Assessment of the efficacy of phage therapy and its influence on gut microbiota in the prevention of neonatal meningitis caused by Escherichia coli K1 | Evaluation de l'efficacité de la phagothérapie et de ses répercussions sur le microbiote intestinal dans la prévention des méningites néonatales liées à Escherichia coli K1

English. Extra-intestinal Escherichia coli (ExPEC) differ from their intestinal counterparts in their ability to invade, colonize and cause infections outside the gastrointestinal tract. ExPEC can cause serious infections in both humans and animals, including urinary tract infections, bloodstream infections and meningitis (J. R. Johnson et al., 2001; Kim, 2017; King et al., 2015). A specific pathotype among ExPEC associated with neonatal meningitis (NMEC) is responsible for catastrophic consequences in affected newborns, with mortality rates of up to 50% in developing countries, as well as significant sequelae in survivors (Furyk et al., 2011; Houdouin et al., 2008). NMECs displaying the capsular type K1 are of particular importance because this important virulence factor facilitates colonization and survival in a variety of host environments (Kim et al., 2003; Quartu et al., 2008). The ability of E. coli K1 to cause neonatal meningitis, combined with the emergence of antibiotic-resistant strains, highlights the need to develop alternative therapeutic strategies. This thesis focuses on the assessment of the therapeutic potential of phage therapy targeting specifically E. coli K1 for the prevention of neonatal meningitis. The first objective of this thesis was to isolate new phages against E. coli K1 and characterize them both in-vitro and in-vivo in the Galleria mellonella larvae model, in order to determine their therapeutic potential. The phages were isolated against the E. coli O18:K1:H7 (APEC45) strain, responsible for avian colibacillosis and sharing similarities with human NMEC strains. The results led to the selection of four phages specifically targeting the K1 capsular type and presenting several interesting characteristics from a therapeutic point of view, such as their strictly lytic infection cycle and their resistance to wide temperature and pH ranges. In addition, all four phages were able to replicate efficiently in the G. mellonella model, while having a positive impact on the survival of the infected larvae. However, phage treatment did not completely eliminate the bacterial load. The second objective was to evaluate the potential of phage therapy to prevent neonatal meningitis by targeting E. coli K1 in the gut microbiota of pregnant women using the SHIME® model. Indeed, extra-intestinal E. coli, including E. coli K1, are often commensal in the gut microbiota of pregnant women and healthy individuals. However, their ability to colonize the vagina during pregnancy and subsequently the gastrointestinal tract of the newborn at birth is an early step in the development of neonatal E. coli meningitis. The study therefore focused on the assessment of the efficacy and dynamics of the phage K1_ULINTec4 in an intestinal environment using the SHIME® system, and to explore the consequences of phage therapy on the gut microbiota of pregnant women. The monitoring of the microbiota involved multiple and complementary analyses, including metagenetics, qPCR, short-chain fatty acid (SCFA) analysis and aryl hydrocarbon receptor (AhR) induction. Three experiments were carried out, involving phage K1_ULINTec4 alone, E. coli K1 alone, and E. coli K1 and phage together over a 6-week period. The results showed that the phage K1_ULINTec4, administered alone, exhibited excellent persistence in the system. Moreover, phage concentrations were higher in the presence of their host, indicating effective phage replication in the model. Furthermore, no significant negative impact was observed on bacterial populations. However, phage treatment did not completely eliminate E. coli K1, suggesting the presence of potential resistance mechanisms. The third objective focuses on understanding the development and consequences of resistance to phage K1_ULINTec4 in two clinical E. coli K1 strains used in studies 1 and 2, which are involved in avian colibacillosis and neonatal meningitis, respectively: APEC 45 and C5. Resistant colonies were analyzed following exposure to phage K1_ULINTec4. Genomic analysis revealed alterations in the kps gene cluster associated with capsule biosynthesis. Phenotypic modifications at the capsule level were observed through a rapid immunoassay targeting the K1 capsule and microscopy. Additionally, the clustered organization of the bacteria observed under microscopy led to the examination of biofilm production. These tests showed increased biofilm production levels compared to the original strains. In parallel, the resistant isolates demonstrated lower virulence in the Galleria mellonella model. This thesis confirm and expand knowledge on phages specifically targeting E. coli K1 as well as their therapeutic potential. An assessment to prevent neonatal meningitis in pregnant women was conducted using the K1-ULINTec4 phage in the SHIME® model. This experiment demonstrated the effective action of the phage in the intestinal environment while preserving the surrounding microbiota. However, the emergence of resistant bacteria both in G. mellonella larvae and in the SHIME® model highlights the need to develop well-designed phage treatments, through notably phage cocktails.

Inteliphages