Bacterial competition assays.

Bacterial cultures were grown overnight in liquid LB medium at 37°C with shaking. Overnight cultures were back-diluted and incubated in liquid LB medium at 37°C with shaking for 3 h. Bacterial cultures were then normalized to an optical density at 600 nm (OD₆₀₀) of 1. If strains harbored plasmids, cultures were grown overnight with antibiotics to maintain plasmids and 100 μM isopropyl-β-d-thiogalactopyranoside (IPTG) if plasmids contained an inducible promoter. If strains were grown in medium containing antibiotics, liquid cultures were then washed three times with fresh LB medium before they were cocultured. For bacterial competitions performed on LB agar medium with 0.4% glucose, overnight and back-diluted cultures were also grown in LB medium supplemented with 0.4% glucose. A 50-μl mixture aliquot of a 10:1 killer/target cell ratio was spotted on a 0.22-μm-pore-size filter paper, which was placed on LB agar medium and incubated at 37°C. After 3 h, filters were vortexed in 5 ml of sterile LB medium for 30 s. One hundred microliters of serial dilutions were then spread (or 3 μl or serial dilutions were spotted) on plates containing the required antibiotic to select for target cells. Data from three cocultures were used to determine significance. Results are representative of at least two independent experiments.