Bacterial strains, growth conditions, and A. thaliana inoculation

C. metallidurans CH34 was obtained from the Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (Braunschweig, Germany). It was routinely grown in Dorn minimal saline medium (Dorn et al., 1974) containing 10 mM gluconate as a sole carbon and energy source in an orbital shaker (150 rpm) at 30 °C. We collected suspensions of cells that were fully grown to the mid-exponential phase (OD_600 nm = 0.6), adjusted them to approximately 10⁸ colony-forming units per milliliter (CFU mL⁻¹), and diluted them in agar medium just prior to solidification in order to obtain 10⁴ CFU mL⁻¹ for A. thaliana inoculation of a gnotobiotic system. CFU were calculated based on colony counts from serial one mL dilutions (usually 10⁻² to 10⁻⁸) of 100-µl aliquots plated on three R2A agar plate replicates. Ultimately, the CFU values corresponded to the average of colony counts x corresponding dilution × 10 (100 aliquot/1,000 uL dilution volume). To assess the effect of heat-inactivated bacteria, we heated an inoculum suspension at 95 °C for 20 min prior to the final dilution in agar medium. This temperature was high enough to kill all bacterial cells without destroying them (Poupin et al., 2013). Bacterial cell viability was routinely confirmed using R2A medium agar plate counting.