E. coli adaptation to the extraintestinal niche: using natural variation to explore mechanisms of pathogenesis

Abstract

Escherichia coli can function as part of the normal human gut flora, or it can act as a virulent pathogen, either within the intestinal tract or without. While intestinally pathogenic E. coli tend to share particular traits and can be distinguished from commensal strains, extraintestinally pathogenic E. coli (ExPEC) are not readily distinguishable from commensal strains. The fecal-oral transmission of ExPEC means that these strains are regularly subjected to the same selection pressures as intestinal strains and may be isolated from the stool of healthy individuals, thereby being labeled as commensal despite their pathogenic potential. Additionally, the adaptations that render a strain more able to cause extraintestinal infection do not appear to be a particular suite of changes, as with Shigella, but are more likely to be some combination from a much larger set of pathoadaptive mutations. My research is interested in exploring this larger set of pathoadaptive mutations. Part 1 (chapters 2-3) explores naturally occurring variation within the fim operon across a number of strains of varying pathotypes. The fim operon codes for a long adhesive filament which is involved in normal transmission and persistence of commensal E. coli and which has also been shown to be involved in migration to and persistence in the urinary tract. Our lab has identified a number of mutations affecting the structure and secretion of the adhesin, FimH, which are associated with uropathogenic E. coli. However, many uropathogenic strains have no such mutations; my research explores whether other variations in the fim operon are associated with extraintestinal pathogenicity. Such mutations could help us understand the role of type 1 fimbriae in extraintestinal infections and potentially illuminate novel characteristics of T1F. While many of the characteristics of the fim operon were as expected given previous research, variation in the short-term is significantly biased towards nonsynonymous and noncoding positions. Additionally, relatively high levels of variation were observed upstream of fimH, with one position under positive selection in a lineage including many ExPEC strains. Part 2 (chapters 4-5) explores the evolution and diversification of an E. coli strain over the course of a long-term chronic infection, in this case in the airways of an individual with cystic fibrosis (CF). Patients with CF develop chronic lung infections that are frequently polymicrobial. Work by our collaborators demonstrated that E. coli can not only colonize the lungs of cystic fibrosis patients, it can persist for months to years and undergoes characteristic phenotypic changes (mucoidy, small colony variants, and antibiotic-resistance) seen in typical CF pathogens during chronic infection. We sequenced the genomes of clonal isolates with varying phenotypes from a single patient and identified mutations specific to each. These mutations allowed us to distinguish and follow lineages that arose over the course of infection, and one mutation, a truncation in mdoH, was definitively shown to be the cause of the mucoid phenotype.