Cells were harvested at a middle logarithmic growth phase and washed with phosphate-buffered saline. Bacterial cells Y-27632 and MnO2 particles were separated by a Percoll (GE Healthcare) density-gradient centrifugation according to a method described elsewhere (Page & Huyer, 1984). An iron

content of bacterial cells was determined according to a colorimetric method described elsewhere (Page, 1995) with modifications. Cells were suspended in 25 μL of 7% perchloric acid and extracted overnight at room temperature, followed by the extraction for 4 h at 90 °C. The extract was mixed with 5 μL of 0.1 M ascorbic acid, 140 μL of 2 mM ferrozine solution, and 30 μL of 0.1 M NaOH. An iron content was normalized to a total protein concentration determined using a Micro BCA protein-assay kit (Pierce). After Shewanella cells were grown in LMM under a MnO2-reducing condition, they were lysed in a detergent solution containing 5% (v/v) Triton X-100 and 50 mM HEPES (pH7.4). Cell lysates were subjected to a spectrometric assay to determine c-cyt contents (Myers & Myers, 1992). A content was estimated from a difference in absorbances of the α peak (at 552 nm) between dithionite-reduced and air-oxidized samples, and a specific content was estimated by normalizing a protein content. Shewanella cells were grown anaerobically in LMM under fumarate- or MnO2-reducing condition, and cells were harvested in exponential

log phases. RNA was extracted using a Trizol reagent (Invitrogen) and subsequently purified using an RNeasy Mini kit and RNase-Free Lumacaftor chemical structure DNase set (Qiagen). RT-PCR and subsequent quantitative PCR were carried out using a LightCycler 1.5 instrument (Roche) with PCR primers listed in Table S1. Standard curves were drawn using dilutions of PCR fragments of target genes (omcA, mtrC, SO3032, and 16S rRNA gene). A specificity of the quantitative PCR was verified by dissociation-curve analyses. An mRNA level of a target gene (omcA, mtrC, or SO3032) was normalized to that of the 16S rRNA gene. After screening of approximately 5000 random Tn-insertion mutants, we obtained one mutant (N22-7) that

generated a smaller halo around its colony (the reduction of brown MnO2 to colorless Mn2+ resulted in the formation of a halo) than the wild-type Rutecarpine MR-1 (WT). An ability of N22-7 to reduce MnO2 was also analyzed in liquid cultures and compared with that of WT (Fig. 1). Figure 1a presents appearances of 96-h cultures in the LMM/MnO2 liquid medium, showing that N22-7 was deficient in MnO2 reduction. Figure 1b shows time courses of MnO2 reduction in the liquid cultures (initial OD600 nm of 0.01), indicating that a MnO2-reduction rate of N22-7 (117 ± 15 μM h−1) was approximately half that of WT (230 ± 30 μM h−1). In contrast, when MnO2-reduction assays were initiated by inoculating with higher concentrations of cells (initial OD600 nm of 0.