Published online before print November 17, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Inserm, UMR756 "Signalisation et Physiopathologie des Cellules Epithéliales," and Université Paris-Sud XI, Faculté de Pharmacie, Châtenay-Malabry, France;
AP-HP, Hôpital Antoine Béclère, Service de Microbiologie et Immunologie, Clamart, France; and
AP-HP, Hôpital Bichat Claude-Bernard, Unité d’Immunologie "auto-immunité et hypersensibilités," Paris, France
1 Correspondence: UMR756, Faculté de Pharmacie, 5 rue J. B. Clément, 92296 Châtenay-Malabry Cedex, France. E-mail: sylvie.chollet-martin{at}u-psud.fr
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, IL-1β, and IL-8 synthesis. The use of wild-type (IH11128, C1845, LF82), control (AAEC185), and recombinant (AAEC185 bearing Dr or F1845 fimbriae, AdLF82, or type 1 pili) bacterial strains allowed us to demonstrate that late IL-8 hyperproduction is triggered by type 1 pili but not by Dr or F1845 fimbriae; MAPKs (p38, ERK, Src) and NF-
B activations are implicated in this response. Thus, in the course of Afa/Dr DAEC intestinal infection, epithelium- and neutrophil-derived IL-8 could, at least in part, control the flow of neutrophils through the lamina propria. Afa/Dr DAEC-induced IL-8 hyperproduction by PMN might thus be important for inducing and perpetuating local inflammation, and this self-amplifying loop might play a role in the pathogenesis of inflammatory bowel diseases such as Crohn’s disease.
Key Words: cytokines • bacteria • inflammation • protein kinases
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Diffusely adherent Escherichia coli expressing Afa/Dr fimbriae (Afa/Dr DAEC) [9
] belong to Group 6 of pathogenic E. coli [10
] and can promote PMN transepithelial migration [3
]. In vitro, the Afa/Dr DAEC strains C1845 (bearing F1845 fimbriae) and IH11128 (bearing Dr fimbriae) induce IL-8 synthesis by intestinal T84 cell monolayers via activation of MAPKs, including ERK, p38, and Src. This IL-8 secretion promotes transepithelial migration of PMN in vitro, which in turn, induces epithelial synthesis of TNF- and IL-1β, up-regulation of apical and basolateral decay-accelerating factor (DAF), and increased adhesion of Afa/Dr DAEC [3
, 11
]. These pathogenic E. coli can cause childhood diarrhea [12
, 13
] and are responsible for one-third of recurrent urinary tract infections (UTIs) in adults [14
]. Afa/Dr DAEC express fimbriae encoded by a family of afa/dra/daa-related operons [9
, 15
]; these fimbriae allow the bacteria to bind to epithelial cell membrane-associated receptors, including DAF and some carcinoembryonic antigen-related cell adhesion molecules (CEACAMs) [9
].
Interactions between apical, enterovirulent E. coli colonizing the intestinal brush border and PMN are poorly documented. Brest et al. [16 ] found that Afa/Dr DAEC accelerated apoptosis of PMN isolated from human blood and led to decreased phagocytosis by transmigrated and nontransmigrated PMN, which are the first line of innate immune defenses, ingesting and killing invading microbes through several coordinated phenomena. The latter include release of preformed antimicrobial molecules stored in granules, generation of reactive oxygen species (ROS) by the NADPH oxidase system [17 ], complex ion disequilibrium and charge compensation across the vacuolar wall [18 ], formation of extracellular neutrophil traps [19 ], and release of numerous cytokines facilitating bidirectional and multicompartmental interactions with immune cells and tissues [20 ]. In subjects with inflammatory bowel disease (IBD), this response can be excessive and deleterious [21 ].
Here, we analyzed E. coli-induced PMN responses by using the human PLB-985 cell line differentiated into mature neutrophils and a set of E. coli strains. We found that these bacteria can interact directly with PLB-985 cells, leading to coordinated cell activation and mediator release, including a strong oxidative burst and rapid release of preformed myeloperoxidase (MPO) and IL-8, followed by late synthesis of IL-1β, TNF-, and IL-8. Using wild-type (WT) enterovirulent, control, and recombinant E. coli strains, we showed that late IL-8 hyperproduction was triggered by type 1 pili but not by Dr or F1845 fimbriae, via MAPK (p38, ERK) and Src and NF-B activation.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
For granulocytic differentiation, exponentially growing cells at a starting density of 2 x 105/ml were cultured in RPMI-1640 medium, supplemented with 0.5% N,N-dimethyl formamide (DMF; Carlo Erba, Rodano, Italy), 1% Nutridoma-SP (Boehringer-Mannheim, Basel, Switzerland), and 0.5% FCS as described in ref. [23 ]. The medium was changed once on Day 3 during the 6-day differentiation period.
On Day 6, granulocytic differentiation was checked by morphological analysis of cytocentrifuged cells stained with May-Grünwald-Giemsa on CD11b expression at the cell surface and on superoxide anion (O2–.) production as described below.
Isolation of human blood PMN
Blood samples were obtained from informed, healthy volunteers. As described previously [24
], leukocytes were isolated by sedimentation on a separating medium containing 9% Dextran T500® (Pharmacia, Uppsala, Sweden) and 38% Radioselectan® (Schering, Lys-lez-Lannoy, France). After red-cell sedimentation, the leukocyte-rich suspension was centrifuged on a Ficoll density gradient (Eurobio, Les Ulis, France). Contaminating erythrocytes were then removed by hypotonic lysis, and leukocytes were resuspended in HBSS (Ca2+ Mg2+, Life Technologies, San Diego, CA, USA) or PBS (Biomerieux, Marcy l’Etoile, France), depending on the experiment.
Bacteria and growth conditions
We used two WT Afa/Dr DAEC strains (IH11128 and C1845) [25
, 26
] and two recombinant strains (AAEC185pDr expressing Dr fimbriae and AAEC185pF1845 expressing F1845 fimbiae; see Table 1
) [27
, 28
]. The laboratory strain AAEC185 lacking type 1 pili was used as a nonvirulent control [29
]. Recombinant strain AAEC185pSH2 and mutant strain AAEC185pUT2002 [30
] were generous gifts from Scott J. Hultgren (Washington University School of Medicine, St. Louis, MO, USA). The WT strain LF82 [31
] and the recombinant strain AAEC185pAdLF82 [32
] were a generous gift from Arlette Darfeuille-Michaud (USC INRA 2018, Université d’Auvergne, Clermont-Ferrand, France).
|
View this table: [in a new window] |
Table 1. Bacterial Strains and Recombinant Plasmids
|
LPS extraction
LPS was extracted from strains WT IH11128 (LPS-IH11128), WT C1845 (LPS-C1845), and AAEC185 (LPS-AAEC185) by using the hot phenol-water protocol as recommended by the manufacturer (LPS extraction kit, Intron Biotechnology, Korea). Briefly, strains were subcultured in LB medium for 18 h at 37°C and washed twice in PBS. Bacterial cells were lysed for 1 h at 60°C in a lysis buffer containing 0.42 mg/ml proteinase K. To separate LPS from other cell components, preparations were incubated for 10 min at –20°C in purification buffer. The LPS pellet was then washed with 70% ethanol to remove salt impurities. Endotoxin was quantified with the Limulus amebocyte lysate (LAL) chromogenic assay, as recommended by the manufacturer (LAL QCL-1000®, Cambrex BioSciences).
Immunofluorescence
Differentiated PLB-985 cells plated on glass coverslips (5x106 cells/ml) in 24-well tissue-culture plates were fixed in 3% paraformaldehyde-PBS for 15 min at room temperature. After washing three times with PBS, cells were treated with 50 mM NH4Cl for 10 min to neutralize aldehyde functions, followed by blocking with PBS containing 0.2% gelatin. Cells monolayers were incubated with mouse mAb 8D11, GM8G5, and 9A6 (Genovac, Freiburg, Germany), specific for DAF, CEACAM1, and CEACAM6, respectively, for 1 h at room temperature. After washing, cells were incubated for an additional 45 min with Alexa fluor-conjugated goat anti-mouse antibody (Alexa 488M, Molecular Probes, Interchim, Montluçon, France). Finally, cells were washed three times in PBS, mounted in Dako antifading reagent (DakoCytomation, Trappes, France), and examined by conventional epifluorescence microscopy using a Leitz Aristoplan microscope (Leica, Rueil-Malmaison, France) and by a confocal laser-scanning microscope [LSM 510 equipped with an air-cooled argon ion laser (488 nm) and a helium neon laser (543 nm)], configured with an Axiovert 100 M microscope using a Plan Apochromat 63x/1.40 oil objective (Carl Zeiss S.A.S., Le Pecq, France).
Flow cytometry
Expression of β2 integrin, DAF (CD55), and CEACAM6 by differentiated PLB-985 cells was measured after 30 min of incubation at 4°C with, respectively, PE-conjugated anti-CD11b (DakoCytomation), PE-conjugated anti-CD55 (BD PharMingen, San Jose, CA, USA), and anti-CEACAM6 (Genovac) mouse mAb. After one wash with ice-cold PBS, PE-conjugated goat anti-mouse Ig was added to the latter samples for an additional 30 min at 4°C. After one wash with ice-cold PBS, the cells were resuspended in 1% paraformaldehyde-PBS and kept on ice until flow cytometry. Nonspecific binding was determined by using irrelevant antibodies of the same isotypes. We used a FACSCalibur flow cytometer equipped with a 15-mV, 488-nm argon laser (Becton Dickinson, San Jose, CA, USA). The data were analyzed with CellQuest software. Fluorescence was recorded with a constant photomultiplier gain, and the results were expressed as the mean fluorescence intensity (MFI) on a four-decade logarithmic scale.
Microscopic evaluation of bacteria adhesion to PLB-985-differentiated cells
Bacteria (AAEC185 and the recombinants AAEC185pDr and AAEC185pF1845) were adjusted at the final concentration of 5 x 108/ml in a buffer containing 50 mM sodium carbonate, 100 mM NaCl, and 100 µg/mL FITC (Sigma-Aldrich Chimie SARL, L’Isle d’Abeau Chesnes, France) for 30 min at 37°C, while protected from light. The labeled bacteria were then washed three times, adjusted in HBSS at the concentration of 2 x 108-labeled bacteria/ml, and used immediately.
Differentiated PLB-985 cells were washed in HBSS and adjusted at 2 x 106 cells/ml HBSS. This suspension (500 µl) was put on glass coverslips in 24-well tissue-culture plates and incubated for 1 h at 37°C to ensure PLB-985 adherence. The medium was then removed, and 500 µl freshly prepared FITC-labeled bacteria were added to the adherent cells. Plates were centrifuged slightly at 140 g for 5 min to synchronize infections and incubated at 37°C in a humidified atmosphere containing 7% CO2 during 1 h. Glass coverslips were then washed three times with PBS to remove nonadherent bacteria, fixed with 3% paraformaldhyde, and washed again twice in PBS. DNA-binding Hoechst (Sigma-Aldrich Chimie SARL) was then added (2.5 µg/ml) for 30 min. After washings, the coverslips were mounted in florescent mounting medium (DakoCytomation) and examined by conventional epifluorescence microscopy using a Leitz Aristoplan microscope (Leica). The number of bacteria per cell was determined for each bacteria strain.
Degranulation experiment
Cells (5x106/ml) in HBSS were incubated with the bacterial strains (5x108 CFU/ml) or LPS (1 µg/ml) at 37°C for 20 min in a water bath with gentle agitation. As a positive control to ensure total degranulation, cells were kept at 37°C for 5 min, incubated with 5 µg/ml cytochalasin B (Sigma-Aldrich Chimie SARL) for 5 min, and then incubated with 10–6 M fMLP (Sigma-Aldrich Chimie SARL) for 10 min as described previously [33
]. Unstimulated control cells were kept at 37°C for 20 min in HBSS alone or maintained at 4°C. After centrifugation, all of the cell-free supernatants were stored at –80°C until IL-8 and MPO assays.
O2–. production
O2–. production was measured with the superoxide dismutase-inhibitable reduction of the ferricytochrome C method, as described previously [23
]. O2–. production by PLB-985 cells (5x105 cells/ml in HBSS) was triggered at 37°C in the presence of ferricytochrome C (0.4 mg/ml) and the stimulus (10–6 M fMLP or 5x108 CFU/ml). Absorbance was monitored continuously for 5 min at 550 nm on a Tecan GENios spectrophotometer (Salzburg, Austria), equipped with a thermostated plate holder, and then converted to nmol reduced cytochrome C by using the extinction coefficient of 2.1 x 104 M–1cm–1.
H2O2 production
Differentiated PLB-985 cells or freshly purified blood neutrophils (106/ml HBSS) were preincubated for 15 min with 1.25 µM 2',7'-dichlorofluorescein diacetate (DCFH-DA; Sigma-Aldrich Chimie SARL) in a water bath with gentle shaking at 37°C. Cells were then stimulated for 15 min with 50 ng/ml PMA (Sigma-Aldrich Chimie SARL), bacteria (5x108 CFU/ml), LPS extracted from the corresponding bacteria (1 µg/ml), or inert polystyrene beads (Molecular Probes, Interchim). In some experiments, cells were primed for 20 min with 10 ng/ml IL-8 (R&D Systems, Abingdon, UK) before incubation with bacteria or PMA. H2O2 production was measured with a FACSCalibur as described above.
Cytokine synthesis and signaling analysis
Isolated blood neutrophils or differentiated PLB-985 cells (5x106 cells/ml) were cultured for 2 h, 4 h, or 8 h in RPMI-1640 culture medium supplemented with 10% FCS, L-glutamine (2 mmol/ml), and gentamycin (200 µg/ml) at 37°C with 5% CO2 in the presence of the following stimulating agents: 100 µg/ml LPS derived from E. coli O55:B5 (Sigma-Aldrich Chimie SARL) plus recombinant human IFN-
at 250 IU/ml (R&D Systems), LPS (0.1–1000 ng/ml) isolated from strain AAEC185, IH11128, or C1845, or the different E. coli strains described in Table 1
(5x108 CFU/ml).
In some experiments, cells were preincubated with the following inhibitors for 30 min at 37°C prior to stimulation with bacteria: 1% D-mannose (pili 1-binding antagonist; Sigma-Aldrich Chimie SARL), 20 µM MG-132 (NF-B inhibitor; AG Scientific Inc., San Diego, CA, USA), 10 ng/ml staurosporine [protein kinase C (PKC) inhibitor; Sigma-Aldrich Chimie SARL], 50 µM LY294002 (PI-3K inhibitor; Alexis, Carlsbad, CA, USA), 100 µM genistein (tyrosine kinase inhibitor), 3 µM SB203580 (p38 MAPK inhibitor; Cell Signaling Technology, Beverly, MA, USA), 20 µM PD98059 (MEK/ERK inhibitor; Cell Signaling Technology), and 5 µM PP2 (Src family kinase inhibitor; Calbiochem, VWR International, Fontenay sous Bois, France). After the indicated times, cell-free culture supernatants were collected and stored at –80°C until cytokine assay.
To better understand the mechanisms of type 1 pili-induced IL-8 synthesis, differentiated PLB-985 (2x106 cells/ml) were preincubated for 30 min with 3 µM SB203580, 20 µM PD98059, or 5 µM PP2 and then stimulated with WT and recombinant strains (5x108 CFU/ml) for 5, 15, 30, or 60 min or with 25 ng/ml PMA for 30 min (positive control). Cells were then lysed in lysis buffer as described, and then proteins were resolved by SDS-PAGE and subjected to Western blot analysis.
SDS-PAGE and Western blot analysis
Purified blood neutrophils and PLB-985 cells were resuspended in lysis buffer (107 cells/ml) containing 50 mM Tris, 1% Nonidet P-40, 150 mM NaCl, 2 mM EDTA, 1 mM Na3VO4, 1 mM NaF, 6.25 mM NaPPi, and a cocktail of protease inhibitors (Sigma-Aldrich Chimie SARL) at pH 7.5 for 15 min at 4°C with gentle agitation. Cell lysates were then centrifuged at 13,000 g for 15 min at 4°C to remove insoluble material. The protein concentration was determined with the bicinchoninic acid assay (Pierce, Rockford, IL, USA). Protein (25 µg) was denatured by boiling in reducing SDS sample buffer. Proteins were resolved by SDS-PAGE and then transferred to polyvinylidene difluoride (PVDF) membranes (Amersham Pharmacia Biotech, UK). For immunoblotting, membranes were washed with TBS containing 0.1% Tween 20 and blocked in TBS containing 0.1% Tween 20, 3% BSA, and 0.5% gelatin. Membranes were probed overnight with the following specific antibodies: anti-CEACAM1, anti-DAF (R&D Systems), anti-CEACAM6 (Genovac), antiphosphorylated (anti-p)ERK, anti-pp38 MAPKs, anti-pSrc family kinases (Cell Signaling Technology), anti-ERK, anti-p38, and anti-Src (Cell Signaling Technology). Blots were then incubated with HRP-linked secondary antibodies (Amersham, Orsay, France), followed by chemiluminescence detection with the ECL Plus kit (Perkin Elmer, Les Ulis, France).
Cytokine and MPO assays
IL-8, IL-1β, and TNF- were quantified by using commercial ELISA kits (Quantikine®, R&D Systems), as recommended by the manufacturer. The detection limits were 5 pg/ml for IL-8 and TNF- and 1 pg/ml for IL-1β. MPO was assayed with an ELISA kit from Oxys Bioxytech® International Inc. (Portland, OR, USA; detection limit, 0.2 ng/ml).
Statistical analysis
Results are expressed as means ± SEM. Bacteria- and LPS-treated and untreated groups were compared by using ANOVA followed by multiple comparisons of means with Fisher’s least significance procedure. The effects of LPS and whole bacteria were compared by using Wilcoxon’s paired test. P values of <0.05 were considered significant.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The expression of receptors for Afa/Dr fimbriae at the PLB-985 cell surface was shown by indirect immunofluorescence, confocal laser-scanning microscopy analysis, flow cytometry, and Western blot analysis. Flow cytometry showed that DAF and CEACAM6 were expressed at the cell surface, as the MFI were, respectively, 86 ± 19 and 15 ± 6 (mean±SEM; n=3). Western blot analysis showed that DAF and CEACAM6 proteins were present at their known molecular masses of 75 kD and 90 kDa, respectively (Fig. 1A ). As shown in Figure 1B , punctuate DAF and CEACAM6 staining was observed at the surface of PLB-985 cells, a pattern characteristic of GPI-anchored proteins. However, DAF was concentrated mainly at the cell periphery, whereas CEACAM6 showed a more polar distribution.
 View larger version (34K): [in a new window] |
Figure 1. Differentiated PLB-985 cells express DAF and CEACAM6. (A) Western blot analysis of PLB-985 cell lysates showing expression of DAF and CEACAM6 proteins by using anti-DAF mAb 8D11 and anti-CEACAM6 mAb 9A6. (B) Immunodetection and confocal laser-scanning microscopy using anti-DAF mAb 8D11 and anti-CEACAM6 mAb 9A6 and an Alexa fluor-conjugated goat anti-mouse antibody. These data are representative of three to five experiments.
|
Altogether, these results indicate that differentiated PLB-985 cells provide a neutrophil-like model suitable for studying neutrophil functional responses following infectious challenge with Afa/Dr DAEC.
The oxidative burst of differentiated PLB-985 cells is triggered by bacterial recognition
We determine H2O2 and O2–. production in response to bacteria. As shown in Figure 2
, differentiated PLB-985 cells produced H2O2 in response to PMA (Fig. 2A)
and to WT IH11128 bacteria (Fig. 2B)
. WT IH11128 and WT C1845 bacteria triggered a significant increase in H2O2 production but less efficiently than PMA (Fig. 2C)
. Similar results were obtained when using isolated blood neutrophils (data not shown). O2–. production increased significantly following challenge with WT IH11128 and C1845 bacteria but less markedly than after challenge with fMLP (Fig. 2D)
. IL-8 did not prime the bacteria-induced oxidative burst: Short exposure of PLB-985 cells to IL-8 prior to bacterial stimulation did not increase H2O2 production significantly, whatever the bacterial strain, as IL-8 increased the PMA-induced H2O2 production (data not shown).
 View larger version (19K): [in a new window] |
Figure 2. Quantification of the oxidative burst H2O2 and O2–. production of differentiated PLB-985 cells in response to E. coli. (A–C) H2O2 production: PLB-985 cells (5x106cells/ml) were preincubated with 1.25 µM DCFH-DA and then stimulated with 50 ng/ml PMA or bacteria (5x108 CFU/ml). The MFI were recorded by flow cytometry. The results presented in C are the mean ± SEM of three independent experiments. Representative histograms are shown in A and B: negative autofluorescence control (dark, shaded histogram), basal production (gray, shaded histogram), PMA-induced production (solid line, A), and WT IH11128-induced production (solid line, B). *, P < 0.05, compared with unstimulated cells. FL1-H, Fluorescence 1-height. (D) O2–. production: PLB-985 cells (5x106 cells/ml) were stimulated with 10–6 M fMLP or bacteria (5x108 CFU/ml). O2–. was monitored for 5 min in the ferricytochrome C reduction assay. Results are expressed as nmol O2–./106 cells/ml and are the mean ± SEM of three independent experiments. *, P < 0.05, compared with unstimulated cells.
|
Preformed mediators (IL-8 and MPO) are released from Afa/Dr DAEC-infected PLB-985 cells
Short-term degranulation experiments were conducted with PLB-985 cells maintained at 4°C or for 20 min at 37°C with inducers of degranulation (PMA or cytochalasin B plus fMLP). As shown in Table 2
, PLB-985 cells treated with PMA or with cytochalasin B plus fMLP (to ensure total degranulation) released large amounts of IL-8 and MPO, contrary to unstimulated cells. WT IH11128 and C1845 bacteria also induced the release of IL-8 and MPO from preformed stocks. The recombinant strains AAEC185pDr and AAEC185pF1845 were about as efficient as the WT strains (Table 2)
. A similar effect was observed following infection of PLB-985 cells with the nonpathogenic strain AAEC185. Thus, like the oxidative burst, the release of IL-8 and MPO preformed stocks from PLB-985 cells in response to E. coli does not appear to be strain-specific.
|
View this table: [in a new window] |
Table 2. Release of Preformed IL-8 and MPO by Infected PLB-985 Cells
|
(122±14, 50±3, 76±10, and 30±8 pg/ml/5x106 cells with AAEC185, WT IH11128, C1845, or control cells, respectively). In the presence of the nonpathogenic AAEC185 control strain and also in basal conditions, differentiated PLB-985 cells released low levels of IL-8 after as little as 4 h of culture, with a small, gradual increase up to 8 h (Fig. 3 A and B
). Challenge with WT IH11128 or C1845 bacteria increased IL-8 production significantly after 8 h of culture as compared with the control strain AAEC185 (Fig. 3 A-C)
.
 View larger version (15K): [in a new window] |
Figure 3. WT IH11128- and C1845-induced IL-8 hyperproduction is not related to Dr or F1845 fimbriae. PLB-985 cells (5x106 cells/ml) were cultured for the indicated times in the presence of each bacterial strain (5x108/ml) or in complete medium alone. (A and B) Time course (2–8 h) of IL-8 release. Effect of WT strains IH11128 and C1845, respectively, together with AAEC185 recombinants bearing Dr or F1845 fimbriae. (C) IL-8 production after 8 h of stimulation by AAEC185 (control), WT IH11128, WT C1845, and AAEC185 recombinants bearing Dr, F1845 fimbriae, and type 1 pili. IL-8 was assayed in the cell-free supernatants by ELISA. Results are expressed as pg/ml/5 x 106 cells (mean±SEM of four independent experiments). *, P < 0.05, as compared with strain AAEC185.
|
Type 1 pili are a major trigger of late IL-8 hyperproduction by PLB-985 cells
We then search to identify the virulence factor(s) present in WT IH11128 and C1845 bacteria, which could be involved potentially in late IL-8 hyperproduction by PLB-985 cells.
As LPS is known to induce IL-8 synthesis [34
], we examined whether LPS, isolated from nonpathogenic E. coli strains AAEC185, WT IH11128, and C1845, could promote late IL-8 production by PLB-985 cells. As shown in Figure 4
, IL-8 production increased in a concentration-dependent manner when PLB-985 cells were treated with LPS-IH11128 and LPS-C1845 but also with LPS from the nonpathogenic strain AAEC185. The levels of IL-8 induced by LPS-IH11128 and LPS-C1845 were ten- to 20-fold lower than those induced by the corresponding whole WT bacteria (Figs. 3
and 4)
. Together, these results indicate that WT LPS is not responsible for the late IL-8 hyperproduction by PLB-985 cells following challenge with the WT E. coli strains IH11128 and C1845. Interestingly, LPS from the three strains failed to induce IL-1β or TNF- production, regardless of the concentration used (data not shown); it also failed to induce significant release of preformed IL-8 and MPO or to stimulate ROS production (H2O2 and O2–.; Table 3
).
 View larger version (31K): [in a new window] |
Figure 4. LPS from strains WT IH11128, WT C1845, and AAEC185 induce late IL-8 production by PLB-985 cells (5x106 cells/ml), which were incubated for 8 h in the presence of LPS-IH11128, LPS-C1845, or LPS-AAEC185 (0.1–1000 ng/ml). IL-8 was assayed in the cell-free supernatants by ELISA. Results are expressed as pg/ml/5 x 106 cells (mean±SEM of four independent experiments).
|
|
View this table: [in a new window] |
Table 3. LPS Fails to Induce the Oxidative Burst or Release of Preformed IL-8 and MPO in PLB-985 Cells
|
 View larger version (15K): [in a new window] |
Figure 5. WT IH11128- and C1845-induced IL-8 hyperproduction is related to type 1 pili. PLB-985 cells (5x106 cells/ml) were cultured for the indicated times in the presence of each bacterial strain (5x108/ml) or in complete medium alone. (A) Effect of 1% D-mannose on IL-8 production after 8 h of stimulation by AAEC185 (control), WT IH11128, C1845, and LF82. (B) Time course (2–8 h) of IL-8 release after stimulation by the recombinant strain AAEC185pSH2 bearing the type 1 pili, in comparison with the AAEC185 control strain. (C) Type 1 pili are involved in IL-8 hyperproduction after 8 h of stimulation by the AAEC185 recombinants bearing AdLF82, pSH2 (type 1 pili), or mutated, nonfunctional type 1 pili (pUT2002), in comparison with AAEC185 control strain (solid bars). The effect of 1% D-mannose is represented in shaded bars. IL-8 was assayed in the cell-free supernatants by ELISA. Results are expressed as pg/ml/5 x 106 cells (mean±SEM of four independent experiments). *, P < 0.05, compared with strain AAEC185; #, P < 0.05, compared with D-mannose-untreated cells.
|
MAPKs and Src family kinases are involved in type 1 pili-induced, late IL-8 production by PLB-985 cells
IL-8 production induced by enterovirulent E. coli is known to be controlled by the MAPK pathway [3
, 5
, 8
]. As shown in Figure 6A
, we observed a rapid increase (after 5 min of infection) in pERK and pp38 MAPKs in WT C1845-infected PLB-985 cells. WT C1845-induced pERK was abrogated by the ERK inhibitor PD98059, and pp38 was abrogated by the p38 inhibitor SB203580. In addition, pSrc was observed in WT C1845-infected cells and decreased after treatment with the Src family kinase inhibitor PP2 (Fig. 6A)
. A similar increase in pERK and pp38 MAPK was observed with WT strains IH11128 and LF82 and was abrogated by specific inhibitors (Fig. 6B)
.
 View larger version (50K): [in a new window] |
Figure 6. MAPK and Src are activated in PLB-985 cells challenged with type 1 pili-bearing E. coli. (A) PLB-985 cells (5x106 cells/ml) were pretreated with 3 µM SB203580, 20 µM PD98059, or 5 µM PP2 for 30 min at 37°C and then stimulated for 5, 15, and 30 min with WT C1845 (5x108 CFU/ml) or 25 ng/ml PMA. (B) PLB-985 cells (5x106 cells/ml) were pretreated with 3 µM SB203580 or 20 µM PD98059 for 30 min at 37°C and then stimulated for 30 min with WT C1845 (5x108 CFU/ml), WT LF82 (5x108 CFU/ml), or WT IH11128 (5x108 CFU/ml). (C) PLB-985 cells (5x106 cells/ml) were pretreated with 3 µM SB203580 for 30 min at 37°C and then stimulated for 30 min with control strain AAEC185 or AAEC185 recombinants bearing AdLF82, pSH2 (type 1 pili), or mutated, nonfunctional type 1 pili (pUT2002). In all of these experiments, cells were then lysed, and protein was loaded on SDS-PAGE and transferred to PVDF membranes. After blocking nonspecific sites, the membranes were probed as indicated with antibodies against pERK, pp38, and pSrc. They were then stripped, washed, and reprobed with antibodies against total forms of ERK, p38, and Src, incubated with HRP-conjugated antibodies, and revealed by adding a chemiluminescent reagent. The figure is representative of five experiments. NS, non-stimulated.
|
Finally, to demonstrate that type 1 pili-dependent activation of MAPKs and Src kinase controls late IL-8 hyperproduction in this setting, we investigated the effects of specific inhibitors of ERK and p38 MAPKs and of Src family tyrosine kinases on WT C1845- and AAEC185pSH2-mediated IL-8 production by PLB-985 cells after 8 h of challenge (Fig. 7 ). The ERK inhibitor PD98059 and the p38 inhibitor SB203580 strongly reduced WT C1845- and AAEC185pSH2-induced late IL-8 production. Genistein, which binds to and inhibits protein tyrosine kinases, and the Src family kinase inhibitor PP2 almost completely inhibited late IL-8 production induced by the two E. coli strains. In conjunction with our other results, these findings strongly suggest that ERK, p38 MAPK, and Src family kinase activation is involved in the IL-8 hyperproduction triggered by type 1 pili-bearing bacteria.
 View larger version (22K): [in a new window] |
Figure 7. Inhibitors of MAPKs and Src family kinases inhibit AAEC185 control strain WT C1845- and AAEC185pSH2-induced late IL-8 production by PLB-985 cells (5x106 cells/ml), which were pretreated with 20 µM PD98059, 3 µM SB203580, 5 µM PP2, 100 µM genistein, 10 ng/ml staurosporine, 50 µM LY294002, or 20 µM MG-132 for 30 min at 37°C and then stimulated for 8 h with strains AAEC185, WT C1845, or AAEC185pSH2 (5x108 CFU/ml). IL-8 was then assayed in the cell-free supernatants by ELISA. Results are expressed as the mean percentage inhibition obtained in four independent experiments.
|
B, strongly inhibited WT C1845- and AAEC185pSH2-induced late IL-8 production (Fig. 7)
, indicating NF-B pathway activation. As the AAEC control strain can also induce a low IL-8 production, the effect of inhibitors was investigated in parallel, and similar percentages of inhibition were observed (Fig. 7) . This suggests that type 1 pili activates intracellular pathways also involved in other types of interaction between the AAEC185 strain and the cells.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B activation. The first neutrophil response to bacteria is the production of ROS and particularly, O2–. and H2O2 in a phenomenon known as the oxidative burst. The neutrophil oxidative burst is dependent on activation of NADPH oxidase (NOX2/gp91phox), a member of the NOX family [17 , 18 ]. The PMN oxidative burst can be "primed" for optimal responsiveness by several proinflammatory cytokines and by LPS [39 ]. Our observation that pathogenic and nonpathogenic E. coli strains induced an early response is consistent with the enhanced PMN oxidative burst previously reported in response to uropathogenic E. coli (UPEC) [40 , 41 ].
Differentiated PLB-985 cells contain various types of rapidly mobilizable granules and secretory vesicles. MPO is stored in azurophilic granules and is necessary to transform the O2–. into H2O2 (a more stable ROS) during the oxidative burst [24 ]. IL-8 is stored in secretory vesicles [42 ]. We show here that short exposure of PLB-985 cells to pathogenic and nonpathogenic E. coli strains triggers the release of large amounts of preformed MPO and IL-8. This rapid mobilization of preformed IL-8 stores upon E. coli stimulation might play a key role in amplifying local neutrophil recruitment and activation. In turn, the release of MPO, IL-8, and ROS in response to E. coli might be involved in intestinal epithelial damage in certain pathological circumstances.
The second step of our study was then to evaluate the prolonged exposure of PLB-985 cells to the bacteria using cytokine production as a functional marker. Indeed, neutrophils modulate inflammatory processes by producing a wide range of cytokines [20
, 24
, 42
]. We found that after 8 h of challenge with WT E. coli strains IH11128, C1845, or LF82, PLB-985 cells produced small amounts of TNF- and IL-1β and large amounts of IL-8.
Experiments with WT, recombinant, and mutant E. coli strains, with and without functional and nonfunctional type 1 pili and a type 1 pili variant, and treatment with D-mannose (a type 1 pili antagonist) showed that type 1 pili were responsible for this strong response and that Dr and F1845 fimbriae were not involved. Similar conclusions could be obtained when using blood neutrophils, confirming that differentiated PLB-985 cells constitute a suitable neutrophil model.
Type 1 pili are common, filamentous appendages consisting of several subunits mediating bacterial adhesion and invasion and triggering certain host cell responses [43 ]. Type 1 pili control the cellular processes necessary for the persistence and recurrence of UTIs, including the formation of intracellular bacterial communities with biofilm-like properties that are protected from host defenses [44 ]. Receptors for type 1 pili or type 1 pili variants have been detected on intestinal cells (CEACAM6) [45 , 46 ]. As these receptors are also present at the surface of PMN and PLB-985 cells, we can suggest that CEACAM6 is a good candidate for mediating the type 1 pili-mediated effect reported in our study. The involvement of type 1 pili in the inflammatory response to E. coli is controversial. In a model based on bladder epithelial cells, type 1 pili were found to cooperate with LPS to promote IL-6 production in vitro [47 ]. However, in a human challenge model with E. coli 83972 fim+, Bergsten et al. [48 ] recently failed to observe type 1 pili-induced urinary tract inflammation.
We identified the signaling pathways that control type 1 pili-induced, late IL-8 hyperproduction in PLB-985 cells. Rapid P-ERK and P-p38 MAPKs were observed in response to the WT strains C1845 and IH11128. In addition, for the first time, we observed MAPK activation and IL-8 production in response to the Crohn’s disease-associated, adherent-invasive WT strain LF82 [31
], expressing a type 1 pili variant [32
]. MAPK phosphorylation was also observed when PLB-985 cells were infected with the recombinant strains AAEC185pAdLF82 expressing the type 1 pili variant from LF82 E. coli and the strain AAEC185pSH2 expressing UPEC type 1 pili [30
]. In contrast and as expected, a low level of MAPK phosphorylation was observed with strain AAEC185pUT2002, expressing a nonfunctional, mutated type 1 pili [47
]. It is interesting to note that Afa/Dr DAEC triggers MAPK activation through different mechanisms in human epithelial intestinal cells and PLB-985 cells. Indeed, in intestinal cells, Betis et al. [3
] have reported that IL-8 production involves DAF recognition by Afa/Dr fimbriae, whereas we show here that the PLB-985 cell IL-8 response is type 1 pili-dependent but does not require Afa/Dr fimbriae. In addition, our results suggest that Src family kinase activation is probably required for type 1 pili-induced, late IL-8 hyperproduction by PLB-985 cells. Finally, we provide evidence that this response is also regulated by the NF-B pathway, as pretreatment with a NF-B inhibitor profoundly inhibited WT C1845- and AAEC185pSH2-induced late IL-8 production. When using the AAEC185 control strain as a stimulus, low levels of IL-8 were released through the same activation pathways as type 1 pili-bearing strains. Our results are in keeping with reports showing the role of MAPKs and NF-B for the production of several inflammatory cytokines by stimulated human neutrophils or PLB-985 [42
, 49
].
Our findings provide new insights into the proinflammatory response to Afa/Dr DAEC infection. In particular, they reveal a role for type 1 pili expressed by Afa/Dr DAEC, and all other reported cellular responses are Afa/Dr adhesin-dependent [9
]. What might be the consequences of Afa/Dr DAEC-induced proinflammatory responses? A recent epidemiological study showed that Afa/Dr DAEC strains inducing IL-8 production by epithelial cells were associated with sporadic diarrhea in children <5 years of age [13
, 50
], confirming the association of Afa/Dr DAEC strains with age-dependent diarrhea [12
]. Moreover, cell invasion [50
] and flagellae [51
] have been implicated in the inflammatory response to clinical isolates expressing Afa/Dr fimbriae or type 1 pili [52
]. Experimental studies point to a role of nonpathogenic and pathogenic E. coli in the induction and/or aggravation of IBD [21
, 53
], but a role of Afa/Dr DAEC in IBD remains to be demonstrated clinically. Other recent experimental findings suggest that the Afa/Dr DAEC-induced proinflammatory response causes cellular changes compatible with IBD lesions. In a T84 cell model, Afa/Dr DAEC-induced PMN migration promotes TNF- and IL-1β hyperproduction and DAF up-regulation [11
], mimicking certain mucosal features in IBD [54
]. In a Caco-2 cell model, the interaction between AfaE-III-positive bacteria and DAF results in IFN--dependent up-regulation of MHC class I chain-related gene A molecules, which are strongly expressed at the surface of colonic epithelial cells in patients with IBD and which bind NK cells via NK-G2D receptors [55
]. Moreover, in Caco-2 cells forming monolayers, Afa/Dr DAEC infection increases paracellular permeability drastically, leading to the expression of the Sat toxin by Afa/Dr DAEC [56
]. This toxin is highly prevalent in IBD-associated E. coli strains belonging to phylogenetic group B2 + D [57
], which also includes Afa/Dr DAEC. Finally, it was reported recently that C1845 could promote vascular endothelial growth factor synthesis by cultured intestinal cells [58
], a cytokine also reported by us to be chemotactic for neutrophils [59
].
Thus, in the course of Afa/Dr DAEC intestinal infection, epithelium- and neutrophil-derived IL-8 could control, at least in part, the flow of neutrophils through the lamina propria. Afa/Dr DAEC-induced IL-8 hyperproduction by PMN might thus be important for inducing and perpetuating local inflammation, and this self-amplifying loop might play a role in the pathogenesis of IBD. Moreover, despite their key beneficial role in bacterial killing, Afa/Dr DAEC-induced ROS generated by PMN could cause structural and functional damage to infected epithelia, thereby disrupting the epithelial barrier. For example, following exposure to ROS, intestinal cells are primed for Fas-mediated cell death, and mucosal permeability and electrolyte transport are enhanced. We are currently attempting to unravel these mechanisms and their possible involvement in IBD.
Received June 11, 2008; revised October 18, 2008; accepted October 24, 2008.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B and MAP kinase pathways and the upregulated expression of interleukin 8 Cell. Microbiol. 4,635-648[CrossRef][Medline]B and AP-1 in T84 cells Infect. Immun. 70,2304-2310B and MAP kinase pathways in the generation of inflammatory cytokines by human neutrophils J. Leukoc. Biol. 81,567-577 up-regulates decay-accelerating factor gene expression in human intestinal epithelial cells Immunology 90,358-363[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
