Platelets play an important role in defense against pathogens; however, the interaction between Escherichia coli and platelets has not been well described and detailed. Our goal was to study the interaction between platelets and selected strains of E. coli in order to evaluate the antibacterial effect of platelets and to assess bacterial effects on platelet activation. Washed platelets and supernatants of pre-activated platelets were incubated with five clinical colistin-resistant and five laboratory colistin-sensitive strains of E. coli in order to study bacterial growth. Platelet activation was measured with flow cytometry by evaluating CD62P expression. To identify the difference in strain behavior toward platelets, a pangenome analysis using Roary and O-antigen serotyping was carried out. Both whole platelets and the supernatant of activated platelets inhibited growth of three laboratory colistin-sensitive strains. In contrast, platelets promoted growth of the other strains. There was a negative correlation between platelet activation and bacterial growth. The Roary results showed no logical clustering to explain the mechanism of platelet resistance. The diversity of the responses might be due to strains of different types of O-antigen. Our results show a bidirectional interaction between platelets and E. coli whose expression is dependent on the bacterial strain involved.

Apart from their involvement in hemostasis, platelets have been recognized for their contribution to inflammation and defense against microbial agents. The interaction between platelets and bacteria has been well studied in the model of Staphylococcus and Streptococcus but little described in Gram-negative bacteria, especially Escherichia coli. Being involved in the hemolytic uremic syndrome as well as sepsis, it is important to study the mechanisms of interaction between platelets and E. coli. Results of the published studies are heterogeneous. It appears that some strains interact with platelets through the toll-like receptor-4 (TLR-4) and others through the Fc gamma glycoprotein. E. coli mainly uses lipopolysaccharide (LPS) to activate platelets and cause the release of antibacterial molecules, but this is not the case for all strains. In this review, we describe the different mechanisms developed in previous studies, focusing on this heterogeneity of responses that may depend on several factors; mainly, the strain studied, the structure of the LPS and the platelet form used in the studies. We can hypothesize that the structure of O-antigen and an eventual resistance to antibiotics might explain this difference.

In addition to their role in haemostasis, platelets are also involved in the inflammatory and antimicrobial process. Interactions between pathogens and platelets, mediated by receptors can lead to platelet activation, which may be responsible for a granular secretion process or even aggregation, depending on the bacterial species. Granular secretion releases peptides with bactericidal activity as well as aggregating factors. To our knowledge, these interactions have been poorly studied for Escherichia coli (E. coli). Few studies have characterised the cellular organization of platelet-E. coli aggregates. The objective of our study was to investigate the structure of platelet aggregates induced by different E. coli strains as well as the ultrastructure of platelet-E. coli mixtures using a scanning and transmission electron microscopy (SEM and TEM) approach. Our results show that the appearance of platelet aggregates is mainly dependent on the strain used. SEM images illustrate the platelet activation and aggregation and their colocalisation with bacteria. Some E. coli strains induce platelet activation and aggregation, and the bacteria are trapped in the platelet magma. However, some strains do not induce significant platelet activation and are found in close proximity to the platelets. The structure of the E. coli strains might explain the results obtained.