Aggregation and Dispersion of Uropathogenic Escherichia coli 536

Abstract

Uropathogenic Escherichia coli (UPEC) are the most common pathogens associated with urinary tract infections (UTIs). The formation of, and the dispersal from, intracellular biofilm-like communities (IBCs) is now recognised as an important part of the aetiology of recurrent UTIs. When grown in iron restricted tissue culture medium RPMI 1640, UPEC forms aggregates that are reminiscent of IBCs. We believe RPMI closely mimics a host environment in which biofilm formation is promoted as a survival mechanism. Aggregation and dispersal can be quantified using the Aggregation Index (AI) = (OD Total cells - OD Planktonic cells / OD Total cells). Aggregates are separated from planktonic cells by centrifugation. In culture the aggregates are dispersed by the provision of iron in a process that is inhibited by antibiotics that prevent transcription or translation, suggesting that new gene expression is required to effect dispersal. Aggregates examined by cryo-scanning electron microscopy show a two-dimensional sheet arrangement with bacteria embedded within a matrix suggesting exopolysaccharide production may be the basis of aggregate formation. Aggregates do not form in the presence of cellulase and cellulase rapidly disperses preformed aggregates in the absence of iron. The aggregate matrix stains strongly with Calcofluor White, a fluorescent stain for cellulose. Mutation of the bcsA gene encoding the catalytic subunit of cellulose synthase inhibits aggregation leading to the conclusion that the major aggregate exopolymer is cellulose. We hypothesise that successful iron acquisition by UPEC is an important step in recurrent UTIs, providing signals that favour dispersal. The signal to disperse is not transduced by Fur, a recognised iron-dependent regulator of gene expression; fur mutants aggregate and disperse normally. A mutation in the phosphodiesterase YhjK, immediately downstream of the cellulose biosynthesis operon severely compromises the ability of aggregates to disperse upon provision of iron. YhjK is hypothesised to affect the balance of intracellular cyclic-di-GMP where low levels favour dispersal. Comparison of UPECΔbcsA (no cellulose) and wild type bacteria in antibiotic sensitivity and Galleria mellonella infection demonstrated possible advantages of a cellulose matrix in resistance to antibiotics and innate immunity in urinary tract infection. Comparison of UPECΔyhjK and wild type bacteria further demonstrated the advantages of co-ordinating the regulation of cellulose mediated aggregation and dispersal in the face of antibiotic challenge and G. mellonella infection.