Uropathogenic Escherichia coli (UPEC) are the most common bacterial pathogens causing urinary tract infections (UTIs). Driven by the development of antibiotic resistant UPEC strains, UTIs have become a major public health issue and generate substantial healthcare costs. Even in the absence of antibiotic resistance, UTIs are notoriously difficult to treat and have high rates of recurrence, mostly due to: 1. the ability of UPEC to adhere to and invade bladder epithelial cells, which protects them from the action of antimicrobial agents and the immune system 2. the presence in antibiotic-susceptible UPEC populations of a small fraction of so-called persister cells, which are killed at a slower rate than the bulk of the population and can therefore survive prolonged antibiotic exposure These two features allow some bacteria to remain viable in a protected niche for the whole duration of the antibiotic treatment, potentially leading to infection relapse when the treatment is terminated. High-content screening (HCS), which uses microscopy as a screening tool, allows the visualization of bacterial phenotypes at the single-cell level and can therefore provide a powerful and unbiased tool to understand the genetic basis of these phenotypes. The present thesis describes two complementary high-content screening approaches aimed at identifying UPEC mutants with altered adhesion to bladder cells and with altered antibiotic persistence respectively. In a first chapter, we report preparatory steps that led to the construction of a transposon insertion library of UPEC mutants that is suitable for high-content imaging, using the CFT073 clinical UTI isolate as background strain. In the second chapter, we performed fluorescence microscopy-based high-content screening to identify mutants with defects in early adhesion to bladder epithelial cells. We recovered 82 mutants with decreased adhesion and 54 mutants with increased adhesion. Unexpectedly, nine low-adhesion €œhits€ mapped to the two P pili operons encoded by CFT073, which are thought to mediate adhesion to kidney cells rather than bladder cells. Additionally, six high-adhesion hits mapped to the operon coding for F1C pili, and we show that this phenotype is linked to increased P pili synthesis. These results therefore reveal a critical role for P pili in UPEC adhesion to bladder epithelial cells, which may inform the development of novel anti-adhesion therapies to prevent UTI recurrence. In the third chapter, we combined microfluidics with time-lapse microscopy to identify mutants with altered persistence to fosfomycin (FOS) —€“ a cell-wall inhibitor used as first-line agent in the treatment of UTIs —€“ in synthetic human urine (SHU), which mimics the physiology of UPEC in real urine. We recovered four mutants with decreased persistence, three of which had defects in lipopolysaccharide (LPS) synthesis. These results identify the outer membrane of UPEC as a key component for survival to FOS treatment, thus opening up new avenues in the fight against antibiotic persistence.