Identification of Different Putative Outer Membrane Electron Conduits Necessary for Fe(III) Citrate, Fe(III) Oxide, Mn(IV) Oxide, or Electrode Reduction by Geobacter sulfurreducens

OmcBC or ExtABCD is sufficient during Fe(III) citrate reduction; deletion of all clusters eliminates Fe(III) citrate reduction. Shown is growth using 55 mM Fe(III) citrate as an electron acceptor by mutants with single conduit cluster deletions (A), triple mutants lacking all but one cytochrome conduit, as well as the Δ5 strain lacking all five cytochrome conduits (B), mutants in an ΔomcBC background (C), and Δ5 mutants expressing omcB or extABCD or carrying an empty expression vector as control (D). All experiments were conducted in triplicate, and curves are averages ± standard deviations (SDs) from ≥3 replicates.

No single outer membrane cluster is essential but all are necessary for wild-type levels of electron transfer to Fe(III) and Mn(IV) oxides. Shown is growth of single-cluster-deletion mutants and triple mutants lacking all but one cytochrome conduit cluster, as well as the Δ5 mutant lacking all clusters utilizing 70 mM Fe(III) oxide (A and B) or 20 mM Mn(IV) oxide (C and D) as the terminal electron acceptor. All experiments were conducted in triplicate, and curves are averages ± SD from ≥3 replicates.

OmcBC and ExtEFG have additive roles in Fe(III) and Mn(IV) oxide reduction. Shown is reduction of 70 mM Fe(III) oxide (A) or 20 mM Mn(IV) oxide (B) by the ΔomcBC strain and additional deletions in an ΔomcBC background. All experiments were conducted in triplicate, and curves are averages ± SD from ≥3 replicates.

Partial complementation by single conduit clusters supports the hypothesis that multiple conduit complexes are necessary for wild-type levels of metal oxide reduction. Shown is reduction of 70 mM Fe(III) oxide (A) or 20 mM Mn(IV) oxide (B) by the Δ5 mutant expressing extABCD or the omcB cluster compared to the empty vector control. All experiments were conducted in triplicate, and curves are averages ± SD from ≥3 replicates.

Only the ExtABCD conduit cluster is necessary for electrode reduction. Shown is current density produced by single- (A) and multiple (B)-cluster-deletion mutants on graphite electrodes poised at +0.24 V versus SHE. All mutants were grown in at least two separate experiments, and curves are representative of results from ≥3 independent replicates per experiment. Similar results were obtained at lower (−0.1 V versus SHE) redox potentials.

Effect of kanamycin on final current density and comparison of ExtABCD and OmcBC complementation. (A) Final current density of wild-type G. sulfurreducens compared to the wild type carrying an empty vector in the presence of increasing kanamycin concentrations. (B) Current density produced by the Δ5 strain plus either extABCD or omcB cluster-containing vectors in the presence of 5 μg/ml residual kanamycin. Wild-type and Δ5 strains carrying the empty vector were used as controls. All experiments were conducted in duplicate, and curves are representative of results from ≥3 replicates per experiment.

Transcriptomic analysis comparing fumarate versus electrode growth for extABCD⁺ and wild-type strains. (A) Comparison of expression levels of wild-type exponentially growing cells under fumarate- and electrode-respiring conditions, showing no significant up- or downregulation of ext clusters (orange triangles) or most other known electron transfer proteins (red circles). Dark and light gray dotted lines represent thresholds of 4 and 2 log₂, respectively. (B) RPKM and log₂ change of open reading frames with largest expression changes as well as genes studied in this work (for additional data, see Table S2). (C and D) Comparison of the transcriptomes of wild-type and extABCD⁺ cells exponentially growing using fumarate (C) or electrode poised at +240 mV (D) as the terminal electron acceptor, showing no changes to electron transfer proteins due to deletion of omBC, extEFG, and extHIJKL clusters. Averages of biological replicate samples are shown.