The life cycle of a biofilm is a continuous, highly dynamic process. Biofilms develop through several stages including microbial attachment on surfaces or interfaces, biofilm maturation and biofilm dispersal (Costerton, 1999; O'Toole et al., 2000; Watnick and Kolter, 2000; Flemming et al., 2016). Similar to other biofilm-forming bacteria, exoelectrogens first physically contact and attach to the surfaces of external electron acceptors, such as metal particles, electrode surfaces, or partner bacteria. Attachment stages of Geobacter and Shewanella biofilms are mediated by functional flagellar system, type IV pilus system and extracellular DNA (eDNA) (Thormann et al., 2004; Gödeke et al., 2011a; Gödeke et al., 2011b; Cologgi et al., 2014; Bennett et al., 2016). Then, biofilms become mature, during which cells form a fixed EPS matrix (Allison, 2003; Flemming and Wingender, 2010). Different from that of nonelectrogenic bacteria, the biofilm matrix of Geobacter and Shewanella spp. is electrically conductive and contains extracellular electron transfer components, such as cytochromes and other electron mediators (Borole et al., 2011; Reguera, 2018; Xu et al., 2018). This is required for the growth of daughter cells or new cell layers at a distance from external electron accepters. The final stage of a biofilm life cycle is biofilm dispersal in which the planktonic cells are released from a mature biofilm, migrate and attach to new surfaces, and subsequently start a new life cycle (McDougald et al., 2012; Rumbaugh and Sauer, 2020).

Cells in a mature biofilm are enclosed in a self-produced EPS matrix that occupies most of biofilm biomass (Allison, 2003). EPS matrix is composed of soluble, gel-forming components such as polysaccharides, proteins and extracellular DNA (eDNA) (Flemming and Wingender, 2010), as well as insoluble components including cellulose, amyloid, flagella and pili (Hobley et al., 2015). By filling and shaping the space between the cells of biofilms, EPS molecules offer a microenvironment for biofilm cells and provide a structural scaffold for biofilm stability (Persat et al., 2015). EPS matrix may also form an external digestion system due to its ability to accumulate enzymes (Tielen et al., 2010). These enzymes, such as protein-, polysaccharide-, oligosaccharide-, or lipid-degrading enzymes and phosphomonoesterases, can degrade diverse matrix components or other nutrients (Allison, 2003; Flemming and Wingender, 2010). After degradation, the products in close proximity to cells are efficiently captured and taken up by cells. Another important role of EPS matrix is to serve as a protective shield against toxins, antimicrobials and desiccation (Flemming and Wingender, 2010). Thus, the matrix is a functionally crucial component of microbial biofilms.

Shewanella matrix contains extracellular polysaccharides, proteins and eDNA (Figure 2) (Mukherjee et al., 2020). Proteomics analysis of EPS extracted from Shewanella sp. HRCR-1 biofilm found the presence of 58 extracellular and outer membrane proteins, such as redox active proteins MtrC and OmcA (Cao et al., 2011). The presence of c-Cyts in the biofilm matrix of S. oneidensis MR-1 is further confirmed by a recent study that detected the electrochemical activities and redox peaks of c-Cyts using voltammetry measurement (Xiao et al., 2017). This study also clearly showed the redox peaks of flavins in EPS of S. oneidensis MR-1. The EPS matrix of Shewanella biofilm likely affects extracellular electron transfer through biofilm via redox proteins and electron shuttles.

It has been reported that Geobacter spp. are among the major phylotypes in the attached fraction of environment samples (Dong et al., 2021). Geobacter biofilms have an electroactive EPS matrix with matrix-associated c-Cyts and pili that spatially organize cells well for growth and electron transfer to mineral oxides and electrodes (Figure 2). The studies that showed the reversible reduction and oxidation of the cytochromes demonstrated their functionality as electron carriers for electron transfer in electricity-producing G. sulfurreducens biofilms (Liu et al., 2011; Liu and Bond, 2012). The roles of outer membrane cytochromes in biofilm electroactivity were further investigated by the studies that constructed diverse cytochrome-defective mutants and tested their current outputs. Deleting outer membrane c-Cyt gene omcE or omcB from the G. sulfurreducens genome has no effect on electricity outputs of MFCs (Holmes et al., 2006; Nevin et al., 2009; Richter et al., 2009). Outer membrane c-Cyt OmcS is required for optimum current production in fuel cells in which electrochemical restrictions limit electricity outputs (Holmes et al., 2006), whereas deleting omcS has no impact on current production when there is no electrochemical limitation (Nevin et al., 2009). However, the genetic deletion of OmcZ leads to a significant reduction in biofilm electroactivity and electricity output (Nevin et al., 2009; Richter et al., 2009). OmcZ preferentially locates at or closer (<10 µm) to the electrode surface in the anodic G. sulfurreducens biofilm (Inoue et al., 2011), suggesting a critical role of OmcZ in the last step of EET, i.e., electron transfer from biofilms to electrodes. Moreover, a recent study discovered that the production of OmcZ can be simulated by electric field (Yalcin et al., 2020). OmcZ has been identified to be an essential matrix-associated c-Cyt for electron transfer in thick biofilms. Other matrix-associated c-Cyts may be required, but compensatory effects observed in encoding gene-defective strains make their genetic identification challenging (Kim et al., 2005; Mehta et al., 2005). For example, the compensatory effect of OmcZ nanowires probably makes ΔomcS G. sulfurreducens biofilms show high current production when the biofilms mature. Thus, OmcZ and OmcS nanowires may have overlapping roles in G. sulfurreducens MFC performance (Nevin et al., 2009). Based on these previous studies, Yelcin et al. proposed a model that OmcS nanowires play an important role in EET during the early stage of biofilm growth, while the high conductivity of OmcZ nanowires enables G. sulfurreducens to form thick biofilms on the electrode surface of MFC (Figure 2) (Yalcin and Malvankar, 2020). On the other hand, Steidl et al. (2016) found that G. sulfurreducens pili have critical impacts on the formation of relatively thick (>10 µm) biofilms on the electrodes.

Steep chemical gradients exist in the EPS matrix of microbial biofilms (Chang et al., 2015). The microbial cells living in these heterogenous environments display different metabolic activities for their adaption to different local environmental conditions by expressing distinct set of genes (Stewart and Franklin, 2008). Cao et al. monitored the spatiotemporal metabolic responses of metabolically active S. oneidensis MR-1 biofilms to contaminant exposure, which demonstrated that metabolic responses were spatially stratified based on local microenvironments (Cao et al., 2012). In an aerobic heterotrophic biofilm, organisms are stratified based on oxygen and nutrient availabilities, which become depleted in the lower layers of the biofilm when the consumption rates of oxygen and nutrients in the upper layers of the biofilm are faster than their diffusion rates (Stewart, 2003). The cells in the upper layers of the biofilm often have higher metabolic activities because they have optimal access to nutrients and electron acceptors.

Unlike a biofilm on a surface, the cells in the upper layers of the anode-respiring biofilm have optimal access to soluble electron donors, nutrients and buffering agents, whereas cell layers close to electrodes have optimal access to electron acceptors (Bond et al., 2002). Optimizing electrode-respiring biofilm systems for practical applications greatly requires the understanding of their internal chemical and electrical fluxes including nutrient diffusion, electron transfer and the movement of protons and pH-buffering compounds (Renslow R. S. et al., 2013). Babauta et al. used microelectrodes to quantify pH and redox potential variations in electrode-respiring S. oneidensis MR-1 and G. sulfurreducens biofilms (Babauta et al., 2011; Babauta et al., 2012). Their results demonstrated that observable pH variation was not detected, and that the redox potential decreased toward the bottom of the biofilm in the S. oneidensis MR-1 biofilm. It implies that there is no proton transfer limitation in S. oneidensis MR-1 biofilms. On the other hand, they found that pH continued to decrease in the G. sulfurreducens biofilm through different growth phase, suggesting that pH was not always a limiting factor for biofilm growth. Physiological stratification in electricity-producing G. sulfurreducens biofilms was further investigated by the study showing that the progressive accumulation of cells did not contribute to current production after a maximal current was achieved (Schrott et al., 2014). Furthermore, Chadwick et al. revealed metabolic stratification within current-producing G. sulfurreducens biofilms using stable isotope probing coupled to nanoscale secondary ion mass spectrometry (nanoSIMS) (Chadwick et al., 2019). The results demonstrated that cells at the anode surface were most active, and that cell activity decreased with an increase in distance from electrodes. This work also found that the cells nearest the electrode could keep their maximum growth rate in 80-μm-thick biofilms, suggesting that nutrient or buffer diffusion into the biofilm was not rate-limiting. In another interesting study, azo dye was used as an alternative electron acceptor in MFCs with Shewanella decolorationis (Yang et al., 2015). They observed electron acceptor-dependent physiological stratification within anode biofilm in the presence of multiple electron acceptors. Collectively, these findings demonstrate the physicochemical and physiological stratifications within electrode-respiring biofilms of Geobacter and Shewanella spp. It helps improve understanding of the electron transfer mechanism within anode-respiring biofilm.

Bis-(3′-5′)-cyclic dimeric GMP (c-di-GMP) is a key regulator in biofilm formation (Hengge, 2009). It is respectively synthesized and degraded by diguanylate cyclases (DGCs) containing GGDEF domain and phosphodiesterases (PDEs) harboring EAL/HD-GYP domain (Figure 3) (Jenal et al., 2017). The dynamic change of c-di-GMP level is converted ultimately into specific cellular responses by mediating the responsive effectors, such as riboswitches and c-di-GMP receipting proteins (Römling et al., 2005; Hengge, 2009).

Environmental cues, such as nitric oxide, affect biofilm formation through the turnover of c-di-GMP signaling network, which has been discovered in both Shewanella woodyi and S. oneidensis (Liu et al., 2012; Plate and Marletta, 2012). It indicates that c-di-GMP is a crucial intracellular regulator for controlling formation of Shewanella biofilm. Moreover, a high abundance of genes encoding DGCs and PDEs in Shewanella genomes further emphasize the importance of c-di-GMP signaling network in the regulation of physiological processes of Shewanella spp. (Mukherjee et al., 2020). For example, S. oneidensis are predicted to carry 51 GGDEF-containing proteins, 27 EAL-containing proteins, and 20 proteins that contain both GGDEF and EAL/HD-GYP. In Shewanella spp., c-di-GMP mediates biofilm formation by interacting with polysaccharide biosynthetic enzymes (e.g., MxdB glycosyltransferase) to activate the biosynthesis of extracellular polysaccharides (Gambari et al., 2019). Previous study showed that c-di-GMP-responsive transcriptional factor FlrA controls the expression of genes for the synthesis of flagella and an adhesin protein BpfA (Figure 3) (Cheng et al., 2017; Gao et al., 2018). In addition to mediating the biofilm development, a recent study showed that c-di-GMP signaling network regulates the expression of c-Cyts, such as CymA, OmcA and MtrCAB, in S. oneidensis MR-1, which provides new insights into the physiological functions of c-di-GMP in Shewanella spp. (Ng et al., 2020) (Figure 3). Notably, a cyclic adenosine 3′,5′-monophosphate (cAMP) also commonly exists in Shewanella spp. Previous study demonstrated that cAMP-CRP (cyclic adenosine 3′,5′-monophosphate receptor protein) regulatory system regulates multiple bidirectional EET-related pathways (Saffarini et al., 2003; Charania et al., 2009). Collectively, cyclic dinucleotides (CDNs) regulate both Shewanella biofilm formation and the expression of EET-related genes or pathways, such as c-Cyts and flavin biosynthesis pathway.

Numerous genes that encode GGDEF, EAL, and both GGDEF and EAL/HD-GYP domain-containing proteins are found in Geobacter genomes. Deletion of esnD gene encoding a DGC reduces c-di-GMP levels in the cells of G. sulfurreducens. This disrupts the growth of G. sulfurreducens biofilm on −0.1 V vs. SHE electrodes, and decreases the current levels (Hallberg et al., 2019). Leang et al. deleted the gene GSU1240 of G. sulfurreducens, which was predicted to encode a protein with the PilZ domain (Figure 3). They found that the GSU1240-deficient mutant formed a highly cohesive and more conductive biofilm and produced 70% higher power densities than the wild-type strain (Leang et al., 2013). Moreover, Tremblay et al. (2011) found that a nucleotide insertion in a GEMM (Genes for the Environment, for Membranes and for Motility) riboswitch of G. sulfurreducens increased the expression of its downstream pgcA gene. This results in an enhanced Fe(III) reduction rate (Figure 3). GEMM riboswitch is a typical c-di-GMP binding riboswitch. Similar to G. sulfurreducens, the G. metallireducens genome also contains GEMM riboswitches and a high abundance of genes encoding the proteins with GGDEF, EAL, and both GGDEF and EAL domain (Kellenberger et al., 2015), indicating the important role of c-di-GMP signaling network in Geobacter spp. Despite these advances, our understanding of how c-di-GMP impacts the formation of Geobacter biofilm is far from complete.

Geobacter spp. produce another CDNs, i.e., cyclic AMP-GMP (cGAMP or 3′, 3′-cGAMP). Thirty years ago, the GGDEF domain-containing enzymes were considered as c-di-GMP synthases. However, Hallberg et al. recently found that the hybrid promiscuous GGDEF enzyme GacA specifically produced cGAMP in G. sulfurreducens (Hallberg et al., 2019). Moreover, GEMM riboswitches preferentially recognize cGAMP (Kellenberger et al., 2015; Nelson et al., 2015). GEMM riboswitches are conserved upstream of EET-associated genes, such as omcST, pilMNPPQ and omcAHG in G. sulfurreducens (Hallberg et al., 2019). These observations suggest a possible role of cGAMP in regulating EET process of G. sulfurreducens. Hallberg et al. found that the ΔgacA deletion strain of G. sulfurreducens was defective in the reduction of insoluble Fe(III) oxide particles, demonstrating the importance of cGAMP in the growth of G. sulfurreducens with Fe(III) oxides as electron acceptors (Hallberg et al., 2019).

Microbes or microbial communities can synchronize behaviors in response to population density by detecting the accumulation of self-produced extracellular chemical signals. This process is known as quorum sensing (QS). A high cell density allows accumulation of the signals in a local environment to the levels that are required to activate signal-responsive transcriptional factors via specific binding. The activated transcriptional factors then regulate the expressions of the genes with QS responsive promoters (Mukherjee and Bassler, 2019; Mukherjee et al., 2020). Quorum sensing occurs in both Gram-positive and negative bacteria via a diversity of chemical signals typically classified to acyl homoserine lactones (AHLs), fautoinducer-2 (AI2), diffusible signal factor (DSF) and peptides (Papenfort and Bassler, 2016). Abundant literatures show that QS participates in different stages of biofilm development for different species (O'Toole et al., 2000; Flemming et al., 2016; Rumbaugh and Sauer, 2020).

One of the most well-studied QS system in Gram-negative bacteria are AHL-based QS system that is characterized by two core proteins, the LuxI-type protein (i.e., the AHL synthase), and the LuxR-type protein (i.e., the AHL-responsive transcriptional factor) (Papenfort and Bassler, 2016). LuxI and LuxR homologs were found in Geobacter uraniireducens genome (Gelencsér et al., 2012). Chen et al. found that the addition of endogenous AHLs enhanced the electroactivity of a mix-culture biofilm grown on the anodes of MFCs most likely by increasing the relative abundance of electroactive bacteria assigned to Geobacter spp. (Chen et al., 2017). The role of AHL signal in a pure culture of Geobacter spp. was further investigated. The results showed that the detectable self-secreted AHLs of Geobacter soli GSS01 enhanced the formation and electrochemical activity of the biofilm grown on the anode of MFC. Exogenous addition of AHLs further facilitated EPS production (Jing et al., 2019). They also showed that AHL addition increased the biomass, cell viability and EPS abundance of G. soli GSS01 biofilm grown on the cathode. Their studies demonstrated, for the first time, the critical role of AHL-type signal in the formation and electroactivity of Geobacter biofilm (Jing et al., 2019). Moreover, AHL was detected in cocultured aggregates of G. metallireducens and G. sulfurreducens, indicating AHL-based QS system may regulate Geobacter syntrophic growth and aggregation (Wei et al., 2017).

The QS systems in Shewanella spp. are well summarized in a recent review (Mukherjee et al., 2020). Briefly, several LuxR homologous genes are predicted in the genomes of Shewanella spp. including S. woodyi MS32 and S. baltica (Li J. et al., 2020; Hayek et al., 2020). The genome of S. oneidensis is predicted to contain seven putative LuxR family proteins. However, they are not LuxR-type QS transcriptional factors (Learman, 2008). Different from Geobacter spp., detectable AHL signals are rarely present in Shewanella spp. (Zhang et al., 2017; Mukherjee et al., 2020). Moreover, there has been no evidence showing the involvement of QS systems in formation of Shewanella biofilm.

MFCs show potential to power electronic devices in remote environments such as sea-floors and to remove organic wastes from contaminated areas in conjunction of electricity generation (Lovley, 2006b; Logan, 2009). But most perceived practical applications are limited by the low current output of MFC. The previous study has shown that the conductivity of the biofilm grown on anodes greatly affected G. sulfurreducens MFC performance (Malvankar et al., 2012a). The conductivity through G. sulfurreducens biofilm is mainly attributed to cytochromes and nanowires that mediate the short- and long-range electron transfer between cells and electrodes (Reguera et al., 2006; Richter et al., 2009; Malvankar et al., 2011). The expression of G. sulfurreducens pili in Pseudomonas aeruginosa resulted in a dramatically increased bioelectricity output of P. aeruginosa MFCs (Liu et al., 2019a). In contrast, Liu et al. expressed P. aeruginosa type IV Pili into G. sulfurreducens, resulting in the decreased electricity output (Liu X. et al., 2014). Yi et al. used a selective strategy to obtain a strain of G. sulfurreducens KN400 that showed a lower internal resistance and generated a higher electricity output than G. sulfurreducens DL-1. The increased electricity output is due to the increased production of pili (Yi et al., 2009). Additionally, the expression of flagella enables G. sulfurreducens KN400 to form a thicker biofilm on the anode as compared to G. sulfurreducens DL-1 without flagella. Although they are nonconductive, flagella also contribute to the conductivity of G. sulfurreducens biofilm. Liu and his collaborators found that flagella can act as Geobacter biofilm scaffolds to stabilize the biofilm and anchor extracellular cytochromes with an orderly arrangement, which facilitates extracellular electron transfer (Liu et al., 2019b). In a recent study, overexpression of GSU1501 (a gene for ATP-dependent exporter within an operon of polysaccharide biosynthesis) promoted polysaccharide secretion. The enhanced production of polysaccharides stimulated the formation of the biofilm with higher cell viability. This enhanced the current generation of G. sulfurreducens MFCs (Zhuang et al., 2020). Collectively, these studies demonstrate that biofilm matrix is a desirable target to engineer Geobacter biofilm for improving MFC performance.

The switch of bacterial lifestyle between planktonic and sessile modes is often mediated by sophisticated signaling networks that control intracellular levels of small molecules, such as c-di-GMP and quorum sensing signaling (Parsek and Greenberg, 2005; Valentini and Filloux, 2016). Therefore, the manipulations of these signaling networks can efficiently engineer biofilms to improve biofilm-based bioprocesses. Liu et al. and Hu et al. enhanced the formation of S. oneidensis biofilm on the anodes by overexpressing a DGC YdeH and a near-infrared-light responsive DGC BphS in S. oneidensis, respectively (Liu T. et al., 2015; Hu et al., 2017). The overexpression of these two genes enhanced the biofilm formation and increased the electricity generation of S. oneidensis MFCs. Cheng et al. functionally expressed the gene for an adenylate cyclase in S. oneidensis and used the engineered strain in microbial electrochemical systems (Cheng et al., 2020). The engineered strain exhibited the enhanced bidirectional EET capacities and the enhanced rate of reducing Cr(VI). The biofilm of G. sulfurreducens was engineered by the manipulation of c-di-GMP effectors. The deletion of the gene encoding a c-di-GMP binding effector resulted in the formation of a thicker and much more conductive G. sulfurreducens biofilm, which subsequently increased the power density of MFCs (Leang et al., 2013).

QS systems are also targeted to engineer S. oneidensis biofilms. Hu et al. used LuxR/AHL QS system originated from Vibrio fischeri to construct a gene circuit of AND logic gate. This circuit includes three parts: an isopropyl β-D- thiogalactopyranoside (IPTG)-inducible module, a QS module and an output module (Hu et al., 2015). IPTG inducible module is composed of an IPTG inducible promoter to control the expression of the transcriptional factor gene luxR. The expressed LuxR is activated by AHL to drive the expression of the gene, i.e., mtrA, in the output module. They introduced this gene circuit into S. oneidensis mtrA-deletion mutant and applied the engineered strain in MFCs. The presence of both IPTG and AHL activates MtrA expression, which restores the EET pathway and increases electricity output (Hu et al., 2015). In another study, the lux QS system was used to construct a population-state decision system for intelligently reprogramming the EET regulation system (Li F.-H. et al., 2020). The EET-related genes or pathways, such as cymA, shuttle exporter genes or flavin biosynthesis pathway, were located at the downstream of QS system and could be expressed at a certain population-state threshold. This system permitted the bacterial growth first and then shifted the metabolic flux form bacterial growth to EET. The engineered strain exhibited a higher EET efficiency and an improved pollutant reduction ability (Li F.-H. et al., 2020).

Probably because of nutrient diffusion limitation and insufficient interaction between bacteria and electrodes, the thickness of the biofilms on the anode is often limited to several tens of micron (Renslow R. et al., 2013). To circumvent this limitation, the biofilms are prepared by embedding microbial cells in a matrix made of synthetic polymers. The biofilms prepared in this way is often referred to as “artificial electroactive biofilm” (Adachi et al., 2008). The materials for the building up of an artificial electroactive biofilm are required to provide a high cell viability and ensure an efficient electrical contact between cells and materials. Yu et al. (2011) designed an artificial S. oneidensis biofilm with a conductive matrix interlocked with the conductive polymer chains of polypyrrole. Similarly, ready-to-use artificial bioelectrodes were constructed through the immobilization of G. sulfurreducens cells in composite materials associating with silica gel and carbon felt fibers (Estevez-Canales et al., 2018). In another study by Yong et al. (2014) a self-assembled 3D reduced graphene oxide (rGO)/S. oneidensis hybrid biofilm was formed by a “fishing” process where the GO nanosheets serving as nets caught the bacterial “fish.” Compared to a natural biofilm, this artificial electroactive biofilm exhibits a dramatical increase in power densities due to enhanced electrical conductivity of biofilms. However, the further improvement of BES performance is largely limited by sluggish biotic/abiotic interfacial electron exchange between cells and materials. To improve electron exchange between living cells and abiotic surfaces, a recent study developed a single cell electron collector. In this collector, FeS nanoparticles are confined in the periplasm and on the outer membrane surface to in-situ build an interconnected intact conductive layer on and cross the individual cell membrane (Yu et al., 2020). A single cell electron collector permits intimate association of the conductive nanoparticles with the cellular electron transfer machinery and maximizes the interfacial area between microbial cells and electrodes, which substantially enhanced high electron transfer rate and BES performance. Similarly, by introducing trans-outer-membrane silver nanoparticles, Cao et al. (2021) obtained the highest power density of Shewanella MFCs reported to date. Collectively, all of these results demonstrate a promising and new platform for wiring-up cells with abiotic interface.

Biofilm is the prevalent lifestyle of microbial growth in natural and engineered environments (Flemming et al., 2016). The electroactive Geobacter and Shewanella biofilms have attracted increasing attention from academic and industrial communities because of their promise in various biotechnological applications, such as bioelectricity generation via MFCs. However, the low electricity output largely limits the practical applications of MFCs. The improved understanding of their EET molecular mechanisms and biofilm biology have greatly helped Geobacter and Shewanella biofilm engineering for improving MFC performance. The rapid advance of synthetic biology has also significantly helped the genetical modification of Geobacter and Shewanella spp. (Leang et al., 2013; Li et al., 2018; Cao et al., 2019). Furthermore, conductive metal nanoparticles have been used to improve bacteria loading capacity and the conductivities of electrodes and microbial cells. However, few studies couple genetically modified living cell catalysis and material catalysts. Thus, future investigation should focus on the genetically engineered cell-nanoparticle hybrid systems to further push the limits of MFCs.

Most studies of Geobacter and Shewanella biofilms focus on their biofilm formation on the anodes of MFCs for electricity generation. Although it is still at proof-of-principle stage, Geobacter and Shewanella spp. as the two best studied groups of exoelectrogens also show the promise to grow on the cathodes for the synthesis of valuable chemicals by using electrode-feeding electrons as reducing power and CO₂ as a feedstock (i.e., electrobiosynthesis) (Dumas et al., 2008; Ross et al., 2011; Soussan et al., 2013; Ueki et al., 2018). This technology is attractive for both the alleviation of the world demands for fossil-based fuel and the decrease of CO₂ emissions. The future research should also focus on the Geobacter and Shewanella biofilms on the cathodes. Previous studies demonstrated that both Geobacter and Shewanella cells express different cytochromes and regulate metabolism pathways when they are grown at different electrode potentials (Levar et al., 2014; Hirose et al., 2018). Thus, one of future directions is to investigate the mechanism of electron transfer from cathodes to the cells. Furthermore, the strategies to engineer Geobacter and Shewanella biofilms formed on the cathodes for the enhanced performance of electrobiosynthesis should also be explored.