Tuning toxicity profiles of graphene oxide through imidazole-oxime modification: zebrafish as a model system

Abstract

The increasing use of nanotechnology, especially in agriculture and the food industry, has raised concerns about the possible adverse effects of nanomaterials (NMs) on human health and the environment. This study investigates the effects of synthesized graphene oxide (GO) and its derivatives on zebrafish exposed for 96 hr, focusing on morphological changes in brain tissue, histopathology, and immunofluorescent markers such as 8-hydroxy-2'-deoxyguanosine (8-OHdG) and nucleolar protein 10 (NOP10). Exposure to GO resulted in malformations, DNA damage, and increased NOP10 expression, and it reduced hatching and survival rates. Our results demonstrated that exposure to GO, graphene oxide-oxime (GO-OX), and OX exerted dose-dependent inhibitory effects on hatching and promoted malformations in zebrafish larvae. Histopathological analysis revealed that higher doses led to more pronounced tissue damage, with GO 50 causing severe degeneration and necrosis, while high doses of GO-OX and OX resulted in moderate tissue changes. This was further supported by the increased expression levels of 8-OHdG (marker of oxidative DNA damage) and NOP10 (marker of nucleolar stress), which aligns with the histopathological findings and confirms the neurotoxic effects. Notably, GO-OX treatments consistently mitigated both morphological and neurotoxic effects at all doses, suggesting that oxime functionalization reduces the inherent toxicity of GO. In contrast, treatment with different concentrations of GO-OX derivatives mitigated these adverse effects, reducing them to mild or moderate levels.

Introduction

Nanotechnology holds significant promise in agriculture for monitoring and improving environmental conditions, designing advanced systems for food and pesticide production, enhancing yield and nutritional value through a better understanding of crop biology, and creating new high-value-added products (Tarafdar & Raliya, 2013). Its potential extends to increasing productivity, developing new agrochemicals, diagnosing and treating plant diseases, reducing pesticide use, creating functional tools for molecular and cellular studies, detecting pathogens, protecting the environment, and developing innovative materials applicable across these areas (Alak et al., 2023a, 2023b, 2024; Köktürk et al., 2022; Ucar et al., 2023).

With the developing nanotechnology, nanomaterials (NMs; <100 nm) have a widespread use in every field with their potential to provide advantages as new technological products. Nanomaterials are classified as metal-based NMs (transition metals and metal oxides), carbon-based NMs (graphene, single and multi-walled carbon nanotubes, and fullerene), nanocrystals, dendrimers (nanoscale polymers), and cadmium quantum particles. Nanometer-sized inorganic compounds are different from other substances due to their size-specific structural properties, such as the electron scavenging effect (Demir, 2016). However, despite these anticipated benefits, there is growing concern about the potential hazards of nanomaterials on human and environmental health (Iavicoli et al., 2017). The residues of NMs in soil, agricultural products, and food raise concerns about their bioaccumulation and potential risks to the environment and the food chain. Although some ecotoxicological studies have begun to explore these concerns (Köktürk et al., 2023; Yiğit et al., 2024), the mechanistic basis of NM exposure and its effects, especially in freshwater and marine environments, is still poorly understood.

In aquatic environments, NMs can reach organisms directly, interact with other pollutants, and persist in stable or aggregated forms, leading to both acute or chronic effects (Ates et al., 2017). Among graphene-based NMs, 2D materials like graphene are widely used, especially in environmental protection and biomedical applications, due to their unique properties and derivatives (Hashemi et al., 2024; Wu et al., 2024). Graphene oxide (GO), one of the most common highly oxidized graphene derivatives based on its microstructural properties, has a large number of oxygen functional groups and a large surface area. This makes it highly reactive. However, it has other functional groups that can be used to prepare various derivatives, and these functional groups support GO solubility by providing high affinity for water molecules (Sargin et al., 2024). Although GO offers many potential benefits, its toxicity remains a concern, warranting thorough environmental toxicity assessments, especially through in vivo models like zebrafish (Wu et al., 2024).

Nanoparticles (NPs) pose significant risks to aquatic ecosystems, some research on their behavior, uptake by aquatic organisms, and toxic effects (Alak et al., 2024, 2025; Yiğit et al., 2024). A detailed investigation of NPs’ effects on aquatic species is essential to accurately assess the risks they pose. While the environmental hazards of NPs have been identified, the pathogenesis of NM exposure remains unclear and requires a multifaceted evaluation approach. Histopathological features and immunohistochemical markers, such as 8-hydroxy-2'-deoxyguanosine (8-OHdG), a commonly measured biomarker of oxidative stress, and nucleolar protein 10 (NOP10), involved in cellular stress, are critical for assessing NM toxicity (Yeni et al., 2024).

Today, while studies in the field of nanotechnology are increasing rapidly, there are some concerns about the possible effects of the nanomaterials developed on the environment and human health. While these concerns have focused on human toxicity, an additional nanosafety component to consider in the risk assessment of this class of materials is their ecotoxicity. However, the presence of NMs in water bodies can alter water quality and threaten the survival of aquatic species through accumulation and biomagnification along food chains (Atamanalp et al., 2022). The imidazole ring forms the core structure of many important natural and synthetic molecules (Menges, 2023). It is also known that the ecotoxicity of these molecules has been studied (Spasiano et al., 2016). In this study, we aimed to investigate the ecotoxicological profile of GO when modified with an imidazole ring. Therefore, our objective was both to functionalize the surface of GO and to examine the toxicity of this material.

Here, we aim to assess the ecotoxicological and physicochemical properties of GO and oxime-modified GO (GO-OX) NPs using zebrafish larvae as model. Additionally, we evaluate their physiological and immunohistochemical toxic effects through a toxicological test system designed to ensure a homogeneous mixing environment. Specifically, the objectives are: (1) characterization of the physicochemical properties of GO-OX NPs, and (2) evaluation of the physical and biological responses of zebrafish larvae exposed to GO-OX NPs through dose-response experiments. Furthermore, we examine the histopathological alterations and changes in the expression of 8-OHdG and NOP10 using immunofluorescence.

Material and methods

Materials and synthetic methods

Thionyl chloride (SOCl₂), dimethylformamide (DMF), 1,4-dioxane, phosphorus pentoxide (P₂O₅), selenium dioxide (SeO₂), ammonium acetate (CH₃COONH₄), anhydrous tetrahydrofuran (THF), ethylenediamine, graphite, sodium nitrate (NaNO₃), sulfuric acid (H₂SO₄), ethanol, potassium permanganate (KMnO₄), hydrogen peroxide (H₂O₂), and hydrazine hydrate (NH₂NH₂) were analytical-grade reagents purchased from different suppliers.

Synthetic methods

Synthesis of GO

Graphene oxide was synthesized using the modified Hummers method (Bulut et al., 2016). In a round-bottom flask, 5 g of graphite, 2.5 g of NaNO₃, and 115 ml of 98% H₂SO₄ were combined and stirred at 600 rpm for 30 min, ensuring the temperature remained below 20 °C. Subsequently, 15 g of KMnO₄ was added gradually over 30 min in an ice bath. The mixture was then stirred for 1 hr. Afterward, the temperature was raised to 45 °C, and 230 ml of distilled water was added, followed by another 15 min of stirring.

After this, an additional 230 ml of water was added, and the mixture was heated to 45 °C for 30 min. Following this step, 600 ml of water and 150 ml of 9% H₂O₂ were added, and the reaction was stirred for another 1 hr. The resulting mixture was then filtered, and the solid product (GO) was washed successively with 5% HCl and distilled water. The GO was dried in a low-temperature oven (Figure 1).

Surface modification of GO

In a 10 ml glass flask, 50 mg of previously synthesized GO was combined with 10 ml of SOCl₂. The mixture was thoroughly stirred, followed by addition of 1 ml of dry DMF. The reaction was stirred at 500 rpm for 24 hr at 70 °C. After the reaction, the mixture was filtered using a Gooch crucible (No. 4). During filtration, the solid (GO-Cl) was washed sequentially with anhydrous diethyl ether and anhydrous THF. The resulting GO-Cl was then dried in a low-temperature oven as described previously (Amudi et al., 2023).

Synthesis of imidazole-oxime derivative (Ox)

The synthesis of oxime (Ox) was based on the method described by Kuzu et al. (2018). Briefly, 1 mmol of aryl methyl ketone was dissolved in 7 ml of 1,4-dioxane in a 10 ml glass flask. To this solution, 2.5 mmol (0.275 mg) SeO₂ was added, and the mixture was heated in an oil bath for 24 hr. The progress of the reaction was monitored by thin-layer chromatography (TLC). After heating, the reaction mixture was filtered to remove SeO₂ and the filtrate was cooled to room temperature.

Separately, 5 mmol (0.385 g) of ammonium acetate (CH₃COONH₄) was dissolved in 10 ml of ethanol. The two solutions were combined and stirred at room temperature for 1 hr. Subsequently, 20 ml of ice-cold water was added and the mixture was stirred for an additional 1 hr. The resulting precipitate was collected using a Büchner funnel (No. 4) and dried (Kuzu et al., 2017).

In a separate reaction, 1 mmol (0.370 g) of (3,4-dimethoxyphenyl)(4-(3,4-dimethoxyphenyl)-1H-imidazol-2-yl)methanone compound was dissolved in 15 ml of n-propanol. To this solution, 1 mmol (0.14 ml) triethylamine and 1 mmol (0.0695 g) hydroxylamine hydrochloride (NH₂OH-HCl) were added. The mixture was refluxed overnight. Completion of the reaction was confirmed by TLC. The mixture was extracted three times with 10 ml ethyl acetate and 20 ml water. The organic layer was evaporated, yielding a gel-like material.

Synthesis of GO-OX

A 25 ml glass flask was charged with 50 mg of GO-Cl and 4 ml of DMF. The mixture was stirred at room temperature, and then 50 mg of OX was added. The reaction mixture was heated at 153 °C while stirring at 500 rpm for 24 hr. After heating, the mixture was filtered using a Gooch crucible (No. 4). The filtered solid (GO-OX) was washed sequentially with diethyl ether and THF, and then dried in an oven at 70 °C for 24 hr for further characterization (Figure 2).

Sample characterization

Fourier transform infrared spectroscopy

Chemical bonding changes were analyzed using a PerkinElmer Spectrum Two Fourier transform infrared (FT-IR) spectrophotometer with KBr pellet technique.

X-ray diffraction

The crystalline structure was analyzed using a Panalytical EMPYREAN X-ray diffractometer, with Cu Ka radiation (k = 1.540598 nm) generated at 45 kV and 40 mA with a scanning rate of 4°/min between 10° and 90°.

Field emission scanning electron microscopy

Surface morphological changes were imaged using a Zeiss GeminiSEM 500 field emission scanning electron microscopy (FE-SEM) at 10 kV 2500× magnification.

Transmission electron microscopy

Particle morphology and structure were examined using a Hitachi HT7800 transmission electron microscopy (TEM) operating at 100 kV, with a magnification of 150,000×.

Surface analysis and porosity

Specific surface area and porosity of the samples were measured using a Micromeritics Tristar II Brunauer–Emmett–Teller (BET) to determine the surface area and pore distribution.

Zebrafish maintenance and exposure to GO, GO-OX, and OX

Zebrafish embryos were obtained from wild-type AB zebrafish broodstock at the Izmir Biomedicine and Genome Center (İzmir, Turkey). Zebrafish were maintained following standard procedures (Köktürk et al., 2023) under the following conditions: 28 °C, 14:10-hr light:dark cycle, and fed with flake food and artemia. After mating, embryos were transferred to Petri dishes containing estriol (E3) medium (0.17 mM KCl, 0.33 mM MgSO₄, 5 mM NaCl, and 0.33 mM CaCl₂) and kept at 28 °C until the experiment commenced. Because the larvae were under five days old, ethical approval was not required (Directive 86/609/EEC and EU Directive 2010/63/EU).

For this study, the concentrations of GO, GO-OX, and OX were selected based on relevant literature on GO toxicity (Chen et al., 2020). Concentrations of 1, 10, and 50 mg/L were used for GO, GO-OX, and OX, respectively. Stock solutions (1,000 mg/L) were prepared using E3 medium and sonicated for 10 min in a sonicator (Minichiller Diagenode, 50/60 kHz, Huber, Germany). In each of the three replications independent experiments, 20 embryos were used per treatment group, with exposure lasting for 96 hr at 28 °C.

Toxicity assessment of GO, GO-OX, and OX

Toxicity was assessed by monitoring mortality rates at 24 to 96 hr, hatchability rates at 48 to 96 hr, and malformation rates at 24 to 96 hr. Observations were made and photographed under a stereomicroscope (SZX16 Olympus microscope + SC50 Olympus camera; Alak et al., 2025).

Histopathologic examination

Tissue samples were fixed in 10% formaldehyde for 48 hr, followed by embedding in paraffin. Sections of 4 µm thickness were taken from each block, stained with hematoxylin-eosin, and examined under a light microscope (Olympus BX 51, Japan). Histopathological features were graded as absent (–), mild (+), moderate (+++), or severe (++++; Alak et al., 2025).

Double immunofluorescence examination

Paraffin sections were placed on poly-L-lysine-coated slides, deparaffinized, and dehydrated. Endogenous peroxidase was blocked using 3% H₂O₂ for 10 min. After antigen retrieval in citrate buffer (pH 6, 1%), sections were incubated with a protein block for 5 min. Primary antibody 8-OHdG (Cat No.: sc393871, 1/100, US) was applied and incubated as per instructions. Immunofluorescence was performed using fluorescein isothiocyanate-conjugated secondary antibody (Cat. No.: ab6785, 1/1000) and kept in the dark for 45 min. Next, the sections were incubated in a second primary antibody, NOP10 (Cat. No.: ab134902, 1/100, UK). The second immunofluorescence staining was performed using a Texas Red-conjugated secondary antibody (Cat. No.: ab6719, 1/1000, UK) in the dark for another 45 min. A DAPI mounting medium (Cat. No.: D1306, 1/200, UK) was applied for 5 min for nuclear staining. Stained sections were analyzed using a fluorescence microscope (Zeiss AXIO, Germany).

Data analysis

Data were analyzed using GraphPad Prism 8. Results were presented as mean ± standard deviation (SD). Statistical significance was set at *p < 0.05 (****p < 0.0001, ***p < 0.001, **p < 0.01) and assessed via one-way analysis of variance (ANOVA) followed by Tukey’s multiple-comparison test for mortality and malformation rates, and two-way ANOVA for hatchability rates. Histopathological data were analyzed using the Duncan test for group comparisons, Kruskal-Wallis test for group interactions, and Mann-Whitney U test for pairwise comparisons. Immunofluorescence images were quantified using Zeiss Zen Imaging Software, selecting five random areas per image. Positive immunoreactive cells were compared using one-way ANOVA followed by Tukey’s test, with *p < 0.05 considered significant.

Results and discussion

Characterization result

According to the FT-IR spectra of OX, the presence of hydroxyl (-OH) and NH at 3,000 and 3,500 cm⁻¹ vibration bands was revealed and indicated these functional groups. Methyl (-CH3) groups were detected between 2,900 and 2,800 cm⁻¹, while carbonyl (-C = N-OH) groups exhibited a strong vibration band around 1,600 cm^-1. (C = C) bonds appeared at 1,509 cm-1 and (C-O) group was present at 1,150 cm⁻¹. Also, strong peaks are observed in the FT-IR spectrum of GO synthesis. These spectra were matched with the vibration band (-OH) groups observed at 3,189 cm⁻¹ and the vibration band carbonyl peak (–C = O), respectively (Çiplak et al., 2015; Figure 3). According to reports, partial or total elimination of oxygen-containing functional groups as a result of the heating or modification procedures results in partial or total flattening of the peaks (Sengupta et al., 2018).

The X-ray diffraction (XRD) analysis of GO and GO-OX provided insight into their crystalline structures (Figure 4). The XRD spectrum for GO showed peaks at 11.42° and 42.56°, 77.73°, corresponding to the oxygen-containing functional groups in GO (Li et al., 2019; Wang et al., 2011). After OX modification, new peaks were observed at 27.2° and 43.82°, 79.18°. Also, the d-spacing values for GO and GO-Ox layers were obtained by fitting the curves according to Bragg’s law. According to the XRD patterns, the GO pattern consists of three prominent peaks with d-spacing around 7.74, 2.10, 1.22 A˚, respectively (Kumar et al., 2022), while peaks with d-spacing of 3.27, 2.06, 1.20 A˚ were observed for GO-Ox. Although similar peaks or reflections are given on the 2θ side, a difference in spacing is observed between the planes. While the overall structure of GO remains intact, it can be said that there is an increase in the interlayer spacing, probably due to the intercalation of oxime groups (Chen et al., 2013; Huh, 2011; Hou et al., 2019; Zhang et al., 2018).

Images of FE-SEM revealed that GO possessed a porous, spongy structure. After modification to GO-OX, shrinkage and curling were observed along the edges, although the characteristic layered structure of GO was maintained (Figure 5). Notably, plate separation occurred after modification, indicating changes in the material’s physical properties (Köktürk et al., 2022; Lavakusa et al., 2017; Meng et al., 2011; Sharma et al., 2023).

Elemental analysis using FE-SEM-energy dispersive X confirmed the modification of GO. The carbon and oxygen ratios for GO were 62.8% and 37.2%, respectively. For GO-OX, the C, O, and N ratios were 86%, 12.4%, and 1.6%, respectively, highlighting the incorporation of nitrogen through imidazole-oxime modification (Esmaeili et al., 2020; Swami et al., 2017; Figure 6).

Analysis of TEM revealed structural details of both OX and GO-OX. In the TEM analysis of OX, both opaque and translucent images were observed (Figure 7). This can be attributed to two main factors. First, organic molecules tend to exhibit a stable appearance under strong electron radiation due to their immobility. Second, the limited energy resolution in TEM for analyzing carbon content in organic molecules hinders detailed observation of carbon bonds. High-energy electron beams used in devices like TEM and SEM have also been reported to induce changes in chemical bonding (Katrinak et al., 1992; Li & Shao, 2009; Li et al., 2011; Moffet et al., 2010). In the TEM images of GO-OX, a characteristic two-dimensional layered structure with wrinkles, typical of graphene oxide morphology, was prominent (Figure 7). An increase in layer thickness was also observed, which is attributed to the modification of the organic molecule within the graphene layer (Qi et al., 2017; Sungjin & Daniel, 2009; Yang et al., 2010). Flake-like, thin, transparent images were visible in some areas (Xu et al., 2017). Importantly, the organic modification did not significantly disrupt the layer-by-layer structure, a defining characteristic of graphene oxide (Figure 7).

The N₂ adsorption–desorption isotherms for the OX and GO-OX samples revealed the surface area and porosity through BET analysis. From the linear region of BET graphs, the surface areas of OX, GO, and GO-OX were determined as 5.0959 m²/g, 2.2167 m²/g, and 10.6471 m²/g, respectively (Figure 8). Adsorption pore volumes were 0.053192 cm³/g, 0.011276 cm³/g, and 0.066279 cm³/g for OX, GO, and GO-OX, respectively, while desorption pore volumes were found as 0.053437 cm³/g for OX, 0.010026 cm³/g for GO, and 0.056362 cm³/g for GO-OX. According to these values, it is noticeable that the surface area increases after the modification of the carbon-based organic compound on the GO surface (Yadav et al., 2021; Köktürk et al., 2022; Karataş et al., 2023; Figure 8). Additionally, adsorption and desorption volumes apart from specific surface areas were given in Table 1.

Embryos and larvae toxicity of GO-OX and OX

The toxicity of GO, GO-OX, and OX was evaluated using zebrafish embryos and larvae (Figure 9). Both GO and GO-OX exhibited minimal toxicity, with mortality rates comparable to the control group at all tested concentrations (1, 10, and 50 mg/L). In contrast, OX displayed dose-dependent toxicity, with significant increases in mortality at 10 mg/L (*p < 0.05) and 50 mg/L (***p < 0.001). This suggests that OX alone is more toxic than when incorporated into a GO structure, highlighting the potential benefit of nanomaterial modification in reducing toxicity. This study represents the first investigation into the toxicity of imidazole-oxime-modified graphene NPs.

Both GO-OX and OX reduced hatchability and increased mortality in zebrafish embryos, with OX exerting stronger effects on these parameters compared to GO-OX. Similar findings have been reported in previous studies, where GO was shown to cause developmental toxicity in zebrafish larvae, potentially related to the lateral dimensions of GO (Chen et al., 2020; Yiğit et al., 2024). In this context, higher concentrations of GO may impair zebrafish hatching by altering the mechanical properties of the chorion, inducing hypoxia within the chorion, reducing spontaneous movements and/or inhibiting the activity of hatching enzymes (Chen et al., 2024; Köktürk et al., 2022).

Zebrafish embryos typically exit the chorion synchronously at 48 hr post-fertilization. However, our study shows that the exit was delayed with increasing concentrations of GO, GO-OX, and OX nanostructures (Figure 10). A significant delay in chorion exit was observed in the GO 10, GO 50, GO-OX 10, GO-OX 50, and OX 50 treatment groups at both 48 and 72 hr compared to the control group (Figure 10). By the 96th hr, significant delays persisted in the GO 50, OX 10, and OX 50 groups. Similar findings were reported by Yang and colleagues, where 10 mg/L GO was found to reduce the hatching rate in zebrafish embryos (Yang et al., 2019). According to another study, low concentrations of GO can penetrate the chorion via its pore channels (0.5–1.0 μm), whereas higher concentrations form a GO “shield” around the chorion, blocking the pores and creating a low-oxygen microenvironment for the embryos (Chen et al., 2016).

Hatching in zebrafish is a complex process regulated by proteins and enzymes that work together to degrade the chorion, maintain osmotic balance, and support proper embryonic development (Alak et al., 2023a; Ucar et al., 2023). Disruption to these processes, due to environmental stress or chemical exposure, can lower hatching rates. Our results indicate that OX at 50 mg/L has a more pronounced effect on delaying hatching. Interestingly, GO-OX NPs at the same concentration did not exhibit similar effects. This suggests that the substitution of the hydroxyl group (OH) on the OX functionality by the graphene surface may reduce its toxicity. It implies that the free OH group in OX might be responsible for interacting with and potentially disrupting key proteins essential for hatching. Further studies are necessary to explore this hypothesis in more detail.

Zebrafish embryos and larvae displayed morphological abnormalities following treatment with GO, GO-OX, and OX, including pericardial edema, spinal curvature, and tail deformities (Figure 11). Control larvae exhibited normal morphology without visible abnormalities. In groups with malformation rates exceeding 10%, GO 10 (12.7%) showed that even at a low concentration (10 mg/L), GO can impact zebrafish morphology, although the effects were not severe (Figure 12). In contrast, GO 50 (15.1%) significantly impaired normal development, leading to skeletal malformations and physiological stress, such as yolk sac edema. Oxime 50 (14.2%) also demonstrated significant toxic effects, disrupting normal development and potentially affecting multiple physiological processes (Figure 12).

In this study, we monitored morphological deformities, hatching, and mortality to assess the developmental toxicity of GO, GO-OX, and OX in zebrafish. Notably, increasing doses of OX induced pronounced morphological abnormalities, such as spinal curvature, pericardial edema, and cardiac malformation (Figure 12). Our findings suggest that while GO NPs are toxic at higher doses, OX alone exerts more severe detrimental effects on zebrafish morphology. Graphene oxide-oxime, however, appears to offer greater developmental safety compared with OX, possibly due to the protective effect of graphene modification. For instance, larvae treated with GO-OX 50 did not show significant abnormalities and resembled the control group, indicating that functionalization of GO with OX mitigates the toxic effects observed with GO alone. These results underscore the importance of carefully considering NP concentrations and modifications, as overexposure can severely impair development in aquatic organisms.

The data in Figure 12 highlight that while both GO and OX exhibit dose-dependent toxicity, the functionalization of GO-OX effectively reduces these harmful effects. These findings emphasize the critical role of NP surface modifications in minimizing developmental toxicity. Further studies are recommended to explore the mechanisms through which GO-OX reduces toxicity, because this could provide valuable insights for designing safer NPs for biological applications.

The lower toxicity of GO-OX compared to free OX can be attributed to several interconnected factors. In GO-OX, OX groups are covalently bound to the GO surface, which limits their free reactivity and reduces the likelihood of direct interactions with key biological targets, such as DNA, enzymes, or structural proteins essential for cellular processes and development. In contrast, free OX molecules are more reactive due to their unbound hydroxyl groups and can interfere with essential cellular functions, including those of hatching proteins in zebrafish. In addition, the anchoring effect of GO in the GO-OX complex reduces the bioavailability of OX groups, minimizing their absorption into cells and preventing cellular degradation. Free OX is likely to exhibit higher solubility and diffusibility, making it easier to penetrate and accumulate in cells, thereby exerting toxic effects. The bulkier structure of GO-OX not only limits cellular uptake but also influences its biodistribution, resulting in fewer toxic interactions with intracellular targets. In addition, GO’s surface properties may play a role in stabilizing the OX groups and potentially exhibit antioxidant properties, which could counteract the oxidative stress induced by reactive OX molecules. The aromatic rings in the GO-OX complex may further enhance its stability, creating a more biocompatible nanostructure. This combination of reduced reactivity, lower cellular uptake, improved stability, and potential antioxidant effects likely explains why GO-OX NPs exhibit a significantly safer toxicological profile compared with free OX, particularly in developmental toxicity studies involving zebrafish larvae.

Our current findings highlight the developmental toxicity of OX, effectively assessed using the zebrafish model. Moreover, the study showed that GO-OX caused less developmental toxicity in zebrafish compared to OX at the same concentrations. However, once the epoxy rings in GO are opened, more active carbon radicals with unpaired electrons are formed, leading to formation of C-OH groups within the GO structure. This can result in additional lipid peroxidation, as observed previously (Chen et al., 2024). In our study, the elevated mortality rates and observed malformations in the GO and OX groups, alongside immunohistochemical findings (notably increase in 8 OHdG), suggest that the roles of C-O-C and C-OH in GO-induced lipid peroxidation may contribute to the accumulation of reactive oxygen species (ROS) in embryos.

Histopathological and immunofluorescence findings

The results of histopathological and immunofluorescent staining of zebrafish brain tissues are shown in Table 2 and Figure 13, with statistical analysis data presented in Figure 14. Nanoparticle toxicity has attracted increasing attention, and, in this study, microscopic images revealed that NPs adhered to or entered zebrafish embryos.

Interestingly, our literature review found no previous studies investigating NOP10 expression in response to GO and its derivatives in zebrafish. This indicates that further research is needed to fully elucidate the underlying mechanisms involved. Table 1 highlights several key observations. The control group displayed normal brain tissue with no signs of degeneration or necrosis, and no expression of oxidative DNA damage marker 8-OHdG or nucleolar protein marker NOP10, indicating healthy conditions. In contrast, the GO-treated groups demonstrated dose-dependent toxicity, with increasing tissue degeneration and biomarker expression. Graphene oxide 1 caused mild tissue damage and biomarker expression, while GO 10 led to moderate degeneration, necrosis, and oxidative stress. At GO 50, severe tissue damage, intense oxidative DNA damage, and nucleolar stress were detectable. This suggests that increased ROS production contributes to lipid peroxidation and DNA damage, perpetuating a cycle of damage (Lai et al., 2021).

In comparison, GO-OX treatments resulted in significantly less damage, indicating that OX modification reduces GO’s toxicity. Even at higher doses, GO-OX groups only showed mild to moderate degeneration and controlled biomarker expression. Graphene oxide-OX 50, in particular, demonstrated reduced toxicity compared to GO 50. Similarly, OX treatments showed increasing toxicity with dose. While OX 1 had minimal effects, OX 50 induced moderate degeneration and oxidative stress, although less severe than in the GO 50 group. These findings underscore the role of oxime functionalization in mitigating the toxic effects of GO and highlight the potential of chemical modification to improve nanomaterial biocompatibility.

The surface functionalization of GO with hydroxyl groups likely increases the negative surface charge and the critical coagulation coefficient, reducing hydrophobicity and preventing aggregation. This results in better dispersibility and stability of the nanomaterial, which enhances its bioavailability and uptake by aquatic organisms (Jung et al., 2024; Köktürk et al., 2022). The observed effects could be attributed to the interplay between ROS production, the increased surface area of GO, decreased metal impurities, and the effects of surface functionalization in aquatic environments (Jung et al., 2024).

The observed morphological changes in these groups further support our findings. Previous research has suggested that graphene may cause minimal or no toxicity (Hashemi et al., 2024). The differences in toxicity outcomes are attributed to factors such as the physicochemical properties of graphene (e.g., size, shape, surface chemistry), exposure conditions, and the specific biological models used (Hashemi et al., 2024). Another study reported that NOP10, which plays a crucial role in processes such as ribosomal RNA processing, small nuclear RNA modification, and telomerase stabilization, exhibited significantly elevated expression levels in tumor tissues exposed to a different xenobiotic (Sulukan et al., 2022). This increased expression is also recognized as an important marker of metastasis formation. In our study, severe DNA damage, marked by the presence of 8OHdG, was especially evident in GO 50 group, indicating that oxidative stress could be a major factor contributing to developmental toxicity in zebrafish. This suggests that oxidative damage plays a key role in GO-induced developmental defects. A common cellular event that can lead to mutations, cancer, aging, and ultimately cell death is DNA damage. Deoxyribonucleic acid is continuously modified by metabolites such as ROS and other external agents. These modifications can cause cell death in single-celled organisms or degeneration and aging in multicellular organisms (Yılmaz Sezer et al., 2024). The toxicity observed in our study from both the single and combined applications of OX might be attributed to the stronger ability of carbonyl (=O) and hydroxyl (-OH) groups to form complexes with OX (Sengottuvelu et al., 2024). Additionally, our findings suggest that the oxygen-containing groups on graphene auxin nanosheets contribute to an inherent oxidative potential (Chen et al., 2024). This oxidative potential was found to be higher in GO compared to GO-OX and OX at environmentally relevant concentrations, further supporting the role of oxidative stress in toxicity.

Conclusion

The findings of the present study are aligned with previous research that has indicated that GO induces oxidative stress, inflammation, and apoptosis in biological systems, with greater toxicity observed at different doses. The lower toxicity of GO-OX observed in this study suggests that OX functionalization stabilizes the GO surface and minimizes interactions with essential cellular components. The role of oxygen-containing groups in NP toxicity is well established, with numerous studies highlighting the significance of surface functionalization and chemistry in determining biocompatibility and toxicity.

Our results underscore the need for further in vivo studies to clarify the complex relationship between oxygen functionalization and toxicity. Such research is crucial to understanding the mechanistic pathways underlying GO-based NP toxicity and to defining safe exposure thresholds. To ensure a comprehensive safety assessment, it is essential to adopt holistic monitoring approaches that use multiple biomarkers, particularly for biomedical or environmental applications.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Author contributions

Serkan Yildirim (Formal analysis, Methodology, Resources, Supervision, Validation, Writing—review & editing), Mine Köktürk (Conceptualization, Formal analysis, Methodology, Resources, Supervision), Aybek Yigit (Conceptualization, Formal analysis, Methodology, Resources, Supervision), Ayse Sahin (Investigation, Methodology), Metin Kiliçlioglu (Investigation, Resources), Muhammed Atamanalp (Methodology, Supervision, Writing—review & editing), Berrah Gözegir (Investigation, Resources), Dilek Nazli (Investigation, Methodology), Gunes Ozhan (Methodology, Resources, Supervision, Validation, Writing—review & editing), Nurattin Menges (Conceptualization, Methodology, Resources, Writing—review & editing), and Gonca Alak (Formal analysis, Methodology, Resources, Supervision, Validation, Writing—review & editing)

Funding

This research has no external funding.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this study.

Ethics statement

Because the study involved larvae younger than five days old, ethical approval was not required, in accordance with Directive 86/609/EEC and EU Directive 2010/63/EU.

References