Aerobic biogenesis of selenium nanoparticles by Enterobacter cloacae Z0206 as a consequence of fumarate reductase mediated selenite reduction

In the present study, we examined the ability of Enterobacter cloacae Z0206 to reduce toxic sodium selenite and mechanism of this process. E. cloacae Z0206 was found to completely reduce up to 10 mM selenite to elemental selenium (Se°) and form selenium nanoparticles (SeNPs) under aerobic conditions. The selenite reducing effector of E. cloacae Z0206 cell was to be a membrane-localized enzyme. iTRAQ proteomic analysis revealed that selenite induced a significant increase in the expression of fumarate reductase. Furthermore, the addition of fumarate to the broth and knockout of fumarate reductase (frd) both significantly decreased the selenite reduction rate, which revealed a previously unrecognized role of E. cloacae Z0206 fumarate reductase in selenite reduction. In contrast, glutathione-mediated Painter-type reactions were not the main pathway of selenite reducing. In conclusion, E. cloacae Z0206 effectively reduced selenite to Se° using fumarate reductase and formed SeNPs; this capability may be employed to develop a bioreactor for treating Se pollution and for the biosynthesis of SeNPs in the future.

particulate SeNPs display special physical characteristics, such as photoelectric, semiconducting and X-ray sensing properties 11,22 . They also possess an adsorptive ability, antioxidant functions, and due to their high surface area-to-volume ratio, marked biological reactivity [23][24][25][26] . However, there is now growing concern about the environmental impact of nanoparticle synthesis based on physico-chemical methods that require high pressures and temperatures, are energy consuming, use toxic chemicals, and generate hazardous byproducts 27 . Consequently, applications using biological systems such as microbial cultures for the production of metal nanoparticles, including SeNPs, are becoming an increasingly realistic alternative 27,28 .
Reduction of selenite to Se° has been shown to be mediated by thiols (Painter-type reactions) in the cytoplasm as part of a microbial detoxification strategy 29 . Selenite reacts with GSH and forms selenodiglutathione (GS-Se-SG), which is further reduced to glutathione selenopersulfide (GS-Se − ) by NADPH-glutathione reductase. GS-Se − is an unstable intermediate and undergoes a hydrolysis reaction to form Se° and reduced GSH. In addition to Painter-type reactions, a number of terminal reductases for anaerobic respiration, two nitrite reductases, an inducible sulfite reductase and a fumarate reductase, have also been reported to be able to carry out selenite reduction in cells [30][31][32][33] .
Enterobacter cloacae SLD1a-1, a selenate-respiring facultative anaerobe, has been demonstrated to catalyze the reduction of both selenate and selenite to Se°1 2, 34, 35 , but the selenite or selenate concentrations adopted in these studies were extremely low (less than 1.5 mM). The reduction of selenate was shown to be mediated by a membrane-bound molybdoenzyme 36,37 , but the mechanism of selenite reduction in this strain has not been elucidated. Moreover, all of the selenite-reducing assays involving E. cloacae in these studies were performed in an anaerobic environment, and the selenite-reducing ability of E. cloacae has not previously been investigated under aerobic conditions. E. cloacae Z0206, a strain that we isolated from Reishi mushroom (Ganoderma lucidum) "meat" 38 , was found to possess excellent selenite resistance, tolerating more than 100 mM selenite. In the present investigation, we studied (i) the selenite-reducing ability of E. cloacae Z0206 under aerobic conditions, (ii) the characteristics and location of the produced SeNPs, and (iii) the mechanism of selenite reduction in the Z0206 strain.

Results and Discussion
Growth profile and selenite-reducing ability of E. cloacae Z0206 under different selenite concentrations. To determine the toxicity of selenite to the microorganism, the growth profile of E. cloacae Z0206 was studied under various concentrations of selenite (0, 0.5, 1, 5, 10, 15 mM). According to the apparent changes in the spent broth shown in Fig. 1A, we found that the strain formed a reddish cell suspension, which indicated its ability to reduce the toxic, colorless, soluble selenite ions to the non-toxic, red, insoluble elemental form of Se (Se°). It is worth noting that the red color of the broth darkened, and viscosity increased with the increase of the selenite concentration. The results regarding the growth profile (Fig. 1B) showed that the addition of selenite strikingly inhibited the growth rate of E. cloacae Z0206, and the inhibitory effect strengthened with the increase of the selenite concentration during the log phase. However, the final cell density in the presence of selenite (0.5 mM-15 mM) was comparable to that of the control without selenite addition, as verified by the result of a bacteria counting assay performed at 96 h (Fig. 1C). Evaluation of the selenite-reducing ability of the bacterium (Fig. 1D) showed that selenite was rapidly reduced by this strain, with 10 mM selenite being completely reduced in 144 h. As shown in Fig. 1E, the rates of selenite reduction were modeled using the Michaelis-Menten kinetic equation (see section 1.1, in Supplementary Information). A nonlinear least-square analysis of the data yield a K m value of 4.37 mM and a V max of 59.32 μmol/h/g. Scanning electron microscopy (SEM) analysis of the morphology of the bacterium and reduced selenite (Fig. 1F) revealed that as the selenite concentration increased, the rod-shaped cells tended to become shorter. Selenite at concentrations greater than or equal to 1 mM significantly stimulated the secretion of extracellular polymeric substances. SeNPs ranging from 100-300 nm were observed scattered around the cells and occurred as aggregates attached to the bacterial biomass in the presence of selenite, and the particle density grew with the increase of the selenite concentration. These results indicated that selenite merely reduced the growth rate of E. cloacae Z0206 rather than decreasing the final amount of bacteria. Additionally, the bacterium detoxified selenite by rapidly reducing it to Se° and formed SeNPs, highlighting the species as a promising exploitable option for setting up of low-cost biological treatment units for the bioremediation of Se-laden effluents.
Characterization and localization of SeNPs in Z0206 cultures. Energy Dispersive X-ray Spectrum (EDX) flat scanning of the area in Fig. 2A revealed a strong Se atom signal, accounting for 2.69% of the total component elements (Fig. 2B). EDX elemental mapping was used to detect the Se distribution. Four elemental maps of Se, carbon, oxygen and nitrogen were obtained and are shown in different colors based on the scanning area encompassing both the contents of E. cloacae Z0206 and the surrounding area (Fig. 2C). The map of elemental oxygen and nitrogen showed the cell shape and distribution of biomass. In contrast, element carbon was distributed both within and outside of cells because the cells were embedded by carbon-containing Epon plastic. However, the Se strong signals, shown in green, perfectly matched the profile of the extracellular nanoparticles, verifying that these particles were SeNPs. Moreover, Se, oxygen and nitrogen overlapped in the SeNPs distribution area, implying that these SeNPs may be coated with biomass. In addition, there was a weak Se signal inside the bacteria, suggesting that SeNPs may also exist inside the cells. However, Transmission electron microscopy (TEM) analysis revealed the presence of electron-dense nanoparticles both within and outside of the cells (Fig. 2D), which was not found in cell cultures without selenite (see Figure S2, in Supplementary Information). The EDX spectra of these nanospheres clearly indicated the presence of Se, as specific absorption peaks at 1.37 and 11.22 keV were recorded (Fig. 2E). The lack of peaks corresponding to other metals indicated that Se occurred in its elemental state (Se°) rather than as a metal selenide. This was confirmed by X-ray photoelectron spectroscopy (XPS) analysis, which shows clearly the 3D spectral peak of Se° (Fig. 2F). These results suggested that E. cloacae Z0206 reduces Se(IV) to Se(0) and assembles it into nanoparticles.

Selenite-reducing ability of different cellular fractions of E. cloacae Z0206 cells. Different mech-
anisms have been proposed for the reduction of selenite to Se° in microorganisms, including (i) Painter-type reactions 29 , (ii) the thioredoxin reductase system 39 , (iii) siderophore-mediated reduction 40 , (iv) sulfide-mediated reduction 6 and (v) dissimilatory reduction [30][31][32] . Among these potential mechanisms, i, ii and v occur inside the cell, while the other reactions occur extracellularly. To help determine how selenite was reduced by E. cloacae Z0206, subcellular fractions, including cytoplasm, membrane proteins and the supernatant from liquid cultures, were isolated after 12 h of growth without selenite, and the selenite-reducing ability of each fraction was evaluated. As shown in Fig. 3, selenite was reduced to orange-colored Se by only the membrane-associated proteins after the addition of NADH, whereas little orange-colored Se occurred without NADH. However, boiling the membrane fraction samples resulted in a complete loss of reduction activity (data not shown). These results indicated that the reduction process mediated by membrane-associated proteins was an enzymatic reaction and was NADH dependent. Interestingly, the cytoplasmic fraction and supernatant failed to change the color to orange, suggesting that these two fractions possess no selenite-reducing ability.
iTRAQ analysis of E. cloacae Z0206 proteins in response to selenite. It is well established that α, β and γ proteobacteria possess high GSH levels in the cytoplasm, which drive Painter-type reduction to reduce selenite 29 . Although E. cloacae Z0206 is a γ proteobacterium, the above results indicated that it may reduce selenite through a membrane-bound enzyme rather than through a thiol-related reaction. To identify the probable mechanism involved, we undertook a large-scale proteomic analysis (iTRAQ) to examine the modification of protein expression in response to selenite (complete data on protein quantification are shown in Table S1, in Supplementary Information). The analysis focused on the proteins whose expression varied by more than a factor of 1.2. According to this criterion, 172 proteins were induced after selenite treatment, and 212 proteins were significantly repressed. All of the significantly differentially regulated proteins identified were subjected to gene ontology analysis. Among the 287 proteins involved in biological processes, 274 proteins and 175 proteins were dedicated to metabolic processes and cellular processes, respectively. For each of the three ontologies, the annotated data revealed that the proteins were mainly distributed among two or three of the general term categories. In the biological process category, the three most abundant categories included proteins that are involved in metabolic processes (244), cellular processes (230), and single-organism processes (198). In the molecular process category, the two main categories involved in catalytic activity (207) and binding (172). In the cellular component category, the three largest categories were cell (153), macromolecular complex (63), and membrane (61) (see Figure S3, in Supplementary Information).
Among the identified proteins, selenite induced a 2.42-fold increase in fumarate reductase abundance (Table 1). Li, et al. 33 demonstrated that the reduction of selenite in Shewanella oneidensis MR-1 is mediated by fumarate reductase, indicating that fumarate reductase may play a role in the selenite-reducing process in E. cloacae Z0206. However, the most expected antioxidant proteins, such as glutathione synthetase, glutathione reductase, glutathione-disulfide reductase, thioredoxin, thioredoxin reductase and thioredoxin-dependent thiol peroxidase, showed no significant change in response to selenite treatment (see Table S2, Support Information).
Analysis of the mRNA abundance of selected genes. To verify the results of the iTRAQ analysis, the mRNA abundance of the enzyme fumarate reductase (frd) and the antioxidative enzymes glutathione synthetase (gshA), glutathione reductase (gor), thioredoxin (trxA) and thioredoxin reductase (trxB) was assessed. As shown in Fig. 4, the mRNA expression of gshA, gor, trxA and trxB in cells after stimulation by selenite was not different from that before selenite treatment. However, selenite treatment promoted the mRNA expression of frd in a time-dependent manner. These data verified that E. cloacae Z0206 may reduce selenite to Se 0 through fumarate reductase instead of via a GSH-mediated Painter-type reaction.

Effect of buthionine sulfoximine (BSO) on the selenite reduction rate in E. cloacae Z0206.
To confirm that a GSH-mediated Painter-type reaction was not involved in the selenite-reducing process in E. cloacae Z0206, BSO, an inhibitor of glutathione synthetase, was used. As shown in Fig. 5A, BSO doses of 1.5 mM, 3.0 mM and 5.0 mM slightly decreased the growth rate during the exponential phase. In the stationary phase, the cell densities of all BSO-treated groups were slightly lower than that of the control group (all P < 0.05). Therefore, doses  Effect of fumarate on the selenite reduction rate in E. cloacae Z0206. The effect of fumarate on the selenite-reducing ability of the bacterium was evaluated to investigate whether selenite reduction in E. cloacae Z0206 was mediated by fumarate reductase. First, the impact of fumarate on the growth of this strain was studied. As shown in Fig. 6A, different concentrations of fumarate did not affect the growth rate during the exponential phase, whereas a dose of 50 mM led to an evident increase in cell density during the stationary phase. Thus, a dose of 20 mM fumarate was selected to ensure that the cell density was similar to that of control cells without fumarate treatment. Based on the results shown in Fig. 6B, a dose of 20 mM fumarate significantly decreased the reduction rate. After 72 h, 41% and 72% remaining selenite was detected in the cultures without and with fumarate, respectively. These results indicated that competition existed between selenite and fumarate for fumarate reductase 33 , and fumarate reductase may present the main pathway of selenite reduction in E. cloacae Z0206.
Effect of fumarate reductase knockdown on the selenite-reducing ability of E. cloacae Z0206. We constructed the fumarate reductase-mutated strain Δfrd to confirm that enzyme's role in selenite reduction in E. cloacae Z0206. As shown in Fig. 6C, mutation of fumarate reductase did not significantly influence the growth of the cell. However, the mutant strain Δfrd exhibited a markedly repressed selenite-reducing capacity compared with the wild-type strain. After 72 h of reduction, 68.25% of the original selenite remained in cultures of the Δfrd strain, while only 42.58% of selenite could be detected in cultures of the wild-type strain (Fig. 6D). Therefore, E. cloacae Z0206 reduces selenite mainly through fumarate reductase.
In conclusion, many bacterial strains have been demonstrated to reduce selenite to Se 0 and form SeNPs under anaerobic conditions, whereas bacteria grown in aerobic conditions possess the ability to rapidly generate more bacterial cells within a short time period and under less stringent culture conditions. In the present study, E. cloacae Z0206 was found to effectively reduce selenite to Se 0 under aerobic conditions, and form monodispersed nanosized (approximately 100-300 nm in diameter) particles, which were observed both within and outside of the cells. Moreover, the selenite-reducing factor of E. cloacae Z0206 was demonstrated to be a membrane-localized fumarate reductase rather than a GSH-mediated Painter-type reaction. Biosynthesis of SeNPs under aerobic conditions presents advantages over the chemical process, in which SeNPs are produced under environmentally harmful conditions. Thus, E. cloacae Z0206 may be used to develop a bioreactor for the treatment of Se pollution and biosynthesis of SeNPs. Further studies may focus on the properties of biogenic SeNPs, compared with chemically synthesized SeNPs and the potential applications in the fields of nanotechnology and biotechnology.

Methods
Bacterial strain Z0206 and culture conditions. E. cloacae Z0206, a strain that we previously isolated 38   reactions. An overnight culture of E. cloacae Z0206 was adjusted to OD 600 = 1 and diluted into fresh broth (1%). When the culture reached the stationary phase, a 2 ml sample of the culture was removed (as a Se-free control). A 500 μL aliquot of the sodium selenite stock solution (1 M) was then added, and the culture was incubated for another 120 min. Samples were collected at 30 min, 60 min and 120 min after the addition of selenite. *P < 0.05, **P < 0.01 (n = 3).
was supplemented with increasing concentrations of selenite (0 mM, 0.5 mM, 1 mM, 5 mM, 10 mM and 15 mM), and an overnight-grown bacteria culture were adjusted to OD 600 = 0.5 and inoculated (1% inoculum size) to the above mentioned broth containing various concentrations of selenite, followed incubation at 32 °C, 250 rpm for 96 h.
Determination of selenite concentration. The culture was collected and centrifuged at 10,000 × g for 10 min. The supernatant was collected to detect residual selenite through the 2,3-diaminonaphthalene fluorimetric method 41 . Please see section 1.2 in Supplementary Information for introduction of the method to detect selenite. Se particle valence analysis. E. cloacae Z0206 was cultured in the presence of 5 mM selenite at 32 °C at 250 rpm for 72 h. Selenium particles were separated from the culture according to the protocol developed by Dobias, et al. 42 . with some modification. Briefly, the culture was decanted, followed by simple centrifugation at 4 °C at 8,000 ×g for 10 min to separate the suspended biomass. The collected Se particles present in the supernatant from the previous centrifugation step were concentrated via centrifugation at 4 °C at 20,000 ×g for 15 min. The pellet was then lyophilized and analyzed using XPS.

Separation of cellular fractions and determination of selenite-reducing activity. An E. cloacae
Z0206 culture grown to the exponential phase without selenite was collected and centrifuged at 4 °C at 10,000 × g for 10 min. The supernatant was then collected and filtered using a 0.2-µm filter. Next, the pellet was washed with 10 mM Tris-HCl (pH 7.5) twice and re-suspended in the same buffer for sonication, followed by centrifugation at 4 °C at 6,000 × g for 10 min to separate unbroken cells. Then, the supernatant was centrifuged at 4 °C at 25,000 × g for 40 min to separate the cytoplasm (supernatant) and membrane (pellet) fractions. The protein concentration was measured with a BCA assay kit. RT-PCR analysis. RNA was extracted using the RNeasy Protect Bacteria Mini Kit. Total RNA was subjected to a reverse transcription reaction using a Quanti Tect Reverse Transcription Kit. Quantitative PCR was performed with a StepOne Plus ™ Real Time PCR System using a FastStart Universal SYBR Green Master (ROX).

Determination of intracellular GSH concentrations.
Cells, collected from the experiment on the "Effect of BSO on selenite reduction in E. cloacae Z0206" at the time points of 24 h, 48 h and 72 h, were concentrated via centrifugation at 10,000 × g, 4 °C for 10 min and resuspended in 50 mM Tris-HCl (pH 7.5), after which the cells were disrupted using sonication. The disrupted cells were centrifuged at 20,000 × g for 15 min, and the supernatants were collected to detect the concentration of GSH using a Total Glutathione Assay Kit.
Construction of the Δfrd mutant. The Δfrd mutant was constructed as reported elsewhere 43 . Briefly, an frd gene fusion fragment was amplified and ligated via PCR, then ligated with pLP12 and subsequently transformed into Escherichia coli β2163. The resulting plasmids were introduced into E. cloacae Z0206 through conjugation with E. coli β2163. After two rounds of selection, the mutant carrying the frd gene deleted was validated through PCR using primers corresponding to sequences upstream and downstream of the deletion (see Figure S4, in Supplementary Information) and subsequent sequencing.

Statistics.
One-way analysis of variance (ANOVA) followed by an LSD multiple comparison test was used to determine the statistical significance for multiple comparisons, and Student's t-test was used for pairwise comparisons. P < 0.05 was considered statistically significant. All statistical tests were carried out with SPSS 22 software. All data are expressed as the mean ± SD. Data availability. The data that support the findings of this study are available from the authors on reasonable request.