1. Introduction
Phototrophy and chemotrophy are two well-known types of microbial metabolism
[1],
[2]. Photosynthetic microorganisms capture light to produce biomass with energy-rich chemical bonds
[3],
[4], whereas chemotrophic microorganisms harness energy from a series of chemical reactions (organic and inorganic) for cellular functions
[5],
[6]. However, a variety of microbial communities have been discovered in the subsurface, where microbes are expected to be limited by these two common sources of energy
[7],
[8],
[9],
[10], such as, weak light irradiation and scarce inorganic/organic electron donors. Remarkably, electroactive microorganisms that are capable of electrical interactions within communities and with extracellular matrices in diverse electromicrobiomes
[11],
[12] have been identified across a spectrum of extreme environments
[13],
[14]. Therefore, it is imperative to explore novel and previously unrecognized energy sources for these electroactive microorganisms, considering their pivotal roles in multiple biogeochemical cycles
[15],
[16],
[17].
Mechanical energy is ubiquitous and versatile in the natural environment and can be generated via various geodynamic processes, such as earthquakes, fault movements and water currents
[18]. Until recently, the potential of renewable mechanical energy to support or enhance microbial survival and growth had received little attention, partly because of documented evidence showing that mechanical forces tend to physically disrupt the integrity of microbial cells
[19],
[20]. However, piezoelectric transduction, such as, mechanical-to-electrical conversion with piezoelectric materials
[21],
[22], may provide promising opportunities for the cooperative interaction of microbial metabolism and mechanical energy. Specifically, electrons transferred from piezoelectric surfaces could act as a viable source of reducing equivalents for microorganisms. Consequently, a robust biohybrid piezoelectric effect (BPE) may be achieved by integrating the precise biological catalysis processes of living microorganisms with the strong piezoelectric and electromechanical coupling properties of piezoelectric materials. However, to unequivocally demonstrate the possibility of such a BPE process, solid evidence regarding electron production, transport and utilization is needed.
Here, we provide evidence in support of a previously unreported geobiological process in which mechanical energy is converted into electron energy by piezoelectric materials, and these electrons are taken up via extracellular electron transport chains to support the survival and growth of electroactive microorganisms. Using the purple nonsulfur bacterial species
Rhodopseudomonas palustris (
R. palustris) CGMCC 1.2180 as a model electroactive microorganism and barium titanate (BTO) as a model piezoelectric ceramic, we constructed
R. palustris-BTO biohybrid system. BTO, one of the most widely used piezoelectric materials
[23], is a robust dielectric material with a high piezoelectric coefficient, low dielectric losses, and remarkably high chemical and thermal stability
[24]. Using this system, it was possible to use mechanical energy to drive the piezoelectric transduction of BTO, producing extracellular electrons that were continuously delivered to
R. palustris for intracellular carbon fixation coupled with nitrate (NO
3−) reduction (
Fig. 1). Research has demonstrated that various other electroactive microorganisms can also utilize mechanical energy to facilitate the growth and biogeochemical cycling of essential elements. Based on these data, we propose that the BPE represents a source of energy that is fundamentally different from that used by conventional chemotrophic and phototrophic organisms and that BPE may play unique roles in microbial survival in energy-limited environments and may be intimately involve in multiple biogeochemical cycles on Earth and the origin and early evolution of life.
2. Materials and methods
2.1. Microorganisms and culture media
R. palustris was acquired from the China General Microbiological Culture Collection Center (China) and cultivated in mineral media
[25].
Geobacter sulfurreducens (
G. sulfurreducens) ATCC 51573,
Shewanella oneidensis (
S. oneidensis) ATCC 700550,
Moorella thermoacetica (
M. thermoacetica) ATCC 39073,
Desulfovibrio desulfuricans (
D. desulfuricans) ATCC 27774,
Escherichia coli (
E. coli) BAA-3219, and
Bacillus subtilis (
B. subtilis) ATCC 6051 were procured from the American Type Culture Collection (ATCC; USA) and cultured in mineral-based media
[25],
[26].
Thiobacillus denitrificans (
T. denitrificans) DSM 12475 and
Methansarcina barkeri (
M. barkeri) DSM 800 was purchased from the Leibniz Institute DSMZ-German Collection of Microorganisms and Cell Cultures (Germany), and cultured in mineral-based media
[27],
[28]. All culture processes were performed in a glovebox with high-purity N
2 gas.
2.2. Construction and characterization of R. palustris-BTO biohybrid system
After
R. palustris reached its logarithmic growth phase, BTO was added to generate an
R. palustris-BTO biohybrid system. After two days of cultivation with mechanical stirring at 180 r·min
−1, the mixed suspension of
R. palustris-BTO was sequentially centrifuged, washed, and resuspended in normal saline three times to prepare it for subsequent applications. Other biohybrid systems were prepared via the same method, except either
R. palustris or BTO was replaced by a different component. A SU8020 scanning electron microscope (SEM; Hitachi, Japan) and Tecnai G2 F30 scanning transmission electron microscope (STEM; FEI, USA) were used to observe the surface morphologies of the
R. palustris-BTO system. Prior to this, it was necessary to immobilize the biohybrids with 2.5 wt% glutaraldehyde and to dehydrate them with an ethanol gradient (25%, 50%, 75%, 90%, and 100%). Freeze-dried samples were obtained via vacuum freeze-drying (Alpha 1-4 LDplus, Christ, Germany). An XRD-6000 X-ray diffractometer (Shimadzu, Japan) was used to obtain X-ray diffraction (XRD) patterns. X-ray photoelectron spectroscopy (XPS) spectra and valence-band XPS spectra were obtained via an ESCALAB 250XI spectrometer (Thermo Fisher Scientific, USA). The size distribution of the biohybrids was determined with a Zetasizer Nano S potentiometer (Malvern, UK). The band gap of BTO was measured via an ultraviolet-visible (UV–vis) spectrometer (UV2600, Shimadzu, Japan). The components of bacterial extracellular polymeric substances (EPS) were detected and analyzed via multiple fluorescence staining methods
[29].
Characterization of biohybrid ferroelectric properties was performed at room temperature via a ferroelectric analyzer (TF2000E, aixACCT, Germany) in combination with a high-voltage amplifier (0–10 kV) with a gradient electric field (2, 3, and 4 kV·cm
−1). Piezoelectric signals were probed via Asylum Research Cypher S piezoresponse force microscopy (PFM; Oxford Instruments, UK). During piezoelectric transduction, the current intensities for the
R. palustris-BTO biohybrids were gauged via a single-chamber cell linked to a PalmSens4 electrical measurement system (PALMSENS, the Netherlands). In this system, graphite and indium tin oxide (ITO) plates were employed as the working and counter electrodes, respectively
[30].
2.3. Piezoelectric catalytic experiments
Piezoelectric catalytic assays were performed using 25 mL of autotrophic medium (Table S1 in Appendix A) and different concentrations of
R. palustris-BTO. To maintain anaerobic conditions, the headspace of the reaction vials was purged with a filter-sterilized N
2:CO
2 gas mixture for 20 min. To remove residual electron donors from microbial cells, the mixed suspensions were statically cultured for three days. The samples were subsequently placed on an MS H S10 magnetic stirrer (DLAB, China). Control tests were carried out under consistent conditions, except
R. palustris cells, BTO nanoparticles or mechanical stirring processes were systematically eliminated. The influence of BTO concentration, stirring rate, sacrificial reagent (ascorbate, a ubiquitous and important intermediate metabolite that is nontoxic and environmentally friendly
[31],
[32] and a redox mediator (antraquinone-2,6-disulfonate (AQDS) with excellent reversible redox capability and chemical stability
[33]) on the piezoelectric catalytic performance of
R. palustris-BTO was also investigated. To further study the effects of different types of mechanical force on piezoelectric catalysis, serum bottles containing
R. palustris-BTO were placed sequentially in a constant–temperature ultrasonic cleaner (KQ-400KDE, Kunshan Ultrasonic Instruments Co., Ltd., China) and in an oscillator to simulate vibration conditions (ZWY-100H, Zhicheng, China). Similar experiments were conducted using tourmaline, strontium titanate (SrTiO
3), and quartz (SiO
2) instead of BTO to evaluate the effects of different piezoelectric materials on piezoelectric catalysis performance. In addition, other microorganisms, including
G. sulfurreducens,
T. denitrificans,
S. oneidensis,
M. thermoacetica,
M. barkeri,
D. desulfuricans,
E. coli, and
B. subtilis, were combined with BTO to construct biohybrids via the same experimental method used for the
R. palustris-BTO biohybrids, and piezoelectric catalytic experiments were then performed on a magnetic stirrer.
The series of inhibition experiments within the BPE process incorporated several distinct inhibitors
[34],
[35],
[36]. These included antimycin A, which targets cytochrome bc
1. Another inhibitor used was carbonyl cyanide
m-chlorophenyl hydrazine (CCCP), which impedes nanofilament formation by dissipating proton motive force. Additionally, potassium cyanide (KCN) was used, which inhibits hydrogenase activity. Stock solutions of antimycin A, CCCP, and KCN were prepared in pure dimethyl sulfoxide
[34]. All experiments were subsequently performed in triplicate, at minimum. For statistical analysis, paired two-tailed
t tests were utilized.
2.4. Product measurements
To examine the variation in biomass, the samples were collected and washed at least three times. The cellular DNA was first extracted via the TIANGEN Bacteria DNA Kit (China). The DNA concentration was subsequently quantified with a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific). Total protein was extracted via a protein kit (Sangon, China). The protein concentration was then quantified with a SpectraMax iD3 reader (Molecular Devices, USA) and a BCA protein kit (Thermo Fisher Scientific). The concentrations of NO
2− and NO
3− were measured with an ICS 900 ion chromatograph (Thermo Fisher Scientific). Moreover, N
2O levels were assessed with a 7890 gas chromatograph (Agilent, USA)
[27]. The concentrations of O
2 and N
2 were determined with a thermal conductivity detector. To measure nitrogenous compounds, experimental vial headspace was purged with 99.999% helium prior to the denitrification experiments. Dissolved oxygen (DO) concentration was measured by a dissolved oxygen analyzer (JPB-607A, Yuwo Instrument Equipment Co., Ltd., China).
To directly assay the change in the redox potential of bacterial cell membranes in the
R. palustris-BTO system, the redox fluorescent probe C
11-BODIPY
581/591 (Molecular Probes, USA) was added to the medium at a 5 μmol·L
−1 concentration, and the medium was incubated for 1 h
[9], with stationary
R. palustris-BTO used as a control. Subsequently, cellular fluorescence was analyzed via a Zeiss LSM880 confocal laser scanning microscope (CLSM; Germany). Cell viability was measured with a live/dead viability kit (Invitrogen, USA), observed via CLSM, and analyzed via ImageJ software. Adenosine triphosphate (ATP) levels were measured through bioluminescence using an ATP Kit (Beyotime, China).
2.5. Isotope analysis
Isotopic experiments were conducted using
13C-labeled NaHCO
3 (99 atom percent (atom %)
13C) and
15N-labeled KNO
3 (99 atom %
15N) in the medium. The isotope-labeled samples were analyzed via a 7890–5975c gas chromatography-mass spectrometer (Agilent). Analysis was conducted via TOF-SIMS 5 time-of-flight secondary ion mass spectrometry (ToF-SIMS, IONTOF, Germany)
[37]. The ToF-SIMS data were analyzed via SurfaceLab 6.7 software.
2.6. Transcriptomic analysis
For transcriptomic analysis, biohybrid samples were collected and centrifuged at 10 000 ×
g (
g: denoting gravity) for 30 min
[30]. Total RNA was isolated via an Invitrogen TRIzol reagent (USA), and RNA libraries were subsequently prepared via the RNA Library Prep Kit (NEBNext Ultra II Directional, New England Biolabs, China). Subtractive hybridization was then employed to remove ribosomal RNA (rRNA). Messenger RNA (mRNA) was sequenced on the Illumina HiSeq/MiSeq platforms. The raw sequencing data were quality-checked and filtered. All sequence reads matching 16S and 23S RNA genes were removed, and the remaining reads were mapped against the published genome of
R. palustris CGMCC 1.2180 (NZ_CP058907.1). Mapped reads were normalized to fragments per kilobase per million mapped reads (FPKM) values.
3. Results and discussion
3.1. Biohybrid characterization
Polymer bridging between the EPS of
R. palustris and BTO facilitated self-assembly in the
R. palustris-BTO biohybrids. This process was particularly supported by the large quantity of high-molecular-mass polysaccharides within the EPS and was facilitated by various potential interactions, including electrostatic forces, van der Waals forces, hydrogen bonding, and chemical reactions
[38],
[39] (Figs. S1 and S2 in Appendix A). SEM, STEM, and energy dispersive X-ray spectroscopy (EDS; Bruker XFlash 6T/30, Germany) mapping revealed clusters of small particles (20–90 nm, Fig. S3(a) in Appendix A) composed of Ti and Ba on the cell surface (
Figs. 2(a) and
(b)), and this finding was corroborated by survey X-ray photoelectron spectra (Figs. S3(b)–(d) in Appendix A). High-resolution TEM images and XRD patterns further confirmed the identity of BTO: the lattice spacing of 2.83 Å matched well with the (110) crystal planes of the tetragonal phase of BTO (indexed as JCPDS # 79-2264, Figs. S3(e)–(g) in Appendix A)
[40].
Piezoelectric transduction in
R. palustris-BTO biohybrids was first evaluated via polarization–electric field (
P−
E) hysteresis curves (
Fig. 2(c)). The quasi-rectangular loops with stable shapes in different electric fields indicated ferroelectrics with good piezoelectric responses. This was confirmed by PFM. As shown in
Fig. 2(d), the 180
o phase change angle in the local piezoresponse hysteresis loop and the classical amplitude loop represented the polarization switching process
[41]. In addition, a rough surface with clearly poled and repoled areas was also observed on the basis of PFM topography and corresponding amplitude maps, and the amplitude, frequency and phase mappings corresponded well to each other (
Fig. 2(e)), further revealing the ferroelectric domains
[42]. These results demonstrated that piezoelectric BTO inherently possesses excellent ferroelectric properties, as demonstrated in a previous study
[43]. Consistent results were obtained for the short-circuit current densities, which were as high as 1.4 μA when
R. palustris-BTO system were subjected to cyclic mechanical compression and release processes (
Fig. 2(f)). These values were slightly lower than those obtained with bare BTO (Fig. S3(h) in Appendix A), which we attributed to electron capture by the active sites on the cell surfaces of
R. palustris by components such as hydrogenases, c-type cytochromes and nanofilaments (as demonstrated below).
3.2. Mechanical energy-driven microbial growth with efficient NO3− reduction
To evaluate the potential for mechanical energy to be used as an energy source for microbial metabolism, we first examined the autotrophic growth of
R. palustris in biohybrids under stirring at 600 r·min
−1 over five successive 12 d cycles (i.e., a total of 60 d) (
Fig. 3(a)). The sterilized autotrophic minimal medium was refreshed periodically, with CO
2 and NO
3− serving as the electron acceptors (Table S1 in Appendix A). As shown in
Fig. 3(b), the cell biomass of
R. palustris significantly increased with mechanical stirring (∼10 fold,
P < 0.001), and both the ATP and DNA concentrations increased (Figs. S4(a) and (b) in Appendix A). In contrast, no significant increase in biomass was observed in the
R. palustris-stir group (
Fig. 3(b)), indicating that the absence of biomass growth in the
R. palustris-BTO system was not due to the limited CO
2 and NO
3− transformation. The CLSM images revealed higher cell concentrations and a higher live/dead ratio (13:1) after mechanical stirring (
Fig. 3(c) and Figs. S4(c) and (d) in Appendix A). Stable isotope analyses with labeled
13C-NaHCO
3 as the sole source of CO
2 revealed substantial incorporation of
13C into the
R. palustris-BTO biomass (
Fig. 3(d)), reaching a level of δ
13C 2349% after mechanical stirring (Fig. S5(a) in Appendix A). The C incorporated was approximately 28% of that obtained via phototrophic CO
2 uptake, whereas the negative control (i.e., no illumination and no stirring) showed no uptake of labeled carbon, with a δ
13C value almost identical to that of the biohybrids grown in
12C-NaHCO
3 medium (Fig. S5(b) in Appendix A). Taken together, these data confirmed biomass accumulation in
R. palustris via autotrophic CO
2 fixation with mechanical energy-derived electrons.
To evaluate the NO
3− reduction performance of
R. palustris-BTO, a series of experiments were conducted in which
R. palustris, BTO, and mechanical energy were systematically eliminated. As shown in
Fig. 3(e), almost no NO
3− reduction was observed in the absence of
R. palustris. In contrast, the NO
3− concentration decreased by 15% without BTO (
R. palustris-stir) or with mechanical stirring (
R. palustris-BTO), possibly because of residual acetate or biosynthetic intermediates remaining from the cultivation process of
R. palustris [44]. Only
R. palustris-BTO with mechanical stirring led to complete and continuous NO
3− reduction (
Fig. 3(f)), implying the synergistic action of different components. The nitrogenous products were then monitored and found to be N
2O-N (68.2% ± 3.4%), N
2-N (16.0% ± 1.9%), NH
4+-N (4.8% ± 2.0%), and NO
2−-N (0.7% ± 0.1%) after 12 d of mechanical stirring (
Fig. 3(g) and Fig. S5(c) in Appendix A), confirming an almost complete mass balance of nitrogen after 3 d of agitation of
R. palustris-BTO. Nitrate reduction was independent of magnetic energy, since nearly identical rates were observed in the mechanical and magnetically stirred samples (Fig. S5(d) in Appendix A). The NO
3− reduction rate with
R. palustris-BTO increased linearly as a function of increasing BTO concentration and stirring rate (Fig. S5(e) in Appendix A), potentially because more electrons were produced as reducing equivalents for
R. palustris. NO
3− reduction via the BPE process could be further improved with the addition of ascorbate as a sacrificial reagent and AQDS as a redox mediator (Fig. S5(f) in Appendix A), which are expected to increase electron–hole separation and electron transfer efficiency, respectively
[45],
[46]. Therefore, a powerful BPE process with mechanical stirring was shown to drive sustainable carbon fixation coupled with NO
3− reduction by
R. palustris-BTO.
3.3. Transcriptomic analyses of the BPE process
Transcriptomic analyses revealed that genes encoding the “stress” proteins DnaK (Hsp70) and GroEL (Hsp60) did not significantly change (
P > 0.05)
[28],
[30]. For electron transfer, hydrogen abstraction is an important pathway for extracellular electron uptake by
R. palustris, as indicated by the significant change in the transcriptional levels of genes encoding hydrogen-related products such as hydrogenases (
P < 0.001). Moreover, the nanofilaments of
R. palustris, including nanowires and flagella, were also highly active with mechanical stirring (
Fig. 4(a)), which is consistent with previous findings showing that the electrical properties of nanofilaments enhance electron exchange with extracellular electron acceptors/donors
[47],
[48],
[49]. In addition, widely distributed c-type cytochromes, such as the periplasmic cytochrome
c2, which enables the transfer of electrons into Complex III for ubiquinone cycling and the reduction of NAD(P)
+ to NAD(P)H with Complex I for energy metabolism, are considered the primary electron uptake factors due to their high expression with mechanical stirring
[50],
[51]. This was corroborated in our experiments, as the reduction potential of BTO was lower than that required to directly reduce NAD
+/NAD(P)
+ (Figs. S6(a)–(c) in Appendix A). Moreover, the genes encoding the typical F-type adenosine triphosphatases (ATPases) of
R. palustris for ATP synthesis, such as
atp1 and
atp2 [34], were also highly expressed. These results were consistent with the fact that autotrophic growth requires cell energy investment
[34]. To further confirm the active sites for electron transfer, inhibition experiments were conducted with different inhibitors, including antimycin A, CCCP, and KCN
[34],
[35],
[36]. Consistent with the transcriptomic analyses, a decrease in NO
3− reduction was observed with
R. palustris-BTO after the addition of any one of the three inhibitors (Fig. S6(d) in Appendix A), suggesting that all the active sites are involved in electron transfer during the BPE process. Notably, the metabolic pathway of
R. palustris in the presence of mechanical energy differed from that in the presence of light and oxygen, particularly with respect to processes involving electron generation, transfer, and utilization. Specifically, under conditions with sufficient light and oxygen, the reaction center (RC)–light harvesting (LH) complex, a critical component of
R. palustris, plays a crucial role in the efficient conversion of light energy into electrochemical energy
[52]. In contrast, during mechanical stirring, the transcription level of the RC–LH complex did not change significantly (
Fig. 4(a)), indicating the versatile metabolic mechanisms of
R. palustris.
The CBB cycle is an important electron sink for extracellular electron uptake by
R. palustris. As shown in
Fig. 4(b), the transcriptional levels of genes encoding CBB cycle enzymes were highly upregulated with mechanical stirring compared with those in the stationary control. Specifically, we observed a high expression of Form I (
cbbLS; ∼3.1 fold,
P < 0.001) and Form II (
cbbM; ∼8.6 fold,
P < 0.001) ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO) genes encoding enzymes that catalyze the interaction between ribulose 1,5-bisphosphate (RuBP) and CO
2 to produce 3-phosphoglycerate (3-PGA). 3-PGA is then converted into glyceraldehyde 3-phosphate (G3P) by phosphoglycerate kinase (
pgk; ∼1.1 fold,
P < 0.001) and transketolase (
tkl; 2.2–18.5 fold,
P < 0.001) with ATP and NAD(P)H as energy sources
[51],
[53],
[54]. The reactions that follow based on specific enzymes, such as phosphoribulokinase (
prk; ∼2.3 fold,
P < 0.001), further drive RuBP synthesis as a CO
2 acceptor molecule.
The relative transcriptional activities of the genes encoding denitrifying functions (NO
3− → NO
2− → NO → N
2O → N
2) in
R. palustris-BTO were also evaluated (
Fig. 4(c)). Mechanical stirring resulted in significant upregulation of the transcriptional levels of denitrification operons
nar,
nir,
nor, and
nos by 32–156, 3–636, 31, and 10–24 fold, respectively, in comparison with the stationary control. However, N
2O reductase is the most sensitive enzyme among the four N-oxide reductases
[55],
[56], and the oxidative species generated during the piezoelectric transduction of BTO, to some extent, influence its activity for complete N
2O reduction. H
2O serves as the ultimate electron donor, eliminating the need for additional sacrificial reagents during the growth and carbon fixation of electroactive microorganisms with mechanical energy. The O
2 concentration in both the headspace and solution increased under mechanical stirring (Fig. S7 in Appendix A). Both the nitrate- and CO
2-reducing equivalents were equal to the O
2 oxidative equivalent, indicating that organisms may prefer to use O
2 as the terminal electron acceptor because of its higher reduction potential. However,
R. palustris can grow and reduce nitrate under microaerobic conditions
[57],
[58],
[59]. The complex regulatory systems and enzymes in
R. palustris, such as heme, bacteriochlorophyll biosynthesis, NADH dehydrogenase genes and Cbb3 terminal oxidases, which have a high affinity for oxygen
[60],
[61], allow the bacteria to use different electron acceptors in response to changes in oxygen concentration
[59],
[62]. This finding was further supported by reports in previous studies showing that the denitrification process in
R. palustris can be facilitated in anaerobic zones formed under microaerobic conditions
[63].
3.4. Diversity of microbial metabolism based on mechanical energy
We further examined the possible activities of other piezoelectric materials in enhancing the metabolism of
R. palustris by replacing the artificial piezoelectric material SrTiO
3 with natural piezoelectric materials (i.e., tourmaline and quartz). Similar patterns were observed under identical experimental conditions, albeit at slower rates (
Fig. 5). We also showed that other methods of mechanical stimulation, such as ultrasonication and vibration, could be used to drive both carbon fixation and nitrate reduction (
Fig. 5). Finally, we tested other electroactive microorganisms for their ability to utilize piezoelectric stimulation to drive their metabolism via both soluble and insoluble electron acceptors. For example, the electroactive methanogen
M. barkeri and the electroactive acetogen
M. thermoacetica were successfully employed to drive CO
2-to-CH
4 and CO
2-to-acetate conversion with BTO under mechanical stirring, respectively. The electroactive sulfate-reducing bacteria
D. desulfuricans was shown to be capable of driving sulfate (SO
42−) reduction via BPE, whereas the electroactive microorganisms
T. denitrificans,
G. sulfurreducens, and
S. oneidensis were all successfully combined with BTO to effectively drive NO
3− reduction (95% for 6 d), decolorization of methyl orange (99% for 3 d), and Mn(IV) reduction (95% for 15 d) with mechanical stirring, respectively (Fig. S8 in Appendix A). As a control, biopiezoelectric catalysis experiments were also conducted with two typical electron-nonactive microorganisms,
E. coli and
B. subtilis, which are unable to utilize extracellular electrons
[64]. The results revealed no significant biomass growth or nitrate reduction with these microorganisms during continuous mechanical stirring (Fig. S9 in Appendix A). Taken together, these data indicate that the electronic energy produced from mechanical energy by piezoelectric materials can be harvested only by electroactive microorganisms as a universal microbial engine to drive Earth’s biogeochemical cycles.
These findings contrast with those previously reported for microbial disinfection with piezoelectric materials under aerobic conditions, that is, piezoelectric disinfection via generated reactive oxygen species (ROS)
[65],
[66]. This difference can be attributed not only to the shorter lifespan of piezoelectric activity-generated ROS under reducing and anoxic conditions but also to the efficient harvesting of piezoelectric activity-generated electrons for microbial metabolism, which inhibits electron transfer for ROS production. Although the lipid peroxidation experiments with the redox fluorescent probe C
11-BODIPY
581/591 indicated that the increase in the activity of the electron transfer chain led to a higher level of cellular ROS (Fig. S10 in Appendix A), self-regulation of intracellular antioxidant enzymes may help the microorganisms protect themselves from ROS toxicity
[67]. Therefore, oxidative damage to microorganisms in biohybrids may not be as severe as expected under mechanical stimulation. Notably, the productivity of the BPE system was greater than that of well-known phototrophic systems, in which significant loss of catabolic energy is observed during dark cycles; such a phenomenon was absent in the mechanical energy-driven system. Although issues related to low-efficiency dark catalysis could be addressed with microorganism-photosensitizer biohybrids
[27],
[68], that is, the biosynthetic intermediates generated by biohybrids during the light cycle could be used for microbial metabolism in the dark cycle, severe photooxidative damage caused by high light fluxes results in the production of more photogenerated electrons, which remains a substantial challenge.
More importantly, a large fraction of microorganisms on Earth live in the deep subsurface, where energy fluxes are orders of magnitude lower than those in the surface world
[7],
[8]. For example, deep-subsurface lithoautotrophic microbes have been found to thrive under oligotrophic conditions
[69]. However, natural rocks and materials with piezoelectric characteristics are also widespread in these areas
[70],
[71]. For example, piezoelectric quartz is the most abundant end-product of mineral weathering, and a higher than 3.2 Ga record of extensive quartz sandstones was discovered
[72],
[73]. Moreover, II-VI and III-V group natural semiconductors, such as zincite, sphalerite and stilleite minerals, are well known for their piezoelectric properties
[74]. If these natural piezoelectric materials are able to interact piezoelectrically with many microorganisms under various geodynamic processes, such as debris flows, vortices, faulting movements, and water currents (Fig. S11 in Appendix A), they may not only play an unappreciated role in the generation, growth and long-term survival of microbes in energy-limited environments but also elucidate the Earth’s multiple biogeochemical cycles that are driven by these energy-limited populations. For example, rich sulfate is widespread in the marine subsurface (30 mmol·L
−1 or more), where other electron acceptors with a relatively high reduction potential, such as nitrite, nitrate, and oxygen, are generally depleted
[75]. Thus, the electronic energy produced by potential geodynamic processes can potentially be used for the reduction of sulfate to sulfide, which plays a crucial role in the marine sulfur cycle and determines the redox state of the Earth. In addition, acetogens (bacteria) and methanogens (archaea) are important ancient anaerobic microbes present deep in the Earth’s crust, where there is very little energy to harness
[76]. Therefore, the acetate and methane produced via the BPE and metabolized by acetogens and methanogens may have played roles in both the origin and early evolution of life on Earth, in addition to the traditionally assumed H
2 metabolism
[77]. This inference is further confirmed by the results of the microbial growth experiments conducted in this study, which revealed that the mechanical stirring of
R. palustris-BTO, even once a week for 6 h, could significantly prolong microbial life under energy-limited conditions compared with a stationary control (Figs. S12(a)–(c) in Appendix A).
Notably, the lower the stirring speed or the shorter the stirring time was, the more dense/compact the biofilms (Fig. S12(d) in Appendix A). This might have occurred because cells tend to form dense/compact biofilms for efficient storage and utilization of sporadic energy produced by the limited mechanical energy (e.g., with a lower stirring speed or shorter stirring time). Owing to the difference in efficiency between electron production by piezoelectric particles (e.g., 10
−9 seconds for ferroelectric materials)
[78] and electron utilization by microbial metabolism (e.g., 10
−4 to 10
1 seconds for cytochrome-mediated processes)
[68], the stored energy could have a continuous and long-lasting impact on microbial growth and metabolism. Such a microbial adaptation strategy involving the BPE may explain how microorganisms survive in deep subsurface environments after many centuries of existence under extreme energy-limited conditions. The biopiezoelectric catalysis experiments detailed in this manuscript were conducted under strict laboratory experimental conditions to systematically explore mechanisms that remain unknown. However, it is believed that such phenomena are widespread in natural environments. For example, soil aggregates are composed of diverse electroactive microorganisms and mineral particles
[79],
[80], such as quartz, which possesses excellent natural piezoelectric properties
[81]. In addition, struvite, an excellent piezoelectric material, has also been found to form
in situ and may bind tightly with microorganisms during wastewater denitrification
[82],
[83],
[84].
4. Conclusions
Taken together, the results we report here reveal a pathway involving mechanical energy as an energy source that can stimulate growth and metabolism of electroactive microorganisms, and involving the participation of many genes and their gene products. These findings are important to microbial survival in the energy-limited environments, as well as an as-yet unappreciated contribution to the Earth’s multiple biogeochemical cycles (e.g., C, N, and S elements) and even the origin and early evolution of life. In addition, this process also offers a means for innovative, simple and sustainable approach for many applications such as pollutant degradation and biofuel generation.
Acknowledgements
This work was supported by the National Science Fund for Distinguished Young Scholars grant (41925028), and the National Natural Science Foundation of China grant (42322706, 42307176, and 42177206).
Appendix A. Supplementary Data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.eng.2024.08.006.
PDF
(2729KB)
