DNA oxidation damage contributes to carcinogenesis and limits the efficacy of many chemotherapeutics. Although platinum‑based drugs remain crucial in cancer therapy, their clinical use is restricted by toxicity and acquired resistance. However, quinazolinone derivatives have recently attracted attention as scaffolds with promising biological activities. Among these, Schiff bases are of particular interest because they can provide suitable ligands for the complexation with metals, making them a potential substitute for cisplatin. Therefore, in the search for new anticancer agents, Cu(II) complexes based on 2,3-substituted quinazolinones have been studied in detail. The radical scavenging/reducing efficiency and DNA-protective effects were monitored using DPPH, FRAP and DNA topology assays. Ligands and Cu(II) complexes, were screened for their in vitro cytotoxic, genotoxic, and protective activity in human hepatocarcinoma (HepG2) cells using the MTT and Comet assays. The results showed that compounds studied are redox-active agents. Cu(II) complexes exhibited higher DPPH radical-scavenging activity in comparison with their corresponding free ligands; they manifested comparable effects in ferric reducing power as well as in DNA topology assays. The cellular assays revealed markedly higher cytotoxicity of Cu(II) complexes compared with the corresponding free ligands. In the presence of H₂O₂, C1 significantly enhanced DNA damage, whereas C2 did not alter the level of DNA damage induced by H₂O₂ alone. On the contrary, ligands exhibited much lower toxicity and significant protection of HepG2 cells against H2O2. The incorporation of Cu(II) ions into quinazolinone Schiff base ligands substantially alters their biological activity, shifting their profile from DNA‑protective to cytotoxic/pro‑oxidant. DFT calculations were performed for both Cu(II) complexes to investigate molecular features that may contribute to their different biological activity. These results highlight the potential of quinazolinone‑Cu(II) complexes as anticancer candidates and support further studies into their mechanisms of action.
Chronic hyperglycemia in diabetes mellitus (DM) may promote oxidative stress, cytotoxicity, and chromosomal instability. However, evidence regarding exfoliated urothelial cells remains limited. This study evaluated micronucleus (MN) frequency and other nuclear anomalies in urothelial cells from women with type 1 DM (T1DM) and type 2 DM (T2DM) and investigated predictors of MN formation. Twenty women with T1DM, 25 with T2DM, and 22 controls were included. Second-void urine samples were collected, and urothelial cells were Feulgen-stained. MN and other nuclear anomalies were assessed in 1000 cells per participant. Clinical and metabolic characteristics were compared, and multivariable binary logistic regression identified predictors of MN formation. Age and BMI differed significantly among T1DM, T2DM, and control groups. T2DM participants had significantly longer diabetes duration than T1DM participants. HbA1c was significantly higher in both DM groups than in controls. Total cholesterol, HDL, and triglycerides also differed significantly among groups, whereas LDL did not. MN frequency was higher in diabetic participants overall but this increase was not statistically significant. Nuclear buds and degenerative changes were significantly more frequent in both DM groups. In direct comparisons of the DM patient groups, NB, KR, KL, and CC frequencies were significantly higher in T2DM than in T1DM participants. In multivariable analysis, longer DM duration and higher HbA1c levels independently predicted MN formation. DM was associated with increased nuclear anomalies in urothelial cells, although MN frequency was not significantly elevated. Longer disease duration and higher HbA1c independently predicted MN formation, suggesting that chronic metabolic exposure may contribute to urothelial cellular damage.
Radiation dose estimation is fundamental to the effective management of radiological emergencies, where prompt and accurate assessment is critical for medical intervention. Conventional biodosimetry methods, including the dicentric chromosome assay (DCA), premature chromosome condensation (PCC), γ-H2AX foci detection, cytokinesis-block micronucleus (CBMN) assay, and fluorescence in situ hybridisation (FISH) translocation analysis, remain important approaches for biological dose estimation. However, their conventional implementation can involve labour-intensive sample preparation and manual or semi-automated analysis, require specialised expertise, and have limited scalability for mass-casualty events. Recent advances in Artificial Intelligence (AI), particularly machine learning (ML) and deep learning (DL), offer considerable potential to enhance biodosimetry through automated image analysis, multimodal data integration, and accelerated dose assessment. DL approaches, including Convolutional Neural Networks (CNNs), enable high-throughput image analysis and automated identification of radiation-induced cytogenetic damage. Furthermore, integrating explainable AI (XAI) approaches may improve model transparency, interpretability, and confidence in clinical applications. This review examines current advances, challenges, and prospects of AI in radiation biodosimetry, highlighting its potential to support rapid, accurate, reproducible, and scalable dose assessment in radiological emergencies.
Metal oxide nanoparticles (NP) are widely used in industrial and biomedical applications. We have investigated the cytotoxicity, genotoxicity, and intracellular localization of aluminum oxide (Al2O3) and cerium oxide (CeO2) NP in Chinese hamster lung fibroblast (CHL/IU) cells. Cytotoxicity was assessed by cell viability analysis and genotoxicity was evaluated using the MN and chromosomal aberration assays, under continuous or short-term exposure conditions, with or without metabolic activation. Transmission electron microscopy (TEM) and scanning electron microscopy coupled with energy-dispersive X-ray spectroscopy (SEM-EDX) analyses were performed to investigate intracellular localization patterns of the NP. Both Al2O3 and CeO2 NP induced concentration-dependent cytotoxicity in these cells. However, neither NP induced significant increases in MN formation or chromosomal aberrations. TEM analyses showed electron-dense particles within membrane-bound intracellular vesicular structures in Al2O3 and CeO2-treated cells. In the examined TEM images, no apparent ultrastructural abnormalities were detected in mitochondria or nuclei. SEM-EDX analyses supported the presence of aluminum- and cerium-containing material within electron-dense intracellular regions. These findings indicate that Al2O3 and CeO2 NP induced cytotoxicity without detectable genotoxicity in CHL/IU cells. The observed vesicular localization was associated with the absence of detectable genotoxicity, although a causal protective role of vesicular sequestration cannot be established from these findings.
The in vivo erythrocyte micronucleus (MN) test is a standard assay for detecting chromosomal damage; however, conventional manual scoring is labor-intensive and subject to inter-scorer variability. Although automated image-based scoring offers a practical alternative, its application to bone marrow analysis requires effective enrichment of erythrocytes, to reduce interference from non-target nucleated cells. We have established and validated an automated bone marrow MN scoring workflow, using the Metafer slide-scanning system and associated software platform. We have compared a conventional cellulose column-based enrichment method with a syringe-filter-based method. The filter method reduced residual non-erythroid cellular components more effectively than the column method, although both methods yielded preparations suitable for automated analysis. Automated scoring showed no significant differences from manual microscopy in MN frequency or polychromatic erythrocyte ratio, in either negative- or positive-control groups. Repeated analysis of the same slide demonstrated high repeatability. Automated and manual MN frequencies showed positive linear correlations in both column- and filter-enriched preparations, and Bland-Altman analysis showed overall agreement between the two scoring methods. Automated historical control data (HCD) generated from male and female rats and mice were comparable to manually established laboratory HCD. These findings demonstrate that automated image-based scoring combined with appropriate erythrocyte enrichment is a reliable and practical alternative to conventional manual scoring in the in vivo bone marrow MN assay and may improve analytical efficiency, objectivity, and reproducibility in genotoxicity testing.
