Evaluating The Role of Neutrophil Extracellular Traps in Pseudomonas Aeruginosa Infection

Abstract

Introduction Chronic respiratory tract infection with Pseudomonas aeruginosa (Pa) is the predominant cause of death in people with cystic fibrosis (Sly et al.). CF is characterised by a progressive decline in lung function that is closely linked to chronic infection. The immune response to Pa in the lungs is predominantly driven by neutrophils, which typically fail to clear the infection, suggesting either neutrophil effector function is inhibited, or Pa immune evasion is more effective within the environment of the CF lung. Neutrophils kill pathogens through phagocytosis, degranulation or the expulsion of neutrophil extracellular traps (NETs). Excessive NETs can be toxic to the host. Whether NETosis during Pa infection is beneficial or detrimental in the context of the CF lung is currently unknown. Understanding the interaction between Pa and neutrophils in the early stages of lung infection is crucial to developing therapeutic interventions for CF lung infections. Aims The aim of this thesis was to investigate the role of NETs in Pa infection in CF, and in other acute and chronic infection settings, by screening a comprehensive library of Pa isolates that captures much of the genetic and phenotypic diversity of the species. To do this, I quantified the ability of Pa isolates to induce NETs and assessed phenotypic differences between mucoid and non-mucoid isolates, to identify specific factors responsible for inducing NETs. Finally, I aimed to determine how bacteria evade the host response by either direct toxicity towards neutrophils or through NETs degradation. Methods Human Peripheral blood neutrophils were isolated from healthy volunteers. NETs formation was visualised by fluorescent microscopy of extracellular DNA, following 4 hours incubation of Pa isolates (total; n=82, Respiratory clinical; n=47, non-respiratory clinical; n=32 and environmental; n=3) with neutrophils. To investigate whether Pa strains secreted factors that may induce NETs formation, bacterial culture supernatant was incubated with human peripheral blood neutrophils. Purified bacterial factors and mutants from a Pa transposon library were used to elucidate factors responsible for NETs induction. Bacterial colony-forming unit counts were used to assess the role of NETs in killing bacteria. Cellular toxicity of the Pa strains to neutrophils was determined by Annexin V FITC and propidium iodide (PI) uptake and quantified by flow cytometry. NETs degradation was assessed by incubating Pa isolates or supernatant, overnight, with pre-formed NETs (stimulated using PMA). To validate degradation data, Pa isolates were incubated with purified genomic DNA, the degradation of which was quantified using DNA gel electrophoresis. Results Initial screening demonstrated that 59/82 (72%) of the Pa isolates specifically induced NETs formation, while the other 23/82 (28%) did not. The ability of Pa isolates to induce NETs formation was associated with a clinical compared to an environmental origin. Assays performed with bacterial culture supernatant demonstrated that secreted factors from 50/82 (61%) Pa isolates specifically induced NETs formation, while the other 32/82 (39%) did not. These data highlight that induction of NETs formation by Pa is a strain-specific phenotype, and that secreted factors may have potential importance in the pathogenesis of Pa infections. To determine potential factors secreted by Pa that might drive NETs formation, we compared the ability of mucoid and non-mucoid strains to induce NETosis. Mucoid strains induced a higher degree of NETs formation, compared to non-mucoid strains. Specifically, 11/11 (100%) of the mucoid strains induced NETs formation, compared to 25/33 (76%) non-mucoid strains. When supernatants were tested, 91% of mucoid strains induced NETs formation, compared to 48% of non-mucoid strains, suggesting that the ability to form mucoid biofilms is related to increased NET induction and may contribute to increased virulence and pathogenicity. Alginate over-production is the key phenotypic difference between mucoid and non-mucoid isolates. Incubating purified bacterial alginate with neutrophils induced a dose-dependent increase in NETs formation. To confirm this observation, we used Pa (strain PAO1) with transposon insertions in algD. PAO1 lacking a functional algD gene demonstrated significantly (PW6997 and PW6998) reduced NETs formation compared to the parental strain PAO1. We next compared IST27 mucoid and IST27N (an spontaneous non-mucoid revertant that is derived from IST27) isolates and found that we detected alginate in the supernatant of IST27 mucoid, which was undetectable in IST27N. Both IST27 mucoid live bacteria and supernatant taken from cultures of this isolate induced moderate to strong NETs formation, which was not observed using IST27N. To determine whether other exopolysaccharides may contribute to the pathogenicity of Pa the context of NETs formation, we investigated whether Psl- or Pel-deficient PAO1 might also show attenuated NETs induction (PW4798 pslB, PW4801 pslC, PW8402 pslD, PW4803 pslD, PW6130 pelF). All mutants tested affected the ability of Pa to induce NETosis demonstrating that both Psl and Pel deficient mutants had a significant reduction in their ability to stimulate NETs formation compared to the parental PAO1 strain. To determine whether certain Pa isolates have adapted to evade the host response, we investigated whether incubation with neutrophils resulted in the killing of Pa isolates, whether their growth was limited or whether they grew. To do this, we performed bacterial colony counts before and after 4 hours incubation with neutrophils. These data demonstrated that only 3/82 strains grew in the presence of neutrophils (4%), 23/82 showed no consistent change in bacterial numbers (28%) and 56/82 strains were specifically killed (68%) in the presence of neutrophils. To determine the potential mechanism Pa employ to evade killing by neutrophils, we determined whether certain Pa strains are cytotoxic to neutrophils. We found that of the 23/82 that had not induced NETs-formation in our original screen, 10/23 were cytotoxic (43%) to neutrophils, whilst the remaining 13/23 had no effect on neutrophil viability. Moreover, 11/23 (48%) of these strains induced significant NETs degradation of preformed NETs, while the other 13/23 did not. We found that the ability of Pa to degrade NETs is more common in isolates taken from respiratory infections than those from non-respiratory infection sources. This could be due to the fact that Pa chronically colonising the airways are more likely to be exposed to NETs. Finally, to investigate whether the degradation of NETs could be inhibited to prevent the potential of bacterial dissemination, we used CecI3, which converts B-DNA to Z-DNA, making DNA more resistant to degradation. These data demonstrated that 11/11 of the Pa strains that previously degraded NETs could no longer do so in the presence of CecI3, suggesting that the conversion from B-DNA to Z-DNA inhibited the ability of bacteria to degrade NETs. Conclusions These data are the first to comprehensively characterize the ability of Pa from different sources to induce NETs formation. This thesis provides novel mechanistic insights into the mechanism of NETs induction by Pa and specifically identifies alginate secretion as a major driver of NETosis. Finally, we provide evidence of how Pa may evade the host response through direct cytotoxicity and NETs degradation. This thesis highlights the potential for personalised medicine in patients with Pa infections to target strain-specific phenotypes with an aim of improving NETs-related pathologies.