Search

Circulating neutrophil-derived extracellular vesicles (EVs) are acutely increased during sepsis and systemic inflammatory response syndrome. Although in vitro studies have shown that neutrophil-EVs have pro-inflammatory activities, little is known about their roles in the propagation of systemic inflammation. Recent studies indicate that the uptake of circulating EVs by the lungs increases dramatically during systemic inflammation, primarily through their interactions with pulmonary intravascular monocytes. In this study, we hypothesised that circulating neutrophil-EVs are potent long-range mediators of pulmonary vascular inflammation, contributing to the development of indirect acute lung injury. Using human cell culture-based models, my main aims were to: 1) characterise neutrophil-EV uptake by pulmonary vascular cells under resting and inflammatory conditions, 2) develop a whole blood model of lipopolysaccharide (LPS)-induced neutrophil-EV subtype production, 3) isolate and characterise the pro-inflammatory activity of these neutrophil-EVs. For the assessment of neutrophil-EV function, I developed a co-culture model of pulmonary vascular inflammation, consisting of peripheral blood mononuclear cells (PMBCs) or monocytes, co-cultured with human lung microvascular endothelial cells (HLMECs). Neutrophil-EVs were taken up by both HLMECs and monocytes and subject to dynamic changes under physiological flow and inflammatory conditions. LPS stimulation of whole blood revealed acute increases in neutrophil- and platelet-EVs. Using immunoaffinity isolation for EV subpopulations, neutrophil- but not platelet-EVs were found to be pro-inflammatory, inducing TNF-α-dependent activation of HLMECs in the presence of PBMCs or monocytes. Further investigations revealed that neutrophil-EVs activate monocytes via Toll-like receptor 4 (TLR4) signalling, in an EV surface protein-dependent manner. These findings indicate that neutrophil-EVs released under septic-like conditions in vitro are potent mediators of inflammation, with the potential to generate localised inflammation within 3 the monocyte-enriched environment of the pulmonary vasculature. Furthermore, the neutrophil-EV protein-dependent TLR4 signalling activity suggests a novel mechanism for propagation of inflammation within the circulation from local sites of infection.

Patients with acute respiratory distress syndrome (ARDS) have an unacceptably high mortality surpassing 40%. Most patients with ARDS die from multi-organ failure rather than respiratory failure, presumably due to systemic propagation of inflammation between the lungs and peripheral organs. Numerous systemic inflammatory mediators have been identified as potential therapeutic targets for ARDS. However, few specific therapies have been successfully developed and treatment remains primarily limited to supportive care. Microvesicles (MVs) are tiny particles released from cells during cell activation and stress. MVs represent a unique form of inter-cellular communication through their ability to act as 'biological ferries' carrying inflammatory mediators as their cargoes in a packaged, lipid-encapsulated environment resistant to neutralisation in the systemic circulation. We hypothesised that leukocyte-derived MVs play a critical role in systemic inflammatory signalling in ARDS. Through initially focusing on periphery-to-lung signalling, we developed an in vitro bioassay of pulmonary microvascular inflammation and investigated the bioactivity of neutrophil-derived MVs. We found that inflammatory neutrophil-derived MVs induced significant acute pulmonary microvascular inflammation, in a peripheral blood mononuclear cell-dependent manner. We then focused on lung-to-periphery signalling through developing a novel human in vitro model of ventilator-induced pulmonary microvascular inflammation. We found, for the first time, that pathological cyclic stretch induced acute leukocyte activation and release of leukocyte-derived MVs. We developed methodology to separate MV subtypes from the mixed total MV population generated, and demonstrated that these stretch-induced leukocyte-derived MVs are capable of inducing remote organ endothelial injury. Finally, we investigated circulating MV profiles in plasma samples of patients with ARDS. We found significantly increased baseline neutrophil- and endothelial-derived MV levels in patients with hyperinflammatory ARDS. Our findings suggest MVs play a crucial role in the early phase of ARDS pathophysiology and could represent a paradigm shift in our understanding of systemic inflammatory signalling in ARDS.

The National Health Service faces many challenges including the preservation of a healthy workforce and prevention of medical error. Psychological Safety, defined as an environment "safe for interpersonal risk taking" can help mitigate these challenges. This environment, where individuals feel safe to "speak up" can promote organisational safety and individual wellbeing. "Stress" is also implicated in the aetiology of error and low job satisfaction, justifying strategies to minimise its occurrence. This Thesis aims to describe key and previously undescribed features regarding the manifestation of psychological safety and stress in healthcare workers. Existing data regarding psychological safety were explored through a comprehensive evidence synthesis. Three mixed methods studies encompassing qualitative interviews and quantitative surveys were subsequently conducted. These explored the aetiology, presence, impact and sequelae of psychological safety and stress in critical care workers, emergency department staff and hospital managers. Data provided a detailed understanding of the complexity of the workplace environment from an individual worker's perspective. A reassuring climate of psychological safety was demonstrated in clinical participants. This was lower for managers, highlighting how this environment can differ within the same organisation. Data analyses yielded novel information regarding the influence of situational context and personality on psychological safety and demonstrated the potential for negative consequences of such an environment. Both 'typical' stressors and the COVID-19 pandemic were evaluated, highlighting that whilst generally detrimental, challenging events could create positive sequelae including enhanced camaraderie. Models were developed highlighting key factors within the workplace that can be improved to enhance psychological safety and mitigate stress. It is important for staff wellbeing and organisational safety that a climate of psychological safety is cultivated, and that stresses are mitigated and supported. Healthcare leaders can use the models presented in this Thesis to identify modifiable areas for change and implement solutions within their own workplace environments.

Acute lung injury (ALI) resulting from remote 'indirect' causes is a major problem in sepsis and systemic inflammatory response syndrome (SIRS) but the underlying mechanisms are poorly understood. Circulating microvesicles (MVs) have been implicated as long-range mediators of vascular inflammation and their role as biomarkers in sepsis and SIRS has been widely investigated in clinical studies in recent years. However, the in vivo functional roles of MVs in sepsis and ALI have received less attention. Specifically, the role of in vivo MVs in the development of sepsis/SIRS-induced indirect ALI has not been previously evaluated. We hypothesised that circulating microvesicles (MVs) play a crucial role in propagating inflammation to the lungs, contributing to the development of pulmonary vascular inflammation in indirect ALI. The overall aims of this project were to: 1) evaluate MV uptake by pulmonary vascular cells and the mechanisms involved, 2) characterise the intravascular production of MVs in animal models of sepsis and sterile extrapulmonary organ injury, and 3) identify the contribution of in vivo-derived circulating MVs to the development of indirect ALI. The major findings of this work were that during sub-clinical endotoxaemia in mice, lung-marginated Ly6Chigh monocytes become a major target for circulating MV uptake via a phosphatidylserine receptor mechanism1. In mouse models of sepsis and extrapulmonary organ injury, neutrophil- and monocyte-derived MVs were the predominant MV subtypes being produced during endotoxaemia, while platelet- and endothelial-derived MVs were predominant during kidney ischaemia reperfusion injury. When MVs obtained from plasmas of endotoxaemic mice were adoptively transferred to isolated perfused lungs (IPLs), they induced significant increases in lung oedema. Depletion of intravascular lung monocytes by treatment with clodronate liposomes resulted in the reversal of the oedema, demonstrating the role of monocytes in MV-induced ALI. To investigate the contribution of different circulating MV subtypes, we immunoaffinity isolated myeloid (CD11b+) and platelet (CD41+) MVs from endotoxaemic mouse plasmas and transferred these to the IPL. We found that myeloid-MVs induced significant lung oedema and potent release of soluble mediators, whereas platelet-MVs produced a statistically significant, but much lower level of oedema and negligible release of soluble mediators. In summary, these findings indicate an important role of myeloid-derived MVs, particularly those derived from neutrophils and/or monocytes, and their interaction with lung-marginated monocytes in the pathogenesis of pulmonary vascular inflammation in indirect ALI. Further work to elucidate the specific MV molecular effectors mechanism involved will facilitate an enhanced understanding of ALI pathobiology.

Acute lung injury (ALI) and its clinical presentation Acute Respiratory Distress Syndrome (ARDS) has an unacceptably high mortality, surpassing 40% in those with the severest form of the disease. Despite intensive research activity and the identification of a vast number of biological mediators, all of which have been promoted as apparently vital to the pathogenesis of ALI/ARDS, treatment remains principally supportive. Therefore, a re-direction in ALI research is required, moving away from highlighting individual inflammatory mediators and instead investigating how inflammatory cargo are transmitted between cells. Microvesicles (MVs) are membrane-circumscribed extracellular particles of 100-1000nm in size and are derived from eukaryotic cells following direct damage, apoptosis or stimulation. MVs provide an alternative yet crucial role in intercellular communication by carrying a variety of bioactive cellular cargo and have been implicated in the pathophysiology of various inflammatory diseases. Yet, their role in acute lung injury (ALI) remains unknown. We hypothesise that MVs are key to the pathogenesis of ALI as mediators of intercellular communication and our aim in this thesis was to assess the role of MVs in ALI. We demonstrated that there is a dynamic production of MVs from multiple intra-alveolar cells very early during the inflammatory course of ALI with a substantial increase in MVs originating from alveolar macrophages. When instilled into the lungs of mice, these alveolar macrophage-derived MVs significantly increased the parameters of ALI tested, providing compelling evidence for their role in the pathophysiology of ARDS. We found that these MVs packaged TNF initiating inflammation via a TNF dependent mechanism. Mechanistically, we found that TNF is packaged within MVs, in a similar fashion to IL-1β, via an unconventional, ER/Golgi-independent route, despite being a classical cytokine (Golgi-dependant). Using pharmacological and genetic inhibition techniques, we demonstrated that this switch in cellular TNF trafficking is mediated via acid sphingomyelinase. . Furthermore, we established that the TNF within MVs, is the transmembrane TNF isoform (rather than soluble isoform) and localises to the MV membrane in a stable manner, activating distant target cells without cell-to-cell contact per se. Therefore we identified an entirely novel way in which membrane TNF signals in vivo. Next we examined how MVs communicate within the alveolar space. We showed that although both epithelial cells and alveolar macrophages take up MVs, alveolar macrophages take up the overwhelming majority, regardless of the environmental condition. We demonstrate that epithelial cell uptake is dependent on integrin/phosphatidylserine binding while alveolar macrophage uptake is more dependent on scavenger receptors indicating clear mechanistic differences between these cell types. Our results suggested two potential scenarios regarding MV-mediated communication within the alveolus: 1) Activation of epithelial cells may be due to TNF enclosed within MVs, which binds to TNF receptors on the surface of epithelial cells (rather than internalisation_; or 2) ICAM-1 upregulation may be assisted by alveolar macrophages, potentially by 'passing' on proinflammatory material to epithelial cells. We consolidated and translated the findings from our in vitro and in vivo models of ALI/ARDS by characterising different MV populations within 30 patients with another inflammatory lung disease, Chronic Obstructive Pulmonary Disease (COPD). We confirmed the presence of multiple MV populations within bronchoalveolar lavage fluid (BALF) taken from our patients (e.g. leukocyte, neutrophil, monocyte, alveolar macrophage, epithelial). We assessed whether these BALF MV populations could be potential biomarkers for COPD and found that BALF neutrophils strongly correlated with a broad range of indices of disease severity including symptoms (MRC dyspnoea score, St George's Respiratory Questionnaire); exercise tolerance (6 minute walk test); disease severity (FEV1); and mortality (BODE index). Taken together, our findings support a role for MVs in the pathogenesis of ALI and lung inflammation. Furthermore MVs are biomarker of lung inflammation that could be used in clinical medicine. Therefore modulation of MVs within the alveolus may be a promising therapeutic target and our studies may open novel diagnostic and therapeutic perspectives in patients with ARDS.

Acute respiratory distress syndrome (ARDS) is a major cause of intensive care mortality, with no effective treatment. Despite improvements in patient management in intensive care, ARDS mortality remains high, therefore there is significant unmet need for active therapies. Tumour necrosis factor (TNF) is a major inflammatory cytokine that has long been implicated in ARDS pathology. TNF signals through two related but functionally distinct receptors, TNF receptor 1 and TNF receptor 2 (TNFR1 & TNFR2). Previous work has shown that signalling through TNFR1 is deleterious in ARDS caused by ventilation and acid aspiration, whereas signalling through TNFR2 is protective. These data suggest that specific inhibition of TNFR1 may be protective in ARDS. In this thesis I investigate the effects of specific TNFR1 inhibition, by a novel TNFR1 domain antibody (dAb™), in two different models of LPS induced ARDS. Firstly, I showed that TNFR1 inhibition reduces inflammatory cell influx, and macrophage activation when given prior to disease induction. Despite these benefits, physiological parameters remained unaffected. In our second model I combined the LPS stimulus with mechanical ventilation once animals had significant pre-existing injury. Mechanical ventilation is essential for the survival of human ARDS patients, however it can cause iatrogenic ventilator induced lung injury (VILI). In this study I investigated whether specific blockade of intra-alveolar TNFR1 could protect mice with established ARDS from VILI. I showed that administration of the TNFR1 dAb, immediately prior to ventilation, resulted in significant improvements in physiology, alongside significant changes in alveolar macrophage responses. These data provide the first evidence that TNFR1 inhibition can provide protection, in the face of significant pre-existing injury, strongly supporting the further investigation of this approach in the clinic. While TNF signalling via TNFR1 has been well characterised, the signalling pathways associated with TNFR2 are less clear. In the final section of this thesis I focus on the role of TNFR2 in macrophages, and demonstrate the effects of TNFR2 signalling on TNFR1 pathways as well as identifying some potential TNFR2 specific pathways for the first time.

Burn injury-associated pain represents a considerable burden to patients and is inadequately managed in the clinical setting. It is thought to be a consequence of the “burn-injury inflammatory soup” produced in the wound that activates and sensitises peripheral nociceptors. However the composition of this fluid and its interaction with peripheral nociceptors in incompletely defined, limiting the application of a mechanistic-based approach to burn injury-associated pain management. Two human experimental burn injury pain models (thermal injury and UVB injury) represent a potential opportunity to further elucidate these complex peripheral processes. Nonetheless controversy exists regarding their respective somatosensory phenotypes and the associated peripheral inflammatory protein and metabolic profiles remain uncharacterised. This thesis aimed to elucidate a more comprehensive understanding of the somatosensory changes and components of the “burn-injury inflammatory soup” with respect to inflammatory proteins and the metabolome. Forty-five healthy subjects had an experimental pain model imparted on their forearm (thermal injury, capsaicin injury or UVB injury models). Somatosensory testing and dermal microdialysis were performed. A multiplex microbead assay array was used to analyse dermal microdialysates to compare temporal within and between injury model inflammatory protein concentrations. Untargeted ultra performance liquid chromatography-mass spectrometry was used for global profiling of dermal microdialysates with the aim of determining significant compounds that separated injury models. Although some similarities were apparent in the three injury models with respect to secondary somatosensory changes and concentrations of the inflammatory proteins IL6, IL8 and CCL2, overall the three experimental pain models displayed distinct somatosensory, inflammatory protein and metabolic profiles. The UVB injury model appears to represent a less acute inflammatory reaction than is seen in the thermal injury model. Together these results support the use of the thermal injury experimental pain model to further investigate peripheral processes involved in acute burn injury-associated pain.

Burn injuries are the fourth most common type of major trauma and responsible for approximately 300,000 deaths worldwide. Most of these deaths are ascribed to the systemic inflammation and multiple organ failure that commonly manifests following burn injury. The ideology that systemic inflammatory response syndrome develops secondary to the action of systemic cytokines has led to clinical trials aimed at intercepting these blood-borne inflammatory mediators before reaching their target sites, these efforts were however in vain. Microvesicles (MVs) are cell-derived signalling particles emerging as important mediators of local and systemic inflammation. Our objective was to investigate the role of microvesicles in the propagation of inflammation following burn injury, focussing on their induction in skin by extracellular ATP signalling. In vitro models of cell injury demonstrated that high temperature causes the release and stabilisation of ATP, in contrast to ATP degradation following mechanically-induced cell lysis. Treatment of skin explants ex vivo with ATP resulted in microvesicle release, which were macrophage-derived, capable of stimulating inflammatory responses in vitro and packaged with pro-IL-1β. Burn injury to skin explants via scalding (steam or immersion) also resulted in the release of microvesicles that were pro-inflammatory in vitro and following intravenous injection into mice, induced leukocyte activation and margination in the vascular beds of the lung and liver. Using a partial-width contact burn injury in skin explants, we were able to demonstrate ATP-purinergic receptor mediated microvesicle release. In conclusion, the results here demonstrate that burn injury can elicit a significant release of microvesicles from skin tissue and that ATP release from injured cells could play a central role in this process. The stabilisation of extracellular ATP by burn injury would produce a unique pathophysiological environment for microvesicle production, which, if released systemically, could play an important and disproportionate role in promoting systemic inflammation following severe burn injury as compared to other SIRS aetiologies.

Background: In populations working under heat stress, Exertional Heat Illness (EHI) may prove fatal. High core body temperature (Tc) and circulatory insufficiency are implicated. A surrogate for heat strain (e.g. elevated Tc, heart rate) could improve diagnosis, or reflect susceptibility. The glycopeptide copeptin is released during regulation of blood osmolality and volume; 'cut-off' levels (e.g. >20 pmol.L-1) may aid diagnosis and prognosis in a variety of medical disorders. Aims: (1) Characterise EHI risk in UK military personnel (2) Evaluate plasma copeptin responses to heat acclimatisation (3) Assess whether copeptin reflects exertional heat strain. Studies: (1) Epidemiology. i. Analysis of military notifications identified incomplete EHI surveillance. ii. Questionnaire respondents in a hot training environment appeared at low individual risk of EHI, but reported inadequate acclimatisation time. iii. Case reporting review highlighted the absence of recognised risk factors in a significant proportion of EHI casualties and increased hospitalisation risk in supposedly 'low risk' states. (2) Laboratory. Serial heat-exercise tests were conducted in the UK and during acclimatisation to a hot environment (Cyprus). In the UK, post-exercise copeptin >20 pmol.L-1 was associated with higher Tc and adrenocortical responses versus <20 pmol.L-1. With acclimatisation, post-exercise copeptin >20 was associated with higher heart rate and adrenocortical responses. (3) Field. During initial heat acclimatisation under moderate heat stress (Kenya), higher copeptin reflected greater Tc responses. In acclimatised personnel subject to more severe heat stress (Cyprus), higher copeptin associated with increased heart rate. Conclusions: In military personnel, heat illness is under-reported and the time required for heat adaptation may be under-appreciated. Occult susceptibility is common and has potential to undermine preventive practices. Copeptin did not reflect group changes in Tc with acclimatisation, but copeptin >20 pmol.L-1 associated with increased heat strain in all settings. The diagnostic/prognostic value of copeptin should be investigated further in suspected EHI cases.

The host response to a severe burn injury is characterised by exaggerated systemic inflammation. Early clinical manifestations include shock, respiratory failure, renal failure and immunosuppression. The signalling pathways that propagate the inflammatory response are unclear but have traditionally been thought to involve overspill of proinflammatory cytokines. The importance of microvesicles, sub-cellular membrane-bound particles, is increasingly being recognised in the context of intercellular communication. Although circulating microvesicles are elevated in proinflammatory states such as sepsis, their relevance to the post-burn inflammatory response has not previously been evaluated. We hypothesised that circulating microvesicles play a crucial role in propagating the post-burn inflammatory response. Our overall aims were to 1) optimise protocols for the processing and analysis of plasma samples for microvesicle content; 2) characterise the circulating microvesicle profile associated with severe burn injury; 3) develop in vitro techniques to assess microvesicle production and function. The major findings of this work were that microvesicles derived from leukocytes, neutrophils, monocytes and endothelial cells were significantly elevated within 24 hours of burn injury. Microvesicle levels fell rapidly and were significantly decreased by day two post-injury. Total leukocyte- and neutrophil-derived microvesicles were significantly higher in non-survivors of burn injury as compared to survivors. In vitro studies demonstrated that neutrophil microvesicle release could be elicited by incubation with opsonised heat-killed cells. Pilot analysis of burn patient samples, using an endothelial co-culture assay, suggested that microvesicles may regulate the innate immune response to burn injury. These findings indicated that circulating microvesicles are an important component of the post-burn inflammatory response. Their precise activity is likely to be subtype-specific but the association of neutrophil-derived microvesicles with patient outcome alludes to a key role in burn pathophysiology.

Mechanical ventilation is a life-saving intervention, but through exacerbating inflammation, it can contribute to mortality, known as ventilator-induced lung injury (VILI). Patients with acute respiratory distress syndrome (ARDS) are particularly susceptible to VILI, however, it is recognised that over-distension of the lung should also be avoided in patients with healthy lungs, whom are also at risk of VILI. It is important to minimise the adverse effects of mechanical ventilation, yet our understanding of how over-distension promotes the devastating inflammatory response in VILI remains poor. In this project, the kinetics of cellular activation were investigated within the lungs of mice following short periods of high tidal volume ventilation using phospho-flow cytometry. The results revealed that the alveolar epithelium is activated prior to alveolar macrophages, suggesting that the epithelium could be responsible for stimulating their activation. The activation observed was very rapid, leading me to hypothesise that communication between these two cell types could either occur via the release of soluble mediators, or through a direct-contact dependent mechanism. Subsequent experiments led to the theory that alveolar macrophages may become activated in response to stretch through the detection of the damage-associated molecular pattern ATP, which is rapidly released by the alveolar epithelium in response to cyclic stretching. These experiments also revealed that pulmonary endothelial cells may be the most sensitive pulmonary cell type to mechanical ventilation in vivo, indicating that stretch-induced inflammation could be immediately disseminated to the systemic circulation. This concept is in contrast to the current ‘decompartmentalisation’ model in which inflammatory mediators released within the alveolar space leak into the vasculature only following the breakdown of the alveolar-capillary barrier. This discovery has important implications for our understanding of the pathogenesis of multiple organ failure which is the most common cause of death in ARDS patients and therefore warrants further research.

Lung transplantation remains the final option for patients with end-stage lung disease. A significant proportion will develop primary graft dysfunction acutely post implantation, severely impacting on their morbidity and mortality. A key role for marginated monocytes has been identified in several experimental models of lung injury. Whilst other inflammatory leukocytes have been implicated in the pathogenesis of lung ischaemia-reperfusion injury (and primary graft dysfunction), a specific role for vascular monocytes has not yet been fully investigated. We hypothesised that marginated lung monocytes played a critical role in the pathogenesis of lung ischaemia-reperfusion injury. Our overall aims were to: (a) develop, optimise and characterise a murine model of ischaemia-reperfusion, utilising the isolated perfused lung system, (b) elucidate the role of monocytes with liposomal-clodronate depletion and adoptive transfer, and (c) to identify and characterise monocytes in human lung transplant samples and correlate them to patient outcomes. The major findings of this work were that lung marginated monocytes are retained and activated following ischaemia-reperfusion injury, inducing inflammatory mediator release. Clodronate depletion of vascular monocytes resulted in reversal of the injury and abrogated mediator release. Moreover, vascular monocyte repletion (via adoptive transfer), reinstated the injury and inflammation originally observed. Analysis of human lung samples, flushed prior to implantation, also demonstrated significant monocyte retention and activation, which correlated with primary graft dysfunction, in a pilot study of transplant recipients. In summary, this study was able to confirm a critical role for marginated monocytes in lung ischaemia-reperfusion injury. A significant number were retained, and activated, in both murine and human lung samples despite considerable pulmonary flushing. This indicates a propensity for these monocytes to be transferred as ‘passenger’ cells to recipients, where they are capable of the propagation of lung inflammation and injury following transplantation.

Pneumonia and ventilator associated pneumonia (VAP) are a frequent cause for admission to Intensive Care and complication of ventilation respectively. VAP occurs in 10-40% of patients requiring mechanical ventilation and is associated with increased mortality, morbidity and healthcare costs. Diagnosis can be difficult due to poor predictive value of clinical features and low specificity of radiological changes. Bronchoscopic techniques are often invasive, may not be suitable for all patients and are not without complications. New tests are required to improve the diagnosis of these conditions allowing early, appropriate antibiotic treatment. In this study several techniques were used to explore the value of profiling of a range of biofluids obtained from ventilated patients as an aid to diagnosis of pneumonia. Patients were recruited from Intensive Care with either a diagnosis of pneumonia or brain injury. Those with brain injuries were tracked to identify patients who developed VAP. Serum, urine and exhaled breath condensate (EBC) were collected from all patients. Metabonomics, an approach that identifies changes in metabolic profiles associated with disease, was applied using proton nuclear magnetic resonance spectroscopy to both blood and urine and with mass spectrometry (MS) to exhaled breath condensate. Following from the metabonomic work a panel of inflammatory mediators, including cytokines and eicosanoids were measured in serum using MS and flow cytometry to explore the inflammatory changes in these patients. Overall metabolic and inflammatory profiling of serum showed potential as an adjunct to clinical diagnosis especially when combined with clinical data. Analysis of urine and EBC proved more challenging due the number of drug metabolites and low concentration of metabolites they respectively contained. In summary this study has added to the field by demonstrating the potential for profiling techniques of serum from critically ill patients to assist in the diagnosis of both pneumonia and VAP.

Late gestation and labour is associated with inflammatory changes in the myometrium that include an increase in neutrophil and monocyte numbers. The mechanisms regulating altered leukocyte dynamics in pregnancy are not well understood, but may be critical in identifying novel targets for the prevention of preterm labour (PTL). PTL is defined as the onset of labour prior to 37 weeks gestation and often results in preterm birth (PTB). Here we investigated local and systemic leukocyte trafficking during late gestation and labour, focussing on the pro-inflammatory Ly-6Chigh monocyte subpopulation in mouse models of normal, preterm and delayed labour. Using flow cytometry and in vivo cell labelling methods, we demonstrated for the first time that blood-derived Ly-6Chigh monocyte densities increase in the myometrium prior to labour onset. Increases were also observed in the pools of intravascular, lung-marginated Ly-6Chigh monocytes and neutrophils during late gestation, which although not mirrored in the circulation is indicative of a systemic inflammatory response. Using exogenous progesterone supplementation and a progesterone receptor antagonist, we demonstrated that progesterone controls Ly-6Chigh monocyte cell densities within the myometrium, with rapid increases following its functional withdrawal. In a mouse model of infection-induced PTL by laparotomy and intrauterine LPS injection, we unexpectedly found that inflammatory leukocyte migration to the myometrium was low pre- labour. Levels of cytokines and chemokines, including the monocyte chemoattractant CCL2, were high in the myometrium, but also systemically in plasma and organ tissue, indicating a significant systemic inflammatory response. In the surgical controls, in which labour was delayed, there were substantial increases in Ly-6Chigh monocytes in the myometrium, which were shown using knockout mice to be CCR2-dependent. Taken together, our data indicate that both local and systemic inflammation play important roles in labour. Despite consistent observations of Ly-6Chigh monocyte infiltration into the myometrium, there was a lack of clear evidence supporting their role in labour, suggesting their late-gestation myometrium infiltration is more relevant to post-partum uterine repair and remodelling.

The acute respiratory distress syndrome (ARDS) remains a major cause of morbidity and mortality in the intensive care. Despite improvements in intensive care and advances in respiratory support, mortality remains high with no active treatments. Tumour necrosis factor (TNF) is a cytokine that has been implicated in ARDS for over 30 years but its precise roles remain elusive. It signals through two main receptors - the p55 TNF receptor and p75 TNF receptor. The first aspect of this thesis investigated the roles of TNF receptors (TNFR) in the early phase of acute acid-induced lung injury. Using genetically modified mice we discovered that alveolar oedema, as a result of acid aspiration, was specifically mediated through the p55-TNFR, whereas, the p75-TNFR promoted a protective effect. Alveolar oedema formation occurred through an effect independent to the downstream inflammatory events, but instead through the activation of TNF/p55-TNFR/caspase-8 death signalling specifically in the alveolar epithelium. Furthermore, this death-signalling axis led to a reduced alveolar epithelial fluid clearance rate. Epithelial dysfunction occurred prior to epithelial cell death and pharmacological blockade of caspase-8 rescued epithelial function with improvements in gas exchange, suggesting that the activation of caspase-8 per se induced this functional deficit in the alveolar epithelium. The second part of the thesis describes the development of a longer-term model of acid aspiration aimed at extending investigation into the later, arguably more clinically relevant, phases of lung injury (0-10 days). Mice showed respiratory physiology that reached clinical ARDS criteria with significant inflammation and epithelial/endothelial injury, which importantly, resolved facilitating investigation into reparative processes. This model was further characterised using novel flow cytometry protocols to examine the compartmental location of leukocytes during the various phases of ARDS. This model provides a translational platform to allow investigation into the injurious, inflammatory, and resolution mechanisms of ARDS.

Background Peritoneal infection that follows surgery is associated with SIRS and the development of multi-organ dysfunction. It has a high mortality. The dynamics of leucocyte trafficking into the peritoneal cavity and the expression of cytokines in response to either endotoxin or abdominal surgery will effect the development of both local and systemic inflammation. Firstly the leucocyte trafficking in response to intraperitoneal LPS (lipopolysaccharide) is compared to an abdominal incision. Then, using the expression of TNFα on the cell membrane as a marker of pro-inflammatory status, this expression is described in animals undergoing surgery followed by a 'second hit' of ip LPS. Finally the ability of these trafficked cells to trigger SIRS leading to pathophysiological effects in the lungs was assessed. Methods C57/B6 mice were exposed to either an intra-peritoneal dose of 20ng LPS or abdominal surgery and the leucocytes identified and counted in both blood and peritoneal lavage fluid by flow cytometry. A two-hit model was devised, abdominal surgery the primary hit and 24 hours later ip LPS the second hit. The expression of TNFα was measured with flow cytometry. Using two intravenous dyes a lung permeability index was created comparing mice primed by surgery and those just receiving LPS. The data was analysed using student t-test and one-way ANOVA as appropriate. Results: The time course of migration was different for the various leucocyte sub-populations in response to both LPS and surgery. There was a significant increase in the expression of membrane bound TNFα on monocytes and macrophages in the two-hit model than with surgery or LPS alone. There was a trend to greater lung permeability in the two-hit group. Conclusions Leucocytes migrate to the peritoneal cavity after surgery and are primed to respond vigorously to a subsequent dose of LPS. This may have effects on lung permeability.

Ventilator-associated pneumonia (VAP) is the commonest nosocomial infection in Intensive Care and is associated with significant morbidity and mortality. Biomarkers offer the potential to speed up diagnosis and differentiate pulmonary from nonpulmonary infection. We postulated that measurement of cell surface receptors in addition to soluble proteins, in dual sites (blood and BAL) and calculation of an index ratio of BAL / blood would increase the discriminative utility and differentiate pulmonary from non-pulmonary infection. Our body of work included paired blood and BALF obtained from 91 patients in a pilot study: 27 with VAP, 15 ventilated patients with non-pulmonary sepsis, 18 ventilated patients with no evidence of infection and 31 non-ventilated non-infected patients. In each sample, the monocytic and neutrophilic surface proteins TREM-1, CD11b and CD62L were assessed using flow cytometry. Soluble proteins (IL-1β, IL- 6, IL-8) were assayed using ELISA in addition to Procalcitonin, CRP and white cell count. The levels of soluble TREM-1, IL-1β and IL-8 were significantly raised in the BAL of patients with VAP. BAL monocytic surface TREM-1 was also significantly higher in VAP. The BAL/blood ratio increased the discrimination of patients with VAP from non-VAP. Furthermore, the BAL/blood ratio of patients differentiated VAP from non-pulmonary infection. Monocytic and neutrophilic TREM-1 were assessed during the development and resolution phases of VAP. Monocytic surface TREM-1 and its BAL / blood ratio accurately mirrored the changes with infection, indicating them to be putative biomarkers of infection. Finally, we constructed and validated a biomarker panel to discriminate patients with VAP from non-VAP. The panel comprised the BAL/blood ratios of monocytic TREM-1 and CD11b, the BAL levels of soluble TREM-1, IL-8 and IL-1β together with the blood levels of IL-6 and CRP. It had high utility in identifying patients with VAP.

To date no definite treatment exists for acute lung injury (ALI), only supportive therapy, among which mechanical ventilation is an important tool; however ventilation itself may exacerbate the underlying injury, termed ventilator-induced lung injury (VILI). The pro-inflammatory cytokine tumour necrosis factor alpha (TNF) has been consistently implicated in the pathogenesis of ALI/VILI, however clinical trials with anti-TNF therapy failed to improve the outcome. TNF acts on two cell-surface receptors (TNFRs), p55 and p75, and knock-out studies in ALI/VILI indicate they have directly opposing roles, i.e. promoting or protecting against pulmonary oedema, respectively, but it remains unclear where these signalling mechanisms take place, being the endothelial or epithelial side of the alveolar-capillary membrane. This project had two main aims: to characterise the expression profile and roles of TNF receptor subtypes on each side of the alveolarcapillary membrane and to modulate ALI progression in mouse models by targeting individual TNFRs accordingly. Characterisation studies on freshly harvested mouse pulmonary endothelial cells (PECs) using flow cytometry indicated expression of both receptors with dominance of p55. Knock-out and pharmacological inhibition studies in vivo and in vitro, respectively, found p55 to be the major mediator of TNF-induced adhesion molecule expression on PECs, with a similar but less potent role for p75 that varied across adhesion molecules. Endotoxaemia resulted in rapid shedding of p55 on PECs while expression of p75 was preserved, indicating that the relative role of the receptors may change during inflammatory conditions. Specific inhibition of intraalveolar p55 signalling by domain antibodies significantly attenuated both pulmonary oedema and inflammation in mouse models of VILI and acid-aspiration. These data offer new, potentially clinically applicable insights into the involvement of TNFR biology in ALI and the novel domain antibody technology may open a new therapeutic approach for patients who suffer from, or at risk of ALI.

Sepsis consists of both the systemic inflammatory response syndrome (SIRS) and the compensatory anti-inflammatory response syndrome (CARS). How these differential response states are regulated is yet to be fully elucidated. Tumour necrosis factor-alpha (TNF) is one of the principal cytokines involved in mediating SIRS. TNF is released from cells by tumour necrosis factor-alpha converting enzyme (TACE), this enzyme is responsible for the ectodomain cleavage of a number of other substrates relevant to inflammation including both TNF receptors and the adhesion molecule L-selectin. How TACE contributes to, and functions in, SIRS and CARS is not yet known. My objective was to investigate TACE activity and associated substrate shedding in monocytes, specifically how the enzyme behaved in the context of in vitro models that I designed to induce states of priming and tolerance. I then obtained in vivo samples from critically ill patients to determine whether there were similarities between the TACE activity profiles found in patient cells, and volunteer cells placed in the in vitro models. My aims were: 1) Determine how TACE activity profiles were altered when sequential inflammatory stimuli were utilised in a two-hit model of sepsis designed to induce states of priming and tolerance and 2) To perform a clinical study to investigate TACE behaviour in the context of critical illness. I successfully refined a method of isolating primary monocytes from healthy volunteers and patients that allowed determination of TACE activity profiles. Furthermore, I demonstrated that the LPS-TACE axis was reset in the context of a two-hit LPS model and in sepsis. I found evidence of differential signalling pathway reprogramming in monocytes taken from patients with infectious and non-infectious SIRS. Finally, I demonstrated that the monocyte TACE response to LPS is dependent on cell contact. These data provide new insights into monocyte inflammatory function during the immune response.

A lack of a simple index that monitors ventilatory efficiency at the bedside has meant that oxygenation has been the predominant variable that is used to monitor adequacy of ventilatory strategies and disease severity in mechanically ventilated patients. Due to complexities in its measurement, deadspace ventilation, the traditional method to track ventilatory failure, has failed to become integral in the management of mechanically ventilated patients. Ventilatory ratio (VR) is an easy to calculate index that uses variables measured at the bedside: [Mathematical equation appears here. To view, please open pdf attachment] where [Symbols appears here. To view, please open pdf attachment] is taken to be 100 ml.kg-1.min-1 based on predicted body weight and [Symbols appears here. To view, please open pdf attachment] is taken to be 5 kPa. Physiological analysis of VR dictates that it is influenced by deadspace fraction and CO2 production. Physiological analysis of VR was validated in a benchside lung model and a high fidelity computational cardiopulmonary physiology model. The impact of CO2 production on VR was investigated in patients undergoing laparoscopic surgery who received exogenous intraperitoneal CO2. This showed that delta values of the 2 variables were linear. The variability of CO2 production was examined in ICU patients and results of the study showed that variability of CO2 production was small. In an ICU population correlation of VR was stronger with deadspace in comparison to CO2 production. Of these two variables, deadspace had the greater effect on VR. The clinical uses of VR were examined in 4 databases of ICU patients. VR was significantly higher in non-survivors compared to survivors. Higher values of VR were associated with increased mortality and more ventilator days. A rising values of VR over time was also associated with worse outcome. VR is a simple bedside index that provides clinicians with useful information regarding ventilatory efficiency and is associated with outcome.

Acute lung injury (ALI) is a severe inflammatory lung disease with high mortality. Previous studies revealed several important concepts in ALI, including cellular interaction between lung-marginated leukocytes and pulmonary endothelium, and decompartmentalisation of soluble mediators. However, there are inherent limitations within both in vivo and in vitro models to identify the detailed mechanism underlying these concepts. In this PhD project, we attempted to address these unanswered questions, using an in situ isolated perfused mouse lung (IPL). Specifically, we aimed to 1) develop, characterise, and optimise the mouse IPL model; 2) investigate soluble and cellular aspects of two models of ALI that are particularly amenable to study using the IPL, namely ventilator-induced lung injury (VILI) and ischaemia-reperfusion injury. From a physiological viewpoint, VILI consists of 2 primary components, high-stretch and atelectasis. Modelling atelectasis-related injury in vivo is difficult due to negative pleural pressure. We took advantage of the zero pleural pressure of open-chest IPL system to develop an atelectasis-related VILI model. Comparison of this ‘atelectrauma’ and a high-stretch ‘volutrauma’ model demonstrated that both cause lung oedema and pulmonary inflammation, but the inflammatory impact was different between them. Volutrauma, but not atelectrauma, facilitated systemic cytokine release, in which lung-marginated monocytes seem to play an important role. This finding in the VILI model drove us to further investigate the role of these monocytes in an ischaemia-reperfusion model, which is clinically highly relevant and simulates a lung transplantation setting. Our results suggested that lung-marginated monocytes may also contribute to develop ischaemia-reperfusion injury, potentially involving TNF upregulation. Through this PhD project, we have successfully developed a technically very challenging mouse IPL model. We utilised the unique features of the IPL to develop experimental models that we believe will be strong tools to fill the gap between in vivo physiological significance and in vitro mechanistic understanding.

Acute respiratory distress syndrome remains a significant cause of mortality among intensive care patients. Mechanical ventilation is a crucial component of therapy but can exacerbate injury through overstretch of lung units, known as ventilator-induced lung injury (VILI). The mechanisms by which ventilation triggers the characteristic inflammatory cascade and how this pulmonary inflammation progresses to extrapulmonary organs have not been clearly elucidated. In this work I explored the role of mononuclear phagocytes in the initiation (alveolar macrophages) and subsequent progression (monocytes) of VILI, using in vivo mouse models. Initially, the importance of alveolar macrophages in VILI was clarified by pharmacologically depleting these cells, which substantially attenuated injury. The observation that injurious ventilation led to a decrease in macrophage recoverability led us to hypothesise that adhesive interactions between epithelial cells and macrophages may be important in their activation. To address this, I developed a novel flow cytometric methodology to assess intracellular mitogen activated protein (MAP) kinase phosphorylation in discrete pulmonary cell populations. I was able to demonstrate rapid alveolar macrophage activation, but little epithelial activation, during high stretch ventilation. This potentially indicated that alveolar macrophages directly sense stretch through their adhesive contacts with epithelial cells. I attempted to study this using antibodies to block integrin-mediated interactions, although results were inconclusive. In addition to these experiments, I carried out a preliminary study showing, for the first time, that activated leukocytes marginate to extrapulmonary organs during VILI. This supports the possibility that monocytes and neutrophils play a role in the development of extra-pulmonary organ injury, although this remains undefined. This data shows that mononuclear phagocytic cells are likely to play crucial roles in the pathogenesis of VILI. The findings and methodologies developed provide many possibilities for future research that will enhance our understanding of this iatrogenic injury, ultimately in view of potential therapeutic targets.

Although monocyte margination to the lungs is enhanced during endotoxemia, little is known of their role in sepsis-associated acute lung injury (ALI). Moreover, monocytes exists as two functionally distinct subsets ‘inflammatory’ Gr-1high and ‘resident’ Gr-1low, which exhibit heterogeneous inflammatory responses upon LPS stimulation both in vivo and in vitro but, the signalling mechanisms regulating their differential responses are unknown. The objective of this research was to define the role of monocyte-endothelial interactions during pulmonary microvascular injury, in response to LPS challenge, using an in vitro monocyte-endothelial coculture model. The specific aims were to: establish a primary lung endothelial cell line culture; investigate priming and enhancement of monocyte pro-inflammatory response; investigate the role of the p38 MAPK pathway in the differential monocyte subset cytokine response, and investigate the role of LPS-mediated TLR4 signalling for monocyte-endothelial reciprocal activation. Primary mouse lung endothelial cells (MLEC), that were successfully cultured, displayed typical morphological and phenotypic properties of the vascular endothelium. During coculture with MLEC, monocytes produced significantly higher levels of TNF when treated with a single high dose of LPS when compared with monocytes cultured alone. Also, upon LPS stimulation, Gr-1high monocytes expressed significantly higher levels of TNF, IL-6, COX-2, and iNOS in comparison to Gr-1low monocytes. Furthermore, significantly higher levels of p38, a critical regulator of inflammatory cytokines, were induced in Gr-1high monocytes in comparison to Gr-1low monocytes. Finally, using a chimeric system of TLR4+/+ (LPS responsive) and TLR4-/- (LPS unresponsive) monocytes and endothelial cells, it was demonstrated that monocytes could activate endothelial cells through TNF-dependent signalling, while endothelial cells could induce TNF production in monocytes. These data indicate an important role for monocytes, especially the Gr-1high subset in the pathogenesis of ALI, and suggest that the p38 MAPK pathway could play a pivotal role in regulating monocyte-induced pulmonary inflammation during sepsis-associated ALI.

Background: Glutathione-S-Transferase (GST) family of enzymes serve as a key antioxidant system. Clinical evidence suggests that several GST polymorphisms are associated with inflammatory lung diseases. The mechanisms by which these polymorphisms impact on cellular responses to stress remain unknown. To define the specific role of GST isoenzymes, we investigated the influence of GST inhibition on lung cell survival and inflammation, in in vitro and ex vivo models. Methodology: Pulmonary cells were exposed to GST inhibitors ethacrynic acid or caffeic acid alone, or in combination with oxidative stressors hydrogen peroxide, tert-butyl-hydroperoxide or hypoxia and reoxygenation. RNA interference was employed to attenuate specific GST isoform activity. Cell injury was assessed by MTT assay and flow cytometry. Oxidative stress was investigated by: (a) DHR123 and DCF fluorescence, (b) monitoring intracellular redox state by HPLC, (c) measurement of protein carbonyls by spectrophotometry and d) evaluation of the effects of specific and broad-spectrum antioxidants. Mitochondrial energetics was assessed by calculation of ATP/ADP ratios by HPLC, while global metabolic response was evaluated by NMR spectroscopy. An ex vivo mouse isolated perfused lung (IPL) model was used to explore the physiological role of GST. Principal findings: Two GST inhibitors and genetic attenuation of GSTπ significantly increased oxidative stress and induced cytotoxicity of lung epithelial cells and rendered them more susceptible to oxidative stress. While cellular energy state was not compromised, EA potentiated protein carbonylation and caused metabolic derangements. Furthermore, in the ex vivo model, EA significantly increased pulmonary permeability and intra-alveolar oedema, which may have been responsible for the increase in pulmonary elastance and resistance that were also observed. Conclusion: These observations highlight the importance of GST enzymes, specifically GSTπ, in the cellular and whole lung response to stress conditions and may help to understand the metabolic determinants of oxidative lung injury and adaptation.

The systemic inflammatory response syndrome (SIRS) describes the immune response to an insult that is inappropriate for a patient’s biological needs. Tumour necrosis factor-α (TNF) is one of the principal pro-inflammatory cytokines that mediates SIRS. TNF-α converting enzyme (TACE) is responsible for the ectodomain cleavage of numerous immunologically relevant substrates including TNF, both TNF receptors and the cellular adhesion molecule L-selectin. The biological significance of TACE lays in its ability to control the availability of these key mediators of the immune response in their membrane-bound and soluble forms. Little is known concerning the regulation of TACE-mediated ectodomain shedding and it contributes to disease pathophysiology. Our objective was to investigate the regulation of TACE catalytic activity and expression during acute inflammation in vitro and in vivo. Our aims were: 1) define the pathways mediating upregulation of TACE activity in primary human monocytes; 2) compare the regulation of TACE catalytic activity and TACE substrate shedding; 3) investigate the acute inflammatory response in elective surgical patients with particular focus on three potential biomarkers of inflammation: TACE expression, HLA-DR expression and circulating monocyte subset populations. We used a fluorimetric assay of TACE activity to demonstrate that lipopolysaccharide (LPS) stimulation of primary human monocytes promotes rapid upregulation of TACE activity, independently of changes in TACE expression. TACE activity upregulation was mediated by reactive oxygen species-induced activation of the p38 MAPK-MK2 pathway. Dithiol oxidation was found to induce L-selectin shedding but to attenuate LPS-induced TACE activity upregulation. Investigation of this paradox revealed that L-selectin shedding may be independent of TACE activity upregulation. Monocyte TACE expression was not modified by low to intermediate risk surgery. In contrast, surgery resulted in downregulation of monocyte HLA-DR expression and differential trafficking of monocyte subsets. These data provide insight into the regulation of TACE during acute inflammation.

Mechanical ventilation, essential for the support of patients with acute lung injury (ALI), causes exacerbation of the existing pathology, a process termed ventilator-induced lung injury (VILI). The pro-inflammatory cytokine tumour necrosis factor-alpha (TNF) has been consistently implicated in ALI/VILI. TNF activates two receptors, TNFR p55 and p75 that act in opposition during VILI to promote or protect against pulmonary oedema formation, respectively, but the mechanisms underlying this are unknown. Alveolar and plasma soluble TNFR (sTNFR) levels are elevated in ventilated ALI patients and associated with mortality. However, the relevance of these increases is unclear. This project had two main aims: to investigate kinetics and sources of sTNFRs in the alveoli and plasma and investigate mechanisms underlying differential TNFR signalling during VILI, using in vivo mouse models. Investigation of intraalveolar sTNFRs during VILI, and also ALI induced by intratracheal administration of hydrochloric acid or bacterial toxins, showed that intraalveolar sTNFRs are differentially regulated during ALI: VILI/hydrochloric acid induced leakage of sTNFR p55 and p75 from plasma, whereas bacterial toxins induced intraalveolar p75 production. These differences have important implications for TNF signalling and potential use as clinical markers. Investigation of plasma sTNFRs suggests that VILI induces direct production of sTNFRs by the pulmonary vasculature, as opposed to leakage of intraalveolar sTNFRs into the circulation as previously suggested. Development of a flow cytometry technique to study pulmonary TNFR expression was successfully validated using TNFR deficient tissue, but consolidation of data by immunohistochemistry was unsuccessful. Investigation of differential TNFR signalling mechanisms showed that following intratracheal fluid administration, p75 deficient mice exhibit physiological changes consistent with impaired fluid reabsorption, implicating p75 in lung fluid reabsorption during VILI. These data offer new, potentially clinically applicable insights into the involvement of TNFR biology in VILI/ALI and the novel methodologies developed herein constitute useful tools for future research.