Eilidh Mowat

Pseudomonas aeruginosa Population Diversity and Turnover in Cystic Fibrosis Chronic Infections

RATIONALE Pseudomonas aeruginosa isolates from chronic cystic fibrosis lung infections display multiple phenotypes indicating extensive population diversity. OBJECTIVES We aimed to examine how such diversity is distributed within and between patients, and to study the dynamics of single-strain phenotypic diversity in multiple patients through time. METHODS Sets of 40 P. aeruginosa isolates per sputum samples were analyzed for a series of phenotypic and genotypic characteristics. Population differentiation between patients, between samples within patients, and between isolates within samples was analyzed. MEASUREMENTS AND MAIN RESULTS We characterized 15 traits for a total of 1,720 isolates of an important and widely disseminated epidemic strain of P. aeruginosa from 10 chronically infected patients with cystic fibrosis multiply sampled during 2009. Overall, 43 sputum samples were analyzed and 398 haplotypes of the Liverpool Epidemic Strain were identified. The majority of phenotypic diversity occurred within patients. Such diversity is highly dynamic, displaying rapid turnover of haplotypes through time. P. aeruginosa populations within each individual sputum sample harbored extensive diversity. Although we observed major changes in the haplotype composition within patients between samples taken at intervals of several months, the compositions varied much less during exacerbation periods, despite the use of intravenous antibiotics. Our data also highlight a correlation between periods of pulmonary exacerbation and the overproduction of pyocyanin, a quorum sensing-controlled virulence factor. CONCLUSIONS These results significantly advance our understanding of the within-host population biology of P. aeruginosa during infection of patients with cystic fibrosis, and provide in vivo evidence for a link between pyocyanin production and patient morbidity.

Fluctuations in phenotypes and genotypes within populations of Pseudomonas aeruginosa in the cystic fibrosis lung during pulmonary exacerbations

Chronic respiratory infection by Pseudomonas aeruginosa contributes significantly to the morbidity and mortality associated with cystic fibrosis (CF). Using a series of phenotypic and genotypic tests on collections of 40 isolates per sputum sample, we analysed fluctuations within sputum populations of the P. aeruginosa Liverpool epidemic strain (LES) during pulmonary exacerbations. For each of three patients, three sequential sputum samples were analysed: (1) on presentation with exacerbation at the Regional Adult Cystic Fibrosis Unit, Liverpool; (2) a few days into intravenous antibiotic treatment; (3) when the patient had recovered. Fluctuations were observed in morphotype distribution, the production of virulence-associated quorum-sensing-dependent exoproducts (the phenazine compound pyocyanin and the elastase LasA), antibiotic susceptibility profiles and levels of auxotrophy. PCR assays were used to screen isolates for the presence of novel regions of the LES genome (islands and prophages) and to detect free phages. In one patient there was an increase in the prevalence of the LESGI-5 genomic island during the sampling period from 10 to 97.5 % carriage. LES phages 2-4 were detected in either the majority or all sputum samples tested, indicating widespread phage activity during the sampling period. The results of this study are indicative that significant fluctuations occur within P. aeruginosa populations during short periods of pulmonary exacerbation and intravenous antibiotic therapy.

Use of Artificial Sputum Medium to Test Antibiotic Efficacy Against Pseudomonas aeruginosa in Conditions More Relevant to the Cystic Fibrosis Lung

There is growing concern about the relevance of in vitro antimicrobial susceptibility tests when applied to isolates of P. aeruginosa from cystic fibrosis (CF) patients. Existing methods rely on single or a few isolates grown aerobically and planktonically. Predetermined cut-offs are used to define whether the bacteria are sensitive or resistant to any given antibiotic1. However, during chronic lung infections in CF, P. aeruginosa populations exist in biofilms and there is evidence that the environment is largely microaerophilic2. The stark difference in conditions between bacteria in the lung and those during diagnostic testing has called into question the reliability and even relevance of these tests3. Artificial sputum medium (ASM) is a culture medium containing the components of CF patient sputum, including amino acids, mucin and free DNA. P. aeruginosa growth in ASM mimics growth during CF infections, with the formation of self-aggregating biofilm structures and population divergence4,5,6. The aim of this study was to develop a microtitre-plate assay to study antimicrobial susceptibility of P. aeruginosa based on growth in ASM, which is applicable to both microaerophilic and aerobic conditions. An ASM assay was developed in a microtitre plate format. P. aeruginosa biofilms were allowed to develop for 3 days prior to incubation with antimicrobial agents at different concentrations for 24 hours. After biofilm disruption, cell viability was measured by staining with resazurin. This assay was used to ascertain the sessile cell minimum inhibitory concentration (SMIC) of tobramycin for 15 different P. aeruginosa isolates under aerobic and microaerophilic conditions and SMIC values were compared to those obtained with standard broth growth. Whilst there was some evidence for increased MIC values for isolates grown in ASM when compared to their planktonic counterparts, the biggest differences were found with bacteria tested in microaerophilic conditions, which showed a much increased resistance up to a >128 fold, towards tobramycin in the ASM system when compared to assays carried out in aerobic conditions. The lack of association between current susceptibility testing methods and clinical outcome has questioned the validity of current methods3. Several in vitro models have been used previously to study P. aeruginosa biofilms7, 8. However, these methods rely on surface attached biofilms, whereas the ASM biofilms resemble those observed in the CF lung9 . In addition, reduced oxygen concentration in the mucus has been shown to alter the behavior of P. aeruginosa2 and affect antibiotic susceptibility10. Therefore using ASM under microaerophilic conditions may provide a more realistic environment in which to study antimicrobial susceptibility.

Extensive diversification is a common feature of Pseudomonas aeruginosa populations during respiratory infections in cystic fibrosis

Background Populations of the Liverpool Epidemic Strain (LES) of Pseudomonas aeruginosa undergo extensive diversification in the cystic fibrosis (CF) lung during long-term chronic infections. Methods We analyzed sets of 40 isolates from the sputa of five CF patients, each chronically infected with a different non-LES strain of P. aeruginosa. For each sample (two per patient), diversity was assessed by characterizing nine phenotypic traits. Results All P. aeruginosa populations were highly diverse, with the majority of phenotypic variation being due to within-sample diversity. Conclusions Maintenance of diverse populations in the CF lung is a common feature of P. aeruginosa infections.

Effect of antibiotic treatment on bacteriophage production by a cystic fibrosis epidemic strain of Pseudomonas aeruginosa.

ABSTRACT Phage production in response to antibiotics varied among four isolates of a Pseudomonas aeruginosa cystic fibrosis (CF) epidemic strain. Whereas ciprofloxacin induced higher levels of phage production, other CF-relevant antibiotics led to reduced production. We detected free phages directly in CF patient sputum samples by both plaque (40% positive) and PCR (76% positive) assays. Our observations suggest that the choice of antibiotics could influence the number of free phages within the CF lung environment.

P108 Phenotypic changes in Pseudomonas aeruginosa (Psa) populations during exacerbations in adult Cystic Fibrosis patients infected with non-epidemic strains

Background Chronic infection with Psa in most CF patients is due to single unique strains, which over time accumulate mutations resulting in mucoid conversion, loss of motility, auxotrophy and increased antibiotic resistance, in turn leading to multiple phenotypes. However, changes in Psa population morphology related to short term pressures (eg, during IV antibiotic-treated exacerbations), have not previously been studied. We therefore looked at Psa population structure during exacerbations. Methods Sputum samples from four CF patients chronically infected (for at least 4 years) with unique single Psa strains were analysed at the beginning and end of an intravenous antibiotic-treated exacerbation, where every patient had subjective and spirometric improvement. From each sample, 40 single Psa colonies were selected (with every morphological type proportionately represented), and colony morphology, susceptibility to six antibiotics (ciprofloxacin, ceftazidime, colomycin, meropenem, tobramycin, tazocin), hypermutability (rate of spontaneous mutation to rifampicin resistance) and auxotrophy (ability to grow on glucose M9 media) determined. Additionally, the density of Psa (colony forming units (CFU) per ml) in sputum samples was measured. Results Although the predominant colony morphology changed from green, non-mucoid and smooth (mean 67%, range 43–98) to straw-coloured, non-mucoid and smooth (57%, 5–95) (p=0.001), there was no change in mean antibiotic resistance to all antibiotics (21.5% vs 20.3%, p=0.9), prevalence of hypermutable isolates (38% vs 25%, p=0.48) and auxotrophic mutants (66% vs 98%, p=0.17). However, there was an increase in Psa sputum density (mean CFU/ml 1.3 x105 vs 2.0x106, p<0.001), despite the use of relevant antimicrobial therapy. Conclusion The changes in prevalent population composition following antibiotic pressure, associated with clinical improvement, might suggest that some morphotypes alone are responsible for the adverse clinical features. Conversely, the increase in sputum density of Psa despite objective clinical improvement implies that the exacerbation has occurred independently of the presence of the organism, supporting the observation that clinical improvement is often seen in CF patients even where the Psa seems resistant to the administered antibiotics. Further work need to be done to tease out the role of Psa in clinical exacerbations of CF patients with chronic infection.

P96 Haplotype variability in Pseudomonas aeruginosa (Psa) strains in the adult CF population

Introduction Morphological type variation in Psa is thought to aid its survival in hostile environments, including the CF lung. Previous studies using PGFE have shown wide heterogeneity in LES strain, despite this, little is known about their phenotypic diversity (haplotypes). To look at this further, we compared the haplotypes of the most successful and prevalent UK transmissible Psa (the Liverpool epidemic strain, LES) with those of sporadic strains and assessed their response to antibiotic pressure. Methods Sputum from two groups of adult CF patients chronically infected (for at least 4 years) with sporadic single Psa strains or LES was analysed at the beginning and end of an intravenous antibiotic-treated exacerbation, where every patient had subjective and spirometric improvement, and also for LES patients over a period of time (at least 4 months) when they were in a stable state. From each sample, 40 single Psa colonies were selected (with every morphological type proportionately represented), and colony morphology, susceptibility to six antibiotics (ciprofloxacin, ceftazidime, colomycin, meropenem, tobramycin, piperacillin/tazobactam), hypermutability (rate of spontaneous mutation to rifampicin resistance) and auxotrophy (ability to grow on glucose M9 media) determined. Each Psa colony could potentially exhibit any of nine combinations: antibiotic susceptibility patterns, auxotrophy, hypermutability, and six colony morphologies. The resulting haplotypes were compared for LES and unique strains using ‘e-burst’ software. Results Overall, 151 unique haplotypes were defined (96 LES, 75 sporadic, 20 shared), with LES demonstrating the greater number (X2=4.67, p<0.03). Following antibiotic pressure, there was an alteration in haplotypes which was similar in both groups (LES 54%, sporadic 64%, p=0.28): but during the stable state LES demonstrated a significantly greater change in haplotypes (89%, p=0.004). Conclusion A change in haplotypes occurs in all CF Psa strains during treatment for clinical exacerbations, presumably as a defence mechanism against the challenge of hostile antibiotic therapy. However, the enhanced haplotype variability of transmissible strains with time when there is no antibiotic pressure may be a strategy that confers a survival advantage compared to sporadic strains: studies are under way to explore this further.

P98 Accuracy of routine antibiotic susceptibility testing of sputum samples in adult cystic fibrosis (CF) patients colonised with Pseudomonas aeruginosa (Psa)

Background In CF patients with chronic Psa infection, the role of conventional antibiotic susceptibility testing (antibiotic disc diffusion on sputum subculture morphotypes according to BSAC protocols) is controversial, but many centres still rely on these methods when selecting treatment. However, morphotypes may exhibit up to six antibiograms, and this may contribute to the variable clinical response often noted. To look at this further, we performed additional Psa susceptibility testing on 26 sputum samples from 10 chronically infected CF patients provided over a 15-month period, and compared the results with those from the routine laboratory. Methods For each sputum sample, 40 colonies proportionately representative of morphological subtypes (mean 2, range 1–4) on the Psa selective plates were cultured onto Columbia plates. From these, single colonies were mixed with sterile distilled water to attain a standard optical density (10 MacFarland units), and 10 μl spread onto iso-sense plates and incubated overnight with tobramycin, meropenem, colomycin, ceftazidime, ciprofloxacin, and piperacillin/tazobactam antibiotic discs. Antibiotic sensitivity (break point 50% of isolates) was determined by the zone of inhibition as per standardised BSAC protocols. In total, 6240 analyses were performed. Results Although there was 100% concordance with the number of morphological subtypes of Psa with the routine laboratory, on multiple antibiotic susceptibility testing in 260 cases (25%) increased resistance was discovered. Overall, mean concordance between the routine diagnostic lab methodology and multiple antibiotic sensitivity testing was 70% (median 80%, IQR 60–100). However, in 15% of cases concordance was <50%, suggesting that more detailed testing may have altered the choice of antibiotics used. Conclusion This study shows that routine microbiological methodology may under-represent antimicrobial resistance in Psa when patients are chronically infected. This may in part explain the clinical experience, and underlines the need for better microbiological techniques to aid the clinician in caring for these complex patients.