0095-1137/05/$08.00⫹0 doi:10.1128/JCM.43.11.5715–5720.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Distribution of
tccP
in Clinical Enterohemorrhagic and
Enteropathogenic
Escherichia coli
Isolates‡
Junkal Garmendia,
1† Zhihong Ren,
2† Sharon Tennant,
3† Monica Aparecida Midolli Viera,
4†
Yuwen Chong,
5† Andrew Whale,
1† Kristy Azzopardi,
7Sivan Dahan,
1Marcelo Palma Sircili,
6Marcia Regina Franzolin,
6Luiz R. Trabulsi,
6Alan Phillips,
5Taˆnia A. T. Gomes,
4Jianguo Xu,
2Roy Robins-Browne,
3and Gad Frankel
1*
Centre for Molecular Microbiology and Infection, Department of Biological Sciences, Imperial College London, London, United Kingdom1; State Key Laboratory for Infectious Disease Prevention and Control, National Institute for
Communicable Diseases Control and Prevention, China CDC, Beijing, China2; Department of Microbiology
and Immunology; University of Melbourne,3and Murdoch Childrens Research Institute, Royal Children’s
Hospital, Victoria,7Australia; Departamento de Microbiologia, Imunologia e Parasitologia,
Universidade Federal de Sao Paulo, Sao Paulo, Brazil4; Centre for Paediatric Gastroenterology,
Royal Free and University College Medical School, London, United Kingdom5; and
Laboratorio Especial de Microbiologia, Instituto Butanta, Sao Paulo, Brazil6
Received 14 April 2005/Returned for modification 7 July 2005/Accepted 17 August 2005
EnterohemorrhagicEscherichia coli(EHEC) and enteropathogenicE. coli(EPEC) are diarrheagenic patho-gens that colonize the gut through the formation of attaching and effacing lesions, which depend on the translocation of effector proteins via a locus of enterocyte effacement-encoded type III secretion system. Recently, two effector proteins, EspJ and TccP, which are encoded by adjacent genes on prophage CP-933U in EHEC O157:H7, have been identified. TccP consists of a unique N-terminus region and several proline-rich domains. In this project we determined the distribution oftccPin O157:H7, in non-O157 EHEC, and in typical and atypical EPEC isolates. All the EHEC O157:H7 strains tested weretccPⴙ. Unexpectedly,tccP was also
found in non-O157 EHEC, and in typical and atypical EPEC isolates, particularly in strains belonging to serogroups O26 (EHEC), O119 (typical EPEC), and O55 (atypical EPEC). We recorded some variation in the length of tccP, which reflects diversity in the number of the proline-rich repeats. These results show the existence of a class of “attaching and effacing” pathogens which express a combination of EPEC and EHEC virulence determinants.
Enteric bacteria have a conserved core genomic organiza-tion that is widespread in both commensal and pathogenic strains. The core genomic structure equips the bacteria with mechanisms required for survival in the gut and the ability to transfer between hosts and survive in the environment. In pathogenic bacteria, genetic islands are often associated with virulence and are dotted along the conserved core genome (6). Prominent examples of such pathogenic bacteria are the diar-rheal agents enteropathogenic Escherichia coli (EPEC) and enterohemorrhagic E. coli (EHEC), the latter also being known as verocytotoxin-producingE. colior Shiga toxin-pro-ducingE. coli(23).
EHEC strains cause a wide spectrum of illnesses ranging from mild diarrhea to severe diseases, such as hemorrhagic colitis and the hemolytic uremic syndrome (HUS). EHEC-induced HUS is now the leading cause of acute pediatric renal failure in developed countries and is caused by potent bacterial
verocytotoxins (VT), also known as Shiga toxins (Stx), which cross the gut epithelium and spread systemically to damage small blood vessels, in particular, those of the kidney. There are two major types of VT referred to as VT1 and VT2 (26). Strains of EHEC belonging to serogroup O157 are most commonly associated with severe human disease (29), but non-O157 EHEC are an emerging threat to human and animal health. EHEC can be isolated directly from the feces of large farm animals, in particular, calves and adult cattle, which rep-resent an important source of human infection (24). Humans can become infected with EHEC through the consumption of contaminated food or by direct transmission from infected animals or humans.
EPEC is a frequent cause of infantile diarrhea in the devel-oping world (4). EPEC strains are split into two evolutionally distinct lineages known as EPEC 1 (characteristically express-ing flagellar antigen H6) and EPEC 2 (characteristically ex-pressing flagellar antigen H2) (7). EPEC strains are also di-vided into typical (defined by the presence of the EAF plasmid which encodes the bundle-forming pilus) (13) and atypical (lacking the EAF plasmid) (27). The major O serogroups of typical EPEC are O55, O86, O111, O119, O127, and O142, while atypical EPEC strains consist of a large number of O serogroups (27).
The hallmark of EHEC and EPEC infections is their ability to colonize the intestinal mucosa and produce characteristic * Corresponding author. Mailing address: Centre for Molecular
Mi-crobiology and Infection, Flowers Building, Division of Molecular and Cellular Biology, Imperial College London, London SW7 2AZ, United Kingdom. Phone: 44 (0) 207594525. Fax: 44 (0) 2075943069. E-mail: g.frankel@imperial.ac.uk.
† J.G., Z.R., S.T., M.A.M.V., Y.C., and A.W. contributed equally to this study.
‡ This study is in memory of Luiz R. Trabulsi, who passed away in June 2005.
“attaching and effacing” (A/E) lesions (9, 23). A/E lesions are characterized by effacement of the brush border microvilli and intimate attachment of the bacterium to the enterocyte plasma membrane (21). In addition, infected cells produce elongated pedestal-like, actin-rich structures at the site of intimate bac-terial adhesion (9).
The EPEC and EHEC genes encoding A/E lesion formation and actin polymerization map mostly to a pathogenicity island termed the locus of enterocyte effacement (22). In common with other enteric pathogens (e.g., Salmonella, Shigella, and
Yersiniaspp.), EPEC and EHEC employ a type III secretion system (18) to inject effector virulence proteins into host cells which interfere with eukaryotic cell physiology and function. Within the locus of enterocyte effacement, Tir, Map, EspF, EspG, EspH, and EspZ are known effector proteins that are translocated into host cells (11). Importantly, none of the above-named effectors, except Tir (20), plays a direct role in A/E lesion formation and actin polymerization.
Following translocation, Tir is targeted to the plasma mem-brane, where it adopts a hairpin loop topology (16). The ex-tracellular, central, Tir domain binds the bacterial outer mem-brane adhesion molecule intimin (8), while the intracellular amino and carboxyl termini interact with a number of focal adhesion and cytoskeletal proteins linking the extracellular bacterium to the cell cytoskeleton (3). Although both pathways converge on the N-WASP and Arp2/3 complex, Tir-mediated actin accretion by EPEC and EHEC differ in that TirEPEC requires tyrosine (Y474) phosphorylation (19) and the N-WASP activator protein Nck (15), whereas TirEHECis lack-ing a Y474 equivalent and utilizes a bacterial-encoded N-WASP activator—the effector protein TccP (Tir-cytoskele-ton coupling protein) (12) (also known as EspFU) (2). TccP is only the second effector after Tir that plays a direct role in EHEC-induced actin polymerization.
TccP belongs to a growing number of prophage-carried ef-fector proteins that have been identified recently (reviewed in reference 11).tccPis carried on prophage CP-933U, forming an operon with the upstream geneespJ(10), which encodes another effector protein. EspJ was shown to influence the dynamics of clearance of the pathogen from the host’s intesti-nal tract, suggesting a role in host survival and pathogen trans-mission in vivo; EspJ is not required for A/E lesion formation in cultured cells and ex vivo (5). Significantly, whileespJhas been found in all the genomes of the sequenced A/E lesion-forming bacteria,tccPwas exclusively found in the genomes of two sequenced EHEC O157:H7 isolates, EDL933 (25) and Sakai (17) strains. The main aim of this study was to survey a collection of EHEC and EPEC strains (934 isolates in total), isolated from humans, animals, and food, for the presence of
tccPby PCR. The secondary aim of this investigation was to determine the prevalence ofespJamong thetccP⫹strains. To
this end, we formed a network of laboratories in Australia, Brazil, China, and the United Kingdom interested in the epi-demiology of EHEC and EPEC infection and standardized a PCR detection system fortccPandespJ.
MATERIALS AND METHODS
Bacterial strains.The bacterial strains used in this study included 365 EHEC O157, 200 non-O157 EHEC, 105 typical EPEC, and 264 atypical EPEC strains. These strains are part of laboratory collections in the participating institutes.
Detection oftccPandespJby PCR.tccPwas amplified by colony PCR using tccP-F1 (5⬘-ATGATTAACAATGTTTCTTCACTT) and tccP-R1 (5⬘-TCACGA GCGCTTAGATGTATTAATGCC) forward and reverse primers (30 cycles of 94°C for 1 min, 58°C for 1 min, and 72°C for 1 min), which anneal to the 5⬘and 3⬘ends oftccP, respectively. The identity of representative PCR products was confirmed by DNA sequencing.espJwas amplified using primers to detect the EHEC O157:H7espJgene (forward primer 5⬘ATGTCAATTATAAAAAACT GCTTATC and reverse primer 5⬘TTTTTTGAGAGGATATATGTCAAC) (16, 24) (30 cycles of 94°C for 1 min, 60°C for 1 min, and 72°C for 1 min) and the EPEC O127:H6 espJgene (forward primer 5⬘ ATGCCAATCATAAAGAA CTGC and reverse primer 5⬘ TTTTTTGAGTGGGTGGATAT) (http://www .sanger.ac.uk/Projects/Escherichia_Shigella/) (30 cycles of 94°C for 1 min, 52°C for 1 min, and 72°C for 1 min). Genomic DNA from the EHEC O157:H7 EDL933 and EPEC O126:H6 E2348/69 strains were used as controls; of note is the fact that due to sequence variation the primers used in this screen might not amplify all thetccP/espJvariants.
Detection of Tir tyrosine phosphorylation by immunofluorescence. HEp-2 cells were grown in Dulbecco’s modified Eagle medium supplemented with 10% fetal calf serum and 2 mM glutamine at 37°C in 5% CO2. Cells were seeded onto glass coverslips (12-mm diameter) in 24-well plates at a density of 5⫻104cells per well. Bacteria for infection assays were grown in brain heart infusion over-night standing cultures at 37°C and added to HEp-2 cells at a multiplicity of infection of 100:1. Monolayers were infected for 6 h, fixed in formalin for 15 min at room temperature, and permeabilized in 0.1% Triton X-100. The cells were then incubated for 60 min at room temperature with mouse monoclonal an-tiphosphotyrosine clone PT-66 (Sigma) (diluted 1:40), washed with phosphate-buffered saline, and incubated for a further 30 min with fluorescein isothiocya-nate-conjugated goat anti-mouse antibody (Sigma) (diluted 1:40). Coverslips were washed, mounted on Aqua Poly/Mount (Polysciences) and analyzed using a Zeiss Universal fluorescence microscope.
Nucleotide sequence accession numbers.The accession numbers of the dif-ferenttirsequences are DQ007019 to DQ007024.
RESULTS AND DISCUSSION
Prevalence oftccPin EHEC O157:H7.A total of 365 EHEC O157:H7 strains were tested (Table 1), of which 234 were isolated in China (30, 31). Of the 234 strains, 191 carried the VT2 gene, 42 strains carried both VT1 and VT2 genes, and only one strain carried the VT1 gene. All 234 strains were
tccP⫹
/espJ⫹
. However, while 32 of the 191 VT2 gene-positive strains (16.2%) carried atccPgene that was smaller (ranging from 700 to 1,000 bp) than the prototype gene (1,150 bp) present in EHEC strain EDL933 (Z3072) (25), no variability in the length oftccPwas recorded among strains carrying both VT1 and VT2 genes. Interestingly,tccPin EHEC strain Sakai is only 1,014 bp long (ECs2715) (17).
Sequencing the 700-bp (two strains), 1,000-bp (one strain), and 1,150-bp (one strain) PCR products revealed that the predicted proteins consist of the 87 N-terminal amino acids and three, five, and six proline-rich repeats, respectively.
An additional 127 EHEC O157:H7 strains were tested in Australia (Table 1). The strains were isolated in different coun-tries and were of human, animal, or food origin. Nineteen of the strains carried the VT1 gene, 34 strains carried both the VT1 and VT2 genes, and 18 strains carried the VT2 gene. The VT gene type was not determined for 56 strains. All 127 strains weretccP⫹
strains, regardless of their country of origin, source, or VT gene type; all the strains yielded a PCR fragment of the prototype tccP gene (1,150 bp). All of the EHEC O157:H7 strains tested in Australia were alsoespJ⫹.
Four O157:H7 strains (originating in the United States, Ar-gentina, Egypt, and Denmark) were tested in Brazil and found to betccP⫹(1,150 kb)/espJ⫹.
study weretccP⫹
/espJ⫹
, and some variation in thetccPgene length was recorded.
Prevalence oftccPin non-O157 EHEC.tccPwas detected in 28 of 200 (14%) non-O157 EHEC (eae-positive/VT gene-pos-itive) strains, belonging to different serogroups (Table 2). In particular,tccPwas detected in five O26 isolates (four of which yielded a typical 1,150-bp PCR fragment and one of which yielded an 850-bp PCR fragment) and in four O5 strains (two of which yielded a typical 1,150-bp PCR fragment and two of which yielded a 1,000-bp PCR fragment). All ninetccP⫹strains
wereespJ⫹
. O5 and O26 EHEC strains are commonly associ-ated with diarrhea in farm animals, which imposes a significant economic burden on livestock producers. Only one of 47 EHEC O111 isolates examined wastccP⫹, producing a PCR
fragment of 1,000 bp. In contrast, all nine O172 EHEC strains isolated from cows in Hong Kong, and three O98 EHEC strains isolated from the same HUS case (but at different times during the infection), were tccP⫹. The O172 EHEC strains
yielded a PCR product of 700 bp, whereas the O98 EHEC strains produced a PCR product of 850 bp. EHEC O156 (one
isolate), nontypeable (NT) (four isolates), and rough (R) (one isolate) strains were alsotccP⫹(espJ⫹); three of the NT strains
and the R strain yielded an 850-bptccPPCR fragment. Sequencing the PCR products oftccPfrom strains O4-7G-10 (O172 and VT2 gene [cow]) and 9600608 (O156 [human]) revealed that the predicted proteins consist of the N-terminus domain and three and two proline-rich repeats, respectively.
Considering that EHEC O157:H7 is a recently emerged clone, the finding that all the isolates weretccP⫹
was antici-pated. In contrast, the presence oftccPin EHEC strains be-longing to serogroups other than O157 was unexpected, par-ticularly as strains for whichtirsequences are available (e.g., O26 and O111) possess a tyrosine residue homologous to that of EPEC strain O127:H6 Y474. The function of TccP in these strains and its contribution to commensalism, colonization, and infection of either animal or human intestine are intrigu-ing subjects for further studies.
A total of 172 non-O157 EHEC isolates, from similar origins and sources, were surveyed and found to lack tccP. These strains belonged to the following serogroups (the number of TABLE 1. Distribution oftccPamong 361 EHEC O157 strains
No. of
strains Country(s) of origin Source(s) VT type tccP
a espJ
177 China, Australia, Hong Kong Human, pig, sheep, goat, cow, calf, chicken, rabbit, goose, unspecified, pickle
2 ⫹ ⫹
3 China Human, sheep, cow 2 ⫹(1,000 bp) ⫹
1 China Sheep, goat 2 ⫹(850 bp) ⫹
28 China Human, pig, sheep, goat, cow, calf, chicken, dung, beetle, unspecified
2 ⫹(700 bp) ⫹
76 China, Australia, USA,e
New Zealand, South Africa, Swaziland, Finland
Human, pig, sheep, goat, cow, calf, chicken, fly, water, vegetable cooked food, food,c
unknown
1, 2 ⫹ ⫹
20 China, Australia Unspecificied, animal,dfood, human 1 ⫹ ⫹
56 Australia, USA, New Zealand, South Africa, Swaziland
Human, food, animal NDb ⫹ ⫹
aPCR fragments that differ from the prototypetccPgene of EDL933/Sakai are indicated in parentheses. bND, VT type not determined.
cFood category includes water.
dAnimal refers mostly to cattle (three isolates were from sheep; one was from a water buffalo). eUSA, United States.
TABLE 2. Characteristics of 28tccP⫹non-O157 EHEC strains
Serotype
(no. of strains) Country(s) of source VT gene type tccP
aorigin and/or espJ
O5 (2) Australia/USA/cowc 1 ⫹ ⫹
O5 (1) USA/cow NDb ⫹(1,000 bp) ⫹
O5:NM (1) Australia/cow 1, 2 ⫹(1,000 bp) ⫹
O26:H21 (1) Australia ND ⫹ ⫹
O26:H21 (1) Australia ND ⫹(850 bp) ⫹
O26 (1) Australia/HUS 1 ⫹ ⫹
O26:H11 (2) USA ND ⫹ ⫹
O98 (3) Australia/HUS 1, 2 ⫹(850 bp) ⫹
O111:NM (1) USA/cow 1 ⫹(1,000 bp) ⫹
O156:H25 (1) Australia/human ND ⫹(550 bp) ⫹
O172:NM (9) Hong Kong/cow 2 ⫹(700 bp) ⫹
NT (3) Hong Kong/cow 2 ⫹(850 bp) ⫹
NT (1) Australia/food 2 ⫹ ⫹
R (1) Hong Kong/cow 2 ⫹(850 bp) ⫹
aPCR fragments that differ from the prototypetccPgene of EDL933/Sakai are indicated in parentheses. bND, VT gene typing not done.
strains is indicated in brackets): O1 (1), O5 (10), O9 (1), O15 (12), O16 (5), O25 (1), O26 (22), O111 (46), O112 (1), O113 (20), O116 (1), O118 (2), O121 (2), O123 (1), O128 (12), O130 (2), O145 (5), O147 (1), O153 (2), O163 (6), O168 (1), NT (13), and R (5).
Prevalence of tccP in atypical O55:H7 EPEC strain. An atypical EPEC strain belonging to serotype O55:H7 is believed to be the progenitor from which EHEC O157:H7 evolved (28). Therefore, it was of particular interest to determine the distri-bution oftccP among clinical O55:H7 isolates. We tested 18 isolates from Australia, Brazil, Thailand, and the United King-dom (Table 3). Of these, 15 (83.3%) weretccP⫹, giving rise to
PCR products of 850 to 1,300 bp (Table 3). Three O55:H7 strains (isolated in Brazil, Thailand, and United Kingdom) lackedtccPbut wereespJ⫹
.
Consistent with these results, sequence determination of the 3⬘end oftirof the three strains lackingtccPshowed that they all contain an EPEC-like Tir (i.e., having a tyrosine equivalent to Y474 of EPEC O127:H6 strain E2348/69). In contrast, the sequence of the 3⬘end oftirfrom representativetccP⫹strains
revealed that they all contain an EHEC (O157:H7)-like Tir (i.e., the equivalent position was occupied by serine) (Fig. 1A). In addition, application of phaloidine actin-staining and phos-photyrosine-specific antibodies (12) to HEp-2 cells infected with the O55:H7 strains revealed that while all the strains triggered actin polymerization under attached bacteria (data not shown), specific phosphotyrosine staining was seen only with the O55 strains lackingtccP; no phosphotyrosine staining was observed following infection with the O55:H7tccP⫹
strains (Fig. 1B). These results show a correlation between Tir ty-rosine phosphorylation and the absence of tccP and, con-versely, between the absence of Tir tyrosine phosphorylation and possession oftccP. The results also show the existence of distinct EPEC-like and EHEC-like O55:H7 clones; it is likely that only a clone of the latter category was the progenitor of EHEC O157:H7.
Prevalence oftccPin typical EPEC.We screened 105 typical EPEC strains (defined aseaepositive, EAF positive, and VT gene negative) belonging to the classical EPEC serogroups for the presence oftccP(Table 4). Out of the 105 strains 21 (20%) weretccP⫹
strains. Twenty of thetccP⫹
strains were isolated in South America (Brazil and Chile), and one strain was isolated in Australia. In particular, 19 of the typicaltccP⫹
EPEC strains belonged to a single serotype, O119:H6. Two O119:H6 strains yielded a tccP PCR fragment of the prototype length, one isolate yielded a 550-bp PCR fragment, and another isolate yielded a 1-kb PCR fragment. A PCR product of 850 bp was
generated from 15 O119:H6 strains; significantly, two of these strains lacked espJ. In addition, one isolate each of EPEC O55:H- (850 bp) and O86:H34 (850 bp) weretccP⫹
. The latter strain lackedespJ. While the O55:H- strain, like EHEC O157: H7, expressed the intimin gamma type, the O86:H34 strain expresses intimin type delta and was the only tccP⫹ strain
expressing this intimin type in this survey.
FIG. 1. (A) Amino acid alignment of the 12-amino-acid C-terminus region containing the Nck binding site of TirEPECand Y474 (3). EPEC O55:H7 strain CPG123-G58 (lacking tccP) (accession number DQ007022) has a sequence identical to that of EPEC O127:H6 strain E2348/69 (accession number AF02236). The sequence of the equiva-lent region of TirEHECO157:H7 strain EDL93 (24) is identical to that of EPEC O55:H7 strain CPG122-G57 (tccP⫹) (accession number
DQ007019). (B) Phase contrast image showing adhesion of atypical EPEC O55:H7 bacteria withouttccP(i) and withtccP(iii) to HEp-2 cells (arrows). Immunofluorescent staining, using antiphosphotyrosine antibody, shows accumulation of tyrosine-phosphorylated proteins be-neath attached bacteria in cells infected with O556:H7 lackingtccP(ii) (arrow) but not withtccP⫹O55:H7 EPEC (iv). Both strains triggered
actin polymerization under attached bacteria (not shown). TABLE 3. Distribution oftccPin 18 EPEC O55:H7 strains
No. of
strains Origin
eaetype/tir
(no. of strains)b tccPa espJ
2 Australia, UKc ␥
/EHEC (1) ⫹ ⫹
8 Brazil ␥/EHEC (2) ⫹(1,000 bp) ⫹
1 Thailand ␥/EHEC (1) ⫹(1,300 bp) ⫹
4 Australia, UK, Brazil ␥/EHEC (1) ⫹(850 bp) ⫹
3 Thailand, UK, Brazil ␥/EPEC (3) ⫺ ⫹
aPCR fragments that differ from the prototype
tccPgene of EDL933/Sakai are indicated in parentheses.
bCorresponds to the presence of either the
tirEHECortirEPECgene.
Sequencing the PCR product oftccPfrom strain O119:H6 (850 bp) revealed that the predicted protein consists of the N-terminus region and four proline-rich repeats. In addition to TccP, this strain expresses a Tir that contains a Y474 equiva-lent.
EPEC O119:H6 is associated with diarrhea endemic in Bra-zil (14). This serogroup belongs to the EPEC 1 lineage, which is typically characterized by H6 and intimin type alpha. How-ever, clinical isolates of O119:H6 are unusual among EPEC 1 strains because, although expressing the flagellar antigen H6, they encode intimin subtype beta (1). The fact that these strains are alsotccP⫹ suggests that they have evolved by
ac-quiring diverse virulence determinants from various sources. Revealing the physiological relevance oftccPin these isolates will show whether EPEC O119:H6 represents a group of strains undergoing specialization through acquisition of new virulence determinants and, if so, in which niches or under which environmental conditions thetccP/espJlocus increases fitness and competitiveness. The fact that 2 of the 18tccP⫹
O119:H6 EPEC strains lackespJsuggests that different selec-tive pressures may operate on these adjacent genes or, alter-natively, that the lack ofespJis a consequence of recombina-tion and/or delerecombina-tions.
A total of 84 typical EPEC isolates, from similar origin and sources, were surveyed and found to lacktccP. The serogroups of these strains were as follows (with the number of strains indicated in brackets): O55 (13), O86 (7), O88 (6), O111 (15), O114 (5), O119 (9), O126 (1), O127 (10), O142 (14), and O145 (4). Importantly, nine O119:H6 lacked tccP, suggesting the existence of several clones within this serotype.
Prevalence oftccPin atypical EPEC.In addition to the 18 O55:H7 strains (Table 3), we screened for the presence oftccP
in 246 isolates belonging to the atypical EPEC category (Table 5). We found 10 (4.1%)tccP⫹atypical EPEC strains; 3
pro-duced a prototype tccP PCR fragment, 5 produced a PCR fragment of 850 bp; 1 produced a PCR product of 1 kb, and another produced a PCR product of 1.3 kb. Three of the strains carried the gene for intimin beta, one possessed the gene for intimin gamma, three carried an NT intimin gene, and in three strains the intimin gene was not classified (Table 5).
A total of 236 atypical EPEC isolates, from similar origin and sources, were surveyed and found to lack tccP. These strains belonged to the following serogroups (with the number of strains indicated in brackets): O2 (6), O4 (4), O5 (1), O8 (2), O11 (1), O15 (5), O16 (1), O20 (1), O25 (2), O26 (11), O28 (1),
O33 (2), O45 (1), O49 (1), O51 (9), O60 (1), O63 (2), O70 (1), O71 (1), O88 (2), O91 (1), O98 (2), O101 (1), O103 (2), O104 (2), O106 (1), O107 (1), O109 (1), O111 (13), O116 (1), O117 (1), O123 (1), O124 (1), O125 (6), O128 (9), O139 (1), O149 (1), O153 (4), O154 (1), O161 (1), O162 (2), O70/172 (3), NT (85), and R (40).
The prototype tccPgene of EDL933 (1,150 bp) encodes a unique 87-amino-acid terminal domain believed to encompass the secretion and translocation signal and six identical proline-rich repeats, each comprising 47 amino acids (141 bp) (12). On the basis of the distinct PCR fragments that diverged from the prototype gene we assumed that the strains producing shorter PCR fragments represent genes missing one (1,000-bp), two (850-bp), or three (700-bp) repeats, while in strains harboring a longertccPgene the number of repeats is greater (1,300 bp). Sequencing representativetccPamplicons confirmed this hy-pothesis. We are currently determining the minimum number of repeats needed for activity.
Results from the atypical EPEC serogroups indicate that, with the exception of serotype O55:H7, the prevalence oftccP
is low (⬍5%). This suggests that acquisition of prophage CP-933U is a current process in the evolution of these pathogens or that there is no selective advantage for strains harboring CP-933U. For unknown reasons we found that among the typical EPEC stains,tccPis most prevalent in serotype O119: H6. These results show the presence of a class of A/E bacteria that express a combination of EPEC and EHEC virulence determinants. Although previous studies of A/E lesion forma-tion have emphasized the divergence between EPEC and EHEC, highlighting the different methods of actin accumula-tion, the current study indicates a convergence between the two categories. The convergence is perhaps not too surprising in view of the retention of A/E capacity when virulence factors, e.g., intimin, EspA, and EspB, are exchanged between EPEC and EHEC or of the fact that O157:H7 TccP can substitute for host-derived Nck in EPEC infection of Nck-deficient cell lines (2, 12).
The benefit of having TccP in strains expressing tyrosine (Y474)-phosphorylated Tir (such as typical EPEC O119:H6; data not shown) is not apparent. However, it could be advan-TABLE 4. Characteristics of 21tccP⫹typical EPECastrains
Serotype
(no. of strains) Origin eaetype tccP
b espJ
O55:H- (1) Brazil ␥ ⫹(850 bp) ⫹
O86:H34 (1) Brazil ␦ ⫹(850 bp) ⫺
0119:H6 (1) Brazil  ⫹(550 bp) ⫹
0119:H6 (2) Brazil  ⫹(850 bp) ⫺
0119:H6 (13) Brazil  ⫹(850 bp) ⫹
0119:H6 (2) Chile NDc ⫹ ⫹
0119:H6 (1) Australia ND ⫹(1,000 bp) ⫹
aIdentified from a sample of 105 typical EPEC strains of human origin. bPCR fragments that differ from the prototype
tccPgenes of EDL933/Sakai are indicated in parentheses.
cND,
eaetyping not done.
TABLE 5. Characteristics of 10tccP⫹atypical EPECastrains
Serotype (no. of strains)
Origin eaetype tccPb espJ
O11:H16 (1) Brazil  ⫹(1,000 bp) ⫹
O13:H11 (1) Brazil NTc ⫹(850 bp) ⫹
O55:H6 (1) Australia ␥ ⫹ ⫹
O55 (1) UKe NDd ⫹ ⫹
O55 (1) UK ND ⫹(1,300 bp) ⫹
O111 (1) Australia ND ⫹ ⫹
O153:H7 (1) Brazil  ⫹(850 bp) ⫹
NT:H2 (1) Brazil NT ⫹(850 bp) ⫹
NT:H- (1) Brazil NT ⫹(850 bp) ⫹
NT:H- (1) Brazil  ⫹(850 bp) ⫹
aIdentified from a sample of 246 atypical EPEC strains of human origin. bPCR fragments that differ from the prototypetccPgene of EDL933/Sakai are
indicated in parentheses.
tageous to the bacterium, as A/E lesion formation and coloni-zation might be more efficient if the adaptor linking Tir to the actin cytoskeleton were cotranslocated with Tir or if Tir were coupled to the cytoskeleton simultaneously by Nck and TccP. Alternatively, the selective pressure to maintain the prophage might be driven from another, yet undefined, prophage-carried gene. Further surveys of EPEC and EHEC strains, together with functional analysis of TccP in the EPEC background, are needed in order to elucidate the role played by TccP in these unique bacterial strains.
ACKNOWLEDGMENTS
This project was supported by the Wellcome Trust (GF) (as part of the International Partnership Research Award in Veterinary Epide-miology project entitled “EpideEpide-miology and Evolution of Enterobac-teriaceae Infections in Humans and Domestic Animals”); The Pro-grama de Apoio a Nu´cleos de Exceleˆncia—PRONEX MCT/CNPq/ FAPERJ (T.G.); The National Natural Science Foundation of China (30070042, 30371279) (J.X.); and the Australian National Health and Medical Research Council (R.R.-B.).
REFERENCES
1.Adu-Bobie, J., G. Frankel, C. Bain, A. G. Goncaleves, L. R. Trabulsi, G. Douce, S. Knutton, and G. Dougan.1998. Detection of intimin␣,,␥, and ␦, four intimin derivatives expressed by attaching and effacing microbial pathogens. J. Clin. Microbiol.36:662–668.
2.Campellone, K. G., R. Robbins, and J. M. Leong.2004. EspFUis a translo-cated EHEC effector that interacts with Tir and N-WASP and promotes Nck-independent actin assembly. Dev. Cell7:217–228.
3.Campellone, K. G., and J. M. Leong.2003. Tails of two Tirs: actin pedestal formation by enteropathogenicE. coliand enterohemorrhagicE. coliO157: H7. Curr. Opin. Microbiol.6:82–90.
4.Chen, H. D., and G. Frankel.2005. EnteropathogenicEscherichia coli: un-ravelling pathogenesis. FEMS Microbiol. Rev.29:83–98.
5.Dahan, S., S. Wiles, R. M. La Ragione, A. Best, M. J. Woodward, M. P. Stevens, R. K. Shaw, Y. Chong, S. Knutton, A. Phillips, and G. Frankel.
2005. EspJ is a prophage-encoded type III effector protein of attaching and effacing pathogens that modulates infection dynamics. Infect. Immun.73:
679–686.
6.Dobrindt, U., B. Hochhut, U. Hentschel, and J. Hacker.2004. Genomic islands in pathogenic and environmental microorganisms. Nat. Rev. Micro-biol.2:414–424.
7.Donnenberg, M. S., and T. S. Whittam.2001. Pathogenesis and evolution of virulence in enteropathogenic and enterohemorrhagic Escherichia coli. J. Clin. Investig.107:539–548.
8.Frankel, G., A. D. Phillips, L. R. Trabulsi, S. Knutton, G. Dougan, and S. Matthews.2001. Intimin and the host cell—is it bound to end in Tir(s)? Trends Microbiol.9:214–218.
9.Frankel, G., A. D. Phillips, I. Rosenshine, G. Dougan, J. B. Kaper, and S. Knutton.1998. Enteropathogenic and enterohaemorrhagicEscherichia coli: more subversive elements. Mol. Microbiol.30:911–921.
10.Garmendia, J., and G. Frankel.2005. Operon structure and gene expression of the espJ-tccPlocus of enterohaemorrhagic Escherichia coliO157:H7. FEMS Microbiol. Lett.247:137–145.
11.Garmendia, J., G. Frankel, and V. F. Crepin.2005. Enteropathogenic and enterohemorrhagicE. coliinfections: translocation, translocation, transloca-tion. Infect. Immun.73:2573–2585.
12.Garmendia, J., A. Phillips, Y. Chong, S. Schuller, O. Marches, S. Dahan, E. Oswald, R. K. Shaw, S. Knutton, and G. Frankel.2004. TccP is an entero-haemorrhagicE. coliO157:H7 type III effector protein that couples Tir to the actin-cytoskeleton. Cell Microbiol.6:1167–1183.
13.Giron, J. A., A. S. Ho, and G. K. Schoolnik.1991. An inducible bundle-forming pilus of enteropathogenicEscherichia coli. Science254:710–713. 14.Gomes, T. A., M. A. Vieira, I. K. Wachsmuth, P. A. Blake, and L. R. Trabulsi.
1989. Serotype-specific prevalence ofEscherichia coli strains with EPEC adherence factor genes in infants with and without diarrhea in Sao Paulo, Brazil. J. Infect. Dis.160:131–135.
15.Gruenheid, S., R. DeVinney, F. Bladt, D. Goosney, S. Gelkop, G. D. Gish, T. Pawson, and B. B. Finlay.2001. EnteropathogenicE. coliTir binds Nck to initiate actin pedestal formation in host cells. Nat. Cell Biol.3:856–859. 16.Hartland, E. L., M. Batchelor, R. M. Delahay, C. Hale, S. Matthews, G.
Dougan, S. Knutton, I. Connerton, and G. Frankel.1999. Binding of intimin from enteropathogenicEscherichia colito Tir and to host cells. Mol. Micro-biol.32:151–158.
17.Hayashi, T., K. Makino, M. Ohnishi, K. Kurokawa, K. Ishii, K. Yokoyama, C. G. Han, E. Ohtsubo, K. Nakayama, T. Murata, M. Tanaka, T. Tobe, T. Iida, H. Takami, T. Honda, C. Sasakawa, N. Ogasawara, T. Yasunaga, S. Kuhara, T. Shiba, M. Hattori, and H. Shinagawa.2001. Complete genome sequence of enterohemorrhagicEscherichia coliO157:H7 and genomic com-parison with a laboratory strain K-12. DNA Res.28:11–22.
18.Hueck, C. J.1998. Type III protein secretion systems in bacterial pathogens of animals and plants. Microbiol. Mol. Biol. Rev.62:379–433.
19.Kenny, B.1999. Phosphorylation of tyrosine 474 of the enteropathogenic Escherichia coli(EPEC) Tir receptor molecule is essential for actin nucle-ating activity and is preceded by additional host modifications. Mol. Micro-biol.31:1229–1241.
20.Kenny, B., R. DeVinney, M. Stein, D. J. Reinscheid, E. A. Frey, and B. B. Finlay.1997. EnteropathogenicE. coli(EPEC) transfers its receptor for intimate adherence into mammalian cells. Cell91:511–520.
21.Knutton, S., D. R. Lloyd, and A. S. McNeish.1987. Adhesion of entero-pathogenicEscherichia colito human intestinal enterocytes and cultured human intestinal mucosa. Infect. Immun.55:69–77.
22.McDaniel, T. K., K. G. Jarvis, M. S. Donnenberg, and J. B. Kaper.1995. A genetic locus of enterocyte effacement conserved among diverse enterobac-terial pathogens. Proc. Natl. Acad. Sci. USA92:1664–1668.
23.Nataro, J. P., and J. B. Kaper.1998. DiarrheagenicEscherichia coli. Clin. Microbiol. Rev.11:142–201.
24.Pearce, M. C., C. Jenkins, L. Vali, A. W. Smith, H. I. Knight, T. Cheasty, H. R. Smith, G. J. Gunn, M. E. Woolhouse, S. G. Amyes, and G. Frankel.
2004. Temporal shedding patterns and virulence factors ofEscherichia coli serogroups O26, O103, O111, O145, and O157 in a cohort of beef calves and their dams. Appl. Environ. Microbiol.70:1708–1716.
25.Perna, N. T., G. R. Plunkett, V. Burland, B. Mau, J. D. Glasner, D. J. Rose, G. F. Mayhew, P. S. Evans, J. Gregor, H. A. Kirkpatrick, G. Posfai, J. Hackett, S. Klink, A. Boutin, Y. Shao, L. Miller, E. J. Grotbeck, N. W. Davis, A. Lim, E. T. Dimalanta, K. D. Potamousis, J. Apodaca, T. S. Ananthara-man, J. Lin, G. Yen, D. C. Schwartz, R. A. Welch, and F. R. Blattner.2001. Genome sequence of enterohaemorrhagicEscherichia coliO157:H7. Nature
409:529–533.
26.Thorpe, C. M.2004. Shiga toxin-producingEscherichia coliinfection. Clin. Infect. Dis.38:1298–1303.
27.Trabulsi, L. R., R. Keller, and T. A. Tardelli Gomes.2002. Typical and atypical enteropathogenicEscherichia coli. Emerg. Infect. Dis.8:508–513. 28.Wick, L. M., W. Qi, D. W. Lacher, and T. S. Whittam.2005. Evolution of
genomic content in the stepwise emergence ofEscherichia coliO157:H7. J. Bacteriol.187:1783–1791.
29.Willshaw, G. A., T. Cheasty, H. R. Smith, O. B. S.J., and G. K. Adak.2001. Verocytotoxin-producingEscherichia coli(VTEC) O157 and other VTEC from human infections in England and Wales: 1995 to 1998. J. Med. Micro-biol.50:135–142.
30.Xu, J., Q. Liu, H. Jing, B. Pang, J. Yang, G. Zhao, and H. Li.2003. Isolation ofEscherichia coliO157:H7 from dung beetlesCatharsius molossus. Micro-biol. Immunol.47:45–49.
