, 2009). Putative

mutants were selected on NA with Km at 50 μg mL−1, and verified by Southern blot. Loss of swimming motility was confirmed in soft agar (0.3%) plates and under the microscope (not shown). MFCs were fabricated as described previously (De la Fuente et al., 2007b). Briefly, the chamber body was constructed with polydimethylsilioxane and consisted of two parallel channels measuring 80 μm wide, 3.7 cm long and 50 μm high, separated by a 50 μm wide polydimethylsilioxane ridge. Chamber bodies were then sandwiched between a cover glass and a supporting glass microscope slide. Teflon tubes were attached to inlet and outlet channels, and media were introduced into the channels using syringes controlled by pumps (Pico Plus, Harvard Apparatus). The chambers were mounted on a Nikon Dabrafenib Ti/U E20L80 microscope (Nikon Co.) using 40 × phase-contrast and differential interference contrast optics. Time-lapse images were recorded using a DS-Qi1Mc digital camera and analyzed using nis elements software (Nikon Co.). The adhesion abilities of bacterial cells were evaluated using a modification of a described procedure (De La Fuente et al., 2007b): (1) cells were introduced from side channels, while the flow in the

main channels was stopped, allowing cells to attach; (2) introduction of cells from the side channels was see more stopped and medium flow in the main channels was resumed at a rate of 0.25 μL min−1 to remove unattached Thymidine kinase cells; and (3) the flow rate in the main channels was gradually increased from 0.25 to 0.5, 1, 2, 4, 8, 16, 32 and 64 μL min−1, each rate being maintained

for one minute. Time-lapse movies were captured during the course of the assay and cells attached to the glass surface were quantified using nis elements software. Each repetition of steps 1–3 was considered a replicate. For each strain, at least three replicates in different locations along the channels were measured. For each flow rate, the amount of cells washed from the field of view was calculated as a function of the total number of cells present at the beginning of the assay. At the end of each flow rate, the number of attached cells was determined by averaging the amount of attached cells in the last three frames of that time period (corresponding to the last 15 s of the corresponding flow rate). Adhesion forces were determined according to De La Fuente et al. (2007b). Biofilm formation was monitored inside the MFCs by maintaining a flow rate of 0.25 μL min−1 in the main channel and capturing images at 30-s intervals for a period of 6–24 h. Swimming and twitching were assessed for all strains inside the MFCs. Twitching motility rates were calculated for six bacterial cells according to De La Fuente et al. (2007a). All experiments were repeated at least three times and data were subjected to the Tukey–HSD test using jmp in v3.2.1 (SAS Institute Inc.). For comparison of adhesion forces, one-way anova were performed using statistix 8.