Electron Transfer Pathways and Vanadium Isotope Fractionation During Microbially Mediated Vanadate Reduction

Engineering ›› 2025, Vol. 46 ›› Issue (3) :271 -281. DOI: 10.1016/j.eng.2025.01.001

Research Environmental Engineering—Review

research-article

Author information +

History +

Received	Accepted	Published Online
2024-08-21	2025-01-09	2025-03-14

PDF (2335KB)

Abstract

Microbial vanadate (V(V)) reduction is a key process for environmental geochemistry and detoxification of vanadium (V). However, the electron transfer pathways and V isotope fractionation involved in this process are not yet fully understood. In this study, the V(V) reduction mechanisms with concomitant V isotope fractionation by the Gram-positive bacterium Bacillus subtilis (B. subtilis) and the Gram-negative bacterium Thauera humireducens (T. humireducens) were investigated. Both strains could effectively reduce V(V), removing (90.5% ± 1.6%) and (93.0% ± 1.8%) of V(V) respectively from an initial concentration of 50 mg·L⁻¹ during a 10-day incubation period. V(V) was bioreduced to insoluble vanadium (IV), which was distributed both inside and outside the cells. Electron transfer via cytochrome C, nicotinamide adenine dinucleotide, and glutathione played critical roles in V(V) reduction. Metabolomic analysis showed that differentially enriched metabolites (quinone, biotin, and riboflavin) mediated electron transfer in both strains. The aqueous V in the remaining solution became isotopically heavier as V(V) bioreduction proceeded. The obtained V isotope composition dynamics followed a Rayleigh fractionation model, and the isotope enrichment factor (ε) was (–0.54‰ ± 0.04‰) for B. subtilis and (–0.32‰ ± 0.03‰) for T. humireducens, with an insignificant difference. This study provides molecular insights into electron transfer for V(V) bioreduction and reveals V isotope fractionation during this bioprocess, which is helpful for understanding V biogeochemistry and developing novel strategies for V remediation.

Graphical abstract

Keywords

Vanadate / Bioreduction / Vanadium isotope fractionation / Electron transfer

Cite this article

Download citation ▾

Wenyue Yan, Baogang Zhang, Yi’na Li, Jianping Lu, Yangmei Fei, Shungui Zhou, Hailiang Dong, Fang Huang. Electron Transfer Pathways and Vanadium Isotope Fractionation During Microbially Mediated Vanadate Reduction. Engineering, 2025, 46(3): 271-281 DOI:10.1016/j.eng.2025.01.001

登录浏览全文

4963

注册一个新账户忘记密码

1. Introduction

Vanadium (V), a redox-sensitive metallic element, is ubiquitous in a wide range of natural minerals [1]. Its average content in the Earth’s crust (97 mg·kg⁻¹) is higher than that of common metals such as copper and nickel [2]. Vanadium possesses unique properties [3], including the formation of alloy carbides in steel to improve its toughness and catalyze denitrification in flue gas; therefore, it is intensively utilized in the high-tech industry [4]. Natural weathering and increasing anthropogenic emissions have led to the presence of V in the environment [5]. For instance, the highest dissolved V concentrations reach 58.6 mg·L⁻¹ in groundwater at the Chisman Creek superfund site (USA) [6], far exceeding the notification level (50 μg·L⁻¹) proposed by the California Department of Public Health (USA) for drinking water. Significant V accumulation in soil near industrial sites has also been detected in the North West Province, South Africa, and Panzhihua City (China) [4]. Potential health and ecological risks are also induced by V, as it can cause pulmonary tumors and harm organisms at high doses [7]. Vanadium has become a global contaminant of re-emerging concern [5], [7]. Surficial environments mainly contain trivalent, tetravalent, and pentavalent V (i.e., V(III), V(IV), and V(V)) [8], [9]. V(III) exists only under anoxic and sulfidic conditions [10]. V(V) in the form of vanadate is the most toxic and mobile form, whereas V(IV) is less toxic and is naturally deposited under neutral conditions [7], [11]. The geochemical cycling of V includes diverse processes such as adsorption/desorption and oxidation/reduction [4].

V(V) can be reduced to V(IV) through microbial catalysis under anaerobic conditions [12], which is recognized as a detoxification strategy for ecosystems to cope with V stress [13]. Microbial V(V) reduction has been observed in various environments, including groundwater, surface water, soil, and mine tailing sites [4], [14], where V accumulation occurs due to human activities related to V exploration. The phenomenon of V(V) bioreduction to V(IV) by microbial communities containing Gram-positive and Gram-negative bacteria has been previously observed [4], [15]. However, very few V(V)-reducing microorganisms have been identified and isolated [4], which mainly belong to Proteobacteria with Gram-negative characteristics [15], [16], [17]. This severely restricts the understanding of the V(V) reduction process and application of the resultant bioremediation. Microbial V(V) reduction is achieved through both extracellular and intracellular electron transfers [18]. Specific active compounds have been found in pure cells, such as cytochrome c and nicotinamide adenine dinucleotide (NADH), which mediate electron transfer [18]. Microbial metabolites may also shuttle electrons; however, their contribution to V(V) reduction remains largely unknown.

Vanadium has two stable isotopes, ⁵⁰V (0.24%) and ⁵¹V (99.76%) [19]. Based on theoretical investigations, V isotope fractionation occurs during redox reactions [20]. Vanadium isotope fractionation analysis is highly challenging because the proportional difference between these two stable isotopes is significant, resulting in difficulties in the precise measurement of low ⁵⁰V concentration. Only recent analytical advances have allowed the accurate and precise quantification of V isotope ratios [21]. V isotope fractionation can be used to investigate variations in redox fluctuations in natural environments [19]. At present, study [22] on V isotope fractionation have mainly focused on high-temperature geochemical processes such as mantle melting and mineral crystallization. Recently, systematically distinct V isotopic compositions from environmental media (water, soil, and sediment) and relevant biota (plants, lichens, and mushrooms) have been reported [21], [23], indicating the possible biological fractionation of V isotopes. Microbial reduction processes that induce isotope fractionation of redox-sensitive metals, such as Cr, Se, U, Sb, and Te, have been previously reported [24]. Microbial V(V) reduction is a pivotal step in V geochemistry. Unfortunately, information regarding microbially mediated V isotope fractionation during V transformation is still lacking.

To fill these knowledge gaps, the Gram-positive bacterium Bacillus subtilis (B. subtilis) and the Gram-negative bacterium Thauera humireducens (T. humireducens), the genera of which are widely distributed in V-polluted sites [14], were employed to explore the metabolic mechanisms of microbial V(V) reduction and the associated V isotope fractionation. The ability of these two strains to reduce V(V) was evaluated through process examination and product characterization. Molecular V(V) reduction pathways were comprehensively studied using electron microscopy, inhibitory trials, and metabolomics. Vanadium isotope fractionation was also observed during V(V) bioreduction by these two strains, which may provide insights into V biogeochemical processes.

2. Materials and methods

2.1. Strain incubation and solution preparation

B. subtilis and T. humireducens strains were obtained from Fujian Agriculture and Forestry University, China. Both strains are facultative anaerobic heterotrophs that live in mild environments with Fe(III) and nitrate as electron receptors under anaerobic conditions [25]. They can also detoxify various environmental pollutants [9]. The bacterial suspension (10 mL) was added to Luria–Bertani medium (yeast extract 5 g·L⁻¹, tryptone 10 g·L⁻¹, and NaCl 10 g·L⁻¹), and the strains were cultured in a shaker (135 r·min⁻¹) at 30 ℃. After 24 h of incubation, 40 mL of the bacterial solution was centrifuged (4000 ×g for 20 min, g is denoting gravity). The resultant solid layer was rinsed twice with sterile water and inoculated into 100 mL aqueous solution containing the following components (per L): 0.2460 g CaCl₂, 1.0572 g MgCl₂·6H₂O, 0.4459 g NaCl, 0.0283 g KCl, 0.8082 g NaHCO₃, 0.1557 g NH₄Cl, and 0.0299 g KH₂PO₄, with resultant optical density at 600 nm (OD₆₀₀) of 0.50. The solution pH was adjusted to 7.00 using 1 mmol·L⁻¹ HCl, and oxygen was removed by flushing with nitrogen gas. V(V) was added to the above solution in the form of NaVO₃ of analytical purity (Aladdin, China). Ultrapure water (18.2 MΩ·cm) was produced using an ultra-pure water system (Milli-Q Synthesis, Millipore, USA). All other analytical-grade chemicals were used directly without further purification.

2.2. Experimental procedure

The V(V) reduction performance of B. subtilis and T. humireducens was first studied over a 10 d period, with an initial V(V) concentration of 50 mg·L⁻¹ and a total organic carbon (TOC) concentration of 300 mg·L⁻¹ in the form of ethanol. Sterilized biomass was used for comparison. All experiments were conducted in triplicate sealed serum bottles in an incubator at (30 ± 2) °C. V(V) attenuation, TOC consumption, and biomass growth were monitored every 2 d. Microbial metabolic characteristics without V(V) were monitored as a control. The composition of the blue precipitates produced during the V(V) reduction was characterized after 10 d of incubation. The different cell components were separated, and their V(V) reduction abilities were compared by monitoring V (V) removal [18]. After the cultured strain was harvested by centrifugation, the residual solution was the cell-free extract. The biomass obtained was then ultrasonically broken and centrifuged. The precipitated substances were the membrane fractions. The resulting solution was defined as the cytoplasmic soluble fraction [26]. The distribution of reduction products in the cells was characterized using electron microscopy. Subsequently, electron transfer properties and metabolomic analyses were performed to study the metabolic pathways involved in V(V) reduction. Subsequently, temporal variations in the V isotope ratios (δ⁵¹V) were measured to investigate microbially induced fractionation.

2.3. Physicochemical characterization

Each aqueous sample (2 mL) was collected using a syringe, passed through a 0.22 μm filter, and stored at 4 °C before analysis. The soluble V(V) concentration was measured by spectrophotometry (detection limit of 2.2 μg·L⁻¹) at 601 nm using 2-(5-bromo-2-pyridylazo)-5-(diethylamino)phenol (5-Br-PADAP) as the chromogenic reagent [27]. Vanadium speciation was also simulated using HSC Chemistry (Version 6.0, Outotec, Finland) code with the Visual MINTEQ (Version 3.1, KTH Royal Institute of Technology, Sweden) database. The pH and oxidation–reduction potential (Eh) were measured using a multipurpose instrument (S400, Mettler-Toledo, Switzerland). The TOC was analyzed using a Multi N/C 3000 TOC analyzer (Analytik Jena AG, Germany).

The precipitates were collected by centrifugation (6000 ×g, 10 min) and freeze-dried. Their compositions were analyzed by energy-dispersive X-ray spectroscopy (EDS) using a scanning electron microscope (SEM; JEOL JAX-840, Hitachi Limited, Japan). The structures of the precipitates were determined using an X-ray diffraction (XRD) instrument (Rigaku-D/MAX-PC 2500, Rigaku, Japan) equipped with an X-ray tube (18 kW). The valences of V in the precipitates were analyzed using X-ray photoelectron spectroscopy (XPS; XSAM-800, Kratos, UK). The cell morphologies were examined using field-emission high-resolution transmission electron microscopy (TEM; JEM-2100F, Hitachi, Japan), and mapping was conducted to examine the elemental distributions. The washed cells were fixed in glutaraldehyde solution (2.5%, weight (w)/w) and osmium tetroxide (1%, w/w), and then dehydrated with acetone and polymerized [18]. Ultrathin sections (50–60 nm) of the cells were prepared using an ultramicrotome (EM UC6, Leica, Germany). The distribution of intracellular reduction products was evaluated using a line scan. After bioreduction, intracellular and extracellular V were collected by washing with 2 mol·L⁻¹ H₂SO₄ and ultrasonication, and then measured [17].

2.4. Microbiological analysis

Microbial biomass was detected by measuring the OD₆₀₀ using a spectrophotometer (DR6000, HACH, USA). The contents of extracellular cytochrome C on the bacterial outer membranes and intracellular NADH were measured spectrophotometrically at 415 and 340 nm, respectively. Intracellular glutathione (GSH) was determined using the Ellman method. These were normalized to volatile suspended solids (VSS), which were quantified by measuring the weight loss of bacterial samples after heating at 103 °C for 12 h in a muffle furnace [28]. The electron transport system activity (ETSA) of the strains was evaluated through reducing 2-(p-iodophenyl)-3-(p-nitrophenyl)-5-phenyl tetrazolium chloride, a type of exogenous electron acceptor, to formazan [26]. Inhibitory trials were conducted by adding a specific inhibitor targeting the corresponding active compounds (antimycin A for cytochrome c, rotenone for NADH, and carmustine for GSH) [29]. The functional groups of extracellular polymer substances (EPS) were characterized by Fourier transform infrared (FTIR) spectroscopy (Nicolet 5700, Thermo Fisher Scientific, USA).

To conduct the metabolomic investigation, supernatants from the V(V)-free and V(V)-present cultures were filtered through a 0.22 μm membrane, and metabolites in the supernatants were detected using ultra high-performance liquid chromatography-Fourier transform mass spectrometry (Exactive HF-X, Thermo Fisher Scientific). The analysis was performed on an ACQUITY UPLC HSS T3 column (Waters, USA) with a 2 μL injection at 40 °C. Mobile phase A consisted of 95% water and 5% acetonitrile (containing 0.1% formic acid), whereas mobile phase B consisted of 47.5% acetonitrile, 47.5% isopropanol, and 5% water (containing 0.1% formic acid). The metabolites in the microbial cells were analyzed using gas chromatography-mass spectrometry (GC-MS; 8890B-5977B, Agilent, USA). After derivation, 1 μL samples were injected into the GC-MS system in shunt mode, separated by a DB-5MS capillary column (Agilent, USA), and then detected by mass spectrometry with an electron bombardment ion source. Differential metabolites were screened when the variable importance in the projection value was above 1.00, and the metabolite was identified as a differential metabolite (p < 0.05). The metabolic pathways of the metabolites were obtained from the Kyoto Encyclopedia of Genes and Genomes (KEGG) expression database.

2.5. Vanadium isotope measurement and calculation

Aqueous samples were collected at selected time points during V(V) reduction by the two strains. The obtained solution was purified by chromatography to remove other elements, with primary and subsequent separations using Bio-Rad AG50W-X12 (200–400 mesh) cationic resin and AG1-X8 anionic resin (Biorad, USA), respectively [30]. The chemical purification procedures were conducted in an ultraclean laboratory. High-precision V stable isotopes were then measured using a Neptune Plus multicollector inductively coupled plasma mass spectrometer (MC-ICP-MS; Thermo Fisher Scientific), employing in-house standard University of Science and Technology of China-vanadium (USTC-V) as a blank with negligible mass (< 1.5 ng) compared to the V used for measurement (5–10 μg) [31]. A dry plasma injection method with an Aridus II (CETAC Technologies, USA) membrane dissolving system was employed to select the jet sample and X intercept cones in the medium-resolution mode to improve the sensitivity of the V isotope measurement [24]. On-peak zero correction was performed for each V isotope analysis. Vanadium isotope components were measured on a Faraday cup connected to 10¹⁰ Ω, and all other Faraday cups of MC-ICP-MS were connected to 10¹¹ Ω. The external precision of the employed V isotope (δ⁵¹V) analysis method was better than ±0.08‰ (two standard deviations (2SD)) on the basis of repeated analyses of BDH standard (δ⁵¹V = (–1.24‰ ± 0.08‰), 2SD, n = 292) and USTC-V standard (δ⁵¹V = (0.06‰ ± 0.08‰), 2SD, n = 186) [32].

Vanadium isotope data are reported as relative deviations from a standard reference material in parts per thousand (per mil or thousand):

(1)

δ^{51} V = [\frac{{(\frac{^{51} V}{^{50} V})}_{Sample}}{{(\frac{^{51} V}{^{50} V})}_{AA}}] \times 1000 ‰

where Alfa Aesar (AA) is the widely accepted zero-δ⁵¹V standard for V isotope measurement [33].

The V isotope fractionation factor (α) indicating the fractionation degree between the product and reactant was defined as:

(2)

α = R_{product} / R_{reactant}

where R_product and R_reactant are the V isotope ratios (⁵¹V/⁵⁰V) of the product and reactant, respectively.

A Rayleigh distillation model applied to closed systems such as batch experiments was employed to obtain α using the following equation:

(3)

δ_{t} = (δ_{0} + 1000) f^{α - 1} - 1000

where δ_t is the δ⁵¹V value of aqueous V in the solution at time t during the experiment, δ₀ is the initial δ⁵¹V value, and f is the fraction of aqueous V remaining in the solution.

For convenience, the kinetic fractionation factor α was converted into an isotope enrichment factor (ε) using the following formula:

(4)

ε \approx (α - 1) \times 1000 ‰

Statistical analysis was conducted using a one-way analysis of variance followed by Duncan’s multiple range test (p < 0.05) using SPSS (version 25.0, IBM, USA) software. Uncertainties of 2SD for the isotope data (n = 3) and 1SD for the concentration data (n = 3) are reported.

3. Results

3.1. V(V) bioreduction behaviors and products

The soluble V(V) concentration decreased over time in B. subtilis and T. humireducens (Fig. 1(a)) at an initial V(V) concentration of 50 mg·L⁻¹. In the 10 d incubation period, V(V) removal efficiency reached (90.5% ± 1.6%) and (93.0% ± 1.8%) for B. subtilis and T. humireducens, respectively. Minimal V(V) was removed by sterilizing B. subtilis and T. humireducens after three cycles of incubation. The TOC concentration declined over time, with a gradual increase in OD₆₀₀ (Fig. 1(b)). T. humireducens exhibited higher organic consumption and biomass yield, corresponding to faster V(V) removal, than B. subtilis.

SEM images showed that the precipitates were closely attached to the bacterial cells after incubation (Fig. S1(a) in Appendix A). The EDS analysis revealed that these precipitates contained V (Fig. S1(b) in Appendix A). Minimal obvious characteristic peaks of the V compounds were detected in the XRD spectra (Fig. 2(a)), indicating that the precipitates were amorphous. The valence state of V in the precipitates was analyzed using XPS (Fig. 2(b)). A sub-band with a peak binding energy of 515.9 eV was identified in both strains. There were also V 2p peaks at 517.1 and 523.7 eV detected in XPS spectra.

3.2. Active sites for V(V) bioreduction

The membrane, cytoplasmic soluble, and cell-free extract fractions of both strains showed reduced V(V) (Fig. 3(a)). The cytoplasmic soluble fraction of B. subtilis showed the highest V(V) reduction ability during 10 d of incubation, with a V(V) removal efficiency of (38.0% ± 4.8%). In contrast, the highest V(V) reduction efficiency was obtained for the membrane fraction of T. humireducens, with a V(V) removal efficiency of (37.3% ± 3.2%) in a 10 d cycle.

The TEM images showed obvious precipitates associated with the intact cells of B. subtilis and T. humireducens after V(V) reduction (Fig. 3(b)). The elemental mapping results revealed the deposition of solid V on the cell surface and the surrounding environment. A higher V intensity was observed in intact T. humireducens cells. Small, dark particles were also found in the ultrathin sections of B. subtilis and T. humireducens (Fig. 3(c)). The line scanning results showed that the intracellular energy intensity of V was higher than that outside the cells for both strains. Weaker intracellular V energy intensity was observed in T. humireducens, indicating that it tends to reduce V(V) extracellularly.

3.3. Electron transfer for V(V) bioreduction

Compared to the control group without V(V), the cytochrome c content increased significantly when V(V) was added (p < 0.01; Fig. 4(a)). NADH content also increased with exposure to V(V), especially in the Gram-positive B. subtilis, from (82.0 ± 7.8) to (156.1 ± 7.1) mg·g⁻¹ normalized to VSS. Another intracellular reductive substance, GSH, was detected, although its content decreased by 25.4% and 22.3% upon the addition of V(V) in B. subtilis and T. humireducens, respectively. The ETSA was improved with V(V) addition (Fig. 4(b)), by an average of 2.86 and 3.35 folds for B. subtilis and T. humireducens, respectively. When each inhibitor was added, the V(V) removal efficiencies during the 10 d cycle decreased considerably (p < 0.001; Fig. 4(c)). However, the resistance from these inhibitors differed between the two strains. NADH was more critical for B. subtilis, whereas T. humireducens depended more on GSH for V(V) reduction (Fig. 4(a)). Hydroxyl (–OH) and carboxyl (–COO^–) groups were detected in the FTIR spectra of the EPS from B. subtilis and T. humireducens (Fig. 4(d)).

Metabolomic analysis showed that the expression of metabolic pathways related to intracellular and extracellular electron transfer differed when the two strains were exposed to 50 mg·L⁻¹ V(V) compared to V(V)-free conditions. The ubiquinone and other terpenoid-quinone biosynthesis pathways were differentially expressed (Fig. S2 in Appendix A). Differential metabolites were also detected in this pathway (Fig. 5(a)). For biotin metabolism, the same enriched metabolites were detected in the supernatants of both strains (Fig. 5(b)). Riboflavin metabolism was significantly different between the supernatants of B. subtilis with and without V(V) (p < 0.05). The abundance of flavin mononucleotides (FMN) increased considerably (p < 0.05; Fig. 5(c)). Comparatively, the expression of this pathway was not significant in T. humireducens (p > 0.05). Riboflavin was also significantly enriched (p < 0.05). GSH metabolism also displayed different expression levels in the tested samples, with the exception of the T. humireducens supernatant. The abundance of glutamate increased significantly after the exposure of B. subtilis to V(V) (p < 0.05; Fig. 5(d)). Regarding oxidative phosphorylation, significant accumulation of FMN and succinate was observed when B. subtilis was subjected to V(V) stress (p < 0.05; Fig. 5(e)).

3.4. Vanadium isotope dynamics

The δ⁵¹V value for NaVO₃ from Sinopharm (China) was (–0.97‰ ± 0.05‰). The δ⁵¹V value of the remaining aqueous V displayed upward tendencies during bacterially mediated V(V) reduction, reaching (0.14‰ ± 0.03‰) for B. subtilis and (–0.15‰ ± 0.08‰) for T. humireducens (Fig. 6(a)). Moreover, the V isotope ratios varied in surficial environments, including environmental matrices (e.g., soil and sediment) and biota, ranging from approximately –2.0‰ to 2.0‰ (Fig. 6(b)). Our obtained δ⁵¹V values were within this range. Considering the well-mixed and closed conditions without V(V) input and output in our study, the V isotopic fractionation factor was assumed to be constant according to the Rayleigh distillation model, with ε values of (–0.54‰ ± 0.04‰) for B. subtilis and (–0.32‰ ± 0.03‰) for T. humireducens (Fig. 6(a)). The variation in V isotope fractionation between these two bacterial strains was insignificant (p > 0.05).

The speciation simulation results indicated that HVO₄^2– was the dominant species of the original V(V), whereas the produced V(IV) mainly existed in the form of solid VO(OH)₂ with trace VO²⁺ in the aqueous solution (Fig. S3 in Appendix A), in the measured ranges of pH 7.00–7.93, and Eh from –120.0 to –32.0 mV during V(V) bioreduction.

4. Discussion

4.1. Performance and mechanisms of V(V) bioreduction

The output of V(V) concentration monitoring in the experimental and sterilized groups indicated that B. subtilis and T. humireducens could reduce V(V) through microbial metabolism. Their V(V) removal efficacy was superior to that of previously reported strains. For example, the V(V) removal efficiency of Lactococcus raffinolactis (L. raffinolactis) during a 10 d operation was approximately 86.5% [34]. A comparatively higher V(V) reduction rate was observed for T. humireducens, attributable to the thinner cell walls characteristic of Gram-negative bacteria, which facilitate more efficient electron transfer [35]. The decline in TOC concentration, along with an increase in OD_600, suggested growth metabolism accompanied by V(V) reduction by both strains. The formation of precipitates implied that the reduction products of V(V) were insoluble. The sub-band with peak binding energy at 515.9 eV of the XPS spectrum corresponded to V(IV) [26], indicating that the transformation from V(V) to insoluble V(IV) was facilitated through microbial mediation. The V 2p peaks at 517.1 and 523.7 eV detected in the XPS spectra were attributed to the binding energy of V 2p_3/2 and V 2p_1/2 for V(V), likely due to the V(IV) re-oxidation during testing [36].

The ability of all separated cell components to reduce V(V) and the distribution of solid V in the TEM images implied that V(V) reduction occurred both intracellularly and extracellularly for both strains. In B. subtilis, the higher V(V) reduction ability of the cytoplasmic soluble fraction and stronger V energy intensity within cells indicated that this strain primarily reduced V(V) in cells, consistent with the previously reported Gram-positive L. raffinolactis for V(V) reduction [34]. In contrast, most of the V(V) was reduced extracellularly by T. humireducens, as its membrane fraction exhibited the highest V(V) reduction performance, with a higher V intensity for the intact cells of T. humireducens.

The increase in the content of electron transporters demonstrated that V(V) reduction occurred through electron transfer. Cytochrome c is embedded in the cell membrane and can transfer electrons from the inside to the outside [37]. NADH is oxidized to NAD⁺, and the NADH/NAD⁺ redox couple is a ubiquitous electron transporter in cells [38]. NADH, as a reductant, may also be involved in the intracellular reduction of V(V) to V(IV) [39], [40]. Another intracellular reductive substance, GSH, is oxidized to glutathione disulfide (GSSG) by forming a disulfide dimer [41]. The GSH/GSSG redox couple is an electron transporter involved in diverse cellular metabolic processes [38]. The role of GSH in V(V) reduction was confirmed in a thermophilic microbial consortium [42]. The decrease in GSH content upon exposure to V(V) was probably due to the reciprocal transformation between GSH and NADH [43]. The improved ETSA further indicated that V(V) reduction was realized through enhanced electron transfer. The results of the trials with inhibitors implied that these electron transporters play important roles in V(V) reduction. Microbially secreted EPS can also deliver electrons because their reductive groups (–OH and –COO^–) donate electrons to V(V), facilitating V(V) reduction [26]. Moreover, they can chelate the V(IV) produced.

The differentially expressed pathways with enriched differential metabolites indicated that additional electron transfer pathways might be involved in V(V) reduction, as revealed by metabolomic analysis. Quinones are redox media that participate in the bioreduction of heavy metals via extracellular electron transport [44]. During biotin metabolism, the electron transfer activity of biotin has been observed at the interfaces of biotin-modified gold electrodes [45]. Biotin has been used as a bridge to change the conductance distribution between electrodes, indicating its electrochemical activity [46]. FMNs are electronic carriers involved in microbial metabolism [47]. Riboflavin participates in various redox reactions in metabolic centers and acts as an electron shuttle for the microbial reduction of heavy metals [48]. These pathways may contribute to extracellular electron transfer during V(V) bioreduction. The increased abundance of glutamate in B. subtilis suggested that GSH is synthesized for electron transfer and V(V) reduction within cells. The significant accumulation of FMN and succinate in B. subtilis suggested that electrons are transferred for V(V) reduction via oxidative phosphorylation. These two pathways may contribute to V(V) bioreduction via intracellular electron transfer.

In summary, there were two types of electron transfer for V(V) bioreduction in these two strains. Extracellular electrons were transferred to V(V) by cytochrome c on the outer membrane and soluble metabolites, such as quinone, biotin, and riboflavin. In addition, V(V) could also be reduced to V(IV) intracellularly via electron transfer mediated by NADH and GSH.

4.2. Vanadium isotope fractionation during V(V) bioreduction

The δ⁵¹V value of NaVO₃ from Sinopharm was slightly different from that of vanadium titanomagnetite (–0.72‰ ± 0.10‰) [49], which is the main raw material for V production in China [50]. Vanadium isotope fractionation during ore smelting and reagent production may have contributed to this discrepancy. According to previous studies [21], [23], the highest δ⁵¹V values were found in the fruit bodies of Amanita muscaria (1.7‰), and the lowest values were found in marten liver (–1.7‰), compared to those in soil (–0.7‰) and sediment (–0.5‰), further demonstrating the fractionation of V isotopes under the influence of organisms.

The increase in δ⁵¹V values of the remaining aqueous V in solution along with V(V) reduction indicated that the lighter isotopes reacted preferentially during V(V) reduction through microbial catalysis. Biological uptake, adsorption, and redox reactions are the main low-temperature geochemical processes controlling the transportation of V [2], and are possibly accompanied by V isotope fractionation. Vanadium bioaccessibility and kinetic isotope effects, where the lighter isotope has a faster reaction rate, jointly determine V isotope fractionation during biological uptake [21]. Theoretical calculations have also shown that the light V isotope (⁵⁰V) is preferentially adsorbed on the surface of goethite [20], which is responsible for V isotope fractionation during adsorption. For redox reactions, lighter metal isotopes are favored in metabolic reactions, resulting in heavier isotope compositions in the remaining reactants [51]. This might also be true for V(V) bioreduction, thereby producing remarkable V isotope fractionation. Microbial reduction contributed primarily to V isotope fractionation in our study, whereas V isotope dynamics were likely affected by V(V) adsorption by microbes, which produced precipitates [20]. The slightly lower V isotopic fractionation factor for T. humireducens might be due to the more effective V uptake and V(V) reduction by gram-negative bacteria with thinner cell walls.

The transformation of V(V) to V(IV) is related to changes in the V–O bond that controls V isotope fractionation. Bond breaking involving light isotopic species is favored because it has a higher zero-point vibrational energy, resulting in a slightly smaller potential energy barrier to overcome [52]. When the V–O bonds were broken during the transformation of V(V) to V(IV), the activation energy for the light isotope during the evolution of the V–O bonds in the ground state of V(IV) relative to the transition state was lower. Strong isotope fractionation factors have also been found during environmentally relevant redox reactions of Cr and Se with microbial participation [53], [54]. Considering the discrepancy in the reactive sites of these two strains for V(V) reduction, it can be inferred that metabolic pathways may have less of an impact on V isotope fractionation. Experimental results have demonstrated this during Cr(VI) reduction by Shewanella oneidensis [55].

4.3. Environmental and geochemical implications

Although microbial V(V) reduction has been reported [4], the functional species that catalyze this process are limited. The finding that B. subtilis and T. humireducens can reduce V(V) augments the V(V)-reducing microbial family. These bacteria are common in the environment and their ability to remove pollutants has been demonstrated [56], [57]. Microbial V(V) reduction can be non-negligible in V redox cycling in the Earth’s surficial environments and indicates a promising route for V detoxification. Therefore, microbial agents can be used for the remediation of toxic V(V) polluted environments [4]. Once reduced, V(IV) is fixed within the cells and chelated by EPS, which is expected to have long-term stability, even if cell death occurs. In practical applications, geological structure and geochemical factors should be fully considered to achieve sustainable remediation without secondary pollution. In particular, the redox conditions should be carefully managed to prevent V(IV) re-oxidization. The scheme for injecting functional microorganisms and nutrients should be well designed. The elucidation of V(V) reduction pathways by these strains indicates that diverse compounds can promote electron transfer, especially the previously unidentified biotin. These soluble and solid electron transporters can be added during application [26] to accelerate V(V) reductive detoxification. It is important to note that electron transfer for V(V) bioreduction may be species-dependent; therefore, a broader range of bacterial species should be investigated to further elucidate the mechanisms involved.

This study accurately assessed V isotope fractionation during biochemical reactions through experiments for the first time, which has geochemical implications for Earth’s surficial environments. This supports the notion that large isotope fractionation occurs during the redox processes of variable-valency metals [58]. Therefore, V isotope fractionation can be used to identify pollution sources and trace pollution processes. V(V) reduction can be biotically and microbially mediated; therefore, the resultant V isotope fractionation may differ. Other geochemical and hydrodynamic factors such as electron donors and flow rates may also induce isotope fractionation during V(V) bioreduction, which requires further investigation. Other geochemical processes, such as adsorption/desorption, may also influence V isotope fractionation. Possible low isotope exchanges among the different valence states of V in such bioprocesses should also be experimentally evaluated.

5. Conclusions

This study investigated V(V) reduction by B. subtilis and T. humireducens with the involved electron transfer and concomitant V isotope fractionation. At the initial concentration of 50 mg·L⁻¹, (90.5% ± 1.6%) and (93.0% ± 1.8%) of V(V) was removed during a 10 d incubation by B. subtilis and T. humireducens, respectively. Insoluble V(IV) was the main product of V(V) bioreduction and was distributed both inside and outside the cells. Improved compounds (cytochrome c, NADH, and GSH) and enriched metabolites (quinone, biotin, and riboflavin) mediated electron transfer for V(V) reduction. The δ⁵¹V value increased as V(V) bioreduction proceeded, following a Rayleigh fractionation model. Isotope enrichment factors of (–0.54‰ ± 0.04‰) for B. subtilis and (–0.32‰ ± 0.03‰) for T. humireducens were obtained. This study advances the understanding of electron transfer mechanisms for V(V) bioreduction and V isotope fractionation in V biogeochemical cycling.

Acknowledgment

This research work was supported by the National Natural Science Foundation of China (U21A2033) and the Fundamental Research Funds for the Central Universities (2652022103). We thank Chuanyu Chang from the University of Science and Technology of China for isotopic analysis and the high-performance computing platform of China University of Geosciences Beijing.

Compliance with ethics guidelines

Wenyue Yan, Baogang Zhang, Yi’na Li, Jianping Lu, Yangmei Fei, Shungui Zhou, Hailiang Dong, and Fang Huang declare that they have no conflict of interest or financial conflicts to disclose.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.eng.2025.01.001.

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	Bellenger JP, Wichard T, Kustka AB, Kraepiel AML.Uptake of molybdenum and vanadium by a nitrogen-fixing soil bacterium using siderophores.Nat Geosci 2008; 1:243-246.

[2]	Schlesinger WH, Klein EM, Vengosh A.Global biogeochemical cycle of vanadium.Proc Natl Acad Sci 2017; 114:E11092-E11100.

[3]	Lu Z, Zhang Y, Wang H, Xia C, Liu Y, Dou S, et al.Transparent thermally tunable microwave absorber prototype based on patterned VO₂ film.Engineering 2023; 29:198-206.

[4]	Zhang B, Zhang H, He J, Zhou S, Dong H, Rinklebe J, et al.Vanadium in the environment: biogeochemistry and bioremediation.Environ Sci Technol 2023; 57:14770-14786.

[5]	Watt JAJ, Burke IT, Edwards RA, Malcolm HM, Mayes WM, Olszewska JP, et al.Vanadium: a re-emerging environmental hazard.Environ Sci Technol 2018; 52:11973-11974.

[6]	Jia L, Anthony EJ, Charland JP.Investigation of vanadium compounds in ashes from a CFBC firing 100% petroleum coke.Energy Fuels 2002; 16:397-403.

[7]	Gustafsson JP.Vanadium geochemistry in the biogeosphere–speciation, solid-solution interactions, and ecotoxicity.Appl Geochem 2019; 102:1-25.

[8]	Huang JH, Huang F, Evans L, Glasauer S.Vanadium: global (bio)geochemistry.Chem Geol 2015; 417:68-89.

[9]	Zhang H, Jiao S, Xing Y, Jiang B, Zhou S, Zhang B.Unveiling soil microbiome adaptation and survival strategy under vanadium stress in nationwide mining environments.J Geophys Res: Biogeosci 2024; 129:e2023JG007655.

[10]	Shaheen SM, Alessi DS, Tack FMG, Ok YS, Kim K, Gustafsson JP, et al.Redox chemistry of vanadium in soils and sediments: interactions with colloidal materials, mobilization, speciation, and relevant environmental implications—a review.Adv Colloid Interface Sci 2019; 265:1-13.

[11]	He J, Zhang B, Tan C, Tang Y, Shen Z, Wu S, et al.Distinguishing contributions of diverse sediment components to vanadium transport, immobilization and transformation in aquifer.Water Res 2024; 265:122248.

[12]	Lai C, Dong Q, Chen J, Zhu Q, Yang X, Chen W, et al.Role of extracellular polymeric substances in a methane based membrane biofilm reactor reducing vanadate.Environ Sci Technol 2018; 52:10680-10688.

[13]	Yan G, Sun X, Dong Y, Gao W, Gao P, Li B, et al.Vanadate reducing bacteria and archaea may use different mechanisms to reduce vanadate in vanadium contaminated riverine ecosystems as revealed by the combination of DNA-SIP and metagenomic-binning.Water Res 2022; 226:119247.

[14]	Sun X, Qiu L, Kolton M, Häggblom M, Xu R, Kong T, et al.Reduction by Polaromonas spp. in vanadium mine tailings.Environ Sci Technol 2020; 54:14442-14454.

[15]	Ortiz-Bernad I, Anderson RT, Vrionis HA, Lovley DR.Vanadium respiration by Geobacter metallireducens: novel strategy for in situ removal of vanadium from groundwater.Appl Environ Microb 2004; 70:3091-3095.

[16]	Carpentier W, Sandra K, De Smet I, Brig Aé, De Smet L, Van Beeumen J.Microbial reduction and precipitation of vanadium by Shewanella oneidensis.Appl Environ Microbiol 2003; 69:3636-3639.

[17]	Zhang J, Dong H, Zhao L, McCarrick R, Agrawal A.Microbial reduction of precipitation of vanadium by mesophilic and thermophilic methanogens.Chem Geol 2014; 370:29-39.

[18]	Wang G, Zhang B, Li S, Yang M, Yin C.Simultaneous microbial reduction of vanadium (V) and chromium (VI) by Shewanella loihica PV-4.Bioresour Technol 2017; 227:353-358.

[19]	Wu F, Owens JD, Scholz F, Huang L, Li S, Riedinger N, et al.Sedimentary vanadium isotope signatures in low oxygen marine conditions.Geochim Cosmochim Acta 2020; 284:134-155.

[20]	Wu F, Qin T, Li X, Liu Y, Huang J, Wu Z, et al.First-principles investigation of vanadium isotope fractionation in solution and during adsorption.Earth Planet Sci Lett 2015; 426:216-224.

[21]	Ch Jételat, Nielsen SG, Auro M, Carpenter D, Mundy L, Thomas PJ.Vanadium stable isotopes in biota of terrestrial and aquatic food chains.Environ Sci Technol 2021; 55:4813-4821.

[22]	Chen Z, Ding X, Kiseeva ES, Lin X, Huang J, Huang F.Vanadium isotope fractionation of alkali basalts during mantle melting.Lithos 2023; 107082:442-443.

[23]	Dong L, Wei W, Xu L, Lin Y, Liu Z, Pan S, et al.Vanadium isotope evidence for seawater contribution to V enrichment/mineralization in early Cambrian metalliferous black shales.Sci Bull 2024; 69:1006-1010.

[24]

Johnson TM, Jennifer D, Basu A, Jemison N, Wang XL, Kathrin S, et al.A review of the development of Cr, Se, U, Sb, and Te isotopes as indicators of redox reactions, contaminant fate, and contaminant transport in aqueous systems.W.W.S. Kenneth (Ed.), Isotopic constraints on earth system processes, John Wiley & Sons, New York City 2022; 237-269.

[25]	Ma C, Yu Z, Lu Q, Li Z, Zhou S.Anaerobic humus and Fe(III) reduction and electron transport pathway by a novel humus-reducing bacterium, Thauera humireducens SgZ-1.Appl Microbiol Biotechnol 2015; 99:3619-3628.

[26]	Fei Y, Zhang B, Chen D, Liu T, Dong H.The overlooked role of denitrifying bacteria in mediating vanadate reduction.Geochim Cosmochim Acta 2023; 361:67-81.

[27]	Deng J, Huang KH, An L, Hu CH, Ju SQ, Xiao N.Cloud point extraction and simultaneous spectrophotometric determination of V(V), Co(II) and Cu(II) ions in water samples by 5-Br-PADAP using partial least squares regression.J Radioanal Nucl Chem 2014; 300:835-842.

[28]	Bridgewater LL, Baird RB, Eaton AD, Rice EW.American public health association, In: Standard methods for the examination of water and wastewater. 23rd edition. Washington, DC: American Public Health Association; 2017.

[29]	Wang S, Zhang B, Fei Y, Liu H, Zhao Y, Guo H.Elucidating multiple electron-transfer pathways for metavanadate bioreduction by actinomycetic Streptomyces microflavus.Environ Sci Technol 2023; 57:19921-19931.

[30]	Qi YH, Wu F, Ionov DA, Puchtel IS, Carlson RW, Nicklas RW, et al.Vanadium isotope composition of the bulk silicate earth: constraints from peridotites and komatiites.Geochim Cosmochim Acta 2023; 259:288-301.

[31]	Wu F, Qi Y, Yu H, Tian S, Hou Z, Huang F.Vanadium isotope measurement by MC-ICP-MS.Chem Geol 2016; 421:17-25.

[32]	Wu F, Qi Y, Yu H, Tian S, Hou Z, Huang F.Coupled variations in V–Fe abundances and isotope compositions in latosols: implications for V mobilization during chemical weathering.Geochim Cosmochim Acta 2022; 320:26-40.

[33]	Prytulak J, Nielsen SG, Ionov DA, Halliday AN, Harvey J, Kelley KA, et al.The stable vanadium isotope composition of the mantle and mafic lavas.Earth Planet Sci Lett 2013; 365:177-189.

[34]	Zhang B, Li Y, Fei Y, Cheng Y.Novel pathway for vanadium (V) bio-detoxification by Gram-positive Lactococcus raffinolactis.Environ Sci Technol 2021; 55:2121-2131.

[35]	Zhou X, Kang F, Qu X, Fu H, Alvarez P, Tao S, et al.Role of extracellular polymeric substances in microbial reduction of arsenate to arsenite by Escherichia coli and Bacillus subtilis.Environ Sci Technol 2020; 54:6185-6193.

[36]	Elmouwahidi A, Bailón-García E, P AFérez-Cadenas, Fernández-Sáez N, Carrasco-Marín F.Development of Vanadium‐coated carbon microspheres: electrochemical behavior as electrodes for supercapacitors.Adv Funct Materials 2018; 28:1802337.

[37]	Zhao Y, Hsieh HS, Wang M, Jafvert CT.Light-independent redox reactions of graphene oxide in water: electron transfer from NADH through graphene oxide to molecular oxygen, producing reactive oxygen species.Carbon 2017; 123:216-222.

[38]

Henke AH, Laudadio ED, Orbeck JKH, Tamijani AA, Hoang KNL, Mason SE, et al.Reciprocal redox interactions of lithium cobalt oxide nanoparticles with nicotinamide adenine dinucleotide (NADH) and glutathione (GSH): toward a mechanistic understanding of nanoparticle-biological interactions.Environ Sci Nano 2021; 8:1749-1760.

[39]	Carpentier W, Sandra K, De Smet I, Brig Aé, De Smet L, Van Beeumen J.Microbial reduction and precipitation of vanadium by Shewanella oneidensis.Appl Environ Microbiol 2003; 69:3636-3639.

[40]	Islam MK, Tsuboya C, Kusaka H, Aizawa S, Ueki T, Michibata H, et al.Reduction of vanadium (V) to vanadium (IV) by NADPH, and vanadium (IV) to vanadium (III) by cysteine methyl ester in the presence of biologically relevant ligands.BBA-Gen Subjects 2007; 1770:1212-1218.

[41]	Kretzschmar J, Strobel A, Haubitz T, Drobot B, Steudtner R, Barkleit A, et al.Uranium(VI) complexes of glutathione disulfide forming in aqueous solution.Inorg Chem 2020; 59:4244-4254.

[42]	Zheng X, Zhao B, Liu C.Bio-reduction mechanism of V(V) by thermophilic hydrogen-producing bacteria under acidic conditions.Environ Sci Wat Res 2021; 7:1657-1665.

[43]	Maiti M, Murali VP, Selvakumar D, Podder A, Maiti KK, Bhuniya S.NADH-induced “kick-on” fluorescent probe validates crosstalk with redox regulator GSH.Sens Actuators B Chem 2019; 299:126964.

[44]	Newman DK, Kolter RA.Role for excreted quinones in extracellular electron transfer.Nature 2000; 405:94-97.

[45]	Kar P, Tatard F, Lamblin G, Banet P, Aubert PH, Plesse C, et al.Silver nanoparticles to improve electron transfer at interfaces of gold electrodes modified by biotin or avidin.J Electroanal 2013; 692:17-25.

[46]	Zhang B, Song W, Pang P, Lai H, Chen Q, Zhang P, et al.Role of contacts in long-range protein conductance.Proc Natl Acad Sci 2019; 116:5886-5891.

[47]	Light SH, Su L, Rivera-Lugo R, Cornejo JA, Louie A, Iavarone AT, et al.A flavin-based extracellular electron transfer mechanism in diverse Gram-positive bacteria.Nature 2018; 562:140-144.

[48]	Suzuki Y, Kitatsuji Y, Ohnuki T, Tsujimura S.Flavin mononucleotide mediated electron pathway for microbial U(VI) reduction.Phys Chem Chem Phys 2010; 12:10081-10087.

[49]	Huang Y, Long Z, Zhou D, Wang L, He P, Zhang G, et al.Fingerprinting vanadium in soils based on speciation characteristics and isotope compositions.Sci Total Environ 2021; 791:148240.

[50]	Leaching characteristics of vanadium in mine tailings and soils near a vanadium titanomagnetite mining site.J Hazard Mater 2014;264:498–504.

[51]	Wiederhold JG.Metal stable isotope signatures as tracers in environmental geochemistry.Environ Sci Technol 2015; 49:2606-2624.

[52]	DePaolo DJ.Surface kinetic model for isotopic and trace element fractionation during precipitation of calcite from aqueous solutions.Geochim Cosmochim Acta 2011; 75:1039-1056.

[53]	Basu A, Johnson TM, Sanford RA.Cr isotope fractionation factors for Cr (VI) reduction by a metabolically diverse group of bacteria.Geochim Cosmochim Acta 2014; 142:349-361.

[54]	Schilling K, Basu A, Wanner C, Sanford RA, Pallud C, Johnson TM, et al.Mass-dependent selenium isotopic fractionation during microbial reduction of seleno-oxyanions by phylogenetically diverse bacteria.Geochim Cosmochim Acta 2020; 276:274-288.

[55]	Han J, Chen G, Qin L, Mu Y.Metal respiratory pathway-independent Cr isotope fractionation during Cr(VI) reduction by Shewanella oneidensis MR-1.Environ Sci Technol Lett 2017; 4:500-504.

[56]	Wu S, Lu S, Liu J, Yang S, Yan Q, Jiang Z.Physicochemical properties and bioactivities of rice beans fermented by Bacillus amyloliquefaciens.Engineering 2021; 7:219-225.

[57]	Ma J, Wang F, Fan H, Li E, Chu H, Zhou X, et al.Metagenomic insight reveals the microbial structure and function of the full-scale coking wastewater treatment system: gene-based nitrogen removal.Engineering 2024; 36:76-89.

[58]	Wen H, Carignan J.Selenium isotopes trace the source and redox processes in the black shale-hosted Se-rich deposits in China.Geochim Cosmochim Acta 2011; 75:1411-1427.