Potential mechanisms underlying the acute lung dysfunction and bacterial extrapulmonary dissemination during Burkholderia cenocepacia respiratory infection

Abstract

Background: Burkholderia cenocepacia, an opportunistic pathogen that causes lung infections in cystic fibrosis (CF) patients, is associated with rapid and usually fatal lung deterioration due to necrotizing pneumonia and sepsis, a condition known as cepacia syndrome. The key bacterial determinants associated with this poor clinical outcome in CF patients are not clear. In this study, the cytotoxicity and procoagulant activity of B. cenocepacia from the ET-12 lineage, that has been linked to the cepacia syndrome, and four clinical isolates recovered from CF patients with mild clinical courses were analysed in both in vitro and in vivo assays.

Methods: B. cenocepacia-infected BEAS-2B epithelial respiratory cells were used to investigate the bacterial cytotoxicity assessed by the flow cytometric detection of cell staining with propidium iodide. Bacteria-induced procoagulant activity in cell cultures was assessed by a colorimetric assay and by the flow cytometric detection of tissue factor (TF)-bearing microparticles in cell culture supernatants. Bronchoalveolar lavage fluids (BALF) from intratracheally infected mice were assessed for bacterial proinflammatory and procoagulant activities as well as for bacterial cytotoxicity, by the detection of released lactate dehydrogenase.

Results: ET-12 was significantly more cytotoxic to cell cultures but clinical isolates Cl-2, Cl-3 and Cl-4 exhibited also a cytotoxic profile. ET-12 and CI-2 were similarly able to generate a TF-dependent procoagulant environment in cell culture supernatant and to enhance the release of TF-bearing microparticles from infected cells. In the in vivo assay, all bacterial isolates disseminated from the mice lungs, but Cl-2 and Cl-4 exhibited the highest rates of recovery from mice livers. Interestingly, Cl-2 and Cl-4, together with ET-12, exhibited the highest cytotoxicity. All bacteria were similarly capable of generating a procoagulant and inflammatory environment in animal lungs.

Conclusion: B. cenocepacia were shown to exhibit cytotoxic and procoagulant activities potentially implicated in bacterial dissemination into the circulation and acute pulmonary decline detected in susceptible CF patients. Improved understanding of the mechanisms accounting for B. cenocepacia-induced clinical decline has the potential to indicate novel therapeutic strategies to be included in the care B. cenocepacia-infected patients.

Figure 1

Citotoxicity of B. cenocepacia assessed by FACS detection of cell staining with propidium iodide (PI). (A) Means (± SD) of the percentages of PI-stained cells detected in two assays carried out at least in triplicate. **, p < 0.01 and ***, p < 0.001 when data were compared with those from control non-infected cells; (B) Representative histograms showing the PI staining intensity of control and infected cells. y-axis corresponds to cell number whereas x-axis corresponds to log fluorescence intensity.

Figure 2

Modulation of TF expression in infected cultures. (A) Percentage of TF-expressing cells in control and infected cultures, determined by FACS analysis; (B) Concentration of TF in supernatants from control and infected airway epithelial cell cultures. (C) Procoagulant activity in cell culture supernatants. Data are means (± SD) of the results obtained in at least two different assays carried out in tripicate. *, p < 0.05 and **, p < 0.01 and p < 0.001 when data were compared with the results obtained with control non-infected cultures.

Figure 3

Microparticle release from control and infected cells. (A) Number of microparticles in control and infected cell culture supernatants submitted to FACS analysis for 1 min; (B) Percentage of TF positive/annexinV positive microparticles in supernatant from control and infected cultures. Data are means (± SD) of the results obtained in two assays carried out in triplicate. *, p < 0.05 and **, p < 0.01 when data from control and infected cells were compared with each other.

Figure 4

PFGE profiles of genomic B. cenocepacia DNAs and dendogram resulting from computer analysis of PFGE profiles.

Figure 5

(A) Blood leukocyte concentration in control and infected mice. Data are means (± SD) of the results obtained in two assays in which at least 12 animals from each group were analysed. **, p < 0.01 and ***, p < 0.001 when data from control and infected mice were compared with each other. (B) Percentage of mice from each group with positive liver cultures; (C) Bacterial concentration in liver parenchyma. **, p < 0.01 when data from Cl-2- and Cl-4-infected mice were compared with data from the other groups by the Wilcoxon nonparametric test.

Figure 6

(A) Total protein, (B) leukocyte, (C) LDH concentrations and (D) procoagulant activity in BALFs from control and infected mice. Data are means (± SD) of the results obtained two assays in which at least 12 animals from each group were analysed *, p < 0.05, **, p < 0.01, ***, p < 0.001 when data from control and infected mice were compared with each other.