Enterohaemorrhagic Escherichia coli serogroup O111 inhibits NF-κB-dependent innate responses in a manner independent of a type III secreted OspG orthologue

Abstract

Enterohaemorrhagic and enteropathogenic Escherichia coli (EHEC and EPEC) inject a repertoire of effector proteins into host cells via a type III secretion system (T3SS) encoded by the locus of enterocyte effacement (LEE). OspG is an effector protein initially identified in Shigella that was shown to inhibit the host innate immune response. In this study, we found ospG homologues in EHEC (mainly of serogroup O111) and in Yersinia enterocolitica. The T3SS encoded by the LEE was able to inject these different OspG homologues into host cells. Infection of HeLa cells with EHEC O111 inhibited the NF-κB-dependent innate immune response via a T3SS-dependent mechanism. However, an EHEC O111 ospG mutant was still able to inhibit NF-κB p65 transfer to the nucleus in infected cells stimulated by tumour necrosis factor α (TNF-α). In addition, no difference in the inflammatory response was observed between wild-type EHEC O111 and the isogenic ospG mutant in the rabbit ligated intestinal loop model. These results suggest that OspG is not the sole effector protein involved in the inactivation of the host innate immune system during EHEC O111 infection.

The GenBank/EMBL/DDBJ accession numbers for the sequences (between the lom and yciE genes) reported in this paper are AB438937 (strain 12009), and AB438936 (strain 11128).

Edited by: M. P. Stevens

INTRODUCTION

Enterohaemorrhagic Escherichia coli (EHEC) are food-borne pathogens that cause severe diarrhoea and haemorrhagic colitis that may be complicated by thrombotic thrombocytopenic purpura or haemolytic uraemic syndrome (Kaper et al., 2004). EHEC O157 is the predominant serogroup associated with large outbreaks, but the clinical importance of EHEC O111, O26 and O103 strains is increasingly recognized (Ogura et al., 2007). Shiga toxins (Stx) are the major determinants of EHEC pathogenicity (Karmali, 1989). In addition, a repertoire of effector proteins secreted via a type III secretion system (T3SS) is recognized as essential for EHEC virulence as for enteropathogenic E. coli (EPEC) (Garmendia et al., 2005). These effector proteins are coded not only by a pathogenicity island that codes for the T3SS and is called the locus of enterocyte effacement (LEE), but also by other chromosomal loci, mainly exchangeable effector loci (EEL) within lambdoid prophages (Loukiadis et al., 2008; Ogura et al., 2007; Tobe et al., 2006).

Homeostasis of the mammalian intestinal epithelium is maintained by keeping a balance between tolerance and response to luminal micro-organisms. Infection with enteric pathogens disrupts this balance, leading to intestinal inflammation (Sansonetti, 2004). The family of NF-κB transcription factors essentially regulates immune-related gene expression. Without stimulation, NF-κB is in an inactive state, bound to its inhibitor IκB in the cytoplasm. Following activation, IκB undergoes phosphorylation, ubiquitinylation and subsequent degradation by the proteasome, thus allowing NF-κB to translocate to the nucleus and to regulate the transcription of genes involved in the inflammatory response, including interleukin (IL)-8 (Li & Verma, 2002). The host inflammatory response against infections with EHEC or EPEC is the result of a balance between bacterial extracellular pro-inflammatory components and intracellular anti-inflammatory effectors that are injected into host cells (Sharma et al., 2006). Several studies indicate that EHEC and EPEC activate the NF-κB signalling pathway (Dahan et al., 2002; Savkovic et al., 1997) and it was reported that flagellin is a major pro-inflammatory mediator (Berin et al., 2002; Zhou et al., 2003). It was also reported that infection of HeLa cells with EHEC and EPEC suppresses NF-κB activation in a T3SS-dependent manner (Hauf & Chakraborty, 2003). Another study using human colonic epithelial cells confirms a role for the T3SS in the ability of EHEC to modulate the host innate response (Gobert et al., 2005). Ruchaud-Sparagano et al. (2007) reported that infection of differentiated Caco-2 cells with EPEC prevents the activation and translocation of NF-κB upon tumour necrosis factor α (TNF-α) stimulation in a T3SS-dependent manner, but this mechanism is independent of all LEE-encoded effectors and at least four non-LEE-encoded effectors (Orf3, NleA, NleF and NleH). These findings suggest that unknown EHEC/EPEC effector(s) is/are involved in the modulation of inflammatory response.

Several T3SS effectors from animal and plant pathogens have been reported to inhibit the NF-κB signalling pathway (Bhavsar et al., 2007; Espinosa & Alfano, 2004). Among these T3SS effectors, Shigella flexneri OspG binds various ubiquitinylated ubiquitin-conjugating enzymes and negatively controls the host innate response induced by Shigella upon invasion of the epithelium (Kim et al., 2005). YopP/J, produced by Yersinia spp., acetylates and inhibits kinase activity of IKK and consequently blocks the NF-κB signalling pathway (Mukherjee et al., 2006).

In this study, we found that ospG homologues are present not only in Shigella spp. but also in non-O157 EHEC and in Yersinia enterocolitica. We observed that these OspG homologues could be translocated into host cells through the T3SS encoded by the LEE. However, the deletion of the ospG homologue did not affect the inflammatory response to EHEC infection in vitro or in vivo in the rabbit ligated intestinal loop model.

METHODS

Bacterial strains, plasmids and cell lines.

Bacterial strains and plasmids used in this study are shown in Table 1⇓. HeLa (ATCC CCL-2) and RK13 actin-GFP cells (Peralta-Ramirez et al., 2008) were maintained in D-MEM (Invitrogen) supplemented with 10 % (v/v) fetal calf serum (FCS, Eurobio) and 80 μg gentamicin ml⁻¹ at 37 °C in 5 % CO₂. G418 (Invivogen) was added at 400 μg ml⁻¹ to RK13 actin-GFP cells.

Table 1.

Strains, plasmids and primers used in this study

Sequencing and in silico analysis.

The sequences of EELs encoding OspG homologues within prophages in EHEC O111 : H− strain 11128 and EHEC O103 : H2 strain 12009 were determined as a part of the genome sequencing project that is in progress in the laboratory of one of the authors (T. H.). The sequence assembly for each OspG homologue encoding EELs was confirmed by PCR using primers targeting the lom family gene and the yciE gene (Table 1⇑). In silico analyses were done with the ORF Finder software () and blast programs (Altschul et al., 1990) The Mac Vector software (Mac Vector, Inc.) was used to create a phylogenetic tree for OspG homologues using the Kimura two-parameter neighbour-joining method.

Construction of mutants and plasmids.

To remove the kanamycin cassette from E22 escN : : kmlox (Marches et al., 2003), plasmid pJW168 was used (Wild et al., 1998). To delete the escN gene in EHEC strain 13369, a DNA fragment including 500 bp up- and downstream of escN, a kanamycin cassette and flanking FRT (Flp recombination target) sites was prepared by PCR using primers Kmcassette-F and Kmcassette-R (Table 1⇑) and plasmid pKD4 as the template. The fragment was cloned into suicide vector pKNG101 to generate pKNG101-escN : : kmFRT. To delete the ospG homologue in EHEC strain 13369, a DNA fragment was amplified by the overlap extension PCR method (Ho et al., 1989), using primers mutospG1-F, mutospG1-R, mutospG2-F, mutospG2-R, mutospGKm-F and mutospGKm-R (Table 1⇑), with the genomic DNA and pKD4 plasmid DNA as templates. This fragment was cloned into pKNG101 to generate pKNG101-kmospG : : kmFRT. The resulting plasmids were electroporated in SM10λpir, and then transferred into EHEC strain 13369 by conjugation. EHEC mutants were selected at 30 °C on MinA plates (Miller, 1992) and the FRT-kanamycin cassette was removed using plasmid pCP20 (Cherepanov & Wackernagel, 1995). The ospG genes (excluding their stop codons) were amplified by PCR with primers ospGgateway-F and ospGgateway-R (Table 1⇑) using the genomic DNA of EHEC strain 11128, S. flexneri strain SF620 and Y. enterocolitica strain 8081 (a kind gift from Dr Elisabeth Carniel, Institut Pasteur, Paris, France). The genes were cloned into pDONR201 vector (Invitrogen), and derivatives of the pKTEM vector encoding OspG proteins fused to the TEM reporter enzyme, were named pKospGTEM (OspG from EHEC O111), pKshiospGTEM (S. flexneri) and pKyerospGTEM (Y. enterocolitica), respectively. These plasmids were electroporated into EPEC strains E22 and E22ΔescN, and EHEC strains 13369 and 13369ΔescN. For complementation, plasmid pABBospG, encoding the EHEC OspG homologue, was constructed by using the Gateway cloning system with pDONR201, encoding OspG, and pABB-99ccdB.V5His6 (a kind gift from Dr Akio Abe, Kitasato University, Tokyo, Japan). The pABBospG and the pUC18-OspG plasmid encoding OspG from S. flexneri (a kind gift from Dr Claude Parsot, Institut Pasteur, Paris, France) were electroporated into the EHEC ospG-deletion mutant.

Translocation assay.

The production and secretion of TEM fusion proteins were carried out as described previously (Charpentier & Oswald, 2004; Jubelin et al., 2009), except that bacteria were grown in D-MEM/HEPES (Invitrogen) with kanamycin at 37 °C for 6.5 h without adding IPTG. The data presented (normalized 450/520 nm emission ratio) are the mean values of the results from triplicate wells from three independent experiments and bars indicate standard error of the mean (sem). For statistical analyses, two-sided Student's t-test was used with independent samples. P-values <0.05 were considered statistically significant.

RT-PCR.

RNA was isolated from bacteria grown in D-MEM/HEPES supplemented with 5 % (v/v) fetal calf serum (FCS) at 37 °C, 5 % CO₂ to OD₆₀₀ ∼0.5, using TRIZOL (Invitrogen) and then treated with the TURBO DNA-free kit (Ambion). One microgram of total RNA and random primers were used for reverse transcription (RT) using the SuperScript III First-Strand Synthesis System (Invitrogen). The PCR amplification was done with RT reactant and primers ospG-F and ospG-R or primers 16S rRNA-F and 16S rRNA-R (Table 1⇑).

Fluorescent-actin staining (FAS) assay.

RK13 actin-GFP cells were inoculated with the bacteria (m.o.i. 50) and incubated for 6 h, then F-actin was visualized as described previously (Peralta-Ramirez et al., 2008).

Analysis of NF-κB (p65) translocation and IκB-α degradation.

Two days before the infection, 2×10⁶ HeLa cells were seeded on 100 mm dishes with 10 ml D-MEM 10 % (v/v) FCS 0.1 mM non-essential amino acids (NEAA, Invitrogen). Twenty hours before infection, the culture medium was replaced with D-MEM 1 % (v/v) FCS and NEAA. Overnight bacterial LB cultures were inoculated (1 : 100) in D-MEM/HEPES and grown at 37 °C for 3 h. The bacteria were added to the HeLa cells (m.o.i. 50) and incubated at 37 °C, 5 % CO₂ for 1–6 h. The cells were washed three times with Hanks' balanced salt solution (HBSS), and 10 ml D-MEM containing TNF-α (20 ng ml⁻¹, Peprotech) was added to the cells and incubated for 30 min. Cells were lysed and fractionated into cytoplasmic and nuclear fractions using a Nuclear Extract kit (Active motif). Ten micrograms of nuclear proteins was used for detection of NF-κB (p65) nuclear translocation using the TransAM NFκB (p65) kit (Active motif). The cytoplasmic fractions were subjected to Western blot analysis with anti-IκB-α antibody.

Western blot analyses.

Proteins were separated on NuPAGE gels and blotted on PVDF membranes (Invitrogen). Western blot analyses were performed using anti-TEM antibody (Gentaur), anti-IκB-α antibody (Upstate) and anti-actin antibody (Sigma), combined with peroxidase-conjugated secondary antibodies. To strip and reprobe the membranes, the Restore Western Blot Stripping Buffer (Pierce) was used. Proteins were detected by chemiluminescence using ECL plus (GE Healthcare) and the ChemiDoc XRS system (Bio-Rad).

Ligated intestinal loop assay.

The assay was performed as previously described (Goffaux et al., 1997), with some modifications. All strains (EHEC 13369, 13369ΔospG, 13369ΔospG hosting the pABBospG plasmid, 13369ΔescN, and the non-EPEC/EHEC negative control strain T282) were tested in four 12-week-old rabbits (Charles River Laboratories). In each rabbit five loops were injected with the bacteria and one additional loop was injected with sterile culture medium. The rabbits were anaesthetized with isoflurane and euthanized after 9.5 h (two rabbits) or 12.5 h (two rabbits). The rabbits received one injection of buprenorphone (0.02 mg kg⁻¹) at the end of the surgery. After euthanasia the intestines were immediately removed and fixed with 4 % formalin/1 % glutaraldehyde, then trimmed and embedded in paraffin wax according to standard laboratory procedures. Transverse intestinal sections of 4 μm thickness were stained with haematoxylin and eosin and examined by light microscopy. All tissue sections were examined for congestion, oedema and infiltration of mononucleated cells and of heterophils in the mucosa and submucosa and for the height of intestinal villi with adherent bacteria. Eight microscopic fields were scored in a blinded manner by an experienced histopathologist. The protocol was approved by the Ethical Committee of the University of Liège (21 November 2006, N 563).

RESULTS

ospG homologues are present in Enterobacteriaceae species with different T3SSs

OspG was originally identified in Shigella spp. The ospG gene resides on the large virulence plasmids of Shigella spp. and codes for a T3SS effector protein that negatively controls the host innate immune response (Kim et al., 2005). We identified ospG homologues in the draft genomic sequences of EHEC strain 11128 of serogroup O111 and EHEC strain 12009 of serogroup O103. While the ospG homologue of strain 11128 of serogroup O111 encodes a full-length protein, that of strain 12009 is likely a pseudogene with a 9 bp deletion at the 5′ end of the coding sequence. The presence of ospG homologues in EHEC strains suggested a wide distribution of ospG homologues in pathogenic bacteria. By searching in the DNA databases, we found additional ospG homologues in the genome of Y. enterocolitica strain 8081 and the plasmid of EIEC strain 53638 (Table 2⇓). The deduced amino acid sequences of OspG homologues from E. coli, Shigella and Yersinia strains are highly conserved (>90 % identity to each other) and the three catalytic motifs involved in the autophosphorylation activity of OspG from Shigella (Kim et al., 2005) are conserved (Fig. 1a⇓). These OspGs are distantly related (about 30 % identity) to another T3SS effector family (Fig. 1b⇓), which comprises NleH of Citrobacter rodentium (Garcia-Angulo et al., 2008) and NleH1-1 and/or NleH1-2 of EHEC O157 strain Sakai, EHEC O111, rabbit EPEC strain E22, and human EPEC strain B171-8 (Loukiadis et al., 2008; Tobe et al., 2006).

Figure image not available in archive

Fig. 1.

In silico analysis of OspG effector gene distribution among EHEC O111, Shigella and Yersinia. (a) Alignment of deduced amino acid sequence of EHEC O111, Shigella and Yersinia OspGs. Asterisks and dots indicate identical and similar amino acids, respectively. Three conserved catalytic motifs are indicated by grey boxes. The residue K53 (shown in red) of Shigella OspG is involved in autophosphorylation activity of OspG (Kim et al., 2005). (b) Phylogenetic analysis of ospG homologues. The following ospG homologues were analysed: ospG genes in Y. enterocolitica strain 8081 (accession no. AM286415), S. flexneri plasmid pWR100 (AL391753), EIEC strain 53638 plasmid p53638_226 (CP001064), EHEC O103 strain 12009 and EHEC O111 strain 11128 (both sequenced in this study); nleH1 gene in Citrobacter rodentium (AY559032); nleH1-1 genes in EHEC O111 strain 11128 (AB303061), EHEC O103 strain 12009 (AB303062), EHEC O157 Sakai (BA000007, ECs 0804), EPEC O111 strain B171 (AAJX01000049) and rabbit EPEC O103 strain E22 (AAJV01000058); nleH1-2 genes in EHEC O157 Sakai (BA000007, ECs 1814). The ABC1 gene of Clostridium acetobutylicum strain ATCC 824 (NC_003030), which shares a serine/threonine protein kinase catalytic domain, was used as the outgroup. (c) Organization of OspG-encoding region in EHEC O111, Y. enterocolitica and S. flexneri. The sequences of Y. enterocolitica strain 8081 (AM286415) and S. flexneri plasmid pWR100 (AL391753) were used for comparison. The regions indicated by orange shading showed over 80 % nucleotide-sequence identity among the three strains.

Table 2.

ospG loci in some ospG-positive strains

The ospG genes have likely spread in Enterobacteriaceae by horizontal gene transfers because the ospG genes of Shigella spp. reside on plasmids, those of EHEC on lambdoid prophages, and that of Yersinia on a genomic island. The EHEC prophages carrying ospG genes are both integrated in the ompW gene. Like other non-LEE-encoded effector genes, the ospG genes are localized in the EEL of the prophages (Fig. 1c⇑). The ospG gene of Y. enterocolitica strain 8081 resides on a genomic island specific to Y. enterocolitica. Interestingly EHEC, Shigella and Yersinia ospG genes are always associated with IS elements, suggesting that IS-mediated mechanisms have been involved in the acquisition of the ospG genes by each mobile genetic element. The 5′ non-coding regions show >80 % nucleotide-sequence identity (68 bp sequence between EHEC, Shigella and Yersinia strains, and 130 bp sequence between EHEC and Yersinia strains) and the 3′ non-coding regions also show 80 % identity (240 bp sequences between EHEC and Yersinia strains) (Fig. 1c⇑).

EHEC, Shigella and Yersinia OspGs are recognized and injected into mammalian cells by the EPEC T3SS

To test the T3SS-dependent secretion and translocation of EHEC, Shigella and Yersinia OspGs, each protein was cloned and fused to the reporter TEM (β-lactamase) enzyme. The plasmids encoding OspG-TEM fusions were introduced into rabbit EPEC strain E22 and its isogenic T3SS mutant (EPECΔescN). Each OspG-TEM fusion protein was detected in the supernatants of the wild-type EPEC by Western blotting analysis, but not in the supernatants of EPECΔescN (Fig. 2a⇓), indicating that OspGs from EHEC, Shigella and Yersinia were recognized by the T3SS encoded by the EPEC LEE. In addition, the translocation of these OspG-TEM fusion proteins into infected HeLa cells was determined by monitoring the intracellular cleavage of a fluorescent β-lactamase substrate (Charpentier & Oswald, 2004). In this experiment, Tir and Cif were used as EPEC LEE-encoded and non-LEE-encoded effector controls, respectively. As expected, the TEM protein (negative control) was not translocated into infected HeLa cells whereas Tir-TEM and Cif-TEM translocation were readily detected (Fig. 2b⇓). Each OspG effector protein fused to TEM was efficiently translocated into infected HeLa cells, as cleavage of the substrate was detectable 30 min after the onset of infection, reaching a plateau after 1.5 h (Fig. 2b⇓ and data not shown). The percentage of translocation of OspGs from EHEC and Yersinia was ∼75 % compared to the Tir-TEM translocation level, whereas the translocation of Shigella OspG was significantly lower (∼60 %) (Fig. 2b⇓). These results indicated that OspGs from EHEC, Shigella and Yersinia can be secreted and translocated into host cells by the LEE-encoded T3SS.

Figure image not available in archive

Fig. 2.

OspG-TEM fusion proteins derived from EHEC O111, Shigella and Yersinia expressed in EPEC are efficiently secreted and translocated into cultured cells. (a) Plasmid-encoded OspGs-TEM fusion proteins were expressed in rabbit EPEC strain E22 (wt) or in the escN mutant. Bacterial lysates and supernatants were analysed by Western blotting with anti-TEM antibody. OspGTEM, shiospGTEM and yerospGTEM indicate OspGs-TEM fusion proteins derived from EHEC O111, S. flexneri, and Y. enterocolitica, respectively. (b) HeLa cells were infected for 3 h (m.o.i. 100) with E22 hosting the indicated plasmids. Plasmids encoding Tir-TEM or Cif-TEM fusion proteins were used as controls. The translocation levels are shown relative to that of Tir-TEM. The bars show the standard error of the mean (n=3). Asterisk indicates P<0.05.

Intact ospG genes are mainly present in EHEC of serogroup O111

By testing a collection of 178 EPEC and EHEC strains of various serogroups with a primer set that amplified an internal ospG region, we found 39 ospG-positive strains (21.9 %) (Table 3⇓). We further analysed these ospG-positive strains by PCR using two sets of primers: one amplifying the full-length gene and the other amplifying the pseudogene containing a 9 bp deletion at the 5′ end of the coding sequence, which was observed in the ospG gene of EHEC O103 strain 12009. This analysis revealed that about half the ospG-positive strains contain ospG pseudogenes. Among the strains bearing a full-length ospG gene, 84.2 % (16/19) were of serogroup O111. Most of the O111 strains were EHEC, as they were stx-positive. In contrast, ospG pseudogenes were mainly distributed in the O103 serotype (65 %, 13/20). Thus full-length ospG genes were mainly present in EHEC strains of serogroup O111.

Table 3.

Distribution of ospG gene in an EPEC/EHEC collection

OspG is translocated by EHEC O111 but is not required for attaching and effacing (A/E) lesion formation

We aimed at investigating the role of OspG in EHEC infection. Despite many attempts we were not successful in constructing mutants in the sequenced EHEC O111 strain 11128. Nonetheless we could generate T3SS (ΔescN) and OspG (ΔospG) mutants in EHEC O111 strain 13369 (Ogura et al., 2007). We confirmed the expression of the chromosomal ospG gene by RT-PCR (Fig. 3a⇓). We next tested whether the EHEC OspG-TEM fusion protein was secreted in EHEC 13369 culture supernatants and translocated into infected host cells. The OspG-TEM was secreted in a T3SS-dependent manner, but the amount of secreted protein was lower compared to that observed in EPEC E22 (Fig. 3b⇓ and data not shown). In infected HeLa cells, translocation of the OspG-TEM was detectable at 3 h after infection, but a 6 h infection was required to reach the full translocation level (Fig. 3c⇓). These results indicated that OspG is a bona fide T3SS effector protein in EHEC. We next tested whether OspG is required for EHEC to form the A/E lesion in infected host cells. RK13 cells expressing the actin-GFP fusion protein were infected with EHEC strains. After a 6 h infection, F-actin rearrangements beneath adherent bacteria were observed with all strains except the ΔescN mutant, but no qualitative difference was observed among the wild-type, the ΔospG mutant and the ΔospG mutant complemented by a plasmid-encoded ospG (not shown). Thus OspG was not required for formation of the A/E lesion.

Figure image not available in archive

Fig. 3.

Expression, secretion and translocation of OspG by EHEC O111. (a) Transcription of ospG in EHEC O111 strain 13369 was assayed by RT-PCR. In lane ‘noRT’ the RT step was omitted and in lane ‘NC’ (negative control) the template was omitted. 16S rRNA was used as a control. (b) Plasmid encoded OspG (from EHEC O111)-TEM fusion protein was expressed in EHEC O111 strain 13369 (wt) or in the escN mutant. Bacterial lysates and supernatants were analysed by Western blotting with anti-TEM antibodies. (c) HeLa cells were infected for 3 or 6 h (m.o.i. 50) with EHEC O111 strain 13369 (wt) or the escN mutant hosting plasmids encoding TEM or OspG-TEM. The translocation levels are shown relative to the maximum OspG-TEM translocation level. Bars show the standard error of the mean (n=3).

Transfer of NF-κB to the nucleus during infection with EHEC O111 and its mutants

It has been reported that Shigella OspG binds to host cell ubiquitin E2, prevents the ubiquitinylation and degradation of phosphorylated IκB-α and subsequently reduces the activation of a reporter NF-κB-responding promoter (Kim et al., 2005). The three catalytic motifs involved in the autophosphorylation activity of OspG are conserved in the deduced amino acid sequence of all the OspGs from E. coli, Shigella and Yersinia (Fig. 1a⇑), suggesting that EHEC O111 OspG could have the same function. We infected HeLa cells for 1–6 h with EHEC strain 13369, the ospG and escN mutants, or the ospG mutant complemented with plasmid-encoded ospG, and then treated or not the cells with TNF-α for 30 min. Nuclear extracts were prepared from the cells and the p65 subunit of NF-κB was quantified by ELISA. Without TNF-α stimulation, we observed no difference in the levels of nuclear p65 between infected and non-infected cells (data not shown). Following a 3 h infection with the wild-type strain and TNF-α stimulation, the level of nuclear p65 was similar to that in non-infected cells without TNF-α treatment (Fig. 4a⇓). Similar reduced levels of nuclear p65 were observed in cells infected with the ospG mutant and the ospG mutant complemented with ospG from EHEC O111 or from S. flexneri (Fig. 4a⇓ and data not shown). In contrast, the cells infected with the escN mutant were still able to respond to TNF-α stimulation, indicating that the inhibition of p65 nuclear translocation required a functional T3SS but not the OspG effector.

Figure image not available in archive

Fig. 4.

Infection with EHEC O111 prevents TNF-α-induced NF-κB p65 nuclear translocation and IκB-α degradation. (a) HeLa cells were infected for 1–6 h as indicated with EHEC O111 strain 13369 (wt), the ospG mutant, the ospG mutant hosting the plasmid pABBospG, or the escN mutant. The results are shown relative to the translocation level in non-infected (NI) TNF-α-treated cells. Bars indicate the standard error of the mean (n=3). (b) IκB-α degradation following EHEC O111 infection and TNF-α stimulation. HeLa cells were infected for 1, 2, 3 and 6 h and stimulated with TNF-α for 30 min. NI, not infected.

To confirm these results, we monitored the degradation of IκB-α in the cytoplasmic fractions of HeLa cells infected with the wild-type EHEC strain 13369 and the escN and ospG mutants. A 1 h infection with either strain followed by TNF-α stimulation induced degradation of IκB-α, whereas at 2–6 h after infection with the wild-type EHEC, the ospG mutant or the ospG mutant hosting the plasmid-encoded ospG, IκB-α remained detectable following TNF-α stimulation, indicating that IκB-α degradation was inhibited. In contrast, the escN mutant failed to inhibit IκB-α degradation induced by the stimulation with TNF-α (Fig. 4b⇑). Taken together, these results indicate that unknown T3SS effector(s) was/were injected into host cells and inhibited the NF-κB signalling pathway.

Wild-type EHEC O111 and the ospG mutant induce similar mild inflammatory responses in vivo

We finally examined whether EHEC OspG could modulate the inflammatory response in vivo. We performed a rabbit ligated intestinal loop assay as detailed in Methods. In loops inoculated with the wild-type EHEC O111, the ospG mutant and the complemented ospG mutant, we observed similar mild lesions of congestion, oedema of the submucosa and infiltration of lymphocytes and/or granulocytes in all layers of the wall (especially in the mucosa and submucosa) compared to loops injected with sterile culture medium, or with the escN mutant or a control non-EPEC/EHEC strain (Fig. 5⇓). We did not observe differences in the inflammatory response between the loops infected with the wild-type strain, the ospG mutant, and the ospG mutant expressing plasmid-encoded ospG. We also examined in this assay the rabbit EPEC strain E22 (which has no ospG) hosting plasmid pABBospG or the plasmid vector, but we did not observe any effect on the inflammatory response (data not shown). These results observed in the rabbit ligated intestinal loop model suggest that EHEC OspG by itself does not modulate the inflammatory response in vivo.

Figure image not available in archive

Fig. 5.

Wild-type EHEC O111 and ospG mutant induce a similar mild inflammatory response in rabbit ileal loops. Rabbit ileal loops were inoculated with (a) non-EPEC/EHEC strain T282 (negative control), (b) EHEC strain 13369, (c) the ospG mutant and (d) the ospG mutant hosting the pABBospG plasmid. Tissue samples were examined by light microscopy at magnifications ×400 and ×1000 (insets). The arrowheads indicate adherent bacteria and effaced microvilli.

DISCUSSION

In this study we observed that OspG is a T3SS effector protein distributed in Shigella, Yersinia and E. coli. In E. coli, OspG appeared to be distributed mainly in EHEC of serogroup O111. To our knowledge it is the first T3SS effector that has been found to be shared by Shigella, Yersinia and EHEC. The ospG genes are carried by a lambda-like prophage in EHEC O111, on virulence plasmids in Shigella and on a putative pathogenic island in Yersinia, suggesting that different types of horizontal gene transfer contributed to the spread of the ospG genes in different Enterobacteriaceae species.

OspG was expressed by EHEC O111, and OspGs from EHEC, Yersinia and Shigella were injected into host cells through the EPEC or EHEC T3SS. It has been reported that Shigella OspG interferes with the host innate immune responses by binding to ubiquitin-conjugating enzymes. We observed, in HeLa cells infected with EHEC O111, an inhibition of the TNF-α-induced NF-κB subunit p65 nuclear translocation, associated with an inhibition of IκB degradation. This phenotype was clearly T3SS-dependent and similar to that observed in EPEC-infected Caco-2 cells (Ruchaud-Sparagano et al., 2007). However, OspG was not required for EHEC O111 to block the activation of the NF-κB pathway in infected HeLa cells stimulated with TNF-α. In addition, no difference in the inflammatory response was observed between wild-type EHEC strain and the ospG mutant in the rabbit ileal loop model. These observations strongly suggest that other unknown EHEC O111 T3SS effector(s) inhibited the NF-κB signalling pathway by suppressing IκB phosphorylation and/or by acting upstream of IκB degradation. Still, we can not exclude the possibility that EHEC OspG has a similar activity that would have been masked by the other putative effector(s). It would be interesting to test a Shigella ospG mutant complemented with the allele of Shigella, Yersinia and EHEC O111.

Tobe et al. (2006) described recently the extensive repertoire of T3SS effectors in EHEC O157: 39 of these effectors were confirmed experimentally, including two effectors distantly related to OspG, namely NleH1-1 and NleH1-2. The C-terminal part of NleH is 46 % identical to that of the Shigella OspG, which contains the three catalytic motifs involved in the OspG function. EHEC O111 also contains a nleH1-1 homologue (Loukiadis et al., 2008). However, Ruchaud-Sparagano et al. (2007) observed that IL-8 secretion by Caco-2 cells was still suppressed when the cells were infected with a CesT-defective mutant that is unable to produce the CesT protein, a chaperone required for NleH translocation. This suggests that NleH either is not involved, or is not the sole T3SS effector involved in the modulation of the host inflammatory response in vitro. More systematic analyses based on the complete genome sequence will be required to try to identify the gene(s) encoding the effector protein(s) that could be involved in the inactivation of the host innate system during infection with EHEC O111.

On the other hand, recent in vivo experiments indicated that NleHs from Citrobacter rodentium and EHEC have a role in the colonization of the gut (Garcia-Angulo et al., 2008; Hemrajani et al., 2008). In contrast with the in vitro results, NleH from C. rodentium causes an increase in NF-κB activity in the colonic mucosa of NF-κB reporter mice (Hemrajani et al., 2008). The S. flexneri effector OspG negatively controls the host innate response induced by S. flexneri upon invasion of the epithelium (Kim et al., 2005). However, the physiopathology of EHEC and Shigella infections is different. In contrast to Shigella and Yersinia, which invade and cause inflammatory destruction of the intestinal epithelium (Sansonetti, 2004), EHEC and EPEC colonize and adhere to the epithelium (Kaper et al., 2004). Thus, it may be possible that the OspG effector protein has a role in the colonization of the gut. Further studies will aim at clarifying the role of OspG effector homologues acquired by horizontal transfer in different types of pathogens.

Acknowledgments

We thank Drs Yoshitoshi Ogura and Tadasuke Ooka for their help in the sequence analysis of the EHEC ospG-encoding regions, Drs Didier Marlier and Ioan Mihai Szalo, Mr Anastasio Poulipoulis of the Clinics for Birds, Rodents and Rabbits, Miss Sabrina Labrozzo, Mr Jean-Noel Duprez for their assistance during the rabbit experiments. Rika Nobe was supported by an INRA in-house fellowship and by the French Ministry of Agriculture. This study was supported by the European project EADGENE CT-2004-506416, and a grant-in-aid for Scientific Research on Priority Areas ‘Applied Genomics’.

References

Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. (1990). Basic local alignment search tool. J Mol Biol 215, 403–410.
Berin, M. C., Darfeuille-Michaud, A., Egan, L. J., Miyamoto, Y. & Kagnoff, M. F. (2002). Role of EHEC O157 : H7 virulence factors in the activation of intestinal epithelial cell NF-κB and MAP kinase pathways and the upregulated expression of interleukin 8. Cell Microbiol 4, 635–648.
Bhavsar, A. P., Guttman, J. A. & Finlay, B. B. (2007). Manipulation of host-cell pathways by bacterial pathogens. Nature 449, 827–834.
Buchrieser, C., Glaser, P., Rusniok, C., Nedjari, H., d'Hauteville, H., Kunst, F., Sansonetti, P. & Parsot, C. (2000). The virulence plasmid pWR100 and the repertoire of proteins secreted by the type III secretion apparatus of Shigella flexneri. Mol Microbiol 38, 760–771.
Charpentier, X. & Oswald, E. (2004). Identification of the secretion and translocation domain of the enteropathogenic and enterohemorrhagic Escherichia coli effector Cif, using TEM-1 β-lactamase as a new fluorescence-based reporter. J Bacteriol 186, 5486–5495.
Cherepanov, P. P. & Wackernagel, W. (1995). Gene disruption in Escherichia coli: Tc^R and Km^R cassettes with the option of Flp-catalyzed excision of the antibiotic-resistance determinant. Gene 158, 9–14.
Dahan, S., Busuttil, V., Imbert, V., Peyron, J.-F., Rampal, P. & Czerucka, D. (2002). Enterohemorrhagic Escherichia coli infection induces interleukin-8 production via activation of mitogen-activated protein kinases and the transcription factors NF-κB and AP-1 in T84 cells. Infect Immun 70, 2304–2310.
Datsenko, K. A. & Wanner, B. L. (2000). One-step inactivation of chromosomal genes in Escherichia coli K12 using PCR products. Proc Natl Acad Sci U S A 97, 6640–6645.
Espinosa, A. & Alfano, J. R. (2004). Disabling surveillance: bacterial type III secretion system effectors that suppress innate immunity. Cell Microbiol 6, 1027–1040.
Garcia-Angulo, V. A., Deng, W., Thomas, N. A., Finlay, B. B. & Puente, J. L. (2008). Regulation of expression and secretion of NleH, a new non-LEE-encoded effector in Citrobacter rodentium. J Bacteriol 190, 2388–2399.
Garmendia, J., Frankel, G. & Crepin, V. F. (2005). Enteropathogenic and enterohemorrhagic Escherichia coli infections: translocation, translocation, translocation. Infect Immun 73, 2573–2585.
Gobert, A. P., Wilson, K. T. & Martin, C. (2005). Cellular responses to attaching and effacing bacteria: activation and implication of the innate immune system. Arch Immunol Ther Exp (Warsz) 53, 234–244.
Goffaux, F., Mainil, J., Pirson, V., Charlier, G., Pohl, P., Jacquemin, E. & China, B. (1997). Bovine attaching and effacing Escherichia coli possess a pathogenesis island related to the LEE of the human enteropathogenic Escherichia coli strain E2348/69. FEMS Microbiol Lett 154, 415–421.
Hauf, N. & Chakraborty, T. (2003). Suppression of NF-κB activation and proinflammatory cytokine expression by Shiga toxin-producing Escherichia coli. J Immunol 170, 2074–2082.
Hemrajani, C., Marches, O., Wiles, S., Girard, F., Dennis, A., Dziva, F., Best, A., Phillips, A. D., Berger, C. N. & other authors (2008). Role of NleH, a type III secreted effector from attaching and effacing pathogens, in colonization of the bovine, ovine, and murine gut. Infect Immun 76, 4804–4813.
Ho, S. N., Hunt, H. D., Horton, R. M., Pullen, J. K. & Pease, L. R. (1989). Site-directed mutagenesis by overlap extension using the polymerase chain reaction. Gene 77, 51–59.
Jiang, Y., Yang, F., Zhang, X., Yang, J., Chen, L., Yan, Y., Nie, H., Xiong, Z., Wang, J. & other authors (2005). The complete sequence and analysis of the large virulence plasmid pSS of Shigella sonnei. Plasmid 54, 149–159.
Jin, Q., Yuan, Z., Xu, J., Wang, Y., Shen, Y., Lu, W., Wang, J., Liu, H., Yang, J. & other authors (2002). Genome sequence of Shigella flexneri 2a: insights into pathogenicity through comparison with genomes of Escherichia coli K12 and O157. Nucleic Acids Res 30, 4432–4441.
Jubelin, G., Varela Chavez, C., Taieb, F., Banfield, M. J., Samba-Louaka, A., Nobe, R., Nougayrede, J.-P., Zumbihl, R., Givaudan, A. & other authors (2009). Cycle inhibiting factors (CIFs) are a growing family of functional cyclomodulins present in invertebrate and mammal bacterial pathogens. PLoS One 4, e4855
Kaniga, K., Delor, I. & Cornelis, G. R. (1991). A wide-host-range suicide vector for improving reverse genetics in gram-negative bacteria: inactivation of the blaA gene of Yersinia enterocolitica. Gene 109, 137–141.
Kaper, J. B., Nataro, J. P. & Mobley, H. L. T. (2004). Pathogenic Escherichia coli. Nat Rev Microbiol 2, 123–140.
Karmali, M. A. (1989). Infection by verocytotoxin-producing Escherichia coli. Clin Microbiol Rev 2, 15–38.
Kim, D. W., Lenzen, G., Page, A.-L., Legrain, P., Sansonetti, P. J. & Parsot, C. (2005). The Shigella flexneri effector OspG interferes with innate immune responses by targeting ubiquitin-conjugating enzymes. Proc Natl Acad Sci U S A 102, 14046–14051.
Li, Q. & Verma, I. M. (2002). NF-κB regulation in the immune system. Nat Rev Immunol 2, 725–734.
Loukiadis, E., Nobe, R., Herold, S., Tramuta, C., Ogura, Y., Ooka, T., Morabito, S., Kerouredan, M., Brugere, H. & other authors (2008). Distribution, functional expression, and genetic organization of Cif, a phage-encoded type III-secreted effector from enteropathogenic and enterohemorrhagic Escherichia coli. J Bacteriol 190, 275–285.
Marches, O., Ledger, T. N., Boury, M., Ohara, M., Tu, X., Goffaux, F., Mainil, J., Rosenshine, I., Sugai, M. & other authors (2003). Enteropathogenic and enterohemorrhagic Escherichia coli deliver a novel effector called Cif, which blocks cell cycle G₂/M transition. Mol Microbiol 50, 1553–1567.
Menard, R., Sansonetti, P. J. & Parsot, C. (1993). Nonpolar mutagenesis of the ipa genes defines IpaB, IpaC, and IpaD as effectors of Shigella flexneri entry into epithelial cells. J Bacteriol 175, 5899–5906.
Miller, J. H. (1992). A Short Course in Bacterial Genetics. A Laboratory Manual and Handbook for Escherichia coli and Related Bacteria. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Mukherjee, S., Keitany, G., Li, Y., Wang, Y., Ball, H. L., Goldsmith, E. J. & Orth, K. (2006). Yersinia YopJ acetylates and inhibits kinase activation by blocking phosphorylation. Science 312, 1211–1214.
Nougayrede, J.-P., Marches, O., Boury, M., Mainil, J., Charlier, G., Pohl, P., De Rycke, J., Milon, A. & Oswald, E. (1999). The long-term cytoskeletal rearrangement induced by rabbit enteropathogenic Escherichia coli is Esp dependent but intimin independent. Mol Microbiol 31, 19–30.
Ogura, Y., Ooka, T., Asadulghani, Terajima, J., Nougayrede, J.-P., Kurokawa, K., Tashiro, K., Tobe, T., Nakayama, K. & other authors (2007). Extensive genomic diversity and selective conservation of virulence-determinants in enterohemorrhagic Escherichia coli strains of O157 and non-O157 serotypes. Genome Biol 8, R138
Peralta-Ramirez, J., Hernandez, J. M., Manning-Cela, R., Luna-Muñoz, J., Garcia-Tovar, C., Nougayréde, J.-P., Oswald, E. & Navarro-Garcia, F. (2008). EspF interacts with nucleation-promoting factors to recruit junctional proteins into pedestals for pedestal maturation and disruption of paracellular permeability. Infect Immun 76, 3854–3868.
Ruchaud-Sparagano, M.-H., Maresca, M. & Kenny, B. (2007). Enteropathogenic Escherichia coli (EPEC) inactive innate immune responses prior to compromising epithelial barrier function. Cell Microbiol 9, 1909–1921.
Sansonetti, P. J. (2004). War and peace at mucosal surfaces. Nat Rev Immunol 4, 953–964.
Savkovic, S. D., Koutsouris, A. & Hecht, G. (1997). Activation of NF-κB in intestinal epithelial cells by enteropathogenic Escherichia coli. Am J Physiol 273, C1160–C-1167.
Sharma, R., Tesfay, S., Tomson, F. L., Kanteti, R. P., Viswanathan, V. K. & Hecht, G. (2006). Balance of bacterial pro- and anti-inflammatory mediators dictates net effect of enteropathogenic Escherichia coli on intestinal epithelial cells. Am J Physiol Gastrointest Liver Physiol 290, G685–G694.
Thomson, N. R., Howard, S., Wren, B. W., Holden, M. T. G., Crossman, L., Challis, G. L., Churcher, C., Mungall, K., Brooks, K. & other authors (2006). The complete genome sequence and comparative genome analysis of the high pathogenicity Yersinia enterocolitica strain 8081. PLoS Genet 2, e206
Tobe, T., Beatson, S. A., Taniguchi, H., Abe, H., Bailey, C. M., Fivian, A., Younis, R., Matthews, S., Marches, O. & other authors (2006). An extensive repertoire of type III secretion effectors in Escherichia coli O157 and the role of lambdoid phages in their dissemination. Proc Natl Acad Sci U S A 103, 14941–14946.
Venkatesan, M. M., Goldberg, M. B., Rose, D. J., Grotbeck, E. J., Burland, V. & Blattner, F. R. (2001). Complete DNA sequence and analysis of the large virulence plasmid of Shigella flexneri. Infect Immun 69, 3271–3285.
Viswanathan, V. K., Koutsouris, A., Lukic, S., Pilkinton, M., Simonovic, I., Simonovic, M. & Hecht, G. (2004). Comparative analysis of EspF from enteropathogenic and enterohemorrhagic Escherichia coli in alteration of epithelial barrier function. Infect Immun 72, 3218–3227.
Wild, J., Sektas, M., Hradecna, Z. & Szybalski, W. (1998). Targeting and retrofitting pre-existing libraries of transposon insertions with FRT and oriV elements for in vivo generation of large quantities of any genomic fragment. Gene 223, 55–66.
Yang, F., Yang, J., Zhang, X., Chen, L., Jiang, Y., Yan, Y., Tang, X., Wang, J., Xiong, Z. & other authors (2005). Genome dynamics and diversity of Shigella species, the etiologic agents of bacillary dysentery. Nucleic Acids Res 33, 6445–6458.
Zhou, X., Giron, J. A., Torres, A. G., Crawford, J. A., Negrete, E., Vogel, S. N. & Kaper, J. B. (2003). Flagellin of enteropathogenic Escherichia coli stimulates interleukin-8 production in T84 Cells. Infect Immun 71, 2120–2129.