Evaluation of Cytotoxicity of Pb2+ Ion-Adsorbed Amino-Functionalized Magnetic Mesoporous Silica Nanoparticles: An In Vitro Study

Application of amino-functionalized mesoporous silica-coated core-shell magnetic nanoparticles (Fe3O4@mSiO2-NH2 NPs) for adsorbing heavy metal ions has attracted intensive interest in recent years. Despite the cytotoxicity triggered by the co-exposure of nanoparticles (NPs) and metal ions in relatively high dosages being reported, the effect of the adsorbed heavy metal ions on the cytotoxicity to human cells remains unexplored. Herein, we demonstrated the effect of amino-functionalized Fe3O4@mSiO2 core-shell magnetic nanoparticles before and after adsorbing Pb2+ ions on the cytotoxicity of human kidney cells (HEK293). The surface morphology, viability, and oxidative stress (OS) induction of HEK293 cells incubated with Fe3O4@mSiO2-NH2 NPs and Pb2+ ion-adsorbed Fe3O4@mSiO2-NH2 NPs were assessed, respectively. Transmission electron microscopic (TEM) images of cell sections depicted that Fe3O4@mSiO2-NH2 NPs were internalized by HEK293 cells and gathered mainly in the cytoplasm. Cell viability (MTT) assays revealed the Fe3O4@mSiO2-NH2 NPs could enhance the cell viability to 119.9% and 108.2% compared to the control group, respectively. On contrast, the Pb2+ ion-adsorbed Fe3O4@mSiO2-NH2 NPs were toxic to the cell because when the Pb2+ ion contents were 5.0 and 7.5 μg mL−1, the viabilities of the samples decreased to 97.1% and 84.7%, respectively. Oxidative stress data proved that OS was negatively affected by both dissociative Pb2+ ions and the Pb2+ ion-adsorbed Fe3O4@mSiO2-NH2 NPs. Cytotoxicity may be attributed to the OS induced by Pb2+ ions leaked from the adsorbent. Under the same Pb2+ ion concentration, the cytotoxicity of the adsorbed Pb2+ ions was lower than that of the dissociative Pb2+ ions, indicating that the adsorption by NPs inhibited the cytotoxicity of Pb2+ ions. This work will provide new references for assessing the cytotoxicity of Pb2+-adsorbed nanoparticles.


INTRODUCTION
With the improvement of human life quality, the dependence of humans on lead-containing batteries and electronic products is becoming more and more serious, leading to the serious pollution of lead ions in the environment (Pandey et al., 2020). When these products are eliminated, they will be buried as solid waste, which will flow into rivers and lakes with rainfall and pollute water resources (Sharma et al., 2020). With the discharge of industrial wastewater, the permanent toxicity of lead has been a major concern for human beings (Brännvall et al., 1999). Pb 2+ ions cannot be biodegraded in human organisms and will cause numerous diseases and disorders (Inglezakis et al., 2003;Li et al., 2005;Maria das Graças et al., 2006). In addition, Pb 2+ ions are enzyme inhibitors and accumulate in the bones, brain, kidneys, and muscles (Li et al., 2005). Patrick (2006) pointed out that the Pb 2+ ions of concentrations higher than 10 μg dL −1 were considered to be deleterious. Therefore, it is of great significance to develop a new semiconductor material to remove Pb 2+ ions.
Recently, a variety of engineered magnetic nanoparticles (MNPs) have been applied in magnetic separation photocatalysts, magnetic recording media, medical research, and wastewater treatments (Slowing et al., 2008;Facchi et al., 2018;Wang S. F et al., 2021). To avoid the accumulation of magnetic nanoparticles, core-shell and other composite nanostructures have been developed by a special preparation method. (Liu et al., 2020;Tang et al., 2020;Cheng et al., 2022a;Li et al., 2022a;Li et al., 2022b). The mesoporous silica coatings around the MNPs are applied in the synthesis process, which also provide a convenient base to graft-specific ligands. Therefore, the magnetic mesoporous nanoparticles are extensively exploited in heavy metal ions or dye adsorption because of their high specific surface area, ease of modification, and collection (Zhang C et al., 2014;Almasian et al., 2016;Gao et al., 2021). To enhance the adsorption ability of heavy metal ions (usually are cations), amino groups are modified onto the mesoporous silica coating surface via a silanization process (Ma et al., 2019). However, it gradually draws our attention that the heavy metal ion-adsorbed MNPs are still risky to human health because of their extremely small particle size which is capable of entering into human cells (Bundschuh et al., 2018). Researchers suggest that MNPs can affect the morphology of cells, leading to apoptosis and the damage of DNA through oxidative stress (OS) (He and Park, 2016). This process will be affected by a series of factors, such as concentration, particle size, surface functional group, and the degree of dispersion of nanoparticles (NPs) Di et al., 2021;Krętowski et al., 2021;Sun et al., 2021;Cheng et al., 2022b;Liu et al., 2022;Xian et al., 2022). For example, the grafted amino groups could prevent superparamagnetic iron oxide (SPIO) nanoparticles from aggregating in DMEM and promote the cellular uptake efficiency without increasing the cytotoxicity (Zhu et al., 2012).
Despite researchers noticing the unique toxicity of the coexposure of nanoparticles and Pb 2+ ions (Lu et al., 2015;Feng et al., 2018;Ahamed et al., 2019), the cytotoxicity of Pb 2+ ionadsorbed amino-functionalized magnetic mesoporous silica (Fe 3 O 4 @mSiO 2 -NH 2 ) NPs with size above 100 nm has not yet been reported. In this study, we synthesized Fe 3 O 4 @mSiO 2 -NH 2 NPs with good ability for adsorption of Pb 2+ ions (Scheme 1). Then, the Fe 3 O 4 @mSiO 2 -NH 2 NPs with and without the adsorption of Pb 2+ ions were chosen to prepare the culture media, which were used for the incubation of HEK293 cells. To evaluate the cytotoxicity of Pb 2+ ion-adsorbed NPs, the cell viability (MTT) and morphology were investigated. Reactive oxygen species (ROS) and malondialdehyde (MDA) levels, superoxide dismutase enzyme (SOD), and glutathione peroxidase (GSH-Px) activity were measured to explore the inner mechanism of the Pb 2+ ion-adsorbed Fe 3 O 4 @mSiO 2 -NH 2 NP-induced cytotoxicity.

Synthesis of Amino-Functionalized Mesoporous Magnetic Nanoparticles
The OA-grafted Fe 3 O 4 NPs (OA-Fe 3 O 4 ) were synthesized, according to the reference with a minor modification . Typically, 4.60 g of FeCl 3 .6H 2 O and 1.80 g of FeCl 2 .4H 2 O were added to 35 ml of deionized water. Then, 0.80 ml of OA and 20 ml of NH 3 .H 2 O (14 wt%) were added to the solution. The solution was heated up to 95°C for 2 h under a nitrogen atmosphere. The black product was collected with a magnet and then dispersed in CHCl 3 (70 mg mL −1 ).
Then, 0.6 ml of the prepared CHCl 3 solution was added to 5.5 ml of water containing CTAB (20.0 mg mL −1 ). The solution was then heated up to 62°C to completely evaporate the residual CHCl 3 . Then, 25 ml of water and 65 μg of TEA were added to the mixture and was heated up to 85°C for 20 min. Next, 1 ml of TEOs and 3 ml of ethyl acetate were subsequently added and reacted for 2 h before the as-prepared Fe 3 O 4 @SiO 2 NPs were collected by a magnet followed by washing. Then, 80 μL of hydrochloric acid were added to the solution (20 ml) at 80°C and stirred for 9 h to remove CTAB in the mesoporous silica coating. The final products were washed by ethanol to remove the residual reactants and preserved in ethanol for use (150 mg mL −1 ).
In addition, 0.5 ml of APTES was added to 20 ml of the aforementioned ethanol solution containing Fe 3 O 4 @mSiO 2 NPs. The mixture of the solution kept reacting for 4 h at 70°C. The final production of Fe 3 O 4 @mSiO 2 -NH 2 NPs was washed with water before drying up in an oven at 65°C.

Adsorption of Pb 2+ Ions
For the adsorption assay, we prepared an acidified stock solution of lead nitrate (500 mg L −1 of Pb 2+ ions). The Pb 2+ ion concentration was adjusted from 20 to 400 mg L −1 with pH = 6 adjusted by nitric acid (wt 5%) and sodium hydroxide (wt 5%). Then, 20 mg of Fe 3 O 4 @mSiO 2 -NH 2 NPs was added to each 20 ml of Pb 2+ ion solution, which was then shaken at 200 rpm at 25°C for 6 h. The Fe 3 O 4 @mSiO 2 -NH 2 NPs were removed using a magnet and dried at 25°C.

Characterization
The surface morphologies of NPs and cells were recorded by scanning electron microscopy (SEM) (JSM-6510) and transmission electron microscopy (TEM) (JEOL JEM-2100F). The Fourier transform infrared spectrum (FT-IR) was investigated from 500 to 4,000 cm −1 on a Nicolet Impact 410 FTIR spectrometer. The concentrations of Pb 2+ ions were investigated with inductively coupled plasma-optical emission spectroscopy (ICP-OES) (iCAP 7000 Series). Powder X-ray diffraction (XRD) analysis was determined by using a Rigaku D-Max 2550 diffractometer. Nitrogen adsorption-desorption isotherms were obtained in ASAP 2020. The magnetic properties of as-prepared samples were recorded by using a SQUID-VSM magnetometer. A microplate reader (Bio-Rad 550) was used for the MTT assay.

Cell Culture Conditions
HEK293 cells were cultivated with DMEM with 10% FBS and then cultivated at 37°C (of 5% CO 2 ). The medium was refreshed per day, and the HEK293 cells were subcultured at the ratio of 1:2 every 2 or 3 days for use.

Cellular Uptake
Location and qualitative analyses of intracellular nanoparticles were demonstrated by TEM. HEK293 cells were collected and incubated for 12 h. Then, the initial DMEM of the cells was replaced by the sample DMEM, and the sample cells were cultivated for 12 h. HEK293 cells were fixed with glutaraldehyde before being laid up for 4.5 h in darkness and then made into ultrathin sample sections after being dehydrated. The sample sections dyed with uranyl acetate and lead citrate were obtained.
Moreover, 96-well cell culture plates were applied to cultivate cells which were cultivated at 37°C for over 10 h in a humid chamber containing 5% CO 2 , until they adhered onto the plates. Then, the culture medium was replaced by each of the prepared mediums, followed by cultivating the cells for 24 h. Blank medium wells were set as the negative control group. To measure the cell viability, 10 μL of the MTT reagent was added to each well of cell culture plates, and HEK293 cells were cultivated for 1.5 h. The absorbance of cell samples at 450 nm was obtained by using a microplate reader. SCHEME 1 | Diagram of the synthesis procedure of Fe 3 O 4 @mSiO 2 -NH 2 NPs.

Biochemical Parameters for Oxidative Stress
The intracellular oxidative stress (OS) was assessed by the ROS level, MDA level, activity of SOD, and GSH-Px activity. In brief, HEK293 cells were exposed to the DMEM samples and washed with ice-cold PBS, followed by being lysed. The supernatants were obtained after centrifugation for the measurements of oxidative stress. Finally, MDA, SOD, GSH-Px, and SOD were tested, according to the instructions of the kits.
In brief, the fluorescence intensity of dichlorofluorescein (DCF) can reflect the ROS level of cells. Affected by intracellular lipase, intracellular dichlorodihydrofluorescein diacetate (DCFH-DA) will hydrolyze into dichlorofluorescein diacetate (DCFH), which will be further oxidized into DCF by active oxygen in cells. Under certain conditions, MDA will react with thiobarbituric acid (TBARS). The red light adsorbing products will be measured to reflect the content of MDA. GSH-Px activity assay can be evaluated by the content of a yellow compound with maximum absorption at 405 nm, which forms after GSH-Px reacts with dithionitrobenzoic acid (DTNB). The SOD activity assay is based on a formazan dye color reaction. SOD inhibits the formation of the formazan dye by reacting with O 2-, which will oxidize the WST-1 reagent into formazan dye originally.

Statistical Analysis
SPSS 21.0 was used for all the statistical analyses. Data were expressed as mean ± SD from three independent experiments. The t-test was applied for analyzing the statistical differences. The statistical significance was set as p < 0.05.

Microstructure Observation
Fe 3 O 4 NPs were synthesized by a one-pot co-precipitation method, followed by enclosing mesoporous silica shell and grafting amino groups. The Fe 3 O 4 @mSiO 2 -NH 2 NPs were spherical with a silica shell, as shown in Figure 1A. The TEM image ( Figure 1B) depicts that well-dispersed OA-Fe 3 O 4 NP cores with diameters of less than 10 nm are enclosed by a silica shell, forming Fe 3 O 4 @mSiO 2 -NH 2 NPs with a diameter of 125 ± 25 nm. In addition, there exists a mesoporous structure in Fe 3 O 4 @mSiO 2 -NH 2 NPs.

Structure and Magnetic Properties
The chemical compositions of samples were investigated by FT-IR (Figure 2A). The absorption peak at 560 cm −1 in the spectrum was due to the Fe-O stretch, indicating the existence of Fe 3 O 4 particles in both NPs. The peaks at 806 and 1098 cm −1 corresponded to the stretch and vibration of Si−O−Si and Si−O groups in the silica coatings . The characteristic peak at 1626 cm −1 in both spectra was due to the vibration of Si-OH groups or adsorbed water (Yuan et al., 2014;Li et al., 2021). The peaks that appeared at 1487 and 1553 cm −1 for the Fe 3 O 4 @mSiO 2 -NH 2 were ascribed to the stretch and vibration of NH 2 (Yuan et al., 2014), proving the successful graft of amino groups on the silica shell. The peak at 3373 cm −1 was attributed to the vibrations of amine N-H (Yu et al., 2012). The stretching mode that appeared at 3450 cm −1 was attributed to the H 2 O molecule (Wu et al., 2014;Yuan et al., 2014;. Adsorption peaks at 2858 and 2931 cm −1 corresponded to the -CH 2 -group . Figure 2B shows the XRD result of as-prepared samples. Corresponding to the spinel structure of the magnetite phase, the diffraction peaks of OA-Fe 3 O 4 NPs at 30.4°, 35.7°, 43.3°, 53.7°, 57.5°, and 62.9°were related to their indices of (220), (311), (400), (422), (511), and (440), respectively (Enache et al., 2017). There was a wide diffraction band observed at 22-26°ascribing to the amorphous silica coating, which led to the weakness of Fe 3 O 4 diffraction peaks (Yuan et al., 2014;Enache et al., 2017).
The results of magnetic properties for the OA-Fe 3 O 4 NPs and Fe 3 O 4 @mSiO 2 -NH 2 NPs are shown in Figure 2C. The magnetic hysteresis loops indicated that both the NPs were superparamagnetic . The magnetic saturation (M s ) value of Fe 3 O 4 @mSiO 2 -NH 2 was 6.28 emu·g −1 , which was much lower than that of OA-Fe 3 O 4 (52.63 emu·g −1 ). This result can be ascribed to the silica shell outside the OA-Fe 3 O 4 NPs ( Figure 1B). The texture characteristics of Fe 3 O 4 @mSiO 2 NPs and amino-grafted NPs were measured by N 2 adsorptiondesorption. The N 2 adsorption isotherms of the two kinds of NPs suggested that both of them possessed a mesoporous structure ( Figure 2D). The Brunauer-Emmett-Teller (BET) surface area of Fe 3 O 4 @mSiO 2 NPs was 511.95 m 2 /g, which was larger than that of Fe 3 O 4 @mSiO 2 -NH 2 NPs (377.47 m 2 /g). The average pore diameter of the initial NPs was 3.4 nm, which was larger than that of amino-grafted NPs (3.0 nm, Table 1). These results depicted that the amino groups were modified on the silica shell successfully, which agreed well with the result of FT-IR (Figure 2A).

Adsorption Ability
To evaluate the proper proportion of Pb 2+ ions adsorbed on Fe 3 O 4 @mSiO 2 -NH 2 , the adsorption assays were conducted under the condition of 25°C with a pH value of 6. Langmuir and Freundlich models were used to fit the data of the adsorption. The Langmuir model is expressed as Equation 1: where Q e (mg·g −1 ) is the adsorption capacity of Fe 3 O 4 @mSiO 2 -NH 2 NPs at equilibrium; C e (mg·L −1 ) is the equilibrium concentration of Pb 2+ ions; Q m (mg·g −1 ) is the monolayer adsorption capacity of Fe 3 O 4 @mSiO 2 -NH 2 NPs; and K L (L·mg −1 ) is the Langmuir parameter.  Freundlich model is expressed as Equation 2: where Q e (mg·g −1 ) and C e (mg·L −1 ) are the same as that in the Langmuir isotherm model; K F (mg·g −1 ) (L·mg −1 ) 1/n and n are the Freundlich parameters.
To calculate the adsorption constants of Fe 3 O 4 @mSiO 2 -NH 2 NPs, linearized Langmuir and Freundlich plots were fitted, as shown in Figure 3. As can be seen from Table 2, the Langmuir model-fitted equilibrium adsorption data are better. This result proves that the adsorption of Pb 2+ on Fe 3 O 4 @mSiO 2 -NH 2 NPs belongs to the monolayer adsorption process, which is caused by electrostatic adsorption. Furthermore, the interaction between neighboring Pb 2+ ions is weak. The maximum adsorption capacity (Q m ) reaches 123.46 mg g −1 .
In this work, the Fe 3 O 4 @mSiO 2 -NH 2 NPs with Pb 2+ ion weight of 10% were taken as an example to investigate the effect of the adsorbed Pb 2+ ions on the cytotoxicity of Fe 3 O 4 @ mSiO 2 -NH 2 NPs to HEK293 cells.
To observe the changes in cell morphology, images of cell samples were taken by using an optic microscope ( Figure 4B). Compared to the control ( Figure 4B), no obvious change in the morphology of the HEK293 cells cultivated with Pb7.5 was observed. But the cell density slightly decreased, which was consistent with the results of the MTT assay ( Figure 4A). In contrast, there was almost no effect on the morphology and the density of the HEK293 cells when cultivated with NPs75 and NPs75-Pb7.5 ( Figure 4B), respectively, depicting that the adsorption weakened the effects of Pb 2+ ions to HEK293 cells.

Cellular Uptake
The location and internalization of NPs are shown in Figure 5. Clusters composed of accumulated NPs with a diameter of 300-700 nm were observed in the cytoplasm ( Figures 5C,D), suggesting that the NPs could be internalized by HEK293 cells without damaging the morphology of the cells.

Biochemical Parameters for Oxidative Stress
To further demonstrate the difference in cytotoxicity of Pb 2+ ions, Fe 3 O 4 @mSiO 2 -NH 2 NPs, and Pb 2+ ion-adsorbed Fe 3 O 4 @mSiO 2 -NH 2 NPs to HEK293 cells, the oxidative stress (OS) of HEK293 cells was investigated by measuring ROS, MDA, SOD, and GSH-Px of cells. As shown in Figure 6, most samples exhibit a significant difference compared to the control (p < 0.05), in a dose-dependent manner. When cultured with Pb5 and Pb7.5, the ROS ( Figure 6A) and MDA ( Figure 6B) of HEK293 cells increased, while GSH-Px ( Figure 6C) and SOD ( Figure 6D) activities of HEK293 cells decreased. Meanwhile, NPs50 and NPs75 inhibited the oxidative stress as ROS ( Figure 6A) and significantly decreased MDA ( Figure 6B), but the GSH-Px ( Figure 6C) and SOD ( Figure 6D) activities increased. When cultured by the Pb 2+ ion-adsorbed Fe 3 O 4 @mSiO 2 -NH 2 NPs, the OS level of HEK293 cells was lower than that incubated by the dissociative Pb 2+ ions ( Figure 6). For example, the ROS levels of HEK293 cells cultured with NPs50-Pb5 and NPs75-Pb7.5 were 109.5% and 123.8%, respectively, while those of Pb5 and Pb7.5 increased to 119.9% and 129.8%, respectively.

DISCUSSION
Previous work reported that the NPs could be internalized by cells through endocytic processes and gather in endocytic vesicles (Valsami-Jones and Lynch, 2015). Various factors affected the internalization process including particle size (Nel et al., 2006;Kim et al., 2014), shape (Nel et al., 2006), and surface modification (Nel et al., 2006;Foglia et al., 2017). For instance, the spherical nanoparticles were easier to be internalized than the ones in columnar shape (Chithrani et al., 2006). The particles with larger size (e.g., > 200 nm) were hard to enter cells, thus exhibiting less toxicity than that with smaller size (Dong et al., 2017). Pang et al. (2011) found that particles with sizes less than 40 nm could enter the nucleus. In this work, the NPs were successfully internalized into the cytoplasm without damaging the cell membrane because of the small size (125 ± 25 nm) and spherical morphology ( Figures 5C,D), but they were too large to enter the nucleus.
Frontiers in Materials | www.frontiersin.org June 2022 | Volume 9 | Article 914009 one of the main mechanisms to evaluate the cytotoxicity caused by NPs and the co-contaminants, which could be measured by the ROS level, MDA level, GSH-Px activity, and SOD activity (Wu et al., 2016;Zhu et al., 2019;Ahamed et al., 2020). Wang et al. (2020) proved that silica NPs could increase intracellular ROS levels and decrease the activity of the GSH content, which exhibited a dose-dependent cytotoxicity to mammalian cell lines. In our work, Fe 3 O 4 @mSiO 2 -NH 2 NPs exhibited non-cytotoxicity to HEK293 cells, according to the MTT assays ( Figure 4A, NPs50 and NPs75) and the tests of OS parameters (Figure 6), indicating amino groups made a positive impact on the process of cell proliferation (Manke et al., 2013;Gonzalez et al., 2019). It can be explained that amino groups may act as a protective shell to prevent the NPs from contacting organelles, thus inhibiting the reactions between NPs and organelles (Zhu et al., 2012).
NPs50-Pb5 exhibited non-toxicity in MTT assays compared to control, but NPs75-Pb7.5 decreased the cell viability to 84.7% of the control, which was attributed to the lower dosage of Pb 2+ ions in NPs50-Pb5 ( Figures 4A, 6) (Zhang H et al., 2014). However, OS measurements showed that the adsorbed Pb 2+ ions could affect the generation of ROS and inhibit the activity of GSH-Px even at a lower concentration such as NPs50-Pb5 ( Figures 6A-C), suggesting Pb 2+ ion-adsorbed Fe 3 O 4 @mSiO 2 -NH 2 NPs in fact showed a negative effect on HEK293 cells.
Metal ions leaked from NPs are also considered as one of the main reasons causing cytotoxicity since they would react with enzymes and induce OS in cells (Zhang H et al., 2014;Sharma et al., 2015;Premshekharan et al., 2017;Zhang et al., 2018;Gonzalez et al., 2019). Compared to the initial NPs, the viabilities of cells decreased dramatically ( Figure 4A), and the OS level increased significantly (p < 0.05) along with the increase in the dosage of Pb 2+ ion-adsorbed NPs (Figure 6), indicating that the adsorbed Pb 2+ ions might leak from Fe 3 O 4 @mSiO 2 -NH 2 NPs, which reacted with enzymes (e.g., GSH-Px and SOD enzyme) in HEK293 cells, triggering OS in HEK293 cells eventually. As shown in Figure 7, the possible mechanism of the leakage of metal ions from NPs can be explained that the engulfed NPs are stocked in endocytic vesicles and fused with the lysosome. The degradation of enzymes and the acidic atmosphere in the lysosome will lead to the leakage of metal ions , which induces the generation of OH and ROS metabolites. These products will inhibit the activity of antioxidant enzymes and finally lead to OS of HEK293 cells (Liu et al., 2012).
Although the desorption of Pb 2+ ions might occur in HEK293 cells, the viabilities of NPs50-Pb5 and NPs75-Pb7.5 were higher than those of the cells cultured with the dissociative Pb 2+ ions of 5.0 and 7.5 μg mL −1 , respectively ( Figure 4A), indicating the adsorbed Pb 2+ ions were not completely desorbed from the NPs in HEK293 cells. The ROS, MDA, GSH-Px, and SOD tests ( Figure 6) were consistent with the MTT assays except for the groups of Pb7.5 and NPs75-Pb7.5, suggesting lipolysis was not inhibited when cells were incubated with a higher concentration of Pb 2+ ions ( Figure 6B). All the results demonstrated that the adsorption by Fe 3 O 4 @mSiO 2 -NH 2 NPs could prevent Pb 2+ ions from inducing OS in HEK293 cells and subsequently inhibit the cytotoxicity of Pb 2+ ions to HEK293 cells.

CONCLUSION
Our study proved that Fe 3 O 4 @mSiO 2 -NH 2 NPs with a size of 125 ± 25 nm could be internalized by HEK293 cells and gathered together in the cytoplasm. There was no cytotoxicity to HEK293 cells found when cells were cultivated with Fe 3 O 4 @mSiO 2 -NH 2 NPs. On the contrary, they inhibited the oxidative stress of HEK293 cells. But the viability of HEK293 cells decreased significantly when the dosage of Fe 3 O 4 @mSiO 2 -NH 2 NPs was increased. Pb 2+ ion-adsorbed Fe 3 O 4 @mSiO 2 -NH 2 NPs exhibited significant dose-dependent cytotoxicity to HEK293 cells, which was attributed to the intracellular OS process. For instance, ROS of NPs50-Pb5 and NPs75-Pb7.5 reached 109.5% and 123.9% of the control, respectively. Meanwhile, MDA increased with the decrease in GSH-Px and SOD activities, suggesting that desorption of Pb 2+ ions might occur in cells. But the adsorbed Pb 2+ ions on NPs exhibited lower cytotoxicity to HEK293 cells than dissociative Pb 2+ ions. The Fe 3 O 4 @mSiO 2 -NH 2 nanoparticles showed no cytotoxicity to HEK293 cells, and the adsorption of Fe 3 O 4 @mSiO 2 -NH 2 NPs reduced the cytotoxicity of Pb 2+ ions depending on the dosage. Our work provides a better understanding of the cytotoxicity of Pb 2+ -adsorbed nanoparticles and assesses the risk to human beings when using engineered nanoparticles for wastewater treatments.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material; further inquiries can be directed to the corresponding author.