2.1 Materials

Cerium nitrate hexahydrate (Ce(NO₃)₃·6H₂O), 3-amino-propyl-triethoxysilane (APTES), ethanol (C₂H₅OH), methanol (CH₃OH), human epidermal growth factor (hEGF), fluorescein-5-isothiocyanate (FITC), cell counting kit 8 (CCK-8), 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA), sodium dihydrogen phosphate (NaH₂PO₄), disodium hydrogen phosphate (Na₂HPO₄), RIPA buffer, immobilon-P PVDF membrane, bovine serum albumin (BSA), donkey antirabbit IgG antibody, glutaraldehyde (C₅H₈O₂, Electron Microscopy Sciences, USA), procure 812 resin (Electron Microscopy Sciences, USA), lead citrate (ProSciTech, Australia), uranyl acetate (Electron Microscopy Sciences, USA), osmium tetroxide (OsO₄, Electron Microscopy Sciences, USA), triethanolamine (TEOA, reagent grade, 99 wt%), 1.0 mM P-benzoquinone (BQ, reagent grade, 98 wt%) and 75 mM isopropyl alcohol (IPA, reagent grade, 99.7 wt%) were purchased from Sigma-Aldrich, Australia. AntiEGFR phospho Y1068 antibody was purchased from Abcam, Australia and PD-MidiTrap G-25 was purchased from Cytiva, Australia. Sodium hydroxide (NaOH), ammonium chloride (NH₄Cl), sodium carbonate (Na₂CO₃), and sodium bicarbonate (NaHCO₃) were purchased from Chem-Supply, Australia. Nitric acid was purchased from RCI Labscan Limited, Australia. Protease inhibitor cocktail was purchased from Roche Diagnostics, Australia and phosphatase inhibitor, EGF Receptor rabbit mAb antibody and β-Actin rabbit mAb antibody were purchased from Cell Signaling Technology, USA. RPMI 1640 medium, DMEM medium, penicillin–streptomycin solution, L-glutamine, Trypsin–EDTA, Dulbecco's phosphate buffered saline (DPBS), Hank's balanced salt solution (HBSS), fetal bovine serum (FBS), Hoechst 33342 solution, 16% w/v paraformaldehyde (PFA; methanol-free), micro BCA protein assay kit, NuPAGE 4%–12% bis-tris gels, and SuperSignal West Pico PLUS Chemiluminescent Substrate were purchased from Thermo Fischer Scientific, Australia. HT-1080 and MRC-5 cell lines were purchased from American Type Cell Culture (ATCC), USA.

2.2 Synthesis of nanoceria

Rod-shaped cerium oxide nanoparticles were synthesized using a previously described hydrothermal technique.⁵ In the synthesis process, 8.6844 g of Ce (NO₃)₃·6H₂O and 24 g of NaOH were added to a 100 mL Pyrex beaker to achieve concentrations of 0.5 M and 15 M, respectively, in 40 mL deionized (DI) water. The solution was stirred for 30 min at room temperature using a magnetic stirrer. The solution was transferred to a 50 mL Teflon-lined autoclave reactor for hydrothermal reaction at 60°C for 48 h (heating rate 5°C/min). After synthesis, the nanoparticles were separated from the medium by centrifugation at 5000 rpm (2935 × g) for 20 min and decanting. The nanoparticles then were cleaned by the following steps applied repeatedly 16–18 times by alternate resuspension in 25 mL of DI water and 25 mL ethanol, sonication for 10 min, centrifugation at 5000 rpm (2935 × g) for 20 min, and decanting. This extent of washing was required for complete removal of Na. The nanoparticles finally were transferred to an opaque glass container, air-dried at 60°C for 12 h in an oven, and readied for long-term storage by sealing with a screw-cap lid.

2.3 Functionalization of nanoceria

Nanoceria was coated using APTES as a linker for the conjugation of EGF/EGF-FITC. Silanization was performed by dropwise addition of 1.5 mL APTES to 100 mg of nanoceria suspended in 25 mL of ethanol, sonication for 10 min, and magnetic stirring for 24 h, as described elsewhere.⁵⁴ The APTES-coated nanoparticles were cleaned by the following steps applied repeatedly 5 times by resuspension in 25 mL ethanol, sonication for 10 min, centrifugation at 5000 rpm (2935 × g) for 20 min, and decanting. The nanoparticles finally were transferred to a clear glass container, air-dried at 60°C for 12 h in an oven, and readied for short-term storage by sealing with a screw-cap lid.

The conjugation of EGF onto the APTES-modified nanoparticles was done according to the variant of a protocol described elsewhere.⁵⁵ This which involves the mixing of APTES-coated nanoparticles with the peptide in the presence of 10 vol% ethanol. This allows the binding of the carboxyl groups of the peptide with the amine groups of APTES through covalent attachment.⁵⁵ 100 mg of APTES-coated nanoceria were suspended in 10 vol% ethanol in DPBS, which were conjugated by the addition of 50 μg of EGF in order to achieve an EGF working concentration of 16 nM. The suspension was mixed using a rotary tube mixer (WiseMix RT-10, Korea) at 20 rpm at 4°C for 14–16 h. After functionalization, the EGF-functionalized nanoparticles were separated from the medium by ultracentrifugation (Optima XPN Beckman-Coulter, USA, SW 41 Ti swinging-bucket rotor) at 30,000 rpm (154,000 × g) for 10 min at 4°C and decanting. Subsequent purification involved the following steps applied twice by resuspension in variable amounts of 10 vol% ethanol in DPBS according to the nanoparticle amount, ultracentrifugation at 30,000 rpm for 10 min at 4°C, and decanting. The particles obtained after purification were resuspended in Milli-Q water, lyophilized (Christ Alpha 1–4 LDplus, Germany; −54°C, 1–4 × 10⁻⁶ kPa) and stored at −20°C for further use. Since the potential desorption of proteins from nanoparticles happen in the liquid state, it is assumed that the desorption does not occur during the period in which the materials are frozen.

FITC-labeled EGF was obtained according to a variant of a process described elsewhere⁵⁶ and then conjugated to APTES-nanoceria for confocal imaging studies. EGF was labeled with FITC by mixing FITC and EGF in 1:10 weight ratio in a solution of 0.1 M carbonate buffer at pH 9.0 using a rotary tube mixer at 20 rpm at 4°C under dark conditions for 14–16 h. This was followed by the addition of 1 M NH₄Cl in order to inactivate the residual FITC. The FITC-labeled protein was purified using a Sephadex G-25 column and stored at −20°C for long-term use. For subsequent conjugation to the APTES-nanoceria, the immediately preceding process was used.

2.4 Characterization of nanoparticles

The morphological characteristics (size and shape) of nanoparticles were analyzed using transmission electron microscopy (TEM; FEI Tecnai G2 20 TEM, USA; 200 kV). The nanoparticles (⁓0.5 mg) were dispersed in 0.5 mL of methanol, sonicated for 3 min, loaded on copper grids, and dried at 37°C for 24 h before imaging. The mineralogy of nanoceria was determined by x-ray diffraction (XRD; Philips X'Pert Multipurpose x-ray Diffractometer (MPD), Netherlands; CuKα, step size 0.02° 2θ, scanning speed 5.5° 2θ min⁻¹) and laser Raman microspectroscopy (Renishaw inVia Raman microscope, UK; 35 mW, helium-neon green laser, 514 nm). The chemical bonding natures of the nanoceria, APTES-nanoceria, and EGF-nanoceria were assessed using Fourier transform infrared spectroscopy (FTIR) with attenuated total reflectance (ATR) (PerkinElmer Spectrum Two FT-IR spectrometer, USA; resolution 4 cm⁻¹) and thermogravimetric analysis (TGA; NETZSCH STA 449F1, Germany; temperature range 25 to 500°C in air, 200 mL/min flow rate, aluminium crucible). The surface chemistry of the nanoceria, APTES-nanoceria, and EGF-nanoceria was examined using x-ray photoelectron spectroscopy (XPS; Thermo Fisher Scientific, ESCALAB 250Xi spectrometer, UK; AlKα, 13.8 kV, 8.7 mA, spot size 500 μm). For the suspensions of APTES-nanoceria and EGF-nanoceria, the hydrodynamic size distributions and the zeta potentials were measured using dynamic light scattering (DLS; Malvern Zetasizer Nano ZS, UK; 10 mW, 633 nm He-Ne laser). Although the suspensions were sonicated for 3 min prior to measurement, the nanoparticles were agglomerated. The photocatalytic nature of nanoceria was analyzed using the methylene blue (MB) degradation assay.⁵⁷ 8 mg of nanoparticles were added to 8 mL of MB dye solution in a 50 mL Pyrex beaker, followed by adding 10.0 mM concentrations of triethanolamine, 1.0 mM P-benzoquinone, and 75 mM isopropyl alcohol. The testing was done using magnetic stirring at 300 rpm for 15 min while in the dark in a light-obstructed black box. The sample was exposed to 365 nm UV irradiation from a UV lamp (8 W, 3UV-38, UVP) while being magnetically stirred at 300 rpm for different times up to 1 h in the black box. The UV–Vis testing was done on the solution following nanoparticle removal by centrifugation.

2.5 Cell culture

Human fibrosarcoma cells (HT1080) were cultured in RPMI medium, supplemented with 10 vol% FBS, 1 vol% penicillin–streptomycin, and 2.0 mM L-glutamine, and incubated at 37°C under 5 vol% CO₂ atmosphere. The human fibroblast cell line (MRC-5) was cultured in DMEM medium, supplemented with 10 vol% FBS, 1 vol% penicillin–streptomycin, and 2.0 mM L-glutamine, and maintained at 37°C under 5 vol% CO₂ atmosphere.

2.6 Dose-dependent cytotoxicity of nanoceria

The dose-dependent cytotoxicity of nanoceria on fibrosarcoma cells was assessed by a cytotoxicity assay using the CCK-8. CCK-8 assay is a standard test^58-61 used to measure cell viability based on the activity of the dehydrogenase enzyme present in the cells, which reduces the pink CCK-8 reagent WST-8 [2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4disulfophenyl)-2H-tetrazolium, monosodium salt] to the orange formazan reaction product.⁶² HT1080 cells were seeded in a 96-well plate at a seeding density of 1000 cells/well with 100 μL RPMI per well and incubated for 24 h. The medium in each well was replaced with different concentrations of APTES-nanoceria suspensions prepared in RPMI medium, ranging from 100 to 500 μg/mL (100, 200, 300, 400, and 500 μg/mL). The nanoparticle suspensions were ultrasonicated for 3 min prior to addition to the cells so as to disperse the nanoparticle agglomerates. The success of this de-agglomeration process was shown by the absence of sedimentation over the following 3–5 min. The medium in the control wells were replaced simultaneously with fresh RPMI medium. Six wells were used for each concentration. The plates then were incubated for 48 h. After incubation, 10 μL of CCK-8 reagent were added to each well and incubated for 3 h. RPMI (100 μL) mixed with CCK-8 reagent (10 μL) were used as the blank. The optical absorbance at 450 nm was determined using a microplate reader (CLARIOstar plus, BMG LABTECH, Germany). Nanoceria control wells were employed in the present work in order to determine the absorbance of nanoparticles alone at 450 nm; any absorbance from the nanoparticles was deducted from the absorbance values obtained for the test wells before calculating the percent cell viability.

2.7 Cell viability assay

HT1080 cells were seeded in three 96-well plates at a seeding density of 1000 cells/well and incubated for 24 h. The medium was removed and replaced with 100 μL of fresh RPMI in the control wells and with 100 μL of functionalized/nonfunctionalized nanoceria in the test wells at 200 μg/mL concentration. The nanoparticle suspensions were ultrasonicated for 3 min prior to addition to the cells so as to disperse the nanoparticle agglomerates. The success of this de-agglomeration process was shown by the absence of sedimentation over the following 3–5 min. Six wells per test sample were used. The plates then were incubated for 24, 48, or 72 h. After the respective incubation periods, CCK-8 reagent (10 μL) was added to each well, incubated for 3 h, and the optical absorbance at 450 nm was measured.

2.8 Testing the effect of cell debris on the active sites of nanoceria

In order to generate cell debris, HT1080 cells were seeded in a 96-well plate at a seeding density of 4000 cells/well and incubated for 24 h. The medium was removed and replaced with 100 μL of 1X DPBS and the plate was incubated for a further 4 h in order to kill the cells. The cell debris were collected and diluted with RPMI. In order to perform the cell viability assay, HT1080 cells were seeded in three 96-well plates at a seeding density of 4000 cells/well and incubated for 24 h. The medium was removed and replaced with 100 μL of fresh RPMI in the control wells and with 100 μL of nanoceria (non-functionalized) in six test wells at 200 μg/mL concentration. 50 μL of 400 μg/mL nanoceria and 50 μL of cell debris suspension was added to the remaining six test wells in order to make a final concentration of 200 μg/mL. The nanoparticle suspensions were ultrasonicated for 3 min prior to addition to the cells so as to disperse the nanoparticle agglomerates. Six wells per test sample were used. The plates then were incubated for 48 h. After the incubation time, CCK-8 reagent (10 μL) was added to each well, incubated for 3 h, and the optical absorbance at 450 nm was measured.

2.9 Measurement of cellular ROS

Cellular ROS generation levels in cancer cells were measured using the DCFDA assay as described elsewhere^{7, 52} with a few modifications. HT1080 cells were seeded as described immediately above and treated with 200 μg/mL of functionalized or nonfunctionalized nanoceria and incubated for 24, 48, or 72 h. The nanoparticle suspensions were ultrasonicated for 3 min prior to addition to the cells so as to disperse the nanoparticle agglomerates. The success of this de-agglomeration process was shown by the absence of sedimentation over the following 3–5 min. After the respective incubation periods, the cells were washed with 1X DPBS and then 1X trypsin (100 μL) was added to each well. The plate was incubated for 5 min and then 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) in HBSS (10 μM) were added in order to achieve a final concentration of 5 μM, followed by incubation for 40 min. The fluorescence intensities at 485 and 535 nm then were determined.

2.10 Quantification of cellular uptake of nanoceria

ICP-MS is a highly sensitive technique employed in the quantification of cellular uptake of metal or metal oxide nanoparticles⁶³ and thus was used to quantify the cellular uptake of nanoceria. HT1080 cells were seeded in 24-well plates at a seeding density of 4000 cells/well and incubated for 24 h. The RPMI in the wells was replaced with 200 μg/mL of functionalized or nonfunctionalized nanoceria and incubated for 24, 48, or 72 h. The nanoparticle suspensions were ultrasonicated for 3 min prior to addition to the cells so as to disperse the nanoparticle agglomerates. The success of this de-agglomeration process was shown by the absence of sedimentation over the following 3–5 min. Six wells per test sample were used. After the respective incubation periods, the nanoceria suspension was discarded and the cells were washed thrice with 1X DPBS. Trypsin (500 μL) was added and the plates incubated for 5 min. Trypsinized cells then were digested completely using concentrated nitric acid. The cerium content in each sample was quantified using inductively coupled plasma mass spectrometry (ICP-MS; PerkinElmer quadrupole NexION, USA). Untreated cells were used as controls for quantification.

2.11 Western blotting tests

HT1080 cells were seeded at a seeding density of 0.5 × 10⁶ cells per well in 6-well plates. Three such plates were seeded in order to study the receptor stimulation after 5, 30, or 60 min. The RPMI in the wells was replaced with serum-free RPMI and the plates were incubated for 14–16 h. The cells then were treated with serum-free RPMI with or without 200 μg/mL EGF-nanoceria and incubated for 5, 30, or 60 min. The nanoparticle suspension was ultrasonicated for 3 min prior to addition to the cells so as to disperse the nanoparticle agglomerates. The success of this de-agglomeration process was shown by the absence of sedimentation over the following 3–5 min. The medium was removed and the cells were washed thrice with cold DPBS before being lysed at 4°C with lysis buffer containing protease inhibitor cocktail and phosphatase inhibitor. The protein concentration was quantified for each sample using the Micro BCA protein assay kit. Equal amounts of protein were loaded onto 4–12 wt% bis-tris gels and transferred to PVDF membranes following electrophoresis. The membranes were blocked with gentle agitation for 1 h using 5% (w/v) BSA in TBST (tris-buffered saline with tween-20) and incubated for 14–16 h with primary antibodies against phosphorylated EGFR (phospho Y1068), EGFR and ß-actin. The membranes Then were washed with TBST for three 5-min intervals and incubated (1 h at 25°C with gentle agitation) with donkey antirabbit IgG antibody conjugated with horseradish peroxidase (HRP). The membranes were washed with TBST for three 5-min intervals and visualized with enhanced chemiluminescence reagents (ECL) using an ImageQuant LAS 4000 Biomolecular Imager (GE Healthcare, UK).

2.12 Imaging of cellular uptake of nanoceria using transmission electron microscopy

HT1080 cells were seeded in 24-well plates at a seeding density of 10,000 cells/well, with sterilized cover slips placed inside each well. Following incubation for 24 h, the RPMI medium in the wells was discarded and the cells were treated with functionalized or nonfunctionalized nanoceria at a concentration of 200 μg/mL and incubated for 1, 2, 5, or 30 min. The nanoparticle suspensions were ultrasonicated for 3 min prior to addition to the cells so as to disperse the nanoparticle agglomerates. The success of this de-agglomeration process was shown by the absence of sedimentation over the following 3–5 min. The nanoparticle suspension was discarded from each well and the cells were washed thrice with 1X DPBS. The cells were fixed overnight at 4°C in a fixative comprised of 2.5% (w/v) glutaraldehyde in 0.2 M sodium phosphate buffer (pH 7.4). Fixed cells were rinsed with 0.1 M sodium phosphate buffer and post-fixed in 1% (w/v) osmium tetroxide in 0.2 M sodium phosphate buffer using a BioWave Pro+ Microwave Tissue Processor (Ted Pella, USA). After rinsing with 0.1 M sodium phosphate buffer, the samples were dehydrated in a graded ethanol series (50, 70, 80, and 90 vol% ethanol OH for 10 min each followed by 100 vol% ethanol twice for 10 min each). They then were infiltrated with resin (Procure 812, Electron Microscopy Sciences, USA) and polymerized in an oven at 60°C for 48 h. Ultrathin sections (70 nm) were cut using a diamond knife (Diatome, Switzerland) and collected onto carbon-coated copper-slot TEM grids. The grids were post-stained using uranyl acetate (2% w/v) and lead citrate. Two grids for each sample were imaged by TEM (JEOL 1400, Japan; 100 kV).

2.13 Imaging of cellular uptake of nanoceria using confocal microscopy

For confocal imaging studies, HT1080 cells were seeded in 12-well glass-bottom plates at a seeding density of 40,000 cells/well and incubated for 24 h. Triplicate wells were used for each test. After 24 h of incubation, the RPMI in the wells was replaced with 200 μg/mL FITC-tagged EGF-conjugated nanoceria and the plates were incubated for 5 min, 30 min, or 5 h. The nanoparticle suspension was ultrasonicated for 3 min prior to addition to the cells so as to disperse the nanoparticle agglomerates. The success of this de-agglomeration process was shown by the absence of sedimentation over the following 3–5 min. After incubation, the cells were washed three times with 500 μL of 1X DPBS and stained with 500 μL of Hoechst 33342 DNA stain which was diluted in the ratio 1:1000 using RPMI. The cells were washed with 500 μL of DPBS and fixed with 500 μL of 4% (w/v) PFA solution. After fixation, the cells were washed thrice with 500 μL of DPBS. The cells were imaged using confocal laser scanning microscopy (Zeiss LSM 800, Germany) using a Plan-Apochromat 20X/0.8 M27 and excitation/emission wavelengths were at 488/400–700 nm/1.13 AU for FITC and 405/400–510 nm/1.21 AU for Hoechst with a Notch filter applied for both channels at the respective excitation wavelengths. High resolution images were obtained using a Plan-Apochromat 63X/1.40 Oil DIC M27 and excitation/emission wavelengths were at 488/400–700 nm/0.82 AU for FITC and 405/400–510 nm/0.88 AU for Hoechst with a Notch filter applied for both channels at the respective excitation wavelengths.

2.14 Biocompatibility of nanoceria (cell viability assay – cancer cells vs. normal cells)

HT1080 cells and MRC-5 cells were seeded in 96-well plates and treated with APTES-nanoceria suspension as described in the cell viability assay protocol. RPMI medium (HT1080 cells) and DMEM medium (MRC-5 cells) were used for the seeding and the preparation of the nanoparticle suspensions. The nanoparticle suspensions in the wells were discarded after incubation for 24 h and replaced with fresh medium every 24 h. Untreated cells were used as controls. After incubation for 24, 48, or 72 h, CCK-8 reagent was added and the optical absorbance at 450 nm was measured.

2.15 Statistical analysis

All statistical analyses were performed using GraphPad Prism software and the data are represented in terms of the mean and the standard error of the mean (SEM; error bars). One-way analysis of variance (ANOVA) or two-way analysis of variance was used for the analysis; P values that were less than 0.05 were considered statistically significant: *p ≤ .05, **p ≤ .01, ***p ≤ .001, and ****p ≤ .0001.

3.1 Fabrication, modification, and characterization of nanoparticles

TEM direct imaging of hydrothermally synthesized nanoceria revealed a rod-shaped morphology and extensive agglomeration. The differential shadowing of overlapping nanorods as shown in Figure S1 (Data S1) indicates the general absence of intergrowths, so the nanorods are monodisperse and soft agglomerates formed only during mounting using methanol. A summary of the average dimensions of the nanoparticles determined from TEM images is given in Table 1.

The spread of these data is such that no conclusion concerning the probable locations of the conjugated species can be made.

The mineralogical analyses by XRD and laser Raman microspectroscopy in Figure S2 (Data S1) confirm that hydrothermal synthesis yielded single-phase CeO_2,^64-67 which crystallizes in the face-centered cubic fluorite structure.⁶⁸ The Raman data reveal the presence of a detectable concentration of oxygen vacancies [ ${\mathrm{V}}_{\mathrm{O}}^{••}$], as shown by the peak at 590 cm⁻¹.

The XPS analyses results and the FTIR spectra data of the nanoparticles are explained in the Data S1 (Figures S3 and S4, and Tables S1 and S2). Coating with APTES and functionalization with EGF was further confirmed using TGA analysis as shown in Figure S5 and explained in the Data S1. The photocatalytic property of nanoceria was also tested using the methylene blue degradation assay as shown in Figure S6.

The DLS analyses shown in Figure S7 and Table S3 illustrate the hydrodynamic diameters, which were interpolated visually from the curves, and the zeta potentials of the nonfunctionalized (APTES-nanoceria) and functionalized (EGF-nanoceria) materials. Although the ceria morphology is nanorod, Figure 6 indicates that the morphology of the agglomerates is spherical.

The average hydrodynamic diameters of APTES-nanoceria and EGF-nanoceria were calculated from the digital output to be 134.3 ± 10.1 nm and 165.2 ± 31.7 nm, respectively. The opposite surface charges of these as well as the increase in average hydrodynamic diameter is attributed to the functionalization by the negatively charged EGF protein⁶⁹ onto the surfaces of the APTES-nanoceria. The tendency to agglomerate is consistent with these zeta potentials, which are less than the well-known threshold of <30 mV for agglomeration.⁷⁰

3.2 Dose-dependent cytotoxicity of nanoceria on fibrosarcoma cells

In order to evaluate the impact of nanoceria dosage on the viability of tumor cells and to determine the optimal nanoceria concentration that can be employed for targeting, HT-1080 cells were exposed to APTES-coated nanoceria (nonfunctionalized nanoceria) of different concentrations varying in the range 100–500 μg/mL for 48 h. Figure 1 shows that the viability of HT-1080 cells decreased logarithmically with increasing nanoceria concentration, where the minimal dose that can induce ~50% cell death (IC50 value) was in the range 200–300 μg/mL. Since visualization of the cells was significantly occluded at the higher concentration of 300 μg/mL (53.1% cell death), the lower concentration of 200 μg/mL (43.5% cell death) was used for all further testing.

3.3 Enhanced anticancer activity of EGF-functionalized nanoceria

The influence of EGF functionalization on the ability of nanoceria to kill cancer cells was determined by performing a cell viability assay after treating the HT-1080 cells with nanoceria or EGF-nanoceria for 24, 48, and 72 h. The cell viability results of Figure 2 show that the cytotoxicity of nanoceria toward fibrosarcoma cells at all-time points was superior for the EGF-nanoceria treated cells compared to the APTES-nanoceria treated cells. However, the cytotoxicity decreased with increasing time. This is attributed to the blockage of the catalytically active sites,^71-74 which generally are considered to be oxygen vacancies.^{75, 76} This blockage is likely to have been caused by the organic cell debris created during necrosis and apoptosis.^{77, 78} This conclusion is supported by the XPS analyses, which show the unavoidable adventitious carbon that adsorbs on the surfaces of solids,⁷⁹ Figure S3 (Data S1) confirms the gradual adsorption of carbonaceous species while Figures S4a and S4b confirm the presence of numerous organic species. The effect of cell debris on the activity of nanoceria was confirmed further by performing a cell viability assay using nanoceria in the presence or absence of cell debris, as shown in Figure S8. The data suggest a decrease in the cytotoxicity of nanoceria with the presence of cell debris, as can be seen by the decreased cell death induced by the nanoceria in the samples where cell debris were present.

Further, the trends for both APTES-nanoceria and EGF-nanoceria are exponential with time but the rate of decrease in cytotoxicity is greater for the former. As these data show both extents of cell death and the generation of debris, the absence of identical trends indicates that the nature of the cell fragmentation upon death is different for each type of nanoceria. That is, APTES-nanoceria appears to generate either more debris or debris that contain ionic species more likely to be adsorbed in the oxygen vacancies. Further, the decrease in cell viability for EGF-nanoceria treated cells in comparison with APTES-nanoceria treated cells was observed consistently at all the time points studied. However, viability percentage of treated cells shows a decreasing trend with respect to increasing time points. This might be due to the saturation level of ROS generation already attaining around 48 h.

Another potential factor affecting the performance could be the change in pH caused by (1) cell death and release of debris and/or (2) cell division. In the former case, this would be likely to decrease the pH owing to the lower pH in the TME. In the latter case, the pH again would be likely to decrease as the metabolites released during this process are known to be acidic.⁸⁰ Since the production of the ROS that are responsible for cancer cell apoptosis increases in acidic environments,⁸¹ then this should decrease the cell viability with time, which is opposite of the observations.

Finally, a third type of potential factor is the potential recovery of the cancer cells.⁸² However, recovery is enabled by the use of inadequate dosages of drugs and the present work was done using a concentration approximating the IC50 value. As this cannot be considered to be inadequate, then it is unlikely that the data reflect this effect.

This discussion indicates that EGF as a biocompatible and stable functionalizing agent represents an advantageous strategy for targeted delivery of nanoparticles and subsequent cancer therapy.

Figure 4 shows the HT-1080 cellular uptake levels of nonfunctionalized and functionalized nanoceria. The results demonstrate no significant enhancement in the uptake of EGF-functionalized nanoceria at 24 h although the two longer time points show significant enhancement. The HT-1080 cells appear to have reached the maximal uptake of EGF-nanoceria after 48 h, as indicated by the equivalence of the data. In contrast, the uptake of APTES-nanoceria is exponential, which indicates the slower uptake owing to passive delivery whereas the EGF-nanoceria uptake is accelerated by active delivery owing to the presence of the receptor-mediated endocytosis.⁹⁰

However, the cellular uptake and ROS production are semi-dependent variables, where the latter depends on the former, but the former is essentially independent of the latter. That is, once uptake occurs, then ROS production commences. As discussed above, following uptake, the processes of overproduction of ROS, cell death, debris generation and adsorption, and blockage of active sites can occur. These would affect the population of cells capable of uptake of nanoceria, thus effectively decreasing the uptake. Thus, this mechanism represents an alternative, which is noncontradictory, to that based on maximal uptake by cells.

3.4 Interaction of EGF-nanoceria with EGF receptor

Western blot analysis was performed in order to confirm the EGFR binding specificity of EGF-functionalized nanoparticles. After the treatment with nanoceria for 5, 30, or 60 min, the medium containing the nanoparticles was discarded and the cells were washed thoroughly with DPBS. This washing step ensured the removal of residual nanoparticles and therefore any possible desorbed EGF, thus ensuring the interaction between the EGF conjugated to the nanoparticles and the receptor, thereby providing a measure of the EGF-nanoceria induced phosphorylation of EGFR. Figure 5 confirms the EGF-induced phosphorylation of the receptor, as evidenced by the presence of p-EGFR bands in the samples treated with EGF-nanoceria for 5 and 30 min. The diminishing of the band in the sample treated for 30 min and the absence of p-EGFR band in the cell sample treated for 60 min is attributed to the dephosphorylation of the receptor.⁹¹ The time-dependence of these processes; which include receptor-binding of EGF-nanoceria, phosphorylation, de-phosphorylation, and further downstream signaling; is supported by the observation of the progressively diminishing p-EGFR band intensities with time. The diminishing of p-EGFR band intensities further confirms the absence of any free EGF moieties in the nanoparticle formulation. The intensities of the protein EGFR bands correlate with those of the p-EGFR in the EGF-nanoceria treated cell lysates, thus confirming the stimulation of EGFR by EGF-nanoceria. The EGF-nanoceria interacts rapidly with the HT-1080 cell membranes, as evidenced by the intense band at 5 min. More broadly, the data confirm the EGFR-binding capability of EGF-nanoceria. However, it is unexpected that the intensities of the EGFR bands for the control and APTES-nanoceria are different at 5 min when the absence of EGF in both should have generated similar intensities. This difference is attributed to the nanoparticle-mediated receptor modulation, where the nanoparticles alone can modulate the activity of the receptors.^{92, 93}

3.5 Cellular uptake of nanoceria by fibrosarcoma cells

The TEM imaging shown in Figure 6 demonstrates the cellular uptake of nanoparticles by fibrosarcoma cells. The images from the early time points of 1, 2, and 5 min indicate the presence of nanoparticles in the vicinity of the cell membrane and the possible interaction with the cell membrane. The latter is confirmed by the western blotting data of Figure 5, as shown by the high-intensity band after 5 min of treatment with the EGF-nanoceria. The images after 30 min of treatment illustrate the cellular uptake of both nonfunctionalized and functionalized nanoceria, where the nanoparticles are enclosed within membrane-bound vesicles, which are likely to be endosomes.^{94, 95}

While these data confirm the preceding comments made about agglomeration and de-agglomeration, they also show that cellular uptake occurs within 5 h. Moreover, these data serve as more direct evidence for the demonstration of EGFR-mediated uptake of EGF-nanoceria since the fluorophore was tagged with EGF, thereby confirming the EGFR targeting potential of EGF-nanoceria.

3.6 Cytotoxicity of nanoceria against tumor cells and normal cells

The metabolic activity of cells is known to decrease the pH,⁹⁶ which can induce ROS generation and adsorption, followed by cell death. However, unlike the previous testing protocols, these samples were subjected to a change in medium every 24 h with the aim of maintaining a physiological pH. Consequently, the ROS generation deriving from the cell metabolism may be a cause of the logarithmic decrease in the viability of the normal fibroblast cells (MRC-5) shown in Figure 8. This is likely to be a result of an accumulation effect of the nanoceria following the charge-dependant uptake of APTES-nanoceria, which is likely to have manifested through subcellular redistribution of the nanoceria in the organelles. In support of this, aminated positively-charged nanoceria have been found to be internalized by normal cells, thereby inducing subsequent toxicity through lysosomal localization. Since lysosomes are acidic in nature, they would trigger the pro-oxidant nature of nanoceria.⁹⁷ It is remarkable that the nanoceria induced minimal toxicity toward normal cells up to 48 h. Thus, when translating the system to in vivo conditions, the toxicity arising from the time-dependent accumulation effect and possibly the pH effect as well would be likely to be negligible as exocytosis and nanoparticle clearance mechanisms are relatively rapid and occur in less than this time frame.^{98, 99}

Concerning the cancer cells the trend is different, where there is a significant decrease in the cell viability all three time points. At the first time point of 24 h, the cell viability decreased to 62 ± 7% (Figure 8), which is statistically distinguishable from the 44 ± 8% of Figure 2. In principle, these two sets of results should be identical because the cell viability testing with (Figure 8) and without (Figure 2) medium change applies only to the 48 and 72 h time points. Table 1 shows that the [Ce³⁺] increased by ~19% from October 2018 to February 2022. As Ce⁴⁺ is known to be cytotoxic to cancer cells,^{7, 100} the aging of CeO₂ should decrease the cytotoxicity. Since the data in Figure 8 were obtained on May 2021 and those in Figure 2 were obtained on October 2021 earlier, then these results are not consistent with the effect of aging on the [Ce³⁺]. This disagreement suggests that the role of the active sites in the form of ${\mathrm{V}}_{\mathrm{O}}^{••},$^{75, 76} which are generated by 2(Ce⁴⁺ → Ce³⁺) reductions, for ROS generation from the active sites (producing H₂O₂⁷) dominate the role of the ROS generation from the Ce³⁺ → Ce⁴⁺ oxidation (producing ^•OH and ^•O₂⁻⁷). This is consistent with the observation that H₂O₂ is the most long-lived ROS¹⁰¹ and so induces considerable cytotoxicity on cancer cells.^{102, 103} However, the effect of the medium is clear in that the cell viability at 48 h decreased to 35 ± 2%. (Figure 8), which is statistically distinguishable from the 53 ± 6% of Figure 2. After 72 h, the cell viability decreased to 45 ± 4% (Figure 8), which again is statistically distinguishable from 89 ± 1% of Figure 2. Since the statistically significant decrease in cell viability in the HT-1080 cells is analogous to the trend in the MRC-5 cells, then this suggests that the pH did not have a dominant effect with the cancer cells. Thus, it is likely again that an alternative mechanism provides the dominant mechanism for the trend of cell viability. In this case, it is likely to be the reduction in blockage of catalytically active sites owing to the removal of cell debris during change of medium. Thus, the catalyst surfaces remain less impeded from the catalytic generation of ROS and associated cell death.