Electricity from lignocellulosic substrates by thermophilic Geobacillus species

Given our vast lignocellulosic biomass reserves and the difficulty in bioprocessing them without expensive pretreatment and fuel separation steps, the conversion of lignocellulosic biomass directly into electricity would be beneficial. Here we report the previously unexplored capabilities of thermophilic Geobacillus sp. strain WSUCF1 to generate electricity directly from such complex substrates in microbial fuel cells. This process obviates the need for exogenous enzymes and redox mediator supplements. Cyclic voltammetry and chromatography studies revealed the electrochemical signatures of riboflavin molecules that reflect mediated electron transfer capabilities of strain WSUCF1. Proteomics and genomics analysis corroborated that WSUCF1 biofilms uses type-II NADH dehydrogenase and demethylmenaquinone methyltransferase to transfer the electrons to conducting anode via the redox active pheromone lipoproteins localized at the cell membrane.


Scientific Reports
| (2020) 10:17047 | https://doi.org/10.1038/s41598-020-72866-y www.nature.com/scientificreports/ Here we explore and demonstrate the ability of strain WSUCF1 to generate electricity under thermophilic conditions from corn stover and food waste, the model substrates for lignocellulosic biomass and complex organic wastes, respectively. We used a combination of proteomics and genomics analysis, electrochemistry methods and chromatography tests to discern the EET mechanisms of WSUCF1 biofilms.

Results and discussion
Electricity generation from pure glucose by Geobacillus sp. strain WSUCF1.. Upon inoculation with WSUCF1 cells in the 3-electrode electrochemical cell (3-EC), operated in a fed batch mode that entailed dual cycles, the voltage followed a gradually increasing pattern and stabilized at 0.065 µA (Fig. 1a). Operational details of the cyclic fed-batch processes including length of cycles are provided in Table S1. The peak values of the open circuit voltage at the optimum growth temperature of 60 O C were 400 ± 50 mV and the current density were 45 ± 10 mW/m 2 , respectively (Fig. 1b). The poising potential of − 0.2 V (vs Ag/AgCl) was chosen based on the repeatable stable current output that we observed in the 3-EC tests ( Figure S2, Supplementary Information).
The first derivative analysis of the cyclic voltammograms (CVs) at low scan rate (10 mv/s) were used to determine the peak potentials and inflection points of the glucose oxidation by WSUCF1 (Fig. 1c,d). A dominant inflection point appeared at − 0.272 V versus Ag/AgCl and a minor peak at − 0.652 vesus Ag/AgCl. These peaks appeared consistently in the CV tests with all the other scan rates ( Figure S6, Supplementary Information). The control was devoid of any apparent biocatalytic current (Fig. 1d). The ability of WSUCF1 to transfer the electrons from glucose oxidation to the glassy carbon electrode was corroborated after observing a distinct current response from the cyclic voltammetry (CV) curve (8 days after the inoculation, Fig. 1e).
The Nyquist plot (Fig. 2a) and Bode plots (not shown) for WSUCF1 on Day 7 of Cycle 2 displayed two distinct time-constant regimes, one attributed to the polarization resistance (R p ) in the high frequency region and other  Table S1 for experimental details of 3-EC.
Scientific Reports | (2020) 10:17047 | https://doi.org/10.1038/s41598-020-72866-y www.nature.com/scientificreports/ to substrate oxidation resistance (R s ) in the low frequency region. The R s value for the 3-EC (49.87 kΩ cm 2 ) was 1.5-fold lower than that in the control (76.24 kΩ cm 2 ). R p (0.73 kΩ cm 2 ) was also significantly lower than the control (2.8 kΩ cm 2 ) (Fig. 2b). The ohmic resistance (R Ω ) in 3-EC (0.007 kΩ cm 2 ) was also 1.4-fold lower than the control (0.011 kΩ cm 2 ). Among the three major sources of resistance, R s was found to be the limiting resistance. The lower internal resistance in the WSUCF1 system corroborates the ability of WSUCF1 to generate electricity under thermophilic conditions. The differential pulse voltammetry (DPV) tests corroborated the redox pairs that appeared in the CV analysis (Fig. 2c). The two DPV peaks (− 0.450 V vs Ag/AgCl and 0.67 V vs Ag/AgCl) indicated the presence of redox active center on the surface of WSUCF1 cells. These DPV results were highly reproducible as the peak was always centered at − 0.45 V vs Ag/AgCl (One-way ANOVA, p value = 0.8). DPV tests revealed a broader peak which increased in height with the age of the biofilm (Fig. 2d). The increased height of peak "a'" (8.1 µA) to "a" (3.6 µA) showed the increased activity of WSUCF1 biofilm. Noting that we did not maintain strict anaerobic conditions prior to the inoculation, we attribute the peaks "b" (4.4 µA) and "b'" (2.5 µA) to the presence of dissolved oxygen. Evident from the reduced height of peak "b", the system eventually shifted to anaerobic conditions, confirming the facultative nature of WSUCF1 in MFCs (Fig. 2d).
Versatility of Geobacillus sp. strain WSUCF1 in MFCs. The strain WSUCF1 generated electricity from both corn stover and food waste (Fig. 3a), respectively. The experimental plan for the versatility analysis of WSUCF1, based on fed-batch 3-ECs that entailed dual cycles is shown in Table S2 (Supplementary Information). Corn stover and food waste yielded peak current density of 38 mA/m 2 and 28 mA/m 2 in Cycle 3 of the fed-batch operation, respectively. Although these values are lower than glucose (45 mA/m 2 ) (Figs. 1a, 3a), their current density profiles were devoid of any undesirable lag phase and they both yielded consistent current profiles (Fig. 3b). The CV peaks for corn stover and food waste were centered at − 0.368 V vs Ag/AgCl and − 0.442 V vs Ag/AgCl, respectively (Fig. 3c). Their electrochemical impedance spectroscopy (EIS) profiles were equivalent to that of pure glucose substrate (Nyquist plots, Fig. 3d), which confirmed the ability of the strain WSUCF1 to reduce charge transfer resistance to the Faradaic reactions influencing the oxidation of complex substrates. The three EIS profiles were carried under identical experimental conditions and on Day 5, Cycle 3 of the fed-batch operation.
To examine the reasons that allowed WSUCF1 to treat unprocessed corn stover in absence of any physicochemical pretreatment, we assessed the presence of endogenous hydrolyzing enzymes in WSUCF1 cultures. The methods used for measuring the enzyme activities are given in the supplementary information. The enzyme assay studies revealed cellulase, hemicellulases and laccase in the WSUCF1 cultures (Fig. 3e). The 10-day growth profile indicated highest activity of laccase in supernatant on day six (460 IU/L), endoxylanase on day eight (270 IU/L), and endoglucanase on day ten (110 IU/L). A data mining exercise revealed the presence of genes encoding for hemicellulases (xylan 1,4-β-xylosidase, endo-1,4-β-xylanase, and xylan α-1,2-glucuronosidase), www.nature.com/scientificreports/ cellulases (endoglucanases, -glucosidases) and amylases in the genome of strain WSUCF1 28,30-32 . We observed that these enzymes displayed stable activity even after 10 days of the prolonged exposure at 60 O C, corroborating the thermostability of these enzymes, as reported in previous studies 33,34 (Fig. 3e).

Mechanisms of extracellular electron transfer in Geobacillus sp. strain WSUCF1. Our final goal
is to elucidate the mechanism of extracellular electron transfer (EET) in WSUCF1. We carried out "Beads / No beads" tests, using cyclic fed-batch strategies shown in Table S3 (Supplementary Information) to ascertain the roles of mediated electron transfer (MET) mechanisms in WSUCF1. To avoid the growth of WSUCF1 biofilm on the anode surface in beads-MFC, the WSUCF1 cells were encapsulated in the calcium alginate beads prior to introducing them into the MFCs. Owing to the nano-scale dimensions of typical beads [(Φ = 6.8 to16.6 nm) 11 , extremely small inter-bead distance and high specific surface area (2.486 cm 2 g −1 )], we expect the beads to fully encapsulate the micron-scale WSUCF1 cells (Φ = 1 µm) [15][16][17][18] and prevent them from leaking into the anolyte. Furthermore, the calcium alginate beads remain stable at the operating temperatures considered in this study (60 °C) 19 . Thus the electricity generation in beads-MFC was primarily due to the planktonic cells. As shown, the WSUCF1 cells in no-beads-MFC adhered readily on the anode surface (Fig. 4a) compared to beads-MFC (Fig. 4b) and control (Fig. 4c). The power density (107 mW/m 2 ) in no beads-MFC was 2.9-fold higher compared to beads-MFC (37 mW/ m 2 ) (Fig. 4d). The performance of no-beads-MFC stayed intact even after the media replacement (42 and 46 mW/m 2 ) before and after the media replacement respectively; Fig. 4a, b, S7), whereas beads-MFC experienced an outright reduction in the power output (e.g., 14 to 0.16 mW/m 2 ) (data not shown). All of these results suggest the presence of endogenous redox mediators in WSUCF1-analyte. To study the nature of these mediators, CV tests using the spent-anolyte from no-beads MFC were carried out in a 3-EC cell (Details in Supplementary   Figure 3. Substrate versatility analysis of strain WSUCF1 using corn stover, food waste and control (a) Temporal current density; (b) Temporal current variation; (c) Representative differential pulse voltammetry scan; (d) Nyquist plots for the entire frequency range (10 kHz to 10 MHz); (e) Lignolytic enzyme activities in Geobacillus sp. WSUCF1 in presence of 1% corn stover. Notes: All experiments were carried out in a 3-electrode electrochemical cell. Experimental details of the cyclic fed-batch processes are shown in Table S2.

.2 (Supplementary Information).
When the biofilm forming capability of WSUCF1 was obviated by encapsulating the cells in beads, WSUCF1 secreted redox-active riboflavin to trigger the MET process. We used proteomics and genomics analysis to study the electron carrier proteins in the cytoplasm and outer membrane responsible for the MET processes. The anolyte samples used, and methods of proteomics and genomic analysis are described in Sects. 5 and 6, respectively (Supplementary Information). Among the analyzed proteins from the cytosolic and membrane protein fractions, a total of 11,427 unique peptides corresponding to 1720 different proteins were revealed. This study revealed various known proteins including cytochrome proteins involved in EET (Table 1) and nearly 154 uncharacterized proteins (Table S5, Supplementary Information) that are likely affiliated to the intracellular electron transport chain. These proteins were screened by identifying the subcellular localization of proteins using the PSORTb v.3.0.2 tool 35 and the screened proteins were assigned a function using the PROSITE database tool (prosite.expasy.org and predictprotein.org). PSORTb tool refers to the protein subcellular localization tool. The EET proteins identified from genomics analysis of WSUCF1 are shown in Table S6 ( Supplementary Information).
The proteomics analysis revealed three key proteins including type II NADH (T2NADH) dehydrogenase, demethylmenaquinone methyltransferase (DMM), pheromone lipoprotein (plp), and an anionic glycopolymer (teichoic acid). Here we establish the putative MET pathways in strain WSUCF1 based on the four key proteins (Fig. 5). The electrons from the glucose oxidation enter the electron transport chain through NADH ubiquinone oxidoreductase, ubiquinone pool, and transmembrane cytochrome proteins. The EET is initially aided by the www.nature.com/scientificreports/ T2NADH dehydrogenase 36 because T2NADH diverts the intracellular electrons away from aerobic respiration and funnel them into the quinone pool localized within the cytoplasmic membrane (Table 1). Quinones such as demethylmenaquinone, menaquinone and ubiquinone then selectively funnels the electrons into riboflavin 37 . We attribute the DMM protein (Table 1) to play a key role in facilitating the MET by riboflavin. The electrons from DMM to plp on the cell surface occur via some putative redox active proteins and cytochromes (R a and R b ) positioned along the teichoic acids anchored to the peptidoglycan layer. The mining of of WSUCF1 genome confirmed the presence of all the four genes of riboflavin operon which were previously reported for Bacillus subtilis 38,39 , a native riboflavin producer (Table S4, Supplementary Information). The presence of plp (Table 1) annotated as having FMN transferase activity transfers the electrons through riboflavin to the graphite electrode 36,40 .
In summary, the current study established the extracellular electron transfer capabilities of a thermophilic, Gram positive Geobacillus sp. strain WSUCF1 which allow them to liberate the electrons from the complex carbon substrates onto conductive solid electrodes. Here a Gram positive Geobacillus sp. strain WSUCF1 generated electricity directly from corn stover using its unique pathways of lignocellulolytic reaction, biofilm formation, Table 1. Proteins in WSUCF1 identified by LC-MS/MS analysis. a Only those proteins were present which had an exclusive unique peptide count of 2 or more. Proteins present had 100% matching at a peptide threshold of 95% and protein threshold of 95%. Exclusive unique peptide is the number of peptide sequences exclusively unique to a protein group.

Exclusive Unique peptides a Protein identified from LC-MS/ MS analysis Protein Subcellular Location GO Annotation/Function 41, 42 Membrane Fraction Cytoplasmic fraction
Menaquinone-cytochrome C reductase iron-sulfur subunit Cytoplasmic Membrane Oxidoreductase activity; oxidation reduction process 4 3 Menaquinol-cytochrome C reductase Cytoplasmic Membrane/Multi-pass membrane protein Although a clear evidence of electricity generation by WUSCF1 was shown, a series of gene-knock out studies are warranted to fully understand the metabolic network involved in the electron flow from complex wastes to the electrode. To discern the EET responses of WSCUCF1 cells exposed to steadystate physicochemical conditions, a series of studies based on continuous flow bioelectrochemical reactors are needed.

Methods
Methods and any associated references are available in the "Supplementary Information" of the manuscript. License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.