A novel method for investigating Burkholderia cenocepacia infections in patients with cystic fibrosis and other chronic diseases of the airways
© The Author(s). 2016
Received: 24 March 2016
Accepted: 18 August 2016
Published: 1 September 2016
Burkholderia cenocepacia is a Gram-negative, opportunistic pathogen that is a cause of morbidity and mortality in patients with cystic fibrosis (CF). Research efforts over the past few decades contributed to our understanding of these infections by identifying virulence factors. However, little is known about how this pathogen adapts to the harsh environment found inside the CF airways, which is characterized by a unique mucus containing high concentrations of inflammatory markers. The current study developed a novel model to further investigate this phenomenon.
Monolayers of human A549 lung carcinoma cells (HLCCs) were exposed to a mixture of artificial CF sputum medium (ASMDM) in tissue culture growth medium, and subsequently infected with B. cenocepacia K56-2 for 24 h. The data showed that this model supported B. cenocepacia growth. In addition, consistent with similar studies using current models such as CF airway tissue samples, HLCC viability was reduced by more than 70 % when grown in 60 % ASMDM and infected with B. cenocepacia compared to mock-infected controls and medium alone. Furthermore, the amount of B. cenocepacia cells associated with the HLCC monolayer was more than 10 times greater in 60 % ASMDM when compared to medium controls.
These findings suggest that HLCC monolayers in 60 % ASMDM serve as a valid alternative to study B. cenocepacia infections in patients with CF, and possibly other chronic diseases of the airways. Furthermore, the results obtained in this study suggest an important role for CF sputum in B. cenocepacia pathogenesis.
Mucus plays an important protective role in our host defenses by trapping inhaled particles such as pathogens that are subsequently transported out of the airways by ciliated epithelial cells. For this mucociliary clearance to work effectively, the cilia must constantly be moving in the thin, moist layer of mucus. Some diseases such as cystic fibrosis (CF) and primary ciliary dyskinesia (PCD) impair this clearance mechanism resulting in chronic lung infections. More specifically, mutations in the cystic fibrosis conductance regulator cause a disruption in the transfer of Cl- across the cell membrane. Consequently, the mucus layer becomes thick and dehydrated preventing the successful removal of the particles. PCD patients have defective cilia structure causing the cilia to beat abnormally. Previous research indicated that there exists striking similarities in terms of mucus biophysical and chemical properties between PCD and CF sputum [1–3]. In addition to impaired mucociliary clearance, both diseases have a neutrophil-dominated inflammation in the airways. These conditions are challenging for many bacteria, however highly adaptable pathogens such as Pseudomonas aeruginosa, Staphylococcus aureus, Haemophilus influenza survive in the CF and PCD airways (see review [2, 4]) and B. cenocepacia thrives in the CF airways (see review [4–6]).
Burkholderia cenocepacia is a member of the Burkholderia cepacia Complex (BCC) that comprises 18 species that are Gram-negative opportunistic pathogens . It is innately resistant to a wide array of antibiotics including aminoglycosides, quinolones, and β-lactams , (see review ). It possesses a variety of virulence factors such as cable pili, biofilm formation, degradative enzymes (see review ), (see review ), and it is transmitted from person-to-person  and from the environment . While the majority of CF patients infected with B. cenocepacia experiences a gradual deterioration of lung function [6, 8], nearly 20 % of infected CF patients experience “cepacia syndrome”. This is a fatal case of necrotizing pneumonia sometimes associated with septicemia that may lead to death within one year [6, 13]. Because of these attributes, this pathogen poses a serious threat to CF patients.
Over the past decade progress has led to a greater understanding of how B. cenocepacia causes disease in CF patients (see review ). However, because there is still no definite cure, further research is needed, on how this pathogen adapts to the harsh CF airways as such information could lead to new forms of treatment. Unfortunately, for investigative purposes, it can be very difficult to acquire sputum or tissue samples from CF patients, and there exists a need for an alternative, readily available method to investigate B. cenocepacia infections in CF patients and patients with similar chronic diseases of the airways.
Fung et al.  developed an artificial sputum medium (ASMDM) that approximates the sputum found in CF patients in terms of components, concentrations of the components, and physical properties. In their study, they found that Pseudomonas aeruginosa, another opportunistic pathogen commonly found in CF airways (see review [4, 15]) grew normally and deeply invaded the ASMDM, suggesting that ASMDM mimicked the CF lower airway mucus well . In addition, the fact that ASMDM seems to be a good substitute for CF mucus and a good growth medium for P. aeruginosa suggests that ASMDM would be a suitable growth medium for B. cenocepacia.
Nevertheless, using ASMDM by itself is not sufficient to simulate the in vivo conditions in the lungs of CF patients, as this model lacks live host cells that can respond to a B. cenocepacia infection. Saijan et al. investigated B. cenocepacia infections in the presence of well-differentiated human CF cells and in non-CF human lung epithelial cells . They found that infected CF cells contained significantly more bacteria in both the mucosal layer and inside the cell layer when compared to infected non-CF cells . This study showed that B. cenocepacia penetrated the mucosal and cellular layers which is comparable to its activity in the CF airways.
Our study investigated the accuracy and effectiveness of ASMDM on a monolayer of human A549 lung carcinoma cells, as a model to investigate B. cenocepacia infections in CF and PCD patients. Although the literature indicates that B. cenocepacia has never been isolated from PCD airways, similarities in mucus biophysical and chemical properties and bacterial flora between the sputa seem striking enough to suspect that B. cenocepacia may be able to colonize PCD airways as well. Therefore, this model could be a valid alternative to investigate B. cenocepacia infections in CF patients, and it might be used to investigate B. cenocepacia or other infections in PCD patients as well.
Cell line and propagation
Human A549 lung carcinoma cells (HLCCs) (American Type Culture Collection, Manassas, VA) were maintained in 1 mL aliquots at −80 °C. Prior to experimental use, 1 mL HLCCs was thawed in a 37 °C water bath, transferred to a Falcon T25 vial (Fisher Scientific, Hanover Park, IL) containing 5 mL Ham’s F-12 complete medium (American Type Culture Collection), and incubated at 37 °C (7 % CO2). At confluency, HLCCs were detached from the flask surface with 2 mL 1X trypsin-EDTA (Mediatech, Inc., Manassas VA), and added to 10 mL Ham’s F-12 complete medium in a Falcon T75 vial (Fisher Scientific, Hanover Park, IL).
Bacterial strain and culture conditions
Burkholderia cenocepacia K56-2  was grown in 10 mL Luria-Bertani broth. Cultures were incubated for 24 h at 37 °C with shaking (150 rpm).
Preparation of artificial sputum medium (ASMDM)
ASMDM was prepared in 30 mL aliquots as described by Fung et al. . Porcine stomach mucin (Sigma-Aldrich, St. Louis, MO), DNA from salmon testes (Sigma-Aldrich, St. Louis, MO), potassium chloride (Fisher Scientific, Hanover Park, IL), sodium chloride (Fisher Scientific, Hanover Park, IL), diethylene triamine pentaacetic acid (Fluka Analytical, St. Louis, MO), casamino acids (Acros, Hanover Park, IL), and bovine serum albumin (Sigma-Aldrich, St. Louis, MO) were dissolved in 25 mL sterile dH2O (final concentrations: 10 mg/mL, 1.4 mg/mL, 2.2 mg/mL, 5 mg/mL, 5.9 mg/mL, 5 mg/mL, and 10 mg/mL, respectively). Egg yolk emulsion (0.15 mL) (Becton Dickinson) and the antibiotics ampicillin (1 ug/mL; (Fisher Scientific, Hanover Park, IL) and penicillin (1 ug/mL; Sigma-Aldrich, St. Louis, MO) were added to revent contamination. Tetracycline was omitted because B. cenocepacia is sensitive to this antibiotic . The suspension was stirred for 5 min at room temperature to dissolve the DNA and mucin. The pH was adjusted to 6.5, and sterile dH2O was added to 30 mL ASMDM was stored at 4 °C, but warmed to 37 °C before addition to cells.
Determination of optimal concentration of ASMDM for HLCCs and B. cenocepacia
HLCCs (0.5 mL) were added to gas-permeable 24-well plates (Coy Laboratory Products, Inc., Grass Lake, MI) at 4x105 cells/mL. The plates were incubated at 37 °C (7 % CO2) for 24 h to form a monolayer. The medium was discarded, cells were washed once with 0.5 mL 1X HBSS (Mediatech, Manassas, VA), and then exposed to either 0.5 mL Ham’s F-12 complete medium, ASMDM suspended in Ham’s F-12 complete medium at varying concentrations (20 %, 40 %, 60 %, or 80 %), or 100 % ASMDM. Two wells were not manipulated (i.e. not washed, and medium not replaced), and were designated as the T24 control. Plates were then incubated at 37 °C (7 % CO2). After 24 h, the supernatant was removed from all wells (including T24 wells), washed with 0.5 mL 1X HBSS, and then washed with 0.5 mL 1X trypsin-EDTA. Cells were detached from the well surface with 0.5 mL 1X trypsin-EDTA at 37 °C (7 % CO2) for 10 min. The cell density (cells/mL) was determined. The cell viability (% live cells) was measured with Trypan blue staining.
To determine the optimal concentration of ASMDM for B. cenocepacia viability, 1 mL of the overnight culture was centrifuged at 14,000 rpm for 2 min, washed in 1 mL saline, and then suspended in 1 mL saline. This suspension (10 μL) was added to non-gas permeable 24-well plates (Fisher Scientific, Hanover Park, IL) containing either 0.5 mL Ham’s F-12 complete medium, ASMDM suspended in Ham’s F-12 complete medium at varying concentrations (20 %, 40 %, 60 %, or 80 %), or 100 % ASMDM. The plate was incubated at 37 °C (7 % CO2). After 24 h, CFU/mL per well was determined via viable cell plating. The CFU/mL of the saline suspension was determined via viable cell plating; this is the T0 control.
Assessment of the effects of B. cenocepacia on HLCCs with 60 % ASMDM
HLCCs suspended in Ham’s F-12 complete medium (0.5 ml) was added to a gas-permeable 24-well plate at a density of 3 × 105 cells/mL and incubated at 37 °C (7 % CO2) for approximately 24 h until confluent. The medium was discarded. Cells were washed once with 0.5 mL 1X HBSS, and then exposed to 0.5 mL medium or 60 % ASMDM in Ham’s F-12 complete medium. Wells without HLCCs were filled with 0.5 mL medium or 100 % ASMDM. Wells were either treated with 10 μL of B. cenocepacia in saline at MOI 0.3-5 or mock-infected with saline. Two wells were not manipulated for the duration of the experiment to assess the cell density (T24 controls). The plates were incubated at 37 °C (7 % CO2). After 24 h, damage to the monolayer was assessed with an inverted Zeiss axiovert 40 CFL microscope (W. Nuhsbaum, Inc., McHenry, IL), a microscope-mounted camera (AmScope, Irvine, CA), and Leica LAS v4.6.2. software. Damage was defined as the presence of gaps in the monolayer. To assess the cell density and viability of HLCCs, half the wells were treated as follows. The remaining supernatant was discarded, cells were rinsed twice with 0.5 mL 1X HBSS, and once with 0.5 mL 1X trypsin-EDTA. Wells were then filled with another 0.5 mL 1X trypsin-EDTA and stored at 37 °C (7 % CO2) for 10 min. The cell density and viability were determined.
The remaining wells were used to measure bacterial density. To quantify the bacteria in the supernatant, 0.2 ml was plated for cell viability. To quantify internal and adherent bacteria, the wells were treated as follows. The remaining supernatant was removed, cells were washed twice with 0.5 mL 1X HBSS, and once with 0.5 mL sterile dH2O. Wells were then filled with 1 mL sterile dH2O and incubated at 37 °C (7 % CO2) for one hour to degrade the monolayer and lyse the HLCCs. Samples (0.2 mL) were removed for viable cell plating.
ASMDM concentration for optimal HLCC viability
The effect of ASMDM on the viability of B. cenocepacia
The effect of B. cenocepacia on HLCCs in 60 % ASMDM
Our goal was to develop an alternative, accessible model that mimics the environment of CF airways and PCD--another chronic airway disease--to investigate B. cenocepacia infections. In order to establish this model, we first needed to choose a cell line. CF epithelial cell lines are not economically feasible, so we used the human lung carcinoma A549 cell line (HLCCs). Next, we needed to show that HLCCs could survive in ASMDM. The ASMDM used in this study mimics the harsh conditions in CF sputum that bacterial pathogens would encounter; it is extremely viscous, acidic, and contains components detected in CF sputum . Initially, we exposed HLCCs to 100 % ASMDM in regular tissue culture wells, but we noted a substantial drop in both cell viability and density (data not shown). We hypothesized the thick ASMDM prevented appropriate cellular gas exchange so we grew the HLCCs on gas permeable plates in 350 uL 100 % ASMDM. Again, viability and density were very low with a 100 % mortality. Instead, we tested a range of ASMDM diluted in Ham’s complete medium in order to find an appropriate balance of nutrients and gas exchange for the cells. We chose the 60 % ASMDM because it was the closest concentration that provided a healthy cell viability. Interestingly, there was a drop (although not statistically significant) in HLCC density in the presence of ASMDM, but this decrease was not detected by visual inspection because the center of the monolayers were intact. During the process of removing the ASMDM for subsequent washes, we noted that some of the periphery monolayer had detached. We think that the weight and thickness of the ASMDM caused the detachment. An additional image file shows this in more detail [see Additional file 2]. These results indicate that the cells survived ASMDM, but the ASMDM affected the robustness of the cell density.
Our second task was to show that B. cenocepacia could grow in the ASMDM. Not only did B. cenocepacia grow well, but the density increased over 100-fold in the presence of any ASMDM. The density increased the most (nearly 1000-fold) in the gas permeable plates which is understandable because these plates are designed to facilitate growth. These results align with the fact that Pseudomonas aeruginosa, another CF pathogen, is also able to grow well in this artificial sputum medium .
Finally, we wanted to create a model that shows B. cenocepacia damages the HLCCs. In the presence of 60 % ASMDM, B. cenocepacia associated with the HLCC monolayer in large numbers, compared to HLCCs in the presence of Ham’s F-12 complete medium alone. Similar results were obtained by Sajjan et al. , who found that B. cenocepacia was able to deeply invade the cell layer of infected CF airway tissue samples in the presence of a CF mucosal layer. However, B. cenocepacia largely remained on the surface of the mucosal layer in infected normal airway tissue samples in the presence of a normal mucosal layer . The fact that B. cenocepacia associated with the cell layer in our model suggests that HLCC monolayers in 60 % ASMDM approximates the environment found in the CF airways, and that this model may be a good starting point to study B. cenocepacia infections in these airways.
Aside from the development of a new model to study B. cenocepacia infections in CF and PCD airways, the present study also indicates that CF sputum may play an important role in the pathogenesis of B. cenocepacia. Keeping in mind that ASMDM is similar to CF sputum, there are two pieces of evidence that support this idea. One, there was low viability of infected HLCCs in 60 % ASMDM compared to infected HLCCs in Ham’s F-12 complete medium. Two, there was a large number of bacterial cells found associated with the HLCC monolayers in 60 % ASMDM compared to HLCC monolayers in Ham’s F-12 complete medium. Collectively, these suggest that CF sputum, unlike normal sputum, may enable B. cenocepacia to reach and grow in or on the cell layers. In other words, unlike normal sputum, which serves as a protective layer against B. cenocepacia infection , CF sputum seems to enhance the pathogenesis of B. cenocepacia.
One possible explanation for this phenomenon is that a component of the ASMDM may activate B. cenocepacia motility by inducing the expression of genes such as the rpoN gene, which are required for motility , or flagellar structural genes. Drevinek et al. , for example, demonstrated that the expression of flagellar structural genes such as the fliF and fliS are upregulated in B. cenocepacia when grown in CF sputum, indicating that CF sputum properties may enhance B. cenocepacia motility . CF sputum may also affect B. cenocepacia motility by enhancing its cepRI quorum sensing system, which controls certain motility genes [22, 23]. Moreover, previous studies suggest that some B. cenocepacia quorum sensing genes may be upregulated in CF sputum as well . The signal that leads to increased B. cenocepacia motility in ASMDM may also originate from the eukaryotic host cells, rather than from the sputum. Eukaryotic host cells secrete certain signaling molecules in response to the inflammatory stress induced by initial B. cenocepacia infection or the presence of CF sputum. B. cenocepacia is known to elicit the secretion of inflammatory signals such as IL-8 by lung cells, for example [24, 25]. This signal may increase B. cenocepacia motility in a manner similar to the one described above, or it may even attract B. cenocepacia to the eukaryotic cell layer by means of chemotaxis. In this study, we found that the fold-increase in bacterial density is greater in infected wells containing HLCCs in either 60 % ASMDM or Ham’s F-12 complete medium than in infected wells void of cells, although these differences were not statistically significant. This supports the notion that the HLCCs could be releasing a signal detected by the bacteria.
Whether B. cenocepacia motility is increased randomly or directionally towards the eukaryotic host cells by way of chemotaxis, increased motility will likely lead to more bacteria being associated with the host cell layer. Once it reaches the host cell layer, B. cenocepacia could bind to and cause damage to eukaryotic cell layers, which could explain the decreased HLCC viability and degradation of the infected monolayer observed in 60 % ASMDM.
The results of this study suggest that monolayers of HLCCs in 60 % ASMDM grown in gas-permeable plates serve as a viable starting point to investigate B. cenocepacia infections in patients with CF and PCD. This model might be used in the near future to further our understanding of this particular pathogen, and to develop new ways to eradicate this detrimental infection that, to this day, is still a significant cause of morbidity and mortality in CF patients.
We thank Dr. Ayana Hinton for her advice on handling tissue culture cells.
This work was funded by Denison University.
Availability of data and materials
The datasets supporting the conclusions of this article are included within the article and as additional files. The data generated and analyzed during the study are available from the corresponding author upon request.
CW and CW conceived the experiments, analyzed the data, and wrote the manuscript. Christiaan Wijers performed the experiments and prepared figures. RV repeated some experiments, analyzed data, and prepared a figure. All authors reviewed drafts of the manuscript. All authors read and approved the final version of the manuscript.
The authors declare that they have no competing interests.
Consent for publication
Ethics approval and consent to participate
This research is approved as exempt from IRB review by the Denison University IRB (IRB00008374 Denison University IRB #1 IORG # 0007000). Exempt Approval: DU IRB SU14 #8-241.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Bush A, Payne D, Pike S, Jenkins G, Henke MO, Rubin BK. Mucus properties in children with primary ciliary dyskinesia: comparison with cystic fibrosis. Chest. 2006;129:118–23.View ArticlePubMedGoogle Scholar
- Noone PG, Leigh MW, Sannuti A, Minnix SL, Carson JL, Hazucha M, Zariwala MA, Knowles MR. Primary ciliary dyskinesia: diagnostic and phenotypic features. Amer J Respir Crit Care Med. 2004;169:459–67.View ArticleGoogle Scholar
- Afzelius BA. A human syndrome caused by immotile cilia. Science. 1976;193:317–9.View ArticlePubMedGoogle Scholar
- Mahenthiralingam E, Urban TA, Goldber JB. The multifarious, multireplicon Burkholderia cepacia complex. Nat Rev Microbiol. 2005;3:144–56.View ArticlePubMedGoogle Scholar
- LiPuma JJ. Burkholderia cepacia: management issues and new insights. Clin Chest Med. 1998;19:473–86.View ArticlePubMedGoogle Scholar
- Isles A, Maclusky I, Corey M, Gold R, Prober C, Fleming P, Levison H. Pseudomonas cepacia infection in cystic fibrosis: an emerging problem. J Pediatr. 1984;104:206–10.View ArticlePubMedGoogle Scholar
- Peeters C, Zlosnik JEA, Spilker T, Hird TJ, LiPuma JL, Vandamme P. Burkholderia pseudomultivorans sp. nov., a novel Burkhoderia cepacia complex species from human respiratory samples and the rhizosphere. Sys Appl Microbiol. 2013;36:483–9.View ArticleGoogle Scholar
- Quinn JP. Clinical problems posed by multiresistant nonfermenting gram-negative pathogens. Clin Infect Dis. 1998;1:S117–24.View ArticleGoogle Scholar
- Loutet SA, Valvano MA. A Decade of Burkholderia cenocepacia virulence determinant research. Infect Immun. 2010;78:4088–100.View ArticlePubMedPubMed CentralGoogle Scholar
- McClean S, Callaghan M. Burkholderia cepacia complex: epithelial cell-pathogen confrontations and potential for therapeutic intervention. J Med Microbiol. 2009;58:1–12.View ArticlePubMedGoogle Scholar
- LiPuma JL, Dasen SE, Nielson DW, Stern RC, Stull TL. Person-to-person transmission of Pseudomonas cepacia between patients with cystic fibrosis. Lancet. 1990;336:1094–6.View ArticlePubMedGoogle Scholar
- Baldwin A, Mahenthiralingam E, Drevinek P, et al. Environmental Burkholderia cepacia complex isolates from human infections. Emer Infect Dis. 2007;13:458–61. doi:10.3201/eid1303.060403.View ArticleGoogle Scholar
- Govan JRW, Hughes JE, VanDamme P. Burkholderia cepacia: medical, taxonomic and ecological issues. J Med Microbiol. 1996;45:395–407.View ArticlePubMedGoogle Scholar
- Fung C, Naughton S, Turnbull L, Tingpej P, Rose B, Arthur J, Hu H, Harmer C, Harbour C, Hassett DJ, Whitchurch CB, Manos J. Gene expression of Pseudomonas aeruginosa in a mucin-containing synthetic growth medium mimicking cystic fibrosis lung sputum. J Med Microbiol. 2010;59:1089–100.View ArticlePubMedGoogle Scholar
- Vonberg RP, Gastmeier P. Isolation of infectious cystic fibrosis patients: Results of a systematic review. Infect Control Hosp Epidemiol. 2005;26:401–9.View ArticlePubMedGoogle Scholar
- Sajjan U, Keshavjee S, Forstner J. Responses of well-differentiated airway epithelial cell cultures from healthy donors and patients with cystic fibrosis to Burkholderia cenocepacia infection. Infect Immun. 2004;72:4188–99.View ArticlePubMedPubMed CentralGoogle Scholar
- Darling P, Chan M, Cox AD, et al. Siderophore production by cystic fibrosis isolates of Burkholderia cepacia. Infect Immun. 1998;66:874–7.PubMedPubMed CentralGoogle Scholar
- Tomlin KL, Clark SRD, Ceri H. Green and red fluorescent protein vectors for use in biofilm studies of the intrinsically resistant Burkholderia cepacia complex. J Microbiol Methods. 2004;57:95–106.View ArticlePubMedGoogle Scholar
- Burns JL, Jonas M, Chi EY, Clark DK, Berger A, Griffith A. Invasion of respiratory epithelial cells by Burkholderia (Pseudomonas) cepacia. Infect Immun. 1996;64:4054–9.PubMedGoogle Scholar
- Duff C, Murphy PG, Callaghan M, McClean S. Differences in invasion and translocation of Burkholderia cepacia complex species in polarised lung epithelial cells in vitro. Microb Pathog. 2006;41:183–92.View ArticlePubMedGoogle Scholar
- Saldias MS, Lamothe J, Wu R, Valvano MA. Burkholderia cenocepacia requires the RpoN sigma factor for biofilm formation and intracellular trafficking within macrophages. Infect Immun. 2008;76:1059–67.View ArticlePubMedPubMed CentralGoogle Scholar
- Drevinek P, Holden MTG, Ge Z, Jones AM, Ketchell I, Gill RT, Mahenthiralingam E. Gene expression changes linked to antimicrobial resistance, oxidative stress, iron depletion and retained motility are observed when Burkholderia cenocepacia grows in cystic fibrosis sputum. BMC Infect Dis. 2008;8:1–16.View ArticleGoogle Scholar
- Lewenza S, Visser MB, Sokol PA. Interspecies communication between Burkholderia cepacia and Pseudomonas aeruginosa. Can J Microbiol. 2002;48:707–16.View ArticlePubMedGoogle Scholar
- Reddi K, Phagoo SB, Anderson KD, Warburton D. Burkholderia cepacia-induced IL-8 gene expression in an alveolar epithelial cell line: signaling through CD14 and mitogen-activated protein kinase. Pediatr Res. 2003;54:297–305.View ArticlePubMedGoogle Scholar
- Palfreyman RW, Watson ML, Eden C, Smith AW. Induction of biologically active interleukin-8 from lung epithelial cells by Burkholderia (Pseudomonas) cepacia products. Infect Immun. 1997;65:617–22.PubMedPubMed CentralGoogle Scholar