Elucidating energy conversion pathways at biotic/abiotic interfaces in microbe-semiconductor hybrids

Abstract

Biotic/abiotic hybrid systems integrating microbes with light-absorbing semiconductor materials offer promising solutions for sustainable energy conversion and value-added chemical production. In this perspective, we discuss the mechanistic insights into upstream energy conversion processes at the biotic-abiotic interfaces, underscoring their pivotal roles in determining biohybrid performance. We explore how biological, physicochemical, and electrochemical characterization techniques have advanced our understanding of energy conversion pathways and electron transport mechanisms within these complex systems. Moreover, we emphasize the growing importance of spatiotemporally resolved imaging in linking biological activity with physicochemical dynamics at the single-cell level. Moving forward, we propose that interdisciplinary collaborations and innovative methodologies will be critical in deepening mechanistic understanding and unlocking the full potential of artificial photosynthetic biohybrid systems.

Graphical Abstract

1. INTRODUCTION

As global sustainability challenges intensify, the need for clean and sustainable energy is becoming increasingly urgent.¹ To address this issue and achieve environmentally friendly energy conversion, hybrid systems composed of green biocatalysts and inorganic materials have emerged as promising candidates.^2–6 Solar energy or electricity is harnessed and provided to biocatalysts for the synthesis of a wide range of value-added products without producing excess waste. Typically, these hybrid systems are categorized into three types: single-enzyme^{7, 8}, multi-enzyme^{9, 10}, and whole-cell type^11–18. For single or multi-enzyme hybrids, metals such as copper and cobalt are combined with enzymes like cytochrome P450 to facilitate electron transfer and catalysis.^{19, 20} While these hybrids exhibit efficient electron transfer and simple configurations, they require laborious protein purification, suffer from low enzyme stability, and rely on limited inorganic supports such as ITO and Mo-doped BiVO₄. In contrast, whole-cell hybrids integrate living microbes with electrodes or semiconductor nanoparticles (e.g., CdS) to form versatile microbe–inorganic complexes. This design leverages microbial self-regeneration and avoids enzyme isolation, reducing operational complexity and cost.

Microbe–semiconductor hybrids can be classified as decoupled or integrated systems with distinct energy conversion pathways. In decoupled hybrids, hydrogen is produced by water electrolysis via inorganic catalysts and subsequently used by microbes for biosynthesis.²¹ In integrated hybrids, light-activated semiconductors generate electron–hole pairs, and photoexcited electrons are transferred to microbes such as Geobacter sulfurreducens¹¹ and Shewanella oneidensis MR-1¹⁴ facilitating reactions such as N₂ reduction to NH₃ or CO₂ to acetate via acetyl-CoA. Decoupled hybrids mainly rely on the metabolism of hydrogen gas, which may pose a safety concern once scaled up in manufacture; the integrated hybrids, on the other hand, require highly efficient electron transfer at the inorganic-microbe interface for the eventual energy conversion, which necessities an in-depth mechanistic understanding and careful design.

Extracellular electron transfer (EET), particularly transmembrane transport from semiconductors to microbes, is a key determinant of biosynthesis efficiency. Electron transport occurs via either indirect or direct pathways. In the indirect pathway, redox mediators facilitate electron exchange at the bacteria-semiconductor interface. In the direct pathway, electrons are directly transferred by conductive protein structures, e.g., outer membrane pilis, cytochromes such as OmcA/MtrC or hydrogenases.^{14, 16} Mechanistic studies—employing genetic, electrochemical, and spectroscopic tools—are essential for improving hybrid design. However, most studies are conducted at the ensemble level and are limited by system heterogeneity, complicating molecular-level understanding.

Herein, this perspective aims to provide an overview of recent progress and key insights on the mechanistic understanding of transmembrane electron transport in the microbe-semiconductor hybrid systems, describing current approaches and their respective features. We first outline energy conversion steps including charge generation, transfer, and intracellular biosynthesis. We then summarize key tools for studying EET, including biological (transcriptomics, proteomics, metabolomics) and (photo)electrochemical approaches involving material design, electrochemical characterization, and spectroscopy.^15–18 Finally, we highlight a multimodal imaging method that measures electron transfer at the single-cell level, offering improved analysis of system heterogeneity.¹² We hope that the comprehensive overview of current methods and the gained critical insights on the interfacial electron transport pathways of microbe-semiconductor hybrid systems will inspire further innovations in mechanistic understanding, ultimately enhancing their energy conversion efficiency.

2. Mechanisms of energy conversion in the biohybrid system

Combining the specificity, cost-effectiveness, and self-repair abilities of non-photosynthetic microbes with the efficient light harvesting of inorganic semiconductors presents a promising hybrid approach to solar-to-chemical conversion. Such semi-artificial photosynthesis in microbe-inorganic biohybrids involves three key processes (Figure 1), as detailed below:

Figure 1. The process of charge generation, charge transfer, value-added chemicals production, and energy conservation in the biohybrid system.

Cys, cysteine; CySS, cystine; Med, mediator. Cyto: Cytochrome; H₂ase: hydrogenase; Ox enzyme: hydrogen oxidation enzyme complex.

(1). Generation and separation of charge carriers in the semiconductor:

This initial step begins with the absorption of photons by the semiconductor material, leading to the creation of electron-hole pairs. As photons are absorbed, electrons transition from the valence band to the conduction band, leaving positively charged holes behind, typically undergoing hole scavenging reactions (e.g., cysteine (Cys) oxidation to cystine (CySS)).^{22, 23} The alignment of these bands determines the redox potential of the generated electrons and holes, crucial for facilitating favorable electron transfer to bacteria and hole transfer to electron donors.²⁴ The semiconductor’s biocompatibility with bacteria is also critical for long-term biohybrid performance.²⁵

(2). Interfacial charge transfer from semiconductor to microbes:

Following charge carrier generation, separated electrons must be transferred into bacteria across the biotic and abiotic interface; this charge transfer process will significantly influence the solar-to-chemical conversion efficiency. Two primary EET modes have been investigated: direct and indirect pathways.²⁶ In the direct EET pathway, electrons are directly transferred from semiconductor surfaces to membrane-bound bacterial electron acceptors (e.g., cytochromes, hydrogenases). For instance, bacteria like Geobacter sulfurreducens and Shewanella oneidensis MR-1, possess outer membrane conductive pili structures and/or c-type cytochromes (e.g., OmcA/MtrABC), which facilitate direct electron transfer (Figure 1).²⁷ Additionally, soluble redox mediators (Meds) enable electron exchange at the bacteria-semiconductor interface, with redox shuttle species such as flavin derivatives and quinones electrochemically regenerated at the electrode-solution interface, providing an indirect electron transfer pathway.²⁸ However, the efficiency of semi-artificial photosynthesis is often limited by factors including the low concentration of bacterial electron shuttles, inefficient transmembrane electron transport due to restricted mediator diffusion, and energy losses from overpotential in redox reactions.²⁹ In addition to microbially self-excreted redox mediators, H₂ produced on the photocatalytic semiconductor surface can also act as an electron carrier, generating both ATP and reducing equivalents through oxidation by the enzyme complex (e.g., HydABC complex). Overall, the detailed electron transfer pathways between bacteria and semiconductors remain understudied due to the convolution of multiple pathways within biohybrids^{30, 31}, which can also evolve over time during operation².

(3). Intracellular electron transfer and biosynthesis:

Upon entering microbial cells, electrons are directed towards specific enzymes responsible for catalyzing desired chemical transformations. These enzymes, typically embedded within the cellular membrane or located in the cytoplasm, convert small inorganic molecules, such as carbon dioxide or nitrogen, into more complex organic compounds.^{32, 33} The transferred electrons supply the necessary energy for these enzymatic reactions, facilitating the synthesis of target chemicals. For instance, in the Wood–Ljungdahl pathway (WLP), CO₂ is enzymatically reduced to acetyl-CoA, which can be utilized for protein biosynthesis or oxidized to acetate to produce ATP.³⁴

3. Biological manipulations to identify key protein players in electron transport

As the interfacial electron transport process involves both microbes and semiconductor, it’s critical to explore their respective roles and interactions. From the biological perspective, various proteins and other species from the bacteria can undergo redox state changes at the physiological pH, and are critical for an efficient electron transport chain, especially those located on the cell membrane and the ones that are secreted out of the cells. Various biological approaches provide important insights into the identity and significance of key proteins for interfacial electron transfer. In this section, we will discuss those approaches and several representative works.

3.1. Gene-specific manipulations

Genetic tools, such as knockouts, overexpression, mutation, etc., are powerful ways to reveal the role of specific proteins in electron transport by comparing the transport efficiencies of different strains. Many electroactive bacteria, such as Shewanella, Aeromonas, and Vibrio spp., maintain an Mtr (metal-reducing) protein complex (comprising MtrABC and OmcA proteins) on the outer membrane, commonly known to be responsible for EET, transporting electrons out of bacteria to reduce extracellular metals ions or other electron acceptors.³⁵ Yet, their capabilities of inward electron transport for biosynthesis are recently observed but not fully explored and requires further investigation. Xiong et al. constructed an inorganic-biological hybrid system by coupling electroactive Shewanella oneidensis MR-1 with biogenic CdS nanoparticles (NPs) and observed the reversed EET phenomenon, while realizing light-driven hydrogen production.¹⁴ CdS NPs are shown by both transmission electron microscopy (TEM) and element mapping to locate mostly outside and attached to the cell outer membrane, while a small fraction enters the periplasmic space. The conduction band of the biosynthesized CdS is measured to be negative enough (−0.68 V vs SHE) to reduce protons to H₂, and OmcA and MtrC make it thermodynamically feasible for an inward electron flow. The pathways involved in CdS nanoparticles biogenesis was also investigated, and interestingly, in contrast to previous CdSe biosynthesis pathways involving Mtr proteins,³⁶ the CdS nanoparticle biosynthesis here does not require outer membrane cytochromes, i.e., OmcA or MtrC. This shows that the thiosulfate precursor reduction to S²⁻, catalyzed by periplasmic enzymes PsrC-PsrB-PsrA and SirD-SirC-SirA,³⁷ is independent of outer membrane cytochromes.

They further explored the protein identities involved in the reversed EET pathway. H₂ production yield and a series of bioelectrochemical parameters on wild-type (WT) and knockout strains were measured for comparison. S. oneidensis MR-1 alone or biosynthesized CdS nanoparticles alone (under light illumination) is able to generate a small amount of hydrogen, demonstrating the enzymatic activity of periplasmic hydrogenase and the photocatalytic property of CdS, respectively. Importantly, integrating S. oneidensis MR-1 with CdS under light illumination shows a significant increase of hydrogen production, revealing a synergistic effect between semiconductors and hydrogenases (Figure 2a, blue bars). To specifically determine the role of the two periplasmic hydrogenases in light-driven hydrogen generation, a ΔhydAΔhyaB mutant strain was constructed where the [Fe-Fe] and [Ni-Fe] hydrogenases were impaired. Knocking out hydA and hyaB genes had no apparent effect on the biosynthesis of CdS nanoparticles, in contrast to a previous work that proposed a hydrogenase-dependent biogenesis of Pd nanoparticles. The hydrogen yield decreased to 26.5% of that produced by WT-CdS, revealing the critical role of hydrogenases for light-driven hydrogen production (Figure 2a, green bars). Combined with the fact that light is essential, the authors concluded that hydrogenases indeed can receive the photoexcited electrons from CdS to catalyze proton reduction.

Figure 2. Light-driven hydrogen production in biotic−abiotic hybrid systems.

(a-c) Hydrogen production by the S. oneidensis MR1-CdS photosynthetic system of WT-CdS, WT (dead)-CdS, ΔhydAΔhyaB-CdS, and ΔhydAΔhyaB (dead)-CdS after 72 h illumination (a); WT-CdS, ΔomcAΔmtrC-CdS, pYYDT::omcAmtrC-CdS (plasmid encoding OmcA and MtrC genes), and ΔomcAΔmtrC-pYYDT-CdS (with empty plasmid) (b); WT, WT (dead)-CdS, and WT-CdS in the hydrogen-producing medium of 20 mM glucose or 20 mM acetic acid (c). (d) Schematic pathways of the light-driven hydrogen production by the S. oneidensis MR-1-CdS photosynthetic system. (a-d) are reprinted with permission from ref 14. Copyright 2022 American Chemical Society.

The contribution of outer membrane c-type cytochromes in transferring the photoexcited electrons across cell membrane was also studied. A ΔomcAΔmtrC mutant lacking two outer membrane cytochromes, OmcA and MtrC, generated only 30.8% of the hydrogen compared to wildtype in the presence of biogenic CdS after illuminated by light for 72 hours, confirming the important role of outer membrane cytochromes during reverse EET. Furthermore, complementing the strain with plasmid-encoded OmcA and MtrC (pYYDT::omcAmtrC) can fully rescue the hydrogen production. As a control, when the mutant strain with empty plasmid (ΔomcAΔmtrC-pYYDT-CdS) showed no improvement, confirming that OmcA and MtrC are critical for electron transport (Figure 2b), serving as an effective transfer channel to transport the extracellular photoexcited electrons from CdS across the membrane to the periplasmic hydrogenases. To explore the contribution from metabolic electron transfer pathways, a respiration inhibitor, rotenone, was added to the photosynthetic hybrid system, and only 22% decrease of hydrogen production was observed. Additionally, when lactic acid was replaced by glucose or acetate, which cannot be oxidized by S. oneidensis MR-1, hydrogen production remained unchanged (Figure 2c). Altogether, these results support that the photoexcited electrons are delivered from CdS to hydrogenases mostly through the reversed EET pathway involving OmcA and MtrC, rather than the respiratory chain. In summary, inward electron transport is indeed feasible in the S. oneidensis MR-1-CdS photosynthetic system and is the major energy source for H₂ production under light illumination (Figure 2d). The dominant pathway is the reversed EET (3) of photo-generated electrons from CdS nanoparticles to the periplasmic hydrogenases via outer membrane cytochromes OmcA and MtrC, while the direct catalytic proton reduction from extracellular CdS (1) or hydrogenases (2), and the direct coupling of CdS and hydrogenases in the periplasmic space (4), contribute just a small fraction.

Introduction of exogenous gene elements can alter the electron transport capability of the original organism, offering a practical method to probe the electron transport pathway and establish new biohybrid systems. Besides H₂, ammonium generated from nitrogen fixation is also an important product. The engineered cyanobacterium S. elongatus PCC 7942 named Se-nif can realize bioelectrochemical nitrogen fixation (e-BNF) through transmembrane electron transfer (TET).³⁸ However, without mediators like methyl viologen, such bioelectrochemical nitrogen fixation is dramatically blocked due to the electrical and physical barrier of the outer membrane. To overcome this, Minteer et al.¹³ inserted OmcS, an outer membrane protein cytochrome from Geobacter sp.³⁹ that aids extracellular electron transfer, to create a modified strain called Senifom (Figure 3a). Examination using ¹H NMR, showed Senifom produced significantly more NH₄⁺ than Se-nif, with distinct doublet signals, confirming NH₄⁺ presence (Figure 3b). In addition, senifom’s NH₄⁺ levels reached 225.0 ± 17.3 μM with a ~23.3% Faradaic efficiency-13 times higher in production and 4 times more efficient than Se-nif (Figure 3c). This demonstrated that combining nif gene cluster and TET conduit omcS indeed enables more efficient e-BNF without using any diffusible mediators. In the e-BNF pathway, OmcS accept electrons from the cathode directly or mediated by biocompatible redox polymers. After the electrons are transported through OmcS, they are delivered to nitrogenase with the aid of some cytoplasmic electron carriers and finally result in superior nitrogen fixation (Figure 3d).

Figure 3. Bioelectrochemical nitrogen fixation in engineered Synechococcus elongatus PCC 7942 strain.

(a) Genetic engineering strategy of S. elongatus PCC 7942 containing nif gene cluster and omcS gene through homologous recombination. (b) ¹H NMR depicting the ammonia production by nitrogen reduction catalyzed by Senifom (green), Se-nif (orange), Se-em (purple), denatured Senifom strain (black), and blank Cc-PAA/alginate electrode without any bacteria loading (magenta) for 20 h. (c) Measurement of ammonium accumulation and Faradaic efficiency in Se-nif strain and Senifom strain. (d) Schematic e-BNF pathway with transmembrane electron uptake through OmcS protein. (a-d) are reprinted with permission from ref 13. Copyright 2021 American Chemical Society.

Overall, gene knockouts or insertions have provided strong evidence that in various microbe-inorganic hybrids, the most critical protein players for interfacial electron transport and the final bioproduct synthesis. In S. oneidensis MR-1-CdS hybrids, reverse EET for light-driven H₂ production relies heavily on outer membrane cytochromes OmcA and MtrC. Knocking out these cytochromes reduced H₂ yield, while complementation restored activity. Metabolic contributions were minimal, as substrate substitutions and respiratory inhibitors had negligible effects. In parallel, engineering S. elongatus PCC 7942 with the omcS cytochrome from Geobacter significantly enhanced e-BNF by enabling direct transmembrane electron uptake, bypassing diffusible mediators. These findings underscore that outer membrane redox proteins and periplasmic enzymes are pivotal for bidirectional electron transport. Enhancing the expression of these proteins may lead to a higher overall energy conversion efficiency for H₂, NH₃, or other bioproducts.

3.2. Transcriptomic, proteomic and metabolomic analyses

Gene-specific knockouts and overexpression only study one or a few genes at a time, therefore the throughput is low. To quickly map out the overall gene regulation network involved in the electron transport pathway to gain a holistic understanding, genome- and transcriptom-level analysis is indispensable. Tian et al. established a periplasmic biohybrid with the aggregate form of CdS and E.coli.¹⁷ Such biohybrids could accomplish semi-artificial photosynthesis of melate under illumination and the electron transfer pathways are studied. Firstly, proteins involved in CdS nanoparticle biogenesis are revealed. Cysteine from the culture medium was converted into hydrogen sulfide by desulfhydrase in the cytosol and then diffused into the periplasm. H₂S then reacted with Cd²⁺, which was transported into the periplasm by the ion efflux pump, to form biomineralized CdS nanoparticles (Figure 4a). Scanning transmission electron microscopy (STEM) verified the stability of cells and nanostructured CdS clusters, and confirmed the non-toxicity of the medium.

Figure 4. Periplasmic biomineralization for photosynthesis.

(a) Schematic pathways describing how CdS nanoclusters are synthesized in the periplasm of E. coli. (b) Heatmap of transcriptomic study confirming the up-regulation of genes related to H₂S production and ion efflux pumps. E_1, E_2, and E_3 represent three independent experimental groups, while C_1, C_2, and C_3 are the control groups. The number of the color bar represents fragments per kilobase of exon per million mapped fragments (FPKM). (c) Production of ATP in E. coli cells with biomineralized CdS nanoclusters under light. (d) Schematic showing the electron transport chain with up-regulated genes highlighted in red letters. (e) Heatmap of transcriptomic analysis indicating how CdS biomineralization up-regulated genes related to the electron transport chain, including dehydrogenases, terminal reductases and oxidases, and ATP synthases. (a-e) are reprinted with permission from ref 17. Copyright 2023 American Association for the Advancement of Science.

Transcriptomic analysis was conducted to probe the transcriptional regulation at whole genome scale. Comparing Cd²⁺/cysteine-treated samples (E_1, E_2, E_3) to Cd²⁺-treated samples (control groups C_1, C_2, C_3) in Figure 4b, genes related to H₂S production, cysK, dcyD, cysM, metB, metC, and sufS, were all up-regulated, indicating cysteine was converted into H₂S. Moreover, additional qPCR experiments verified synergistic effects between Cd²⁺ and cysteine further boosted cysM expression. Genes related to Cd²⁺ efflux into the periplasm, cueO, macA, and tolC are also up-regulated. Additionally, 20 different dehydrogenases were up-regulated in the presence of periplasmic CdS, suggesting an accelerated electron transport in the respiratory chain (Figure 4d). Oxidative stress associated with CdS formation was also studied through transcriptomic analysis. Genes related to the respiratory chain like cydA and cydB were transcribed with an increase of ~19 and ~24 fold, respectively. This result aligned with need for cydA and cydB to maintain respiration and energy production when oxygen availability is limited. Genes related to NADH dehydrogenases like nuoE were upregulated ~6 fold (Figure 4e).⁴⁰ Finally, seven ATP synthases-related genes were found highly up-regulated (Figure 4c–e) in Cd²⁺/cysteine-treated cells compared to the control group, indicating increased ATP production and bacterial respiratory efficiency. Meanwhile, ATP production accelerated ~8 times in cells with periplasmic CdS nanoclusters under light compared to darkness. In summary, such biohybrid system can be a promising cell factory to yield high-value bioproductions like malate.

In another study, Liu et al. established microbe-quantum dots (QDs) biohybrids composed of Xanthobacter autotrophicus and CdTe to maximize the cellular light-driven CO₂ and N₂ fixation for biomass and NH₃ synthesis.¹⁶ This hybrid system reached the theoretical internal quantum yield (IQY) for the first time. To explore the mechanism of this high energy conversion efficiency, the authors first verified the fast charge transfer at the microbe-QDs interfaces. The close contact between QDs and microbes was revealed by high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) and energy-dispersive X-ray spectroscopy (EDS). Time-correlated single-photon counting indicated a fast interfacial charge transfer with the addition of cysteine which worked as the sacrificial hole scavenger. The photon emission intensity decreased steadily to almost none with a significantly shorter lifespan of 4.4 ± 0.3 ns compared to when cysteine is absent.

Proteomic and metabolomic analyses revealed the species involved in the electron transport pathways and light-driven CO₂ and N₂ fixation (Figure 5a). For proteomics, among 2743 proteins, 727 was dramatically up-regulated in the hybrid (e.g., nitrogenases for NH₃ synthesis) while 53 down-regulated. hupV protein, known to be repressed by H₂, was also upregulated. This indicates that H₂ is unlikely a redox mediator in the interfacial charge transfer process and a direct, non-diffusive pathway is more probable. Moreover, the redox-signaling two-component RegA/RegB system that regulates electron-transfer cytochromes was dramatically up-regulated, boosting IQY (Figure 5c). In metabolomic analyses, the lower level of ATP and ADP in hybrid was consistent with the down-regulated complex IV and ATP synthase and the dramatically up-regulated complex I and complex II for oxidative phosphorylation in proteomic analysis. Furthermore, a higher level of 12 amino acids and 3 nucleobases (nitrogen-containing building blocks) reflected energy redirected toward biomass production (Figure 5d). Collectively, the underlying mechanism of light-driven CO₂ and N₂ fixation in biohybrids can be depicted in Figure 5b: the photogenerated electrons from CdTe QDs were transported into cells facilitated by different electron transport chain (ETC) enzymes and provided energy for biomass accumulation. The comprehensive transcriptomic analysis showed that in this hybrid system, electron transfer is likely to be mediated by non-diffusive cytochromes, rather than diffusive hydrogen molecules.

Figure 5. CO₂ and N₂ fixation in microbe-semiconductor hybrids.

(a) Principal component analysis (PCA) and volcano plot of the results of the proteomic analysis and metabolomic analysis. (b) Detailed steps in the photocatalytic hybridization of CdTe QDs and the CO₂- and N₂-fixing bacterium X. autotrophicus. (c, d) Heatmaps of significantly regulated proteins (c) and metabolites (d). (a-d) are reprinted with permission from ref 16. Copyright 2022 Springer Nature.

A similar hybrid system between Moorella thermoacetica and CdS nanoparticle was constructed by Qiao et al. to synthesize acetate with high efficiency.¹⁵ Again, proteomic and metabolomic analyses revealed distinct differences between bacteria interacting with CdS nanoparticles (Bio-CdS) and bacteria only (Bio) (Figure 6a). In the Bio-CdS group, proteins critical for cell wall synthesis (PBP 1A, MurC, and Moth_2365)⁴¹, regulating cell growth (IMPDH), energy production (PstB), and stress response (Moth_1994⁴² Moth_1389,⁴³ Moth_1507⁴⁴) were upregulated, indicating that CdS photoexcitation enhanced the interfacial electron transport pathway, and facilitated cells to minimize the photodamage by upregulating related proteins. Notably, rubrerythrin, an important enzyme for tolerating oxidative stress, was found to be down-regulated in Bio-CdS (~1/3 of that in Bio). This demonstrated the inevitable oxidative stress from electron movement despite the presence of sacrificial reagent like cysteine. The schematic pathways were described in Figure 6c, where photoexcited electrons were imported into cells through membrane-bound ferredoxin, and passed on to CO₂ to synthesize acetate catalyzed by acetyl coenzyme A in the Wood-Ljungdahl pathway.

Figure 6. Light harvesting by Bio-CdS hybrid system with proteomic and metabolic analysis.

(a) Proteomic and metabolic PCA and volcano plots of M. thermoacetica-CdS (BioCdS) in comparison to M. thermoacetica alone (Bio). (b) Relative abundance of nine vital proteins in Bio-CdS and Bio. (c) Schematic pathways of charge transfer and CO₂ fixation in Bio-CdS system. (a-c) are reprinted with permission from ref 15. Copyright 2019 Elsevier.

Overall, transcriptomic, proteomic and metabolomic analyses provide a holistic picture of the electron transport pathway and the above works have clearly shown that upon light stimulation and interaction with photocatalytic semiconductor, the gene regulation profile changes significantly. In E. coli-CdS hybrids, transcriptomics revealed upregulation of H₂S biosynthesis genes and ion efflux pumps critical for periplasmic CdS biomineralization, alongside respiratory chain components and ATP synthases, correlating with enhanced ATP production under light. Proteomics in X. autotrophicus-CdTe hybrids showed upregulation of nitrogenases and the RegA/RegB redox-sensing system, enabling efficient light-driven CO₂ and N₂ fixation, while metabolomics confirmed elevated amino acid and nucleobase synthesis. Similarly, in M. thermoacetica-CdS systems, proteomics highlighted stress-response adaptations and electron transport upregulation, despite oxidative stress from photoexcitation. These studies collectively reveal that biohybrids rewire central metabolism to prioritize electron transport, redox balancing, and stress mitigation, with key enzymes (e.g., dehydrogenases, nitrogenases) and membrane proteins (e.g., cytochromes, ferredoxins) driving interfacial electron transfer. Specific roles of key proteins can be further revealed in detail with targeted knockout or overexpression strains as shown in section 3.1.

4. Key insights from physicochemical characterization

Beyond biological manipulation of microbes, semiconductor materials, as a key component of biohybrids, significantly influence electron transfer pathways and efficiency in biohybrids. Additionally, the intrinsic properties of semiconductor materials, such as conduction and valence band positions, along with the spatial distribution of microbes, also determine the preferred electron transfer pathways. In this section, we summarize how the design of semiconductors, combined with multimodal electrochemical and spectroscopic characterizations, has enhanced our understanding of energy conversion in biohybrids.

4.1. Semiconductor design for extracellular electron transfer

The selection of semiconductors with suitable conduction (E_CB) and valence band edge (E_VB) is crucial for facilitating electron transport across cell membranes by driving redox reactions with varying potentials, including protons, membrane-bound proteins, and cellular redox mediators.⁴⁵. For example, the photo-generated electrons from the conduction bands of Cu₂WS₄ and CdS can facilitate proton reduction in electrolytes (Figure 7a).¹² However, the E_CB of BiVO₄ is below the proton reduction potential, indicating that H₂-mediated electron-transport pathways will not occur when bacteria are interfaced with BiVO4 (Figure 7b). Although Ralstonia eutropha, which can fix CO₂ into the bioplastic polyhydroxybutyrate (PHB), can form a stable contact with semiconductor films composed of Cu₂WS₄, BiVO₄, or CdS particles, different steady-state photocathodic currents suggest diverse electron uptake pathways. The electron uptake capability, indicated by the significant photocurrent of R. eutropha on the BiVO₄ film, suggests that a non-H₂-mediated electron transport mechanism is involved, with redox mediator-based electron transport likely dominating the electron uptake process.

Figure 7. Band gap engineering of photocatalysts.

(a) Energy levels of the conduction band and valence band edges of semiconductors (Cu₂WS₄, BiVO₄ and CdS) and the redox potential ranges of possible redox mediatiors in R. eutropha, all referenced to the RHE (H⁺/H₂). (b) Scanning electron microscopy (SEM) images show cells on top of BiVO₄ semiconductor films. (a-b) reprinted with permission from ref 12. Copyright 2023 Springer Nature. (c) Z-scheme systems using bio-abiotic hybrids and co-catalysts for water splitting and CO₂ to formate conversion. (d) SEM and elemental mapping images of Z-scheme systems. (c-d) reprinted with permission from ref 18. Copyright 2022 Springer Nature.

In addition to the rational design of semiconductor band edges, the selection of wide bandgap semiconductors is crucial for driving photoreduction and photooxidation simultaneously, without the need for electron/hole scavengers or external bias. Artificial Z-scheme systems—comprising two complementary light harvesters—are promising for biohybrids due to spatial separation of redox sites and efficient charge separation.⁴⁶ A recent advancement involved a biohybrid system featuring non-photosynthetic S. ovata paired with a Z-scheme photocatalysis sheet (Cr₂O₃/Ru-SrTiO₃:La,Rh|ITO|RuO₂-BiVO₄:Mo), demonstrating an unprecedented solar-to-acetate conversion efficiency of 0.7% (Figure 7c)¹⁸. Ru and RuO₂ nanoparticles were loaded onto La and Rh co-doped SrTiO₃ and Mo-doped BiVO₄ particulate semiconductors to provide active sites for H₂ and O₂ evolution. Indium tin oxide (ITO) nanoparticles were introduced into the Z-scheme systems as solid-state charge mediators for electrical connections (Figure 7d). It’s noteworthy that selective CH₃COO⁻ production by S. ovata and O₂ evolution by Mo-doped BiVO₄ were achieved using only sunlight, CO₂, and H₂O, without the need for any sacrificial reagents (e.g., cysteine/cystine redox couple). Furthermore, a hybrid semiconductor nanowire–bacteria system, consisting of Z-scheme Si/TiO₂ nanowire arrays and cellular catalysts S. ovata, has successfully achieved unassisted solar-powered acetic acid production with low overpotential (~200 mV) and high Faradaic efficiency (~90%)⁴⁷. The TiO₂ nanowire electrode was utilized for water-oxidization reactions and to prevent possible photodamage in S. ovata through a layer-by-layer design. S. ovata formed an interconnected network among the Si nanowires after an initial incubation period and directly used the reducing equivalents generated from Si nanowires to reduce CO₂ to biosynthetic intermediates.

The limited electron transport pathways and contact areas at the biotic-abiotic interface pose significant scientific obstacles for enhancing extracellular electron transfer, thereby impeding the rate of photoreduction in biohybrids²³. Whole-cell photosensitization by semiconductors, such as quantum dots, functional polymers, and membrane-intercalated molecules, emerges as an effective strategy to maximize electron transport pathways and interfacial contact areas in biohybrid systems. In-situ precipitation offers an efficient method for anchoring light-absorbing abiotic components on non-photosynthetic bacterial membranes (e.g., Moorella thermoacetica) with short charge transport distances.⁴⁸. Typically, metal ions (e.g., Cd²⁺) and cysteine are added to cultures, where cysteine decomposition by desulfhydrase in M. thermoacetica yields sulfide to react with Cd²⁺ and form uniform CdS nanoparticles on the cell surface (Figure 8a). Upon excitation of membrane-anchored CdS, M. thermoacetica utilizes the photogenerated reducing equivalents to activate its innate CO₂-fixing Wood–Ljungdahl pathway. The M. thermoacetica–CdS biohybrids achieved a high quantum yield of 2.44 ± 0.62% to generate acetate from CO₂ in three days under simulated sunlight. Introducing CdS nanoclusters into the bacterial periplasm can further promote electron transport by directly activating enzymes such as phosphoenolpyruvate synthase and pyruvate carboxylase¹⁷. STEM imaging and elemental mapping confirmed the formation of CdS nanoparticles (~29.3 nm) in the E. coli periplasmic space, resulting from the reaction of metal ions and intracellular H₂S decomposed from cysteine (Figure 8b). Periplasmic CdS nanoparticles, positioned in close proximity to the electron transfer chain within the inner membrane of E. coli, enhance malate production from 1.87 to 12.10 mg L⁻¹ under light excitation. In addition, biomineralization serves as a general method to construct microbial biohybrids with periplasmic nanoparticles (e.g., CdS, PbS, HgS and high-entropy semiconductor clusters) for intracellular biosynthesis and solar-to-chemical production.

Figure 8. Whole-cell photosensitization by semiconductor designs.

(a) Biologically precipitated CdS nanoparticles on M. thermoacetica: High-angle annular dark field STEM image and elemental mapping demonstrate CdS clusters covering the entire cell surface. Reprinted with permission from ref 48. Copyright 2016 American Association for the Advancement of Science. (b) Biomineralization of CdS occurs in the periplasm, confirmed by elemental mapping showing Cd and S composition located between the inner and outer membranes. (c) Energy level diagram illustrating the photo-generated electron transfer mechanism from poly(fluorene-alt-phenylene) (PFP) to A. Chroococcum membrane in biohybrids. The light-harvesting PFP electrostatically binds to the surface of A. Chroococcum. Reprinted with permission from ref 50. Copyright 2020 John Wiley &Sons. (d) Scheme of conjugated oligoelectrolyte (COE)-A.vinelandii biohybrid for N₂ fixation. Confocal laser scanning microscopy image reveals COE molecules intercalated within the lipid bilayer membrane. Reprinted with permission from ref 51. Copyright 2023 John Wiley &Sons.

While self-photosensitization using inorganic semiconducting nanomaterials can significantly enhance electron transport at the abiotic-biotic interface, the oxidative stress, cytotoxicity, and environmental hazards associated with inorganic nanomaterials may limit the long-term stability of biohybrid systems. Organic semiconducting photosensitizers (e.g., polymers, small molecules) with good biocompatibility can also interact with cell membranes and form self-photosensitized hybrids⁴⁹. For example, light-harvesting cationic poly(fluorene-alt-phenylene) (PFP) was coated onto Azotobacter Chroococcum via electrostatic interaction with the cell membrane (Figure 8c)⁵⁰. The well-aligned orbital energy levels of PFP with redox proteins on the A. Chroococcum surface enable the photoexcited electron transfer pathway and activate the nitrogen fixation process. These polymer-coated biohybrids showed a 260% increase in nitrogenase activity and a 47% rise in L-amino acid production.

To further reduce the electron transport distance between artificial semiconductors and redox proteins, photo-active conjugated oligoelectrolytes (COEs) are specially designed⁵¹. These COEs consist of hydrophobic conjugated backbones and polar ionic pendant groups, allowing them to intercalate into lipid bilayers of Azotobacter vinelandii (Figure 8d). The “halo-like” emission from COEs surrounding the cells confirms the membrane intercalation of COEs in Azotobacter vinelandii, as verified by confocal laser scanning microscopy. These modified hybrids enable photoelectron transfer from membrane-localized COEs to ferredoxin/flavodoxin and nitrogenase, facilitating N₂ fixation. Through the chemical design of COEs, an acceptor-donor-acceptor structure in the π-conjugated backbone exhibits broad absorption across the visible light spectrum (300–700 nm), which is beneficial for the light-harvesting and charge generation processes.

4.2. Investgation of electron transport pathway via multimodal electrochemical characterizations

Each bacterial species has distinct intrinsic electron transport pathways, making it essential to identify these pathways and quantify electron transport capabilities between bacteria and abiotic materials to optimize biohybrids⁵². Electrochemical methods are widely used to evaluate interfacial electron transfer, including electron uptake rate, faradaic efficiency, and bacteria-material contact quality. However, the measured electrochemical signals often reflect the combined effects of multiple electron pathways, making it challenging to precisely identify specific EET pathways and their capabilities. Integrating electrochemical measurements with other characterization techniques—such as electron transport inhibitors, UV-Vis spectrophotometry, gene editing, and isotope labeling—offers a promising platform to decouple and analyze individual electron pathways, providing comprehensive insights into the electron transport mechanisms within biohybrids.

In microbes, both the photosynthetic and respiratory chains typically involve several redox-active compounds, including the quinone pool, NAD(P)H-dehydrogenase, cytochrome c oxidase, ATP synthase, and others. While these compounds are proposed to play key roles in receiving and transporting electrons, their real functions after the integration of abiotic materials remain elusive. The functions of various redox-active compound in the Synechococcus elongatus transformed with a nitrogen fixation gene cluster and omcS plasmid (Senifom strain) can be decoupled by measuring the electron uptake rates of biohybrids with different inhibitors that selectively inhibit the cross-membrane electron transfer process (Figure 9a)¹³. Four kinds of inhibitors, including dicumarol, rotenone, antimycin A, and carbonyl cyanide m-chlorophenylhydrazone (CCCP), are introduced to the biohybrid system to specifically inhibit the electron transfer from the quinone loop, from NADH dehydrogenase to the quinone pool, from cytochrome c oxidoreductase to the ubiquinol complex, and from ATP synthase, respectively. As shown in Figure 9b, upon the addition of 200 mM and 600 mM dicumarol (a quinone loop inhibitor), the current uptake of the Senifom biocathode exhibited a 36% and 85% decrease from the initial value, respectively. This suggests that the quinone pool plays a crucial role in intracellular electron transfer. Conversely, the addition of up to 1500 μM rotenone only had a slight influence on the current uptake (Figure 9c), indicating that NADH dehydrogenase is not essential to the inward electron transfer pathway of Senifom. Furthermore, selective inhibition measurements demonstrate that cytochrome c oxidoreductase, the ubiquinol complex, and ATP synthase are involved in extracellular electron uptake. Decoupling the functions of each redox-active compound reveals that the electron uptake pathway begins at the outer membrane via OmcS proteins, then transfers to cytochrome c, followed by flow into the quinone pool. Subsequently, it proceeds into the cell metabolism cycle and ATP synthase, ultimately contributing to intracellular nitrogenase turnover.

Figure 9. Selective inhibition the cross-membrane electron transfer.

(a) Mechanism of extracellular electron uptake in Senifom-his strain. (b-c) Impact of electron transport inhibitors on the current output of inward extracellular electron transport between the bacteria and cathode. (a-c) reprinted with permission from ref 13. Copyright 2021 American Chemical Society.

Measuring differences in electron uptake rate and product yield between wild-type and mutant strain biohybrids clarifies the function of the EET chain between abiotic materials and intracellular redox proteins. For instance, while the outer membrane c-type cytochromes (OmcA/MtrABC) in the EET chains of S. oneidensis MR-1 are theoretically responsible for electron transport across the membrane, the impact of CdS nanoparticles on the extracellular surface of S. oneidensis MR-1 on EET remains unknown. The ΔomcAΔmtrC strain, lacking two crucial outer membrane c-type cytochromes (MtrC and OmcA), exhibits significantly reduced current when a cathodic potential is applied, compared to biohybrids composed of wild-type strains (Figure 10a)¹⁴. The near absence of inward electron flow from the electrode into the ΔomcAΔmtrC strain suggests a critical role of outer membrane c-type cytochromes in EET. Although the current magnitude can suggest inward electron flow, the downstream cellular processes involved in electron transfer remain unknown. Gas chromatography was used to probe the function of hydrogenases and quantify the final product H₂ in CdS-S. oneidensis MR-1 biohybrids (Figure 10b). The hydrogen production of ΔomcAΔmtrC biohybrids showed a similar trend. No hydrogen was detected in the ΔomcAΔmtrC strain and mutant strain without [Fe−Fe] hydrogenase and [Ni−Fe] hydrogenase (ΔhydAΔhyaB mutant). Respiration inhibition tests were conducted to investigate the electron flow in CdS-S. oneidensis MR-1 biohybrids (Figure 10c). Upon the addition of rotenone (a respiration inhibitor), only a 22% reduction in hydrogen production was observed. This finding suggests that metabolic hydrogen contributed to only a small portion of the total hydrogen production in the hybrid photosynthetic system. By integrating gene editing, gas chromatography, inhibition tests, and electrochemical tests, the authors demonstrated that the Mtr pathway played a crucial role in facilitating electron transfer from the extracellular environment to hydrogenases for hydrogen generation.

Figure 10. Identification of the redox-active compound via multimodal physical characterizations.

(a) Current densities of wild-type CdS-S. oneidensis MR-1 biohybrids and mutant CdS-S. oneidensis MR-1 biohybrids during the hydrogen generation process. (b) Time-resolved quantification of hydrogen production in wild-type strain and mutant strain composed biohybrids. (c) Hydrogen production quantification coupled with respiration inhibition tests of wild-type strain composed biohybrids. (a-c) reprinted with permission from ref 14. Copyright 2022 American Chemical Society. (d-f) Coupled cyclic voltammetry (CV) and UV–vis spectroscopy are used to identify redox-active compound in Anabaena variabilis. (d-f) reprinted with permission from ref 53. Copyright 2017 Elsevier.

Understanding the identity of unknown redox mediators in bacteria is crucial for elucidating electron transport at the abiotic–biotic interface. CV and UV–vis spectroscopy are valuable tools for this, leveraging characteristic redox and absorbance peaks⁵³. Cyanobacteria, comprising vegetative and heterocyst cells containing enzymes such as nitrate reductase, nitrite reductase, and nitrogenase, play a pivotal role in nitrogen fixation for ammonia production (Figure 10d). In the presence of cyanobacteria, CV shows a reductive peak at 0.2 V and an oxidative peak at +0.15 V vs. SCE (Figure 10e). Exposure to N₂ and light leads to an increase in the reductive peak and a decrease in the oxidative peak, indicating the oxidation of predominant redox moieties at the electrode surface. Furthermore, UV–vis spectra further reveal that the spent electrolyte overlaps with commercial ferredoxin, with a similar peak near 250 nm (Figure 10f), suggesting ferredoxin as the redox signal origin.

Beyond multimodal physicochemical characterizations (e.g., inhibitor tests, gene editing, product quantification, absorption spectra), isotope labeling offers insight into photosynthesis mechanisms in biohybrids⁵⁴. For instance, deuterium labeling experiments can ascertain whether electrochemically generated H₂ acts as a redox mediator in Clostridium ljungdahlii (C. lj.) biofilms by quantifying the deuterated acetate product under different electrochemical potentials. Deuterium (D₂ or HD) is electrochemically produced in situ at −0.15 V (vs. RHE). However, at 0.04 V (vs. RHE), where the hydrogen evolution reaction (HER) is negligible, an alternative EET mechanism would be required. Surprisingly, deuterated acetate yield was lower at −0.15 V than at 0.04 V, indicating a non-H₂-mediated EET mechanism in C. lj. In addition, ¹⁵N and ¹³C labeled substrates were used as nitrogen and carbon sources fed to the biohybrid systems allowing for the distinction of reductive products (¹⁵NH₄⁺ or ¹³C) generated from background cellular metabolic activities^{13, 54}.

4.3. Spectroscopic elucidation of electron transport in biohyrids

Time-resolved pump-probe spectroscopy has become a powerful tool for elucidating charge transfer kinetics and mechanisms in semiconductor–microbe interactions⁵⁵. Yang et al. used pump-probe transient absorption (TA) and time-resolved infrared (TRIR) spectroscopy to examine competing electron transfer pathways in M. thermoacetica–CdS biohybrids². These include a direct enzymatic pathway for acetic acid production and a hydrogenase-mediated pathway generating H₂ (Figure 11a). TA revealed that hydrogenase-enriched biohybrids (incubated with H₂) exhibited shorter band edge bleach lifetimes than glucose-incubated controls or CdS alone (Figure 11b), indicating faster electron transfer or greater availability of acceptors. Triexponential decay analysis showed shorter τ₁ and τ₂ lifetimes with increasing hydrogenase activity, supporting a charge-transfer route to hydrogenase (Figure 11c). TRIR spectroscopy also indicated rapid decay in the 1,760–1,880 cm⁻¹ spectral window, corresponding to the vibrational range of CO and CN double and triple bonds, characteristic of amino acids. This decay occurred on the picosecond scale of electron transfer, consistent with the observations from the TA data (Figure 11d). Interestingly, highest acetic acid yield occurred with lowest hydrogenase activity during the first 3 hours of photosynthesis, suggesting the presence of a direct electron-transport pathway favoring electron injection to a membrane-bound electron carrier (e.g., Fd, Fp, cytochrome, MK). TA also confirmed fast electron transfer to hydrogenase in S. ovata with intracellular InP quantum dots⁵⁶.

Figure 11. Utilization of transient absorption and time-resolved infrared spectroscopies to elucidate energy transfer in biohybrids.

(a) Possible dual pathway of charge transfer in M. thermoacetica-CdS biohybrids. (b) Transient absorption plots of cell-free samples (CdS only) and biohybrids incubated with glucose or H₂. (c) Exponential lifetimes τ1 and τ2 exhibit decreasing trends as the incubation time under H₂ increases. (d) Time-resolved infrared spectroscopies analysis of M. thermoacetica-CdS biohybrids incubated with H₂ reveals bleaching of multiple peaks. (a-d) reprinted with permission from ref 2. Copyright 2016 National Academy of Science.

Analyzing the time-resolved photoluminescence intensity of biohybrids over time provides valuable insights into the photon-excited states of the semiconductors which are quenched by photogenerated reducing equivalents¹⁶. In CdTe-X anthobacter autotrophicus biohybrids with cysteine, the photoluminescence intensity at 566 nm under 450 nm excitation gradually decayed to near zero within 8 hours, suggesting non-diffusive static quenching at the microbe–semiconductor interface facilitated by cysteine. Furthermore, the photoluminescence decay rate was significantly faster in CdTe and CdTe-X autotrophicus samples. A static quenching mechanism is proposed to achieve near-unity efficiency in extracting photoexcited carriers from the CdTe and delivering reducing equivalents to the microorganisms, thus explaining the fast charge transfer at microbe–semiconductor interfaces.

5. Bulk-level vs. single-cell measurement

As stated in the above sections, electron transfer mechanistic studies often report ensemble-averaged results from numerous cells at the nano-bio interface, such as the densely packed S. ovata in the 3D Si nanowires (Figure 12a).⁵⁷ In these measurements, the number of cells contributing to the signals is hard to quantify accurately furthermore, cells have large heterogeneity in terms of electron transporting and energy conversion efficiency, making dissecting the mechanism more challenging. Recent advanced high-resolution imaging and measurement enable single-cell analysis for revealing electron transfer mechanisms. For instance, Chen et al. constructed a bottom-up microbe-semiconductor hybrid system, where single bacteria is well separated and probed individually (Figure 12b).¹² This novel approach resolved the inevitable large heterogeneity in biological and nanomaterial systems, and deciphered the electron transport pathway more quantitively, providing new mechanistic insights unattainable through bulk measurements.

Figure 12. Bulk and single-cell characterization of hydrogenases in biohybrids.

(a) SEM image of nanowire-S. ovata hybrids. Reprinted with permission from ref 57. Copyright 2020 Cell Press. (b) Single R. eutropha integrated with CdS film, Cu₂WS₄ film and BiVO₄ particle. (c) Acetic acid synthesis rate at first 3 hours with different H₂ incubation time. (d) Acetic acid synthesis rate over 24 hours with different H₂ incubation time. (c-d) reprinted with permission from ref 2. Copyright 2016 National Academy of Science. (e) Cellular hydrogenase concentration (1–3: HoxG; 4–6: HoxY) under different growth condition. (f) Both MBH and SH show cell envelope localization. (g) CV of R. eutropha supernatant cultured in autotrophic condition. (b, e-g) reprinted with permission from ref 12. Copyright 2023 Springer Nature.

The functions and roles of key proteins involved in electron transfer can be identified and quantified through characterization techniques across different spatial scales. For example, hydrogenases are shown to be critical in electron transport and energy conversion in various microbe-material biohybrids. Yang et al. observed that longer H₂ incubation time induced higher hydrogenases level in M. thermoacetica, confirmed with bulk enzyme activity measurement.² Surprisingly, averaged acetic acid production rates negatively correlated with H₂ incubation early in growth (<3 hours) while positively correlated in the late growth stage (Figure 12c, d). This suggests two competing pathways: a non-hydrogenase mechanism dominates early, while hydrogenases takes over later. This study also highlights the large heterogeneity and the evolving nature of cellular biological properties among a population, whereas single-cell investigation would be a valuable addition.Chen et al. used fluorescent proteins to endogenously label hydrogenases (membrane-bound hydrogenase MBH, or soluble hydrogenase SH) in R. eutropha and quantified their expressions at single-cell resolution.¹² Consistent with earlier ensemble results, both hydrogenases were significantly upregulated when cells were cultured under H₂ compared to fructose, and higher H₂ concentration (30%) induced more expression than lower H₂ concentration (9%) (Figure 12e). Furthermore, the two hydrogenases expression were positively correlated with each other at single-cell level, attributing to that both hydrogenases were regulated by the same promoter. More importantly, protein spatial distribution as well as diffusive behavior in live cells were resolved at high resolution using super-resolution imaging. MBH were found to be located at the cell envelope (Figure 12f, left), as expected. Surprisingly, the authors found that SH, which were previously thought to be uniformly cytosolic, also showed envelope distribution (Figure 12f, right). This observation indicated that soluble hydrogenases are recruited close to the inner membrane or may even be secreted into the periplasmic space to facilitate an efficient electron transfer. CV on cell culture supernatant showed a catalytic reduction peak consistent with hydrogenase HER, further supporting the hypothesis that hydrogenases may be secreted to the extracellular space (Figure 12g).

To quantify energy conversion efficiency, product yield – the end point of energy flow – is calculated by quantifying the amounts of synthesized products. Previous bulk measurements mainly compared the overall product yields with different genetic manipulations (Figure 13a) to imply the energy flow pathway.¹¹ With single-cell information, Chen et al. quantified the synthesized products from individual cells with cell-to-cell heterogeneity resolved (Figure 13b). Specifically, polyhydroxylbutyrate (PHB) was the final product of R. eutropha when metabolizing H₂. A reporter protein, PhaP1, which decorates the PHB granules, was tagged with mVenus fluorescent protein and the fluorescence intensity directly represents PHB amount. Through this single-cell quantification, the authors reproduced the previous bulk results and showed that SH (HoxY) is essential for PHB production while MBH (HoxG) is not (Figure 13b). More importantly, the positive correlation between SH and PHB obtained by single-cell two-color imaging confirms the critical, quantitative role of SH for energy conversion (Figure 13c), which is not easily achieved in bulk analysis.

Figure 13. Bulk and single-cell characterization of biosynthetic products and interfacial electron transport in biohybrids.

(a) Quantification of biosynthesis reactant and product, and Faraday efficiency in IO-ITO|G. sulfurreducens electrodes. (b) Single-cell quantification of PHB synthesis in different strains at 9% (red) and 30% (blue) H₂+CO₂ growth condition. (c) PHB is positively correlated with SH in autotrophically grown R. eutropha. (d) Cathodic current density of an IO-ITO|G. sulfurreducens electrode catalyzing fumarate. (a,d) Reproduced or adapted from ref 11. Available under a CC-BY 4.0 license. Copyright 2016 National Academy of Science. (e) Cathodic current density of bulk Cu₂WS₄ thin-film with and without R. eutropha at different gas environment. (f) Schematic of laser beam focused at on-cell and off-cell location (top) and representative local photocurrent-time trajectories (bottom). (g) Histograms of single-cell Δi_ph (blue) obtained on Cu₂WS₄ and a control Cu₂WS₄ sample that contains no cells (black). (h-i) Correlations of single-cell Δi_ph obtained on a Cu₂WS₄ photocathode with MBH or SH concentrations. (b-c, e-i) reprinted with permission from ref 12. Copyright 2023 Springer Nature.

Current density, a key metric for electron transport at the interface of microbes and semiconductors, is commonly measured at bulk level with multiple cells contributing to the signal (μA-mA/cm² scale) (Figure 12d, e). Since the number of cells varies in different systems and is difficult to quantify, it is challenging to evaluate the single-cell electron transport capacity and the quantitative contribution of related proteins, or to compare the electron transfer efficiency across different hybrid systems. To probe electron transfer of individual cells and therefore reveal the mechanism more clearly, Chen et al. took advantage of focused laser beam to map the photocurrent from individual bacteria (Figure 12f)¹². The focused laser beam is smaller than the size of one bacterium and by comparing on- and off-cell locations, the electron transport capacity of an individual cell was measured to be at nA magnitude (Figure 12g). This value is surprisingly large considering the theoretical hydrogenase turnover rate, indicating that non-H₂-mediated pathways should dominate the electron transfer process. Moreover, both MBH and SH were positively correlated with the current magnitude, in contrast to that only SH is essential in metabolizing H₂ (Figure 13h, i). Overall, an electron transport pathway from photocatalytic semiconductor to R. eutropha is proposed, where both hydrogenases contribute quantitatively.

These results demonstrated that single-cell and bulk experiments complement each other and provide a more complete understanding of the electron transfer mechanisms. Bulk measurements have the advantage of measuring absolute final products, while currently the single-cell analysis can only perform indirect measurement as the product from one cell is too little. At the bulk level, yields from different products can be measured with high throughput analytical tools, which is challenging to achieve at the single cell level. On the other hand, single-cell analysis resolves the large heterogeneities among cell populations and provides a cleaner system for mechanistic study. As an example, Chen et al. measures the electron transfer capability of R. eutropha in a biohybrid system for the first time and quantitatively demonstrate the roles of hydrogenases in this process. They showed the colocalization of MBH and SH at the cell envelope and secretion into the periplasm, challenging the assumption that SH is uniformly cytosolic. Additionally, single-cell photocurrent mapping further demonstrated that both MBH and SH contribute quantitatively to electron transfer. We believe single-cell level studies have the potential to provide more quantitative information such as the respective importance of each electron transfer protein/redox mediator and how much they contribute to the overall energy conversion. This new information can guide the design strategy to enhance overall product yield by genetic engineering (optimize gene expression) and material engineering (increase charge separation).

6. Outlook

Despite substantial progress in elucidating EET pathways and enhancing the performance of biohybrid systems (Table S1), critical challenges remain—particularly in understanding mass and charge transport, dynamic biotic–abiotic interactions, and spatiotemporal correlations. Addressing these gaps is essential for advancing rational design and scalable deployment of biohybrids for solar-to-chemical conversion and sustainable production.

Electron and mass transfer mechanisms:

Integrating semiconductor materials with microbial hosts reshapes native transport processes and establishes new EET routes with distinct efficiencies, though underlying mechanisms remain unclear. For instance, localization-dependent EET efficiency has been observed in M. thermoacetica–CdS biohybrids (Figure 8a, b), where CdS nanoparticles at membrane interfaces enhance electron uptake, and in Au–Rhodobacter capsulatus systems, where periplasmic Au nanoparticles with cysteine outperform surface-bound ones due to their proximity to redox-active components.⁵⁸ Nanomaterial properties, such as size and crystallinity, also influence EET. In vivo voltammetry showed that 14 nm Au nanoparticles outperform 35 nm particles by lowering the interfacial electron transfer barrier.⁵⁹ Ultimately, key performance metrics—quantum efficiency, energy conversion, and operational stability—are tightly linked to interfacial transport.

Dynamic interactions at biotic–abiotic interfaces:

Biohybrids are dynamic, with both microbial and abiotic components evolving over time. These dynamics arise from feedback interactions between microbial metabolism, gene expression, material properties, and environmental stimuli (e.g., light, redox potential). However, the time-resolved evolution of these interfaces is still poorly understood. For example, in M. thermoacetica-CdS biohybrids, time-dependent electron transport pathways (Figure 11a) are mediated by membrane-bound electron carriers or hydrogenase activity. These pathways may shift due to changes in microbial metabolism or material properties, influencing electron uptake efficiency and overall system performance. Without temporal resolution, controlling or predicting electron flux is challenging. Thus, developing time-resolved, operando techniques to probe structural, chemical, and functional changes under realistic conditions—is essential for capturing the transient behaviors that govern biohybrid performance.

Spatiotemporal heterogeneity and characterization challenges:

Biohybrids exhibit heterogeneity across scales, including variations in nanoparticle distribution, enzyme expression, redox mediator concentration, and membrane integrity, all affecting EET. Single-parameter techniques often fail to capture these intricate interactions and interdependencies. For instance, spatial heterogeneity in redox mediator distribution or enzyme activity can create subcellular variations in electron transfer efficiency that ensemble-averaged measurements fail to detect. Addressing these limitations requires integrating multiple complementary techniques. For example, current measurements coupled with inhibitor studies (Figure 9) can isolate and identify specific components involved in electron transfer pathways. Similarly, current measurements combined with strain-specific gene manipulations (Figure 10a) enable the functional dissection of microbial contributions to electron transfer. Additionally, the fluorescent labeling of hydrogenases integrated with photoelectrochemical measurements (Figure 12) allows for spatiotemporal and functional mapping of hydrogenase activity and electron uptake capability at the single-cell level. This integrative approach not only clarifies the roles of specific microbial components in EET processes but also demonstrates how their activity and spatial distribution influence overall biohybrid system efficiency. These insights provide a robust framework for optimizing biohybrid designs, enhancing electron transfer performance, and enabling more targeted applications.

Technology and interdisciplinary integration:

Advancing biohybrid systems requires coordinated development in both experimental and computational methodologies. High-resolution biological techniques—such as super-resolution fluorescence microscopy, fluorescence lifetime imaging, and multi-omics profiling (genomics, transcriptomics, proteomics, and metabolomics)—enable detailed insight into cellular responses and regulatory networks influenced by abiotic materials. When combined with materials characterization tools like synchrotron-based X-ray spectroscopy, transient absorption spectroscopy, and in situ electrochemical measurements, these approaches provide multiscale understanding of charge transport, interfacial chemistry, and structural dynamics. Computational approaches such as density functional theory (DFT) and molecular dynamics (MD) offer predictions of band energetics and atomistic interfacial behavior. For instance, DFT can estimate conduction and valence band positions to evaluate electron transfer feasibility,⁶⁰ while MD offers atomistic insights into material–microbe interactions and transport pathways.⁶¹ These models help identify materials with optimal energetics and biocompatibility. In parallel, machine learning tools are essential to manage multimodal data, extract localization patterns, and guide mechanistic understanding and system design.

Together, these approaches pave the way for precise optimization of electron transfer pathways, enabling enhanced performance in biohybrid systems. Crucially, cross-disciplinary insights into electron transport pathways are reshaping our understanding of biohybrids. For example, the discovery of metabolic rewiring in S. ovata induced by water-splitting electrochemistry challenges traditional views of microbial metabolism⁶². Such insights underscore the transformative potential of biohybrids, revealing the complex interplay between biology and abiotic materials. These insights deepen our understanding of EET in biohybrid systems and inspire innovative strategies for their design and optimization, which hold immense promise for renewable energy and sustainable chemical production.

Supplementary Material

The Supporting Information is available free of charge at.

Summary of characterization techniques for biohybrids