All chemical reagents for synthetic materials were obtained from Sinopharm Chemical Reagent Co. Reagents used in cell culture such as phosphate-buffered saline (PBS), Dulbecco’s modified Eagle’s medium (DMEM), fetal bovine serum (FBS), penicillin and streptomycin were provided by Gibco, Invitrogen. Cell count kit-8 (cck-8) and Fluo-4AM were purchased from Beyotime Biotechnology. 2'7'-dichlorofluorescein diacetate (DCFH-DA), Mitosox red, and 4', 6-diamidine-2-phenylindole dihydrochloride (DAPI) were provided by Sigma–Aldrich. Antibodies of Caspase-9/Caspase-3 were obtained from a protein technology company. The secondary antibody used Alexa Fluor 488 conjugated goat anti-mouse and Cy3 conjugated goat anti-rabbit, which were purchased from Servicebio. Calcein-AM/PI and Annexin-V/PI double staining kits were purchased from Dojindo laboratories.

Cell viability was measured by cck-8. HBE cells and BEAS-2B cells were seeded in 96-well plates overnight and Fe₃O₄/GO in DMEM was added to each well at a dose of 0, 10, 20, 50, 100, and 200 μg/ml for 3 replicates per group. HBE cells were tested after 6, 12, 24, and 48 h of co-incubation and BEAS-2B cells were tested after 12, 24 h of co-incubation later according to the manufacturer’s instructions.

Oxidative stress changes are represented by reactive oxygen species (ROS). The fluorescence intensity of DCFH-DA is the most commonly used method to detect intracellular ROS levels. HBE cells were seeded in confocal dishes overnight. Fe₃O₄/GO (0, 100, 200 μg/ml) in DMEM was added and then coincubated after 24 h. Then, it was cleaned with PBS three times, then prepared DCFH-DA staining solution was added and incubated for 15 min. It was cleaned with PBS three times, observed, and analyzed by using a confocal microscope (LSM 900, ZEISS, Germany) (Ex: 505 nm Em: 525 nm).

Mitosox red is a mitochondrial superoxide indicator. HBE cells were seeded in confocal dishes overnight. Fe₃O₄/GO (0, 100, 200 μg/ml) in DMEM was added and then coincubated after 24 h. Then, it was cleaned with PBS three times, prepared Mitosox red and DAPI staining solution was added and incubated for 15 min. It was cleaned with PBS three times, observed, and analyzed via a confocal microscope (LSM 900, ZEISS, Germany) (Mitosox red, Ex: 510 nm Em: 580 nm; DAPI, Ex: 350 nm Em: 461 nm).

Ca²⁺ levels were detected by Fluo-4AM. HBE cells were seeded in confocal dishes overnight. Fe₃O₄/GO (0, 100, 200 μg/ml) was added and then co-incubated for 24 h. Then, it was cleaned with PBS three times and then prepared Fluo-4AM staining solution was added, and incubated for 15 min. It was cleaned with PBS three times, observed, and analyzed under a confocal microscope (LSM 900, ZEISS, Germany) (Ex: 494 nm Em: 516 nm).

Dead/live cells were analyzed by using a calcein-AM/PI double staining kit. HBE cells were seeded in confocal dishes overnight. Fe₃O₄/GO (0, 100, 200 μg/ml) in DMEM was added and then coincubated for 24 h. Then, it was cleaned with PBS three times and prepared calcein-AM/PI staining solution was added and incubated for 15 min. It was cleaned with PBS three times, observed, and analyzed under a confocal microscope (LSM 900, ZEISS, Germany) (Calcein-AM, Ex: 490 nm Em: 515 nm; PI, Ex: 530 nm Em: 617 nm).

HBE cells were seeded in confocal dishes overnight. Fe₃O₄/GO (0, 100, 200 μg/ml) in DMEM was added and then coincubated for 24 h. Then, cells were stained with Annexin-V FITC/PI and analyzed by confocal microscope. (LSM 900, ZEISS, Germany) (Annexin-V FITC, Ex: 488 nm Em: 515 nm; PI, Ex: 530 nm Em: 617 nm).

HBE cells were seeded in confocal dishes overnight. Fe₃O₄/GO (200 μg/ml) in DMEM was added and then co-incubated for 24 h. Then, cells were immunofluorescence stained by Caspase-3 antibody and Caspase-9 antibody and observed under a confocal microscope (LSM 900, ZEISS, Germany).

It could be found that GO showed transparent sheet-like gauze with folds at the edge of the sheet. The research of Ajayan et al. suggested that folding was majorly attributable to the destruction of the C=C double bond caused by Sp2 hybrid oxygen-containing functional groups on the graphite oxide (Datta et al., 2018; Sadighian et al., 2021). Fe₃O₄ particles prepared by a coprecipitation method have a diameter of approximately 10 nm. However, due to the interaction between coulomb force and van der Waals force among the nanoparticles, a few of the nanoparticles exhibit the agglomeration phenomenon (Dyer et al., 2015; Fu and Li, 2014). Therefore, some Fe₃O₄ nanoparticles in the composite have a particle size of 30–50 nm after agglomeration in Figure 1 (Narayanaswamy and Srivastava, 2017). However, the overall dispersion is favorable.

As displayed in Figure 2, Fourier transform infrared spectroscopy of Fe₃O₄/GO was analyzed. The absorption peak at 3,434 cm⁻¹ was attributed to the stretching vibration of OH from GO, and the absorption peak near 2,926 cm⁻¹ was attributed to the stretching vibration of CH₂ from synthetic Fe₃O₄/GO (Hanh et al., 2018). The absorption peak at 1,624 cm⁻¹ was accounted for by the C=O stretching vibration at the GO edge (Narayanaswamy and Srivastava, 2017). The absorption peak at 1,379 cm⁻¹ was caused by the C-O-C stretching vibration on the GO surface (Seyyed et al., 2021). The absorption peak at 580 cm⁻¹ was induced by the stretching vibration of Fe-O-Fe (Zhang et al., 2017). In summary, it was indicated that Fe₃O₄/GO nanoparticle complex with high purity was prepared.

The hydration particle size of Fe₃O₄/GO was analyzed in deionized (DI) water and DMEM, respectively by DLS. As shown in Figure 3A, Fe₃O₄/GO was 1,325 nm in dH₂O. There is a certain aggregation in DI water, so the measured particle size is larger. Due to the presence of serum in the medium, the serum protein such as albumin could form a corona which stabilization of the materials in the suspension (Wang et al., 2016). Thus, Fe₃O₄/GO had better dispersion in the medium with a particle size of 1,164 nm in Figure 3B. The Zeta potential of materials was in the range of 3.96 mV and −13.6 mV in DI water and DMEM, respectively, which changes the positive charge in DI to negative charge in DMEM (Figure 4).

It was evaluated at different concentration and different times of Fe₃O₄/GO cytotoxicity via cck-8 assay, respectively. HBE cells were cultured with different concentrations (0, 10, 20, 50, 100, and 200 μg/ml) of Fe₃O₄/GO, and cell viability was detected after 6, 12, 24, and 48 h by cck-8 assay. As shown in Figure 5A, the cell survival rate still reached more than 60% after 6 h Fe₃O₄/GO exposure at 200 μg/ml. In short periods, even high concentrations of Fe₃O₄/GO can have a rational biosafety profile. After 12 h, we can significantly infer that cell viability was decreased to 47.55% after the highest concentration of Fe₃O₄/GO exposure shown in Figure 5B. Compared with Figure 5A, the results indicated that cytotoxicity of Fe₃O₄/GO was time-dependent. In Figures, 5C,D, HBE cell viability was only around 40% after 24 and 48 h Fe₃O₄/GO stimulation at the concentration of 200 μg/ml. It is worth noting that cell viability was less than 60% after 48 h, even at low concentrations (20 μg/ml). However, other reports indicated that these nanoparticles had the potential to produce toxic effects in cells, and their toxic effects were related to their size, concentration, time, shape, and the cell type (Wang et al., 2020; Zhang S. et al., 2020). Thus, we also verified the cytotoxicity effects of Fe₃O₄/GO in BEAS-2B cells, which also belong to human lung epithelial cells. As shown in Supplementary Figure S1, with an increasing concentration of Fe₃O₄/GO, the BEAS-2B cell’s cytotoxicity was Fe₃O₄/GO nanosheet concentration-dependent and time-dependent after 12 and 24 h co-incubation. This study showed that the cytotoxicity effects of Fe₃O₄/GO on cells was time-dependent and concentration-dependent.

The imbalance between oxidants and antioxidants favors oxidants and may culminate in so-called “oxidative stress” (Salvador et al., 2021). As a product of oxidative stress reaction, ROS produced by the interaction between nanomaterials and cells has been reported as one of the pivotal causes of cell damage. It has been previously reported by several researchers that ROS [such as superoxide anions (O₂•⁻), hydroxyl radicals (HO•), and hydrogen peroxide] levels could enhance in human cells after exposure to Fe₃O₄/GO nanosheets. Literature also unravels that iron oxide nanoparticles can induce cytotoxicity by activating oxidative stress responses (Ahamed et al., 2020). To further detect whether Fe₃O₄/GO induces ROS production in HBE cells, DCFH staining was used. Oxidative stress levels were expressed as ROS levels and detected by DCFH fluorescent probe. As shown in Figure 6A, it can be found that the green fluorescence (DCFH fluorescent probe) in the cells increased significantly in response to Fe₃O₄/GO concentration after 24 h of co-incubation, indicating that the oxidative stress level of the cells increases significantly after Fe₃O₄/GO stimulation. In addition, our quantitative data (Figure 6B) showed that ROS levels in the Fe₃O₄/GO group were significantly higher than in other control group in a concentration-dependent manner. These results suggested that Fe₃O₄/GO induced ROS production and resulted in oxidative stress in HBE cells.

Fluo-4AM was used to detect the calcium ion levels in cells by means of confocal observation. As shown in Figures 6C, D, fluorescence intensity enhanced significantly with the increase of concentration after coincubation with Fe₃O₄/GO for 24 h. Most studies elucidated that ROS in the mitochondrial respiratory chain could elevate Ca²⁺ levels (Zhang et al., 2022). Therefore, intracellular free Ca²⁺ may be implicated in the mechanisms of apoptosis (Zhang et al., 2015). In this study, it was confirmed that Fe₃O₄/GO induced ROS generation and Ca²⁺ influx in HBE cells. Considering that calcium signaling is a crucial manipulator of cell function, ER is the dominant source of intracellular calcium and assumes an essential role in the process of cell apoptosis. Following Fe₃O₄/GO exposure, intracellular calcium levels were increased by releasing Ca²⁺ of ER. The disruption of intracellular calcium homeostasis contributes to calcium metabolism disorders and impairs protein folding. The long-term accumulation of misfolded proteins in ER results in ER stress-mediated apoptosis. Thus, we can infer that Fe₃O₄/GO may lead to HBE apoptosis via ROS-induced Ca²⁺ influx, which was verified in subsequent experiments.

Mitochondrial dysfunction was verified by Mitosox red, which is a mitochondrial superoxide indicator (Wang et al., 2013). Mitochondria are stimulated to produce mitochondrial superoxide, which leads to impaired mitochondrial function. As shown in Figure 7, after Fe₃O₄/GO exposure, red fluorescent was enhanced with concentration increased, which indicated mitochondrial generated superoxide. We can infer that ER-Ca²⁺ may induce mitochondrial dysfunction, after Fe₃O₄/GO exposure.

The state of dead and alive cells was verified by calcein-AM/PI staining, where calcein-AM represented living cells (green fluorescent) and PI (red fluorescent) represented dead cells (Zheng et al., 2020). Figure 8 showed that after 24 h coincubation with Fe₃O₄/GO, part of the HBE cells died and the amount of dead cells response to increased Fe₃O₄/GO concentration.

Apoptosis detection was performed to verify the death mode of cells after incubation with Fe₃O₄/GO, which was characterized via Annexin-V/PI in Figure 9. After Fe₃O₄/GO exposure, fluorescence of Annexin-V and PI were obviously increased response to Fe₃O₄/GO nanoparticle concentration, which were concordant with the results of the cck-8 assay. It indicated high concentration Fe₃O₄/GO can lead to HBE cell apoptotic (Chen M. et al., 2021).

So far, we have confirmed that Fe₃O₄/GO can stimulate HBE cells to produce oxidative stress, calcium influx, and mitochondrial superoxide, ultimately leading to apoptosis, and these phenomena are concentration-dependent. In order to explore the specific pathway through which Fe₃O₄/GO stimulates HBE cell genesis, we characterized the expression levels of Caspase-9 and Caspase-3.

After coincubation with Fe₃O₄/GO, the fluorescence of Caspase-9 and Caspase-3 was significantly enhanced in Figures 10, 11, confirming that both Caspase-9 and Caspase-3 were activated. All these results indicated that Ca²⁺-ER stress led to mitochondrial dysfunction, which promoted the activation of Caspase-9 and activation of Caspase-3, resulted in cell apoptosis (Chen et al., 2005).

ROS are produced in cells via a variety of mechanisms. High intracellular ROS increased Ca²⁺ levels, which can trigger a series of mitochondrial related events, including endoplasmic reticulum stress, and mitochondrial dysfunction, then activated Caspase-9/Caspase-3 relate apoptosis, these proteins are important in apoptotic pathway (Figure 12) (Hailan et al., 2022). Our study demonstrated that after HBE cells co-incubation with Fe₃O₄/GO, ROS levels increased and Ca²⁺ levels enhanced, lead to mitochondrial dysfunction, and then, resulting in Caspase-9/Caspase-3 related apoptotic of HBE cells.