Bacterial Diversity in Cases of Lung Infection in Cystic Fibrosis Patients
KeywordsCystic Fibrosis Bacterial Community Cystic Fibrosis Transmembrane Conductance Regulator Cystic Fibrosis Lung Pulmonary Exacerbation
Cystic fibrosis (CF) is a common autosomal recessive genetic disorder, which stems from mutations in the cystic fibrosis transmembrane conductance regulator gene. While this disorder impacts on many body systems, the predisposition to airway infection for individuals with CF is particularly important. These infections, and associated host immune response with neutrophil-driven inflammation, lead to progressive lung damage. The respiratory failure that follows is the leading cause of death for these individuals whose median age of survival is in their late 30s (Cystic Fibrosis Foundation 2013; UK CF Registry 2012). Maintaining lung function is therefore critical to the well-being of individuals with CF. The aim of this overview is to detail the state of existing knowledge and advances made in the past decade from studies of the bacterial component of the microbiota of the airways of individuals with cystic fibrosis.
CF lung disease is characterized by relentless cycles of mucus obstruction, infection, and inflammation with the airways, and current therapeutic strategies target one or more of these individual components. There have been marked improvements in treatment, mostly antibacterial agents and mucolytics, and this has been paralleled by significant improvements in survival. Key then to improved therapy is defining what species are present and are driving lung damage in this context. A range of bacterial species have been associated by conventional diagnostic microbiology with airway infection in CF. These species have included known human pathogens such as Haemophilus influenzae and Staphylococcus aureus, opportunistic pathogens including Achromobacter xylosoxidans, Pseudomonas aeruginosa, and Stenotrophomonas maltophilia, as well as species less commonly associated with human infections such as Burkholderia cepacia complex (Lipuma 2010). While the role of many of these species in lung disease progression is not clear, certain species such as P. aeruginosa have been associated with poor clinical outcomes. P. aeruginosa is viewed as of particular importance when first detected in the pediatric CF airways, with robust attempts to eradicate the species employed clinically.
More generally, the means by which these species are currently detected however is important. Traditional culture-based microbiology, which employs selective media incubated aerobically to detect pathogens, remains the standard means by which the bacteriological content of CF airway samples is assessed.
Culture-based diagnostic microbiology has been of tremendous benefit in CF and other infections, but is bounded by the ability to detect particular species. Molecular methods, however, have consistently demonstrated that many more species were present in CF respiratory samples than those focused on by culture-driven diagnostic microbiology. Though many advances were made in CF microbiology prior to 2002, that year saw an important first step in the process of widening our understanding of the microbes in the CF lung. Coenye et al. (2002) reported a collection of “unusual bacteria” that had been cultured from the respiratory secretions of individuals with cystic fibrosis. The following year, Rogers et al. (2003) applied for the first time a culture-independent means of analysis, developed in environmental microbiological studies, to focus on the species present in airway samples from adults with CF. This study supported both the view of P. aeruginosa as being common by this stage of life in the CF airways. It also, however, drew attention to a wide range of other bacterial species, including many that at that time had not been identified in the CF lung. Among them were species that required anaerobic conditions for growth – under conventional diagnostic microbiological procedures for the analysis of CF airway samples, these anaerobic species would not have been detected. Since these studies, other cross-sectional studies have addressed the bacteria in adult CF airways and pediatric CF airways (Harris et al. 2007). The trend of detecting novel and often anaerobic bacterial species in the CF airways (Tunney et al. 2008) has continued through the application of next generation sequencing strategies (e.g., Guss et al. 2011). While these advances add to our understanding of the complexity of CF airway microbiology, the importance of these species to the pathophysiology of lung disease is not clear.
Possible New Pathogens
These studies showed therefore that a complex mix of species was present in the airways of individuals with CF. They also demonstrated that a range of species were present that were considered to be pathogens in other infectious scenarios. The first longitudinal analysis of species present in the CF airways proposed a new candidate pathogenic bacterial species, namely, the Streptococcus milleri group (Sibley et al. 2008). Evidence from this work, based on culture-independent analysis of airway samples from a patient collected over the course of a 6-year period, associated the presence of this species with periods of pulmonary exacerbation. Through culture-independent analyses, several other bacterial species, including the anaerobe Prevotella intermedia, have also been suggested to be of potential importance in lung disease progression (Ulrich et al. 2010). This theme – of pathogens undetected by current culture-based approaches – is common to many of the publications highlighted already. One may conclude from these consistent findings that CF airway infections are phylogenetically complex, and it would thus be wise to consider more than just single microbial species to understand lung disease pathogenesis. The importance of the entire community, or at least parts of the community, acting in a “pathogenic” manner has, for example, also been raised.
Following these first studies, a range of methodological issues were identified as important to consider in defining the bacterial microbiota present in CF airway samples. At a fundamental level, the issue of which clinical sample best reflects the contents of CF airways has been a focus of discussion in many studies. While expectorated sputum is the most readily collected, bronchoalveolar brushing and lavage strategies may have advantages, for example, in minimizing contamination by oral and upper respiratory tract microbiota. The availability of these samples is in part influenced by the clinical status and (typical) age of patient (particularly in relation to sputum production) and what can be justified on a clinical basis (particularly with respect to bronchoscopic sampling). This is an important debate and one that balances a range of practical and ethical issues. The issue of sampling has been considered in other ways too. Rogers et al. (2010a) showed that one sample did not give a complete picture of the microbiota present. Taking multiple samples may of course be impractical in most clinical contexts, but these results question the current reliance on a single sample for diagnostic purposes. One likely explanation for this finding is that there is heterogeneity in the distribution of bacterial species across the CF airways. At another fundamental level, we know that the bacteria present within the CF airways are subject to many challenges, including a robust host immune response and frequent antibiotic treatments, resulting in a mixture of live and dead bacterial cells in the airways, regardless of sampling technique. Efforts therefore have been directed at removing the signal of nonviable cells from culture-independent studies by using photochemical cross-linking of nucleic acids (Rogers et al. 2010b). Thus, technical approaches may have a dramatic impact on the results of culture-independent analyses. Moreover, in a recent study, Zhao et al. (2012a) showed how the choice of method for DNA extraction from respiratory samples leads to very different sensitivity for the detection of the common CF airway pathogen, S. aureus, as above. Combining these points, it is clear that what is sampled and how that sample is processed has a bearing on the findings that emerge.
Bacterial Microbiota Linked to Clinical Parameters
Despite these technical concerns, dramatic advances have already been made in our understanding of the CF lung microbiota. Here, studies initially cross-sectional in design have been important in relation to their identification of associations between microbiota and clinical characteristics. Cox et al. (2010), for example, showed that there was a progressive loss of airway microbial diversity observed as the age of patient sampled increased. This study also identified specific species as either early or late colonizers of the CF lung. Klepac-Ceraj et al. (2010) similarly showed correlations between microbiota present in pediatric respiratory samples and clinical markers such as CF genotype. This study also identified inverse correlations between community complexity and not only age but also presence of P. aeruginosa and antibiotic use. van der Gast et al. (2011) partitioned species detected in the adult CF lung into core and satellite species by using ecological statistical tools. Correlations were found here again with CF genotype, antibiotic use, and lung function. Causal relations cannot be inferred from these correlations, but they help build at very least important hypotheses for future study. Other authors have shown that the stage of disease is an important predictor of bacterial community complexity. Rudkjøbing et al. (2012) showed in end-stage disease that P. aeruginosa was the “sole pathogen.” This study also resolved the bacteria in samples by using a version of fluorescence in situ hybridization; this is an important means of studying the bacterial component of the microbiota.
Important questions remain about the stability of CF respiratory microbiota and how they change during periods of exacerbation and/or antibiotic treatment. Tunney et al. (2011) examined the impact of antibiotic therapy on patients with CF with samples collected at the start and end of treatment of exacerbation. The antibiotic treatment (targeted against aerobes) was found to have only a small impact on the abundance of anaerobic bacteria and the composition of the community. A recent study by Carmody et al. (2013) studied the differences in the bacterial component of the microbiota at times of exacerbation, a time of heightened pulmonary symptoms, and found that the changes in community structure observed were dependent on the complexity and composition before exacerbation was triggered. Over a longer term, our interpretation of the potential clinical importance of CF lung microbiota would be vastly different if the species detected were shown to be present over longer periods of time or alternatively only transiently. This question formed the focus of two recent studies. Stressmann et al. (2012) assessed the stability of the species present in sputa from adult patients over the course of a year. While exceptions were noted, the most common finding was that the species present were typically stable over this period when the patient was clinically stable. Given the repeated antibiotic therapies that these patients received, this was to some extent unexpected. Zhao et al. (2012) examined similar issues of bacterial community dynamics over an even longer time frame of up to 9 years. Here, bacterial community diversity was found to decrease over time with patients showing more progressive lung disease; antibiotic exposure was considered to be the prime reason for the observed decrease in diversity. As in Stressmann et al. (2012), Zhao et al. (2012) showed that the airway bacterial communities were resilient when challenged by antibiotic therapy, and the authors identified no particularly marked change in bacterial community at the time of onset of pulmonary exacerbation.
Directions of Research
The above overview demonstrates that the bacterial species component of the microbiota associated with the CF lung is complex; this renders generalization difficult. Nevertheless, an appraisal of these recent advances in CF microbiology highlights several important areas in which more research is needed. The explosion of information generated by these studies begs new ways to manage and analyze sequence data. In parallel with this expansion of information, the impact of these strategies is only likely to increase in this and other studies that focus on human-associated microbiota. Knowing what roles “nontraditional” species are playing is also important. In this effort, we may benefit greatly from conceptual and practical work that is emerging from the Human Microbiome Project.
As highlighted above, studies have already detected associations between particular components of the microbiota. These findings suggest the potential for microbial interspecies interactions, both positive and negative, in shaping the CF airway microbiome, as well as in driving CF lung disease. The consequences of interacting species on pathogenicity in model systems need to be addressed in much more detail. Such interspecies interactions could impact the efficacy of antibiotics targeting individual pathogens, such as P. aeruginosa, or on other outcomes. Studies linking microbiota with the host are needed. Many studies, by necessity, have focused on only one component of the microbiota present; so far, the focus has been on the bacteria comprising the CF airway microbiota. More is needed to broaden this scope by examining the relationships between host inflammation, host disease status, and the structure of the host airway microbiota as well.
We consider the longer-term goal of this field to be the rational design of interventional studies using the more complete picture of the microbiome at different stages in CF lung disease progression. To get the most benefit from this work, a firmer grasp of the link between molecular diagnostics and lung pathophysiology is needed in this and other respiratory conditions. So far though, it is clear that the bacterial species component of the microbiota of CF airway infections is complex and much more diverse than typically considered by conventional diagnostic microbiology. Many of the species detected by culture-independent means were first reported as detected in the CF airways as such with many species present requiring anaerobic conditions for growth. A series of practical, clinical, and ethical issues surround sampling and the analysis of respiratory microbiota. More studies are needed to address these issues though it is clear from existing studies that there are clear and strong associations between bacterial species and clinical parameters. Longitudinal studies are showing the impact of antibiotic use and longer-term dynamics of the bacteria present in these airway infections. Through these efforts, it is hoped that the next few years will witness the translation of our emerging understanding of the microbiomes of diseased and healthy airways into benefits for CF patients.
- Coenye T, Goris J, Spilker T, et al. Characterization of unusual bacteria isolated from respiratory secretions of cystic fibrosis patients and description of Inquilinus limosus gen. nov., sp. nov. J Clin Microbiol. 2002;40:2962–1069.Google Scholar
- Cystic Fibrosis Foundation. http://www.cff.org/AboutCF/ (2013). Accessed 23 Sept 2013.
- Rogers GB, Hart CA, Mason JR, et al. Bacterial diversity in cases of lung infection in cystic fibrosis patients: 16S ribosomal DNA (rDNA) length heterogeneity PCR and 16S rDNA terminal restriction fragment length polymorphism profiling. J Clin Microbiol. 2003;41:3548–458.PubMedCentralPubMedCrossRefGoogle Scholar
- Rogers GB, Skelton S, Serisier DJ, et al. Determining cystic fibrosis-affected lung microbiology: comparison of spontaneous and serially induced sputum samples by use of terminal restriction fragment length polymorphism profiling. J Clin Microbiol. 2010a;48:78–86.PubMedCentralPubMedCrossRefGoogle Scholar
- UK CF Registry. Annual data report 2012. https://www.cysticfibrosis.org.uk/media/31676/Scientific%20Registry%20Review%202012.pdf (2012). Accessed 20 Dec 2013.