© 2004 American Thoracic Society
Burkholderia cepacia Complex and Cystic FibrosisIn Search of the Smoking Gun
a University of British Columbia Vancouver, British Columbia, Canada The host–bacterial interaction in cystic fibrosis (CF) results in a highly unusual chronic respiratory inflammatory process, the nature of which has been an area of intense investigative scrutiny. Theories abound to explain the prototypic chronic endobronchial infection in CF, but a single unifying explanation remains elusive. Among the bacterial pathogens infecting the CF lung, members of the Burkholderia cepacia complex are most feared because of their high-level intrinsic antibiotic resistance, their propensity for patient-to-patient spread, and their capacity to cause profound inflammation and fatal invasive disease (1). The study reported by De Soyza and coworkers (2) in this issue of the Journal (pp. 70–77) attempts to identify bacterial factors responsible for the catastrophic outcome associated with infection by B. cepacia in some patients with CF. B. cepacia complex (formerly B. cepacia) is a class of several related but genetically distinct species (3). Infection in CF is usually caused by bacteria from one of two species—B. cenocepacia (formerly genomovar III) and B. multivorans (formerly genomovar II) (4, 5)—and infection with the former is most often associated with patient-to-patient spread and fatal invasive infection (the so-called "cepacia syndrome") (6). Molecular differences among the genomovars have been suggested as the cause of the differential pathogenicity, and their LPS has been proposed as the most likely phlogistic agent. The report in this issue of the Journal provides important new information on the potential differences in inflammatory potential among bacterial strains within the B. cepacia complex and demonstrates that whole-cell lysates from certain strains elicit a more robust proinflammatory cytokine (tumor necrosis factor and interleukin-1) response than others from a human monocytic cell line (2). Importantly, the strains with greatest proinflammatory potential were from the B. cenocepacia species, the group of organisms with the greatest potential for devastating invasive disease in patients with CF, particularly in those undergoing lung transplantation (7, 8). There were great differences, however, among strains within the different species, and indeed among different isolates from the same genetic lineage (ET12, the cable-piliated epidemic strain which has been the dominant strain in Ontario, Canada). In fact, the ET12 isolates in this study that were LPS-smooth (containing O antigen side chains) were less inflammatory than those that were LPS-rough (with no O antigen side chains). This heterogeneity among strains from the same species illustrates the importance of examining multiple clinical bacterial isolates and not generalizing the results from studies of a single strain.
De Soyza and coworkers (2) attempted to identify the bacterial product responsible for the proinflammatory difference among whole-cell lysates from different strains in the B. cepacia complex. Because they demonstrated that cytokine induction was CD14-dependent in three strains of different species, LPS was investigated. Here, the authors extracted the LPSs of only two strains from patients with CF. LPS was isolated from one strain of B. cenocepacia, which is the most problematic species in patients with CF, and one strain of B. multivorans, which infects these patients but which is associated generally with less severe pulmonary disease. Differences between the two LPSs were demonstrated in their inflammatory potential (consistent with the effects of the whole cell lysates) and biochemically. These assays showed that the B. cenocepacia LPS induced increased tumor necrosis factor- Matrix-assisted laser desorption–time of flight (MALDI-TOF) mass spectrometric analysis of the lipid A moieties from these two strains was performed. Each apparently has a disaccharide backbone that was acylated at 2, 3, 2', and 3' positions by 3-hydroxy fatty acids. The negative ion mode spectral analysis showed both similarities and differences in the lipid As between the two strains. Both the B. cenocepacia and the B. multivorans contained lipid A with both single and double aminoarabinose modifications. The B. cenocepacia also included lipid A entirely devoid of aminoarabinose. In data "not shown" mass spectrum of lipid A from a B. multivorans strain that was a potent cytokine inducer was apparently the same as that of the B. cenocepacia strain, while the B. multivorans that was a weak inducer had a lipid A species that was deacylated, lacking a hydroxylated myristic acid and an aminoarabinose. Differences in proinflammatory capacity of the different strains were inferred from studies with the extracted LPS; however, such conclusions should probably be considered preliminary until several more strains with differences in inflammatory capacity can be examined. In fact, comparing the LPS from strains of the same lineage that gave differing responses might have given clearer results than comparing LPS of different species. Alternatively, chemical modification of the lipid A could have been performed to directly correlate the structure and function of these molecules. The observed differences between the lipid A portion of two LPSs could be important, particularly in the context of the evolving understanding of the inflammatory capacity of similar molecules from Pseudomonas aeruginosa (9, 10). In that species of bacteria (which is the predominant pulmonary pathogen in CF), the degree of acylation, and other modifications of lipid A, is directly related to the inflammatory potential of the bacteria. Adaptation of the bacteria to chronic CF lung infection could therefore be in part responsible for the unusual host–bacterial détente characteristic of the disease. Induction of inflammation via stimulation of the innate immune system is an area of intense investigation. The different Toll-like receptors and their adaptor proteins appear to activate a range of inflammatory programs with heterogeneous effects on the host (11). It is possible that LPS activates host inflammation by multiple mechanisms, which remain to be elucidated. Moreover, inflammation can be induced by many different classes of cells in the conducting airway (epithelial cells, polymorphonuclear leukocytes, alveolar macrophages, and dendritic cells). The studies by De Soyza (2, 7) and coworkers provide an important first step in understanding the differences in inflammatory potential among the different species of bacteria within the B. cepacia complex. Studies with products from a wider range of bacterial strains and their effects on primary human airway cells, however, will provide the essential next steps in investigation of the virulence of this important class of CF pathogens. FOOTNOTES Conflict of Interest Statement: D.P.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. J.B.G. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
and interleukin-6 compared with the B. multivorans LPS.