In Vitro Modelling of the Formation of Inflammatory Platelet- Leukocyte Aggregates and their Adhesion on Endothelial Cells (an Early Event in Sepsis)

Sepsis, a severe systemic inflammatory response to an infection, is a global health problem with significant economic burden. Platelet leukocyte aggregates (PLAs) are extensively formed in sepsis, and correlate with severity of disease. Molecular interactions that lead to the formation and adherence of these cellular aggregates to endothelial cells might represent novel targets to use for therapy. Whole blood stimulation assays and flow cytometry are widely used to study the formation of PLA aggregates in an in vitro approach to understand the acute inflammatory reaction in the bloodstream to the presence of pathogen associated molecular patterns (PAMPs, most commonly LPS) during septicaemia. However, these assays are limited by the lack of robust and physiologically relevant conditions. Most importantly, the extent of spontaneous aggregate formation is unclear in most assays, as cells outside the body may aggregate, forming PLA. The aim of this work was to assess extent of spontaneous aggregate formation in whole blood stimulation assays and to compare the effects of endotoxin and of heat killed, clinically relevant, pathogens on aggregate formation and adhesion of complex aggregates to TNFα stimulated endothelial cells. Endotoxin from E.coli or Salmonella enteritidis was not a suitable stimulus to provoke sepsis relevant platelet leukocyte aggregates in vitro, as it did not further increase the extent of aggregates formed spontaneously in stasis of hirudin anticoagulated blood. By contrast, heat killed Klebsiella pneumoniae or Staphylococcus aureus produced significantly enhanced and complex cellular aggregates which adhered to TNFα stimulated endothelial cells. These were reliably captured by scanning electron microscopy. Adhesion of cellular aggregates could be blocked by incubation of endothelial cells with a commercial P-Selectin antibody and a novel angiopoietin-2 ligand trap. In conclusion, a method was developed that models the acute inflammatory reaction in whole blood in the presence of relevant pathogen surfaces.