Are Electrical Characterizations Consistent with the Cytochrome Structures ofGeobacter‘Nanowires’

Abstract

AbstractElectrically conductive filaments fromGeobacter sulfurreducenswere reported to be pili with metallic-like conductivity, and yet were later shown to be redox-active cytochromes by cryogenic electron microscopy. It has recently been argued that the filaments were simply misidentified, implying that key observations formerly used to refute the involvement of cytochromes in conductivity now must be ascribed to them. Herein, the temperature, pH, voltage, crystallinity, charge propagation, and aromatic density-related dependencies of the conductivity reported for putative pili are re-examined in light of the CryoEM structures of cytochrome filaments. It is demonstrated that:Electrons flow through cytochrome filaments in a succession of redox reactions for which the energetics are physically constrained and the kinetics are largely independent of protein identity for highly conserved heme packing geometries. Computed heme-to-heme electron transfer rates in cytochrome filaments agree, on average, within a factor of 10 of rates experimentally determined in other multi-heme proteins with the same heme packing geometries.T-stacked heme pairs, which comprise nearly or exactly half of all heme pairs in cytochrome filaments are electronic coupling-constrained bottlenecks for electron transfer that set the rate-limiting reaction to the µs timescale, which isfast enoughcompared to typical ms enzymatic turnover. Tuning the conductivity of cytochromes over the reported ∼107-fold range for filaments fromG. sulfurreducensstrains with pili variants seems both physically implausible and physiologically irrelevant if those filaments are supposed to be cytochromes.The protein-limited flux for redox conduction through a 300-nm filament of T- and slip-stacked heme pairs is predicted to be ∼0.1 pA; aG. sulfurreducenscell discharging ∼1 pA/s would need at least 10 filaments, which is consistent with experimental estimates of filament abundance. The experimental currents for the Omc- S and Z filaments at a physiologically relevant 0.1 V bias, however, are ∼10 pA and ∼10 nA, respectively. Some of the discrepancy is attributable to the experimental conditions of a dehydrated protein adsorbed on a bear Au- electrode that contacts ∼102hemes, and in the case of conducting probe atomic force microscopy, is crushed under forces known to deform and change the electron transport mechanism through more highly-structured proteins.Previously observed hallmarks of synthetic organic metallic-like conductivity ascribed to pili are inconsistent with the structurally resolved cytochrome filaments under physiological conditions, including (I) increased crystallinity promoting electron delocalization, (II) carbon nanotube-like charge propagation, and (III) an exponential increase-then-decrease in conductivity upon cooling, which was only explain by a model predicted on redox potentials known to be experimentally false. Furthermore, spectroscopic structural characterizations of OmcZ that attest to a huge acid-induced transition to a more crystalline state enhancing conductivity either strongly disagree with CryoEM analyses at higher pH values or give inconclusive results that can be overly interpreted.Overall, a significant discrepancy currently exists—not between theory and experiment—but between the CryoEM cytochrome filament structure in one hand and the other functional characterizations ofGeobacter‘nanowires’ in the other. The CryoEM structures, theoretical models, biological experiments, and kinetic analyses are all in agreement about the nature and rate of electron transfer in multi-heme architectures under physiological conditions, and stand opposed to the solid-state functional characterizations ofGeobacterfilaments reported to date. The physiological relevance and/or physical plausibility of some experiments should be examined further.