Size-Dependent Cytotoxicity and Reactive Oxygen Species of Cerium Oxide Nanoparticles in Human Retinal Pigment Epithelia Cells

Introduction

Nanomaterials have unique properties, such as small size and enlarged surface area, that enhance regenerative and catalytic enzyme activities and, consequently, their biological effects. Cerium (Ce), a critical rare earth element with a unique f-electron configuration that gives its compounds special properties, has been called a universal new material.¹ In cerium oxide nanoparticles (CeO₂ NPs), there are two valence states-Ce³⁺ (reduced state) and Ce⁴⁺ (oxidation state)—and these two states can be converted to each other. The transition between Ce⁴⁺/Ce³⁺ on the crystal surface results in catalytic and antioxidant effects.^1–3 Pezzini et al reported that CeO₂ NPs serve as antioxidants in primary cultured skin fibroblasts.⁴ As free radical scavengers, CeO₂ NPs can treat various diseases induced by oxidative stress.⁵ Recent studies have found that CeO₂ NPs act as ROS scavengers in diabetic nephropathy,⁶ rheumatoid arthritis,⁷ and ischemic stroke.⁸ Furthermore, CeO₂ NPs are widely used in single-phase or multiphase drug carriers or delivery devices to solve cancer drug resistance and mistargeting, and to achieve a synergistic anti-tumor activity with drugs.^9–11 In addition, some studies have reported that CeO₂ NPs are used in the treatment of eye diseases, such as to reduce light-induced retinal degeneration¹² and photoreceptor death rate.¹³ Moreover, CeO₂ NPs have been reported to act as antioxidants in the retina and protect against retinal nerve damage induced by high-intensity light exposure.¹⁴

Therefore, the widespread use of CeO₂ NPs has raised human health concerns. It has been suggested that CeO₂ NPs lead to ROS generation, DNA damage, and apoptosis in human lung cells.^15,16 Moreover, CeO₂ NPs induce cytotoxicity of the human hepatoma cell line SMMC-7721 through oxidative stress and activation of the MAPK signaling pathway.¹⁷ In addition, CeO₂ NPs induce cytotoxicity and oxidative stress in human skin keratinocytes¹⁸ and genotoxicity in human intestinal Caco-2 cells.¹⁹ However, despite being an important and sensitive organ, the eyes have been ignored in evaluating the toxicity of CeO₂ NPs; whether or not CeO₂ NPs exert toxicity to other organs is largely unknown.

ROS are natural byproducts of normal oxidative metabolism and of free radicals such as the highly reactive hydroxyl radical (·OH) or superoxide anion radical (O₂·^–).²⁰ ROS are unstable and highly reactive compounds that can strip electrons from nearby molecules and induce significant oxidative damage to cellular structures if the amount of ROS exceeds the system’s antioxidant capacity.^21,22 The cytotoxicity effect is referred to as “oxidative stress,” which leads to a change in the mitochondrial membrane potential.²³ In the whole life cycle of cells, mitochondria use oxidable substrates to produce an electrochemical proton gradient on the mitochondrial membrane, which is used to produce ATP and generate energy for cellular activities.²⁴ The evaluation of the mitochondrial membrane potential (Δ_{Ψ m}) of intact cells can provide the necessary information to assess their physiological and pathological status.^25,26

This study evaluated the toxicity of CeO₂ NPs with different particle sizes in ARPE-19 cells, which are a type of human retinal pigment epithelial cell. We also determined the role of CeO₂ NPs in ROS generation. In addition, we explored the change in mitochondrial membrane potential in response to CeO₂ NPs treatment.

Materials and Methods

Chemical and Reagents

CeO₂ NPs were purchased from Shanghai Xiangtian Nanomaterials Co., Ltd. (Shanghai, China); fetal bovine serum (FBS), DME/F−12 medium, and penicillin/streptomycin were obtained from Life Technologies (Carlsbad, CA, USA); N-acetylcysteine (NAC), and 2′,7′-Dichlorofluorescin diacetate (H₂DCF-DA) were purchased from Sigma-Aldrich (St. Louis, MO, USA); and a mitochondrial membrane potential assay kit with JC-1 was purchased from Beyotime Biotechnology Co., Ltd. (Shanghai, China).

Characterization of Different Sizes of CeO₂ NPs

The size and morphology of CeO₂ NPs were examined using a transmission electron microscope (TEM). Briefly, a drop of CeO₂ NPs suspension at 50 µg/mL was tested under a TEM (200 kV, Tecnai F20, Philips, The Netherlands).

Cell Culture

The human retinal pigment epithelial cell line (ARPE-19 cells) was from the Fu Heng Cell Center (Shanghai, China). It was cultured with 10% FBS, penicillin (50 U/mL), and streptomycin (50 U/mL at 37°C in a humidified atmosphere with 5% CO₂. Next, CeO₂ NPs were dispersed in ultrapure water to prepare stock solutions (200 mg/mL). The stock solution was sonicated using a probe sonicator (Ningbo Xinzhi Biotechnology Co., Ltd., China) at 600 W for 40 min and diluted to different concentrations with culture medium and penicillin/streptomycin just before cell exposure. The cells were adjusted to a concentration of 1×10⁵ cells/mL in a volume of 100 μL per well in 96-well plates for toxicity assays.

Cell Morphology

ARPE-19 cells were collected and seeded into 96-well plates at a density of 1×10⁴ cells/well and cultured overnight in a CO₂ incubator. Cells were exposed to CeO₂ NPs at different concentrations (1–100 µg/mL) for 24 and 48 h. The cell morphological changes were examined using a phase-contrast microscope (Leica DM16000B, Germany).

Cell Viability Assay

The Cell Titer 96 Aqueous One Solution Cell Proliferation Assay (Promega Corporation, Madison, WI, USA) was employed to examine cell viability. Following 24 or 48 h of exposure to CeO₂ NPs at various concentrations, 10 μL of cell titer agents was added to each well. The 96-well plate was incubated in an incubator for 2 h. Absorbance was measured at 490 nm with the Synergy H4 Hybrid microplate reader (Bio Tek Instruments, Inc., Winowinsky, VT, USA).

Measurement of Cellular ATP Levels and Lactate Dehydrogenase (LDH) Release

Cell Titer-Glo^® Luminescent Cell Viability Assay (Promega, Madison, WI, USA) was used to assess the ATP levels in CeO₂ NPs-treated ARPE-19 cells according to the manufacturer’s instructions. Luminescence was recorded using a Synergy H4 Hybrid microplate reader.

The cytotoxicity of CeO₂ NPs was examined using the LDH Release Assay (Beyotime, Beijing, China) as described previously.²⁷

Measurement of Intracellular ROS

Intracellular ROS concentration was measured using the fluorescent dye H₂DCF-DA.²⁷ Briefly, ARPE-19 cells were treated with 10 μM H₂DCF-DA for 30 min in the cell culture incubator. The cells were washed twice with PBS and then treated with 3.125–100 μg/mL CeO₂ NPs in phenol-red-free medium. The cells were continuously incubated, and the fluorescence intensities were measured at 6, 12, 24, and 48 h time points with a Synergy H4 Hybrid microplate reader. Meanwhile, the oxidation of H₂DCF-DA was detected using a confocal laser scanning microscope (Leica TCS SP5, Germany) at the 24- and 48-h time points.

Detection of Mitochondrial Membrane Potential

JC-1 Staining Kit (Beyotime, Beijing, China) was used to assess mitochondrial membrane potential changes. Cells were seeded on dishes at a density of 1×10⁵ cell/mL and stored overnight. The cells were treated with different concentrations of CeO₂ NPs for 24 and 48 h. At the end of treatment, cells were removed from the medium, washed three times with PBS, and then incubated with JC-1 staining kit (20 μM) for 15 min. The JC-1 staining solution was removed, and the cells were washed three times. PBS was added for imaging by confocal laser scanning microscopy (Leica TCS SP5, Germany).

Statistical Analysis

Results were presented as the mean ± standard deviation (SD). Analyses were performed using Graph Pad Prism 6 (Graph Pad Software; La Jolla, CA, USA). Statistical significance was determined by one-way analysis of variance (ANOVA) followed by the Dunnett’s tests for comparisons between different concentrations to vehicle control or two-way ANOVA followed by the Sidak’s multiple comparisons test for comparisons of two treatment groups in NAC pretreatment experiments. The differences were considered statistically significant when the p value was <0.05.

Results

Characterization of CeO₂ NPs

The TEM images of the different sizes of CeO₂ NPs are presented in Figure 1. The images showed that the average sizes of CeO₂ NPs were about 15 ± 5 nm, 30 ± 5 nm, and 45 ± 5 nm (Figure 1A–C, left panels). The enlarge images showed shapes of CeO₂ NPs were mostly irregular spheres (Figure 1A–C, right panels).

Figure 1 Characterization of CeO2 NPs by transmission electron microscopy (TEM), showing nanoparticles with average diameters of (A) 15 ± 5 nm, (B) 30 ± 5 nm, and (C) 45 ± 5 nm. (Left image scale bar: 50 nm and right image scale bar: 10 nm).

Cytotoxicity of CeO₂ NPs in ARPE-19 Cells

A previous study showed that CeO₂ NPs induced toxicity in human lung cells.¹⁶ In our present study using ARPE-19 cells we compared the cytotoxicity of CeO₂ NPs of different diameters, including 15, 30 and 45 nm. The cytotoxicity was determined using three parameters, namely ATP content, LDH release and MTS viability assay (Figure 2). ARPE-19 cells were treated with different sizes of CeO₂ NPs at concentrations of 3.125 μg/mL to 100 μg/mL for 24 and 48 h. As shown in Figure 2A and B, CeO₂ NPs caused a time-dependent decrease in ATP content (Figure 2A and B). In addition, our results indicate that CeO₂ NPs lead to increased LDH release (Figure 2C and D) and growth inhibition (Figure 2E and F), and that ARPE-19 cells have the highest sensitivity to 15 nm CeO₂ NPs. Therefore, in the current study, we focused the following studies on 15 nm CeO₂ NPs.

Figure 2 CeO2 NPs induced cytotoxicity in ARPE-19 cells. ARPE-19 cells were exposed to different concentrations (3.125–100 μg/mL) of CeO2 NPs for (A, C and E) 24 h and (B, D and F) 48 h before measurements of (A and B) ATP content, (C and D) LDH release and (E and F) cytotoxicity determined using the MTS assay. Data points represent the mean ± SD from three independent experiments with three samples per concentration in each experiment. *p < 0.05 compared to controls.

Figure 3 CeO2 NPs induced morphological changes in cells. Morphological changes of ARPE-19 cells were observed via microscopy following 24 h and 48 h of exposure to CeO2 NPs with indicated concentrations. (Scale bar: 25 μm.).

CeO₂ NPs Induce Morphological Changes in Cells

The morphology of ARPE-19 cells changed with the increase of CeO₂ NP concentration. Morphological analysis of ARPE-19 cells exposed to CeO₂ NPs showed that the morphology of ARPE-19 cells became irregular starting from the concentration of 25 μg/mL after 24 h of exposure (Figure 3). At 48 h, the changes of cell morphology became more prominent with increasing concentration. At 100 μg/mL, most cells detached, and the density was reduced.

CeO₂ NPs Induce ROS Generation

Cytotoxicity can result from ROS accumulation, so it is of interest to investigate whether CeO₂ NPs induce ROS generation in human retinal pigment epithelial cells. Therefore, ARPE-19 cells were treated with CeO₂ NPs at concentrations ranging between 3.125 μg/mL and 100 μg/mL, and ROS production was monitored at 6, 12, 24 and 48 h (Figure 4A). ROS generation was observed as early as during the 12-h treatment with 6.25 μg/mL. The maximum ROS induction was two-fold that of the control at 24 h and 100 μg/mL CeO₂ NPs treatment. At 48 h, ROS levels were less remarkable compared to those at 12 and 24 h even though the production of ROS at 48 h remained significantly elevated compared to the corresponding control. For instance, at the concentration of 12.5 μg/mL, ROS generation was 1.19-fold higher at 48 h for CeO₂ NPs, whereas the ROS levels were 1.35-fold higher and 1.23-fold higher at 12 and 24 h, respectively. The reduction in ROS may result from decreased cell viability (Figure 2). To further verify the ROS generation results, we performed ROS fluorescence staining. ARPE-19 cells were treated with CeO₂ NPs at concentrations of 6.25, 25 and 100 μg/mL. Confocal laser scanning microscopy (CLSM) images showed that ROS levels increased (Figure 4B) as the incubation time increased. The increased intensity of the ROS indicator suggested that oxidative stress resulted from CeO₂ NPs treatment.

Figure 4 CeO2 NPs induced ROS generation. (A) ROS levels were measured at 6, 12, 24 and 48 h after exposure to various concentrations (3.125–100 μg/mL) of CeO2 NPs by H2DCF-DA staining. (B) ROS levels were monitored under CLSM, which showed that ROS levels increased following 6, 12, 24, and 48 h of exposure to CeO2 NPs with concentrations of 25 and 100 μg/mL. Data points represent the mean ± SD from three independent experiments with three samples per concentration. *p < 0.05 compared to controls. (Scale bar: 25 μm.).

CeO₂ NPs Induce Mitochondrial Dysfunction

Mitochondrial dysfunction can lead to cellular energetic depression, which may result in cell death. In addition, mitochondria are the main sites of ATP and ROS generation. Mitochondrial depolarization (the ΔΨm decrease) can lead to ROS accumulation and decreased ATP level. Next, we explored whether CeO₂ NPs cause mitochondrial depolarization in ARPE-19 cells. The ΔΨm was accessed by JC-1 dye; the accumulation of JC-1 in organelles leads to the formation of red J-aggregates (emission maximum at 590 nm) at higher mitochondrial concentrations, reflecting higher mitochondrial potential, which, in addition to the typical green fluorescence of J-monomers (emission maximum of 529 nm) at lower mitochondrial concentrations, indicates loss of membrane potential. The ARPE-19 cells were treated with CeO₂ NPs at concentrations of 6.25, 25 and 100 μg/mL for 24 and 48 h. The decreased ΔΨm of ARPE-19 cells was observed as early as the 24-h treatment with 6.25 μg/mL(Figure 5A). JC-1 staining images showed that the transition from red fluorescence to green fluorescence became more obvious at 48 h (Figure 5B), which suggested that CeO₂ NPs induced a significant time- and concentration-dependent decrease of ΔΨm.

Figure 5 CeO2 NPs induce mitochondrial dysfunction. ARPE-19 cells were treated with three concentrations (6.25, 25 and 100 μg/mL) of CeO2 NPs for 24 h (A) and 48 h (B). JC-1 staining was performed to assess mitochondrial membrane potential. (Scale bar: 25 μm.).

CeO₂ NPs-Induced Cytotoxicity is Attenuated by the ROS Scavenger

To investigate further the role of ROS generation in the cytotoxicity of CeO₂ NPs, we used a ROS scavenger, NAC, to suppress intracellular ROS levels. Pretreating ARPE-19 cells with 10 mM NAC for 1 h, prior to exposures of 3.125 −100 μg/mL CeO₂ NPs for 12h, significantly attenuated ROS induction and confirmed the effectiveness of NAC pretreatment (Figure 6A). The NAC pretreatment alleviated CeO₂ NPs-induced cytotoxicity, as evidenced by reductions in both ATP content (Figure 6B) and LDH release (Figure 6C) in NAC pretreated groups. These results demonstrated that CeO₂ NPs cytotoxicity was partially mediated by ROS generation.

Figure 6 NAC pretreatment alleviates CeO2 NPs-induced cytotoxicity. (A) Intracellular ROS levels were measured after a 12-h CeO2 NPs treatment with and without 1-h pretreatment of 10 mM NAC. (B and C) ATP content and LDH release were evaluated after a 24-h CeO2 NPs treatment with and without 1-h pretreatment of 10 mM NAC. The data points represent the mean ± SD from at least three independent experiments. *,#p < 0.05 compared to the vehicle control without or with NAC pretreatment, respectively. &p < 0.05 between the treatments with and without NAC pretreatment at the same concentration of CeO2 NPs.

Discussion

Given the wide application of CeO₂ NPs in the biomedical field raises safety concerns for human health. The toxicity of CeO₂ NPs has been studied previously by some investigator.^15,28,29 Characteristics of nanomaterials, including synthesis methodologies, size, and coating, affect their toxicity.³⁰ Unfortunately, these previous studies ignored some of these characteristics on CeO₂ NPs-induced cytotoxicity. In our present study, we investigated whether the size can affect the cytotoxicity of CeO₂ NPs. We assessed the toxicity of CeO₂ NPs of different sizes (15 ± 5 nm, 30 ± 5 nm and 45 ± 5 nm) in ARPE-19 cells. Lin et al¹⁶ and Mittal et al¹⁵ studied the fate of 20 and 177 nm CeO₂ NPs in human lung cells. They found that CeO₂ NPs can induce oxidative stress, DNA damage and apoptosis in A549 cells. In SMMC-7721 cells, exposure to hexahedral CeO₂ NPs with a size of 20–30 nm induced apoptosis and oxidative stress by activation of MAPK signaling pathways.¹⁷ Recently, the therapeutic effects of CeO₂ NPs in the retinal degenerative process were reported.^13,14 With the widespread application of CeO₂ NPs to the treatment of ocular diseases, its ocular toxicity requires the attention of scientists and ophthalmologists. Therefore, we used ARPE-19 cells to study the ocular toxicity of CeO₂ NPs.

Previous studies indicate that the size of nanoparticles significantly alters their toxicity potential. For example, AgNPs exhibit cytotoxicity and genotoxicity in a size-dependent manner in L5718Y cells.³⁰ Kim et al studied the toxicity of silica nanoparticles with diameters of 20–200 nm in A549 epithelial cells, HepG2 epithelial cells and NIH/3T3 fibroblasts. They found that the cytotoxicity changed in a size-, dose- and cell type-dependent manner.³¹ Interestingly, among a group of silica nanoparticles ranging in size from 20 to 200 nm, the 60 nm silica nanoparticles exhibited the highest toxicity. However, whether the toxicity of CeO₂ NPs is size dependent has not been reported. In this study, we focused on the toxicity of CeO₂ NPs of three sizes. First, the particle morphology and average size of CeO₂ NPs were examined by TEM (Figure 1). The images showed three dimensions: 15 ± 5 nm, 30 ± 5 nm and 45 ± 5 nm. Next, we compared the cytotoxicity of these three kinds of CeO₂ NPs (Figure 2). All three of the tested CeO₂ NPs induced different magnitudes of cytotoxicity in ARPE-19 cells. Among them, the 15 nm CeO₂ NPs showed the highest cytotoxicity. Thus, in the subsequent toxicity studies, only 15 nm CeO₂ NPs were used.

Mittal et al found that 8–20 nm CeO₂ NPs accumulated in the cytoplasm of A549 cells, resulting in cell morphology changes.¹⁵ We tested whether the cytotoxicity of 15 nm CeO₂ NPs affects the morphology of ARPE-19 cells. The density of ARPE-19 cells treated with 15 nm CeO₂ NPs became lower in a time- and concentration-dependent manner, and the shape of cells became ambiguous (Figure 3). In the past decade, CeO₂ NPs have been reported as one kind of antioxidants for scavenging ROS in many diseases,³² including cancer,³³ ocular diseases¹⁴ and neurodegenerative diseases.³⁴ However, some studies reported that the small particle size and large reactive surface of nanomaterials can lead to toxicity through the production of ROS and oxidative stress.^35,36 It has been reported that ROS accumulation is an upstream event that triggers cytotoxicity.³⁷ Thus, we investigated whether CeO₂ NPs induce ROS overproduction in ARPE-19 cells. We found that ROS generation resulted from CeO₂ NPs treatment and occurred at a lower concentration and earlier (Figures 4 and 6). Studies have reported ROS overproduction accompanied by mitochondrial membrane potential depolarization.^15,38,39 In this study, CeO₂ NPs exhibited the ability to induce mitochondrial dysfunction (Figure 5). The result suggested that the excessive production of ROS leads to mitochondrial dysfunction. Our previous study showed it is likely that ROS overproduction is the upstream event triggering cytotoxicity.³⁷ In the present study, we confirmed this phenomenon also existed in the CeO₂ NPs-induced cytotoxicity. This assertion was evidenced by the following assays: inhibition of ROS significantly diminished LDH release and cellular ATP depletion caused by CeO₂ NPs (Figure 6). It is worth noting that NAC only showed modest protective effect on CeO₂ NPs-induced cytotoxicity (Figure 6B and C); therefore, CeO₂ NPs-induced ROS may not be the only cause of cytotoxicity.

Conclusion

In summary, the current study suggests that CeO₂ NPs induce morphological alteration and cytotoxicity in a time- and dose-dependent manner. Oxidative stress, including ROS overproduction and mitochondrial membrane potential depolarization, may be part of the cause of CeO₂ NP-induced toxicity. Our results provide new insights into the toxicity of CeO₂ NPs in ophthalmologic research and improve our understanding of potential hazards associated with the application of CeO₂ NPs for treating eye diseases. However, this study did not explore molecular pathways related to the CeO₂ NP-induced toxicity in ARPE-19 cells. For a better understanding of which signal pathways play a crucial role in CeO₂ NP-induced ocular toxicity, additional studies are needed.