Infection and Immunity, September 2004, p. 5530-5533, Vol. 72, No. 9
0019-9567/04/$08.00+0 DOI: 10.1128/IAI.72.9.5530-5533.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Unité Prévention et Thérapie Moléculaires des Maladies Humaines,1 Unité Postulante Cytokines & Inflammation, Institut PasteurParis,3 Centre Hospitalier Intercommunal Créteil, Service de Pneumologie, Creteil, France2
Received 3 March 2004/ Returned for modification 19 April 2004/ Accepted 10 May 2004
|
|
|---|
|
|
|---|
Interactions between bacteria and respiratory epithelial cells can modulate the host immune response through the secretion of proinflammatory cytokines (16) and induce respiratory inflammation as observed in the lungs of patients who died of whooping cough (25). Respiratory pathogens, such as Pseudomonas aeruginosa, Streptococcus pneumoniae, Haemophilus influenzae, Mycoplasma pneumoniae, and Chlamydia pneumoniae, produce a variety of molecules (such as lipopolysaccharide [LPS], adhesin, and flagellum) that interact with epithelial cells and cause secretion of proinflammatory cytokines such as interleukin-6 (IL-6), IL-8, and tumor necrosis factor alpha (7, 9, 13-15, 17-19, 22, 27, 33-35). It was previously shown that infection of a human bronchial epithelial cell line, BEAS-2B, with B. pertussis upregulates mRNA expression of a number of proinflammatory cytokine-encoding genes (genes for IL-6 and IL-8, etc.) and induces IL-6 and IL-8 protein secretion in the supernatants of infected cells (3). However, the bacterial factors responsible for this effect were not identified. In this study, we examined the effect of B. pertussis as well as that of purified toxin or adhesin on secretion of IL-6 and IL-8 by a human tracheal epithelial (HTE) cell line.
HTE cells infected with Bvg wild-type B. pertussis secrete IL-6 and IL-8 cytokines. Simian virus 40-immortalized HTE cells were cultured on a collagen gel (Collagen G; Poly Labo) as described previously (11). Briefly, HTE cells were cultured in Dulbecco's modified Eagle medium (GIBCO Laboratories) supplemented with 2% Ultroser G (GIBCO Laboratories), streptomycin (100 µg/ml), and ampicillin (100 µg/ml; Sigma). Confluent monolayers of HTE cells were grown in 24-well plates to obtain 2.5 x 105 cells per well in 0.5 ml the day of the infection. The experiments were performed using either the parental Bvg wild-type strain (BP338) (32) or the Bvg mutant (BP359) (31). Bacteria were grown on Bordet-Gengou (BG) blood agar at 36°C for 5 days. Isolated colonies were then replated onto BG medium and incubated for 24 h before each experiment. Finally, the day of the infection, the bacteria were resuspended in the complete medium without antibiotics to an optical density at 650 nm of 1.0. Except when indicated, the following infection scheme was used: 2.5 x 107 CFU in 0.5 ml was added to each cell well to reach a bacterium/cell ratio of 100:1 and incubated at 37°C in 5% CO2 for 24 h under static conditions. The bacterial inoculum was evaluated by plating dilutions of the bacterial suspension onto BG blood agar and counting CFU after 3 days at 37°C. As a positive control, the cells were stimulated with IL-1ß (20 ng/ml; Sigma) (9), and as a negative control, the cells were incubated with culture medium to detect the basal level of cytokines secreted into the culture supernatant. Each condition was tested in triplicate. Culture supernatants were collected after 24 h of interaction between HTE cells and bacteria, centrifuged at 15,900 x g for 30 min, frozen, and stored at 80°C until assayed for IL-6 and IL-8 content by using homemade enzyme-linked immunosorbent assays (6, 26). Briefly, microtitration plates (96 wells; Maxisorp; Nunclon, Rockeville, Denmark) were coated with 100 µl of mouse monoclonal anti-human IL-6 or IL-8/well for 2 h at 37°C or overnight at 4°C. After washings and the protein-blocking step, cell supernatants from activated cell cultures in Dulbecco's modified Eagle medium, tested at 1:10 in phosphate-buffered saline-Tween-bovine serum albumin, were added in duplicate. A 2-h incubation was carried out at 37°C. After washings, rabbit polyclonal anti-human IL-6 or IL-8 was added for 90 min at 37°C. After further washings, peroxidase-labeled goat anti-rabbit immunoglobulin (Silenus-AMRAD Biotech, Victoria, Australia) was added. After incubation at 37°C for 1 h, plates were washed and enzymatic activity was revealed by the addition of ortho-phenylene-diamine substrate (Sigma). The reaction was stopped by the addition of 3 N HCl. Optical densities at 490 and 630 nm were measured with a microplate reader spectrophotometer (Dynex Technologies, Chantilly, France). Titration curves were determined with recombinant human IL-6 and IL-8, respectively.
We first determined a dose response curve with different bacterium/cell ratios. As shown in Fig. 1, incubation of the parental strain BP338 with HTE cells led to a significant increase in levels of secreted IL-6 (P < 0.05) in supernatants of cell cultures compared to levels in supernatants of negative controls, and this increase was bacterial-dose dependent (Fig. 1). The ratio of 100 bacteria per cell was then chosen for all other experiments, as usually used (2, 10). Parental strain BP338 killed by heating at 56°C for 30 min did not stimulate cytokine production at a ratio of 100 bacteria per cell (Fig. 1) or at a ratio of 1,000 bacteria per cell (data not shown). Furthermore, the amount of IL-6 secreted was dependent on the expression of vir-activated genes since a Bvg mutant (BP359) was unable to induce IL-6 release even at a high bacterium/cell ratio (Fig. 1).
![]() View larger version (12K): [in a new window] |
FIG. 1. IL-6 secretion by HTE cells exposed to B. pertussis (live and heat-killed) parental Bvg wild-type (BP338) and Bvg mutant (BP359) bacteria for 24 h. Black columns represent BP338, and white columns represent BP359. Ratios of bacteria to cells were 1,000:1, 100:1, and 10:1 as indicated. IL-1ß (20 ng/ml) was used as a positive control. The results are expressed in picograms of IL-6 per milliliter of culture fluid. Error bars represent the standard errors of the means of results from three to six experiments. One-way analysis of variance was used to identify differences between the groups. If the analysis of variance was significant, a nonparametric Mann-Whitney test was used to compare the means. A P value of <0.05 was considered significant. *, P of <0.05 versus BP338 (100 bacteria per cell). , P of <0.05 versus negative control.
|
Bacterial uptake is not involved in IL-6 secretion by HTE cells. It was previously demonstrated that B. pertussis invades HTE cells (2). To determine whether this invasive capacity is necessary to induce IL-6 and IL-8 secretion by HTE cells, we measured cytokine production in the presence of cytochalasin D. An invasion assay was performed as described previously (2) with 24-h contact between HTE cells preincubated or not with cytochalasin D (5 µg/ml in 0.5% dimethyl sulfoxide) and parental strain BP338 (ratio of 100 bacteria per cell). The number of intracellular bacteria decreased from 0.5 ± 0.1 (mean ± standard error of the mean; three experiments) to 0.08 ± 0.02 bacteria per cell with cytochalasin D, confirming that this reagent blocks the internalization of more than 80% of B. pertussis bacteria by HTE cells.
The amounts of IL-6 produced by the cells after interaction with the bacteria were similar in the presence and absence of cytochalasin D (6,023 ± 2,022 and 6,346 ± 2,105 pg/ml, respectively). A similar observation was made for IL-8 (848 ± 261 and 711 ± 126 pg/ml, respectively). This result suggests that actin-mediated internalization of either whole organisms or ligands is not required to stimulate the epithelial IL-6 response. Similarly, the internalization of P. aeruginosa is not required to induce the secretion of IL-8 by the same cells used in this study (8).
AC-Hly is the bacterial factor involved in IL-6 secretion by HTE cells. To characterize the bacterial factors involved in the secretion of IL-6 and IL-8 by the HTE cells, we first performed cytokine assays (as described above) with different mutants deficient in the expression of adhesins or toxins (at a ratio of 100 bacteria per cell). The strains and mutants used were the parental Bvg wild-type strains (BP338 and 18323) and their derived mutants: BPMC (FHA) (24), BBC42 (PRN) (29), BPRA (PT) (1), BP348 (AC-Hly) (32), SK16 (FHA) (23), and 18HS19 (AC-Hly) (20). The amounts of IL-6 secreted by the cells were not significantly different after interaction either with the parental strain BP338 or with its corresponding FHA-, PRN-, or PT-deficient mutants or with the parental strain 18323 or its corresponding FHA-deficient mutant (Fig. 2A).
![]() View larger version (24K): [in a new window] |
FIG. 2. IL-6 secretion by HTE cells exposed for 24 h to B. pertussis parental strains BP338 and 18323 and mutants (A) or to purified bacterial factors (B). (A) Black columns represent the BP338 background, and grey columns represent the 18323 background. Results shown are those from three to six experiments, except in the case of BPRA, for which experiments were done twice (realized in triplicate for each strain and mutant). The bacterium/cell ratio was 100:1. To combine data from different experiments, concentrations of IL-6 were normalized by setting the level of IL-6 secreted by HTE cells infected with the parental strain Bvg wild-type BP338 or 18323 24 h after infection at 100% within each experiment. One-way analysis of variance followed by a paired t test was used to identify differences between the groups. A P value of <0.05 was considered significant. P, <0.05 versus parental B. pertussis strain, BP338 Bvg wild type (*) or 18323 Bvg wild type (&). (B) The results are expressed in picograms of IL-6 per milliliter of culture fluid. ^, P of <0.05 versus negative control.
|
To confirm the results obtained with the different mutants, we examined the role of purified bacterial proteins. The infection scheme described above was carried out using purified FHA and PT (both a kind gift from M. J. Quentin-Millet of Aventis-Pasteur), PRN (a kind gift from P. Denoel of GlaxoSmithKline), and recombinant AC-Hly (4) incubated for 24 h at two concentrations (1 and 5 µg/ml). Recombinant AC-Hly was purified from Escherichia coli XL1 carrying the plasmid pCACT3 and analyzed as described earlier. Briefly, the toxin was purified after urea extraction, DEAE filtration, and affinity chromatography on calmodulin affigel (4). The data obtained with purified bacterial factors confirmed those obtained with the different mutants. In fact, purified PRN and PT did not elicit IL-6 secretion by the epithelial cells compared to the control (Fig. 2B). Purified FHA induced low amounts of IL-6 expression at 5 µg/ml (Fig. 2B) or at 25 µg/ml (data not shown) but with no statistical significance compared to the control. However, purified AC-Hly was able to trigger an IL-6 response after 24 h of incubation with the cells (Fig. 2B). In our study, we used recombinant AC-Hly produced and purified from E. coli. The purified enzyme was analyzed for purity by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. We tested the effect of preincubation of AC-Hly with polymyxin B (2 µg/ml) for 1 h at 37°C. The amounts of IL-6 induced by the cells stimulated by AC-Hly or by AC-Hly in the presence of polymyxin B were the same (data not shown), indicating that eventual E. coli LPS contamination does not play a role in the AC-Hly activation. Furthermore, the action of purified E. coli LPS was tested. No induction of IL-6 or IL-8 was observed after incubation of the cells with E. coli LPS at concentrations ranging from 1 to 10,000 ng/ml (data not shown). The action of AC-Hly on human epithelial cells seems to be specific. On the contrary, a recent study demonstrated that the expression of IL-6 by murine macrophages and dendritic cells is also enhanced by the action of AC-Hly but that this action synergizes with that of E. coli LPS (30).
Similar experiments were performed in order to determine the amounts of IL-8 secreted by the infected cells, but no significant differences were found whatever the mutant used (data not shown), confirming the results obtained with the Bvg mutant.
In conclusion, B. pertussis induces the synthesis of high levels of IL-6 but very low levels of IL-8 by human epithelial cells. The secretion of IL-6 is linked to the expression of AC-Hly. This finding confirms the important role of AC-Hly in the induction of the inflammatory response (12, 21).
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»