, 2003; Spiers & Rainey, 2005). The involvement of iron in the control of dispersal suggests a role for genes regulated via Fur (Pennella & Giedroc, 2005); however, the impact of other metal ions suggests the involvement of an additional regulatory mechanism. It is tempting to speculate that there may be a role for a metal ion-activated Bortezomib purchase phosphodiesterase (which would degrade cyclic di-GMP) (Bobrov et al., 2005). In addition, we note that UPEC 536 does aggregate in both R and RF during the first few hours of growth following inoculation (Fig. 1), suggesting that in addition to the provision of iron there may be regulatory input associated with entry into the stationary phase (Hengge, 2008) and/or

quorum sensing (Walters & Sperandio, 2006). Iron has been shown to regulate biofilm equilibrium in other bacterial species in ways different from that observed here with UPEC. For example, iron is necessary for biofilm formation by Pseudomonas aeruginosa, and increasing the FeCl3 concentration

up to 10 μM (the concentration used in this study) steadily increases biofilm formation Selleckchem FDA-approved Drug Library (Banin & Greenberg, 2008). In P. fluorescens, the provision of iron (0.1–5 μM), and of copper, lead, and manganese, promotes the production of a cellulosic biofilm matrix with a weak, easily disrupted structure. The provision of higher levels of iron (50 μM) results in the production of a stronger cellulose matrix (Koza et al., 2009). The divergent effects of biofilm responses to the available iron may reflect important differences in the ecology of the bacterium and its host interaction that we do not yet understand. In a UPEC infection scenario where bacteria are successful in the struggle for iron, our results support the initiation of dispersion from the cellulose matrix, which we propose may involve the release Cyclic nucleotide phosphodiesterase of glucose from cellulose that can be used an energy source and that may also affect subsequent gene expression mediated by cAMP–CAP (Weyand et al., 2001). To conclude, it is important

to consider the implications of cellulose as part of the IBC/QIR matrix. Iron starvation of UPEC induces the synthesis of the cellulose matrix with properties that will aid attachment (Wang et al., 2006; Saldaña et al., 2009), and presumably provide a protective matrix favouring survival in the presence of both antibiotics and immune defences. Rosen et al. (2007) found a clear association with the presence of IBCs in urine and a UPEC UTI. IBCs were not found in the urine of uninfected individuals, and those infected with other species of bacteria. We found that seven of 12 clinical isolates (two of the nonaggregative isolates were from asymptomatic patients) formed aggregates when cultured in R, in agreement with the finding that many pathogenic isolates form cellulose at 37 °C (Bokranz et al., 2005; Da Re & Ghigo, 2006; Monteiro et al., 2009). Given that Reisner et al. (2006) found no association with the ability of E.