Lipopolysaccharide with long O-antigen is crucial for Salmonella Enteritidis to evade complement activity and to facilitate bacterial survival in vivo in the Galleria mellonella infection model.

Bacterial resistance to serum is a key virulence factor for the development of systemic infections. The amount of lipopolysaccharide (LPS) and the O-antigen chain length distribution on the outer membrane, predispose Salmonella to escape complement-mediated killing. In Salmonella enterica serovar Enteritidis (S. Enteritidis) a modal distribution of the LPS O-antigen length can be observed. It is characterized by the presence of distinct fractions: low molecular weight LPS, long LPS and very long LPS. In the present work, we investigated the effect of the O-antigen modal length composition of LPS molecules on the surface of S. Enteritidis cells on its ability to evade host complement responses. Therefore, we examined systematically, by using specific deletion mutants, roles of different O-antigen fractions in complement evasion. We developed a method to analyze the average LPS lengths and investigated the interaction of the bacteria and isolated LPS molecules with complement components. Additionally, we assessed the aspect of LPS O-antigen chain length distribution in S. Enteritidis virulence in vivo in the Galleria mellonella infection model. The obtained results of the measurements of the average LPS length confirmed that the method is suitable for measuring the average LPS length in bacterial cells as well as isolated LPS molecules and allows the comparison between strains. In contrast to earlier studies we have used much more precise methodology to assess the LPS molecules average length and modal distribution, also conducted more subtle analysis of complement system activation by lipopolysaccharides of various molecular mass. Data obtained in the complement activation assays clearly demonstrated that S. Enteritidis bacteria require LPS with long O-antigen to resist the complement system and to survive in the G. mellonella infection model.