Intracellular Exposure Dose-Associated Susceptibility of Steatotic Hepatocytes to Metallic Nanoparticles

Non-alcoholic fatty liver disease (NAFLD), mainly characterized by the accumulation of excess fat in hepatocytes, is the most prevalent liver disorder afflicting ~25% of adults worldwide. In vivo studies have shown that adult rodents with NAFLD were more sensitive to metallic nanoparticles (MNPs) than healthy MNPs. However, due to the complex interactions between various cell types in a fatty liver, it has become a major challenge to reveal the toxic effects of MNPs to specific types of liver cells such as steatotic hepatocytes. In this study, we reported the susceptibility of steatotic hepatocytes in cytotoxicity and the induction of oxidative stress to direct exposures to MNPs with different components (silver, ZrO2, and TiO2 NPs) and sizes (20–30 nm and 125 nm) in an oleic acid (OA) -induced steatotic HepG2 (sHepG2) cell model. Furthermore, the inhibitory potential of MNPs against the process of fatty acid oxidation (FAO) were obvious in sHepG2 cells, even at extremely low doses of 2 or 4 μg/mL, which was not observed in non-steatotic HepG2 (nHepG2) cells. Further experiments on the differential cell uptake of MNPs in nHepG2 and sHepG2 cells demonstrated that the susceptibility of steatotic hepatocytes to MNP exposures was in association with the higher cellular accumulation of MNPs. Overall, our study demonstrated that it is necessary and urgent to take the intracellular exposure dose into consideration when assessing the potential toxicity of environmentally exposed MNPs.


Introduction
The widespread applications of nanomaterials in various areas such as consumer products [1], agriculture [2], biomedicine [3] and environmental remediation [4] result in their higher environmental release and human exposure. Metallic nanoparticles (MNPs) are among the largest classes of engineered nanoparticles (NPs) found in daily life. For example, commercial products containing MNPs account for 70% of the listed entries with identified NPs supplemented [5]. Environmentally exposed MNPs may enter the human body, interact with biomolecules, and perturb various physiological systems [6,7]. Such adverse outcomes in response to MNP exposure may be further aggravated in susceptible populations characterized by underdeveloped protection mechanisms, impaired self-repair ability, and/or compromised immunity [8][9][10]. Therefore, it is urgent to reveal the potential toxic effects of engineered MNPs to various susceptible populations.
The liver is the main target of various xenobiotic substances including MNPs [11]. Individuals with hepatic disorders (e.g., hepatitis [12] and hepatic steatosis [13]) are, thus, suggested to be more sensitive to environmentally exposed MNPs. Non-alcoholic fatty liver disease (NAFLD) is the most prevalent liver disorder characterized by excess accumulation 2 of 13 of fat in hepatocytes, afflicting approximately 25% of the global adult population [14,15]. Exposure to MNPs resulted in aggravated liver injuries in animal models of NAFLD, with enhanced hepatic inflammation and hepatocyte steatosis as the major manifestations [13,[16][17][18][19]. However, the complex interplay among different kinds of liver cells in animals has become a limitation for exploring the direct toxic effects of MNPs to the specific type of liver cells.
Hepatocytes are the major parenchymal cells of the liver, comprising 70-85% of liver volume and playing critical roles in metabolism, detoxification, protein synthesis, and innate immunity [20]. In vitro studies showed that exposure to MNPs led to oxidative stress, inflammation, and eventually different types of cell death outcomes in various hepatic cell lines [21]. While the different sensitivities of normal liver cells and hepatic cancer cells to MNP exposure have been reported in several recent studies, with cancer cells more sensitive than normal cells [22,23], the susceptibility of hepatocytes with physiological abnormalities, i.e., steatotic hepatocytes, to direct MNP exposure, remains to be elucidated.
The oleic acid (OA)-induced steatotic HepG2 cell model shows morphological similarities to the hepatocytes of patients with hepatic steatosis and, thus, has been applied as an in vitro model of steatosis for revealing pathogenesis and evaluating the effects of medical or dietary interventions of hepatic steatosis [24][25][26]. In the present study, the toxic potential of several MNPs with high environmental exposure risks to human non-steatotic and steatotic hepatocytes was determined. Both composition and particle size are critical factors in determining the cytotoxicity of MNPs [27,28]. While silver nanoparticles (Ag NPs) and titanium dioxide NPs (TiO 2 NPs) are the top two MNPs that have the most prevalent usage in consumer products [5], zirconium dioxide NPs (ZrO 2 NPs) have been widely used for the removal of various pollutants from wastewater, environmental water, and even drinking water [29,30]. Thus, four NPs that have distinct compositions and/or particle sizes were selected to reveal the differential toxicity of MNPs to non-steatotic and steatotic hepatocytes.

Characterization of MNPs
All four MNPs selected were spherical in shape, with transmission electron microscope (TEM) sizes of 23.58 ± 6.30, 28.15 ± 6.31, 21.16 ± 5.04, and 125.28 ± 41.16 nm for Ag NPs, ZrO 2 NPs, small-sized TiO 2 (sTiO 2 ) NPs, and large-sized TiO 2 (lTiO 2 ) NPs, respectively ( Figure 1). The hydrodynamic diameters of these NPs were generally much larger than their TEM sizes, which were further enlarged in minimum essential medium (MEM) cell culture medium supplemented with 10% fetal bovine serum (FBS) due to the adsorption of proteins on the surface ( Figure 1E). The formation of protein corona also shifted the surface charges, generally increasing the zeta potential values of these four NPs to similar levels ( Figure 1E).

Steatotic Hepatocyte Modeling
The dose-dependent effects of OA treatments on the accumulation of lipid and growth of HepG2 cells were determined for the selection of a proper dose for steatotic hepatocyte modeling. There was almost an absence of intracellular lipid in cells without the treatment of OA (Figure 2A). OA treatments led to the accumulation of lipid droplets in the cytoplasm of HepG2 cells ( Figure 2B-F), at a concentration of ≥0.5 mM. While treatments with 0.1 or 0.5 mM OA showed little effect on the growth of cells, OA exposure at concentrations higher than 1.0 mM resulted in a decreased cell index, suggesting significant alteration of cell growth ( Figure 2G). Thus, an OA dose of 0.5 mM, which effectively induced steatosis without affecting the normal growth of HepG2 cells, was selected for steatotic hepatocyte modeling in the following experiments. Following steatotic hepatocyte modeling, the question of whether steatotic hepatocytes were more susceptible to MNP exposure was further investigated. , small-sized TiO2 (sTiO2) NPs (C), and large-sized TiO2 (lTiO2) NPs (D). (E) Summarized characteristics of these four MNPs, including TEM size, hydrodynamic diameter, and zeta potential in DI water and MEM cell culture medium supplemented with 10% fetal bovine serum (FBS). In (E), data are shown as means ± s.d., n = 3.

Steatotic Hepatocyte Modeling
The dose-dependent effects of OA treatments on the accumulation of lipid and growth of HepG2 cells were determined for the selection of a proper dose for steatotic hepatocyte modeling. There was almost an absence of intracellular lipid in cells without the treatment of OA (Figure 2A). OA treatments led to the accumulation of lipid droplets in the cytoplasm of HepG2 cells ( Figure 2B-F), at a concentration of ≥0.5 mM. While treatments with 0.1 or 0.5 mM OA showed little effect on the growth of cells, OA exposure at concentrations higher than 1.0 mM resulted in a decreased cell index, suggesting significant alteration of cell growth ( Figure 2G). Thus, an OA dose of 0.5 mM, which effectively induced steatosis without affecting the normal growth of HepG2 cells, was selected for steatotic hepatocyte modeling in the following experiments. Following steatotic hepatocyte modeling, the question of whether steatotic hepatocytes were more susceptible to MNP exposure was further investigated. , small-sized TiO 2 (sTiO 2 ) NPs (C), and large-sized TiO 2 (lTiO 2 ) NPs (D). (E) Summarized characteristics of these four MNPs, including TEM size, hydrodynamic diameter, and zeta potential in DI water and MEM cell culture medium supplemented with 10% fetal bovine serum (FBS). In (E), data are shown as means ± s.d., n = 3.

Higher Cytotoxicity of MNPs to sHepG2 Than to nHepG2
Ag NPs are the most toxic of the four MNPs, causing a decrease in viability of nonsteatotic HepG2 (nHepG2) cells at concentrations ≥ 10 µg/mL ( Figure 3A). ZrO 2 NPs caused a slight decrease in viability of nHepG2 cells at concentrations of 200 µg/mL ( Figure 3B), and TiO 2 NPs of any of these two sizes showed little effect on the viability of nHepG2 cells at concentrations of as high as 50 µg/mL ( Figure 3C,D). These results indicated the NP composition-dependent cytotoxicity of MNPs, which has been well recognized in previous studies [31][32][33].

Higher Cytotoxicity of MNPs to sHepG2 Than to nHepG2
Ag NPs are the most toxic of the four MNPs, causing a decrease in viability of nonsteatotic HepG2 (nHepG2) cells at concentrations ≥ 10 μg/mL ( Figure 3A). ZrO2 NPs caused a slight decrease in viability of nHepG2 cells at concentrations of 200 μg/mL (Figure 3B), and TiO2 NPs of any of these two sizes showed little effect on the viability of nHepG2 cells at concentrations of as high as 50 μg/mL ( Figure 3C,D). These results indicated the NP composition-dependent cytotoxicity of MNPs, which has been well recognized in previous studies [31][32][33].
In steatotic HepG2 (sHepG2) cells, the toxicity of MNPs is much higher than that in nHepG2 cells. While the viabilities of nHepG2 cells treated with 10 and 20 μg/mL Ag NPs were 88.5% and 61.4%, the same treatments reduced the viabilities of sHepG2 cells to 58.5% and 44.9%, respectively ( Figure 3A). In an OA-induced steatotic human hepatocellular carcinoma (HCC) Bel7402 cell model, Ag NP exposure also led to a much higher cytotoxicity than in the non-steatotic ( Figure S1). Similar enhanced cytotoxicity to the steatotic hepatocytes, relative to the non-steatotic, was also found for ZrO2 NP exposure, as indicated by a much lower cell viability of sHepG2 (68.2%) compared with that of nHepG2 (84.9%) in response to 200 μg/mL ZrO2 NPs exposure ( Figure 3B). Treatments with 50 μg/mL sTiO2 and lTiO2 NPs decreased the cell viability to 35.9% and 69.0% only in sHepG2 cells ( Figure 3C,D). Despite the similarity in cell type-, dose-, composition-and particle size-dependent cytotoxicity to non-steatotic and steatotic liver cells, our above results demonstrated that the steatotic hepatocytes were more susceptible in cytotoxicity to MNP exposure than the non-steatotic. Using an antioxidant responsive element (ARE) reporter HepG2 cell line (as described in the Materials and Methods), we further investigated the ARE gene expression in nHepG2 and sHepG2 with NP treatments. Ag NPs and sTiO2 NPs showed little alteration on the expression of ARE reporter in nHepG2 cells, with the relative units ranging from ~0.8 to ~1.0 ( Figure 4A). The ARE reporter expressions in sHepG2 cells exposed to 4, In steatotic HepG2 (sHepG2) cells, the toxicity of MNPs is much higher than that in nHepG2 cells. While the viabilities of nHepG2 cells treated with 10 and 20 µg/mL Ag NPs were 88.5% and 61.4%, the same treatments reduced the viabilities of sHepG2 cells to 58.5% and 44.9%, respectively ( Figure 3A). In an OA-induced steatotic human hepatocellular carcinoma (HCC) Bel7402 cell model, Ag NP exposure also led to a much higher cytotoxicity than in the non-steatotic ( Figure S1). Similar enhanced cytotoxicity to the steatotic hepatocytes, relative to the non-steatotic, was also found for ZrO 2 NP exposure, as indicated by a much lower cell viability of sHepG2 (68.2%) compared with that of nHepG2 (84.9%) in response to 200 µg/mL ZrO 2 NPs exposure ( Figure 3B). Treatments with 50 µg/mL sTiO 2 and lTiO 2 NPs decreased the cell viability to 35.9% and 69.0% only in sHepG2 cells ( Figure 3C,D). Despite the similarity in cell type-, dose-, composition-and particle size-dependent cytotoxicity to non-steatotic and steatotic liver cells, our above results demonstrated that the steatotic hepatocytes were more susceptible in cytotoxicity to MNP exposure than the non-steatotic.
Using an antioxidant responsive element (ARE) reporter HepG2 cell line (as described in the Materials and Methods), we further investigated the ARE gene expression in nHepG2 and sHepG2 with NP treatments. Ag NPs and sTiO 2 NPs showed little alteration on the expression of ARE reporter in nHepG2 cells, with the relative units ranging from~0.8 to~1.0 ( Figure 4A). The ARE reporter expressions in sHepG2 cells exposed to 4, 8, or 12 µg/mL Ag NPs were 1.3-, 1.7-, and 2.6-fold that of vehicle treatment ( Figure 4B). Such upregulated expressions of ARE reporter were much lower in sTiO 2 NP treated sHepG2 cells, with the relative units ranging from 1.2 to 1.4 ( Figure 4B). Using an antioxidant responsive element (ARE) reporter HepG2 cell line (as described in the Materials and Methods), we further investigated the ARE gene expression in nHepG2 and sHepG2 with NP treatments. Ag NPs and sTiO2 NPs showed little alteration on the expression of ARE reporter in nHepG2 cells, with the relative units ranging from ~0.8 to ~1.0 ( Figure 4A). The ARE reporter expressions in sHepG2 cells exposed to 4, 8, or 12 μg/mL Ag NPs were 1.3-, 1.7-, and 2.6-fold that of vehicle treatment ( Figure 4B). Such upregulated expressions of ARE reporter were much lower in sTiO2 NP treated sHepG2 cells, with the relative units ranging from 1.2 to 1.4 ( Figure 4B).

Alterations of Fatty Acid Oxidation (FAO)-Related Genes in sHepG2 Cells
The accumulation of lipid droplets in hepatocytes is one of the most basic pathological characteristics of clinical hepatic steatosis. Here, the alteration of NP exposures on the contents of intracellular lipids and the expression of lipid metabolism-related genes was further investigated. OA treatments led to lipid accumulation in cells and increased intracellular contents of both total cholesterol (T-CHO) and triacylglycerol (TG) ( Figure 5). Exposure to MNPs caused few alterations in the intracellular lipids of either nHepG2 or sHepG2 cells ( Figure 5).
The accumulation of lipid droplets in hepatocytes is one of the most basic pathological characteristics of clinical hepatic steatosis. Here, the alteration of NP exposures on the contents of intracellular lipids and the expression of lipid metabolism-related genes was further investigated. OA treatments led to lipid accumulation in cells and increased intracellular contents of both total cholesterol (T-CHO) and triacylglycerol (TG) ( Figure 5). Exposure to MNPs caused few alterations in the intracellular lipids of either nHepG2 or sHepG2 cells ( Figure 5).  Despite the negative regulatory effects of MNPs on the intracellular lipids, the expression of several fatty acid metabolism-related genes was further detected in nHepG2 and sHepG2 cells in response to Ag NP exposure at non-lethal doses. The expressions of peroxisome proliferator activated receptor alpha (Ppara), peroxisome proliferator activated receptor delta (Ppard), peroxisome proliferator activated receptor gamma coactivator 1alpha (Pgc1a), and carnitine palmitoyltransferase-1b (Cpt1b) in sHepG2 cells were 1.12-, 1.30-, 1.77-, and 2.09-fold those in nHepG2 (Figure 6), suggesting enhanced FAO in response to OA internalization in sHepG2 cells. While showing little alteration on the expressions of these four FAO-related genes in nHepG2 cells, Ag NPs exposure (4 µg/mL) upregulated the expression of Ppard by 112% but downregulated the expression of its coactivator Pgc1a and target gene Cpt1b by 16% and 20%, respectively, in sHepG2 cells ( Figure 6B-D). tor 1-alpha (Pgc1a), and carnitine palmitoyltransferase-1b (Cpt1b) in sHepG2 cells were 1.12-, 1.30-, 1.77-, and 2.09-fold those in nHepG2 (Figure 6), suggesting enhanced FAO in response to OA internalization in sHepG2 cells. While showing little alteration on the expressions of these four FAO-related genes in nHepG2 cells, Ag NPs exposure (4 μg/mL) upregulated the expression of Ppard by 112% but downregulated the expression of its coactivator Pgc1a and target gene Cpt1b by 16% and 20%, respectively, in sHepG2 cells (Figure 6B-D).

Higher Accumulation of MNPs in sHepG2 Than in nHepG2
At non-lethal dose of 4 μg/mL, MNPs showed higher accumulation in sHepG2 cells in comparison with that in nHepG2 cells. The amounts of Ag NPs, ZrO2 NPs, sTiO2 NPs, and lTiO2 NPs in sHepG2 were 2.8-, 1.9-, 1.4-, and 1.3-fold those in nHepG2 after the cells were incubated with these NPs for 24 h ( Figure 7A). Meanwhile, neither the removal of OA from culture medium for sHepG2 cells nor the supplement of OA in medium for nHepG2 cells significantly shifted the accumulation of Ag NPs and sTiO2 NPs in these two cells ( Figure 7B). These results demonstrate that it is the steatotic status but not the

Higher Accumulation of MNPs in sHepG2 Than in nHepG2
At non-lethal dose of 4 µg/mL, MNPs showed higher accumulation in sHepG2 cells in comparison with that in nHepG2 cells. The amounts of Ag NPs, ZrO 2 NPs, sTiO 2 NPs, and lTiO 2 NPs in sHepG2 were 2.8-, 1.9-, 1.4-, and 1.3-fold those in nHepG2 after the cells were incubated with these NPs for 24 h ( Figure 7A). Meanwhile, neither the removal of OA from culture medium for sHepG2 cells nor the supplement of OA in medium for nHepG2 cells significantly shifted the accumulation of Ag NPs and sTiO 2 NPs in these two cells ( Figure 7B). These results demonstrate that it is the steatotic status but not the presence of OA that contributes to the higher accumulation of MNPs in sHepG2 than in nHepG2. presence of OA that contributes to the higher accumulation of MNPs in sHepG2 than in nHepG2.

Discussion
Liver toxicity is a pivotal concern for risk assessment of environmentally exposed MNPs, as the liver is the major organ where MNPs are deposited and metabolized. In healthy adult rodents, MNP exposure caused hepatic inflammation, oxidative stress, DNA damage, and hepatocyte apoptosis and/or necrosis [34][35][36]. The above adverse outcomes in response to MNP exposure may be further aggravated in rodent models of NAFLD [13], suggesting the susceptibility of NAFLD individuals to MNPs. Additionally, MNPs aggravated hepatocyte steatosis in NAFLD animal models [16,17]. Hepatic inflammation is a main driver of liver injury, accelerating the transformation of NAFLD from hepatic steatosis to steatohepatitis [37]. Meanwhile, it plays a critical role in regulating the lipid metabolism in hepatocytes. For example, pro-inflammatory cytokine tumor necrosis factor α (TNF-α) disturbed the metabolism of hepatic fatty acid, contributing to the pathogenesis of hepatic steatosis [38,39]. While the induction of hepatic inflammation, as indicated by the increased production of pro-inflammatory cytokines such as TNF-α, interleukin-6 (IL-6), and IL-1β, is one of the most common toxic outcomes of MNPs in both healthy and NAFLD animals [21], the susceptibility of steatotic hepatocytes to direct MNP exposures, especially in regulating the metabolic activity of intracellular lipids, remains to be elucidated. In the present work, the susceptibility of steatotic hepatocytes to four MNPs with different components and particle sizes was first determined in vitro using an OA-induced steatotic HepG2 cell model. Despite the particle size-and component-dependent cytotoxicity of these MNPs to both nHepG2 and sHepG2 cells, MNPs were more cytotoxic to sHepG2 than to nHepG2 (Figure 3), suggesting the susceptibility of steatotic hepatocytes in cytotoxicity to direct MNP exposure.
Induction of oxidative stress is one of the most well-recognized toxic mechanisms of nanosized particles [40]. The Kelch-like ECH-associated protein 1 (Keap1)/NF-E2-related factor 2 (Nrf2)/ARE signaling system is the main antioxidant mechanism against even the

Discussion
Liver toxicity is a pivotal concern for risk assessment of environmentally exposed MNPs, as the liver is the major organ where MNPs are deposited and metabolized. In healthy adult rodents, MNP exposure caused hepatic inflammation, oxidative stress, DNA damage, and hepatocyte apoptosis and/or necrosis [34][35][36]. The above adverse outcomes in response to MNP exposure may be further aggravated in rodent models of NAFLD [13], suggesting the susceptibility of NAFLD individuals to MNPs. Additionally, MNPs aggravated hepatocyte steatosis in NAFLD animal models [16,17]. Hepatic inflammation is a main driver of liver injury, accelerating the transformation of NAFLD from hepatic steatosis to steatohepatitis [37]. Meanwhile, it plays a critical role in regulating the lipid metabolism in hepatocytes. For example, pro-inflammatory cytokine tumor necrosis factor α (TNF-α) disturbed the metabolism of hepatic fatty acid, contributing to the pathogenesis of hepatic steatosis [38,39]. While the induction of hepatic inflammation, as indicated by the increased production of pro-inflammatory cytokines such as TNF-α, interleukin-6 (IL-6), and IL-1β, is one of the most common toxic outcomes of MNPs in both healthy and NAFLD animals [21], the susceptibility of steatotic hepatocytes to direct MNP exposures, especially in regulating the metabolic activity of intracellular lipids, remains to be elucidated. In the present work, the susceptibility of steatotic hepatocytes to four MNPs with different components and particle sizes was first determined in vitro using an OA-induced steatotic HepG2 cell model. Despite the particle size-and component-dependent cytotoxicity of these MNPs to both nHepG2 and sHepG2 cells, MNPs were more cytotoxic to sHepG2 than to nHepG2 (Figure 3), suggesting the susceptibility of steatotic hepatocytes in cytotoxicity to direct MNP exposure.
Induction of oxidative stress is one of the most well-recognized toxic mechanisms of nanosized particles [40]. The Kelch-like ECH-associated protein 1 (Keap1)/NF-E2related factor 2 (Nrf2)/ARE signaling system is the main antioxidant mechanism against even the lowest level of NP-triggered oxidative stress [41]. The repressor protein, Keap1, acts as a sensor of intracellular oxidative stress, dissociating the transcription factor Nrf2 from the Keap1-Nrf2 complex in response to oxidative stress, and the released Nrf2 then regulates the induction of gene encoding antioxidant proteins via ARE [42,43]. Thus, the ARE reporter has been applied as the most sensitive strategy for measuring the NPinduced oxidative stress response in cells [44]. Using an ARE reporter HepG2 cell line, we compared the induction of oxidative stress in nHepG2 and sHepG2 cells in response to MNP exposure. Treatments with MNPs upregulated the expression of the ARE reporter only in sHepG2 cells (Figure 4), suggesting that the induction of oxidative stress was involved in the susceptibility of steatotic hepatocytes in cytotoxicity to MNP exposure. Previously, the induction of oxidative stress in response to MNP exposure was associated with the inhibition of various detoxification enzymes. One great example is that the inhibition of antioxidant defense systems was shown to be responsible for the gold NPinduced oxidative stress in various human cells [45,46]. Thus, the potential role of the regulatory effects of MNPs on the differential induction of oxidative stress in nHepG2 and sHepG2 cells remains to be elucidated.
The pathogenesis of NAFLD is complex and multifactorial. Peroxisome proliferatoractivated receptors (PPARs) are a class of nuclear receptors that have been proven to play critical roles in regulating the metabolism of glucose and lipids, contributing to the development and progression of NAFLD [47]. Among the three PPAR isoforms existing in mammals, both PPARα and PPARδ have high expression in the liver, while PPARγ is highly expressed in adipose tissue [48]. In response to Ag NP exposures, our previous study has shown that the downregulated expression of Ppard, and its co-activator Pgc1a and target genes contributed to the aggravated hepatic steatosis in overweight mice [17]. However, the question of whether the increased accumulation of lipids in hepatocytes is caused by the internalized Ag NPs in hepatocytes remains to be answered. In the present work, Ag NP exposure showed little effect on the expression of Ppara. Meanwhile, treatments with Ag NPs upregulated the expression of Ppard but downregulated its co-activator Pgc1a and target gene Cpt1b ( Figure 6). Despite the disagreement of the Ag NP-regulated expression of Ppard in vitro and in vivo, the inhibitory potential of Ag NP exposure on the process of FAO is suggested to be the same, as Ppard and its co-activator Pgc1a together regulate the process of FAO [49]. In contrast, significantly increased accumulation of lipids was not detected in either nHepG2 or sHepG2 cells in response to Ag NP exposure ( Figure 5), probably because an exposure time of 24 h was too short.
The intracellular amounts of MNPs increased with the exposure doses, resulting in dose-dependent cytotoxicity of various MNPs, which has been well-recognized in previous studies [50][51][52]. Thus, the MNP-induced cytotoxicity is primarily determined by the intracellular dose of MNPs. In this work, we found that both Ag NPs and sTiO2 NPs had much higher accumulation in sHepG2 cells than in nHepG2 cells (Figure 7a), which was suggested to be responsible for the susceptibility of sHepG2 cells in response to MNP exposure. Even though we excluded the contribution of OA on the differential internalization of MNPs in nHepG2 and sHepG2 cells (Figure 7b), the underlying mechanisms of the enhanced accumulation of MNPs in steatotic hepatocytes remain to be elucidated from how the steatotic status affects the internalization, subcellular localization, and excretion of MNPs in steatotic hepatocytes.

Cell Culture and Steatotic Hepatocyte Modeling
Human hepatocellular carcinoma HepG2 cells kindly provided by Stem Cell Bank, Chinee Academy of Sciences were cultured in minimum essential medium (MEM, HyClone, UT, USA) containing 10% FBS (ExCell Bio, Shanghai, China), 1% Penicillin-Streptomycin Solution (HyClone, UT, USA), 1% Sodium Pyruvate (Gibco, New York, NY, USA), 1% MEM Non-Essential Amino Acids (Gibco, New York, NY, USA), and 1% GlutaMAXTM-1 (Gibco, New York, NY, USA). The cells were maintained in a 95% humidity and 5% CO 2 incubator at 37 • C for cultivation. For steatotic hepatocyte modeling, HepG2 cells were seeded in a 12-well plate (60,000 cells per well). After incubation for 24 h, cell culture medium in each well were replaced with fresh culture medium containing 0, 0.1, 0.5, 1.0, 1.5, or 2.0 mM oleic acid (OA) for another 24 h. Cells were washed with PBS three times, stained with oil red O, and analyzed using ECLIPSE Ti2-U Inverted Microscope System (Nikon Instruments Inc., Tokyo, Japan). Real-time growth status of cells in response to OA treatments (0, 0.1, 0.5, 1.0, and 2.0 mM) was determined using an xCELLigence real-time cell analysis system (OLS OMNI Life Science GmbH & Co KG, Bremen, Germany) and reflected as a normalized cell index.

Cell Viability Assay
Cells were seeded in 96-well plates (6000 cells per well). After 24 h incubation, the cells were treated with 0. 5

Real-Time Quantitative Reverse Transcription PCR (qRT-PCR)
Total RNA of HepG2 cells treated with 4 µg /mL NPs for 24 h was extracted with Trizol reagent (Ambion, Carlsbad, CA, USA). The concentration and quality of RNA was determined using a NanoDrop One spectrophotometer (Thermo Scientific, Waltham, MA, USA). RNA to cDNA were reversely transcribed using PrimeScript TM RT Master Mix (Takara Bio Inc., Shiga, Japan). TB Green TM Premix Ex Taq TM II (Takara Bio Inc., Shiga, Japan) was used to perform quantitative PCR on a LightCycler ® 96 Instrument (Roche Molecular Systems, Inc., Pleasanton, CA, USA). The relative expression of genes was quantified using the 2 −∆∆CT method [53]. Primers (Table 1) used in this study were synthesized by Sangon Biotech (Shanghai, China).

Total Cholesterol and Triacylglycerol Levels
Logarithmic growth cells were seeded into 12-well plates. After incubation for 24 h, the normal cell culture medium was replaced by the cell culture medium with/without 0.5 mM OA for cell modeling for another 24 h. Then, the cells were treated with metalbased nanoparticles 24 h. The cells were collected, and the total cholesterol (T-CHO) and triacylglycerol (TG) levels of HepG2 cells after various treatments were determined by T-CHO/TG Assay Kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China).

Cellular Uptake
HepG2 cells were seeded in 12-well plates at a density of 60,000 cells per well. After a 24 h incubation and a 24 h steatotic hepatocyte modeling, the cells were treated with 4 µg/mL NPs for another 24 h. Cells were then washed three times with PBS, detached by 0.25% trypsin-EDTA solution (Gibco, New York, NY, USA), collected, and counted. The collected cells were dissolved with nitric acid and analyzed using the Inductively Coupled Plasma Mass Spectrometry (ICP-MS, Agilent 7900, Agilent Technologies, Inc., Santa Clara, CA, USA). To explore the effect of OA on the cellular uptake of NPs, cells were exposed to 4 µg/mL Ag NPs or sTiO 2 NPs for 3 h with/without the supplement of OA, and the intracellular amounts of NPs were determined using the above methods.

Conclusions
Using an OA-induced steatotic HepG2 cell model, we found that the steatotic hepatocytes were more sensitive to MNP exposures than the non-steatotic in both cytotoxicity and induction of oxidative stress. MNPs showed the potential of perturbing the process of FAO at extremely low doses only in steatotic hepatocytes. The susceptibility of steatotic hepatocytes in response to MNP exposure was in association with the higher cellular accumulation of MNPs, suggesting the necessity and urgency to take the intracellular exposure dose into consideration in the health risk assessment of environmentally exposed MNPs.