Promoting Extracellular Electron Transfer of Shewanella oneidensis MR-1 by Optimizing the Periplasmic Cytochrome c Network

Abstract

The low efficiency of extracellular electron transfer (EET) is a major bottleneck for Shewanella oneidensis MR-1 acting as an electroactive biocatalyst in bioelectrochemical systems. Although it is well established that a periplasmic c-type cytochrome (c-Cyt) network plays a critical role in regulating EET efficiency, the understanding of the network in terms of structure and electron transfer activity is obscure and partial. In this work, we attempted to systematically investigate the impacts of the network components on EET in their absence and overproduction individually in microbial fuel cell (MFC). We found that overexpression of c-Cyt CctA leads to accelerated electron transfer between CymA and the Mtr system, which function as the primary quinol oxidase and the outer-membrane (OM) electron hub in EET. In contrast, NapB, FccA, and TsdB in excess severely impaired EET, reducing EET capacity in MFC by more than 50%. Based on the results from both strategies, a series of engineered strains lacking FccA, NapB, and TsdB in combination while overproducing CctA were tested for a maximally optimized c-Cyt network. A strain depleted of all NapB, FccA, and TsdB with CctA overproduction achieved the highest maximum power density in MFCs (436.5 mW/m²), ∼3.62-fold higher than that of wild type (WT). By revealing that optimization of periplasmic c-Cyt composition is a practical strategy for improving EET efficiency, our work underscores the importance in understanding physiological and electrochemical characteristics of c-Cyts involved in EET.

Keywords: EET; MFC; Shewanella; cytochrome c; genetic engineering.

FIGURE 1

c-Type cytochrome (c-Cyt)-involved extracellular electron transfer (EET) pathways in Shewanella oneidensis. Electrons generated from oxidation of electron donors are conserved in NADH and fed into the quinol pool composed of menaquinone (MQ) and ubiquinone (UQ), which link to different quinol oxidases. CymA, the predominant quinol oxidase for EET via Mtr, is promiscuous for electron acceptors (EAs). This promiscuity is a likely cause of low EET efficiency. Proteins whose names are in black are c-Cyts whereas in white are not. F, fumarate; S, succinate.

FIGURE 2

Physical and bioelectrochemical characterization of MR-1 overproducing CcmFGH. (A) Cell pellet color and levels of heme c of indicated strains grown anaerobically on TMAO. Cells in late-exponential phase cultures were pelleted, photographed, and then lysed for quantification of heme c levels. The average heme c level in MR-1 [the wild type (WT)] was set to 1, to which the heme c levels in other strains were normalized, giving the relative abundance (RA). pccmFGH, controlled expression within plasmid pHGE-Ptac under the control of isopropyl β-D-1-thiogalactoside (IPTG)-inducible promoter Ptac. (B) Output voltage of microbial fuel cells (MFCs) incubated with indicated recombinant and the control (WT, carrying empty vector) strains across the 2,000 Ω external resistor. (C) Polarization curves and power density curves determined by measuring the stable output voltage generated across various external resistances. In panels (B,C), IPTG, 0.5 mM. Three independent operations (biological replicates, throughout the study) were performed, and data are presented by the means ± standard deviation (SD). In panel (A), asterisks indicate statistically significant difference when compared with the WT values (**p < 0.01; ***p < 0.001). In both (B,C), the differences in the peak values between two strains were <0.05.

FIGURE 3

The periplasmic c-Cyts in overabundance interfere with EET efficiency. (A) Cell pellet color and levels of heme c of indicated strains grown anaerobically on TMAO with 0.5 mM IPTG. Data were processed and presented as described in Figure 2A. (B) Methyl orange (MO) kinetic traces of the indicated strains grown with 0.5 mM IPTG. (C) MO kinetic traces of indicated periplasmic c-Cyt single mutants grown with 0.5 mM IPTG. Three independent operations were performed, and data are presented by the means ± SD. In panel (A), asterisks indicate statistically significant difference when compared with the WT values (ns, not significant; *p < 0.05; ***p < 0.001). In both (B,C), the differences in the peak values between two strains were <0.05.

FIGURE 4

Bioelectrochemical characterization of periplasmic c-Cyt mutants in dual-chamber MFCs. Data from WT and indicated periplasmic c-Cyt mutants w/o complementation with 0.5 mM IPTG are shown. (A,C) Output voltage of MFCs incubated with the indicated strains across the 2,000 Ω external resistor. (B,D) Power density curves determined by measuring the stable output voltage generated across various external resistances. Three independent operations were performed, and data are presented by the means ± SD. In all panels, the differences in the peak values between WT and each of periplasmic c-Cyt mutants under test were <0.01.

FIGURE 5

Optimized MFC performance of engineering strains. Data from mutants are given in panels (A–F) show triple mutants overexpressing cctA with 0.5 mM IPTG. (A,C) Output voltage of MFCs incubated with the indicated strains across the 2,000 Ω external resistor. (B) Power density curves determined by measuring the stable output voltage generated across various external resistances. (D) Polarization curves and power density curves (dashed lines) determined by measuring the stable output voltage generated across various external resistances. (E) Cyclic voltammetry (CV) characterization of the MFC with inoculations of the control strain and indicated recombinant strains under turnover condition, respectively. (F) Nyquist plots (0.01 Hz to 100 kHz at the bias potential 0.2 V vs. Ag/AgCl with a 5-mV perturbation signal of different strains in MFCs. Three independent operations were performed, and data are presented by the means ± SD. In all panels, the differences in the peak values between WT and each of mutants under test were <0.01.

FIGURE 6

Impacts of mutations with overexpressed CctA on growth and biofilm structure. (A) Growth of WT and Δtriple/pcctA on TMAO. Cells growing exponentially were inoculated into the fresh MS with 30 mM TMAO as the EA. Growth was monitored by recording the optical density at 600 nm. (B) Biofilm mass of Δtriple/pcctA and WT formed on the anode of MFC. At each time point, the total biomass of the biofilms grown on the anodes was determined by using bicinchoninic acid (BCA) measurement as described in the section “Materials and Methods.” (C,D) Characterization of the anodic biofilms of MFC after output voltage reaching the peak. SEM images of the biofilm of WT (C) and Δtriple/pcctA (with induction of 0.5 mM IPTG) (D). The right panels represent a portion of the images on the left panels enlarged. Three independent operations were performed with data presented in panels (A,B) by the means ± SD and representative images shown in panels (C,D).

FIGURE 7

Impacts of riboflavin (RF) on MFC performance of the optimized strain. (A) Output voltage of MFCs incubated with the Δtriple/pcctA and WT, across the 2,000-Ω external resistor without or with the addition of RF. (B) Polarization curves and power density curves (dashed lines) determined by measuring the stable output voltage generated across various external resistances. (C) MO kinetic traces of the Δtriple/pcctA without or with addition of RF. Three independent operations were performed, and data are presented by the means ± SD. In all panels, the differences in the peak values between the strains under test without or with addition of RF were <0.01.