Effects of Modified Magnetite Nanoparticles on Bacterial Cells and Enzyme Reactions

Current paper presents biological effects of magnetite nanoparticles (MNPs). Relations of MNP’ characteristics (zeta-potential and hydrodynamic diameters) with effects on bacteria and their enzymatic reactions were the main focus. Photobacterium phosphoreum and bacterial enzymatic reactions were chosen as bioassays. Three types of MNPs were under study: bare Fe3O4, Fe3O4 modified with 3-aminopropyltriethoxysilane (Fe3O4/APTES), and humic acids (Fe3O4/HA). Effects of the MNPs were studied at a low concentration range (< 2 mg/L) and attributed to availability and oxidative activity of Fe3+, high negative surface charge, and low hydrodynamic diameter of Fe3O4/HA, as well as higher Fe3+ content in suspensions of Fe3O4/HA. Low-concentration suspensions of bare Fe3O4 provided inhibitory effects in both bacterial and enzymatic bioassays, whereas the MNPs with modified surface (Fe3O4/APTES and Fe3O4/HA) did not affect the enzymatic activity. Under oxidative stress (i.e., in the solutions of model oxidizer, 1,4-benzoquinone), MNPs did not reveal antioxidant activity, moreover, Fe3O4/HA demonstrated additional inhibitory activity. The study contributes to the deeper understanding of a role of humic substances and silica in biogeochemical cycling of iron. Bioluminescence assays, cellular and enzymatic, can serve as convenient tools to evaluate bioavailability of Fe3+ in natural dispersions of iron-containing nanoparticles, e.g., magnetite, ferrihydrite, etc.


Introduction
Magnetite nanoparticles (MNPs) have been widely used as adsorbents to extract ecotoxicants from waste and natural waters [1][2][3]. Thus, iron oxide nanomaterials, including magnetite (Fe 3 O 4 ) can be found in engineered and natural aqueous environments due to their numerous applications. Hence, fate of these particles in aquatic systems is an issue of a considerable concern. However, given that bare MNPs are susceptible to air oxidation [4] and are easily aggregated in aqueous systems, the stabilization of the iron oxide particles is mandatory. Functionalization of MNP surfaces is usually Similar to cellular bioassay, the enzymatic assays can assess a general toxicity in test samples. This toxicity type integrates all interactions of toxic compounds with the bioluminescent assay system: redox processes, polar and nonpolar binding, etc. Moreover, the enzymatic bioassay is specific to oxidizers due to its additional kinetic parameter, induction period, which depends on redox potentials of toxicants in solutions [44]. Therefore, the enzymatic bioassay can be additionally applied to monitor a toxicity of oxidative type, which is attributed to the redox properties of the toxic compounds only. Due to such features, the enzymatic bioassay can be used to reveal the ability of MNPs to interfere the redox reactions, as well as stimulate/inhibit biochemical processes as a result of surface hydrophobic/hydrophilic interactions.
Previously [45,46], we used the bioluminescent enzymatic assay system to evaluate the toxicities of general and oxidative types in solutions of organic and inorganic oxidizers (quinones and polyvalent metals, respectively). Changes in general toxicity and oxidative toxicity of aqueous media under exposure to humic substances were studied previously in [47][48][49]. Later [41,[50][51][52], the toxicity and antioxidant activity of a series of fullerenols (i.e., carbon nanostructures, water-soluble derivatives of fullerenes, perspective pharmaceutical agents) were evaluated and compared. Effects of MNPs and MNPs with modified surface on bacterial enzyme reactions have not been studied yet.
Current paper compares the biological effects of MNPs with their zeta potentials and hydrodynamic diameters. Chemical precipitation method in situ and sol-gel synthesis were used to produce bare and modified MNPs. The MNPs were characterized by XRD, SEM, and DLS analysis. Toxicity effects of MNPs were evaluated and compared using a model cellular bioassay-luminous marine bacteria. Ability of MNPs to affect the biochemical processes was studied using a bioluminescence assay based on enzyme reactions of the luminous bacteria. A vital role of a ligand type (HA or APTES) in the effects of MNPs on the enzymatic activity was under consideration.

Preparation of the Fe 3 O 4 MNPs
The bare Fe 3 O 4 MNPs were prepared by the coprecipitation of Fe (II, III) salts in base solution as described in detail in [53]. Briefly, 7.56 g of FeCl 3 ·6H 2 O (Sigma-Aldrich Chemie GmbH, Steinheim, Germany) and 2.78 g of FeCl 2 ·4H 2 O (Sigma-Aldrich Chemie GmbH, Steinheim, Germany) were dissolved in 70 mL H 2 O, then added to 40 mL of 25% solution of ammonium hydroxide (Sigma-Aldrich Chemie GmbH, Steinheim, Germany) at 50 • C under argon flow with vigorous stirring at 600 rpm. The formed Fe 3 O 4 nanoparticles were washed five times with Millipore water and ethanol to eliminate the impurities from synthesis followed by drying at 70 • C in vacuum.

Preparation of the Silica-Coated Fe 3 O 4 /APTES Sample
To prepare the aminosilica-functionalized MNPs the surface modification of Fe 3 O 4 was performed using 3-aminopropyltriethoxysilane (APTES, (Sigma-Aldrich Chemie GmbH, Steinheim, Germany)) as a silylation agent according to [54]. Briefly, 3.21 g of Fe 3 O 4 was dispersed in 150 mL of ethanol/water (volume ratio, 1:1) solution. Then, 13.6 g of APTES was added into the solution under argon atmosphere, at 40 • C for 2 h; molar ratio of APTES to the Fe 3 O 4 was applied as 4:1. After that, the solution was cooled down to a room temperature. The prepared Fe 3 O 4 /APTES MNPs were collected with a magnet (Nd, 0.3 T) and rinsed three times with ethanol and deionized water. Finally, the Fe 3 O 4 /APTES sample was dried in vacuum at 70 • C for 2 h.

Preparation of the Humic Acids-Coated Fe 3 O 4 /HA Sample
The Fe 3 O 4 /HA sample was prepared by in situ method, which consists of the nucleation and growth of magnetite nanoparticles into humic acids matrix. The sample of HA was prepared from sodium humate (Powhumus (Humintech GmbH, Grevenbroich, Germany)). The low H/C atomic ratio (0.85) suggests high content of aromatic structures in the HA preparation. The total acidity of the sample was 5.3 mmol/g of acidic COOH and OH-groups, weight-average molecular weight (Mw) was 9.9 kD. The MNPs modified with HA (as Fe 3 O 4 /HA ratio 80:20 wt %) were prepared by the procedure described in [53,55].

Characterization of the MNPs
The phase composition and particle size of the MPNs samples were determined by X-ray diffraction analysis (XRD) in Bragg-Brentano geometry using a Philips X-Pert Diffractometer (Cr-Kα radiation, λ = 2.29106 Å (Philips Analytical, Eindhoven, The Netherlands)). The full width at a half maximum (FWHM) of the all reflections was used for particle size determination with the Scherrer equation. In order to quantify oxidation progress, the (440) reflection was fitted with five different functions in Origin 2019 Pro (OriginLab Corporation, Northampton, MA, USA). The morphology of the nanomaterials was investigated by scanning electron microscopy (SEM) using a Zeiss LEO SUPRA 25 microscope (ZEISS, Jena, Germany). The average particle diameter of the synthesized MNPs was determined by measuring the diameters of more than 100 particles with ImageJ software ± standard deviation. The data were fitted with Lognormal functions (95% confidence interval) using Origin 2019 Pro (Table S1, Supplementary Materials).
Dynamic light scattering (DLS) measurements were conducted with a NanoZS apparatus (Malvern Panalytical Ltd., Malvern, UK) at a wavelength of 633 nm with a solid-state He-Ne laser at a scattering angle of 173 • at 25 • C. The intensity, volume, and number average diameters of MNP samples were calculated with the Zetasizer Nano 7.11 software (Brookhaven Instruments, Holtsville, NY, USA) utilizing an algorithm based on the Mie theory, which transforms time-varying intensities to particle diameters [56]. For the DLS analysis, each sample was diluted to approximately 0.1 g/L. Since the colloidal stability of the samples varies with the solution conditions, the particles may aggregate; all were measured at a given kinetic state achieved by 10 s of ultrasonication followed by 100 s of standstill. The average values of the hydrodynamic diameter were calculated from 3rd order cumulant fits of the correlation functions. The experiments were carried out in a disposable zeta cell (DTS 1070). The range of pH was between~3 and~10. The values of pH before and after the study were measured, subsequently; the value of pH after measurement by the method of dynamic light scattering was used. The experiments were performed at constant ionic strengths (0.01 M) set by NaCl. Data of zeta potential from pictures (dots) was fitted with Boltzmann functions (lines) and 95% confidence band in Origin 2019 Pro (

Bioluminescence Assay Systems and Experimental Data Processing
Effects of the MNPs on microbiological and biochemical processes were evaluated using model bioluminescence assay systems, both cellular and enzymatic, i.e., luminous marine bacteria and a system of coupled enzymatic reactions of the marine bacteria, respectively.
The bacterial assay, i.e., Microbiosensor 677F, was based on the lyophilized luminous bacteria Photobacterium phosphoreum from the collection of the Institute of Biophysics SB RAS (CCIBSO 863), strain 667F IBSO. The enzyme preparation was based on the system of coupled enzyme reactions catalyzed by NADH:FMN-oxidoreductase from Vibrio fischeri (0.15 a.u.) and luciferase from Photobacterium leiognathi (0.5 mg/mL) [57]. The enzyme reactions (1) and (2) are presented in Introduction. All biological preparations were produced at the Institute of Biophysics SB RAS, Krasnoyarsk, Russia.
Preparation of the MNPs suspensions for the bioluminescence analyses: Solid samples were ground in a mortar and dissolved in 3% aqueous NaCl solution or 0.05 M potassium phosphate buffer for bacterial or enzymatic assays, respectively. To obtain homogeneous suspensions, the 1000 mg/L stock suspensions were exposed to the ultrasound using an Elmasonic EASY 10 bath (Elma Schmidbauer GmbH, Singen, Baden-Wurttemberg, Germany) for 10 min. Analyzed MNPs concentrations (5-10 −13 mg/L) were prepared from the stock suspensions, with additional 5 s ultrasonic treatment at every dissolution stage. Before the bioluminescence measurements, the dispersions were additionally treated in the ultrasonic bath for 5-10 s.
To evaluate the effect of optic filter on the MNPs' suspensions, absorption spectra of the suspensions were recorded using UVIKON 943 Double Beam UV/VIS Spectrophotometer (Kontron Instruments, Milan, Italy). The suspensions of optical density higher than 0.1 (at 490 nm) were excluded, hence, the effect of "optic filter" [58] did not skew the results the bioluminescence measurements.
Measurements of bioluminescence intensity were carried out with bioluminometer Luminoskan Ascent (Thermo Electron Corporation, San Diego, CA, USA).
To assess the toxic/inhibitory effect of MNPs on the bacterial and enzymatic assays, the relative bioluminescence intensity, I rel , was calculated as: where I contr and I MNP are maximal bioluminescence intensities in the absence and presence of the MNPs, respectively. The experimental error for I rel did not exceed 8% in the bacterial assay and 12% in the enzymatic assay. The data for I rel processing were obtained with five samplings from all control and MNP solutions.
The dependence of I rel on MNP concentrations was studied. To characterize toxic effects of the MNPs on the bacteria and enzyme reactions, their effective concentrations that inhibited luminescence intensity by 20% (I rel = 0.8), EC-20, were determined and compared.
Additionally, the effects of the MNPs on the enzyme system were studied under conditions of an oxidative stress. An organic oxidizer, 1,4-benzoquinone (Bq), was chosen for these experiments to model conditions of the oxidative stress. The effective concentration of Bq that inhibited the bioluminescence intensity by 50% (8 × 10 −6 M), was applied in all experiments.
To study and compare the effects of the MNPs under conditions of the oxidative stress, the relative bioluminescence intensity, I rel Bq , was calculated as: where, I MNP+Bq and I Bq are maximal bioluminescence intensities in Bq solutions in the presence and absence of MNPs, respectively (see Figure 1). To characterize the MNP efficiency to suppress the enzymatic activity under the oxidative stress, the effective MNP concentrations that inhibited bioluminescence intensity by 50% (I rel Bq = 0.5), EC-50, were determined and compared. Additionally, the specific effects of MNPs on redox processes in the enzymatic assay system under conditions of the oxidative stress were studied, i.e., the bioluminescence induction periods (T 0.5 ) Bq in Bq solutions were measured ( Figure 1). The T rel Bq values were calculated as: where (T 0.

Microstructure of MNPs
The crystalline structures of the nanoparticles were identified with XRD analysis ( Figure 2). The XRD patterns are similar for all samples and can be interpreted as a face-centered cubic (fcc) lattice with the parameters of 8.383(2), 8.372(1), and 8.382(6) Å for the Fe3O4, Fe3O4/APTES, and Fe3O4/HA samples, respectively (Table 1). The lattice parameters determined for all samples formulated in this study are smaller than those reported for magnetite 8.396-8.400 Å (ICDD-PDF 19-629), but larger than those for maghemite 8.33-8.34 Å (ICDD-PDF 39-1346). A plausible explanation of this phenomenon could be the process of partial oxidation of Fe 2+ during drying and storage resulting in the nonstoichiometric Fe3-δO4 formation where δ can range from zero (stoichiometric magnetite) to 1/3 (completely oxidized) [59]. For magnetite with an ideal Fe 2+ content (assuming the Fe3O4 formula), the mineral phase is known as stoichiometric magnetite (x = 0.50). As magnetite becomes oxidized, the Fe 2+ /Fe 3+ ratio (formula 4) Experimental error for I rel Bq and T rel Bq did not exceed 12%. The data for I rel Bq and T rel Bq processing were obtained in two experiments with five samplings from all control and nanoparticle solutions.
Statistical processing of the biotesting results was carried out; p-values were calculated with RStudio using ANOVA and presented in Supplementary Materials, Tables S2-S5. The p-values were assessed by Student's t-test of two independent sample distributions.

Microstructure of MNPs
The crystalline structures of the nanoparticles were identified with XRD analysis ( (Table 1).

Microstructure of MNPs
The crystalline structures of the nanoparticles were identified with XRD analysis ( Figure 2). The XRD patterns are similar for all samples and can be interpreted as a face-centered cubic (fcc) lattice with the parameters of 8.383(2), 8.372(1), and 8.382(6) Å for the Fe3O4, Fe3O4/APTES, and Fe3O4/HA samples, respectively (Table 1). The lattice parameters determined for all samples formulated in this study are smaller than those reported for magnetite 8.396-8.400 Å (ICDD-PDF 19-629), but larger than those for maghemite 8.33-8.34 Å (ICDD-PDF 39-1346). A plausible explanation of this phenomenon could be the process of partial oxidation of Fe 2+ during drying and storage resulting in the nonstoichiometric Fe3-δO4 formation where δ can range from zero (stoichiometric magnetite) to 1/3 (completely oxidized) [59]. For magnetite with an ideal Fe 2+ content (assuming the Fe3O4 formula), the mineral phase is known as stoichiometric magnetite (x = 0.50). As magnetite becomes oxidized, the Fe 2+ /Fe 3+ ratio (formula 4)  1,3 D XRD and D SEM -average particle size calculated by the Scherrer equation and SEM, respectively, ± standard deviation, nm. 2 CV-coefficient of variation characterizing the polydispersity of the system, %.
The lattice parameters determined for all samples formulated in this study are smaller than those reported for magnetite 8.396-8.400 Å (ICDD-PDF 19-629), but larger than those for maghemite 8.33-8.34 Å (ICDD-PDF 39-1346). A plausible explanation of this phenomenon could be the process of partial oxidation of Fe 2+ during drying and storage resulting in the nonstoichiometric Fe 3−δ O 4 formation where δ can range from zero (stoichiometric magnetite) to 1/3 (completely oxidized) [59]. For magnetite with an ideal Fe 2+ content (assuming the Fe 3 O 4 formula), the mineral phase is known as stoichiometric magnetite (x = 0.50). As magnetite becomes oxidized, the Fe 2+ /Fe 3+ ratio (formula 4) decreases (x < 0.50), with this form denoted as nonstoichiometric or partially oxidized magnetite [59]. The stoichiometry can easily be converted to the following relationship: The coherent-scattering region size was derived from powder XRD data by Scherrer's method. According to the formulas of Scherrer, Wolfe-Bragg, and formula for calculating the unit cell parameter, the value of the unit cell parameter a and the average particle size is strongly influenced by the angle at which the reflexes were detected (Q) and the full width at half maximum of XRD reflexes (FWHM). The original data is a set of points, for which it is rather difficult to determine Q and FWHM exactly. To calculate these parameters more accurately, the baseline points were described using five mathematical parameters using to R 2 and χ 2 as standard statistics parameters for fitting. The full width at half maximum of the reflections was used for particle size determination. The reflections were fitted with five models: Gauss function, Lorenz function, Voigt function, Pseudo-Voigt function, and Pearson VII function in Origin 2019 Pro ( Figure S1, Supplementary Materials). According to R 2 and χ 2 , data for Fe 3 O 4 and Fe 3 O 4 /HA samples were most accurately fitted by the Pseudo-Voigt function (R 2 = 0.9098 (9) and χ 2 = 11.3893 (5), R 2 = 0.9361 (2), and χ 2 = 10.2240 (8), Table S7 in Supplementary Materials), Fe 3 O 4 /APTES-by the Pearson VII function (R 2 = 0.9116 (4) and χ 2 = 49.9908 (7), Table S7 in Supplementary Materials).
Changes in size of nanoparticles upon coating studied by SEM and XRD correlated. Although the spherical particle shape remained constant for all modification routes, a slight particle growth can be observed (Figure 3). For XRD, we used a spherical shape factor (0.94), and as a result, the smaller particle diameters were determined from SEM images. The bare magnetite particles exhibit a median particle diameter of 32.1 and 6.9 nm for SEM and XRD analysis, respectively. Such magnetite particles for the magnetic iron oxide nanoparticles are known to be for superparamagnetic MNPS with a high saturation magnetization and a high specific surface area [60,61].

Characteristics of Surface and Hydrodynamic Size
Characteristics of surface charging and hydrodynamic size of MNPs define their biological activity. The interval of pH 6-7 is particularly important for our study as it provides proper conditions for bioassay procedures described in Section 3.3.
The particle charge measured as zeta potential is presented in Figure 4 for the Fe3O4, Fe3O4/APTES, and Fe3O4/HA nanoparticles. The charge of native magnetite reverses from positive to negative at pH ∼6.3, which may be considered as isoelectric point (IEP) in accordance with literature values [65]. The reactions of surface Fe-OH sites of magnetite, which can lead to formation of positive (Fe-OH2 + ) and negative (Fe-O − ) surface charges, were observed earlier [5]. In the absence of any steric stabilizing layer, the naked Fe3O4, the uncoated sample is only prone to electrostatic stabilization. The zeta potential values were measured between +15 and −30 mV for pH 6-8, and the naked sample showed worse stability than the MNP samples coated with APTES and humic acids (Figure 4). The Fe3O4/APTES sample showed IEP value at pH 7.1, whereas the Fe3O4/HA sample showed IEP value at pH 3. According to [13], APTES has an IEP at pH 10.05. The variations in data for analogous samples in different studies probably indicate the dependence of the zeta potential value on the synthesis approach. The diameter of the Fe 3 O 4 /APTES particles changed to 9.6 and 24.18 nm compared to bare MNPs, whereas the Fe 3 O 4 /HA particles demonstrated a diameter gain to 10.3 and 34.75 nm for SEM and XRD, respectively. The particle growth can be explained by the phase distortion of the original inverse spinel, migration of iron ions to the surface, and subsequent oxidation [62]. Indeed, the XRD patterns confirm preservation of the spinel structure during both modification processes: content of magnetite decreases from~80% to~75% and~60% for Fe 3 O 4 /HA and Fe 3 O 4 /APTES, respectively [63]. According to the coefficient of variation CV and standard deviation value σ of samples for SEM, Fe 3 O 4 /APTES has the smaller size distribution (11.6%, 2.8) than Fe 3 O 4 (13.5%, 4.33) and Fe 3 O 4 /HA (12.45%, 4.33); all samples are considered to be polydisperse [64].

Characteristics of Surface and Hydrodynamic Size
Characteristics of surface charging and hydrodynamic size of MNPs define their biological activity. The interval of pH 6-7 is particularly important for our study as it provides proper conditions for bioassay procedures described in Section 3.3.
The particle charge measured as zeta potential is presented in Figure 4  and negative (Fe-O − ) surface charges, were observed earlier [5]. In the absence of any steric stabilizing layer, the naked Fe 3 O 4 , the uncoated sample is only prone to electrostatic stabilization. The zeta potential values were measured between +15 and −30 mV for pH 6-8, and the naked sample showed worse stability than the MNP samples coated with APTES and humic acids (Figure 4). The Fe 3 O 4 /APTES sample showed IEP value at pH 7.1, whereas the Fe 3 O 4 /HA sample showed IEP value at pH 3. According to [13], APTES has an IEP at pH 10.05. The variations in data for analogous samples in different studies probably indicate the dependence of the zeta potential value on the synthesis approach.
Nanomaterials 2020, 10, x FOR PEER REVIEW 9 of 20 It is known that the presence of amino groups on the surface of Fe3O4 should shift the IEP towards higher pH values [12]. The shift of IEP value of Fe3O4/APTES to pН 7 (Figure 4) was attributed to the presence of protonated amino groups on the SiO2 surface [66]. Additionally, the position of zeta potential curve shows fewer negative charges in the alkaline region on the surface of Fe3O4/APTES, as compared to bare Fe3O4. The reactions of surface Fe3O4-O-Si-NH2 sites with H + and OH − ions lead to the formation of positive (Fe-O-Si-NH3 + ) surface charges. The zeta potential of the amino-functionalized MNPs sample via APTES varied drastically between +20 and -10 mV with pH in the given region. However, despite the low surface zeta potential as 0 to -10 mV at pH 7-8, aggregation of MNPs is negligible according to the magnitude of the hydrodynamic diameter varying from 250 to 300 nm (see in Figure 5). Presumably, the ethyl groups of APTES provide steric stabilization, generating an electrosterically stabilized system that confers proportionally larger stability as compared to the pure electrostatic stabilization by the protonated amino groups attached chemically to MNPs' surface at low pH.  amino-functionalized MNPs sample via APTES varied drastically between +20 and −10 mV with pH in the given region. However, despite the low surface zeta potential as 0 to −10 mV at pH 7-8, aggregation of MNPs is negligible according to the magnitude of the hydrodynamic diameter varying from 250 to 300 nm (see in Figure 5). Presumably, the ethyl groups of APTES provide steric stabilization, generating an electrosterically stabilized system that confers proportionally larger stability as compared to the pure electrostatic stabilization by the protonated amino groups attached chemically to MNPs' surface at low pH. The humic acids-coated MNP S are characterized by a negative zeta potential (between −35 and −40 mV) in the bioassay pH interval of 5.8-7.6 ( Figure 4). For this pH range, the zeta potential value (between −35 and −40 mV) for humic acids-coated MNP S shows the high degree of repulsion between the charged nanoparticles coating by anionic polyelectrolyte humic acids (pK 1 = 4.8 and pK 2 = 10, the total acidity of the sample is 5 mmol/g of acidic COOH and OH-groups with pK 1 = 4.8 and pK 2 = 10, respectively) in the dispersion. Thus, Fe 3 O 4 /HA MNPs with high zeta potential values provide the system's stability in a way of combined steric and electrostatic effects integrated with steric hindrance due to the thick layer of macromolecular humic acids adsorbed on MNPs. A negative charge on the surface of Fe 3 O 4 /HA in the pH range of 3-10 indicates the complete coverage of the surface of magnetic nanoparticles with humic acids. in the given region. However, despite the low surface zeta potential as 0 to -10 mV at pH 7-8, aggregation of MNPs is negligible according to the magnitude of the hydrodynamic diameter varying from 250 to 300 nm (see in Figure 5). Presumably, the ethyl groups of APTES provide steric stabilization, generating an electrosterically stabilized system that confers proportionally larger stability as compared to the pure electrostatic stabilization by the protonated amino groups attached chemically to MNPs' surface at low pH.

Effects of MNPs on Bioluminescence of Cellular and Enzyme Assay Systems
We examined bioeffects of three types of MNPs, i.e., the bare magnetite nanoparticles (Fe 3 O 4 ), and those modified by silica (Fe 3 O 4 /APTES) and humic acids (Fe 3 O 4 /HA). Cellular and enzymatic bioluminescent assay systems (Sections 3.3.1 and 3.3.2, respectively) were used to study the bioeffects. Suppression of the bioluminescence is concerned with inhibition of membrane and intracellular processes in the bacterial cells or with inhibition of chemical and biochemical reactions in the enzyme system. The results might be considered as a model of MNP effects on microorganisms and their enzymatic processes in natural ecosystems. Additionally, we studied MNP effects under conditions of an oxidative stress, i.e., in the presence of a model oxidizer, Bq (Section 3.3.2.2). A role of ligand type (APTES or HA) in the catalytic activity of the bacterial enzymes was one of our main focuses.

Effects of MNPs on Bacterial Cells
Luminescence intensity of bacterial cells was studied in the presence of MNPs; concentrations of bare Fe 3 O 4 , Fe 3 O 4 /APTES, and Fe 3 O 4 /HA varied as shown in Figure 6. The effects for low MNP concentrations (<2 mg/L) were analyzed. The analysis of higher concentrations of MNPs was limited by their solubility in water and the dispersion stability during the time of the experiments. Additionally, we had to avoid the effect of "optic filter" which limits application of the luminescence signal registration in high optical density solutions and/or light scattering suspensions [58].  Figure 6 shows that no significant inhibition of the bacterial bioluminescence intensity was observed for the presented samples of MNPs; EC-50 values were not found experimentally. This indicates that MNPs did not result into a high toxic impact on bacterial cells in the concentration range applied. Therefore, the native limitation of MNP content in the water dispersions provides a low inhibitory effect on the bacterial luminescence, in contrast to fine solutions of compounds with higher solubility studied earlier, i.e., metals, their oxides, and organic chemicals [39,67]. The authors reported EC-50 values as 3.5-21 mg/L for soluble rare earth elements, ≤3 mg/L for dopant metals (Ni 2+ , Fe 3+ ), and no toxicity for metal oxides (EC-50 > 500 mg/L). Figure 6 shows that 20% decrease of the bacterial luminescence intensity was observed for bare Fe 3 O 4 (p < 0.05 at C > 10 −2 mg/L, Table S4). This indicates a moderate toxicity of bare Fe 3 O 4 for the bacteria. This effect can be explained by a release of surface Fe 2+ to media [68,69] with its following oxidation to Fe 3+ in water solutions. The similar oxidative effect of Fe 3+ on luminous bacteria was demonstrated earlier in [48,67].
The absence of inhibitory effect of silica-coated MNPs (Fe 3 O 4 /APTES) is evident from Figure 6 (p > 0.05). This result might be related to the specific characteristics of the silica material used for modification of Fe 3 O 4 surface: higher hydrophobicity and low redox activity (due to low surface charge with zeta potential interval of +15 to −10 mV). Probably, the silica cover can neutralize the toxic effect of Fe 3 O 4 surface by preventing it from dissolution of surface iron due to covalent binding. A confirmation of this hypothesis was presented previously in [69]: Swindle et al. showed that organic coatings of the MNP surface protects the particles from the dissolution.
Humic acids-coated MNPs, Fe 3 O 4 /HA, demonstrated 20% inhibition of the bioluminescence intensity (p < 0.05 at C > 10 −3 mg/L), similar to bare Fe 3 O 4 , Figure 6. Hence, modification of magnetite surface with HA did not protect the bacteria from the toxic effect, in contrast to the silica-modified magnetite. The peculiarity of its effect is a wide concentration interval of 20% inhibition (ca. 5 × 10 −3 -1 mg/L).
Bioeffects of humic substances were observed earlier [47,70]. In [48,49], the inherent inhibitory effect of humic substances on luminous bacteria was found at concentrations higher than 10 mg/L. This value shows that the toxic effect of humic acids-coated MNPs on bacterial cells is not defined by the HA, but by the composite of Fe 3 O 4 /HA.
A membrane activity of Fe 3 O 4 /HA probably can be responsible for its toxic effect on the bacteria: the hydrodynamic diameter of Fe 3 O 4 /HA (100 nm) is smaller than the other systems' diameters (300 nm for Fe 3 O 4 /APTES and 500 nm for the bare Fe 3 O 4 ) providing a higher penetrating ability for the humic acids-coated nanoparticles. Unlike the other systems, the surface charge of the Fe 3 O 4 /HA sample is constantly high and negative (−35 to 40 mV) in the biotesting pH range of 6-7 ( Figure 4). However, remarkable variations in zeta potential are not observed likely due to the lower sensitivity of the electrophoretic light scattering method in comparison with the cellular sensitivity. Sensitive Mossbauer spectroscopy showed minor changes in composition of solid Fe 3 O 4 -HA due to iron ions release [71,72].
As an outline, a moderate toxicity of bare Fe 3 O 4 was found at concentrations < 2 mg/L; it was supposedly attributed to the oxidative effect of the surface and dissolved Fe 3+ in the solutions. High negative surface charge, lower size, and membrane activity, as well as iron ions release, are potential reasons for the moderate effect of  Inhibition of the enzyme bioluminescence system by Fe 3 O 4 was statistically reliable (p < 0.05, Table S2), and the inhibitory effect was not concentration dependent in the used range (p < 0.05; Figure 7). Probably, interactions of the enzymes with the surface of Fe 3 O 4 promote the inhibition of the enzymatic reactions. Additionally, a release of iron ions to water media, discussed before [68,69], can contribute to the specific inhibition of the bioluminescence intensity. Earlier [46,48], the inhibition of the bioluminescence enzymatic system by Fe 3+ was demonstrated.
At low concentrations (<10 −1 mg/L), no inhibition effects were found in the solutions of modified MNPs: Fe 3 O 4 /APTES and Fe 3 O 4 /HA (p > 0.05). Hence, coating the MNP surface with polyelectrolytes APTES and HA prevents from the inhibition of the enzyme reactions, probably, by excluding the enzyme interactions with iron atoms of the MNP surface.
It is evident from the data presented that inhibitory ability of Fe 3 O 4 /HA differed in cellular and enzymatic assay systems: humic acids coating protected from the enzyme inhibition, but does not protect from the cell suppression (compare Figures 6 and 7). This comparison likely supports a suggestion on the membrane-active mechanism of toxicity of Fe 3 O 4 /HA for bacterial cells. On the other hand, the difference in results could be additionally explained with the acidity regimes in the assay solutions: the enzymatic assay was conducted in a phosphate buffer (pH 6.8), while the bacterial assay was carried out in NaCl solution. The pH regimes might be critical for the toxicity of HA-coated MNPs suspension, and they should be further studied.

Effects of MNPs on the Bioluminescence System of Coupled Reactions under Conditions of Oxidative Stress
Since the conditions of water purification with MNPs might vary, the biological activity of MNP surface under the oxidative stress is of a special interest.
The oxidative stress was chemically modeled in the bioluminescence enzymatic assay. To provide this, the bioluminescence kinetics of the enzyme system was studied in solutions of a model organic oxidizer (1,4-benzoquinone, Bq, 8 × 10 −6 M) and MNPs. Concentration of MNPs varied at a low concentration range of 10 −13 -5 mg/L; the effect of "optic filter" was excluded.
The bioluminescence intensity and the induction period were monitored (see Figure 1 in Section 2.3). The relative bioluminescence intensity, I rel Bq , and relative induction period, T rel Bq , were calculated at different concentrations of MNPs and presented in Figures 8 and 9.   Only humic acids-coated MNPs, Fe 3 O 4 /HA (Figure 8), demonstrated low, but reliable deviations from the control samples (T rel Bq < 1) (p < 0.05 at C > 5 × 10 −9 mg/L). This effect was constant in a wide range of low concentrations, similar to this presented earlier in Figures 6 and 7. Hence, the results demonstrate that Fe 3 O 4 /HA additionally increased the oxidative feature of the Bq solutions. This effect might be explained with a partial hydrolyzing of HA coating in the Bq solutions and, as a result, higher content of oxidized iron (Fe 3+ , surface and/or free) in the suspensions of Fe 3 O 4 /HA, as compared to bare Fe 3 O 4 . Peculiarities of these processes are previously discussed in [71][72][73][74][75][76][77][78], as summarized below: Dissolution of the solid phase and an increase in dissolved Fe 2+ and/or Fe 3+ as a result of incubation of MNPs with humic substances was demonstrated by Sundman et al. in [71]. Authors revealed that the magnetite incubated with native humic substances becomes more oxidized as compared with a control magnetite [71]. The reason for iron ions release from Fe 3 O 4 /HA matrix is the supramolecular nature of HA that means associations via weak hydrophobic (van der Waals, π-π, CH-π) and hydrogen bonds [73]. As a result of different treatments (including an oxidative stress), HA can be hydrolyzed, thus leading to the destruction of the protective layer of Fe 3 O 4 /HA. Derivatives of HA contribute into the iron ions release from magnetite. Both iron ions, Fe 2+ and Fe 3+ , form strong mixed ligand complexes with HA [72,74] and/or enzymes [75]. Therefore, thermodynamically driven dissolution and subsequent complexation reactions between HA, enzymes, and iron ions can be an important reason for Fe 3 O 4 /HA dissolution. The dissolution of Fe 3 O 4 /HA was also supported by hydrodynamic particle size analysis via DLS showing an increase in the particle size [76,77].
The studies of the Fe 3 O 4 /HA bioeffects provides an understanding of a role of humic substances in the biogeochemical cycles of iron (Fe 2+ /Fe 3+ ). These are naturally important processes, since humic substances are main components of dissolved organic matter and, therefore, play an important role in a complex formation of colloidal magnetite particles, redox activity of iron in these particles, and their effect on microbiota. In nature, humic substances are known to be sorbed onto MNPs, change MNP' surface charges, affect the aggregation and bioavailability [79], interfere with mineral dissolution/precipitation reactions [80], and drive redox reactions [81].
In contrast to Fe 3 O 4 /HA, water solutions of HA did not show any low-concentration inhibitory effects under the conditions of the oxidative stress, but opposite, revealed an antioxidant activity and mitigation of the oxidative effect of Bq (i.e., T rel Bq > 1) [49,52,58]. Hence, humic acids-coated MNPs, Fe 3 O 4 /HA, show higher oxidative toxicity than its components Fe 3 O 4 or HA under the conditions of the oxidative stress. Figure 9 presents the I rel Bq values at different concentrations of the MNPs. This parameter (Equation (2)) analyses the bioluminescence intensity, I, of the enzymatic system (Figure 1 in Section 3.2); it depends on the combination of physicochemical mechanisms (redox, polar, and apolar interactions) under conditions of the model oxidative stress.
The I rel Bq values demonstrated the inhibition of the bioluminescence intensity at higher concentrations of all MNPs (p < 0.05; concentration intervals for statistical confidence are presented in Table S3; Figure 9). All MNPs did not show a reliable activation of the bioluminescence enzyme system (i.e., no mitigation of the oxidative stress) at low concentrations; even Fe 3 O 4 /APTES showed the p > 0.05 at 10 −7 -10 −1 mg/L.
No statistical difference was found between the effects of bare Fe 3 O 4 and Fe 3 O 4 /APTES under the conditions of the oxidative stress (p > 0.05).
The parameters of I rel Bq , Figure 9, allowed us to determine the effective concentrations of EC-50 using the experimental data, in contrast to the bioluminescence parameters analyzed before (I rel in the bacterial assay, I rel in the enzymatic assay, and T rel Bq in Figure 5, Figure 7, and Figure 8, respectively). So, I rel Bq appeared to be most sensitive bioluminescence parameter to monitor the effects of MNPs. The EC-50 were determined as 1, 1, and 5 × 10 −2 mg/L for the Fe 3 O 4 , Fe 3 O 4 /APTES, and Fe 3 O 4 /HA, respectively. It is seen that humic acids-coated MNPs, Fe 3 O 4 /HA, showed maximal inhibitory activity under the oxidative stress. Additionally, in a wide interval of low concentrations (5 × 10 −6 -5 × 10 −1 mg/L), the Fe 3 O 4 /HA demonstrated higher ability to inhibit the bioluminescence intensity (Figure 9), similar to the other kinetic bioluminescence parameter of the enzyme system, T rel Bq (Figure 8), discussed before. Comparison of Figures 8 and 9 reveals that the parameter I rel Bq is more sensitive to MNPs than T rel Bq . Since the T rel Bq is attributed to redox characteristics of solutions only [44,45,82], the sensitivity of the I rel Bq parameter might be explained by contribution of hydrophobic interactions to the enzymatic activity under the conditions of oxidative stress.
As an outline for Section 3.3.2.2: under the conditions of the model oxidative stress (i.e., in the presence of the model organic oxidizer, Bq), all MNPs, bare and modified, did not reveal an antioxidant activity. Moreover, humic acids-coated MNPs demonstrated the additional inhibitory effect on the enzyme reactions in a wide concentration range (Figures 8 and 9), which can be explained with HA hydroxylation, a destruction of the protective layer, and a higher content of iron Fe 3+ in the samples of Fe 3 O 4 /HA, as compared to bare Fe 3 O 4 [78].

Conclusions
We studied effects of MNPs on luminous marine bacterium and its enzymatic reactions. Additionally, the enzymatic assay was applied to study inhibitory ability of the nanoparticles under conditions of a model oxidative stress (i.e., in the solutions of model oxidizer, 1,4-benzoquinone). We demonstrated that the oxidative stress increases a sensitivity of the enzymatic assay system to MNPs: the bioluminescence intensity determined under the conditions of an oxidative stress, I rel Bq , was found to be most sensitive parameter to monitor the inhibitory effects of MNPs. Humic acids-coated MNPs (Fe 3 O 4 /HA) showed maximal inhibitory effect on the enzyme reactions, probably due to the partial dissociation of HA coating in the oxidant' solutions and, as a result, higher content of oxidized iron Fe 3+ in the suspensions.
Silica surface modification made the magnetite nanoparticles (Fe 3 O 4 /APTES) inert for both bacteria and their enzymes (Figures 6 and 7). Since silica in free and combined forms is a dominant component of many solid soils and soil solutions/dispersions, our results elucidate the biological function of silica in the biogeochemical cycling of iron.
Our results show that the bioluminescence assays, cellular and enzymatic, can serve as promising and highly convenient tools to evaluate and compare the bioavailability of Fe 3+ ions in natural dispersions of iron-containing nanoparticles-magnetite, ferrihydrite, so on. The following studies should reveal the quantitative dependences of the bioluminescence response vs. a release of Fe 3+ from these MNPs. We have already previously detected the significant iron dissolution in the presence of Fe 3 O 4 coated by other ligands [83].
Author Contributions: Investigation, software, formal analysis, L.S.B.; experimental studies using bioluminescence methods, data processing, interpretation, writing and editing the manuscript, E.S.K.; experimental studies using bioluminescence methods, A.G.K.; resources, investigation of samples microstructure using SEM, writing-review and editing, XRD, G.I.D.; resources, data curation, methodology, E.I. and E.T.; conceptualization, general leadership of the work, data analysis, writing-original draft preparation, writing-review and editing K.A.K. and N.S.K. All authors have read and agreed to the published version of the manuscript. Acknowledgments: G.I.D. performed this study in accordance with the state tasks, state registration AAAA-A19-119032690060-9.

Conflicts of Interest:
The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.