I am taking an evolutionary approach to integrate high throughput sequencing technologies into the daily discovery process of microbiologists. To do this, I use a model of urinary tract infection (UTI) caused by Escherichia coli
. The primary and most well-studied virulence factor enabling E. coli
to cause UTI in the mammalian bladder is type 1 pili, a fibrous structure created by E. coli
that allows them to bind to the bladder epithelium. Natural variations in the genes encoding type 1 pili exist and affect the ability of E. coli
to cause disease.
By sampling the variation in type 1 pili genes and studying it from an evolutionary perspective, we can distinguish between variations that affect fitness (i.e. are under selection, meaning evolution "cares" about these) and those that do not (i.e. are evolving neutrally). Furthermore, these computational
predictions about fitness effects correspond precisely to what is observed in an in vivo
experimental mouse model of UTI. Moreover, these evolutionarily selected mutations influence the course of disease via a mechanism that was not detected by previous genetic, biochemical, structural, and epidemiological studies. Therefore, sequence-based evolutionary tools represent a powerful and complementary approach to studying virulence factor function, allowing us to specifically identify a small subset of functional
amino acids within key proteins.
My lab is extending and generalizing these methods and initial results to uncover the molecular these selected mutations act in during disease; to discover new virulence factors affecting in vivo
infections caused by E. coli
; and to explore the evolution of regulatory sequences and mobile elements in bacterial genomes. The approach is a dynamic hybrid of computational and experimental work, spanning high throughput sequencing, molecular evolution, bacterial genetics, in vitro
cell culture, and in vivo
models of infection.