteriophage. PAIs generally contain mobile elements that may supply them with portable regions of homology and promote RecA-independent integration of foreign DNA (Hacker et al., 1997). If the DNA sequences within PAIs contribute to the fitness or pathogenicity of the host organism, bacteria in which the mobility elements are inactivated or deleted to enhance stability of the PAI may be at a relative advantage. Hence, over time, successful PAIs will contain only the genes required for their essential functions (Hacker and Kaper, 2000). The observation that the LEE PAI of 84/110-1 is significantly larger than other LEE PAIs and mobilizable suggests that it may be less evolved than the LEE of E2348/69, which is less than half the size of the LEE PAI of 84/110-1 and is stable.
- Top of page
- Experimental procedures
- Supplementary material
- Supporting Information
Bacterial strains, plasmids and growth conditions
The bacterial strains and plasmids used in this study are listed in Tables 3 and 4. Strains were routinely grown at 37°C in Luria–Bertani (LB) broth (1% tryptone, 0.5% yeast extract, 171 mM NaCl) (Ausubel et al., 1991) supplemented with 100 μg ml−1 of ampicillin (CSL), 50 μg ml−1 of kanamycin, and 25 μg ml−1 of chloramphenicol (Sigma-Aldrich) when necessary.
|Bacterial strain||Relevant characteristicsa||Reference or source|
|83/39||Wild-type REPEC strain, O15:H-||Robins-Browne et al. (1994b)|
|84/110-1||Wild-type REPEC strain, O103:H2||Robins-Browne et al. (1994b)|
|RDEC-1||Wild-type REPEC strain, O15:H-||Zhu et al. (2001)|
|E2348/69||Wild-type EPEC strain, O127:H6||Elliott et al. (1998)|
|J96||Wild-type uropathogenic E. coli strain, O4||Swenson et al. (1996)|
|EDL933||Wild type EHEC strain, O157:H7||Perna et al. (1998)|
|DH1||Laboratory strain, F−supE44 recA1 endA1 gyrA96 (NalR) thi1 hsdR17(rK−mK+) relA1||Hanahan (1983)|
|HB101||Laboratory strain, F−(gpt-proA)62 leuB6 supE44 ara-14 galK2 lacY1(mcrC-mrr) rpsL20 (StrR) xyl-5 mtl-1 recA13||Boyer and Roulland-Dussoix (1969)|
|W3110||Laboratory strain, K-12 F−λ− IN(rrnD-rrnE)1 rph-1||Bachmann (1972)|
|SY327||Laboratory strain, F−araD(lac pro) argE recA56 nalA [λ−pir] KmR 84/110-1||Miller and Mekalanos (1988)|
|MT5||84/110-1 eae::pMT30, ApR||This study|
|XL1-Blue||Laboratory strain, F::Tn10 proA+B+lacIq(lacZ)M15/recA1 endA1 gyrA96 (NalR) thi hsdR17(rK−mK+) supE44 relA1 lac||Stratagene|
|MT10||84/110-1 eae::pMT30 yqgA::pMT32, ApR KmR||This study|
|MT11||84/110-1 carrying sacB and KmR gene in non-coding region downstream of espF, KmR||This study|
|Plasmid||Relevant characteristicsa||Reference or source|
|pBluescript-II KS+||Cloning vector, ApR||Stratagene|
|pCactus||pSC101-based low copy number vector with sacB, CmR||Van den Bosch et al., 1997|
|pHC79||Cosmid cloning vector, ApR||Hohn and Collins (1980)|
|pST76-A||Temperature-sensitive suicide vector derived from pSC101, ApR||Posfai et al. (1997)|
|pSG76-K||Suicide vector, R6K γ-ori, KmR||Posfai et al. (1997)|
|pSG76-A||Suicide vector, R6K γ-ori, ApR||Posfai et al. (1997)|
|pCVD442||Suicide vector, ori R6K, mob RP4, sacB, ApR||Miller and Mekalanos (1988)|
|pUC4-KIXX||pUC4K derivative containing KmR and BmR genes from Tn5||Pharmacia|
|pGEM-T Easy||Cloning vector, ApR||Promega|
|pMT30||0.7 kb PCR fragment containing part of yqgA from 84/110-1 cloned into EcoRI/BamHI sites of pSG76-K, KmR||This study|
|pMT32||0.7 kb PCR fragment containing part of eae from 84/110-1 cloned into EcoRI/BamHI sites of pST76-A, ApR||This study|
|pMT35||1.4 kb PCR fragment containing downstream region of espF from 84/110-1 cloned into EcoRI/BamHI sites of pSG76-A, ApR||This study|
|pMT36||2.6 kb fragment containing sacB from pCVD442 cloned into PstI site of pBluescript-II KS+||This study|
|pMT38||pMT36 containing 1.4 kb KmR gene from pUC4-KIXX cloned into SmaI site, KmR||This study|
|pMT40||pMT35 containing 4.0 kb fragment of pMT38 bearing sacB and KmR gene cloned into EcoRV site, KmR||This study|
|pMT23||1.6 kb PCR fragment containing int-phe from 84/110-1 cloned into pGEM-T Easy, ApR||This study|
|pMT24||pMT23 containing 1.4 kb KmR gene from pUC4-KIXX cloned into EcoRV site, KmR, ApR||This study|
|pMT25||pMT23 containing 1.4 kb KmR gene from pUC4-KIXX cloned into Smal site, ApR KmR||This study|
|pMT27||pCactus containing 3.0 kb fragment encompassing int-phe from pMT25 cloned into Xbal/Sacl sites, CmR KmR||This study|
|pMT29||pCactus containing 3.0 kb fragment encompassing int-phe disrupted with KmR gene from pMT25 and cloned into Xbal/Sacl sites, CmR KmR||This study|
Preparation and manipulation of DNA
Plasmid DNA was isolated using the Wizard Plus SV DNA Purification System (Promega), and cosmid DNA was isolated using the Qiagen large-construct kit (Qiagen). Standard restriction digestion and cloning procedures using DNA-modifying enzymes supplied by Promega or New England Biolabs were used (Sambrook et al., 1989; Ausubel et al., 1991). Electrocompetent Escherichia coli XL1-Blue, DH1, 84/110-1 and MT5 cells were obtained by growing the bacteria to mid-log phase (OD600 0.5–0.8) at 37°C. Cells were then washed three times with sterile, cold 10% (v/v) glycerol in distilled water and resuspended in 1:75 of the original culture volume. Transformation was achieved using a Bio-Rad Gene Pulser (Bio-Rad Laboratories) and electroporation conditions, 1.80 kV, 200 3Ω, and 25 μF.
DNA amplification, sequencing and analysis
PCR primers used in this study are listed in Table 5. Genomic DNA of E. coli strains was prepared by the boiling lysis method (Ausubel et al., 1991), and PCR amplifications were routinely performed with AmpliTaq polymerase (Applied Biosystems) in a reaction volume of 25–50 μl in a Gene Amp PCR System 9700 thermal cycler (Applied Biosystems). The PCR conditions involved denaturation for 2 min at 94°C, followed by 30 cycles of 30 s at 94°C, 30 s at 50–65°C (depending on the Tm of the primers), and 1–4 min at 72°C, with a final extension for 5 min at 72°C. The method referred to as colony PCR was performed as follows: one bacterial colony from an agar plate was transferred with a sterile toothpick to a 0.5 ml PCR tube containing 100 μl of sterile distilled water. The sample was heated to 98°C for 6 min to lyse the bacterial cells and release their DNA. 1 μl of this solution was used as template DNA in a PCR.
Nucleotide sequencing was performed using an ABI PRISM Big Dye Terminators v3.0 Cycle Sequencing Kit (Applied Biosystems). Reactions were analysed on an Applied Biosystems ABI PRISM 377 DNA sequencer. Nucleotide sequence data were edited and assembled into contiguous sequence with the SEQUENCHER program (Gene Codes). Searches for nucleotide and amino acid homology with sequences in the public databases were performed by using the BLASTN, BLASTP and BLASTX programs (Altschul et al., 1990) available at the National Center for Biotechnology Information website (www.ncbi.nlm.nih.gov).
Isolation of LEE+ cosmids
pHC79 cosmid libraries of E. coli 83/39 and 84/110-1 (maintained in E. coli HDH1) were screened by colony hybridization (Sambrook et al., 1989) for LEE-specific sequences under stringent conditions at 65°C. For these studies, bacteria were grown for 6 h at 37°C on Hybond-N nylon membranes (Amersham Biosciences) and probed with LEE A, C and D (McDaniel et al., 1995) labelled with [α-32P]-dATP using the Megaprime DNA labelling system (Amersham Biosciences). Bound probe was detected by autoradiography with Kodak X-Omat Film (Eastman Kodak). Restriction digestion analysis of cosmid DNA from probepositive clones was used to identify cosmids containing overlapping inserts. The ends of these cloned inserts were sequenced using primers pHC79(308) and pHC79(468), which flank the BamHI insertion site of pHC79. Physical maps were generated and orientated with respect to the direction of transcription of the eae gene.
Fluorescent-actin staining (FAS) assay
Actin rearrangements were detected by fluorescence microscopy of fixed cells after staining with fluorescein isothiocyanate-coupled phalloidin, as described by Knutton and colleagues (Knutton et al., 1989). E. coli strains were grown in 10 ml of LB broth without shaking at 37°C overnight and incubated with a semiconfluent monolayer of HEp-2 cells for 6 h. After removal of non-adherent bacteria by washing, cells were examined using fluorescence microscopy and then by phase-contrast microscopy to confirm that fluorescent areas corresponded to attached bacteria, using a Leica DM-LB HC microscope with a I3 fluorescein filter (excitation 450–490 nm) (Leica Microsystems).
Construction of a double insertion mutant to determine size of the LEE PAI of E. coli 84/110-1
Two suicide plasmids containing I-SceI and NotI sites were inserted into target sites that flank an unknown segment of genomic DNA. The first site was the eae gene. Primers P129 (including an EcoRI site) and P130 (including a BamHI site) were used to amplify a 694 bp fragment of the eae gene. The fragment was digested with EcoRI and BamHI and inserted into EcoRI- and BamHI-digested pSG76-K, an R6K γ-ori suicide vector, which requires the II protein, the product of the pir gene, for replication. This generated pMT30 which was maintained in E. coli SY327(λpir). The second target site was yqgA, the first ORF homologous to E. coli K-12 sequence adjacent to the pheV tRNA locus. Primers P131 and P132, incorporating an EcoRI and a BamHI site, respectively, were used to amplify a 706 bp fragment containing part of the yqgA gene. This fragment was digested with EcoRI and BamHI and ligated into EcoRI- and BamHI-digested temperature-sensitive suicide vector, pST76-A, to generate pMT32, which was maintained in E. coli DH1 at 30°C. pMT30 was electroporated into E. coli 84/110-1, which was then plated on LB agar containing ampicillin and incubated overnight. Transformants were investigated by PCR to confirm that pMT30 had recombined by a single crossover into the eae gene; one such positive recombinant was designated MT5. Electrocompetent MT5 cells were then transformed with pMT32, plated on LB agar containing ampicillin and kanamycin, and incubated at 30°C overnight. Transformants of MT5 were then incubated at 42°C for 7–9 h and at 37°C for a further 12–24 h to obtain cells cured of free plasmid. PCR was used to identify transformants in which pMT32 had recombined by a single crossover into the yqgA gene of MT5. One such recombinant was designated E. coli MT10 (ApR, KmR)
Pulsed-field gel electrophoresis (PFGE)
Agarose-embedded DNA suitable for separation by PFGE was prepared in accordance with instructions provided by the manufacturer of the CHEF-DR III variable angle electrophoresis system (Bio-Rad Laboratories). Low-melting temperature agarose plugs containing approximately 2 μg of DNA were equilibrated with digestion buffer (150 μl) and then incubated at 37°C for 16 h with 30 U of NotI (New England Biolabs) in a 50 μl reaction mixture; or with 4 U of I-CeuI (Roche Molecular Biochemicals) in a 50 μl reaction for 3 h at 37°C; or with 10 U of I-SceI (Roche Molecular Biochemicals) in a 40 μl reaction mixture for 1 h on ice, followed by the addition of 30 μl 25 mM MgCl2 and further incubation for 1 h at 37°C. After digestion the agarose plugs were equilibrated in TE buffer (10 mM Tris [pH 8.0], 1 mM EDTA) for 30 min at room temperature before PFGE in a CHEF DR III system apparatus. Gels contained 1% agarose (or 0.7% for I-CeuI-digested DNA) in 0.5× TBE buffer (44.5 mM Tris, 44.5 mM boric acid, 1 mM EDTA). TBE buffer was also used as the running buffer for electrophoresis. Gels were electrophoresed at 13°C at 6 V cm−1, included angle 120°, for either 22 h and switch time 4–40 s (NotI and I-SceI-digested DNA); or for 12 h and switch time 20–120 s followed by a further 12 h and switch time 60–100 s (I-CeuI-digested DNA). After electrophoresis, the gels were stained with ethidium bromide.
Electrophoretically separated DNA fragments were transferred overnight by capillary transfer onto positively charged nylon membranes (Roche Molecular Biochemicals) (Sambrook et al., 1989). DNA probes were prepared from genomic DNA of 83/39 and 84/110-1, and labelled by PCR amplification with digoxigenin, hybridized overnight under conditions of high stringency at 65°C and detected as recommended by the manufacturer (Roche Molecular Biochemicals). The 917 bp eae1 probe was generated using primers P120 and P118 and is specific for the conserved N-terminal region of the eae gene. The 694 bp eae2 probe was generated using primers P129 and P130 and is specific for the C-terminal region of the eae gene for β-intimin. The 914 bp int-phe probe was generated using primers P47 and P114 and is specific for the integrase gene, int-phe. Chemiluminesence detection with CDP-Star (Roche Molecular Biochemicals) was used by exposing membranes to Kodak X-Omat film (Eastman Kodak Co.)
Construction of a LEE::sacB, KmR mutant of 84/110-1 to investigate the capability of the LEE PAI to excise
A non-coding region downstream of espF containing an internal EcoRV site was chosen as the integration site for the counter-selectable marker, sacB. A 1372 bp fragment from this region was amplified with primers P162 (containing an EcoRI site) and P163 (containing a BamHI site). The amplified fragment was digested with EcoRI and BamHI and ligated into EcoRI- and BamHI-digested suicide vector, pSG76-A, to generate pMT35 which was maintained in E. coli SY327. A 2.6 kb PstI fragment containing sacB was excised from pCVD442 and ligated into pBluescript-II KS+ to generate pMT36, after which a 1.4 kb SmaI fragment from pUC4-KIXX containing a kanamycin resistance gene (KmR) was inserted into SmaI-digested pMT36, to generate pMT38, which was electroporated into E. coli XL1-Blue. pMT38 was then digested with SalI and XbaI and the excised 4.0 kb fragment containing sacB and KmR was filled in using Klenow fragment to enable blunt end ligation into EcoRV-digested pMT35. The resulting construct, pMT40, was electroporated into E. coli strain 84/110-1 and the double crossover mutant was selected by screening for ApS, KmR recombinants. A mutant with the appropriate integration of sacB and KmR, as confirmed by PCR, was designated E. coli MT11(KmR).
Selection of spontaneous LEE deletion mutants of E. coli MT11
Spontaneous sucrose resistant, KmS derivatives of E. coli MT11 were isolated by plating serial dilutions of an overnight culture of MT11 grown in LB broth onto LB agar plates containing 10% (w/v) sucrose. Serial dilutions were also plated onto LB agar plates to determine the total number of colony forming units (cfu). The agar plates were incubated at 30°C for 24–48 h to retard growth and establish the toxic effect of the levansucrase. Sucrose-resistant colonies were replica-plated on LB agar containing kanamycin or chloramphenicol (as 84/110-1 is naturally chloramphenicol-resistant) to identify sucrose and chloramphenicol-resistant, kanamycin-sensitive strains. Agglutination with a specific hyperimmune anti84/110-1 antiserum was used to confirm that putative LEE PAI strains were derivatives of 84/110-1. Deletion of the LEE PAI of 84/110-1 was analysed by colony PCR. Primers P74 and P126 were used to amplify the left junction, primers P64 and P86 were used to amplify the right junction region, and inward facing primers, P74 and P86, situated at either side of the 23 bp imperfect direct repeats were used to amplify a product spanning the right and left junctions of the PAI, if it had excised. Primers that spanned the 34 kb core region (rorf2/rorf2, ler1/ler2, escT1/escT2, escV1/ escV2, tir1/tir2, eaeF/eaeR, P162/P163) and the IS3 element (IS3/IS3 and P10/P15) were also used. The deletion frequency was calculated as the quotient of LEE PAI deletant cells to the total number of cfu.
Generation of plasmids and recombinant strains of E. coli DH1 used to analyse integrase function
Genomic DNA of E. coli 84/110-1 was used as a template to amplify the int-phe gene plus a portion of attR (POB′). PCR was performed with Vent DNA polymerase (New England Biolabs), which has proofreading activity. The integrase gene contains an internal EcoRV site, and 20 bp downstream of the putative translation stop codon is a SmaI site. A 1646 bp fragment was amplified using primers P117 (containing an XbaI site) and P119 (incorporating a SacI site). The PCR fragment was A-tailed with AmpliTaq polymerase and ligated into the linearized pGEM-T Easy vector containing single 3′ thymidine overhangs, to generate pMT23. A 1.4 kb SmaI fragment containing a kanamycin resistance gene from pUC4-KIXX was (i) inserted into SmaI-digested pMT23 to generate pMT25 and (ii) inserted into EcoRV-digested pMT23 to generate pMT24. Because the 3.0 kb inserts were approximately the same size as the pGEM-T Easy vector, pMT25 and pMT24 were first digested with PvuI, which liberated a 4.0 kb fragment that was subsequently digested with XbaI and SacI. The resultant 3.0 kb inserts were then ligated into the XbaI- and SacI-digested suicide vector, pCactus, to generate pMT27 and pMT29. pMT27 contained attR, a full copy of the integrase gene, and a kanamycin resistance gene, whereas pMT29 contained attR and the integrase gene insertionally inactivated by the kanamycin resistance gene. pMT27 and pMT29 were then electroporated into E. coli DH1. Bacteria were plated on LB agar containing kanamycin and incubated at 30°C overnight. Transformants of E. coli DH1 were selected and grown in LB broth supplemented with kanamycin overnight at 30°C and serial dilutions were plated onto LB agar (without selection), incubated at 42°C for 7–9 h and at 37°C for a further 12–24 h to obtain cells cured of free plasmid. Serial dilutions were also plated on LB agar and incubated at 30°C to determine the total number of cfu. Bacterial cells were replica-plated on LB agar containing kanamycin. Colony polymerase chain reaction (PCR) using primers adjacent to the phe tRNAs (P86, specific for the pheV tRNA locus and P32, specific for the pheU tRNA locus) and two primers located within the putative integrase gene (P64, P114) was performed on the kanamycin- resistant DH1 clones to determine if pMT27 and pMT29 had integrated into the chromosome, and their site of integration. The integration frequency was calculated as the quotient of E. coli DH1 cells containing plasmid integrated into the chromosome to the total number of cfu.
Nucleotide sequence accession numbers
The GenBank accession numbers for the novel sequences reported here are AF453441, AF453442, AF453829, AF454555, AF461393 and AF461394.
- Top of page
- Experimental procedures
- Supplementary material
- Supporting Information
The following material is available from
Fig. S1. Alignment of the nucleotide and predicted amino acid sequences of espF of REPEC strains 84/110-1, 83/39 and RDEC-1.
Fig. S2. Amino acid sequence alignment of Int-phe and homologous sequences identified in the databases.
- Top of page
- Experimental procedures
- Supplementary material
- Supporting Information
We are indebted to Dr J. B. Kaper, University of Maryland School of Medicine; Dr R. Moronal, University of Adelaide; Dr G. Posfai, Hungarian Academy of Sciences and Dr R. A. Welch, University of Wisconsin-Madison, for the gifts of bacterial strains and plasmids that were used in this study; and to V. Bennet-Wood, R. Good, and Dr H.Y. Li for their advice and technical assistance. This study was supported in part by a grant from the Australian National Health and Medical Research Council. M.T. was supported by an Australian Postgraduate Research Award.
- Top of page
- Experimental procedures
- Supplementary material
- Supporting Information
- 1999) Enteropathogenic Escherichia coli translocated intimin receptor, Tir, requires a specific chaperone for stable secretion. Mol Microbiol 33: 1162–1175. , , , , , et al. (
- 1997) Identification and characterization of a K88- and CS31A-like operon of a rabbit enteropathogenic Escherichia coli strain which encodes fimbriae involved in the colonization of rabbit intestine. Infect Immun 65: 5222–5230. , , , , and (
- 2001) Genetic organization of the she pathogenicity island in Shigella flexneri 2a. Microb Pathog 30: 1–8. , , , , , and (
- 1990) Basic local alignment search tool. J Mol Biol 215: 403–410.DOI: 10.1006/jmbi.1990.9999 , , , , and (
- 1991) Current Protocols in Molecular Biology. Struhl, K. (ed.). New York: John Wiley & Sons. , , , , , and (
- 1972) Pedigrees of some mutant strains of Escherichia coli K-12. Bacteriol Rev 36: 525–557. (
- 1998) Distribution of chromosome length variation in natural isolates of Escherichia coli. Mol Biol Evol 15: 6–16. , and (
- 1997) The complete genome sequence of Escherichia coli K-12. Science 277: 1453– 1474. , , , , , et al. (
- 1994) Excision of large DNA regions termed pathogenicity islands from tRNA-specific loci in the chromosome of an Escherichia coli wild-type pathogen. Infect Immun 62: 606–614. , , , , , , and (
- 1969) A complementation analysis of the restriction and modification of DNA in Escherichia coli. J Mol Biol 41: 459–472. , and (
- 1998) The high-pathogenicity island of Yersinia pseudotuberculosis can be inserted into any of the three chromosomal asn tRNA genes. Mol Microbiol 30: 965–978. , , , , and (
- 1996) cag, a pathogenicity island of Helicobacter pylori, encodes type I-specific and disease-associated virulence factors. Proc Natl Acad Sci USA 93: 14648–14653. , , , , , et al. (
- 2001) Pathogenesis and evolution of virulence in enteropathogenic and enterohemorrhagic Escherichia coli. J Clin Invest 107: 539–548. , and (
- 1997) The locus of enterocyte effacement pathogenicity island of enteropathogenic Escherichia coli encodes secretion functions and remnants of transposons at its extreme right end. Gene 184: 107–114.DOI: 10.1016/s0378-1119(96)00581-1 , , and (
- 1998) The complete sequence of the locus of enterocyte effacement (LEE) from entero-pathogenic Escherichia coli E2348/69. Mol Microbiol 28: 1–4. , , , , , et al. (
- 1999a) Identification of CesT, a chaperone for the type III secretion of Tir in enteropathogenic Escherichia coli. Mol Microbiol 33: 1176–1189. , , , , , et al. (
- 1999b) The cloned locus of enterocyte effacement from enterohemorrhagic Escherichia coli O157: H7 is unable to confer the attaching and effacing phenotype upon E. coli K-12. Infect Immun 67: 4260–4263. , , and (
- 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: 6115–6126. , , , , , et al. (
- 1998) Genotypic and phenotypic changes in the emergence of Escherichia coli O157: H7. J Infect Dis 177: 1750–1753. , , , and (
- 1998) Enteropathogenic and enterohaemorrhagic Escherichia coli: more subversive elements. Mol Microbiol 30: 911–921.DOI: 10.1046/j.1365-2958.1998.01144.x , , , , , and (
- 1999) The integrase family of recombinase: organization and function of the active site. Mol Microbiol 33: 449–456. , and (
- 2000) Pathogenicity islands and the evolution of microbes. Annu Rev Microbiol 54: 641–679. , and (
- 1997) Pathogenicity islands of virulent bacteria: structure, function and impact on microbial evolution. Mol Microbiol 23: 1089–1097. , , , and (
- 1983) Studies on transformation of Escherichia coli with plasmids. J Mol Biol 166: 557–580. (
- 1980) A small cosmid for efficient cloning of large DNA fragments. Gene 11: 291–298. , and (
- 1971) Some morphological properties of P4 bacteriophage and P4 DNA. Virology 44: 67–72. , , , , and (
- 1995) Enteropathogenic Escherichia coli contains a putative type III secretion system necessary for the export of proteins involved in attaching and effacing lesion formation. Proc Natl Acad Sci USA 92: 7996–8000. , , , , , and (
- 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: 7839–7843. , , , and (
- 2001) A novel locus of enterocyte effacement (LEE) pathogenicity island inserted at pheV in bovine Shiga toxin-producing Escherichia coli strain O103: H2. FEMS Microbiol Lett 204: 75–79. , , , , and (
- 1999) Pathogenicity islands and other mobile genetic elements of diarrheagenic Escherichia coli. In Pathogenicity Islands and Other Mobile Virulence Elements. Kaper, J.B., Hacker, J. (eds). Washington, DC: American Society for Microbiology, pp. 33–58. , , and (
- 1997) Cloning of the RDEC-1 locus of enterocyte effacement (LEE) and functional analysis of the phenotype on HEp-2 cells. Adv Exp Med Biol 412: 241–245. , , , and (
- 1989) Infection by verocytotoxin-producing Escherichia coli. Clin Microbiol Rev 2: 15–38. (
- 1985) The association between idiopathic hemolytic uremic syndrome and infection by Verotoxin-producing Escherichia coli. J Infect Dis 151: 775–782. , , , , , and (
- 1997) Enteropathogenic E. coli transfers its receptor for intimin adherence into mammalian cells. Cell 91: 511–520. , , , , , and (
- 2000) A large toxin from pathogenic Escherichia coli strains that inhibits lymphocyte activation. Infect Immun 68: 2148– 2155. , , , , , , and (
- 1989) Actin accumulation at sites of bacterial adhesion to tissue culture cells: basis of a new diagnostic test for enteropathogenic and enterohemorrhagic Escherichia coli. Infect Immun 57: 1290–1298. , , , and (
- 2001) Mega2: Molecular Evolutionary Genetics Analysis Software. Tempe, Arizona: Arizona State University. , , , and (
- 2001) afa-8 Gene cluster is carried by a pathogenicity island inserted into the tRNA (Phe) of human and bovine pathogenic Escherichia coli isolates. Infect Immun 69: 937–948. , and (
- 1989) Dynamic, structural, and regulatory aspects of lambda site-specific recombination. Annu Rev Biochem 58: 913–949. (
- 1987) Escherichia coli that cause diarrhea: enterotoxigenic, enteropathogenic, enteroinvasive, enterohemorrhagic, and enteroadherent. J Infect Dis 155: 377– 389. (
- 1998) Homologous recombination between rrn operons rearranges the chromosome in host-specialized species of Salmonella. FEMS Microbiol Lett 164: 275–281. , and (
- 1993) Genomic mapping with I-Ceu I, an intron-encoded endonuclease specific for genes for ribosomal RNA. Salmonella spp., Escherichia coli, and other bacteria. Proc Natl Acad Sci USA 90: 6874–6878. , , and (
- 1987) Isolation and characterization of IS elements repeated in the bacterial chromosome. J Mol Biol 196: 445–455. , , , and (
- 1997) A cloned pathogenicity island from enteropathogenic Escherichia coli confers the attaching and effacing phenotype on E. coli K-12. Mol Microbiol 23: 399–407. , and (
- 1995) A genetic locus of enterocyte effacement conserved among diverse enterobacterial pathogens. Proc Natl Acad Sci USA 92: 1664–1668. , , , and (
- 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: 296–306. , , , , and (
- 2001) The pathogenicity islands (PAIs) of the uropathogenic Escherichia coli strain 536: island probing of PAI II536. J Infect Dis 183 (Suppl. 1), S17–S20. , , , , , and (
- 1988) A novel suicide vector and its use in construction of insertion mutations: osmoregulation of outer membrane proteins and virulence deter-minants in Vibrio cholerae requires. Toxr J Bacteriol 170: 2575–2583. , and (
- 1983) Attaching and effacing activities of rabbit and human enteropathogenic Escherichia coli in pig and rabbit intestines. Infect Immun 41: 1340–1351. , , , , and (
- 1995) Identification and cloning of a novel plasmid-encoded enterotoxin of enteroinvasive Escherichia coli and Shigella strains. Infect Immun 63: 4721–4728. , , , , , et al. (
- 2000) Identification of a novel locus that is required for in vitro adhesion of a clinical isolate of enterohaemorrhagic Escherichia coli to epithelial cells. Mol Microbiol 35: 288. , , and (
- 2000) Lateral gene transfer and the nature of bacterial innovation. Nature 405: 299–304.DOI: 10.1038/35012500 , , and (
- 1987) Enteric infections caused by non-enterotoxigenic Escherichia coli in animals: occurrence and pathogenicity mechanisms. A review. Vet Microbiol 14: 33–46. (
- 1998) Molecular evolution of a pathogenicity island from enterohemorrhagic Escherichia coli O157: H7. Infect Immun 66: 3810–3817. , , , , , , and (
- 1997) Versatile insertion plasmids for targeted genome manipulations in bacteria: isolation, deletion, and rescue of the pathogenicity island LEE of the Escherichia coli O157: H7 genome. J Bacteriol 179: 4426–4428. , , , and (
- 1997) Use of a novel approach, termed island probing, identifies the Shigella flexneri she pathogenicity island which encodes a homolog of the immunoglobulin A protease-like family of proteins. Infect Immun 65: 4606–4614. , , and (
- 2001) Integrative module of the high-pathogenicity island of Yersinia. Mol Microbiol 39: 407–415.DOI: 10.1046/j.1365-2958.2001.02227.x , , , and (
- 2000) Parallel evolution of virulence in pathogenic Escherichia coli. Nature 406: 64–67. , , , , and (
- 1987) Traditional enteropathogenic Escherichia coli of infantile diarrhea. Rev Infect Dis 9: 28–53. (
- 1994a) Host-specificity of enteropathogenic Escherichia coli from rabbits: lack of correlation between adherence in vitro and pathogenicity for laboratory animals. Infect Immun 62: 3329–3336. , , , and (
- 1994b) Adherence characteristics of attaching and effacing strains of Escherichia coli from rabbits. Infect Immun 62: 1584–1592. , , , , , , and (
- 1996) Enteropathogenic Escherichia coli-associated infections in rabbits: a model of human infection. Rev Microbiol (Sao Paulo) 27 (Suppl. 1), 112–116. , , , and (
- 1993) Site-specific genetic recombination: hops, flips, and flops. FASEB J 7: 760–767. (
- 1989) Mole-cular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press. , , and (
- 1996) Identification and characterization of the linear IS3 molecules generated by staggered breaks. J Biol Chem 271: 197–202. , , and (
- 1998) Characterization of the locus of enterocyte effacement (LEE) in different enteropathogenic Escherichia coli (EPEC) and Shiga-toxin producing Escherichia coli (STEC) serotypes. FEMS Microbiol Lett 164: 133–139. , , , , , and (
- 1996) Two pathogenicity islands in uropathogenic Escherichia coli J96: cosmid cloning and sample sequencing. Infect Immun 64: 3736–3743. , , , and (
- 1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22: 4673–4680. , , and (
- 1999) A novel chromosomal locus of enteropathogenic Escherichia coli (EPEC), which encodes a bfpT-regulated chaperone-like protein, TrcA, involved in microcolony formation by EPEC. Mol Microbiol 33: 741–752. , , , , , and (
- 2001) Nested deletions of the SRL pathogenicity island of Shigella flexneri 2a. J Bacteriol 183: 5535–5543. , , , and (
- 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: 1142–1150. , , and (
- 1997) Regulation of O-antigen chain length is required for Shigella flexneri virulence. Mol Microbiol 23: 765–775. , , and (
- 1997) Insertion site of the locus of enterocyte effacement in enteropathogenic and enterohemorrhagic Escherichia coli differs in relation to the clonal phylogeny of the strains. FEMS Microbiol Lett 156: 49–53. , , , and (
- 1990) Cloning of the genes for AF/R1 pili from rabbit enteroadherent Escherichia coli RDEC-1 and DNA sequence of the major structural subunit. Infect Immun 58: 1124–1128. , and (
- 1997) Site-specific recombination in plane view. Structure 5: 1401–1406. , and (
- 2001) Complete nucleotide sequence and analysis of the locus of enterocyte effacement from rabbit diarrheagenic Escherichia coli RDEC-1. Infect Immun 69: 2107–2115. , , , , , , and (
- Top of page
- Experimental procedures
- Supplementary material
- Supporting Information
Fig. S1. Alignment of the nucleotide and predicted amino acid sequences of espF of REPEC strains 84/110-1, 83/39 and RDEC-1. Fig. S2. Amino acid sequence alignment of Int-phe and homologous sequences identified in the databases.
|MMI_2968_sm.pdf||36K||Supporting info item|
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.