| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
B Activation and Proinflammatory Cytokine Expression by Shiga Toxin-Producing Escherichia coli 1
Institut für Medizinische Mikrobiologie, Justus-Liebig-Universität Giessen, Giessen, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B family of transcription factors forms one of the first lines of defense against infectious disease by inducing the expression of genes involved in inflammatory and immune responses. In this study, we analyzed the impact of Shiga toxin-producing Escherichia coli (STEC) on the NF-B DNA-binding activity in HeLa cells. After a period of weak initial activation, DNA binding of NF-B was actively suppressed by viable, E. coli secreted protein B (EspB)-secreting STEC. Sustained NF-B activity was observed either using an isogenic mutant lacking EspB or after gentamicin-based killing of STEC after allowing bacterial attachment. These observations indicate that the ability of STEC to cause NF-B activation is suppressed by a translocated bacterial effector protein, which is either EspB itself or requires EspB for delivery into the host cell. We found that STEC, enterohemorrhagic E. coli, and enteropathogenic E. coli all interfere with NF-B activation initiated by TNF-
, indicating that suppression of signal-induced NF-B activity is a property common to several attaching and effacing bacteria. As a consequence of NF-B suppression, wild-type STEC induces significantly lower mRNA levels of IL-8, IL-6, and IL-1 upon prolonged infection periods compared with bacteria lacking EspB. For IL-8 and IL-6, the suppressive effect was also reflected at the level of cytokine secretion. Suppression of both basal and signal-induced NF-B DNA binding by attaching and effacing-inducing bacteria appears to be an active strategy to counteract host defense responses, thus favoring intestinal colonization by these pathogens. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The important role of intestinal epithelial cells in activating and regulating the mucosal inflammatory and immune response is well documented (6). Mucosal inflammation is characterized by coordinate expression and up-regulation of a specific array of epithelial gene products, including secreted cytokines with chemoattractant (e.g., IL-8, macrophage-inflammatory protein 1, and monocyte chemoattractant protein-1) or proinflammatory (e.g., TNF- and IL-1) function. For S. dublin, it was clearly shown that most (if not all) genes up-regulated after infection were regulated by NF-B (4).
NF-B comprises a family of closely related transcription factors that play a key role in the expression of genes involved in inflammation and immune responses (7, 8, 9). Usually, the term NF-B is collectively used for homo- and heterodimeric complexes formed by the so-called Rel/NF-B proteins. In mammals, five of such proteins are known, referred to as RelA (p65), RelB, c-Rel, p50 (NF-B1), and p52 (NF-B2). NF-B’s most obvious characteristic is its rapid translocation from the cytoplasm—where it is sequestered in an inactive form, bound to one of several inhibitory proteins (IB, IB
, IB
)—to the nucleus in response to a great variety of extracellular signals. This signal-induced NF-B activation is mainly accomplished by phosphorylation of IB at two specific serine residues followed by polyubiquitination and IB degradation by the 26S proteasome (8, 10). Because degradation exposes the nuclear localization signal on NF-B, the liberated active NF-B translocates to the nucleus, where it modulates gene expression by binding to the B motifs of its target genes.
Shiga toxin-producing Escherichia coli (STEC)3 are an important cause of serious human gastrointestinal disease (11, 12). Central to STEC pathogenesis is the intestinal colonization generally resulting in a striking histopathological feature known as the attaching and effacing (A/E) lesion. After initial adherence to the intestinal mucosa, STEC triggers the localized destruction (effacement) of brush border microvilli and intimately attaches to the epithelial surface. It thereby induces the formation of a pedestal-like actin structure (A/E lesion) directly beneath adherent bacteria, through which STEC remains attached extracellularly to the host cell. A/E lesions are also produced by several other enteric pathogens, including the closely related enteropathogenic E. coli (EPEC), the prototypic A/E organism. The genetic determinant for A/E lesion formation in both STEC and EPEC is a chromosomally encoded pathogenicity island, called the locus of enterocyte effacement (LEE) (13, 14, 15).
Many of the proteins encoded within the LEE locus are part of a type III secretion system comprising proteins required for both the formation of a secretion complex (E. coli secretion and secretion of EPEC proteins) and the secretion of the LEE-located E. coli secreted proteins (Esp proteins). The secretion complex is inserted into the inner and outer membranes of the bacterium and secretes the Esps across the bacterial envelope. It is generally assumed that after initial attachment of EPEC and STEC to the intestinal epithelium, a LEE-encoded type III-secreted protein translocation tube is formed, which connects the pathogen with its target cell (15, 16). EspA seems to be a major component of this tube (17, 18), whereas EspD (19, 20) and EspB (18, 21, 22) appear to be inserted into the host membrane, forming a pore structure. Thus, additional Esp proteins such as EspE/translocated intimin receptor (23, 24) and EspF (25) can be directly injected from the cytoplasm of the bacterium into the host cell to modulate host responses. EspB appears to have a dual function as translocator and effector protein (18, 21, 22, 26, 27, 28).
Previous studies with EPEC have revealed a role for the NF-B family of transcription factors in infected T84 intestinal cells (29, 30). These studies indicate that EspB-dependent activation of NF-B contributes to diarrhea after infection. The underlying mechanism appears to involve NF-B-mediated up-regulation of Galanin-1 receptor expression and increased Cl- secretion in response to galanin binding to the Galanin-1 receptor, thus resulting in enhanced intestinal fluid secretion (30). Because STEC and EPEC appear to share common mechanisms in elaborating diarrhea, we investigated the ability of an O26:H- STEC strain (413/89-1) to induce NF-B in HeLa epithelial cells. Unexpectedly, our studies revealed that STEC and EPEC bacteria actively suppress both basal and signal-induced NF-B DNA-binding activity by a process involving EspB and indicated that EspB is not required for NF-B activation in HeLa cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
E. coli strains used for cell culture infections are shown in Table I. Bacteria were normally grown in Luria-Bertani broth at 37°C with shaking. For infection experiments, bacterial suspension from overnight cultures grown in Luria-Bertani broth were diluted 1/50 in serum-free MEM (Life Technologies, Rockville, MD; STEC and enterohemorrhagic E. coli (EHEC) strains) or 1/40 in serum-free DMEM (Life Technologies; EPEC strain). The cultures were incubated without shaking at 37°C with 5% CO2 to an OD600 of 0.8.
|
A nonpolar in-frame deletion mutation of espB in STEC strain 413/89-1 (
espB) was constructed by allelic exchange. Two fragments containing the 5' and 3' ends of espB and additional flanking regions were generated by two separate PCRs (Expand High Fidelity PCR Kit; Roche, Basel, Switzerland) using chromosomal DNA as template. Fragment B1 was 1069 bp long and was amplified with oligonucleotides B1 (5'-CAG GTC ACT CAT ATG TGA CGC CCT CTG TTG CTG-3') and B2 (5'-AAT TGC ATT GGA TCC GGT AGT ATT CTC CGA AAC-3'), which incorporate NdeI and BamHI restriction sites (underlined), respectively. Fragment B2 was 1011 bp long and was generated using oligonucleotides B3 (5'-GAT ATG ACA GGA TCC GCT CGC GAT CTC ACT GAT-3') and B4 (5'-TTC TAT TAT ACG CGT GGA CCA GAC TGC ATA ACA-3') with incorporated BamHI and MluI restriction sites (underlined), respectively. The corresponding restriction endonucleases were used to enable ligation of both amplification products and for subsequent cloning of the obtained fragment into the temperature-sensitive vector pMAK700oriT (36). The ligation of fragments B1 and B2 at the BamHI site created an in-frame deletion in the espB gene, such that 816 bp were removed from the open reading frame, fusing the 5' region of espB encoding the first 22 N-terminal amino acid residues with the 3' region encoding the last 20 C-terminal amino acid residues of EspB. The resulting recombinant suicide plasmid was introduced by electroporation into the wild-type strain and used to construct the deletion mutant by allelic exchange as described previously (17). Bacteria harboring a mutated espB allele were identified by PCR using primers B1and B4 and were confirmed by sequencing.
For complementation of the STEC 413/89-1 espB mutant, the complete espB gene of STEC strain 413/89-1 was amplified by PCR using oligonucleotides D3 (5'-TCT CAG TTA GGA TCC GAC TCA GCA CGA GTA AAT-3') and eaeB413 (5'-CAG AAT TCT TAC CCA GCT AAG CGA ACC G-3') with incorporated BamHI and EcoRI restriction sites (underlined), respectively. The resultant 1032-bp fragment was subsequently cloned under the control of an arabinose-inducible promoter into the vector pBAD/Myc-HisB (Invitrogen, San Diego, CA), generating plasmid pBAD::espB. The plasmid was transformed into the deletion mutant STEC 413/89-1 espB using electroporation, thereby creating strain STEC 413/89-1 espB + pBAD::espB. DNA sequencing, Esp protein secretion profiles, and Western blot analysis confirmed the generation of the required strain.
The deletion strain STEC 413/89-1 espE/tir was generated analogous to the espB strain using the primer combinations E1/E2 (5'-GAC TAA CCA GCT AGC TCA GGC CGT GGC CAA ACA-3'/5'-AGG TAA TGG CGG CCG AGG TGG AAT TAA AGC TCT-3') and E3/E4 (5'-GCC GCA CCA CGG CCG GGA CCC GCA CGT TTC GTT-3'/5'-CCA GCC TTC ATC GAT TCG CTT TCG GAA CTG TAT) with incorporated NheI, XmaIII, and ClaI restriction sites (underlined), respectively. The resulting in-frame deletion in the espE/tir gene eliminated 1536 bp from the open reading frame by fusing the 5' region of espE/tir (encoding the first 19 N-terminal amino acid residues) with its 3' region (encoding the last seven C-terminal amino acid residues).
Cell lines and culture conditions
Epitheloid human cervix carcinoma cells (HeLa, ATCC CCL2) were a kind gift from Dr. G. Baljer (Institute for Infectious Diseases and Hygiene, Faculty of Veterinary Science, Justus-Liebig-University, Giessen, Germany). HeLa cells were routinely cultured in MEM with glutamine (Life Technologies), supplemented with 10% FCS (Sigma-Aldrich, St. Louis, MO) and 1% nonessential amino acids (Biochrom, Berlin, Germany) at 37°C and 5% CO2. Before stimulation or infection, cells were seeded 48 h in tissue culture plates (Nunc, Roskilde, Denmark) at 1.8 x 106 cells/10-cm2 plate. Twenty hours before infection or stimulation, the tissue culture medium with 10% FCS was changed to a medium containing 1% FCS.
SDS-PAGE of Esp proteins secreted by STEC
Preparation and analysis of STEC-secreted proteins by SDS-PAGE were performed as described previously (17, 24).
Stimulation or infection of HeLa epithelial cells
For stimulation, the cells were treated with human TNF- (Sigma-Aldrich; 10 ng/ml for 15 min), PMA (Sigma-Aldrich; 50 ng/ml for 30 min), or IL-1 (Sigma-Aldrich; 2 ng/ml for 15 min). For infection, the cells were incubated with E. coli (listed in Table I) to give a multiplicity of infection of 50 bacteria per cell for a period of time ranging from 30 min to 8 h. At prolonged infection periods (2–8 h), washing steps with HBSS (Biochrom) were included to remove nonadherent bacteria. For the preinfection experiments, cells were first infected with pathogenic E. coli for different time periods and subsequently were stimulated with 10 ng/ml TNF- for 15 min.
Gentamicin assay
After infection of HeLa cells with STEC (413/89-1) for 1.5 h as described above, STEC-infected monolayers were washed three times with HBSS and exposed to gentamicin (100 µg/ml) for the residual infection periods.
EMSA
Nuclear extracts from cells treated as indicated in the figure legends were prepared according to the method of Schreiber et al. (37) and analyzed by EMSA essentially as described previously (38, 39). In brief, the complementary oligonucleotides 5'-AGC TTC AGA GGG GAC TTT CCG AGA GG-3' and 3'-AGT CTC CCC TGA AAG GCT CTC CAG CT-5' (Sigma-Aldrich) were annealed to generate a double-stranded B probe containing the consensus binding site for NF-B dimeric complexes (shown in bold; Ref.40) and overhanging ends, which were 32P-labeled using Klenow enzyme. To separate unincorporated radioactive nucleotides from the radiolabeled DNA fragment, NucTrap probe purification columns (Stratagene, La Jolla, CA) were used according to the instruction manual. DNA binding reaction mixtures (15 µl) contained 3–4 µg of nuclear extract, 5 µl of 3x binding buffer (60 mM HEPES, pH 7.9; 3 mM DTT; 3 mM EDTA; 150 mM KCl; and 12% Ficoll), 1.6 µg of poly(dI-dC) (Roche), and 20,000 cpm of radiolabeled B probe. Unless specifically indicated, the binding reaction also contained 6 ng of a unlabeled dsDNA fragment encompassing a mutated NF-B DNA-binding site (38). This was added to reduce the formation of two nonspecific protein-DNA complexes, which migrate directly below the p50–p50-B complex (compare Fig. 1 with Figs. 2 and 3; indicated by an arrowhead). The generation of both of these nonspecific complexes could not be prevented by addition of the nonspecific competitor poly(dI-dC) alone. After 25 min of incubation on ice, resultant nucleoprotein complexes were resolved from unbound DNA on a native 5% polyacrylamide gel and were detected by autoradiography of the dried gel.
|
|
|
B probe or DNA fragment mutated in five bases of the B consensus motif (38) was included in the binding reaction mixture. The mixture was incubated on ice for 10 min before the radiolabeled B probe was added. For supershift experiments, 2 µg of a specific antiserum were added to the binding reaction mixture before addition of the radiolabeled B probe and preincubated with the nuclear proteins for 20 min on ice. An Ab that binds specifically to a particular DNA-binding protein can either prevent formation of a specific protein-DNA complex or further retard (supershift) its electrophoretic mobility (41). All Abs used for supershift experiments were rabbit polyclonal IgGs supplied from Santa Cruz Biotechnology (Santa Cruz, CA). RNA isolation and RT-PCR analysis
Total RNA of infected HeLa cells was extracted with TRIzol reagent (Invitrogen) as recommended by the manufacturer’s instructions. RNA samples were quantified spectrophotometrically at 260/280 nm and treated with RNase-free DNase I (Ambion, Austin, TX) before reverse transcription to remove contaminating genomic DNA. One microgram of DNA-free total RNA was reverse transcribed into single-stranded cDNA according to protocols and reagents from Clontech Laboratories (Palo Alto, CA; Advantaq RT-for-PCR kit using oligo(dt)18 primer). One microliter of the 20-µl reverse transcription reaction was amplified by PCR in 50 µl of a mixture containing 5 µl of 10x Titanium Taq PCR buffer (Clontech Laboratories), 1 µl of 10 mM dNTP mix (Amersham Pharmacia Biotech, Piscataway, NJ), 0.2 µmol each of the specific primer pairs listed in Table II, and 1 µl of 50x Titanium Taq DNA polymerase (Clontech Laboratories). For all PCRs, the following conditions were used: an initial denaturation step at 94°C for 1 min, two-step cycles of 30 s at 94°C and 1.5 min at 60°C, and a final extension step at 60°C for 3 min. The number of PCR cycles was adjusted as appropriate to maximize the differences between samples (GAPDH, 23 cycles; IL-6, 28 cycles; IL-1, 30 cycles; IL-8, 33 cycles). To control for contamination with genomic DNA, experiments were performed omitting the enzyme during the reverse transcription step and subsequent PCR amplification using the GAPDH primer pair. PCR products (5–10 µl) were analyzed by 1.8% ethidium bromide-agarose gel electrophoresis. The level of amplified products were normalized to constant amounts of GAPDH mRNA. Each RT-PCR experiment was done in duplicate.
|
Cytokine release from HeLa cells infected with STEC was examined at 6 h, 10 h, and 20 h. To prevent bacterial overgrowth of the epithelial cell cultures, the medium was depleted of nonadherent bacteria at 2 and 6 h. At the 6-h time point, adherent extracellular bacteria were killed by the addition of gentamicin (100 µg/ml). IL-1, IL-6, and IL-8 concentrations were determined in cell supernatants using Biotrak human IL-1, IL-6, and IL-8 ELISAkits (Amersham Pharmacia Biotech) following instructions of the vendor. The results presented are based on three independent experiments that were assayed in quadruplicate.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B (RelA-p50 and RelB-p50) in HeLa epithelial cells
To investigate whether STEC (413/89-1) induces NF-B DNA-binding activity as previously shown for EPEC (29), we first analyzed the kinetics of nuclear translocation of NF-B in STEC-infected HeLa cells. Cells were first challenged with bacteria for specific infection periods ranging from 30 min to 6 h. Nuclear extracts were prepared and subjected to EMSA using a radiolabeled consensus oligonucleotide B probe (see Materials and Methods). Noninfected cells served as negative control. Two clearly detectable changes in protein DNA-binding activity were observed throughout the period of infection investigated (indicated by arrows in Fig. 1A). First, formation of the slower-migrating protein-DNA complex was markedly enhanced by STEC within 0.5–1.5 h after infection. At later time points (3–6 h), this complex disappeared almost completely. Second, the faster-migrating complex was either not induced or only poorly induced at early infection times (0.5–2 h), but unexpectedly at late infection periods, the amount of this complex decreased to below the baseline level of noninfected cells. As depicted in Fig. 1B, the upper band actually comprises two different complexes (C1a and C1b) migrating with nearly the same mobility.
To confirm the specificity of the detected protein-DNA complexes C1a, C1b, and C2 (Fig. 1B), a competition experiment was conducted (Fig. 2). Upon addition of increasing concentrations of the unlabeled B oligonucleotide, a dose-dependent decrease of all three complexes was observed. As expected, an oligonucleotide containing a mutated NF-B DNA-binding motif did not inhibit complex formation at any concentration tested.
To further characterize the composition of the complexes, nuclear extracts were preincubated with Abs directed against all known Rel/NF-B proteins (RelA, RelB, c-Rel, p50, and p52) and were tested by EMSA (Fig. 3). Neither the anti-p52 nor the anti-cRel antiserum impaired the generation of any of the three complexes. Whenever the anti-p50 antiserum was added, the formation of C2 was blocked with concomitant appearance of a weak supershifted complex (S1; Fig. 3, lanes 3, 7-9, 13, 14, and 16). This indicates that C2 is likely to be formed by binding of p50-p50 homodimers to DNA. Moreover, preincubation with the anti-p50 antiserum resulted in a significant reduction of C1a and C1b, suggesting that p50 is also a component of these two complexes. Addition of the anti-RelA antiserum led to the complete disappearance of C1a and appearance of supershift 2 (S2; Fig. 3, lanes 4, 7, 10, 11, and 13-16). Thus, we concluded that C1a represents RelA–p50 heterodimers bound to the B oligonucleotide. Analysis of the components of the C1b complex was found to be difficult because C1b migrates directly below the more prominent C1a complex (Figs. 1B and 3). Because the band comprising C1a and C1b was more reduced in the presence of anti-RelB in comparison with anti-cRel antiserum (Fig. 3, compare lanes 5 and 6 and lanes 8 and 9), we assume that C1b contains RelB additionally to p50. Collectively, these results suggest that C1a represents RelA-p50 heterodimers bound to DNA, that C1b contains RelB-p50 heterodimers complexed with the B probe, and that C2 is generated by binding of p50-p50 homodimers to the oligonucleotide.
EspB is not responsible for transient activation of NF-B in STEC-infected HeLa cells, but is required for inhibition of persistent NF-B activation
Because the EPEC-induced NF-B activation in intestinal epithelial cells was shown to be dependent on the LEE type III-secreted protein EspB (29), we next examined whether this is also true for STEC. Hence, an EMSA was performed with nuclear extracts from HeLa cells infected for different time points with wild-type STEC (413/89-1) and with an isogenic mutant harboring an in-frame deletion in espB (espB) (Fig. 4A). Surprisingly, the espB mutant induced NF-B DNA-binding activity with the same kinetics as the wild type in early infection periods (0.5–1 h). However, at prolonged infection times (2–6 h), espB-infected cells revealed sustained increases of RelA–p50- and RelB–p50-DNA complexes as well as of p50–p50-B complexes. The espB-induced DNA-binding activity of RelA-p50 and RelB-p50 was even higher at later times compared with early times (up to 1 h). Moreover, p50-p50 DNA-binding activity did not decrease as it did with the wild-type strain, but was slightly enhanced within 2–6 h after infection. A STEC mutant strain espE/tir, which did not induce A/E lesions because of its inability to deliver the translocated intimin receptor EspE/translocated intimin receptor into host cells (Ref.23 and unpublished data), behaves essentially like wild-type STEC with regard to transient NF-B activation (Fig. 4B). This result clearly indicates that EspB from STEC actively inhibits sustained activation of these NF-B dimers.
|
B DNA-binding activity is mediated by viable, metabolically active STEC bacteria
Given that the STEC-mediated NF-B suppression is caused by a translocated bacterial effector protein, this effect would require initial bacterial attachment to the host cell, followed by formation of the protein translocation apparatus and subsequent injection of the effector into the host cell. As is shown in Fig. 1A, at 1.5 h postinfection, enhanced NF-B DNA-binding activity was seen, suggesting that the postulated bacterial effector exerting the suppressive effect on NF-B activation either has not yet been translocated or has only inefficiently been translocated into the host cell. Moreover, this time point was critical for initial attachment of STEC (413/89-1) because after washing the cells, adherent bacteria were still present. Therefore, the 1.5-h postinfection time point was chosen for the following experiment. HeLa cells were infected with live STEC that either were killed by gentamicin treatment at 1.5 h or were left untreated as control. At later infection times (4–6 h), gentamicin-treated cells showed considerably higher NF-B DNA-binding activity than did untreated cells (Fig. 5). This result supports the idea that STEC actively causes NF-B suppression by a translocated effector protein.
|
B signaling initiated by TNF-
To examine whether STEC and other A/E-inducing bacteria are able to interfere with NF-B activation caused by a well-established NF-B inducer such as TNF- (7), the following set of experiments was performed. HeLa cells were preinfected with STEC (413/89-1), EHEC (EDL933), or EPEC (E2348/69) for different times and subsequently stimulated with TNF- for 15 min. Nuclear proteins from noninfected or infected cells, which were either stimulated with TNF- or left unstimulated, were extracted and analyzed by EMSA (Fig. 6A). Preinfection of HeLa cells with each of the three strains used resulted in impairment of NF-B activation triggered by TNF-. Interestingly, the block in TNF--induced NF-B activation correlates with the level of bacterial attachment to their target cells. Thus, in the case of EPEC (E2348/69)—harboring a bundle-forming pilus that allows rapid attachment to HeLa cells (15)—complete inhibition of TNF--mediated NF-B activation was already seen 4 h postinfection (Fig. 6A; EPEC, lane 8), whereas this effect was delayed in the more slowly adherent STEC and EHEC strains (STEC > EHEC). These results are consistent with the hypothesis that NF-B suppression by STEC, EHEC, and EPEC is mediated by a translocated bacterial effector molecule, which then interferes at some level with NF-B signaling pathways initiated by TNF-.
|
B suppression, we complemented the espB deletion strain with the pBAD::espB plasmid, which expresses the espB gene from STEC strain 413/89-1 in an arabinose-inducible fashion (Fig. 6C). The experiment was repeated as above using the espB mutant and the complemented strain. As expected, preinfection with the espB mutant did not result in impairment of TNF--induced NF-B activation. In contrast, in cells preinfected with the complemented espB mutant (in the presence of 0.2% arabinose), the inhibitory effect on TNF--induced NF-B activation was even more rapid than that observed with the wild-type STEC strain (Fig. 6, A and B).
Fig. 7 shows that the three virulent wild-type E. coli strains tested induced similar levels of NF-B DNA-binding activity. However, induction levels were substantially less than those caused by the well-known NF-B inducers, IL-1, PMA, and TNF-.
|
NF-B is an important transcriptional regulator of inducible expression of numerous genes involved in inflammatory and immune responses (7). To determine the effect of STEC-mediated NF-B suppression on NF-B-dependent gene expression, we first investigated the mRNA levels of IL-8, IL-6, and IL-1 in response to bacterial infection. The changes in cytokine expression of cells infected with wild-type STEC and the espB mutant were analyzed by RT-PCR. For each experiment, mRNA level was normalized to GAPDH expression, which is unaffected by STEC infection. Using IL-8 and IL-6 gene expression as a readout, a time course of infection was performed in which HeLa cells were infected with wild-type STEC and espB for 3, 4, and 7 h. As shown in Fig. 8A, the espB mutant induced a significantly higher amount of IL-8 mRNA than did the wild type at all time points investigated. In addition, this mutant also exhibited increased IL-6 mRNA expression at 4 and 7 h compared with the wild type (Fig. 8A). However, at 3 h postinfection, both strains induced similar levels of IL-6 mRNA. For subsequent experiments addressing the expression level of the IL-1 gene, cells were only infected for 6 h, when NF-B suppression by wild-type STEC was maximal. At this specific time point, the amount of IL-1 mRNA in wild-type STEC-infected cells is comparable to that expressed in noninfected cells, whereas espB infection clearly induced IL-1 mRNA synthesis (Fig. 8B). Taken together, transcription of different NF-B-regulated inflammatory cytokine genes was substantially reduced upon prolonged infection periods with wild-type STEC compared with the espB mutant.
|
B activation (Fig. 6, A and B). Under the experimental conditions described, at 6 h no significant differences were observed for both IL-8 and IL-6 secretion (Table III). However, 20 h after infection with wild-type STEC, a 10-fold reduction of IL-8 release and a 6-fold inhibition of IL-6 secretion was detected (Table III). IL-1 secretion by noninfected or infected HeLa cells was below the threshold sensitivity (10 ng/ml) of the assay used and thus could not be quantified at any time point investigated. These results clearly show that wild-type STEC inhibits production of IL-8 and IL-6 by HeLa cells in an EspB-dependent manner.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-induced NF-B DNA-binding activity in HeLa cells after an initial induction period. Evidence for suppression of NF-B activation by these bacteria was derived from the following observations. First, wild-type STEC bacteria only transiently induce DNA-binding activity of RelA-p50 and RelB-p50, whereas DNA binding of p50 homodimers was reduced to even below the baseline level of noninfected cells after prolonged infection periods (3–6 h). Second, the STEC espB mutant revealed persistent activation of RelA-p50 and RelB-p50 dimers with slight elevation of DNA binding of p50 homodimers at late infection times (2–6 h). This effect (persistent NF-B activation) was specific, because it was not observed using an isogenic espE/tir mutant strain that behaved essentially like the wild-type strain. Third, gentamicin-based killing of wild-type STEC at 1.5 h postinfection abrogated NF-B suppression, suggesting that this effect was mediated by a translocated bacterial effector protein. Fourth, prior exposure of HeLa cells to STEC, EHEC, EPEC, or the complemented STEC espB mutant interfered with subsequent NF-B activation by TNF-, a well-known NF-B inducer. Collectively, these data indicate that EspB is required for suppression of both basal or signal-induced NF-B DNA-binding activity. In addition, this suppressive effect is a common property of several A/E-inducing bacteria.
These results are reminiscent of suppression of RelA-p50 DNA-binding activity seen with Y. pseudotuberculosis and Y. enterocolitica. In that case, suppression is due to the intracytoplasmic presence of a secreted prokaryotic effector protein (YopJ or YopP, respectively), which is injected into the target cell via a type III secretion system (42, 43, 44). YopJ was shown to bind directly to the superfamily of mitogen-activated protein kinase kinases, thus inhibiting their kinase activity (44). One member of this superfamily is the IB kinase , which is a component of a large cytoplasmic multisubunit complex phosphorylating IB as well as IB. Because phosphorylation of these IBs is blocked, subsequent IB degradation and NF-B activation is inhibited (42, 43). A search for proteins homologous to the Yersiniae YopJ/P proteins in the sequenced genomes of two serotype O157 EHEC strains (including the EDL933 strain examined in this study) yielded no results, indicating that the effect observed here is mediated by a hitherto unknown effector protein in these strains.
Our data are also consistent with the notion that a translocated bacterial effector protein common to STEC, EHEC, and EPEC mediates suppression of both basal and signal-induced NF-B activity by interfering at some level with NF-B signaling. The delayed suppressive activity seen with these strains is in agreement with a sequence of events requiring initial bacterial attachment to the host cell, followed by formation of the type III protein translocation apparatus with subsequent delivery of the putative effector protein into the host cell. Indeed, the rate and level of adhesion of STEC, EHEC, and EPEC is consistent with the respective appearance of NF-B suppression caused by these bacteria. The observation that the STEC mutant lacking EspB, unlike the complemented STEC espB mutant, is unable to suppress TNF--mediated NF-B activation indicates that EspB is required for this activity. EspB appears to have a dual function, acting both as an effector protein in the cytoplasm of the host cell and as a part of the translocation machinery (18, 21, 22, 26, 27, 28).
In a previous study, Savkovic et al. (29) showed induction of NF-B DNA-binding activity in T84 intestinal epithelial cells infected with wild-type EPEC (E2348/69) using EMSA. In this study, we found a similar time course of NF-B activation using HeLa cells with enhanced DNA-binding activity of NF-B dimers detectable at 1 h postinfection and reduction in DNA binding at 3 h postinfection. In the report by Savkovic et al. (29), it was concluded that the low level of NF-B-DNA complexes observed at 3 h simply reflects the termination of the EPEC-induced NF-B activation. However, using prolonged infection periods (3–6 h), we found that the low level of NF-B DNA-binding activity seen at these later time periods is a result of active suppression of NF-B activation. This suppression is a property common to many bacteria producing A/E lesions, because prior exposure of HeLa cells to STEC, EHEC, or EPEC interfered with subsequent NF-B activation by TNF-, a well-known NF-B inducer. The data obtained indicate a crucial role of EspB in NF-B suppression, rather than in NF-B activation as it was previously reported (29).
The NF-B family of transcription factors form one of the first lines of defense against infectious disease and cellular stress by inducing the transcription of numerous genes involved in inflammatory and immune responses. Thus, we monitored the kinetics of expression and secretion of IL-8, IL-6, and IL-1 after infection of HeLa cells using wild-type STEC and mutant bacteria lacking EspB. The transcription of these ILs is thought to involve regulation by NF-B-binding sites in the promoters of the respective genes (7, 29, 45, 46, 47). The chemokine IL-8 is a potent chemoattractant for polymorphonuclear cells and directs recruitment of these cells into the infected epithelium (6, 48). The proinflammatory cytokines IL-1 and IL-6 play a well-documented role in activation of the mucosal inflammatory response and are induced and secreted by epithelial cells in response to different enteric pathogens such as Salmonella spp. and Shigella spp. (3, 6, 49). However, unlike these pathogens, many A/E bacteria inhibit the release of proinflammatory cytokines by suppressing NF-B activation. Thus, A/E-inducing bacteria, such as STEC, behave like Yersinia spp. and comprise a family of pathogens with strong anti-inflammatory properties.
The results of this study are also in agreement with a recent study (50) examining the ability of a diverse group of STEC clinical isolates to induce polymorphonuclear cell transmigration across and IL-8 secretion from T84 intestinal cells. Of the 10 STEC strains investigated, three strains lacking EspB (espB) and the outer membrane adhesin intimin (eae) significantly induced more neutrophil transmigration and higher IL-8 secretion than did their eae- and espB-positive STEC and EPEC counterparts (50). Our finding that STEC lacking EspB exhibits prolonged NF-B DNA-binding compared with wild-type bacteria and consequently induced increased IL-8 expression and secretion could clearly account for these observations.
In summary, our data indicate that STEC, EHEC, and EPEC actively suppress DNA binding of NF-B dimers. Because NF-B plays an important role in the generation of inflammation and immune responses, it is likely that the bacteria-mediated suppression of NF-B activation represents a strategy of holding host defense mechanisms at bay while a beachhead for colonization is established. Put another way, suppression of NF-B DNA-binding activity appears to have evolved as a pathogenic mechanism permitting bacteria to both escape and counteract the ensuing host inflammatory and immune responses, while remaining extracellular to the infected host.
|
Acknowledgments |
|---|
ner and Xenia Alevra for help with some of the experiments described here. We are also indebted to Till Podszadel and Christina Deibel for the generation of the deletion strains and their complementation. N.H. is grateful to Saeid Bouzari for critical reading of the manuscript as well as for stimulating discussions. |
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Trinad Chakraborty, Institut für Medizinische Mikrobiologie, Justus-Liebig-Universität Giessen, Frankfurter Strasse 107, 35392 Giessen, Germany. E-mail address: trinad.chakraborty{at}mikrobio.med.uni-giessen.de
3 Abbreviations used in this paper: STEC, Shiga toxin-producing Escherichia coli; A/E, attaching and effacing; EPEC, enteropathogenic E. coli; LEE, locus of enterocyte effacement; Esp, E. coli secreted protein; EHEC, enterohemorrhagic E. coli.
Received for publication July 2, 2002. Accepted for publication December 6, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B. Annu. Rev. Cell Biol. 10:405.
B and Rel proteins: evolutionarily conserved mediators of immune responses. Annu. Rev. Immunol. 16:225.[Medline]
B and the innate immune response. Curr. Opin. Immunol. 12:52.[Medline]
B activity. Annu. Rev. Immunol. 18:621.[Medline]
-catenin. Cell. Microbiol. 4:213.[Medline]
B in intestinal epithelial cells by enteropathogenic Escherichia coli. Am. J. Physiol. 273:1160.
B in P388D1 macrophage-like cells. Infect. Immun. 62:2740.B (RelA/p50) activation induced by lipoteichoic acid and bacterial phospholipases and mediated by IB and IB degradation. Proc. Natl. Acad. Sci. USA 94:9394.B: affinity, specificity, Zn2+ dependence, and differential half-site recognition. J. Biol. Chem. 266:252.B: involvement in the induction of programmed cell death and in the suppression of the macrophage tumor necrosis factor production. J. Exp. Med. 187:1069.B activation and cytokine expression: YopJ contains a eukaryotic SH2-like domain that is essential for its repressive activity. Mol. Microbiol. 28:1067.[Medline]
promoter by human T-cell leukemia virus type I and type II Tax proteins. Blood 87:3410.This article has been cited by other articles:
|
|
 |
|
  R. Nobe, J.-P. Nougayrede, F. Taieb, M. Bardiau, D. Cassart, F. Navarro-Garcia, J. Mainil, T. Hayashi, and E. Oswald Enterohaemorrhagic Escherichia coli serogroup O111 inhibits NF-{kappa}B-dependent innate responses in a manner independent of a type III secreted OspG orthologue Microbiology, October 1, 2009; 155(10): 3214 - 3225. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Bellmeyer, C. Cotton, R. Kanteti, A. Koutsouris, V. K. Viswanathan, and G. Hecht Enterohemorrhagic Escherichia coli suppresses inflammatory response to cytokines and its own toxin Am J Physiol Gastrointest Liver Physiol, September 1, 2009; 297(3): G576 - G581. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Vareille, F. Rannou, N. Thelier, A.-L. Glasser, T. de Sablet, C. Martin, and A. P. Gobert Heme Oxygenase-1 Is a Critical Regulator of Nitric Oxide Production in Enterohemorrhagic Escherichia coli-Infected Human Enterocytes J. Immunol., April 15, 2008; 180(8): 5720 - 5726. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. P. Gobert, M. Vareille, A.-L. Glasser, T. Hindre, T. de Sablet, and C. Martin Shiga Toxin Produced by Enterohemorrhagic Escherichia coli Inhibits PI3K/NF-{kappa}B Signaling Pathway in Globotriaosylceramide-3-Negative Human Intestinal Epithelial Cells J. Immunol., June 15, 2007; 178(12): 8168 - 8174. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Roxas, A. Koutsouris, and V. K. Viswanathan Enteropathogenic Escherichia coli-Induced Epidermal Growth Factor Receptor Activation Contributes to Physiological Alterations in Intestinal Epithelial Cells Infect. Immun., May 1, 2007; 75(5): 2316 - 2324. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. L. Gyles Shiga toxin-producing Escherichia coli: An overview J Anim Sci, March 1, 2007; 85(13_suppl): E45 - E62. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. N. Bhattacharjee, K.-S. Park, Y. Kumagai, K. Okada, M. Yamamoto, S. Uematsu, K. Matsui, H. Kumar, T. Kawai, T. Iida, et al. VP1686, a Vibrio Type III Secretion Protein, Induces Toll-like Receptor-independent Apoptosis in Macrophage through NF-{kappa}B Inhibition J. Biol. Chem., December 1, 2006; 281(48): 36897 - 36904. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Sharma, S. Tesfay, F. L. Tomson, R. P. Kanteti, V. K. Viswanathan, and G. Hecht 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, April 1, 2006; 290(4): G685 - G694. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Girard, I. P. Oswald, I. Taranu, P. Helie, G. D. Appleyard, J. Harel, and J. M. Fairbrother Host Immune Status Influences the Development of Attaching and Effacing Lesions in Weaned Pigs Infect. Immun., September 1, 2005; 73(9): 5514 - 5523. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Stekel, D. Sarti, V. Trevino, L. Zhang, M. Salmon, C. D. Buckley, M. Stevens, M. J. Pallen, C. Penn, and F. Falciani Analysis of host response to bacterial infection using error model based gene expression microarray experiments Nucleic Acids Res., March 30, 2005; 33(6): e53 - e53. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Menge, I. Stamm, P. M. van Diemen, P. Sopp, G. Baljer, T. S. Wallis, and M. P. Stevens Phenotypic and functional characterization of intraepithelial lymphocytes in a bovine ligated intestinal loop model of enterohaemorrhagic Escherichia coli infection J. Med. Microbiol., June 1, 2004; 53(6): 573 - 579. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Deng, J. L. Puente, S. Gruenheid, Y. Li, B. A. Vallance, A. Vazquez, J. Barba, J. A. Ibarra, P. O'Donnell, P. Metalnikov, et al. Dissecting virulence: Systematic and functional analyses of a pathogenicity island PNAS, March 9, 2004; 101(10): 3597 - 3602. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. M. Ceponis, D. M. McKay, R. J. Menaker, E. Galindo-Mata, and N. L. Jones Helicobacter pylori Infection Interferes with Epithelial Stat6-Mediated Interleukin-4 Signal Transduction Independent of cagA, cagE, or VacA J. Immunol., August 15, 2003; 171(4): 2035 - 2041. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
