Enterohaemorrhagic and enteropathogenic Escherichiacoli use a different Tir-based mechanism for pedestalformation

Rebekah DeVinney,1*† Jose Luis Puente,2

Annick Gauthier,3 Danika Goosney4 and

B. Brett Finlay1,3,4

1Biotechnology Laboratory, University of British Columbia,

Vancouver, BC, Canada, V6T 1Z3.2Molecular Microbiology Department, Instituto de

Biotecnologıa, Universidad Nacional Autonoma de

Mexico, Cuernavaca, Morelos 62250, Mexico.

Departments of 3Biochemistry and 4Microbiology and

Immunology, University of British Columbia, Vancouver,

BC, Canada, V6T 1Z3.

Summary

Enterohaemorrhagic Escherichia coli (EHEC) adheres

to the host intestinal epithelium, resulting in the

formation of actin pedestals beneath adhering

bacteria. EHEC and a related pathogen, enteropatho-

genic E. coli (EPEC), insert a bacterial receptor, Tir,

into the host plasma membrane, which is required for

pedestal formation. An important difference between

EPEC and EHEC Tir is that EPEC but not EHEC Tir is

tyrosine phosphorylated once delivered into the host.

In this study, we assessed the role of Tir tyrosine

phosphorylation in pedestal formation by EPEC and

EHEC. In EPEC, pedestal formation is absolutely

dependent on Tir tyrosine phosphorylation and is not

complemented by EHEC Tir. The protein sequence

surrounding EPEC Tir tyrosine 474 is critical for Tir

tyrosine phosphorylation and pedestal formation by

EPEC. In contrast, Tir tyrosine phosphorylation is not

required for pedestal formation by EHEC. EHEC forms

pedestals with both wild-type EPEC Tir and the non-

tyrosine-phosphorylatable EPEC Tir Y474F. Pedestal

formation by EHEC requires the type III delivery of

additional EHEC factors into the host cell. These

findings highlight differences in the mechanisms of

pedestal formation by these closely related patho-

gens and indicate that EPEC and EHEC modulate

different signalling pathways to affect the host actin

cytoskeleton.

Introduction

Enterohaemorrhagic Escherichia coli (EHEC) O157:H7 is

a significant cause of food- and water-borne illness in

North America, as evidenced by the recent outbreak in

Walkerton, Ontario, which affected close to 2000

individuals and resulted in seven deaths (Griffin and

Tauxe, 1991; Kondro, 2000). EHEC infection initially

results in a watery diarrhoea, which can progress to severe

bloody diarrhoea (haemorrhagic colitis) in patients of all

ages (Nataro and Kaper, 1998). In young children and the

elderly, EHEC can cause a potentially fatal systemic

complication, the haemolytic uraemic syndrome (HUS),

characterized by haemolytic anaemia, thrombocytopenia

and renal failure (Griffin et al., 1988; Nataro and Kaper,

1998). EHEC is a member of a family of pathogens that

cause attaching and effacing (A/E) lesions. A/E lesions are

characterized by the degeneration of the epithelial cell

microvilli and the formation of actin-rich pedestals within

the host enterocytes beneath the adherent bacteria (Moon

et al., 1983; Knutton et al., 1989). This family of pathogens

includes enteropathogenic E. coli (EPEC), a major cause

of infantile diarrhoea in the developing world, rabbit, dog

and porcine EPEC strains and Citrobacter rodentium,

which causes colonic hyperplasia in mice (Cantey and

Blake, 1977; Beaudry et al., 1996; Nataro and Kaper,

1998; Newman et al., 1999).

Central to the process of A/E lesion formation by this

family of pathogens is Tir, the translocated intimin

receptor. Tir is a 72–78 kDa bacterial protein that is

inserted into the host cell plasma membrane, where it is

phosphorylated and becomes the receptor for the bacterial

outer membrane protein, intimin (Kenny et al., 1997; Ebel

et al., 1998; DeVinney et al., 1999). Intimin binding to Tir

leads to the dramatic rearrangement of the host actin

cytoskeleton, which forms a pedestal beneath the

adherent bacteria (Kenny et al., 1997; DeVinney et al.,

1999). Tir delivery into the host cell is mediated by the

type III secretion apparatus and is dependent on the E. coli

secreted proteins EspA, EspB and EspD, which are pro-

posed to form the type III translocon (Kenny et al., 1997;

Taylor et al., 1998; Wolff et al., 1998; Kresse et al., 1999;

Wachter et al., 1999). EHEC and EPEC strains that are

lacking, or are unable to deliver, Tir do not form pedestals

in cultured epithelial cells (Kenny et al., 1997; DeVinney

Accepted 13 July, 2001. *For correspondence. E-mail rdevinne@ucalgary.ca; Tel. (11) 403 220 4095; Fax (11) 403 270 2772.†Present address: University of Calgary Health Sciences Centre, Rm278 HMRB, 3330 Hospital Dr. NW, Calgary AB, T2N 4N1, Canada.

Molecular Microbiology (2001) 41(6), 1445–1458

Q 2001 Blackwell Science Ltd

et al., 1999). The genes required for Tir delivery and A/E

lesion formation are encoded in a chromosomal patho-

genicity island called the LEE (locus of enterocyte

effacement) (McDaniel and Kaper, 1997; Perna et al.,

1998). Once in the host cell, Tir is an integral membrane

protein, with two predicted transmembrane domains.

Recent work has indicated that these regions are required

for the stable insertion of Tir into the host plasma mem-

brane (Gauthier et al., 2000). The amino- and carboxy-

termini of Tir are intracellular (Kenny et al., 1997;

DeVinney et al., 1999), and intimin binds to an extracellular

domain called the IBD (intimin-binding domain) via a large

hydrophobic pocket (de Grado et al., 1999; Kenny, 1999;

Luo et al., 2000).

The ability to form A/E lesions is essential for full

virulence of this family of pathogens (Abe et al., 1998;

Marches et al., 2000; Tacket et al., 2000). Although the

more severe symptoms observed with EHEC (haemor-

rhagic colitis and HUS) are caused by the Shiga toxin

produced by this pathogen, the initial watery diarrhoea is

attributed to the formation of A/E lesions (Nataro and

Kaper, 1998). In EPEC, human volunteer studies have

implicated two proteins that are required for pedestal

formation in vitro, intimin (Donnenberg et al., 1993) and

EspB (Tacket et al., 2000), as important virulence factors.

Strains lacking either of these proteins are severely

attenuated in their ability to cause human disease. Further

evidence for the importance of A/E lesion formation in

virulence comes from work with a natural rabbit pathogen,

REPEC O103:H2. This work demonstrated that EspA

and EspB (Abe et al., 1998) and Tir and intimin (Marches

et al., 2000) are essential for the diarrhoeal disease

caused by this organism. In EHEC, intimin is required

for colonization and the formation of A/E lesions in the

intestines of gnotobiotic piglets and colostrum-deprived

calves and is essential for the disease observed in these

hosts (McKee et al., 1995; Dean-Nystrom et al., 1998).

Additionally, intimins from EPEC and EHEC are function-

ally interchangeable. Heterologous intimin expression

supports both intestinal colonization (Tzipori et al., 1995;

Phillips et al., 2000) and pedestal formation in cultured

epithelial cells (DeVinney et al., 1999).

The process of pedestal formation is best understood for

EPEC. EPEC Tir is tyrosine and serine/threonine

phosphorylated upon insertion into the host cell mem-

brane, and phosphorylation of tyrosine 474 is required for

pedestal formation (Rosenshine et al., 1992; Kenny et al.,

1997; Kenny, 1999). Pedestal formation requires the

recruitment of N-WASP (neural Wiscott–Aldrich syn-

drome protein; Kalman et al., 1999), which is recruited to

the pedestal in a Tir tyrosine phosphorylation-dependent

manner (Goosney et al., 2000). N-WASP functions to

target the Arp 2/3 complex to Tir, which is required for

pedestal elongation (Kalman et al., 1999) Both N-WASP

and the Arp2/3 complex play a role in subversion of the

actin cytoskeleton by other bacterial and viral pathogens

(Suzuki et al., 1998; Frischknecht et al., 1999; Moreau

et al., 2000). Additionally, the amino-terminus of EPEC Tir

binds directly to the host actin binding and cross-linking

protein a-actinin, anchoring Tir to the host actin

cytoskeleton (Goosney et al., 2000). The focal adhesion

proteins talin and vinculin have also been shown to interact

directly with Tir (Freeman et al., 2000).

Considerably less is understood about the mechanism

of pedestal formation by EHEC. Unlike EPEC Tir, EHEC

Tir is phosphorylated on serine or threonine, but not on

tyrosine, and the role of these phosphorylation events is

unknown (DeVinney et al., 1999). Despite these differ-

ences in Tir phosphorylation, EHEC forms seemingly

identical pedestals within the host cell (Ismaili et al., 1995;

DeVinney et al., 1999; Goosney et al., 2001). In order to

define further a role for Tir phosphorylation, we used site-

directed mutagenesis to generate EPEC Tir tyrosine 474

mutants, which we expressed in both EPEC and EHEC tir

deletion strains, and subsequently examined infected cells

for Tir delivery and pedestal formation. We found that

EPEC and EHEC use different Tir-based mechanisms for

pedestal formation. Unlike EPEC, pedestal formation by

EHEC can occur independently of EPEC Tir tyrosine

phosphorylation and requires the type III delivery of

additional bacterial factors into the host cell.

Results

EHEC Tir does not fully complement an EPEC tir deletion

mutant

To determine whether EPEC and EHEC Tirs are func-

tionally interchangeable, a series of cross-complementa-

tion experiments were performed. EPEC and EHEC tir

genes were amplified, cloned into pACYC184 under the

control of the TetR promoter and transformed into EPEC

JPN15Dtir and EHECDtir. This generated the complemen-

ted strains JPN15Dtir/EPECtir and EHECDtir/EHECtir and

the cross-complemented strains JPN15Dtir/EHECtir and

EHECDtir/EPECtir. We used JPN15, an EPEC E2348/69

plasmid-cured derivative that lacks the large virulence

plasmid (EAF) encoding the bundle-forming pilus (BFP)

(Donnenberg et al., 1992), as it adheres diffusely to the

host cell in a similar manner to EHEC, facilitating the direct

comparison of the two pathogens.

We initially examined whether Tir was synthesized,

delivered to and phosphorylated within the host cell by the

complemented and cross-complemented strains. Triton

X-100-soluble fractions were prepared from HeLa cells

infected with each of the four strains as described in

Experimental procedures. Although this method also

extracts bacterial proteins, it remains useful for this

1446 R. DeVinney et al.

Q 2001 Blackwell Science Ltd, Molecular Microbiology, 41, 1445–1458

purpose, as we are monitoring for the change in apparent

molecular weight resulting from Tir phosphorylation, which

occurs upon Tir delivery into the host cell (Gauthier et al.,

2000). Tir phosphorylation was assessed by incubating

the Triton X-100-soluble fractions with the non-specific

phosphatase alkaline phosphatase before resolution on

8% SDS–PAGE and analysis by immunoblotting with

anti-Tir and anti-phosphotyrosine (PY) antisera. In all four

strains, Tir was synthesized by the bacteria, delivered

to the host cell and appropriately phosphorylated

(Fig. 1). EPEC Tir was delivered to the host cell by both

JPN15Dtir and EHECDtir, where it became tyrosine

phosphorylated, as evidenced by the reduction in apparent

molecular weight after treatment with alkaline phospha-

tase and reactivity with anti-PY antisera (Fig. 1A and B,

left). In the case of EHEC Tir, both JPN15Dtir/EHECtir and

EHECDtir/EHECtir synthesized and delivered Tir into the

host cells. In both strains, we observed a decrease in

EHEC Tir apparent molecular weight after treatment with

alkaline phosphatase, indicating that EHEC Tir becomes

phosphorylated upon translocation (Fig. 1A, right). Anti-PY

blotting revealed that EHEC Tir was not tyrosine phos-

phorylated (Fig. 1B, right), supporting previous obser-

vations in wild-type EHEC (Ismaili et al., 1998; DeVinney

et al., 1999).

We next examined whether pedestals form in cells

infected with the cross-complemented EPEC JPN15 and

EHEC strains. HeLa cells were infected with either

the cross-complemented strains JPN15Dtir/EHECtir or

EHECDtir/EPECtir or with the control complemented

strains JPN15Dtir/EPECtir or EHECDtir/EHECtir. Samples

were prepared for immunofluorescence microscopy and

labelled with either anti-Tir or anti-PY antisera and

phalloidin to visualize actin pedestals. EPEC Tir com-

plemented pedestal formation by EHECDtir. The pedestals

formed by EHECDtir/EPECTir were indistinguishable

from those formed by JPN15Dtir/EPECtir (Fig. 2A) and

formed with equal efficiency to those elicited by

EHECDtir/EHECtir (Table 1). In both cases, anti-EPEC

Fig. 1. EPEC and EHEC Tir are expressed and delivered to the hostcell by JPN15Dtir and EHECDtir. Anti-tir (A) and anti-PY (B) Westernblots of Triton X-100-soluble fractions from HeLa cells infected withJPN15Dtir/EPECtir, EHECDtir/EPECtir, JPN15Dtir/EHECtir orEHECDtir/ EHECtir. Samples were prepared, treated with alkalinephosphatase (Alk. Phos.) as described in Experimental procedures,resolved by 8% SDS–PAGE and transferred to nitrocellulose.A. The left blot was probed with mouse-anti EPEC Tir mAb 2A5, andthe right blot was probed with rat-anti EHEC Tir antisera. Arrowheadsindicate the position of phosphorylated and dephosphorylated Tir, andmolecular weight markers are in kDa.

Fig. 2. EHEC Tir does not complement EPECDtir.A. Immunofluorescence micrograph of HeLa cells infected with eitherJPN15Dtir/EPECtir or EHECDtir/EPECtir. Cells were triple labelledwith either mouse anti-EPEC Tir mAb 2A8 or anti-PY, phalloidin (actin)and DAPI (bacteria and nuclei).B. HeLa cells infected with JPN15Dtir/EHECtir or EHECDtir/EHECtirand triple labelled with either anti-PY or rat anti-EHEC Tir, phalloidinand DAPI. Cells were fixed in 2.5% PFA and labelled as described.Secondary antisera used were anti-mouse Alexa 488 (EPEC Tir andanti-PY), anti-rat TRITC (EHEC Tir) and either Alexa 488 or 568phalloidin.

Pedestal formation by EHEC 1447

Q 2001 Blackwell Science Ltd, Molecular Microbiology, 41, 1445–1458

Tir and anti-PY labelling were tightly focused at the tip of

the actin pedestals, corresponding to tyrosine-phosphory-

lated EPEC Tir (Fig. 2A; Kenny et al., 1997). Taken

together with the results in Fig. 1, this indicates that EHEC

is able to stimulate the tyrosine kinases required for EPEC

Tir phosphorylation. In contrast, although JPN15Dtir/E-

HECtir delivered Tir to the host cell, pedestals did not form

beneath the adhering bacteria (Fig. 2B). The EHEC tir

construct expressed by this strain is fully functional, as

evidenced by its ability to complement pedestal formation

by EHECDtir (Fig. 2B, bottom). Collectively, these data

suggest that, although EPEC and EHEC Tirs are

expressed and delivered to the host cell by the cross-

complemented strains, they are not functionally identical

with regard to pedestal formation.

Pedestal formation by EHEC occurs independently of

EPEC Tir tyrosine phosphorylation

One important difference between EPEC and EHEC Tir is

at the level of phosphorylation. EPEC but not EHEC Tir is

tyrosine phosphorylated on residue 474, which is essential

for pedestal formation (Kenny, 1999). To examine further

the role played by EPEC Tir tyrosine phosphorylation in

pedestal formation, we used a polymerase chain reaction

(PCR)-based mutagenesis strategy to change the phos-

phorylated EPEC Tir tyrosine 474 to phenylalanine. The

resulting construct, EPECtir Y474F, was transformed into

JPN15Dtir and EHECDtir, and the strains were examined

for their ability to deliver Tir and form pedestals within

HeLa cells. Infection with JPN15Dtir/EPECtir Y474F did

not result in pedestal formation, although EPEC Tir Y474F

was delivered to the host cell and focused beneath

adhering bacteria (Fig. 3A, top). The inability of this strain

to form pedestals resulted from the absence of Tir tyrosine

phosphorylation, as shown by immunofluorescence and

immunoblot analysis. No 90 kDa band was detected upon

anti-PY immunoblotting of Triton X-100-soluble fractions

from HeLa cells infected with JPN15Dtir/EPECtir Y474F

(Fig. 3B, right, compare with wild-type EPEC Tir in Fig. 1),

indicating that Tir Y474F is not tyrosine phosphorylated.

However, EPEC Tir Y474F is serine/threonine phosphory-

lated after delivery to the host cell, as evidenced by the

mobility shift observed on an anti-Tir blot after treatment

with alkaline phosphatase (Fig. 3B, left l). Taken together,

this suggests that EPEC Tir serine/threonine phosphoryl-

ation alone is not sufficient to support pedestal formation

by EPEC, which requires Tir tyrosine phosphorylation.

In contrast, pedestal formation by EHEC occurs

independently of EPEC Tir tyrosine phosphorylation.

Infection with EHECDtir/EPECtir Y474F resulted in the

formation of actin-rich pedestals beneath the adhering

bacteria, where Tir, but not phosphotyrosine, accumulated

at the pedestal tip (Fig. 3A, bottom), suggesting that Tir is

not tyrosine phosphorylated. Pedestal formation by this

strain was significantly less efficient than that of

EHECDtir/EHECtir (Table 1; 19.0^ 2.0% for EPECY474F

compared with 35.0^ 3.5% for EHEC Tir). Results

from anti-Tir and anti-PY immunoblots of Triton X-100-

soluble fractions confirmed these observations. EHECDtir/

EPECtir Y474F delivers Tir to the host cell (Fig. 3B, left),

which is phosphorylated but not on tyrosine, as evidenced

by the decrease in apparent molecular weight after

phosphatase treatment and lack of reactivity with anti-PY

antisera. There was no difference in the electrophoretic

mobility and alkaline phosphatase sensitivity between

EHECDtir- or JPN15Dtir-delivered EPEC Tir Y474F

(Fig. 3B). Collectively, these data suggest that EHEC

uses a Tir tyrosine phosphorylation-independent mechan-

ism for pedestal formation. In contrast, EPEC Tir tyrosine

phosphorylation is an essential step in pedestal formation

mediated by EPEC, which cannot use either an unphos-

phorylatable EPEC Tir mutant or EHEC Tir to form

pedestals. Based on these results, we addressed two

additional questions: what role does Tir tyrosine phos-

phorylation play in pedestal formation, and how does

EHEC form a pedestal in the absence of Tir tyrosine

phosphorylation?

EPEC Tir Y474E does not rescue pedestal formation by

EPEC

To determine the contribution of the negative charge

introduced upon phosphorylation of Y474, we constructed

a ‘pseudophosphorylated’ mutant of EPEC Tir using site-

directed mutagenesis to change tyrosine 474 to a glutamic

acid residue (EPECtir Y474E) and expressed this con-

struct in JPN15Dtir and EHECDtir. This mutation intro-

duces a negative charge that may mimic some of the

effects of protein tyrosine phosphorylation (Creighton,

1993). Transformation of this construct into JPN15Dtir did

not rescue pedestal formation by this strain. After a 4 h

incubation, bacteria adhered strongly to the HeLa cell

surface and focused Tir, yet pedestals did not form

(Fig. 4A). We infrequently observed actin condensation

Table 1. Efficiency of pedestal formation.

Construct

Percentage of cells with pedestals

JPN15Dtir EHECDtir

EPEC Tir 32.6 ^ 3.7% 35.0 ^ 3.5%EPEC Tir Y474F NPa 19.0 ^ 2.0%b

EPEC Tir Y474E NP 30.3 ^ 3.2%Indel-1 NP 26.3 ^ 3.1%b

Indel-2 28.0 ^ 4.0% 40.3 ^ 8.5%EHEC Tir NP 35.3 ^ 4.0%

a. NP, pedestals not formed by this strain.b. Mean^SD (n¼ 3); P , 0.05 by ANOVA.

1448 R. DeVinney et al.

Q 2001 Blackwell Science Ltd, Molecular Microbiology, 41, 1445–1458

beneath adhering bacteria, but only in areas densely

covered with adherent JPN15Dtir/EPECtir Y474E (Fig. 4A,

arrows). Even after a 7 h incubation, we never observed

the formation of elongated pedestals with this strain (data

not shown). In contrast, EHECDtir/ EPECtir Y474E infec-

tion of HeLa cells resulted in the formation of pedestals

that appeared identical to those elicited by EHECDtir/-

EPECtir Y474F (Fig. 4A, bottom, compare with Fig. 3).

Pedestal formation by this strain occurred at a similar

efficiency to that of EHECDtir/EHECtir (Table 1). Tir, but

not phosphotyrosine, was focused at the tip of the actin

pedestals elicited by this strain (Fig. 3). When Triton

X-100-soluble membranes from HeLa cells infected

with either JPN15Dtir/EPECtir Y474E or EHECDtir/

EPECtirY474E were examined by immunoblotting for

Tir delivery and phosphorylation, we found that EPEC Tir

Y474E was serine/threonine phosphorylated upon delivery

to the host cell, as indicated by the decrease in apparent

molecular weight observed after alkaline phosphatase

treatment and the lack of reactivity on with anti-PY anti-

sera. There was no difference in the electrophoretic

mobility and alkaline phosphatase sensitivity between

EHECDtir- or JPN15Dtir-delivered EPEC Tir Y474E, as

shown in Fig. 4B.

The protein sequence surrounding EPEC Tir Y474 is

essential for pedestal formation by EPEC

Although the EPEC and EHEC Tir protein sequences are

58% identical, the two proteins are quite divergent in their

C-termini (40% identity) (DeVinney et al., 1999). In par-

ticular, the 18-amino-acid motif that contains EPEC Tir

Y474 is completely different with respect to the corre-

sponding region in EHEC Tir (Fig. 5A). We hypothesized

that these unique regions of Tir might play an important

role in pedestal formation by EPEC and EHEC. To

examine a role for these sequences, we constructed two

chimeric Tir proteins as described in Experimental

procedures. Indel-1 is a derivative of EPEC Tir, with the

amino acid region from serine 465 to glycine 482 replaced

with aspartic acid 469 to valine 489 from EHEC Tir

(Fig. 5A), a substitution that eliminates tyrosine 474 and its

Fig. 3. Pedestal formation by EPEC but notEHEC requires EPEC Tir tyrosinephosphorylation.A. Immunofluorescence micrograph of HeLacells infected with eitherJPN15Dtir/EPECtir Y474F (top) orEHECDtir/EPECtir Y474F (bottom). Cells werefixed and triple labelled with DAPI, phalloidinand either mouse anti-EPEC Tir mAb 2A8 oranti-PY.B. Western blot of Triton X-100-soluble fractionsfrom HeLa cells infected with eitherJPN15Dtir/EPECtir Y474F orEHECDtir/EPECtir Y474F and treated withalkaline phosphatase. Samples were resolvedby 8% SDS–PAGE, and blots were probed witheither mouse anti-EPEC Tir mAb 2A5 (left) oranti-PY (right). Arrowheads indicate the positionof phosphorylated and dephosphorylated Tir,and molecular weight markers are in kDa.

Pedestal formation by EHEC 1449

Q 2001 Blackwell Science Ltd, Molecular Microbiology, 41, 1445–1458

flanking amino acids. Similarly, Indel-2 was constructed by

replacing aspartic acid 469 to valine 489 of EHEC Tir with

serine 465 to glycine 482 from EPEC Tir, containing

tyrosine 474 and its flanking amino acids (Fig. 5A). These

chimera were expressed in JPN15Dtir and EHECDtir and

examined for the ability to deliver Tir and form pedestals.

Replacement of the region around Y474 with the

corresponding EHEC sequence in EPEC Tir (Indel-1)

completely abolished pedestal formation by EPEC. In

HeLa cells infected with JPN15Dtir/Indel-1, bacteria

adhered to the host cells and focused Tir, but did not

form pedestals (Fig. 5B). In contrast, EHEC Tir carrying

the EPEC Tir Y474 domain (Indel-2) restored the ability of

EPEC to form pedestals with EHEC Tir (Fig. 5C,

JPN15Dtir/Indel-2). Infection with JPN15Dtir/Indel-2 led

to the formation of pedestals beneath the adhering

bacteria. These pedestals formed with equal efficiency

(Table 1) and appeared to be identical to those formed with

wild-type EPEC Tir, with anti-Tir and anti-PY labelling

focused at the pedestal tip, suggesting that Indel-2 was

tyrosine phosphorylated once delivered to the host cell

(compare Fig. 5C with Fig. 2A). As expected, both Indel-1

and Indel-2 supported pedestal formation when trans-

formed into EHECDtir (Fig. 5B and C). In the case of

EHECDtir/Indel-1, Tir was focused at the tip, but not

tyrosine phosphorylated, as shown in Fig. 5B, and

pedestal formation occurred at a significantly lower

efficiency when compared with wild-type EHEC (Table 1;

26.3^ 3.1% for Indel-1 compared with 35.0^ 3.5% for

EHEC Tir). Pedestals formed by EHECDtir/Indel-2 con-

tained tyrosine-phosphorylated Tir at the tip and formed as

efficiently as those elicited by EHECDtir/EHECtir (Fig. 5C,

Table 1). In all cases, both Indel-1 and Indel-2 were

delivered efficiently into the host cell, as determined by

anti-Tir immunoblotting (Fig. 5D). Collectively, these data

indicate that the region around EPEC Tir Y474 is essential

for pedestal formation by EPEC and that this region

contains the sequences required for EPEC Tir tyrosine

phosphorylation.

Pedestal formation by EHEC requires additional bacterial

factors

Our results indicate that EHEC Tir does not complement

pedestal formation by EPEC JPN15Dtir and that EHEC but

not EPEC can use the non-tyrosine-phosphorylated EPEC

Tir Y474F to form pedestals within the host cell. In these

experiments, there was no difference in either the EHEC

Tir or EPEC Tir Y474F delivered by EPECDtir or EHECDtir

or the HeLa cells used in the infections. The only differ-

ence was in the bacterial strains delivering EPEC Tir

Y474F into the HeLa cell membrane, suggesting to us

that EHEC may be delivering additional bacterial factors

to facilitate Tir tyrosine phosphorylation-independent

pedestal formation. To test this hypothesis, we used a

co-infection strategy. Strain JPN15Dtir/EHECtir/GFP,

which expresses the green fluorescent protein (GFP),

was used to determine whether EHEC delivers additional

Fig. 4. A. Immunofluorescence micrograph ofHeLa cells infected with eitherJPN15Dtir/EPECtir Y474E orEHECDtir/EPECtir Y474E. Cells were fixed andlabelled as in Fig. 3A. Arrows indicate actinaccumulation betweenJPN15Dtir/EPECtir Y474E as described in thetext.B. Anti-EPEC Tir mAb 2A5 (left) and anti-PY(right) Western blots of Triton X-100-solublefractions from HeLa cells infected with eitherJPN15Dtir/EPECtir Y474E orEHECDtir/EPECtir Y474E. Samples wereprepared and alkaline phosphatase treated asdescribed and resolved on 8% SDS–PAGE.Arrowheads indicate the position ofphosphorylated and dephosphorylated Tir, andmolecular weight markers are in kDa.

1450 R. DeVinney et al.

Q 2001 Blackwell Science Ltd, Molecular Microbiology, 41, 1445–1458

bacterial factors to the host cell that are necessary for

pedestal formation. HeLa cells were co-infected with

JPN15Dtir/EHECtir/GFP and EHEC UMD619Dtir. The

latter strain lacks both Tir and intimin, adheres weakly to

the host cell, and the majority of bacteria are removed by

the extensive washing involved in this method. Although

this strain does not form pedestals, it contains an intact

type III secretion and translocation apparatus and can

therefore deliver proteins to the host cell. If any of

these delivered proteins are required for pedestal

formation, we expected to observe pedestals form

beneath JPN15Dtir/EHECtir/GFP.

The results from these experiments are shown in

Fig. 6. In cells infected with JPN15Dtir/EHECtir/GFP

alone, there was no evidence of actin accumulation

beneath the adhering GFP-expressing bacteria (Fig. 6,

top). Only when cells were co-infected with EHEC

UMD619Dtir and JPN15Dtir/EHECtir/GFP was pedestal

formation observed (Fig. 6, second row). In contrast, the

corresponding EPEC JPN15 tir/intimin double mutant

strain UMD207Dtir was unable to rescue pedestal

formation by JPN15Dtir/EHECtir/GFP (Fig. 6, third row).

The additional bacterial factors are probably delivered via

the EHEC type III secretion system, as co-infection with

JPN15Dtir/EHECtir/GFP and CVD451, a weakly adherent

EHEC strain containing a mutation in escN, an essential

component of the type III secretion system, did not result

in pedestal formation (Fig. 6, bottom). Interestingly,

co-infection with EHEC UMD619Dtir restored pedestal

formation by both JPN15Dtir/EPECtir Y474F/GFP and

JPN15Dtir/Indel-1/GFP in a similar manner to that obser-

ved with JPN15Dtir/EHECtir/GFP (data not shown). These

data strongly suggest that pedestal formation by EHEC, in

the absence of tyrosine phosphorylation, requires

additional bacterial factors that are delivered to the host

cell by the type III secretory apparatus and that both EPEC

and EHEC Tir share the sequence information necessary

for this process.

Fig. 5. The protein sequence surrounding EPEC Tir Y474 is essentialfor Tir tyrosine phosphorylation and pedestal formation by EPEC.A. Construction of Indel-1 and Indel-2. Top: protein sequences of theregions of EPEC and EHEC Tir that were exchanged during theconstruction of the Tir chimera. Bottom: diagram of Indel-1 and Indel-2chimera. EPEC Tir sequence is indicated in grey, whereas EHECsequence is white.B. Immunofluorescence micrographs of HeLa cells infected with eitherJPN15Dtir/Indel-1 or EHECDtir/Indel-1. Cells were triple labelled witheither anti-PY or mouse anti-EPEC Tir 2A5, DAPI and phalloidin.C. Immunofluorescence micrographs of HeLa cells infected with eitherJPN15Dtir/ Indel-2 (top) or EHECDtir/Indel-2 (bottom). Cells weretriple labelled with either mouse anti-EPEC Tir mAb 2E8 or anti-PYand DAPI and phalloidin.D. Anti-Tir immunoblot of Triton X-100-soluble fractions from HeLacells infected with JPN15Dtir/Indel-1, EHECDtir/Indel-1 (left) orJPN15Dtir/Indel-2 or EHECDtir/Indel-2 (right). Samples wereprepared and alkaline phosphatase treated as described and resolvedon 8% SDS–PAGE. Arrowheads indicate the position ofphosphorylated and dephosphorylated Tir, and molecular weightmarkers are in kDa.

Pedestal formation by EHEC 1451

Q 2001 Blackwell Science Ltd, Molecular Microbiology, 41, 1445–1458

Discussion

Pedestal formation by EHEC and EPEC is often believed

to occur by a similar mechanism. Despite differences in Tir

tyrosine phosphorylation, the pedestals elicited by EPEC

and EHEC appear to be morphologically identical, with Tir

at the pedestal tip and cytoskeletal components including

actin and alpha actinin along the length (Ismaili et al.,

1995; Cantarelli et al., 2000; Goosney et al., 2001). In this

study, we report the surprising finding that EHEC and

EPEC use different Tir-based mechanisms for pedestal

formation. Our data indicate that pedestal formation by

EHEC and EPEC differs in the requirements for Tir

tyrosine phosphorylation and additional bacterial factors

delivered to the host cell. This information represents a

change in how we view pedestal formation by these two

pathogens and indicates that what we know about the

mechanism of pedestal formation by EPEC is not always

directly applicable to EHEC.

One critical difference between pedestal formation by

EHEC and EPEC is in the requirement for Tir tyrosine

phosphorylation. In EPEC, pedestal formation is absolutely

dependent on Tir tyrosine phosphorylation. Mutation of

EPEC Tir tyrosine 474 to either phenylalanine (Gauthier

et al., 2000; Goosney et al., 2000; this study) or serine

(Kenny, 1999) abolishes the ability of EPEC to form

pedestals within the host cell, without affecting either Tir

delivery or membrane insertion. Additionally, expression of

EHEC Tir in an EPEC tir deletion strain does not result in

pedestal formation. Both EPEC Tir Y474F and EHEC Tir

are still phosphorylated on serine/threonine, indicating that

these modifications alone are not sufficient to support

pedestal formation by EPEC. In contrast, pedestal

formation by EHEC occurs independently of Tir tyrosine

phosphorylation. EHEC Tir is not tyrosine phosphorylated,

and our results from cross-complementation experiments

demonstrate that EHEC expressing EPEC Tir Y474F

forms pedestals within the host cell even though it cannot

be tyrosine phosphorylated. This is not caused by the

inability of EHEC to stimulate Tir tyrosine phosphorylation,

as tyrosine-phosphorylated wild-type EPEC Tir is found at

the tip of the pedestal formed by EHECDtir/EPECTir.

The function of Tir tyrosine phosphorylation in pedestal

formation by EPEC is not understood. Our data indicate

that it does not result only from the introduction of a

negative charge, as the ‘pseudophosphorylated’ tyrosine

to glutamic acid mutant Y474E did not restore pedestal

formation to an EPEC Tir deletion strain. Although we

occasionally observed filamentous actin beneath some,

but not all, adhering JPN15Dtir/EPECtir Y474E, this accu-

mulation appeared to be different morphologically from

that observed in EPEC pedestals. The actin accumulation

observed may result from the recruitment of some but not

all the required cytoskeletal components to EPEC Tir

Y474E or may represent the non-specific actin accumu-

lation we have observed at the site of adherence of

EPECDtir strains (Kenny et al., 1997; Goosney et al.,

2001). Recently, the identity of a number of host proteins

recruited to pedestals formed by EPEC and EHEC has

Fig. 6. Pedestal formation by EHEC requiresadditional bacterial factors.Immunofluorescence micrograph of HeLa cellsinfected with JPN15Dtir/EHECtir/GFP alone(top), co-infected with EHEC UMD619Dtir(second row), co-infected with EPECUMD207Dtir (third row) or co-infected withEHEC CVD451 (DescN, bottom). Cells werefixed and labelled with Alexa 568 phalloidin.Blots show total bacteria/host cell nuclei (DAPI),JPN15Dtir/EHECtir/GFP (GFP) and, if formed,actin pedestals (actin).

1452 R. DeVinney et al.

Q 2001 Blackwell Science Ltd, Molecular Microbiology, 41, 1445–1458

been determined (Goosney et al., 2001), and this

information will provide us with the tools to distinguish

between these two possibilities.

Although EHEC formed pedestals with all the constructs

examined in this study, the efficiency of pedestal formation

was not identical for all strains. Pedestal formation by both

EHECDtir/EPECtir Y474F and EHECDtir/Indel-1 was sig-

nificantly less efficient that that of EHECDtir/EHECtir.

Bacterial adherence and pedestal formation by these

strains was 25–46% lower than that observed with

EHECDtir/EHECtir (Table 1). Both strains delivered Tir

into the host cell at levels equivalent to that observed with

EHECDtir/EHECtir (Figs 3 and 5). In all cases, pedestal

formation was observed under all adherent bacterial

microcolonies, suggesting that the effect on efficiency of

pedestal formation results from differences in adherence

to the host cell. In EHEC, both Tir and intimin are required

for stable adherence to cultured epithelial cells and

intestinal colonization (McKee et al., 1995; DeVinney et al.,

1999). One possible explanation for the difference in

efficiency of pedestal formation is that the mutations

introduced to construct EPECtir Y474F and Indel-1 alter

Tir–intimin binding and therefore affect initial bacterial

adherence. We are presently undertaking studies to

examine the effect of these mutations on Tir–intimin

binding.

Results from our studies with the chimeric Tir proteins

Indel-1 and Indel-2 indicate that the region surrounding

Y474 in EPEC Tir is critical for pedestal formation by

EPEC. Replacement of amino acid sequence surrounding

EPEC Tir Y474 with the non-identical corresponding

sequence from EHEC Tir (Indel-1, Fig. 5A) abolished

pedestal formation by EPEC expressing this construct.

However, the introduction of the same sequence

surrounding EPEC Tir Y474 into EHEC Tir (Indel-2) and

expression in EPECDtir resulted in the formation of

pedestals that appeared to be identical to those elicited by

wild-type EPEC. This sequence contains both the phos-

phorylated Y474 and upstream sequences with homology

to substrate sequences recognized by mammalian

tyrosine kinases. Work from Songyang et al. (1993)

suggested that a number of mammalian tyrosine kinases

preferred to phosphorylate peptides with acidic amino

acids in the 23 and 24 positions before the phosphory-

lated tyrosine and a hydrophobic amino acid at the 13

position. The sequence surrounding the phosphorylated

Y474 in EPEC Tir (EEHI-Y474-DEV) contains amino acids

in this context. In addition, this sequence shares significant

homology with phosphoprotein motifs shown to be impor-

tant for their recognition by SH2 domain-containing pro-

teins (Songyang et al., 1993). Collectively, these data

suggest that this amino acid region of EPEC Tir contains

all the information necessary for tyrosine phosphorylation

and the potential interaction with SH2 domain-containing

proteins that may be involved in linking Tir to the actin

cytoskeleton.

In EPEC, Tir tyrosine phosphorylation is required for the

recruitment of two important cytoskeletal components,

N-WASP and the Arp 2/3 complex, to Tir (Goosney et al.,

2000), which are required for pedestal formation (Kalman

et al., 1999). Both N-WASP and the Arp 2/3 complex play

important roles in cytoskeletal reorganization in response

to extracellular stimuli, but have not been shown to interact

directly with tyrosine-phosphorylated receptors (Mullins,

2000). An important class of proteins, SH2 domain-

containing cellular adaptors, function as molecular linkers

between tyrosine-phosphorylated receptors and the actin

cytoskeleton (Rivero-Lezcano et al., 1995; Carlier et al.,

2000). We have demonstrated recently that EPEC Tir

binds directly to the SH2 domain-containing adaptor

protein Nck and does so in a Tir tyrosine phosphorylation-

dependent manner (Gruenheid et al., 2001). Nck binding is

essential for pedestal formation, as cells that are deficient

in Nck family members (Nck1/Nck2) are unable to support

pedestal formation by EPEC. Not surprisingly, pedestal

formation by EHEC is unaffected in the Nck1/Nck2 double

mutant cell line.

A second major difference between the mechanisms of

pedestal formation by EHEC and EPEC is the requirement

for the delivery of additional bacterial factors into the host

cell. Several lines of evidence support this conclusion.

EHEC, but not EPEC, forms pedestals with EPEC Tir

Y474F, suggesting that EHEC may be delivering addi-

tional bacterial factors to facilitate Tir tyrosine phosphory-

lation-independent pedestal formation. We addressed this

directly by performing co-infection experiments. Although

expression of either EHEC Tir or EPEC Tir Y474F in an

EPEC Tir deletion strain did not result in pedestal forma-

tion, we were able to rescue pedestal formation by co-

infecting with an EHEC strain lacking both Tir and intimin

(EHEC UMD619Dtir), but not with a type III secretion

mutant (EHEC CVD451). Our data strongly suggest that

putative EHEC additional factors are delivered into the

host cell via the EHEC type III secretion apparatus, as co-

infection with the type III secretion mutant CVD451 did not

rescue pedestal formation.

In EHEC, pedestal formation occurs without Tir tyrosine

phosphorylation and requires the delivery of additional

bacterial factors into the host cell. Despite the absence of

Tir tyrosine phosphorylation, N-WASP and the Arp 2/3

complex are still recruited to the EHEC pedestal (Goosney

et al., 2001), suggesting that different signalling pathways

are modulated by EHEC leading to pedestal formation. It

is tempting to speculate that additional EHEC factors

delivered into the host cell may serve as an interface

between Tir and the host cytoskeleton. These factors may

either bind directly to EHEC Tir, leading to recruitment of

Pedestal formation by EHEC 1453

Q 2001 Blackwell Science Ltd, Molecular Microbiology, 41, 1445–1458

the cytoskeletal machinery, or act indirectly to affect

signalling pathways controlling cytoskeletal organization.

The EHEC protein(s) that provide the capacity to form

pedestals in a tyrosine phosphorylation-independent

manner are not yet known. However, one interesting

possibility is that functional differences may rely on the

LEE-encoded proteins, such as the type III secreted EspB

and EspD, which exhibit the most sequence divergence

between EHEC and EPEC (Elliott et al., 1998). This

sequence divergence may confer additional features to

these proteins during pedestal formation by EHEC. EspB

and EspD are delivered to the host cell membrane and

cytosol and are hypothesized to play a role as bacterial

effectors and in the translocation of bacterial proteins into

the host cell (Taylor et al., 1998; Wolff et al., 1998; Kresse

et al., 1999; Wachter et al., 1999). EspF contains sequ-

ences with high homology to mammalian poly proline-rich

SH3-binding domains, which are found in a number of

signalling and cytoskeletal proteins. This protein, although

not essential for pedestal formation by EPEC, has not

been studied in EHEC (McNamara and Donnenberg,

1998). Alternatively, the additional EHEC factors may be

encoded elsewhere in the EHEC genome, which is < 25%

larger than that of E. coli K-12 and contains close to 1400

genes, with many organized in potential pathogenicity

islands, that are not present in E. coli K-12 (Ohnishi et al.,

1999; Nicholls et al., 2000; Tarr et al., 2000; Perna et al.,

2001). Therefore, it is not unexpected that some of these

additional genes could encode EHEC virulence factors

involved in pedestal formation. In addition, EHEC strains

also contain a large virulence plasmid, pO157, which

contains ORFs of unknown function that may also encode

factors involved in pedestal formation (Burland et al.,

1998; Makino et al., 1998). We are presently pursuing

studies to identify and elucidate the function of these

additional EHEC factors.

An intriguing question is why EHEC O157:H7 evolved a

different mechanism for pedestal formation. All the Tir

sequences obtained from non-O157:H7 EHEC strains

published in the data bank to date contain a tyrosine

residue in the similar context to EPEC Tir Y474 (e.g.

O26:H-AJ223063; O111:H-AF070069). Preliminary immu-

nofluorescence microscopy studies from this laboratory

have shown that a number of EPEC and non-O157 EHEC

strains form pedestals and accumulate tyrosine-phos-

phorylated Tir at the tip. Only the closely related EHEC

strain O55:H7 (Reid et al., 2000), which produces a Tir

protein that has a C-terminal sequence almost identical to

that found in EHEC O157:H7, forms pedestals that do not

contain tyrosine-phosphorylated Tir at the tip (unpublished

Table 2. Strains and plasmids used in this study.

Strain Relevant characteristics Antibiotic resistancea Reference

JPN15Dtir EPEC O127:H6 tir deletion mutant of EAF plasmid-curedstrain E2348/69,

Nal This study

EHECDtir EHEC O157:H7 strain 86/24 tir deletion mutant Nal DeVinney et al. (1999)UMD619Ætir EHEC 86/24 eae–tir double mutant Nal DeVinney et al. (1999)CVD451 EHEC 86/24 sepB (escN ) mutant Jarvis and Kaper (1996)JPN15Dtir/EPECtir EPEC JPN15Dtir carrying pEP23 Nal/Cm This studyEHECDtir/EPECtir EHECDtir carrying pEP23 Nal/Cm This studyJPN15Dtir/EHECtir EPEC JPN15Dtir carrying pEH86 Nal/Cm This studyEHECDtir/EHECtir EHECDtir carrying pEH86 Nal/Cm This studyJPN15Dtir/EPECtir Y474F EPEC JPN15Dtir carrying pEP23-Y474F Nal/Cm This studyEHECDtir/EPECtir Y474F EHECDtir carrying pEP23-Y474F Nal/Cm This studyJPN15Dtir/EPECtir Y474E EPEC JPN15Dtir carrying pEP23-Y474E Nal/Cm This studyEHECDtir/EPECtir Y474E EHECDtir carrying pEP23-Y474E Nal/Cm This studyJPN15Dtir/Indel-1 EPEC JPN15Dtir carrying pEP-Indel1 Nal/Cm This studyEHECDtir/Indel-1 EHECDtir carrying pEP-Indel1 Nal/Cm This studyJPN15Dtir/Indel-2 EPEC JPN15Dtir carrying pEH-Indel2 Nal/Cm This studyEHECDtir/Indel-2 EHECDtir carrying pEH-Indel2 Nal/Cm This studyJPN15Dtir/GFP EPEC JPN15Dtir expressing GFP from pUFR047 Nal/ This study

Cm/ApJPN15Dtir/EPECtir/GFP JPN15Dtir/EPECtir expressing GFP from pUFR047 Nal/Ap/Cm This study

PlasmidspEP23 pACYC/EPEC tir Cm This studypEH86 pACYC/EHEC tir Cm This studypEP23-Y474F pACYC/EPEC tir Y474F Cm This studypEP23-Y474E pACYC/EPEC tir Y474E Cm This studypEP-Indel1 pACYC/EHEC carrying Indel-1 Cm This studypEH-Indel2 pACYC/EPEC Carrying Indel-2 Cm This studypUFR047-GFP pUFR047 expressing enhanced GFP from

the aph promoterAp J. Celli

(unpublished)

a. Nal, nalidixic acid; Cm, chloramphenicol; Ap, ampicillin.

1454 R. DeVinney et al.

Q 2001 Blackwell Science Ltd, Molecular Microbiology, 41, 1445–1458

results). Therefore, it is tempting to speculate that, by

forming tyrosine phosphorylation-independent pedestals,

EHEC has evolved to become a more efficient pathogen

that is less dependent on host cell signalling pathways,

such as those involving tyrosine kinases, to form

pedestals.

We have shown that, in EPEC, Tir tyrosine phosphoryl-

ation and the region surrounding the phosphorylated

tyrosine 474 are essential features for pedestal formation.

The role played by EPEC Tir tyrosine phosphorylation is

not the result only of the introduction of a negative charge.

In contrast, in EHEC, pedestal formation occurs indepen-

dently of Tir tyrosine phosphorylation and requires the

delivery of additional EHEC factors into the host cell to

circumvent host tyrosine kinase signalling. These findings

represent a fundamental difference in the mechanisms of

pedestal formation by these two closely related

pathogens.

Experimental procedures

Bacterial strains and HeLa cell cultures

The strains used in this study are listed in Table 2. EHEC andEPEC strains used to infect HeLa cells were grown in Luria–

Bertani broth supplemented with appropriate antibiotics asstanding overnight cultures at 378C. HeLa cells (CCL2;

American Type Culture Collection) were cultured in minimalessential medium (MEM) supplemented with 10% fetal calf

serum (FCS), grown at 378C in 5% CO2.

Construction of plasmids expressing EPEC and EHEC tir

The coding regions of the EPEC and EHEC tir genes were

polymerase chain reaction (PCR) amplified using the primersTir0041 and Tir005–, which introduced Bgl II and Xho I

restriction sites respectively (Table 3). Tir0041 hybridized tothe ribosome binding site and 50 end of both tir genes,

whereas Tir005– hybridized to the 30 end sequences thatspan the stop codons. The amplified products were cloned

into the Bam HI–Sal I sites in pACYC184, leaving the tir genesunder the control of the tetracycline resistance gene (TetR)

promoter, thus generating pEP23 (EPEC) and pEH86(EHEC).

Site-directed mutagenesis of EPEC tir Tyr474

To construct both the Tyr-474Phe and the Tyr-474Glumutants of EPEC Tir, pEP23 was used as a template to

PCR amplify a fragment containing the 30 end of the tir genefrom codon 466A, followed by a vector region containing the

pACYC184 Eag I unique restriction site. Oligos DGK0011 orTIR0131 (Table 3), which span an endogenous Sap I

restriction site and introduce single point mutations in the474 codon to create the Y474F and Y474E mutants,

respectively, together with the reverse oligonucleotideRD131–, which is complementary to pACYC184, were

used. Both pEP23 and the mutagenized fragments weredigested with Sap I 1 Eag I. The pEP23 large Sap I–Eag I

fragment, lacking the unmutagenized insert fragment, wasgel purified and ligated with the corresponding digested

mutagenized fragments to create pEP23-Y474F andpEP23-Y474E. The tir mutant genes in these plasmids were

sequenced to confirm the modifications and electroporatedinto EPEC JPN15Dtir and EHECDtir. Expression and stability

of the mutant proteins was assessed by Western blot using ananti-Tir monoclonal antibody (see below).

Construction of EPEC and EHEC Tir chimeric proteins

EPEC Indel-1 was constructed by inverse PCR amplificationof pEP23 using the primers Tir0101 and Tir011–, which

introduce a silent mutation to create a unique Spe I site(Table 3). This resulted in the deletion of an EPEC tir fragment

Table 3. Oligonucleotides used in this study.

Oligonucleotide Sequence

Tir0041 50-GGA AGA TCT AGG AAA GGA GAR ATT TAT GCC TAT TGG-30

(EPEC and EHEC tir 50 end)Tir005– 50-CGG CCT CGA GGA TAT ATT TAG ACG AAA CG-30

(EPEC and EHEC tir 30end)DGK0011 50-GCT CAT CAG CCA GAA GAG CAT ATT TTT GAT GAG-30

(EPEC Y474F)RD131– 50-GCC CAG CCG CGT CGG CCG CCA TGC CGG CGA-30

(pACYC reverse primer for SDM)Tir0131 50-CTA TCG GCT CAT CAG CCA GAA GAG CAT ATT GAA GAT G-30

(EPEC Y474E)Tir0111 50-GTG CCG ACT AGT AAT TCT AAT ACG TCT GTT CAG AAT ATG GGG AAT ACA GAT TCT GTT GTA TAT AGC GTT

ATT CAG AAT TTT TCA G-30

(Indel-1)Tir010– 50-AGA ATT ACT AGT CGG CAC CTG TAG ACT ATT CCG AGC CCC CCC AAC-30

(Indel-1)RDI21 50-CCC CTG CAG ATC CTG GTT ATA GCA CCA TTC AAC AT-30

(Indel-2)RDI2– 50-GGG CTG CAG CGA CCT CAT CAT AAA TAT GCT CTT CTG GCT GAT GAG CCG ACG AAT CAT GCA GCG ATG T-30

(Indel-2)

Pedestal formation by EHEC 1455

Q 2001 Blackwell Science Ltd, Molecular Microbiology, 41, 1445–1458

that spans codons 465S to 482G and the insertion of anEHEC tir sequence from codons 469Q to 489V. The resulting

fragment was digested with Spe I and religated to generatepEP-Indel-1. Likewise, EHEC Indel-2 was constructed by

inverse PCR amplification of pEH86 using the primers RDI21

and RDI2–, which introduce a silent mutation to create a

unique Pst I site (Table 3). This resulted in the deletion ofEHEC tir codons 469Q to 489V and the insertion of EPEC tir

codons 465S to 482G. The resulting fragment was digestedwith Pst I and religated to generate pEH-Indel-2. pEP-Indel-1

and pEH-Indel-2 were transformed into EPEC JPN15Dtir andEHECDtir.

Cellular fractionation and immunoblotting

Cultured HeLa cells were infected with either the JPN15Dtir

strains for 4 h or the EHECDtir strains for 6 h, with the culturemedia replaced after 4 h, washed and solubilized with lysis

buffer containing 50 mM Tris (pH 7.4)21% Triton X-100supplemented with protease inhibitors (Complete EDTA-free;

Calbiochem) and phosphatase inhibitors (1 mM NaVO4,100 nM Microcystin LR). Triton X-100-soluble fractions

(membranes, cytosol and solubilized bacterial proteins)

were separated from the insoluble fraction containing mostof the bacteria and host cytoskeleton by centrifugation.

Samples for alkaline phosphatase treatment were prepared inlysis buffer without phosphatase inhibitors and treated with 2 U

of calf intestinal alkaline phosphatase (NEB) for 1 h at 378C.Samples were analysed by immunoblotting with anti-EPEC Tir

(mouse monoclonal antibodies 2A5, amino-terminus; 2E8,carboxy-terminus), rat anti-EHEC Tir (DeVinney et al., 1999)

and anti-phosphotyrosine (anti-PY, clone 4G10; UBI) asdescribed previously (Rosenshine et al., 1996).

Immunofluorescence microscopy

One millilitre of 2� 104 HeLa cells was added to each well of a

24-well plate containing a 12 mm glass coverslip and grownovernight. Monolayers were infected with 5ml of broth-grown

EHEC or JPN15 strains and incubated for either 4 h or 4 hfollowed by the addition of gentamicin (50mg ml21) for an

additional 3 h to promote pedestal elongation (Rosenshineet al., 1996). Cells were fixed in 2.5% paraformaldehyde

(PFA), permeabilized with 1% Triton-X100–PBS and blockedin 10% normal goat serum20.1% Triton X-100–PBS before

incubation with primary and secondary antisera as describedpreviously. Antisera were used at the following dilutions: rat

anti-EHEC Tir, 1:200; mouse anti-EPEC Tir mAb clone 2A5(Tir N-terminus) or 2E8 (Tir C-terminus), 1:100; mouse anti-

PY, 1:100; goat anti-mouse Alexa 488 (Molecular Probes)1:400; donkey anti-rat tetramethyl rhodamine isothiocyanate

(TRITC; Jackson Labs) 1:400; Alexa 488 or 568 phalloidin(Molecular Probes) 1:300; DAPI (Sigma) 2mg ml21. Samples

were examined using a Zeiss Axioskop microscope, andimages were acquired with an Empix DVC1300 digital camera

and analysed using NORTHERN ECLIPSE imaging software. Theefficiency of pedestal formation by the complemented and

cross-complemented strains was assessed by determiningthe percentage of cells showing tightly focused actin

pedestals beneath microcolonies of at least five adhering

bacteria. Replicate fields of at least 100 cells were counted for

each strain that formed pedestals. In all cases, pedestalformation was observed under all adherent bacterial micro-

colonies. Data is expressed as the mean^SD of threereplicate experiments, and the differences between groups

were determined by ANOVA.

Co-infection experiments

Enhanced GFP under the control of the aphA3 (kanamycin)promoter was excised from pAT113, ligated into the Sal I–

Pst I sites of the low-copy-number plasmid pUFR047 andtransformed into JPN15Dtir/EHECtir to create JPN15Dtir/-

EHECtir/GFP (Table 2). HeLa cells plated at 2� 104 ml21 on12 mm glass coverslips in 24-well plates were infected with

10ml of EPEC JPN15Dtir/EHECtir/GFP either alone or co-infected with the EHEC Tir/intimin double mutant, UMD619D

tir (3ml), the EPEC BFP, tir/intimin triple mutant UMD207Dtir(3ml) or the EHEC type III secretion mutant CVD451 (3ml) for

4 h. Cells were washed, fixed, permeabilized and blocked asdescribed above, before labelling with Alexa 568 phalloidin

and DAPI.

Acknowledgements

The authors thank Wanying Deng and Jean Celli for providing

the GFP construct used in these studies, and Derek Knoecheland Mitchell Uh for their assistance with the site-directed

mutagenesis. We also thank Samantha Gruenheid for critical

reading of the manuscript. This work was supported by a grantfrom the Canadian Institutes for Health Research (CIHR).

R.D. is supported by AHFMR and the Province of Alberta;J.L.P. is supported by DGAPA and CONACyT and is a

Howard Hughes International Scholar; A.G. is supported by aCIHR doctoral research award; and D.L.G. by a Natural

Sciences and Engineering Research Council Scholarship.B.B.F. is a CIHR Distinguished Scientist. and B.B.F. and

J.L.P. are Howard Hughes International Scholars.

References

Abe, A., Heczko, U., Hegele, R.G., and Finlay, B.B. (1998)Two enteropathogenic Escherichia coli type III secreted

proteins, EspA and EspB, are virulence factors. J Exp Med188: 1–10.

Beaudry, M., Zhu, C., Fairbrother, J.M., and Harel, J. (1996)Genotypic and phenotypic characterization of Escherichia

coli isolates from dogs manifesting attaching and effacinglesions. J Clin Microbiol 34: 144–148.

Burland, V., Shao, Y., Perna, N.T., Plunkett, G., Sofia, H.J.,and Blattner, F.R. (1998) The complete DNA sequence and

analysis of the large virulence plasmid of Escherichia coliO157:H7. Nucleic Acids Res 26: 4196–4204.

Cantarelli, V.V., Takahashi, A., Akeda, Y., Nagayama, K., andHonda, T. (2000) Interaction of enteropathogenic or

enterohemorrhagic Escherichia coli with HeLa cells resultsin translocation of cortactin to the bacterial adherence site.

Infect Immun 68: 382–386.Cantey, J.R., and Blake, R.K. (1977) Diarrhea due to

Escherichia coli in the rabbit: a novel mechanism. J Infect

Dis 135: 454–462.

1456 R. DeVinney et al.

Q 2001 Blackwell Science Ltd, Molecular Microbiology, 41, 1445–1458

Carlier, M.F., Nioche, P., Broutin-L’Hermite, I., Boujemaa, R.,

Le Clainche, C., Egile, C., et al. (2000) GRB2 links signaling

to actin assembly by enhancing interaction of neural

Wiskott–Aldrich syndrome protein (N-WASp) with actin-

related protein (ARP2/3) complex. J Biol Chem 275:

21946–21952.

Creighton, T. (1993) Proteins: Structure and Molecular

Properties. New York: WH Freeman.

Dean-Nystrom, E., Bosworth, B., Moon, H., and O’Brien, A.

(1998) Escherichia coli O157:H7 requires intimin for

enteropathogenicity in calves. Infect Immun 66:

4560–4563.

de Grado, M., Abe, A., Gauthier, A., Steele-Mortimer, O.,

DeVinney, R., and Finlay, B.B. (1999) Identification of the

intimin binding domain of Tir of enteropathogenic E. coli

(EPEC). Cell Microbiol 1: 7–18.

DeVinney, R., Stein, M., Reinscheid, D., Abe, A., Rusch-

kowski, S., and Finlay, B.B. (1999) Enterohemorrhagic

Escherichia coli O157:H7 produces Tir, which is translo-

cated to the host cell membrane but is not tyrosine

phosphorylated. Infect Immun 67: 2389–2398.

Donnenberg, M.S., Giron, J.A., Nataro, J.P., and Kaper, J.B.

(1992) A plasmid-encoded type IV fimbrial gene of

enteropathogenic Escherichia coli associated with localized

adherence. Mol Microbiol 6: 3427–3437.

Donnenberg, M.S., Tacket, C.O., James, S.P., Losonsky, G.,

Nataro, J.P., Wasserman, S.S., et al. (1993) Role of the

eaeA gene in experimental enteropathogenic Escherichia

coli infection. J Clin Invest 92: 1412–1417.

Ebel, F., Podzadel, T., Rohde, M., Kresse, A., Kramer, S.,

Deibel, C., et al. (1998) Initial binding of Shiga toxin-

producing Escherichia coli to host cells and subsequent

induction of actin rearrangements depend on filamentous

EspA-containing surface appendages. Mol Microbiol 30:

147–161.

Elliott, S.J., Wainwright, L.A., McDaniel, T.K., Jarvis, K.G.,

Deng, Y.K., Lai, L.C., et al. (1998) The complete sequence

of the locus of enterocyte effacement (LEE) from

enteropathogenic Escherichia coli E2348/69. Mol Microbiol

28: 1–4.

Freeman, N.L., Zurawski, D.V., Chowrashi, P., Ayoob, J.C.,

Huang, L., Mittal, B., et al. (2000) Interaction of the

enteropathogenic Escherichia coli protein, translocated

intimin receptor (Tir), with focal adhesion proteins. Cell Motil

Cytoskeleton 47: 307–318.

Frischknecht, F., Moreau, V., Rottger, S., Gonfloni, S.,

Reckmann, I., Superti-Furga, G., and Way, M. (1999) Actin-

based motility of vaccinia virus mimics receptor tyrosine

kinase signalling. Nature 401: 926–929.

Gauthier, A., de Grado, M., and Finlay, B.B. (2000)

Mechanical fractionation reveals structural requirements

for enteropathogenic Escherichia coli Tir insertion into host

membranes. Infect Immun 68: 4344–4348.

Goosney, D.L., DeVinney, R., Pfuetzner, R.A., Frey, E.A.,

Strynadka, N.C., and Finlay, B.B. (2000) Enteropathogenic

E. coli translocated intimin receptor, tir, interacts directly

with alpha-actinin. Curr Biol 10: 735–738.

Goosney, D.L., DeVinney, R., and Finlay, B.B. (2001)

Recruitment of cytoskeletal and signalling proteins to

enteropathogenic and enterohemorrhagic, E. coli ped-

estals. Infect Immun 69: 3315–3322.

Griffin, P.M., and Tauxe, R.V. (1991) The epidemiology of

infections caused by Escherichia coli O157:H7, other

enterohemorrhagic E. coli, and the associated hemolytic

uremic syndrome. Epidemiol Rev 13: 60–98.

Griffin, P., Ostroff, S., Tauxe, R., Greene, K., Wells, J., Lewis,

J., and Blake, P. (1988) Illnesses associated with

Escherichia coli O157:H7 infections. Ann Intern Medicine

109: 705–712.

Gruenheid, S., DeVinney, R., Bladt, F., Goosney, D., Gelkop,

S., Gish, G.D., et al. (2001) Enteropathogenic E. coli Tir

binds Nck to initiate actin pedestal formation in host cells.

Nature Cell Biol (in press).

Ismaili, A., Philpott, D.J., Dytoc, M.T., Soni, R., Ratnam,

S., and Sherman, P.M. (1995) Alpha-actinin accumu-

lation in epithelial cells infected with attaching and

effacing gastrointestinal pathogens. J Infect Dis 172:

1393–1396.

Ismaili, A., McWhirter, E., Handelsman, M., Brunton, J., and

Sherman, P. (1998) Divergent signal transduction

responses to infection with attaching and effacing Escheri-

chia coli. Infect Immun 66: 1688–1696.

Jarvis, K., and Kaper, J. (1996) Secretion of extracellular

proteins by enterohemorrhagic Escherichia coli via a

putative type III secretion system. Infect Immun 64:

4826–4829.

Kalman, D., Weiner, O.D., Goosney, D.L., Sedat, J.W., Finlay,

B.B., Abo, A., and Bishop, J.M. (1999) Enteropathogenic E.

coli acts through WASP and Arp2/3 complex to form actin

pedestals. Nature Cell Biol 1: 389–391.

Kenny, B. (1999) Phosphorylation of tyrosine 474 of the

enteropathogenic Escherichia coli (EPEC) Tir receptor

molecule is essential for actin nucleating activity and is

preceded by additional host modifications. Mol Microbiol 31:

1229–1241.

Kenny, B., DeVinney, R., Stein, M., Reinscheid, D.J., Frey,

E.A., and Finlay, B.B. (1997) Enteropathogenic E. coli

(EPEC) transfers its receptor for intimate adherence into

mammalian cells. Cell 91: 511–520.

Knutton, S., Baldwin, T., Williams, P.H., and McNeish, A.S.

(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.

Kondro, W. (2000) E. coli outbreak deaths spark judicial

inquiry in Canada. Lancet 355: 2058.

Kresse, A.U., Rohde, M., and Guzman, C.A. (1999) The EspD

protein of enterohemorrhagic Escherichia coli is required for

the formation of bacterial surface appendages and is

incorporated in the cytoplasmic membranes of target cells.

Infect Immun 67: 4834–4842.

Luo, Y., Frey, E.A., Pfuetzner, R.A., Creagh, A.L., Knoechel,

D.G., Haynes, C.A., et al. (2000) Crystal structure of

enteropathogenic Escherichia coli intimin–receptor com-

plex. Nature 405: 1073–1077.

McDaniel, T., and Kaper, J. (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.

McKee, M., Melton-Celsa, A., Moxley, R., Francis, D., and

O’Brien, A. (1995) Enterohemorrhagic Escherichia coli

O157:H7 requires intimin to colonize the gnotobiotic pig

Pedestal formation by EHEC 1457

Q 2001 Blackwell Science Ltd, Molecular Microbiology, 41, 1445–1458

intestine and to adhere to HEp-2 cells. Infect Immun 63:

3739–3744.

McNamara, B.P., and Donnenberg, M.S. (1998) A novel

proline-rich protein, EspF, is secreted from enteropatho-

genic Escherichia coli via the type III export pathway. FEMS

Microbiol Lett 166: 71–78.

Makino, K., Ishii, K., Yasunaga, T., Hattori, M., Yokoyama, K.,

Yutsudo, C.H., et al. (1998) Complete nucleotide

sequences of 93-kb and 3.3-kb plasmids of an enterohe-

morrhagic Escherichia coli O157:H7 derived from Sakai

outbreak. DNA Res 5: 1–9.

Marches, O., Nougayrede, J.P., Boullier, S., Mainil, J.,

Charlier, G., Raymond, I., et al. (2000) Role of tir and

intimin in the virulence of rabbit enteropathogenic Escheri-

chia coli serotype O103:H2. Infect Immun 68: 2171–2182.

Moon, H.W., Whipp, S.C., Argenzio, R.A., Levine, M.M., and

Giannella, R.A. (1983) Attaching and effacing activities of

rabbit and human enteropathogenic Escherichia coli in pig

and rabbit intestines. Infect Immun 41: 1340–1351.

Moreau, V., Frischknecht, F., Reckmann, I., Vincentelli, R.,

Rabut, G., Stewart, D., and Way, M. (2000) A complex of

N-WASP and WIP integrates signalling cascades that lead

to actin polymerization. Nature Cell Biol 2: 441–448.

Mullins, R.D. (2000) How WASP-family proteins and the

Arp2/3 complex convert intracellular signals into cyto-

skeletal structures. Curr Opin Cell Biol 12: 91–96.

Nataro, J., and Kaper, J. (1998) Diarrheagenic Escherichia

coli. Clin Microbiol Rev 11: 142–201.

Newman, J.V., Zabel, B.A., Jha, S.S., and Schauer, D.B.

(1999) Citrobacter rodentium espB is necessary for signal

transduction and for infection of laboratory mice. Infect

Immun 67: 6019–6025.

Nicholls, L., Grant, T.H., and Robins-Browne, R.M. (2000)

Identification of a novel genetic locus that is required for in

vitro adhesion of a clinical isolate of enterohaemorrhagic

Escherichia coli to epithelial cells. Mol Microbiol 35:

275–288.

Ohnishi, M., Tanaka, C., Kuhara, S., Ishii, K., Hattori, M.,

Kurokawa, K., et al. (1999) Chromosome of the entero-

hemorrhagic Escherichia coli O157: H7; comparative

analysis with K-12 MG1655 revealed the acquisition of a

large amount of foreign DNAs. DNA Res 6: 361–368.

Perna, N.T., Mayhew, G.F., Posfai, G., Elliott, S., Donnen-

berg, M.S., Kaper, J.B., and Blattner, F.R. (1998) Molecular

evolution of a pathogenicity island from enterohemorrhagic

Escherichia coli O157:H7. Infect Immun 66: 3810–3817.

Perna, N., Plunkett, G.I., Burland, V., Mau, B., Glasner, J.,

Rose, D., et al. (2001) Genome sequence of entero-

haemorrhagic Escherichia coli O157:H7. Nature 409:

529–533.

Phillips, A.D., Navabpour, S., Hicks, S., Dougan, G., Wallis,

T., and Frankel, G. (2000) Enterohaemorrhagic Escherichia

coli O157:H7 target Peyer’s patches in humans and cause

attaching/effacing lesions in both human and bovineintestine. Gut 47: 377–381.

Reid, S.D., Herbelin, C.J., Bumbaugh, A.C., Selander, R.K.,and Whittam, T.S. (2000) Parallel evolution of virulence in

pathogenic Escherichia coli. Nature 406: 64–67.Rivero-Lezcano, O.M., Marcilla, A., Sameshima, J.H., and

Robbins, K.C. (1995) Wiskott–Aldrich syndrome proteinphysically associates with Nck through Src homology 3

domains. Mol Cell Biol 15: 5725–5731.Rosenshine, I., Donnenberg, M.S., Kaper, J.B., and Finlay,

B.B. (1992) Signal transduction between enteropathogenicEscherichia coli (EPEC) and epithelial cells: EPEC induces

tyrosine phosphorylation of host cell proteins to initiatecytoskeletal rearrangement and bacterial uptake. EMBO J

11: 3551–3560.Rosenshine, I., Ruschkowski, S., Stein, M., Reinscheid, D.,

Mills, S., and Finlay, B. (1996) A pathogenic bacteriumtriggers epithelial signals to form a functional bacterial

receptor that mediates actin pseudopod formation. EMBO J

15: 2613–2624.Songyang, Z., Shoelson, S.E., Chaudhuri, M., Gish, G.,

Pawson, T., Haser, W.G., et al. (1993) SH2 domainsrecognize specific phosphopeptide sequences. Cell 72:

767–778.Suzuki, T., Miki, H., Takenawa, T., and Sasakawa, C. (1998)

Neural Wiscott–Aldrich syndrome protein is implicated inthe actin-based motility of Shigella flexneri. EMBO J 17:

2767–2776.Tacket, C.O., Sztein, M.B., Losonsky, G., Abe, A., Finlay,

B.B., McNamara, B.P., et al. (2000) Role of EspB inexperimental human enteropathogenic Escherichia coli

infection. Infect Immun 68: 3689–3695.Tarr, P.I., Bilge, S.S., Vary, J.C., Jr, Jelacic, S., Habeeb, R.L.,

Ward, T.R., et al. (2000) Iha: a novel Escherichia coliO157:H7 adherence-conferring molecule encoded on a

recently acquired chromosomal island of conservedstructure. Infect Immun 68: 1400–1407.

Taylor, K., O’Connell, C., Luther, P., and Donnenberg, M.(1998) The EspB protein of enteropathogenic Escherichia

coli is targeted to the cytoplasm of infected HeLa cells.Infect Immun 66: 5501–5507.

Tzipori, S., Gunzer, F., Donnenberg, M., de Montigny, L.,Kaper, J., and Donohue-Rolfe, A. (1995) The role of the

eaeA gene in diarrhea and neurological complications in agnotobiotic piglet model of enterohemorrhagic Escherichia

coli infection. Infect Immun 63: 3621–3627.Wachter, C., Beinke, C., Mattes, M., and Schmidt, M. (1999)

Insertion of EspD into epithelial target cell membranes byinfecting enteropathogenic Escherichia coli. Mol Microbiol

31: 1695–1707.Wolff, C., Nisan, I., Hanski, E., Frankel, G., and Rosenshine, I.

(1998) Protein translocation into host epithelial cells by

infecting enteropathogenic Escherichia coli. Mol Microbiol28: 143–155.

1458 R. DeVinney et al.

Q 2001 Blackwell Science Ltd, Molecular Microbiology, 41, 1445–1458