Binding of intimin with Tir on the bacterial surface is prerequisite for the barrier disruption induced by enteropathogenic Escherichia coli

BBRC Biochemical and Biophysical Research Communications 337 (2005) 922–927 www.elsevier.com/locate/ybbrc

Binding of intimin with Tir on the bacterial surface is prerequisite for the barrier disruption induced by enteropathogenic Escherichia coli Masami Miyake a,*, Miyuki Hanajima a, Takeshi Matsuzawa b, Chiho Kobayashi a, Masayoshi Minami a,1, Akio Abe b, Yasuhiko Horiguchi a a

Department of Molecular Bacteriology, Research Institute for Microbial Diseases, Osaka University, 3-1 Yamadaoka, Suita-City, Osaka 565-0871, Japan b Laboratory of Bacterial Infection, Kitasato Institute for Life Sciences, Kitasato University, 5-9-1 Shirokane, Minato-ku, Tokyo 108-8641, Japan Received 17 September 2005 Available online 30 September 2005

Abstract Enteropathogenic Escherichia coli (EPEC) infects intestinal epithelial cells and perturbs the intestinal barrier that limits the paracellular movement of molecules. The disruption of the barrier is mediated by the effectors translocated into the host cells through the bacterial type III secretion system (TTSS). A previous report has described the importance of a bacterial outer membrane protein, intimin, in EPEC-mediated disruption of the barrier, and proposed that intimin, in concert with a host intimin receptor, controls the activity of the translocated barrier-disrupting effectors [P. Dean, B. Kenny, Intestinal barrier dysfunction by enteropathogenic Escherichia coli is mediated by two effector molecules and a bacterial surface protein, Mol. Microbiol. 54 (2004) 665–675]. In this study, we found that the importance of intimin is in its ability to bind a bacterial intimin receptor, Tir. Additionally, the impaired ability of an intimin-negative mutant was not restored by co-infection with intimin-expressing TTSS mutants. Collectively, the results in this study favor an alternative scenario explaining the importance of intimin, that the binding of intimin with Tir on the bacterial surface triggers or promotes the translocation of factors required for the efficient disruption of the barrier. Thus, the interaction of intimin with Tir may serve as a molecular switch that controls the delivery of virulence factors into the host cells.
2005 Elsevier Inc. All rights reserved. Keywords: Type III secretion system; EPEC; Contact dependent; Effector secretion

Enteropathogenic Escherichia coli (EPEC) is an important cause of severe infantile diarrhea especially in developing countries [1]. The pathogenesis of diarrhea due to EPEC is incompletely understood, although it may involve an ability to cause both morphological and functional alterations of the infected host cells. One characteristic morphological change is the rearrangement of the cytoskeleton to form actin-rich membrane protrusions, termed pedestals, beneath the attached bacteria [2]. The ability to produce these pedestals is mediated by a unique secretory machinery, the type III secretion system (TTSS), that delivers virulence-associated proteins, termed effectors, directly *

Corresponding author. Fax: +81 6 6879 8283. E-mail address: [email protected] (M. Miyake). 1 Present address: National Tokyo Medical Center, 2-5-1 Higashi-gaoka, Meguro-ku, Tokyo 152-8902, Japan. 0006-291X/$ - see front matter
2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2005.09.130

into host cells [3,4]. One of the effectors delivered through TTSS is the translocated intimin receptor (Tir) essential for the formation of the pedestals. Tir inserts into the plasma membrane featuring a central extracellular domain that serves as a receptor for an outer membrane protein, intimin. The binding of intimin to Tir triggers phosphorylation of tyrosine 474 at the cytoplasmic C-terminus of Tir. Phosphorylated Y474 and flanking residues bind the host adaptor protein Nck, an activator of the N-WASP-Arp2/3 pathway of actin assembly in mammalian cells [2]. The intercellular tight junction acts as a paracellular seal between epithelial cells and regulates permeability across intestinal mucosa [5]. EPEC has been reported to disrupt the barrier to increase permeability across an epithelial cell monolayer [6]. The EPEC-induced disruption also depends on TTSS, suggesting that the disruption inducers are TTSS effectors [7–12]. It has also been reported that intimin is

M. Miyake et al. / Biochemical and Biophysical Research Communications 337 (2005) 922–927

important for EPEC to perturb the barrier function [6,8,13,14]. On the other hand, the role of Tir, a binding partner for intimin, in the pertubation is controversial [8,15]. Deans and Kenny have recently described the dispensability of Tir for the EPEC-induced disruption of the barrier, and concluded that intimin might control the activities of barrier-disrupting effectors through a signal elicited by its binding to a host-derived membrane component [8]. In this study, we show the importance both of intimin and of Tir for EPEC to disrupt the epithelial barrier and that the binding between Tir and intimin features in the disruption. Our findings consequently suggest a regulatory role for the binding in the translocation of certain effectors through TTSS. Materials and methods Bacterial strains, plasmids, and bacterial culture conditions. E2348/69 (O127:K63:H6) is a wild-type strain of EPEC [16]. Mutant strains prepared from E2348/69 have been reported elsewhere [17]. All the bacterial strains and plasmids used in this study are listed in sTable 1. Plasmids were prepared as described in the supplementary text using primers listed in sTable 2. EPEC and its derivatives were cultured in LB broth at 37 C overnight and activated in DulbeccoÕs modified EagleÕs medium (DMEM) for 3 h in a CO2-incubator. The culture was diluted with DMEM and subjected to infection assays. In each experiment, the CFUÕs of bacteria in each inoculum were enumerated and the multiplicity of infection (moi) was calculated. Cell culture. Human adenocarcinoma Caco-2 cells were maintained in DMEM supplemented with 20% fetal bovine serum (FBS, Sigma). For determination of a transepithelial electrical resistance (TER) across the Caco-2 monolayer, Caco-2 cells were seeded on collagen-coated permeable membrane supports (Transwell COL, Corning), placed in 24-well plates at a density of 4–5 · 104 cells per well, and incubated for 5 days. A portion of 50 ll of bacterial suspension was added into the apical compartment of the Transwell chamber. The TER across the monolayer was measured with Millicell-ERS (Millipore) every 1–2 h. Initial TERs were between 200 and 350 X cm. The values are comparable to those reported previously [7]. When two strains were co-infected in the same chamber (double-infection experiment), each bacterial inoculum prepared to show the equivalent number of CFUs was mixed in equal volumes prior to infection, and 50 ll was added into the chamber. To examine the effect of soluble forms of intimin on the EPEC-induced loss of TER, a portion of 100 ll of culture medium was removed and a mixture of 50 ll of bacterial suspension and 50 ll of intimin solution was added into the chamber. Preparation of GST fusion proteins. GST and GST fusion proteins were expressed in Escherichia coli BL21(DE3) from plasmids and purified from the bacterial cell lysate with glutathione-coupled resin (GSTrap, Amersham Bioscience) by the method recommended by the manufacturer. The protein content was determined with a Micro BCA protein assay reagent (Pierce Biotechnology) using bovine serum albumin (BSA) as a standard. The purified protein was dialyzed against phosphate-buffered saline, pH 7.0 (PBS), overnight, filter-sterilized, and stored at 80 C prior to use. Detection of the secreted proteins from EPEC and the mutants. EPEC and the mutants activated in DMEM as described above were washed with and resuspended in methionine- and cystein-free DMEM (Gibco) to give a final density at 4–8 · 107 CFU/ml. The bacterial suspension (3.5 ml) was incubated with 170 lg GST or GST–Tir, in the presence of 3.7 MBq of redivue Pro-mix L-[35S] in vitro cell labelling mix (Amersham Bioscience) for 2 h in a CO2-incubator. The culture supernatant was centrifuged, filtered (0.22 lm pore size), and subjected to SDS–PAGE. After being stained with Coomassie brilliant blue, the gel was dried on a filter paper, and the radioactivity was detected with BAS 1000 (Fujifilm, Japan). Microscopy. Caco-2 cells cultured for 5 days in the Transwell Clear (Corning) were infected with bacteria and incubated for a period in a CO2-

923

incubator. After the incubation, the cells were washed and fixed with 3.7% formaldehyde in PBS for 20 min, permeabilized with 0.1% Triton X-100 for 10 min, and subjected to the staining of F-actin with rhodamine– phalloidin (Invitrogen), and of intimin with anti-intimin rabbit serum and Alexa350-conjugated anti-rabbit IgG antibody. Immunostained samples were visualized using an epifluorescence microscope (BX50, Olympus). Anti-intimin serum was prepared according to the standard protocol [18], by injecting recombinant full-length intimin fused with GST subcutaneously into female Japanese white rabbits.

Results and discussion Binding of intimin with Tir is required for EPEC-induced loss of TER We initiated this study to examine if intimin-deficient (Deae) and Tir-deficient (Dtir) mutants can disrupt the epithelial barrier by measuring TER across Caco-2 monolayers [6]. The infection of Caco-2 monolayers by wild-type EPEC strain, E2348/69, caused a decrease of TER at 4 h, and the decline continued with time, until finally the background was reached in 8 h. In contrast, neither Dtir nor Deae caused a significant decrease in TER within 9 h (Fig. 1A). Even a 10-fold higher dose (moi = 30) did not induce a significant loss of TER within 8 h (Deae) or 6 h (Dtir) post-infection (data not shown). Complementation of these mutants with the corresponding genes restored their TER-lowering activity (sFigs. 1A and B), suggesting that the phenotypes of these mutants were not a result of polar effects or of unwanted mutations. The results suggest that the intimin-mediated binding of EPEC to Tir on the host membrane plays an important role in the EPEC-induced loss of TER. The necessity of the binding was demonstrated utilizing a recombinant C-terminal extracellular domain (D555–K939) of intimin fused with GST at its C-terminus (GST-Int0123, Fig. 1B). The ability of the fusion protein to bind to Tir was confirmed with a CellELISA [19], in which Tir was expressed on the Caco-2 cells with Deae and the binding was detected using anti-GST antibody (sFig. 2). Addition of GST-Int0123 into the Transwell caused a delay in the onset of the E2348/69-induced decrease in TER in a dose-dependent manner (Fig. 1C). It seems unlikely that the inhibitory effect is due to the disturbance of EPEC binding to the monolayer, because E2348/69 adhered to a similar extent even in the presence of GST-Int0123 (sTable 3). Furthermore, the inhibitory effect of GST-Int0123 was dependent on the Tir-binding region of the fusion protein, which is located in the C-terminal lectin-like D3 domain among the four domains of the extracellular part of intimin (Fig. 1B) [20]. Truncation of the N-terminal D0 did not affect the ability of the fusion protein to inhibit the EPEC-induced decrease in TER (Fig. 1D), whereas a deletion of C-terminal D3 abrogated the ability both to bind to Tir (sFig. 2) and to inhibit the decrease in TER. The results indicate that competitive inhibition of the interaction between bacterial surface intimin and Tir prevents the EPEC-induced disruption of the epithelial barrier.

924

M. Miyake et al. / Biochemical and Biophysical Research Communications 337 (2005) 922–927

Fig. 1. Binding of intimin with Tir is required for EPEC-induced loss of TER. (A) Time course of changes in TER in Caco-2 monolayers induced by EPEC, Deae, or Dtir. Caco-2 monolayers were infected with wild-type E2348/69 (closed square), Deae (closed circle), or Dtir (open triangle) at a moi of 1–3. After addition of the bacterium, the monolayers were incubated in a CO2-incubator, and TER across the Caco-2 monolayer was determined at the indicated time. TER of the uninfected monolayer is also shown (open square). Data represent means (±SD) of three to four independent experiments. (B) Structures of intimin and GST–intimin fusion proteins used in this study. (C) GST-Int0123 or GST was added into the Transwell simultaneously with E2348/69 (moi = 50) and incubated in a CO2-incubator. Open circle, uninfected control; closed square, E2348/69 alone; open square, GST-Int0123 alone (50 lg per well); closed circle, E2348/69 + GST-Int0123 (50 lg per well); closed triangle, E2348/69 + GST-Int0123 (25 lg per well); closed diamond, E2348/69 + GST-Int0123 (12.5 lg per well); open triangle, E2348/69 + GST (25 lg per well). Data represent means (±SD) of three independent experiments. (D) Inhibition of the EPEC-induced decrease in TER by soluble intimins is dependent on the Tir-binding region. Caco-2 monolayers were infected with E2348/69 at a moi of 50, in the presence or absence of 25 lg GST or GST–intimins and incubated in a CO2-incubator. TERs at 4 h postinfection are shown. Data represent the means (±SD) of three to four independent experiments. (*) Significantly different (p < 0.002) against Ônone.Õ

Tir–intimin interaction plays a role in the efficient translocation of barrier-disrupting effectors into the host cell cytoplasm One interpretation of the intimin-dependent disruption of the barrier is that intimin regulates the activity of the translocated barrier-disruption inducers through its binding with a non-Tir host receptor, as proposed by a previous study [8]. This seemed unlikely in the current study, because Dtir was not competent for the induction of a loss of TER (Fig. 1A). Another possibility is that the interaction of intimin with Tir activates the translocated effectors through stimulation of a parallel signaling pathway. To examine this possibility, we infected Caco-2 monolayers with Deae simultaneously with intimin-expressing mutants and observed a significant decrease in TER (Fig. 2A). The possible explanation that Tir–intimin interaction on the bacterial surface activates the translocated effectors delivered from Deae turned out to be unlikely, because other mutants, DespA, DespB, and DespD, which are defective in TTSS function but express intimin on their surface [4], caused no decrease in TER even in cells co-infected with Deae (Fig. 2B, and data not shown). With either of the double-infection assays, actin

polymerization beneath the attached bacteria was observed (Figs. 2C–F), indicating the binding of intimin with Tir and the generation of a downstream signal triggering actin polymerization in both double-infection experiments. The requirement of a functional TTSS on a bacterial cell in addition to the Tir–intimin interaction raises the possibility that the intimate attachment of EPEC to the host cells through the binding between intimin and Tir regulates the translocation of barrier-disrupting effectors. Because the translocation of effectors through TTSS is generally energy-dependent [21], we then examined if metabolism in Dtir is important for the loss of TER. Caco-2 monolayers were primed with Deae for 2 h and then challenged with Dtir after Deae was washed out. As shown in Fig. 3, TER significantly decreased at 4 h post-challenge to 27.5% (Fig. 3, columns C). Even when the metabolism in Deae was disturbed with gentamicin for 30 min prior to the challenge, Dtir caused a similar loss of TER (Fig. 3, columns D), suggesting that the metabolism in Deae is no longer required, once sufficient priming was achieved. In contrast, no significant decrease of TER was observed when gentamicin was present simultaneously with Dtir (Fig. 3, columns E), indicating that Dtir must be metabolically active to induce the efficient loss of

M. Miyake et al. / Biochemical and Biophysical Research Communications 337 (2005) 922–927

925

Fig. 2. Tir, intimin, and TTSS are necessary for EPEC to induce a decrease in TER across Caco-2 monolayers. Caco-2 monolayers were mono-infected with E2348/69 (A,B, closed square), Deae (A,B, open square), Dtir (A, closed diamond), or DespA (B, closed circle), or were double-infected with Deae and Dtir (A, closed triangle), or Deae and DespA (B, open triangle). The infected cells were incubated in a CO2-incubator and the TER was determined at the indicated time. Multiplicity of infection; moi of 3 for E2348/693; moi of 5 for other EPEC mutants. Data represent means (±SD) of three independent experiments. Actin polymerization beneath the infected bacteria is shown in (C–F). Bacterial infections were performed under the same conditions as shown in (A,B). Caco-2 monolayers infected either by E2348/69 (C), by Deae with Dtir (D), by Deae with DespA (E), or by Deae with DespB (F) were subjected to the staining of intimin and F-actin at 4 h post-infection. Blue represents the bacterial microcolonies expressing intimin. Note that the blue microcolonies co-localized with red F-actin aggregates. Scale bars, 10 lm.

TER under the experimental conditions. The results support our idea that the binding of intimin with Tir regulates the secretion/translocation of certain factors necessary for a maximal induction of the decrease in TER. Binding of soluble Tir with intimin on the bacterial surface triggers secretion of a protein in a TTSS-dependent manner Next, we tried to identify the effector that is secreted in response to the binding of Tir with intimin. EPEC and

EPEC mutants were incubated in DMEM with [35S] amino acids to label the secreted proteins. During the incubation, a soluble recombinant GST–Tir fusion protein was used as a secretion stimulator. When wild-type EPEC was incubated in the absence of the stimulator, we detected four major secreted proteins in the culture supernatant (Fig. 4A). Three of the secreted proteins were identified as EspA, EspB, and EspD, using the respective deletion mutants, and by their predicted molecular masses (data not shown). Addition of GST in the culture medium did not affect the

926

M. Miyake et al. / Biochemical and Biophysical Research Communications 337 (2005) 922–927

Fig. 3. The treatment of Dtir with gentamicin disturbs the ability of the bacteria to induce a loss of TER in double-infection experiments. Caco-2 monolayers were primed with Deae at a moi of 40–50 (A,C–E) or left unprimed (B) for 2 h. After the priming, the initial TER (Ôpre-challengeÕ) was determined (white bars). The monolayer was washed and challenged with Dtir at a moi of 50 in the presence (E) or absence (B,C) of 100 lg/ml of gentamicin or further incubated without the challenging bacteria (A). A Deae-infected monolayer was treated during priming with 100 lg/ml of gentamicin for 30 min prior to the challenge (D). TER at 4 h postchallenge was determined (gray bars). Data represent means (±SD) of three independent experiments. Significance was examined with StudentÕs t test comparing TER Ôpre-challengeÕ and Ô4 h post-challenge.Õ p values are shown above the columns.

overall band pattern, except that a broad 30-kDa radioactive band appeared. The band was suspected to be a byproduct generated from GST and [35S]cystein in the medium, because GST protein was detected at the corresponding position in the gel by the staining with Coomassie brilliant blue. Notably, a 24-kDa protein band emerged when GST–Tir was added during the culture. A TTSS-defective mutant, DespC, did not produce the 24-kDa band (Fig. 4B), indicating the TTSS-dependent secretion of the 24-kDa protein in the culture medium. The results suggest that the interaction of soluble Tir with intimin on the bacterial cell surface induces the secretion of a 24-kDa protein into the culture medium in a TTSS-dependent manner. In summary, our observations do not favor the notion that intimin regulates the activity of the translocated barrier-disruption inducers [8]. Rather, we prefer an alternative model, that Tir–intimin interaction results in the delivery of the factors necessary to disrupt the barrier more efficiently. The results may also suggest a new role for intimin and Tir, as a bacterial sensory system to recognize the presence of the host cells. The interaction between the proteins may consequently trigger the effective delivery of a barrier-disrupting factor, or a regulator controlling the barrier-disrupting factor, into the host cell through TTSS. Actually, several reports have described contact-dependent effector secretion in EPEC [22], and that the efficient translocation of effectors can be controlled by the Tir–intimin interaction [10,15]. Further study has to be performed to identify the factors delivered into the host cytoplasm in an intimate attachmentdependent manner. So far, three LEE-encoded effectors, EspF, Map, and EspG, and a non-LEE-encoded EspG homolog, Orf3, are known to be required for the disruption [8–10], among which the translocation of EspG into the host cells may be contact-dependent [10]. Our prediction is that EspF is not the effector that is secreted intimin-dependently, because EspF was translocated efficiently from EPEC into host cytoplasm even in the absence of Tir and intimin (our unpublished observations). This was shown by a translocation monitoring system using Bordetella adenylate cyclase toxin as a reporter [9,23]. Whether other effectors are translocated in an intimin-dependent manner is currently under investigation. Acknowledgments

Fig. 4. The treatment of EPEC with recombinant GST–Tir induced the de novo secretion of a TTSS effector. Wild-type EPEC, E2348/69 (A), or EPEC mutants, Dtir or DespC (B) were activated in DMEM for 3 h, and were washed and resuspended in methionine- and cystein-free DMEM supplemented with [35S]methionine and [35S]cystein. After addition of PBS, GST, or GST–Tir, the suspension was incubated in a CO2-incubator for 2 h. The filter-sterilized culture supernatant was collected and subjected to SDS–PAGE and autoradiography using BAS 1000. Positions of the molecular mass markers are shown at the left side of the gel (A).

We are grateful to Katsuya Hirata, Sami Fujihara, and Kuniko Ishiguro for technical and secretarial assistance. This work was supported by Grants-in-Aid for Scientific Research on Priority Areas (No. 12144208 and No. 16017255), and for Scientific Research (B) (No. 15390142) from the Ministry of Education, Culture, Sport, Science and Technology of Japan, and by a grant from Ôthe 21st Century COE Program, Combined Program on Microbiology and ImmunologyÕ from the Japan Society for the Promotion of Science.

M. Miyake et al. / Biochemical and Biophysical Research Communications 337 (2005) 922–927

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/ j.bbrc.2005.09.130. References [1] J.P. Nataro, J.B. Kaper, Diarrheagenic Escherichia coli, Clin. Microbiol. Rev. 11 (1998) 142–201. [2] J.B. Kaper, J.P. Nataro, H.L.T. Mobley, Pathogenic Escherichia coli, Nat. Rev. Microbiol. 2 (2004) 123–140. [3] A. Blocker, K. Komiyama, S.-I. Aizawa, Type III secretion systems and bacterial flagella: insights into their function from structural similarities, Proc. Natl. Acad. Sci. USA 100 (2003) 3027–3030. [4] K. Sekiya, M. Ohishi, T. Ogino, K. Tamano, C. Sasakawa, A. Abe, Supermolecular structure of the enteropathogenic Escherichia coli type III secretion system and its direct interaction with the EspAsheath-like structure, Proc. Natl. Acad. Sci. USA 98 (2001) 11638– 11643. [5] S.-I. Tsukita, M. Furuse, M. Itoh, Multifunctional strands in tight junction, Nat. Rev. Mol. Cell Biol. 2 (2001) 285–293. [6] C. Canil, I. Rosenshine, S. Ruschkowski, M.S. Donnenberg, J.B. Kaper, B.B. Finlay, Enteropathogenic Escherichia coli decreases the transepithelial electrical resistance of polarized epithelial monolayers, Infect. Immun. 61 (1993) 2755–2762. [7] G.K. Collington, I.W. Booth, S. Knutton, Rapid modulation of electrolyte transport in Caco-2 cell monolayers by enteropathogenic Escherichia coli (EPEC) infection, Gut 42 (1998) 200–207. [8] P. Dean, B. Kenny, Intestinal barrier dysfunction by enteropathogenic Escherichia coli is mediated by two effector molecules and a bacterial surface protein, Mol. Microbiol. 54 (2004) 665–675. [9] B.P. McNamara, A. Koutsouris, C.B. OÕConnell, J.-P. Nougayre´de, M.S. Donnenberg, G. Hecht, Translocated EspF protein from enteropathogenic Escherichia coli disrupts host intestinal barrier function, J. Clin. Invest. 107 (2001) 621–629. [10] F.L. Tomson, V.K. Viswanathan, K.J. Kanack, Rajani, P. Kanteti, K.V. Straub, M. Menet, J.B. Kaper, G. Hecht, Enteropathogenic Escherichia coli EspG disrupts microtubules and in conjunction with Orf3 enhances perturbation of the tight junction barrier, Mol. Microbiol. 56 (2005) 447–464. [11] S.J. Elliott, E.O. Krejany, J.L. Mellies, R.M. Robins-Browne, C. Sasakawa, J.B. Kaper, EspG, a novel type III system-secreted protein

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

927

from enteropathogenic Escherichia coli with similarities to VirA of Shigella flexneri, Infect. Immun. 69 (2001) 4027–4033. S.J. Elliott, C.B. OÕConnell, A. Koutsouris, C. Brinkley, M.S. Donnenberg, G. Hecht, J.B. Kaper, A gene from the locus of enterocyte effacement that is required for enteropathogenic Escherichia coli to increase tight-junction permeability encodes a chaperone for EspF, Infect. Immun. 70 (2002) 2271–2277. D.J. Philpott, D.M. McKay, P.M. Sherman, M.H. Perdue, Infection of T84 cells with enteropathogenic Escherichia coli alters barrier and transport functions, Am. J. Physiol. 33 (1996) G634–G645. G.K. Collington, I.W. Booth, M.S. Donnenberg, J.B. Kaper, S. Knutton, Enteropathogenic Escherichia coli virulence genes encoding secreted signaling proteins are essential for modulation of Caco-2 cell electrolyte transport, Infect. Immun. 66 (1998) 6049–6053. M.M. Muza-Moons, A. Koutsouris, G. Hecht, Disruption of cell polarity by enteropathogenic Escherichia coli enables basolateral membrane proteins to migrate apically and to potentiate physiological consequences, Infect. Immun. 71 (2003) 7069–7078. M.M. Levine, E.J. Bergquist, D.R. Nalin, D.H. Waterman, R.B. Hornick, C.R. Young, S. Sotman, Escherichia coli strains that cause diarrhoea but do not produce heat-labile or heat-stable enterotoxins and are non-invasive, Lancet (1978) 1119–1122. T. Matsuzawa, A. Kuwae, S. Yoshida, C. Sasakawa, A. Abe, Enteropathogenic Escherichia coli activates the RhoA signaling pathway via the stimulation of GEF-H1, EMBO J. 23 (2004) 3570– 3582. E. Harlow, D. Lane, Immunizations, in: E. Harlow, D. Lane (Eds.), Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory publications, Cold Spring Harbor, 1988, pp. 53–137. H. Liu, L. Magoun, S. Luperchio, D.B. Schauer, J.M. Leong, The Tir-binding region of enteropathogenic Escherichia coli intimin is sufficient to trigger actin condensation after bacterial-induced host cell signaling, Mol. Microbiol. 34 (1999) 67–81. Y. Luo, E.A. Frey, R.A. Pfuetzner, A.L. Creagh, D.G. Knoechel, C.A. Haynes, B.B. Finlay, N.C.J. Strynadka, Crystal structure of enteropathogenic Escherichia coli intimin–receptor complex, Nature 405 (2000) 1073–1077. A.P. Tampakaki, V.E. Fadouloglou, A.D. Gazi, N.J. Panopoulos, M. Kokkinidis, Conserved features of type III secretion, Cell Microbiol. 6 (2004) 805–816. C. Wolff, I. Nisan, E. Hanski, G. Frankel, I. Rosenshine, Protein translocation into host epithelial cells by infecting enteropathogenic Escherichia coli, Mol. Microbiol. 28 (1998) 143–155. N. Dautin, G. Karimova, D. Ladant, Bordetella pertussis adenylate cyclase toxin: a versatile screening tool, Toxicon 40 (2002) 1383–1387.