Nanoceria’s Silent Threat: Investigating Acute and Sub-Chronic Effects of CeO2 Nanopowder (≤50 nm) on the Human Intestinal Epithelial Cells

TL;DR

This study explores how nanoceria (CeO2) affects human intestinal cells over short and long-term exposures, revealing potential health risks.

Contribution

The novelty is the sub-chronic in vitro exposure assessment of CeO2 nanopowder on intestinal cells over 35 days.

Findings

CeO2 NP caused dose- and time-dependent cytotoxicity and pro-oxidant effects up to 14 days.

Mitochondrial impairment peaked at 7 days but showed recovery, while genotoxicity suggested DNA repair saturation.

Sub-chronic exposure led to irreversible cell damage and reduced cell replication, with overexpression of mitophagy-related genes.

Abstract

The increased mobilization of Rare Earth Elements (REEs), due to emerging technologies, could impact human health. The study assessed the effects of CeO2 nanopowder (100 μg/mL) in human intestinal cells (HT-29) following both acute (24 h) and, a novelty for in vitro study, sub-chronic exposure, treating subcultures of exposed cells to CeO2 NP up to 35 days. Recovery was also examined in exposed cells’ progeny. CeO2 NP internalization and acute cytotoxicity were dose and time dependent. A significant pro-oxidant effect was observed for up to 14 days. The highest mitochondrial impairment was detected after 7 days, but in post-exposure experiments the recovery was observed. Conversely, genotoxicity highlighted the saturation of the DNA repair mechanisms. The irreversible cell damage of sub-chronic exposure was highlighted by the percentage of death cells (p = 0.011) and by the weekly cell replication index (5.68 vs. 7.41). The homeostatic mitophagy pathway was able to counteract ROS-induced mitochondrial dysfunction, as shown by overexpression of ATG5, LC3, and BECN1 genes throughout the examined times. Instead, the overexpression of the pro-apoptotic gene Bax was very brief, highlighting that prolonged exposure might cause more widespread adverse effects, also involving cells that are not directly exposed to nanoceria.

Figures

Full text

1. Introduction

Emerging technologies require massive use of the Rare Earth Elements (REEs), so-called because their extraction and purification are challenging. Due to their metallic nature, the increased mobilization of REEs could have a significant impact on both ecosystems and human health since, once released, they remain permanently in the environment.

Among these elements (also known as lanthanides), cerium (atomic number 58) is abundantly present in the lithosphere, in particular, as dioxide (CeO_2_), and its concentration is higher or overlapping with those of some elements that have always been extracted (Ce 66 mg/kg vs. 55 and 70 mg/kg for Cu and Zn) [1,2]. CeO_2_, a metal-based nanomaterial capable of self-assembling into crystalline nanoparticles, is known as nanoceria. These particles can be utilized in their native state or be strategically functionalized [3]. As with other nanomaterials, the nanoscale dimensions of CeO_2_ elicit unique physicochemical properties that are not observed in its bulk counterpart [4,5]. Their nanoscale dimensions greatly increase the surface area, making them highly reactive per unit of mass [6].

Due to its unique physicochemical attributes—most notably its exceptional catalytic activity—nanoceria has seen widespread integration across diverse industrial sectors. Its applications range from semiconductor planarization and precision optics to energy technologies like fuel cells, solar cells, and fuel additives. In addition, it plays a critical role in environmental and consumer technologies, such as automotive catalytic converters (ACC), diesel particulate filters (DPF), and UV-shielding materials. Recently, it was also used in crude petroleum cracking [7,8,9].

Thanks to its electronic configuration, CeO_2_ exhibits the ability to be oxidized and reduced very rapidly and efficiently (redox behavior), changing the oxidation state between Ce^3+^ and Ce^4+^. Although nanoceria can effectively scavenge reactive oxygen species (ROS) thanks to the redox cycle, the precise impact of Ce^3+^ ions should be carefully taken into account. Similarly to transition metals, Ce^3+^ can exert a pro-oxidant effect in the presence of H_2_O_2_ by catalyzing the generation of hydroxyl radicals (°OH) via Fenton-like or Haber–Weiss reactions [10]. These redox transitions occur exclusively at the particle surface and exhibit an inverse correlation with particle size. This behavior is primarily influenced by the Ce^3+^/Ce^4+^ ratio, which is determined during synthesis through the formation of oxygen vacancies that modulate the material’s overall redox potential [4].

Given their strong ROS scavenging activity, nanoceria have also been explored for biomedical applications, particularly after pre-engineering steps aimed at improving biocompatibility [6]. Unlike traditional antioxidants, CeO_2_ NPs could theoretically act as ideal antioxidants due to their ability to spontaneously switch from the oxidized (4+) to the reduced (3+) state [11]. However, the dual role of nanoceria (i.e., oxidant/antioxidant) makes its use in medicine controversial, as evidenced by the results of numerous studies that have also observed pro-oxidant effects [12,13] in addition to its potent antioxidant properties [14,15,16,17,18].

The widespread use of nanoceria in emerging technologies inevitably causes environmental release during the disposal of nanoceria-containing devices, with potential long-term effects on ecosystems and human health. In 2019, the volume of waste from electrical and electronic equipment, with a high lanthanide content, reached 53.3 million tons [19].

However, considering the lanthanides’ natural origin, it should be noted that both occupational and unintentional human exposure, albeit at lower amounts, has always occurred. This is confirmed by the presence of CeO_2_ nanoparticles and other REEs, in coal fly ash and in other combustion by-products [20,21], causing the exposure by inhalation of workers and residents in industrial areas where oil refining plants, metallurgical industries, etc., are present. Internalization into organisms, including humans, is a common feature of many nano-sized particles, both natural and those released into the environment as a result of anthropic activities [22,23,24].

Moreover, the potential use of nanoceria in agriculture to improve crop yields could further increase human exposure through ingestion. Although this practice is currently more common in China, several studies have demonstrated the fertilizing potential of nanoceria, encouraging its wider adoption [25,26,27].

Given the uptake through the roots and translocation of nanoceria to the entire plants, including the edible parts, and both its use as a fertilizer and environmental contamination due to inappropriate waste management, the assessment of this unintentional human exposure is essential.

This route of exposure to nanoceria is in addition to the more well-known inhalation route, which is both occupational and unintentional, the latter being nowadays primarily due to the use of nanoceria as a diesel fuel additive [28,29].

Recently, we highlighted the detrimental effect of nanoceria in human alveolar cells, observing both metabolic changes and cytotoxicity due to pro-oxidant effects in almost all cell compartments [30]. As reported by Moreno et al. [31], the monitoring of urban areas detected non-negligible levels of Ce and other lanthanides, mainly in the coarse fraction of the particulate matter (PM10). Critically, significantly higher Ce and La concentrations were detected in indoor environments. This increase is largely attributed to the presence of REEs in tobacco smoke [32], which confirms REEs absorption by plant biomass and, significantly, highlights the potential role of ingestion as a pathway for involuntary human exposure.

While toxicokinetic studies indicate efficient fecal excretion, the possibility of localized damage to the intestinal epithelium induced by the xenobiotic cannot be ruled out [33]. To address this critical gap, we performed an in vitro study by exposing a human colon-derived cell line to CeO_2_ nano-powder (NP).

Given that exposure to persistent environmental pollutants is prolonged over time, we assessed the impact of sub-chronic exposure in addition to short-term effects. To accurately simulate realistic exposure scenarios by an in vitro investigation, often reserved exclusively for acute exposure evaluation, we utilized a technique we devised and previously applied to analyze the impact of sub-chronic exposure to different xenobiotics [30,34]. In particular, we extended the nanoceria treatment up to 35 days in subsequent subcultures of the exposed cells to evaluate the capacity of the cell progeny to recover basal physiological conditions through parallel experiments conducted using sub-chronically exposed cells that were subsequently maintained in a nanoparticle-free medium for a one-week recovery period. The effects of CeO_2_ NP were assayed by canonical toxicological tests and by detecting the expression of some genes involved in specific homeostatic mechanisms such as phagocytosis/autophagy and apoptotic pathways.

2. Materials and Methods

2.1. Exposure Conditions and Cell Model

A commercially available cerium oxide nanopowder (CeO_2_ NP; median primary particle diameter: 50 nm; Sigma-Merck, catalogue no. 700290, Milan, Italy) was employed for all experiments. A stock suspension (10 mg/mL) was prepared in phosphate-buffered saline (PBS), and working dilutions were freshly prepared in cell culture medium immediately prior to use. All suspensions were subjected to sonication for 20 min in an ice bath (frequency 40 kHz) to avoid nanoparticle agglomeration. As previously reported [30], under physiological pH and osmotic pressure (i.e., in growth cell medium), the zeta potential test indicated an absence of surface charges on the nanoceria particles, underscoring their high hydrophobicity and, therefore, an augmented ability to interact with biological membranes. The lack of surface charges resulted in colloidal instability, and DLS spectra revealed the presence of nano/micro aggregates with the average particle diameter of the CeO_2_ NP suspension in growth cell medium at 60 µg/mL measuring 116.6 nm.

The human colorectal adenocarcinoma-derived epithelial cell line HT-29 (ATCC^®^ HTB-38™, Manassas, VA, USA) was utilized as an in vitro model of the intestinal epithelium. Cells were cultured at 37 °C in a humidified atmosphere of 5% CO_2_/95% air, using McCoy’s 5A Modified Medium (ATCC^®^ 30-2007™, Manassas, VA, USA), supplemented with 10% (v/v) fetal bovine serum (FBS) and 1% antibiotic-antimycotic solution (penicillin/streptomycin/amphotericin B). Cells were subcultured weekly to maintain subconfluent growth.

For short-term exposure studies, cells were treated with CeO_2_ NP for 24 or 48 h. For sub-chronic exposure, cells previously treated for 24 h were subcultured in fresh CeO_2_ NP containing growth medium. The nanoceria suspension was replenished twice weekly with fresh growth medium at the established CeO_2_ NP concentration, subsequent to rinsing the monolayers with PBS to eliminate any residual particles. The cells were split weekly at a constant seeding density (i.e., 2 × 10^4^/cm^2^), continuing the treatment for a total of five weeks. At each time point (days 7, 14, 21, 28, and 35), aliquots of cells were collected to assess the xenobiotic’s effects. In parallel, control cells underwent identical treatment replacing CeO_2_ NP with PBS [34,35,36].

Additional experiments were conducted on cell progeny derived from the 28-day sub-chronic exposure group. One subculture of these cells was maintained for an additional 7 days in CeO_2_ NP-free medium to evaluate the potential for recovery of basal cellular conditions following the exposure to the xenobiotic stressor.

2.2. Cytotoxicity Assay and Cellular Internalization

The acute cytotoxic potential of CeO_2_ NP was assessed across a concentration range of 1.65–200 μg/mL using the colorimetric MTT assay, which measures mitochondrial metabolic activity via the reduction of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) by cellular dehydrogenases. The abiotic interference of CeO_2_ NP in MTT assays was assessed by absorbance measurements. These were performed in duplicate on both MTT and, following solubilization, in formazan salt produced by cellular dehydrogenases activity. In one series, CeO_2_ NP suspension was added, while in the other, it was substituted with PBS. The lack of variation in absorbance values at 540 nm ruled out nanoceria interference.

HT-29 cells were seeded in 96-well plates and exposed to CeO_2_ NP for 24 or 48 h. CeO_2_ NP working concentrations were prepared by diluting the 10 mg/mL stock suspension into the culture medium (100 µL per well). Following the exposure period, cell viability was assessed according to our validated protocol [34] through an absorbance measure at 540 nm using a microplate reader (Tecan, Milan, Italy). Dimethyl sulfoxide (10%) and PBS were used as positive and negative controls, respectively.

Considering that different endocytosis mechanisms are involved in particle internalization, the endocytic apparatus (late endosomes and lysosomes) was examined by employing metachromatic fluorophore Acridine Orange (AO). This dye emits red fluorescence when sequestered within acidic compartments, whereas it appears green in the cytosol and nucleus.

Therefore, by the red/green ratio, it detected both the internalization of foreign material and, by the loss of cytoplasmatic green fluorescence, the leakage of the acidic compartment [22]. The analyses were performed in semi-confluent HT29 monolayers grown in chamber slides and treated for 24 and 48 h at 37 °C with CeO_2_ NP suspensions at 25 and 100 µg/mL. Following medium removal and multiple PBS washes, the cells were incubated with an AO solution (5 µg/mL). CLSM observations were performed using the Leica TCS SP2 instrument (Leica Microsystems, Wetzlar, Germany), with Leica Confocal software (version 2.0) and a Leica DM IRB fluorescence microscope.

To quantify the acid compartment, the image-processing program Image (imagej.nih.gov/ij/index.html, accessed on 15 January 2025) was used to calculate the cellular area which emitted red fluorescence and its intensity. At least 100 cells were analyzed for each slide. In comparison to the negative controls (cells treated in free CeO_2_ medium), the percentage change (%Δ) in the product of area by emission intensity was taken into account.

2.3. Assessment of ROS Generation Following Acute and Sub-Chronic Exposure

To investigate both acute and sub-chronic effects on intestinal cells, HT-29 cell aliquots were exposed to 100 μg/mL of CeO_2_ NP for the time points specified in the experimental design. The fluorogenic probe 2′,7′-dichlorodihydrofluorescein diacetate (DCF-DA; Merck Life Science S.r.l., Milan, Italy) was used to measure the amount of ROS produced inside cells. After being exposed, cells were rinsed with PBS and incubated with a 1 µM DCF-DA solution prepared in PBS supplemented with 10 mM D-glucose (pH 7.4). Fluorescence was measured using a microplate reader (Tecan Italia, Milan, Italy) with excitation and emission wavelengths set at 485 nm and 535 nm, respectively, after 30 min at 37 °C. Results were expressed as %Δ relative to negative control cells.

2.4. Detection of ROS-Induced Effects

To assess CeO_2_ NP-induced mitochondrial dysfunction, changes in mitochondrial membrane potential (ΔΨm) were monitored using rhodamine 123 (R123; Invitrogen Molecular Probes, Eugene, OR, USA), a cationic fluorochrome that accumulates selectively in the matrix of intact mitochondria. Cell aliquots (CeO_2_NP-treated cells and PBS-treated cells) were incubated with 10 µM R123 for 10 min at 37 °C. Fluorescence was recorded, after PBS washing, at excitation/emission wavelengths of 535/595 nm using a microplate reader (Tecan Italia, Milan, Italy).

DNA damage and, in particular, single-strand breaks and base damage induced by CeO_2_ NP, were evaluated via the alkaline Comet assay through the performance of duplicate tests using 2 × 10^4^ cells per spot. Electrophoresis was carried out at 25 V and 300 mA (0.86 V/cm) for 30 min. After staining with ethidium bromide (20 µg/mL), samples were visualized using a DMIRB fluorescence microscope (Leica Microsystems Heidelberg, Mannaheim, Germany), equipped with a digital camera (Power Shot S50; Canon, Milan, Italy) at 400× magnification. Images of 100 randomly selected cells per slide were analyzed using the CASP (Comet Assay Software http://www.casp.sourceforge.net, accessed on 25 January 2026) image analysis tool. Percentage of DNA in the tail (%TDNA) was used to quantify DNA damage.

Additionally, oxidative DNA damage was evaluated by detecting 8-oxo-2′-deoxyguanosine (8-oxo-dG) via FITC-labeled avidin binding. This is possible because avidin exhibits high affinity for 8-oxo-dG due to structural similarity with biotin [37,38]. This fluorimetric qualitative assay, previously utilized by us [24], was employed to evaluate variations in emission levels compared to the negative control. Specificity of the assay was validated using positive controls treated with 300 µM H_2_O_2_.

Using an avidin–FITC conjugate, fluorimetric analyses were performed in control and treated cells permeabilized in methanol for 15 min at −20 °C. After PBS washes and rehydration, cells were then incubated for 1 h at 37 °C with 1 µM avidin–FITC in PBS containing 0.1% FBS. Fluorescence was measured at excitation/emission wavelengths of 485/535 nm.

Finally, Cytotoxic effects resulting from sub-chronic CeO_2_ NP exposure were evaluated by fluorimetric detection of cells detached from monolayers using the DNA intercalating dye propidium iodide (PI). Detached cells were incubated with 3 µg/mL PI for 5 min at 4 °C, and fluorescence was detected at 535 nm excitation and 615 nm emission wavelengths.

2.5. Gene Expression Analyses

The changes in the expression of ATG5, LC3, and BECN1, genes involved in phagocytosis/autophagy, and BAX and Bcl-2, genes involved in regulation of cell survival, were assayed by the Sybr Green Real-Time PCR using β-actin as an endogenous control.

Following the manufacturer’s instructions, TRIzol reagent (Life Technologies, Milan, Italy) was used to extract total RNA from treated and untreated (negative control) cell pellets. For each sample, 2 µg of isolated RNA were reverse-transcribed into cDNA using the High-Capacity cDNA Archive kit (Life Technologies). Subsequently, the expression levels of ATG5, LC3, BECN1, BAX, and Bcl-2 were quantified in triplicate. A 7900HT Fast Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) was used to perform quantitative real-time PCR (qPCR) in 10 µL reaction volumes with 1 × SYBR Green Master Mix (Life Technologies), 0.1 µM specific primers, and cDNA equivalent to 25 ng of total RNA. The primers used are reported in Table 1.

Amplification was carried out using the following conditions: 50 °C (2 min), 95 °C (10 min), and 40 cycles of 95 °C (15 s) and 60 °C (1 min). Primer specificity was verified through a post-amplification dissociation stage. β-Actin was utilized as the endogenous control for normalization (primer sequences provided in Table 1).

Data were analyzed using the 2^−ΔΔCt^ relative quantification method, and values are presented as fold changes relative to controls.

2.6. Statistical Analysis

All experimental results are expressed as mean ± standard deviation (SD) from a minimum of three independent biological replicates. Statistical analyses were performed using GraphPad Prism software (version 8.0, GraphPad Software, San Diego, CA, USA). Pearson correlation coefficients were calculated to evaluate the strength and direction of associations between variables. Comparisons between groups were conducted using unpaired Student’s t-tests. A p-value < 0.05 was considered statistically significant.

3. Results

3.1. Nanoceria Cytotoxicity and Cellular Internalization

The MTT test was performed on HT-29 monolayers at 24 and 48 h for the assessment of acute cytotoxicity of CeO_2_ NP. Nanoceria suspension, assayed in the range 1.65–200 µg/mL, had a moderate cytotoxicity (Figure 1A) which was positively related to the exposure dose and time (r 0.761 and 0.783, respectively, p < 0.05).

The metachromatic fluorophore AO for the CLSM observations was used to evaluate the endocytic apparatus in cells treated with nanoceria suspension (25 and 100 µg/mL) for 24 and 48 h (Figure 1B).

It is nevertheless important to underline that internalization into the phagolysosomal compartment does not rule out the possible presence of individual particles or aggregates of CeO_2_ in the cytosol, due to passive diffusion across the phospholipid bilayer.

The microscopic observations (Figure 1C) clearly highlighted the consistent nanoceria dose- and time-dependent internalization in mature endosomes that, at least up to 48 h in cells exposed to 25 µg/mL, maintained their integrity. However, at the highest dose, both at 24 and especially at 48 h, alongside an increase in the acidic compartment (i.e., a larger cell surface emitting in red), a decrease in green emission was also observed (%Δ −6 and −17, respectively), indicating partial cytoplasmic acidification attributable to damage to the phagolysosomal membranes with leakage of the engulfed material.

3.2. Nanoceria Effects After Acute and Sub-Chronic Exposure

Based on the cytotoxicity data, for the assessment of biological effects after acute and sub-chronic exposure, a concentration of 100 µg/mL of CeO_2_ NP was used. This, in order to learn more about the processes that cause cytotoxicity and to evaluate the homeostatic capabilities of our cell model, consequently, highlighting a possible adaptation of the cellular progeny sub-chronically exposed to a high dose of nanoceria.

The pro-oxidant effect of acute and sub-chronic nanoceria exposure is reported in Figure 2A. ROS levels, expressed as Δ% compared to the control cells, were already significantly higher after acute exposure (Δ% 111.2 p < 0.05). ROS overproduction continued to progressively increase in subsequent subcultures, reaching a maximum after exposure for 14 days and then moderately decreasing. On average, for all the times examined, ROS values were four-fold higher in comparison to control cells (p < 0.01). Despite the marked ROS decrease, the absence of nanoceria for 7 days was not sufficient to restore basal conditions observing a Δ% equal to 34.0 (p n.s, i.e., not significant) in this cell progeny.

To verify the xenobiotic’s capabilities to trigger mitochondrial impairment, transmembrane potential (ΔΨm) was measured. Figure 2B reports the observed values expressed as %Δ. Despite the high dose assayed, the acute exposure caused a negligible reduction in the ΔΨm (%Δ value 8.03 at 24 h). Increasing exposure time, the lowest levels of ΔΨm were observed at 7d when a more marked nanoceria-induced mitochondrial impairment was observed in cell progeny (%Δ value −34.8). Apparently, a clear tendency to restore mitochondrial function was observed for longer exposure times, highlighting a significant inverse association between exposure time and mitochondrial impairment (r = −0.99, p = 0.002). The trend was confirmed in post-exposure experiments, since in the cell progeny grown in the absence of the xenobiotic for 7 days, ΔΨm was superimposable to the control cells.

The association between ROS overproduction and mitochondrial dysfunction was confirmed by the trend of both variables. During the initial 14 days of exposure, the lowest mitochondrial potential values were seen in conjunction with the greatest amounts of reactive oxygen species (ROS). In later subcultures, the partial restoration of mitochondrial activity was associated with a better regulated overproduction of reactive oxygen species (ROS).

The pro-oxidant effects CeO_2_NP also caused DNA damage, as shown in Figure 2C that reports the results of alkaline Comet assay. The %Δ of TDNA after acute exposure underlined the genotoxic effect of the xenobiotic. The values tended to decrease prolonging exposure time. It seems reasonable to assume that in the subcultures exposed for 7, 14, and 21 days, the activation of repair mechanisms, although not entirely efficient, contained the genotoxic effects. Unlike what was observed for mitochondrial function, these cellular capabilities were transient, and further prolonged exposure showed saturation of the homeostatic mechanisms with % Δ values significantly higher than controls, demonstrating how DNA damage increases with exposure duration. DNA integrity was completely restored only in the cell progeny grown for seven days without CeO_2_NP.

As expected, the results of 8-oxo-2′-deoxyguanosine (8-oxo-dG), performed spectrophotometrically to assess oxidative DNA damage via FITC-labeled avidin binding, confirmed the acute and sub-chronic pro-oxidant effect at the nucleus level of CeO_2_NP. As shown in Figure 2C, the %Δ of 8-oxo-dG was strongly related to those of %TDNA (p < 0.01), highlighting the same trend. This confirmed that sub-chronic exposure to the toxic dose assayed overloaded the repair mechanisms, increasing the extent of genotoxic damage after prolonged exposure time.

The cytotoxicity elicited by sub-chronic nanoceria exposure was quantified spectrofluorimetrically by analyzing propidium iodide (PI)-stained cells detached from the monolayers. Figure 3 illustrates the comparative results between acute and sub-chronic treatments. Consistent with previous MTT assay findings, acute exposure resulted in a 1.5-fold increase in the percentage of detached cells relative to the control group. After sub-chronic exposure, nanoceria caused a steady rise in cellular mortality with increasingly higher values as exposure times lengthen (r = 0.900; p = 0.015). In comparison to control, the % of death cells was on average three-fold higher (p = 0.011), highlighting the irreversible cell damage due to sub-chronic exposure. According to the Pearson test, a very significant correlation was observed between DNA damage—expressed both as %TDNA and more specifically as the presence of 8-oxo-dG—and the % of cell mortality (p < 0.001).

However, the absence of nanoceria in the cell progeny restored basal levels of cell mortality.

During the sub-chronic exposure, the weekly cell replication index (WCRI) was also evaluated. Due to the high dose assayed, in comparison to control cells, the nanoceria exposure caused a marked decrease in the WCRI value (5.68 vs. 7.41), recording on average a 23.3% drop. Given the high rate of cell death, the decrease in WCRI is most likely due to fewer cells that are able to divide rather than a slowdown in replication.

3.3. Gene Expression Regulation

The acute and sub-chronic effects of CeO_2_ NP exposure were also assessed by studying the changes in the expression of some key genes in homeostatic mechanisms, such as the ones involved in phagocytosis/autophagy pathways (ATG5, LC3 and BECN1) and regulation of cell survival (BAX and Bcl-2). This is in order to assess the cells’ ability to contain damage after exposure to the toxic dose of CeO_2_ NP without triggering cascade processes involving cells not directly exposed. The effects of acute and sub-chronic exposure in the expression of the key genes involved in phagocytosis/autophagy pathways are reported in Figure 4A. As clearly shown, all genes were overexpressed after both acute and sub-chronic exposure, and the mRNA levels were positively strongly related (p < 0.01). On average the mRNA levels increased by 70% after 24 h of exposure, and then further increased, reaching peak expression after 7 days, when mRNA levels for ATG5 and LC3 were at least three times higher compared to controls. By extending the observation time, expression levels were inversely related with exposure time, although ATG5 and LC3 remained significantly higher than controls. Instead, BECN1 expression overlapped with the controls over longer periods.

The expression of the genes that regulate death/survival processes is reported in Figure 4B. The acute effects of nanoceria exposure appeared with the overexpression of BAX, with a 62% increase in mRNA, accompanied by a slight under expression of BCL2 (BAX/BCL2: 1.88). Prolonging the exposure time, the pro-apoptotic signal, given by the overexpression of BAX, further increased, reaching its peak at 7 days when the BAX/BCL2 ratio was equal to 7.2. Subsequently, the expression of the two genes normalized, and no significant differences were observed compared to the controls. Based on this, the BAX/Bcl2 ratio of mRNA implicitly validated that the pro-apoptotic effect of CeO_2_ NP was limited to the early subcultures. Consequently, by not always allowing the programmed death process, a more massive damage, with the involvement of cells not directly exposed to the xenobiotic, could be hypothesized. Moreover, the expression of BAX was positively related to the one of ATG5, LC3, and BECN1, highlighting that these homeostatic mechanisms were co-expressed during some phases of the exposure to contain the CeO_2_-induced damage.

4. Discussion

To improve our understanding of the potential health effects on humans of ingested nanoceria, we used an in vitro model resembling both acute and sub-chronic exposure. We think our method is especially helpful because all persistent environmental contaminants cause long-term exposure.

However, pollution levels of this emerging xenobiotic are poorly known, and in this preliminary study, to better highlight the underlying mechanisms of nanoceria pathogenesis, we focused exclusively on an exposure dose exhibiting a non-negligible cytotoxicity in our cell model. It is likely that the assayed dose does not allow us to extrapolate the results obtained to the general population. CeO_2_ fate after being mobilized is even less understood, although recent data demonstrate its bioavailability and the plant’s capacity to sequester nanoceria in their tissues [39,40,41]. It is still unknown if this results in bioaccumulation, while biomagnification in the trophic chain is improbable given the efficient fecal clearance observed in vivo studies [13].

The results highlighted the CeO_2_ NP internalization into the phagolysosomal compartment of colon cells, already widely demonstrated on other human cell models for engineered nanoparticles and for nanoplastics as well [22,35]. This mechanism of nanoceria, through phagocytosis, was previously confirmed in other non-professional phagocytic cells (i.e., human keratinocytes and lung epithelial cells), and it happens through clathrin- and/or caveolae-mediated endocytosis [42,43,44]. However, as previously stated, single nanoceria particles or nano/micro aggregates could also diffuse directly through the membrane [5]. Both these factors, playing a key role in the internalization process, regulate the balance between endocytosis and passive diffusion. As previously reported [30], the pronounced hydrophobicity of nanoceria, resulting from the lack of surface charges, augmented its ability to interact with biological membranes while simultaneously contributing to colloidal instability, thereby promoting particle aggregation. A limit of our study was that we did not evaluate the possible adsorption of medium proteins onto the surface of the nanoparticles and nano/micro aggregates, leading to the formation of protein corona, which, already observed for nanoplastics, hinders aggregation [45,46,47,48]. Even if phagocytosis of particles is the major mechanism of uptake of insoluble materials and, at the same mass concentration, it is directly related to particle size, damage to the phagolysosomal compartment can subsequently cause the leakage of particles that, in their free form, spread into different cellular compartments. This was demonstrated by the partial acidification of the cytosol and was observed primarily at 48 h, when a further increase in the acidic compartment was observed, demonstrating the simultaneity between uptake and damage to the endocytic apparatus.

The strong CeO_2_ NP-induced prooxidant effect was observed both after acute and sub-chronic exposure, assessed by fluorometric detection of all intracellular ROS. Up to the second cell progeny exposed to nanoceria, the rise in ROS was time-dependent while, only in subsequent subcultures, the cells’ ability to reduce, even if partially, ROS overproduction was observed.

The prolonged oxidative stress caused irreversible redox imbalance as confirmed in the cell progeny grown in the absence of the xenobiotic. As expected, ROS overproduction triggered mitochondrial impairment and DNA oxidative damage. The observed trend between ROS overproduction and mitochondrial dysfunction highlights how the pro-oxidant effect of CeO_2_ NPs triggers a vicious cycle, characterized by further overproduction of ROS from damaged organelles [34]. Nonetheless, the function of mitochondria in the redox imbalance was temporary, owing to the activation of the homeostatic process of autophagy. This appears to be corroborated by the more reduced levels of ROS and the partial recovery of mitochondrial function observed in the later stages of exposure. Furthermore, mitophagy allowed for complete recovery of basal mitochondrial function in the progeny of exposed cells, grown without nanoceria.

Conversely to this homeostatic pathway triggered by sub-chronic exposure to nanoceria, HT-29 cells do not appear to be able to initiate an efficient enzymatic DNA repair pathway, which for oxidative damage occurs mainly by Base Excision Repair (BER) [49].

The rise in cell mortality seen at various times during sub-chronic exposure, which, in turn, is responsible for the marked reduction in WCRI, is certainly caused by this steady increase in DNA oxidative damage, as highlighted by the significant positive correlation.

Although only indirectly assessed through the assessment of gene expression, without quantifying the synthesized proteins, the results of canonical toxicological tests matched the one obtained by molecular tests. The overexpression of the main genes involved in the phagocytosis/autophagy pathway confirmed, as stated above, its ability to restrict mitochondrial impairment, even after prolonged exposure. This homeostatic pathway is finely orchestrated by several proteins. Among these, LC3 is required to the formation and maturation of the autophagosome [50,51], ATG5 is essential in autophagosome biogenesis for the expansion and elongation stage [52,53], while BECN1/Beclin 1 controls the production of the endocytic vesicle’s double membrane, which will contain the damaged molecules or organelles [54,55]. In particular, nanoceria exposure caused, throughout the examined period, a marked overexpression of the genes encoding these proteins, with the exception of BECN1/Beclin, whose mRNA levels as early as the third subculture were, on average, only 15% higher than the negative control.

Instead, the pro-apoptotic gene BAX was overexpressed for a much shorter period of time, indicating that prolonged exposure might cause more widespread adverse effects also involving cells that are not directly exposed to nanoceria.

5. Conclusions

Overall, the different biological effects documented in human colonic cells shed light on the potential health risks posed by exposure to this emerging environmental pollutant. Since exposure triggered homeostatic pathways that might be sufficient for lower exposure doses, further studies are certainly needed for a more in-depth risk assessment.

Based on previous examples of the unregulated discharge of persistent environmental contaminants, both natural and synthetic, in-depth research is essential to mitigate the impact of CeO_2_ nanoparticles on human health and ecosystems, especially considering the expected rise in industrial applications and the widespread future utilization of nanoceria.