Humin was extracted from the surface soil of the Kamajima paddy field (Endoaquept, Yatomi City, Aichi Prefecture, Japan) using a previously described method (Zhang et al., 2015a) with some modifications. In brief, the soil was air dried and sieved using a 1-mm mesh sieve. The sieved soil (100 g) was repeatedly washed by shaking with a chemical solution (24 h), followed by centrifugation (8,000 g, 15 min, 24°C) and decantation. Chemical solutions for washing were as follows: Ten times with 150 ml of 2% HF, 10 times with 150 ml of 0.1 M NaOH, 10 times with 150 ml of 2 % HF, and 20 times with 150 ml of ultrapure water. After washing, the pH was adjusted to 7.0, using 0.1-M NaOH. The pH-adjusted humin was freeze dried, ground using a ceramic mortar and pestle, and subjected to the experiments.

Thirteen N-fixing bacterial type strains of different phyla and one N-fixing archaeon were used: Azorhizobium caulinodans JCM 20966, Ensifer fredii JCM 20967, Rhodobacter sphaeroides JCM 6121, Pelomonas saccharophila JCM 15912, Derxia gummosa JCM 20996, Rubrivivax gelatinosus JCM 21318, Azotobacter vinelandii JCM 21475, Pseudomonas stutzeri JCM 5965, Geobacter sulfurreducens DSMZ 12127, Paenibacillus macerans JCM 2500, Clostridium pasteurianum JCM 1408, Clostridium tyrobutyricum JCM 11008, Nocardia cellulans JCM 9965, and Methanosarcina barkeri JCM 10043. These strains were purchased from Japan Collection of Microorganisms (JCM) at the RIKEN BioResource Research Center (Tsukuba, Ibaraki, Japan), through the National Bio-resource Project of the Ministry of Education, Culture, Sports, Science, and Technology (MEXT), Japan, and from the DSMZ-German Collection of Microorganisms and Cell Cultures GmbH, Germany. They were maintained as cultures in the media recommended by JCM or DSMZ, as shown in Table 1.

Three modified nitrogen-deficient Ashby (MNDA) media (Ashby, 1907) were used to examine the N-fixation ability of the individual-type strains. The anaerobic medium was composed of MgSO₄·7H₂O (0.2 g/L), K₂SO₄ (0.1 g/L), NaCl (0.2 g/L), K₂HPO₄ (0.2 g/L), CaCO₃ (5 g/L), trace elements, vitamins, and an organic carbon source. The trace elements were provided by the addition of 1 ml/L of trace element solution SL-10 (Widdel et al., 1983) and vitamins by 10 ml/L vitamin solution (Sanford et al., 1996). The anaerobic condition of the medium was achieved by sparging with N₂ gas for 90 min, followed by flushing the headspace for 5 min. The different organic carbon sources used for the individual strains along with the recommended incubation temperatures, are summarized in Table 1, along with the recommended incubation temperature. The anaerobic MNDA HEPES medium was obtained by replacing the CaCO₃ buffer with 30-mM HEPES, and the anaerobic MNDA HCO3- medium was prepared by replacing CaCO₃ with NaHCO₃ (0.85g/L) and N₂ gas with a mixture of N₂ and CO₂ gases (80:20) for 90 min followed by flushing the headspace for 5 min.

To examine the effect of humin on the N-fixation activity of the individual diazotrophs, washed and starved cultures were prepared. The maintained individual diazotrophs grown in the recommended medium were washed three times, suspended in organic carbon source-free MNDA HEPES medium, and by decanting the supernatant after centrifugation at 2,500 g for 5 min at 4°C. The washed individual microbial biomass was transferred into 20 ml of anaerobic MNDA HEPES medium with an organic carbon source in 50 ml-volume vial and incubated under anaerobic conditions at the recommended temperature (30, 35, or 37°C depending on the diazotrophs, as shown in Table 1) for 2 weeks. Subsequently, the cultured individual diazotrophs were rewashed three times to remove the residual organic carbon source. The washed culture was then transferred to the organic carbon source-free anaerobic MNDA HEPES medium and incubated for 2 weeks for starvation, after which the washed and starved individual cultures were subjected to further experiments.

The effect of the different redox states of humin on the N-fixation activity of the type strains was examined. An oxidized and a reduced humin were prepared in 200 ml of 0.5-M Na₂SO₄ solution using an electrochemical system consisting of a potentiostat (Automatic Polarization System HSV-110, Hokuto Denko, Osaka, Japan), two twisted platinum electrodes (1 m in length and 0.8 mm in diameter) as the working and counter electrodes, and an Ag/AgCl reference electrode [+0.199 V vs. the standard hydrogen electrode (SHE)], under anaerobic conditions using a vinyl anaerobic chamber (Coy-7450000, COY, Grass Lake, MI, USA). Two grams of autoclaved humin were reduced or oxidized by maintaining the redox potential at −0.4 V or +0.4 V (vs. SHE) for 24 h, respectively. After reaching the potential at the equilibrium state, the reduced and oxidized humin were collected by filtration through filter paper, dried using a vacuum pump, and subjected to the experiments.

The ARA assay is a widely used method for examining N-fixation activity (Hardy et al., 1968, 1973). Therefore, an ARA assay was performed to assess the activity of the nitrogenase enzyme of the individual diazotrophs, which reduces N₂ to NH₃ and other substrates, including the reduction of acetylene to ethylene (Burris, 1991). The ARA assay was conducted using a 10 ml glass vial (referred to as the test vial). The test vial was autoclaved and sealed with a butyl rubber stopper and aluminum seal to ensure that no air exchange occurred. When the effect of humin was examined, 0.03 g (15 g/L) of humin (intact, oxidized, or reduced) was added anaerobically before sealing the test vial. The test vial was flushed with He gas aseptically to expel the air inside the vial, and 2 ml of the anaerobic culture medium (organic carbon source-free MNDA medium) was added. Then, 2 ml of washed and starved culture was aseptically transferred to each test vial, followed by a 200-μl acetylene gas injection. The ARA assays of two strains, A. vinelandii and C. pasteurianum, were also performed in the medium with vanadium as sodium orthovanadate dihydrate, Na₃VO₄·2H₂O, by replacing Na₂MoO₄·2H₂O at the same concentration in SL-10 solution (MNDA V medium). The amounts of acetylene and ethylene on day 0 (before incubation) and day 7 (after incubation) in the headspace were determined using a GC-14B gas chromatograph (Shimadzu, Kyoto, Japan) equipped with a Molecular Sieve-5A column (60/80 mesh, 3-mm inner diameter, and 2-m length), a Porapak N column (50/80 mesh, 3-mm inner diameter, and 3-m length) in a series, and with a thermal conductivity detector and a flame ionization detector, following the procedure described in a previous study (Dey et al., 2021).

Nitrogen fixation activity of the individual diazotrophs was also determined by measuring the total Kjeldahl nitrogen using the Kjeldahl digestion method (Kjeldahl, 1883) followed by spectrophotometric measurement of ammonium ion using the indophenol blue method (Keeney and Nelson, 1982). Serum vials (50 ml volume) were prepared with 25 ml of organic carbon source-free anaerobic MNDA medium. The medium was supplemented with intact humin (15 g/L), oxidized humin (15 g/L), or the reduced humin (15 g/L) before the vials were sealed. Medium without humin was also provided. For M. barkeri, an organic carbon source-free MNDA HCO3- medium was used. The medium was inoculated with 2.5 ml of washed and starved cultures of the individual diazotrophs, and incubated for 14 days at an appropriate temperature (30, 35, or 37°C depending on the diazotrophs; Table 1). Before and after 14 days of incubation, 5 ml of the culture sample was collected using a syringe and subjected to Kjeldahl nitrogen determination. In addition, H₂ in the headspace was measured on day 0 (before incubation) and day 14 (after incubation) using a GC-14B gas chromatograph (Shimadzu, Kyoto, Japan) equipped with a thermal conductivity detector, as described above. The experiment was performed at least in triplicate for each type of strain.

To examine whether humin works as a carbon source for promoting the N-fixation activity of the individual diazotrophs, the total CO₂ was analyzed along with the ARA assay. The serum vials with 50-ml volume were filled with 35 ml of organic carbon source-free anaerobic MNDA HEPES medium in an anaerobic chamber and inoculated with the washed and starved culture 3.5 ml after the headspace was flushed with He gas for 5 min. One set of vials was supplemented with 15 g/L of the reduced humin and another set without humin. All vials were spiked with 1 ml of acetylene gas, and the prepared cultures were incubated for 14 days at an appropriate temperature (Table 1). Before and after the incubation, CO₂, acetylene, and ethylene in the headspace were determined using a GC-14B gas chromatograph (Shimadzu, Kyoto, Japan) equipped with a thermal conductivity detector and a flame ionization detector (described above). Aqueous phase HCO3- determination was performed by taking 15 ml of sample from the culture using a syringe before (0-day sample) and after incubation (14th-day sample). The samples were filtered using a membrane filter with 0.22-μm pore size (Millex^®-Gv, Durapore™ PVDF membrane), and the HCO3- concentration in the filtered sample was determined using a total organic carbon analyzer (TOC–VCPH, Shimadzu, Kyoto, Japan). For A. vinelandii and C. pasteurianum, the organic carbon source-free anaerobic MNDA HEPES V medium was also provided by replacing Na₂MoO₄·2H₂O in the SL-10 solution with Na₃VO₄·2H₂O at the same concentration. For other diazotrophs, Na₂MoO₄·2H₂O containing the SL-10 solution was used.

To determine the carbon saturation condition for maximum ARA of the individual diazotrophs, anaerobic MNDA medium containing different concentrations (up to 6 times of the standard concentration) of organic carbon was prepared. For the ARA assay, 2 ml of anaerobic MNDA medium with different organic carbon concentrations was added to six sets (with at least three replicates each) of test vials, and an ARA assay was performed as described above after the inoculation of 2 ml of washed and starved culture. The organic carbon concentration, defined as the organic carbon-saturated concentration at which diazotrophs exhibited the maximum ARA, was determined. The standard concentration of organic carbon (1 ×) was 0.125 g/L of glucose, 0.125 g/L of soluble starch, and 0.075 g/L of Na-pyruvate for A. vinelandii; 1 g/L of glucose and 0.125 g/L of soluble starch for C. tyrobutyricum and C. pasteurianum; for E. fredii, it consisted of 5 g/L of mannitol; and for P. macerans, it was 1 g/L glucose. For A. vinelandii and C. pasteurianum, the medium used was anaerobic MNDA V medium supplemented with Na₃VO₄·2H₂O, replacing Na₂MoO₄·2H₂O in SL-10 solution with the same concentration of Na₃VO₄·2H₂O; these diazotrophs had vanadium nitrogenase. For other diazotrophs, an SL-10 solution containing Na₂MoO₄·2H₂O was used for molybdenum nitrogenase.

The ARA assay of the individual diazotrophs was performed by incubation with the inoculation with washed and starved culture under carbon-saturated conditions with the reduced humin. The autoclaved test vials were provided, allowing five sets (with at least three repetitions each), and the first to fourth sets were supplemented with 15, 30, 45, and 60 g/L of the reduced humin, respectively; the fifth set contained no humin. The test vials were added aseptically with 2 ml of He-bubbled anaerobic MNDA medium with 3- or 4-times organic carbon concentration (saturated concentration), followed by 2 ml of the washed and starved culture inoculation. After injecting 200 μl of acetylene gas into each vial, an ARA assay was performed on day 0 (before incubation) and day 7 (after incubation). For A. vinelandii and C. pasteurianum, a He-bubbled anaerobic MNDA V medium was used. For other diazotrophs, a normal SL-10 solution containing Na₂MoO₄·2H₂O was used.

The bacterial cultures were serially diluted (from 10⁻¹ to 10⁻⁴) with PBSE buffer (130-mM NaCl, 1 mM EDTA, 30 mM K₂HPO₄, 30 mM KH₂PO₄, pH 7.0), filtered using a membrane filter with 0.2 μm pore size, and dried using a vacuum pump. Then, a drop of DAPI solution (ProLong^TM Gold, Thermo Fisher Scientific, Waltham, MA, USA 02451) was placed over the membrane filter, the samples were kept in the dark for 2 h, and observed under a fluorescent microscope (Olympus DP70, Tokyo, Japan) using an excitation filter WU (Blue) with an excitation wavelength of 320–385 nm.

The ATP assay was performed using the CellTiter-Glo^® 2.0 assay (G9241, Promega Corporation, Madison, WI, USA) and Victor nivo™ multimode plate reader (PerkinElmer, Inc. Waltham, MA, USA). Serum vials (50 ml volume) were provided with 20 ml of carbon source-free MNDA HEPES medium, which was bubbled with N₂ for 90 min, and the headspace was flushed with N₂ for 5 min prior to inoculation. For A. vinelandii and C. pasteurianum, the medium used was an organic carbon source-free anaerobic MNDA HEPES V medium, and M. barkeri had an organic carbon source-free MNDA HCO3- medium. One set of vials was supplemented with the 15-g/L reduced humin, and another set was not supplemented. Two milliliters of the washed and starved cultures were aseptically inoculated into each vial. Before incubation (day 0) and after incubation (day 3), an ATP assay was performed with 100 μl of culture according to the manufacturer's protocol. In brief, 100 μL of the sample was transferred to an opaque-walled multiwell plate, and an equal volume of CellTiter-Glo^® 2.0 was added to the sample, and the contents were mixed for 2 min using an orbital shaker. The plate was then incubated at 23°C for 10 min to stabilize the luminescent signal, and luminescence was recorded at 700 nm.

To construct the nitrogenase overexpression mutant, nifDK genes (gene ID: AOZ74834 and AOZ74835), from C. pasteurianum JCM 1408 (Supplementary Figure S1C) that encodes nitrogenase, were amplified using Phusion High-Fidelity DNA Polymerase (New England BioLabs, MA, USA), using nifDK primers nifDK-F_KpnI (5′- GGGGGGTACCTTTAATTTTGATGAGGGGTG−3′) and nifDK-R_XbaI (5′- GGTCTAGATGTCAATTCCATATAGAATGAC−3′), where the italic sequences denote restriction sequences for cloning. The PCR product was purified using a QIAquick PCR Purification Kit (Qiagen, Venlo, Netherlands) according to the manufacturer's instructions. The purified PCR product and cloning vector pBBR1MCS-2 were digested with KpnI and XbaI (New England BioLabs, MA, USA), and the digested PCR product was cloned into the digested pBBR1MCS-2 (pBBRnifDK) (Supplementary Figure S1A). Also, pBBRnifDK was introduced into Escherichia coli JM109λpir, derived from the E. coli K-12 strain. The expression of nifD and nifK in the constructed mutant was confirmed by qRT-PCR to be induced by 100-μM isopropyl β-D-1-thiogalactopyranoside (IPTG). The mutant was designated nifDK-OE.

The nifDK-OE mutant was cultured in a 20-ml N-free M9 medium containing 100 μM or no IPTG at an initial optical density at 600 nm (OD₆₀₀) of 0.5 under anaerobic conditions at 30°C, and harvested 1 h after incubation. The total RNA was extracted using ISOGEN (Nippon Gene Co., Tokyo, Japan) according to the manufacturer's protocol. The extracted RNA was purified using the RNeasy Mini Kit and RNase-free DNase set (Qiagen, Venlo, Netherlands). The concentration and quality of the purified RNA were evaluated using a DeNovix DS-11 spectrophotometer (Denovix Inc., DE, USA) and the Agilent 4150 TapeStation system with RNA ScreenTape and RNA ScreenTape reagents (Agilent Technologies Inc., CA, USA) according to the manufacturer's protocol.

Quantitative reverse transcription PCR (qRT-PCR) was conducted using a LightCycler 1.5 instrument with LightCycler RNA Master SYBR Green I (Roche, Basel, Switzerland) following the manufacturer's protocol (Kouzuma et al., 2012). Briefly, a PCR reaction mixture contained 25-ng total RNA, 1.3 μl of 50-mM Mn(OAc)₂ solution, 7.5-μl of LightCycler RNA Master SYBR Green I (Roche), and 0.15 μM primer sets [16S rRNA: 341F-518R (Luo et al., 2010), nifD: qrt-nifD-F 5′-GGTTGTGCTTATGCAGGATG-3′ and trt-nifD-R 5′-TCCTATAGGTCCGTGTGTGATG-3′, nifK: qrt-nifK-F 5′-GTTGCATTACTTGGAGATCCTG-3′ and qrt-nifK-R 5′-CTTCTGCAAGCATAGCATCG-3′) (Supplementary Figure S1I]. The DNA fragments of target genes (16S rRNA, nifD, and nifK) were amplified by PCR using Ex Taq polymerase (Takara Bio Inc., Shiga, Japan) and the same primer sets and purified by gel electrophoresis using a QIAEXII gel extraction kit (Qiagen, Venlo, Netherlands) according to the manufacturer's instructions to generate a standard curve. Standard curves were generated by amplifying a series of purified DNA fragments of each gene. Expression levels of target genes (nifD and nifK) were normalized based on the reference gene expression levels (16S rRNA).

The nifDK-OE mutant was cultivated in a single-chambered electrochemical reactor (115 ml capacity, Supplementary Figure S1B) under anaerobic conditions. The reactor was filled with 80 ml of modified N-free M9 medium, in which NH₄Cl was removed (LaCroix et al., 2015). The nifDK-OE cells were cultivated in N-free M9 medium at 30°C for 2 h to consume the nitrogen compounds in cells completely and were then inoculated into the medium containing 50 μg/ml kanamycin and 100-μM IPTG at an OD₆₀₀ of 0.01. When necessary, 5 g/L of humin was added to the medium. The reactor was sealed with a cap equipped with a carbon rod (diameter, 5 mm; length, 15 cm) as the working electrode; platinum wire (diameter, 0.8 mm; length, 1 m) as the counter electrode; and Ag/AgCl as the reference electrode, and purged with pure nitrogen gas for 30 min. The working electrode was poised at −0.5 V vs. SHE using a potentiostat (Automatic Polarization System HSV-110, Hokuto Denko, Osaka, Japan).

For the extraction of DNA from the nifDK-OE mutant, nifDK-OE cells incubated in the electrochemical reactor were collected after 0 and 72 h of incubation and subjected to the extraction of total DNA using the FastDNA SPIN Kit for Soil (MP Biomedicals, CA, USA) according to the manufacturer's instructions. The purified DNA was dissolved in 50 μl of DES solution in the kit.

The abundance of the target genes (16S rRNA) in the cultures was measured by quantitative PCR. The standard DNA fragments were amplified using primer sets 341F-518R for 16S rRNA (Dey et al., 2021) and purified using the QIAquick PCR Purification Kit (Qiagen, Valencia, CA, USA). The qPCR was performed using LightCycler systems and a LightCycler DNA Master SYBR Green I (Roche Diagnostics, Basel, Switzerland), as previously described (Dey et al., 2021).

Statistical analysis was performed using t-test, two-tailed Pearson's correlation assignment, or one-way ANOVA followed by post hoc analysis by Tukey's method using the software IBM, SPSS, v.21.

Fourteen heterotrophic diazotrophs from different taxonomic groups were used to test the effect of humin on BNF activity using the washed and starved cells in a modified nitrogen-deficient Ashby (MNDA) medium, based on the acetylene reduction activity (ARA) and total Kjeldahl nitrogen content. Two Clostridium spp. were included in 14 diazotrophs as the representatives of Clostridiales, which were major taxonomic group in the nitrogen-fixing anaerobic consortia promoted by humin in the previous study (Dey et al., 2021). The test was carried out by providing 10⁹ cells/ml of the diazotrophs initially (Supplementary Table S1). Figure 1 shows the effect of humin on ARA (indicator of nitrogenase activity) and total Kjeldahl nitrogen in the cultures of A. vinelandii and C. pasteurianum incubated in the absence of organic carbon source. Only the cultures with the reduced humin showed increases in both ARA and total Kjeldahl nitrogen. No increase was observed in the cultures with intact humin, oxidized humin, or no humin, in the absence of other organic carbon source. Other 12 heterotrophic diazotrophs also showed the increases in ARA (Supplementary Figure S2) and Kjeldahl nitrogen (Supplementary Figure S3) with the reduced humin in absence of other organic carbon source. Table 2 summarizes the specific BNF (nitrogenase) activities, i.e., BNF activity per cell, of the heterotrophic diazotrophs with the reduced humin in the absence of organic carbon source. Since the fixed nitrogen, detected as the increased Kjeldahl nitrogen, and promotion of ARA by the reduced humin showed a significant positive correlation (r = 0.779, p < 0.01) (Supplementary Figure S4), the ARA assay was used for evaluating the N-fixing activity in the subsequent experiments. The MNDA medium contained molybdenum; therefore, molybdenum nitrogenase activity was promoted by the reduced humin. The promotion was also observed in vanadium nitrogenase using A. vinelandii and C. pasteurianum, when vanadium was used in the incubation medium instead of molybdenum (Figure 2). Since BNF generally accompanies H₂ production, this substance was also identified. Most diazotrophs produced slightly more H₂ than that calculated from the stoichiometry of the BNF reaction with the reduced humin (Supplementary Figure S5, Supplementary Table S1). The role of humin as an electron mediator to promote BNF, and not as a source of organic carbon for diazotrophs, was confirmed by no increase in the CO₂ production and the presence of ARA with the reduced humin in organic carbon source-free MNDA media (Figure 3, Supplementary Figure S6). The cultures with the reduced humin did not show an increase in the amount of CO₂ but ARA (ethylene production). While, neither CO₂ nor ethylene production was detected in cultures incubated without humin. These results indicate that the reduced humin promoted BNF activity of a broad spectrum of diazotrophs, including alpha-, beta-, gamma-, and delta-Proteobacteria, Firmicutes, Actinobacteria, and Archaea, which harbor nitrogenases classified into clusters I, II, and III, regardless of whether molybdenum or vanadium, without producing CO₂. It was also confirmed that no other electroactive microorganism was involved in the BNF promotion of the individual heterotrophic diazotrophs with humin. That is, a broad taxonomic and enzymatic spectrum of diazotrophs can be activated using not genetically engineered, wild-type cells by extracellular electron transfer through humin.

The ARA assays of washed and starved diazotrophs were performed with different concentrations of organic carbon sources to determine the saturated concentration of ARA. Figures 4A,C show that ARA increased to a maximum at four times of the organic carbon source in the culture of C. pasteurianum and at three times in E. fredii. There was no significant difference in ARA at higher concentration of the organic carbon sources than the saturated concentration. Obtained with the same trend, the saturated concentration of ARA was determined also for other diazotrophs (Supplementary Figure S7). Another ARA assay was performed to examine the effect of humin in a medium containing a saturated concentration of the carbon source. The results illustrated in Figures 4B,D show that the addition of the reduced humin further increased the ARA of C. pasteurianum and E. fredii, after ARA had previously reached a maximum with the organic carbon-saturated concentration. This further promotion was observed regardless of their taxonomic positions, nifH clusters, and nitrogenase type, as shown in Supplementary Figure S8. The promotion effect increased with an increase in the amount of the reduced humin, probably because of the greater number of supplied extracellular electrons. The promotion reached in the range between 194% of A. vinelandii (the lowest) and 916% (the highest) of E. fredii in the cultures with 60 g/L of reduced humin compared with the BNF activity of the diazotrophs tested with saturated organic carbon concentration but no humin, as 100%. The addition of the reduced humin did not increase the CO₂ production, rather higher concentrations of the reduced humin inhibited the CO₂ production (Figures 4B,D, Supplementary Figure S8B), which indicated the slackening of organic carbon metabolism. These results suggest that the supply of extracellular electrons through the reduced humin can further accelerate ARA and BNF activity.

To elucidate the BNF promotion mechanism of the reduced humin, ATP synthesis in the diazotrophs was examined under conditions of the reduced humin with no organic carbon source. The results (Figure 5, Supplementary Figure S9) showed that all six washed and starved diazotrophs harboring three different nifH clusters (clusters I, II, and III), and containing both molybdenum and vanadium nitrogenases, had increased ATP content in their cells without producing CO₂ (Figure 3, Supplementary Figure S6). Controls without inoculation showed no increase in ATP levels, regardless of the presence of the reduced humin. This indicates that the reduced humin activated oxidative phosphorylation for ATP synthesis in broad taxonomic and nitrogenase spectra of diazotrophs. The extracellular electrons from the reduced humin most likely enter the diazotrophs through the cell membrane upstream of the oxidative phosphorylation process taking place.

Moreover, to examine whether humin can transfer electrons directly to nitrogenase protein without the involvement of nitrogenase reductase, the nifDK gene overexpression mutant (nifDK-OE) was cultured in an N-free M9 medium containing glucose under conditions of intact humin or no humin using an electrochemical system. The atmospheric nitrogen was provided as nitrogen source. The experiment was carried out by supplying electricity as an external electron source (closed circuit) and without the supply of electricity (open circuit) for both intact humin and no humin conditions (Supplementary Figure S1). The nitrogen fixation activity of the mutant was assessed indirectly by measuring the growth of the mutant using copy number of 16S rRNA gene before (0 h) and after 72 h of the incubation. Increase in the relative copy number of 16S rRNA gene, that is, the growth of mutant was observed only under the conditions of humin and a closed circuit; no growth was observed under the other conditions (the conditions of open circuit, and the condition of no humin and closed circuit), which was attributed to nitrogen deficiency of the mutant for growth (Figure 6A). The results indicate that the presence of humin aided the nifDK-OE mutant to receive electrons from external electron supply from the cathode, and these received electrons were utilized by MoFe protein (nitrogenase) of the mutant to fix N₂ necessary for their growth. As a consequence, only the mutant under the condition of humin and closed circuit showed the growth. It is noted that, in natural BNF, MoFe protein receives electrons for nitrogen fixation through Fe protein (nitrogenase reductase), which was not present in the nifDK-OE mutant. In the nifDK-OE mutant, the expression of nifD and nifK was induced in presence of isopropyl β-D-1-thiogalactopyranoside (IPTG), which was added to all samples and confirmed by qRT-PCR (Figure 6B). These results suggest that humin is able to transfer electrons to nitrogenase (MoFe protein) directly because the mutant does not have nitrogenase reductase (Fe protein) genes. The mutant is also not an electroactive microorganism, indicating a direct electron transfer into the cell without relying on a biological mechanism.