Biosynthesized bimetallic nanoparticles containing CeO and ZnO exert shape and size dependent anticancer effects.

Introduction
Nanotechnology has emerged as a transformative force across various scientific disciplines, including medicine, agriculture, and engineering1. Therefore, the synthesis of novel and effective nanomaterials is expected to advance multiple fields of science and technology such as, engineering, agriculture, and medicine. Consequently, the application of nanomaterials is expanding in all dimensions of science2. Metal oxide (MeOx) nanoparticles have many advantages due to their unique properties such as high stability, biocompatibility, and large-scale production capacity for biomedical applications3. Cancer is one of the main leading causes of death worldwide4. It is anticipated that almost 2,001,140 new cases of cancer will be diagnosed that will lead to 611,720 deaths in the United States in 2024. It is predicted that 28 million new cancer cases and 16.2 million deaths will be globally happened by 20405. Various therapeutic strategies including surgery, chemotherapy, radiation therapy, immunotherapy are currently employed. However, cancer continue to pose a major challenge to healthcare systems. Therefore, massive attempts are being made to develop novel agents, such as small molecules and nanomaterials to improve cancer treatment6. It has been revealed that among the nanoparticles, the ZnO and CeO2 NPs are fascinating nanoparticles due to their biocompatibility toward normal cells7–10. Recent studies have shown that the ZnO and CeO2 nanoparticles exhibit cytotoxic effects on cancer cell lines1,11–14. Bimetallic nanoparticles take advantages due to presence two metallic atoms in their structure and show unique physical and chemical properties15–17. Cao et al.. reported that ZnO/CuO bimetallic nanoparticles reduced melanoma cells viability18. It has been recently shown that AgZnO nanoparticles exhibited cytotoxic effect on the lung cancer cell line15. Cheng et al.. observed that the Zn-CeO2 and Cu-CeO2 nanoparticles exhibit cytotoxic effects toward MDA-MB-231 cell line2.
Among the multiple approaches for synthesizing the metal oxide nanoparticles, green synthesis has gained more attention due to its cost-effective and eco-friendly synthesis stages. In this approach, one part of an organism is utilized to produce nanoparticles. Plants are the mostly abundant organisms which their extractions or productions are used for fabricating the nanoparticles19–21. Walnut is one of the most consumed dry fruit worldwide20,22. Walnut fruit includes three parts: walnut shell (WS), walnut husks/hulls (WH), and kernel. The two first parts are usually considered as a waste. However, recent works have shown multiple applications of WS and WH by-products in medical, nutrition, and industrial fields23. WS is a rich source of lignin, cellulose, and hemicelluloses which play essential role in the synthesis of nanoparticles due to presence many hydroxyl groups in their structure24. The hydroxyl-rich biopolymers act as natural chelating and reducing agents, enabling uniform metal cation distribution and controlled nucleation without additional surfactants. This biomass-templating strategy not only ensures eco-friendly synthesis but also yields smaller crystallite sizes (22–29 nm) compared to chemically synthesized ZnO/CeO₂ composites (40–60 nm), enhancing surface reactivity and cellular interaction24–27. In fact, Metal cations are distributed on WS biomass via binding with its hydroxyl groups. Hence, WS acts as a template for the metal oxide precursors. After distribution of the cations on the template and calcination, the template are removed and then metal oxide nanoparticles are formed27,28. Although several studies have reported the synthesis and anticancer effects of ZnO-CeO₂ bimetallic nanoparticles2,15,In this work, we utilize walnut shell powder as a green, low-cost, and lignin-rich biomass template for the simultaneous synthesis of three structurally distinct ZnO-CeO₂ bimetallic nanoparticles (ZnCeO₃, CeO₂@ZnO, and ZnO@CeO₂) in a single calcination step at 500 °C. Unlike chemical co-precipitation or sol-gel methods that require multiple reagents and generate toxic byproducts25,29. Then, we investigate the fabricated nanoparticles anticancer effects on colorectal and breast cancer cell lines.

Materials and methods

Preparation of metal oxide nanoparticles
Zinc nitrate hexahydrate, Zn(NO3)2.6H2O, cerium nitrate hexahydrate, and Ce(NO3)2.6H2O (Merck) were purchased and used. Walnuts were collected from Urmia local walnut trees then the shells was squeezed using a high-speed rotary cutting mill. Particles smaller than 0.45 mm were then separated for use. Pure nanoparticles (ZnONPs and CeO2NPs) were prepared using the following procedure. Initially, in 100mL of deionized water (Millipore, Milli-Q grade) 10 g of walnut shell and 6.9 mmol of corresponding metal nitrate were mixed at room temperature. After 5 h mixing by stirrer, the water of the resulting mixture was removed through evaporation under reduced pressure and the resulting solid was calcined at 500 °C for 4 h under open-air conditions. By the same method but using 6.9 mmol Zn(NO3)2.6H2O and 6.9 mmol Ce(NO3)2.6H2O, ZnCeO3 nanoparticles (ZnCeO3NPs) can be produced.Also, CeO2@ZnONPs and ZnO@CeO2NPs (core/shell nanoparticles) have been prepared by impregnation of ZnO NPs or CeO2NPs(core)and metal nitrate solution (molar ratio of core to metal nitrate: 1:1) and the resulting solid was calcined at the same conditions. The limitation in the synthesis of these nanoparticles is the calcination temperature which must be equal for all types of the nanoparticles.

Characterization of the fabricated bimetallic nanoparticles
The UV-visible spectrometer (Perkin) was used to obtain the absorbance of the fabricated nanoparticles. Fourier transform infrared (FT-IR) spectra were recorded using a Bruker Vector 22 FT-IR spectrophotometer under mild conditions in a KBr/Nujol mull in the range of 400–4000 cm− 1.X-ray diffraction patterns of the fabricated nanostructures achieved from Shimadzu XRD-6000 diffractometer with Cu Kα radiation at room temperature. The morphology and elemental observation of the nanomaterials were characterized by ZEISS Sigma 300 Field Emission Scanning Electron Microscope (FESEM).

Free radical scavenging capacity of the fabricated nanoparticles
The DPPH test was performed to evaluate the bimetallic NPs’ antioxidant capacity. Briefly, the NPs were added to 2 ml of DPPH s (0.1mM, Sigma) dissolved in methanol (Merck) in concentrations ranging from 12.5 to 400 µg/mL and then incubated in dark at 37 °C for 25 min. The methanolic solution was considered as a control, and the Butylated hydroxyl anisole (BHA, Sigma) solution was served as standard. Thereafter, the samples’ absorbance was checked at 517 nm by a spectrophotometer. Subsequently, the bimetallic NPs free radical scavenging activity was evaluated by Eq. 1:
In Eq. 1, the Ac and As are absorbance of control and standard or the the ZnCeO3, CeO2@ZnO, and ZnO@CeO2 NPs in DPPH solution, respectively30.

Cell culture and treatment
HCT-116 (colorectal cancer), MCF-7(breast cancer) cell lines, and HUVEC (Human Umbilical Vein Endothelial cell) cell line as a normal cell line were obtained from Iranian Pasteur institute. The cells were cultured in the DMEM high glucose medium (Gibco) containing penicillin and streptomycin (Biowest) enriched with Glutamine (Biowest) and 10% Fetal Bovine Serum(FBS, Gibco) in an incubator(Memmert, Germany) containing 5% CO2 at 37 °C with 95% humidity. Different concentrations (25, 50, and 100 µg/mL) of the ZnCeO3, CeO2@ZnO, and ZnO@CeO2 nanoparticles were prepared in PBS(pH 7.4). The limitation is concentrations of the nanoparticles, because in the higher concentrations the nanoparticles accumulated in the cell culture medium. Therefore, the concentrations were used to assess cytotoxicity. Cells were treated with the concentrations for 72 h under standard cell culture condition.

Cytotoxicity assay using MTT test
The cytotoxic effects of the nanoparticles were assessed using MTT assay. Briefly, 6 × 103 cells per well of 96-well plates were seeded and treated with different concentrations (25, 50, and 100 µg/mL) of the ZnCeO3, CeO2@ZnO, and ZnO@CeO2NPs and then incubated for 24, 48, and 72 h. Subsequently, 10 µl of MTT dye dissolved in deionized water (5 mg/mL, Sigma) was poured to each well and the plates were kept in incubator for 4 h. After removing the medium, the resulting formazan crystals were dissolved by adding 100 µL of DMSO (Merck) to each well. Finally, the wells absorbance was checked at 570 nm and the untreated and treated cells viability was calculated and the cell viability graphs were drawn and IC50 concentrations of the nanoparticles were assessed using Graph pad prism Ver.9.0.0.

Cell death assay
In order to determine the cell death induced by the nanoparticles, 2 × 105 cells of HCT-116 and MCF-7 cells were seeded in each well of 24-well plate. The cells were divided in two groups, treated and untreated(control) group. Then, the treated groups were exposed to 300 µg/ml of the ZnCeO3, CeO2@ZnO, and ZnO@CeO2 nanoparticles for 72 h. For morphological analysis of apoptosis, the cells were imaged by an invert microscope (Zeiss, Germany). After that, each well medium were depleted and the cells were rinsed once with PBS (pH 7.4). Then, the cells were trypsinized and centrifuged at 1500 rpm for 5 minutes and washed with PBS once. Subsequently, the cells were stained with acridine orange dye solution (1 mg/ml, Sigma). Finally, the cells were observed and imaged using a Fluorescent Microscope (Zeiss, Germany).
For evaluating the apoptosis inducing effect of the nanoparticles, first the 106 of HCT-116 and MCF-7 cells were seeded in 6 well plate and then treated with 300 µg/mL of the nanoparticles for 72 h. Thereafter the cells were trypsinized and washed with PBS, sunbequently the cells were prepared to Annexin V/PI test as the Annexin V/PI kit manufacture(Biolegend) protocol. Shortly, the cells were wash twice with staining buffer and then resuspended by binding buffer. After that, the Annexin V and propidum iodide were added to the cell solution. Finally the cell solutions were analyzed by a flowcytometer (Partec).

Real-time PCR evaluates expression of apoptotic genes
The expression levels of p53, BAX, Bcl-2, and Beta-actin genes were assessed by real time-PCR. Initially, 106 of HCT-116 and MCF-7 cells were cultured in every well of a 6-well plate and treated with IC50 concentration of ZnO@CeO2 NPs for 72 h. Then, the samples’ RNA was extracted by RNX PLUS kit (Sinaclon, Iran) and their concentration was determined by a nanodrop (Biotek). Subsequently, cDNA was synthesized using cDNA EASY synthesis kit (Parstous, Iran). Primers for p53, BAX, Bcl-2 and Beta-actin genes were designed by Oligo 7 software (Table 1). For analysis of the genes expression, the samples cDNA was amplified using SYBER green master mix (Ampliqon, Denmark) by a thermocycler (Applied Biosystems). Finally, the gene expression levels were quantified based on their Ct values and fold change plots were drawn by Graph Pad Prism Ver.9.0.0.

DCFH-DA staining to evaluate ROS level
The DCFH-DA staining was employed to analysis the intracellular ROS generation by the nanoparticles. The DCFH-DA dye was sourced from sigma and dissolved in DMSO(10mM). 106 of the HCT-116 and MCF-7 cells were seeded in 24 well palates and exposed to 300 µg/mL of the nanoparticles for 72 h. Thereafter, the cells were trypsinized and washed with PBS and suspended by DCFH-DA solution(0.1 µM in PBS) for 30 min in dark31. Finally, the cell solutions were analyzed by a flowcytometer(Partec).

Statistical analysis
All experiments were performed three times. The data were statistically evaluated by one and two-way ANOVAs using Tukey post hoc analysis in Graph Pad Prism Version.9.0.0. Then, p-values were determined and the p < 0.05, p < 0.01, p < 0.001, and p < 0.0001 were considered as significant difference between treated and control groups.

Results and discussion

UV-Vis analysis revealed the bimetallic NPs absorbance near to 361 nm
The UV–Vis absorption spectra result showed that the ZnCeO3, CeO2@ZnO, and ZnO@CeO2 NPs displayed strong absorption peaks near to 361 nm (Fig. 1). This is strongly supported by the previous reports which reported the CeO2 and ZnO containing nanoparticles maximum absorbance is seen between 350 and 400 nm. These absorption peaks arise from the bandgap transitions in ZnO (~ 3.37 eV) and CeO2 (~ 3.2 eV), with the slight blue-shift in bimetallic NPs compared to pure oxides indicating quantum confinement effects due to smaller particle sizes. In contrast to sol-gel synthesized ZnO/CeO2 heterostructures (~ 370 nm peak)32–34.Our green-synthesized NPs showed peaks at 361 nm (Fig. 1), suggesting improved optical properties from biomass-derived capping agents, which reduce defects and enhance electron-hole separation for potential ROS generation in anticancer applications35.

FT-IR analysis showed the presence of of Zn-O and Ce-O bonds
The FT-IR spectrum of metal oxide nanostructures is shown in Fig. 2. The spectra showed the presence of metal–oxygen bonds, with broad bands between 440 and 520 cm⁻¹ corresponding to Zn–O and Ce–O stretching vibrations. An additional broad peak above 3400 cm⁻¹ indicated adsorbed surface moisture. The observed Zn–O and Ce–O vibrations confirm the metal-oxygen framework, with peak intensities in core-shell NPs (CeO2@ZnO and ZnO@CeO2) reflecting shell dominance, as seen in similar bimetallic systems36.

X-ray diffraction patterns showed crystalline peaks for ZnO and CeO2
X-ray diffraction (XRD) analysis of the fabricated metal oxide nanoparticles is presented in Fig. 3. The XRD spectra were similar to the JCPDS card No. 043-0002and 34–0394 for ZnO and CeO2, respectively. It can be observed that the XRD pattern reveals well-resolved peaks, 31.8, 34.5, 36.3, 47.7, 56.7, and, 62.9 for ZnO and 28.6, 33.0, 47.5, 56.3, and 59.0for CeO2.The crystallite sizes of grains were determined through the Debye Scherrer formula and full-width half maximum (FWHM) of the most intense peak (Table 2).

Debye Scherrer formula: D = kʎ/βcosθ.
Where k = 0.89, ’D’ represents the average crystallite size (nm), λ is the wavelength of X-ray (0.15406 nm) and β constitutes the FWHM.
The XRD patterns confirm the formation of crystalline phases in the bimetallic nanoparticles, with distinct peaks corresponding to the hexagonal wurtzite structure of ZnO (space group P63mc) and the cubic fluorite structure of CeO2 (space group Fm-3 m), as matched to JCPDS cards No. 043 − 0002 and 34–0394, respectively. In the bimetallic spectra (ZnCeO3, CeO2@ZnO, and ZnO@CeO2 NPs), the coexistence of both ZnO and CeO2 peaks indicates a composite structure without significant phase impurities or alloying, suggesting successful integration of the two oxides during green synthesis37.The broadening of peaks, particularly in ZnO@CeO2 NPs, reflects smaller crystallite sizes (as calculated via the Debye-Scherrer formula: D = kλ/β cosθ, where smaller D correlates with increased FWHM), which may enhance surface reactivity due to higher defect densities38. Comparatively, our crystallite sizes (e.g., ~ 20–30 nm, Table 2) are smaller than those reported for chemically synthesized ZnO/CeO2 composites (~ 40–50 nm)39, likely owing to the templating effect of walnut shell biomass, which restricts grain growth during calcination at 500°C24,27,28. This is supported by literature on plant-mediated synthesis, where organic templates yield finer crystallites with improved photocatalytic or biomedical properties compared to sol-gel methods40. However, the absence of peak shifts suggests minimal lattice strain or doping between ZnO and CeO2 phases, differing from doped systems like Ce-doped ZnO, where shifts indicate ionic substitution41.These structural insights underscore the role of synthesis method in tailoring crystal properties for enhanced anticancer efficacy.

FESEM images revealed the average diameters of ZnCeO3, CeO2@ZnO, and ZnO@CeO2 NPs
Figure 4 presents FESEM images of mixed metal oxide nanomaterials, as well as size distributions derived from SEM images. The reported mean diameter by scanning electron microscopy of the nanoparticles were 29 nm, 26 nm, 22 nm for ZnCeO3NPs, CeO2@ZnONPs, and ZnO@CeO2NPs, respectively. The size ranges of nanoparticles were 20–40 nm for ZnCeO3NPs, CeO2@ZnONPs, and 15–30 nm for ZnO@CeO2NPs. Also, SEM images showed mainly spherical and aggregated rod-like nanoparticles for ZnCeO3NPs, amorphous nanostructures for CeO2@ZnONPs, and spherical nanoparticles for ZnO@CeO2NPs. Amorphous nanoparticles exhibit the highest surface energy due to their disordered atomic arrangement and abundant surface defects. In contrast, spherical nanoparticles show the lowest surface energy, while rod-like and tubular structures possess intermediate values depending on their aspect ratio and exposed crystallographic facets42.

DLS technique provides the hydrodynamic diameter of ZnCeO3, CeO2@ZnO, and ZnO@CeO2 NPs
As shown in Fig. 5, the size of the nanoparticles obtained from DLS analysis is larger than the size calculated using the FESEM histogram (Fig. 5). This discrepancy arises because DLS technique measures the hydrodynamic diameter of clustered nanoparticles rather than the real size of them43.It should be noted that, the comparison of size distribution histograms(Fig. 4) and DLS diagrams (Fig. 5)indicates that the ZnCeO3 NPs exhibited the lowest degree of aggloeration among the bimetallic NPs, whereas ZnO@CeO2NPs showed the highest degree of aggregation.

The observed discrepancy between SEM and DLS size measurements can be attributed to the fundamental differences in the techniques. SEM provides direct visualization of dry, individual nanoparticle morphology under vacuum conditions, yielding primary particle sizes (e.g., 29 nm for ZnCeO3 NPs, 26 nm for CeO2@ZnO NPs, and 22 nm for ZnO@CeO2 NPs). In contrast, DLS measures the hydrodynamic diameter in aqueous suspension, which includes the solvated shell and accounts for particle aggregation or clustering, often resulting in larger apparent sizes44.This aggregation is influenced by factors such as surface charge, ionic strength, and interparticle interactions, leading to polydispersity indices (PDI) that reflect the heterogeneity of particle size distributions. For instance, the higher DLS sizes observed for ZnO@CeO2 NPs (indicating greater agglomeration) may arise from their spherical morphology and lower surface energy, promoting clustering in solution, as evidenced by a broader PDI compared to the more dispersed rod-like and spherical mix in ZnCeO3 NPs25. Such polydispersity can impact biological applications, as aggregated nanoparticles may exhibit altered cellular uptake and bioavailability45.These findings align with previous studies on metal oxide nanoparticles, where DLS sizes were 1.5–3 times larger than SEM due to hydration layers and agglomeration46.

Zeta potential analysis showed the moderate stability of the ZnCeO3 NPs, CeO2@ZnO NPs, and ZnO@CeO2 NPs
The zeta potential measurements of the ZnCeO3 NPs, CeO2@ZnO NPs, and ZnO@CeO2 NPs, conducted using the HORIBA SZ-100 instrument at approximately 25 °C (Fig. 3S; Table 3), reveal distinct colloidal stabilities in distilled water (low ionic strength, conductivity ~ 0.1 mS/cm). The nanoparticles exhibit moderate stability with mean zeta potentials of −22.7 mV (ZnCeO3 NPs), −19.4 mV (CeO2@ZnO NPs), and − 17.9 mV (ZnO@CeO2 NPs), indicating sufficient electrostatic repulsion to prevent immediate aggregation, though values below − 30 mV suggest potential long-term instability without additional stabilizers46,47. The unimodal distributions in all spectra confirm uniform charge profiles, with ZnCeO3 NPs showing the highest stability in water, while ZnO@CeO2 NPs display the lowest, likely owing to differences in shell composition and surface chemistry. These findings highlight the influence of nanoparticle structure on colloidal behavior in low-ionic media, implying the potential need for surface modifications to enhance stability for biomedical applications. The findings indicated that there is not significant difference in surface charges of the nanoparticles.

EDX and SEM-mapping analyses confirmed the presence and uniform distribution of Zn and Ce within the nanostructures
The SEM mapping images of zinc and cerium elements along with the EDX elemental analysis for mixed metal oxide nanoparticles are presented in Fig. 6. Given that the sensitivity to different elements in SEM-EDS mapping varies and heavier elements generally being easier to detect than lighter ones48, nearly equal molar amounts with uniform distribution of zinc and cerium elements were observed in ZnCeO3NPs (Fig. 6). While, as expected, in CeO2@ZnONPs due to cerium oxide is coated by zinc oxide, 16.6 mol% of the elements detected on the surface belong to cerium and 27.4 mol% belong to zinc (Fig. 6). Obviously, in this sample, cerium showed more uniform distribution than zinc, while zinc oxide was mainly seen as separate aggregates. Also, as shown in Fig. 6, ZnO@CeO2NPs consist of 35.3 mol% cerium and 17.7 mol% zinc.

ZnCeO3, CeO2@ZnO, and ZnO@CeO2 NPS did not show significant antioxidant activity
DPPH test was performed to evaluate the antioxidant capacity of ZnCeO3, CeO2@ZnO and ZnO@CeO2 NPs. Ejaz Ahmed et al. reported the CeO2 NPs derived from Abelmoschus esculentus extract showed antioxidant activity. In the other study, Banu et al. showed that the CeO2@ZnO nanocomposites derived from azadirachta indica aqueous leaf extract contains excellent antioxidant activity39. Despite these finding, as shown in Fig. 7, the ZnCeO3, CeO2@ZnO, and ZnO@CeO2NPs showed slightly free radical scavenging activity. Therefore, they cannot be considered as antioxidant nanoparticles. However, the CeO2@ZnO NPs exhibited slightly more antioxidant activity compared with the ZnCeO3 and ZnO@CeO2 NPs (Fig. 7). The less antioxidant activity of the nanoparticles may be concluded from the high temperature condition to synthesize the nanoparticles which leads to degrade the antioxidant organic constituents of ZnCeO3, CeO2@ZnO, and ZnO@CeO2NPs.

The ZnCeO3, CeO2@ZnO, and ZnO@CeO2NPs decreased cell viability
To evaluate the ZnCeO3, CeO2@ZnO, and ZnO@CeO2 NPs cytotoxic effects on HCT-116, MCF-7, and HUVEC cell line, the MTT assay was employed and the cell viability plots were drawn and then the IC50 concentrations were calculated by Graph pad prism Version. 9(Fig. 8; Table 4). As shown in Fig, the nanoparticles reduced the HCT-116, MCF-7, and HUVEC cells viability and their cytotoxic effect was time and dose-dependent. Alabyadh et al. showed that chemically synthesized ZnO@CeO2 nanocomposites exhibit cytotoxic effect on HepG2 cells. Moreover, the nanocomposites displayed lesser cytotoxic effect on NIH3T3 cells (normal cells) in compared to HepG2 Cells49. Similarly, ZnO/CuO NPs synthesized from Annona muricata L. extract showed not only considerable cytotoxic effect on MCF-7 cell line(IC50 = 25 µg/mL) but also exerted no significant cytotoxic effect on fibroblast cell line18. In another study, Al Bitar et al. reported that the ZnO-Ce NPs display cytotoxic effect toward HCT-116 cells (IC50 = 0.52 mM). They concluded that the ZnO pure NPs exerted more cytotoxic effect than ZnO-Ce NPs on the HCT-116 cells29. According to Table 4, ZnO@CeO2 showed lesser cytotoxic effect (IC50 = 487.3 µg/mL) than ZnCeO3 and CeO2@ZnO NPs(IC50 = 176.5 and 82.49 µg/mL, respectively) on HUVEC cell line. Akhtar et al. observed that the CeO2-Zn NPs exhibit cytotoxic effect on HUVEC cell line50, which is similar to our findings that showed CeO2@ZnO NPs display cytotoxic effect on HUVEC cell line. In another hand, Table 4 indicates that the IC50 value of ZnO@CeO2 NPs on HCT-116 and MCF-7 cells is lesser than it on HUVEC cells. Therefore, the nanoparticles exhibited more cytotoxic effect on the cancer cells rather than normal cells. Based on the results, the ZnO@CeO2 NPs were chosen as a promising anticancer nanoparticles for further evaluation of apoptotic genes expression in HCT-116 and MCF-7 cell lines.

Hu et al. reported that amorphous nanoparticles exhibit higher cytotoxicity than spherical nanoparticles26. In another hand, Hadji and Bouchemal suggested that the rod-like nanoparticle exert more cytotoxic effect than spherical nanoparticles51. In this work, the CeO2@ZnO NPs with amorphous shape showed more cytotoxic effect on HUVEC and the cancer cell lines (Fig. 8; Table 4) which indicates the nanoparticles are more cytotoxic compared to others. As shown in Table 4, the NPs exerted shape dependent cytotoxic effect on HUVEC cell line as following order:
CeO2@ZnO NPs (amorphous shape) > ZnCeO3 (rod-like and spherical shape) > ZnO@CeO2 (spherical NPs).
The order of cytotoxic effect of the nanoparticles on cancer cells is as the following order:
CeO2@ZnO NPs (amorphous shape) > ZnO@CeO2 NPs (spherical NPs) > ZnCeO3 NPs(rod-like and spherical shape).
In both normal and cancer cells the irregular nanoparticles showed more cytotoxic effect than others. It has been elucidated that the smaller nanoparticles display more cytotoxic effects due to their larger specific surface area which facilitates their interaction with cell membrane and penetration in to cell52. In this work, the ZnCeO3 NPs showed remarkable lesser cytotoxic effect than the ZnO@CeO2 NPs toward cancer cells, this effect may be related to the size dependent effect of the NPs on the cancer cell line because the size of the ZnO@CeO2(22 nm) is smaller than ZnCeO3 NPs.

All nanoparticles induced apoptotic cell death morphology and ROS level elevation in HCT-116 and MCF-7 cell lines
To evaluate the apoptosis inducing effects of the ZnCeO3, CeO2@ZnO, and ZnO@CeO2 NPs on HCT-116 and MCF-7 cells, the cells treated with the 300 µg/mL of the nanoparticles for 72 h and then the cells and the nuclei morphology were photographed by invert and fluorescent microscopes, respectively. Apoptosis, a programmed cell death, eliminates cancer cells from tissue with minimal damage to surrounding cells. Therefore it is an excellent strategy to cancer treatment53. Dead cells lose their attachment to their environment and are observed as spherical cells. Moreover, in the apoptotic cells, the cellular DNA is degraded to yield single strand DNA that would be seen as red color by acridine orange (AO) staining. In turn, the intact cells and early apoptotic cells nuclei are seen as green and orange color, respectively54. Figures 9 and 10 showed an elevation in the number of spherical and red cells in HCT-116 and MCF-7 cells exposed to the nanoparticles. As observed in Fig. 9, all the bimetallic NPs caused an increase in spherical cells indicating the raised dead cells number. Furthermore, the Fig. 10 showed that the all the NPs elevated the dead cells. According to Fig. 10 the CeO2@ZnO NPs exerted more apoptosis inducing effect on the HCT-116 and MCF-7 cell line which is similar to the MTT test data (Table 4). As shown in Fig. 11, all the nanoparticles induced apoptotic cell death in the cells. the most apoptosis inducing effect was observed in CeO2@ZnO NPs treated cells (43.2% and 45.3%for HCT-116 and MCF-7 cell lines, respectively). The results of the Annexin V/PI test confirmed the AO staining results. The lowest apoptosis inducing effect was seen in the MCF-7 cells treated with ZnCeO3 NPs(15.5%).

ZnO@CeO2 NPs induced cell death by Raising the Bax/Bcl-2 ratio
In order to determine the type of cell death caused by nanoparticles, the apoptotic genes expression was evaluated by real-time PCR. According to the lesser cytotoxic effect of ZnO@CeO2NPs on the normal cell (HUVEC), the nanoparticles were chosen to analyze the apoptotic genes expression. The Bax and Bcl-2 are two members of the Bcl-2 family proteins that their balance determines the cell fate. In the apoptosis process the Bax/Bcl-2 ratio is increased that leads to promote apoptosis. Therefore, in the apoptotic cells the ratio would be raised55. In the other hand, p53 is a main protein in inducing apoptosis, because it induces Bax gene over-expression56. In this work the level of Bax, Bcl-2 and p53 genes expression were evaluated in HCT-116 and MCF-7 cells exposed to IC50 concentration of the ZnO@CeO2 NPs after 72 h. As shown in Fig. 12, in both cancer cell lines, the p53 gene expression was significantly increased. Adeniyi et al.. showed that combining of CeO2 and ZnO NPs causes reactive oxygen species (ROS) generation57. In the other studies, it has been concluded that ZnO and CeO2 NPs can induce ROS production58–61. As observed in Fig. 13, all of ZnCeO3, CeO2@ZnO, and ZnO@CeO2 NPs increased the ROS level in treated cells. The CeO2@ZnO NPs treated cells showed the most ROS generation effect on the HCT-116 and MCF-7 cells (23.2 and 27.1%, respectively). The ZnO@CeO2 NPs induced ROS genereation in the HCT-116 and MCF-7 cells (20.8 and 21.7%, respectively). Therefore. It can be concluded that the nanoparticles increase ROS level in the treated cells and their cytotoxic effect may be related to ROS elevation. The ROS can damage DNA and activate the p53 protein which leads to over-expression of Bax and low-expression of the Bcl-2 genes56,62. Figure 12 showed that the ZnO@CeO2 NPs caused the p53 and Bax genes over expression in the two cancer cells which has led to Bax/Bcl-2 ratio elevation. Therefore, it can be suggested that the ZnO@CeO2 may increase apoptotic gene expression by ROS production (Fig. 13) in the cancer cells.

Conclusion
In this study, three distinct types of Zn–Ce bimetallic nanoparticles—ZnCeO3 (solid mixture), CeO2@ZnO (core–shell with CeO2 core), and ZnO@CeO2 (core–shell with ZnO core)—were successfully synthesized for first time using an eco-friendly method based on walnut shell powder. Comprehensive physicochemical characterizations confirmed the formation of nanostructures with varied morphologies, crystallite sizes, and elemental distributions. Biological assessments revealed that while all nanoparticles exhibited cytotoxic effect. CeO2@ZnO NPs showed shape dependent cytotoxic effect but it reduced the normal cell viability more the cancer cells. ZnO@CeO2 NPs revealed size dependent cytotoxic effect and most favorable anticancer effect toward breast and colorectal cell lines. Additionally, ZnO@CeO2 induced apoptosis, as evidenced by the morphological changes and the upregulation of pro-apoptotic genes p53 and Bax, accompanied by a decrease in Bcl-2, resulting in an elevated Bax/Bcl-2 ratio (Fig. 14). These findings underscore the crucial impact of nanoparticle structure and composition on biological activity. The core–shell architecture of ZnO@CeO2, in particular, appears to offer enhanced selectivity and therapeutic potential, likely due to optimized surface interactions and redox activity. Therefore, ZnO@CeO2 emerges as the most promising candidate among the three tested formulations for further development as a selective and biocompatible anticancer agent. Future in vivo studies and mechanistic investigations are warranted to validate these in vitro results and facilitate clinical translation.

Supplementary Information
Below is the link to the electronic supplementary material.