Neutrophils in bacterial pneumonia: influx and clearance
Lawson Roderick A.
MetadataShow full item record
Despite the advent of powerful modern antibiotics, pneumonia continues to be of great importance. Although most cases of community acquired pneumonia (CAP) will recover fully, some (including those previously fit and well) may die. Many hospital patients will suffer nosocomial pneumonia (NP), with a high mortality rate. This is particularly true on the intensive care unit.The neutrophil granulocyte is of particular importance in the defence of the lung. It contains many substances that are bactericidal. During pneumonia, it is recruited from the intravascular space into the lung interstitium and the air -spaces. There, its bactericidal products are capable not only of damaging and killing bacteria, but also of causing'by- stander' damage to the host lung. During the course of pneumonia it is important that neutrophils are not only recruited rapidly to defend against bacteria, but that their recruitment should cease as soon as adequate numbers are attained or their function is no longer required. Not only that, but the neutrophils that have already been recruited must then be safely and speedily removed from the site of pneumonia.To investigate these dynamic processes, a rabbit model of pneumonia was used. Bacteria were instilled to a localised area via a fibre -optic bronchoscope. This allowed strict definition of the time of onset of pneumonia, and study of subsequent evolving processes. Two organisms were used; 1) Streptococcus pneumoniae, an organism characteristic of CAP and; 2) Escherichia coli, an organism characteristic of NP. It was hypothesised that in pneumonia due to the former (PneuS), the usual remarkably complete recovery witnessed clinically is due to the tissue load of neutrophils being carefully controlled. By contrast, pneumonia due to the latter (pneuE) is clinically much more severe and lung- damaging, due to a larger lung neutrophil burden. This could be because of earlier cessation of neutrophil influx and /or more rapid neutrophil clearance in pneuS than pneuE.Radiolabelled neutrophils from donor animals were injected at intervals after induction of pneumonia to assess the magnitude of ongoing neutrophil recruitment. In pneuS, neutrophil influx was significantly elevated above control levels at 6 hour but not 30 hour or subsequent time points. In pneuE, neutrophil influx was at least as high at 30 hours as at 6 hours. This confirms the hypothesis that neutrophil influx is more prolonged in pneuE than in pneuS.The requirement for the CD18 adhesion molecule, (known to be important in recruitment of neutrophils) has been shown to change with time in the peritoneum. This could be an important facet of control of the development of inflammation. In pneuS, neutrophil recruitment is known to be CD18 independent. The previous finding of others that such antibodies inhibit neutrophil recruitment early (6 hours) in pneuE was confirmed. It was found there was no change in this CD18 dependency later on (at 30 hours).The chemokine IL -8 is thought to be particularly important in the recruitment of neutrophils. An anti -IL -8 blocking antibody was used to assess the importance of IL -8 in neutrophil recruitment at 6 and 30 hours. Although this inhibited IL -8 induced neutrophil shape change in vitro and intradermal IL -8 induced neutrophil influx in vivo, iand reversed the prolonged retention of tracer neutrophils in pneumonic lungs minutes after injection, the antibody failed to block neutrophil influx in either type of pneumonia at 6 or 30 hours. hndeed, there was a trend towards increased influx after treatment. High dose antibody produced the same effect. These surprising results are partly explained by higher broncho- alveolar lavage (BAL) and plasma levels of IL -8 after antibody treatment. This may be due to IL -8 /anti -IL -8 antibody complexes amplifying the inflammatory response, although no endothelially bound anti -IL -8 was detected. Alternatively, it may represent the release of negative feedback on IL -8 production.In both pneus and pneuE, the early peak in BAL neutrophil numbers was followed by an increase in the number of apoptotic neutrophils. This in turn was followed by an increased number of alveolar macrophages containing apoptotic bodies. Trypan blue positive, necrotic neutrophils were rare. This is consistent with neutrophil apoptosis and associated macrophage phagocytosis, (a process that limits the release of toxic neutrophil products), playing an important role in neutrophil clearance in pneumonia.BAL from animals with pneumonia promoted rabbit neutrophil apoptosis in vitro. In pneuS (but not pneuE) this correlated weakly though significantly with the amount of apoptotic neutrophils recovered from BAL and with BAL levels of IL -8, gro and MCP -1. It is suggested that a factor is elaborated within the lung during inflammation that promotes neutrophil apoptosis, thus giving negative feedback control on the lung neutrophil burden. The different relationships found between in vitro and in vivo in pneuS and pneuE may again contribute to the different clinical outcomes in the two diseases.In summary, an animal model demonstrated that neutrophil influx was more prolonged in pneuE than in pneuS. It is suggested this contributes to the more severe manifestations of the former clinically. CD18 was important to the influx of neutrophils in pneuE late as well as early in the disease. Anti -IL -8 antibody failed to inhibit neutrophil influx in either type of pneumonia, which may be due to immune -complex formation or release of negative feedback controls. Evidence that neutrophil apoptosis is involved in their clearance during pneumonia was obtained, together with evidence that a factor is released to promote neutrophil apoptosis and thus provide negative feedback control of inflammation during pneumonia. The control may differ in PneuS and PneuE, affecting outcome.