Assessment of chemical-induced toxicity has traditionally relied on single-endpoint measurements that provide limited mechanistic resolution for organ-specific disease outcomes. This limitation is particularly evident for CMIT/MIT (chloromethylisothiazolinone/methylisothiazolinone), preservatives linked to severe health effects following South Korea's humidifier disinfectant disaster, where prior studies focused predominantly on Th2-skewed respiratory inflammation with limited cytokine resolution. Using a 46-cytokine Luminex panel, we profiled inflammatory responses across four human cell lines - pulmonary (BEAS-2B), hepatic (HepG2), neuronal (SH-SY5Y), and immune (Jurkat T) - and integrated the resulting profiles with the Comparative Toxicogenomics Database (CTD) and AOP-Wiki for disease-contextualized hazard characterization. CMIT/MIT induced dose-dependent ROS elevation across all cell lines, with cell type-specific cytokine signatures: IL-6 suppression with sustained IP-10 and growth factor elevation in BEAS-2B, and chemokine-dominant activation (MIP-3α, IL-8) in HepG2. Partial least squares-discriminant analysis revealed coordinated, pathway-level cytokine reorganization rather than isolated marker changes. Mapping these signatures onto CTD and AOP key events identified disease-relevant hazard signals for pulmonary fibrosis, hepatocellular carcinoma, and hepatic encephalopathy, supporting mechanistic hypotheses rather than disease causation. These findings demonstrate that multi-cytokine profiling combined with curated database integration provides a reproducible strategy for translating early inflammatory perturbations into mechanistically grounded hazard signals.

Ambient fine particulate matter (PM2.5) is a leading global environmental risk factor for cardiovascular disease. Although PM2.5 exposure is known to lead to mitochondrial dysfunction, the precise metabolic mechanisms underlying this effect remain incompletely understood. In this study, we investigated the role of the mitochondrial deacetylase Sirt3 in PM2.5-induced cardiotoxicity using human induced pluripotent stem cell-derived cardiac organoids (hiPSC-COs), a physiologically relevant multicellular model. Our results demonstrated that PM2.5 exposure compromised the contractile function of COs. Specifically, it induced the expression of myocardial injury markers and triggered a series of mitochondrial impairments, including structural damage, inflammatory response, oxidative stress and loss of membrane potential. Mechanistically, exposure of COs to PM2.5 suppressed Sirt3 expression, leading to increased acetylation and decreased activity of malate dehydrogenase 2 (Mdh2), a key enzyme in the tricarboxylic acid (TCA) cycle. This disruption resulted in reduced levels of TCA cycle metabolites and a severe ATP deficit. Pharmacological activation of Sirt3 with dihydromyricetin attenuated the PM2.5-induced mitochondrial dysfunction, restored ATP levels and improved contractility. Our findings elucidate that PM2.5 impairs cardiac energy metabolism via the Sirt3/Mdh2 pathway, identifying Sirt3 as a potential therapeutic target for mitigating PM2.5-related cardiovascular dysfunction.

This study establishes an in vitro protocol using the EpiOral three-dimensional human buccal oral mucosal model to assess potential irritation from oral medical devices and their leachables. EpiOral tissues were exposed to test substances (e.g., methyl acetate, Triton X-100, sodium lauryl sulfate (SLS), phosphoric acid, benzalkonium chloride) at concentrations relevant to dental materials, with exposure time for 50% tissue viability (ET-50) calculated from viability assays at 1, 4, and 18 h. The application volumes of the test substances (40 μL and 100 μL), vehicle controls (saline, sesame oil), and a 50% viability cut-off were validated using linear mixed models (LMM), post-hoc comparisons, and ROC analysis with Youden's index. ET-50 assays correctly classified the test irritants: strong (e.g. SLS, phosphoric acid, ET-50 < 1 h), moderate (hydrogen peroxide, 1.12 h), and non-irritants (e.g., glycerol, polyethylene glycol >18 h). No significant viability differences occurred between 40 μL and 100 μL for most substances (p > 0.05), except hydrogen peroxide. Using a 50% viability cut-off for irritancy classification, all tested substances were correctly classified in this exploratory dataset (sensitivity/specificity = 1.000). The in vitro testing protocol using the EpiOral model provides a structured experimental approach for oral mucosal irritation assessment. Under the conditions evaluated in this study, a 100 μL application volume and a 50% viability threshold were found to be appropriate.

Here, we present a blockchain-enabled, privacy-preserving platform that integrates label-free differential pulse voltammetry (DPV) measurements with image-based cell viability references for rapid and real-time monitoring of doxorubicin (DOX) induced cytotoxicity in L929 cells. A poly-l-lysine surface modified screen-printed carbon electrode was integrated with a petri dish to develop a cytotoxicity assay that converts toxicity-induced changes in cell coverage and viability into time- and dose-sensitive DPV signals. These signals were paired with percentage viability values obtained from a U-Net-based deep learning segmentation model trained using pseudo-masks generated from Live/Dead images via HSV-based rule-driven thresholding. To mitigate the risk of information leakage from individual images in future shared-data settings, the model was trained with Differential Privacy Stochastic Gradient Descent (DP-SGD) and maintained a validation performance of ≈ 0.94 at an ϵ ≈ 2.6 level. The regression model, built with normalized DPV current, duration, and dose, achieved R[2] = 0.943, reflecting time-dose-dependent toxicity. To reduce communication overhead, the data hash was processed on the blockchain via an event-based smart contract.

Inhalation is a major route of human exposure to chemicals, while the fate of inhaled chemicals across the pulmonary barrier remains unclear due to accessibility challenges and animal model limitations. In vitro lung epithelial models offer valuable insights into absorption. However, reliance on nominal concentrations and lack of a standardized model led to inaccurate permeability assessment. This study investigates the transport of five xenobiotics across two human epithelial models: bronchial (Calu-3), and alveolar (h-AELVi), cultured at Air-Liquid Interface. Their ability to form a tight epithelial barrier over a two-week culture period was compared, and along with an assessment of their growth kinetics on inserts. Both Calu-3 and h-AELVi cells cultured at ALI formed stable, tight epithelial barriers, evidenced by low Lucifer yellow permeability, stable high TEER and occludin expression. Despite differences in cell origin and growth kinetics, both models provided nearly identical permeability values for tested xenobiotics. Neglecting chemical losses such as binding to plastic, volatilization and metabolic degradation leads to an underestimation of apparent permeability values. In conclusion, Both barrier models provide reliable apparent permeability values only when effective compound concentrations are accurately determined. A framework based on experimental results was proposed to standardize in vitro lung epithelial permeability assays.

This study investigated the developmental toxicity of four cyanide-containing compounds (gold cyanide, sodium cyanide, zinc cyanide, and amygdalin) in zebrafish (Danio rerio) embryos. Fertilized eggs were exposed to concentrations ranging from 1 to 600 ng/mL for 96 h, following the OECD Test Guideline 236. Acute toxicity was determined by calculating the median lethal concentrations (LC50), while developmental toxicity was assessed through a semi-quantitative scoring system based on six morphological endpoints: eye, heart, circulation, yolk sac, tail, and spontaneous movement. Of the compounds tested, amygdalin proved to be the most acutely toxic (LC50 = 3.2 ng/mL), followed by sodium cyanide (6.3 ng/mL), zinc cyanide (33 ng/mL), and gold cyanide (90 ng/mL). A broadly similar toxicity ranking was observed for the severity of developmental malformations. These comparative data offer valuable insights for the environmental risk assessment of cyanide compounds, especially in areas impacted by active gold mining activities.

Polystyrene micro- and nanoplastics (MNPs) are widely used model particles in in vitro toxicology and represent a relevant fraction of environmental plastic pollution. Due to their small size and modifiable surface properties, MNPs interact with cellular membranes, undergo internalization, and influence intracellular signaling pathways. Reported biological effects range from oxidative stress and inflammation to minimal or absent toxicity. This review summarizes current in vitro evidence on cellular uptake, toxicological outcomes, and exocytosis of MNPs. Biological responses are strongly modulated by particle size, surface functionalization, aggregation state, concentration, and cell type. While smaller nanoparticles generally show higher uptake efficiency, internalization alone does not predict toxicity. Data indicate that the balance between endocytosis and exocytosis is a key determinant of intracellular effects. Efficient clearance may attenuate toxicity, whereas impaired trafficking and lysosomal accumulation promote organelle stress. Overall, MNPs behave as conditional toxicants in vitro, with their biological effects determined by their physicochemical properties and the cellular context. Standardized particle characterization and combined assessment of uptake and clearance are crucial for improving comparability and evaluating toxicological risk.

A new nasal formulation combining xylometazoline, a well-established nasal decongestant, and hyaluronic acid (HA) at a 3 mg/ml concentration and low-molecular-weight has been developed as a topical product intended for use in viral acute rhinosinusitis, with the aim of relieving nasal congestion and supporting epithelial integrity. HA, a moisturizing component of respiratory secretions, has been shown to promote epithelial integrity, stimulate mucociliary clearance and support wound healing. As part of the drug development process, this study investigated the permeation and penetration properties of xylometazoline and HA in the new formulation using a human nasal epithelium in vitro model. Results were compared to those of a xylometazoline mono-formulation and a well-established fixed-dose combination of xylometazoline and dexpanthenol. Morphological, ultrastructural and physiological analyses were performed to test nasal cells tolerability on prolonged drug exposure (6 h). HA in this new nasal formulation seems to be a component that exerts its effect mainly on the mucosal surface, which is the favoured site of action for topical products. HA was not detected in the basolateral compartment, consistent with retention at the mucosal surface. The results of our preclinical study do not indicate that the safety profile of xylometazoline would be altered by the addition of HA (or dexpanthenol). The barrier integrity of the in vitro human nasal epithelium was preserved across all tested formulations. Nanoparticle endocytosis was absent. These findings, obtained in a healthy in vitro nasal epithelial model, provide a valuable foundation for subsequent evaluation of this formulation in human clinical studies.

Environmental pollution has emerged as a major global health concern, with microplastics (MPs) now recognized as pervasive contaminants in the ecosystem and human tissues. Among them, polystyrene microplastics (PS-MPs) is frequently detected in metabolically relevant organs, raising concerns regarding their roles in diabetes development. In this study, we investigated the cellular and functional effects of PS-MP exposure in pancreatic β-cells using the MIN6 cell line. Our results demonstrated that 1 μm PS-MPs were efficiently internalized and predominantly localized within cytoplasmic and perinuclear regions of the cells. Exposure to PS-MP induced a dose- and time-dependent increase in mitochondrial superoxide and intracellular reactive oxygen species, accompanied by significant suppression of functional β-cell markers. This oxidative stress was associated with pronounced necrotic cell death. Moreover, metabolic viability assays revealed a progressive decline in β-cell viability along with suppression of key β-cell identity markers, including PDX1 and MAFA. Collectively, these alterations contributed to early β-cell dysfunction, a hallmark event in the initiation of diabetes. To our knowledge, this study is among the first to demonstrate an association between PS-MPs exposure, oxidative stress-mediated cytotoxicity, necrosis, and functional impairment in pancreatic β-cells. Our findings highlight a plausible environmental route linking chronic microplastic exposure to β-cell dysfunction, impaired insulin production, and heightened susceptibility to diabetes, underscoring the urgent need to evaluate microplastics as emerging metabolic disruptors.

Heated tobacco products (HTPs) and electronic cigarettes (E-cigs) are prominent next-generation tobacco products; However, a toxicological comparison using physiologically relevant air-liquid interface (ALI) models is lacking. We evaluated three HTPs and three E-cigs using human bronchial epithelial cells (NCI-H292) at ALI. The IC50 values for HTPs ranged from 302.8 to 577.6 puffs, suggesting a significantly higher cytotoxicity than E-cigs (IC50: 1065-1277 puffs). All three HTPs induced dose-dependent increases in ROS and decreases in GSH. In contrast, among the E-cigs, only E-cig-1 produced similar dose-dependent responses for ROS and GSH, while the others showed minimal effects. A modest, dose-dependent increase in γ-H2AX expression was observed for HTPs, whereas only one E-cig induced a comparable increase. Overall, under ALI conditions, the examined HTPs exhibited substantially higher toxicological potential than the E-cigs, highlighting differences in their relative health risks.

This study aimed to conduct an in vitro assessment of the concentration- and time-dependent cytotoxic effects of cetalkonium chloride (CKC), a preservative used in ophthalmic preparations, on human corneal keratocytes and to investigate its underlying cellular mechanisms of toxicity. Primary human corneal keratocytes were treated with a range of CKC concentrations (0 to 4.0 × 10[-4]% w/v) for up to 72 h, after which cell viability, intracellular reactive oxygen species (ROS) production, mitochondrial integrity, and the expression of key proteins in cell survival and apoptosis pathways were analyzed. The results showed that CKC induced a significant, concentration- and time-dependent reduction in keratocyte viability and a dose-dependent increase in ROS production. CKC exposure also led to the collapse of the mitochondrial network, cell shrinkage. Mechanistically, CKC suppressed the phosphorylation of pro-survival proteins Akt, mTOR, and ERK, while increasing the pro-apoptotic BAX/Bcl-xL ratio. In conclusion, CKC exhibits toxicity to keratocytes through mechanisms involving increased ROS, mitochondrial damage, cell membrane damage, and alterations of key survival/apoptosis signaling pathways. These findings provide a critical basis for defining safe concentration limits for CKC, guiding the development of ophthalmic preservative systems that minimize stromal keratocyte damage.

The Bovine Corneal Opacity and Permeability (BCOP) assay is a widely adopted non-animal method for identifying substances capable of causing severe eye damage. However, its implementation is limited by the requirement for freshly excised bovine corneas, collected and tested within a narrow post-mortem window. This proof-of-concept study evaluated whether short-term hypothermic preservation in Optisol-GS, a clinically validated corneal storage medium, could extend the use bovine corneas for BCOP assay. Corneas were preserved for up to 14 days and assessed for baseline opacity, control performance, IVIS classification of selected OECD proficiency chemicals, and histological integrity. Optisol-GS maintained corneal suitability for up to seven days, whereas HBSS and EMEM retained tissue acceptability only within the first 24 h. Preserved corneas consistently met BCOP validity criteria, yielding stable negative and positive control responses and expected IVIS classifications for all OECD proficiency substances tested. Histological analyses (HE, PAS, Masson) confirmed preservation of key epithelial, stromal, and endothelial structures, as well as depth-dependent injury patterns consistent with chemical severity. These findings indicate that Optisol-GS effectively extends the operational window for BCOP without compromising assay performance, suggesting a practical solution to logistical constraints associated with fresh tissue availability and supporting broader implementation of validated New Approach Methodologies in ocular toxicology.

Steartrimonium chloride (STC) is a cationic surfactant used in rinse-off hair and hair-dye products (≤2.5%) and leave-on hair products (≤1.0%) in Korea and the EU, and as a preservative in leave-on facial products in Korea (≤0.1%). Since no dermal absorption data are available for risk assessment, an in vitro study was conducted using mini-pig skin in Franz diffusion cell system following MFDS and OECD guidelines. Quantification of STC was performed by validated LC-MS/MS with MRM, showing excellent linearity (r[2] = 0.9945-0.9999), accuracy (84.9-113%), and precision (1.7-11.8%). A 2.5% gel (rinse-off) or 1.0% cream (leave-on) was applied at 10 μL/cm[2], with the gel formulation washed off after 30 min. After 24 h, swabbing, tape stripping, and skin sectioning were performed, and receptor fluid was collected at multiple time points. Total dermal absorption was 5.32 ± 3.75 μg/cm[2] (1.25 ± 0.91%) for the gel and 0.22 ± 0.20 μg/cm[2] (0.10 ± 0.10%) for the cream formulation. Recovery rates were for 97.14 ± 3.42% and 88.04 ± 2.35%, respectively. Estimated systemic exposure doses (SED) were 0.1258 mg/kg bw/day for rinse-off hair and hair dye product and 0.0061 mg/kg bw/day for leave-on hair and hair dye product, supporting refined safety assessments of STC in cosmetics.

This study evaluated the KeraSkin™ reconstructed human epidermis (RhE) model for in vitro skin corrosion testing per OECD Series on Testing and Assessment No. 219. A validation study across three laboratories assessed reliability and predictive performance using OECD reference chemicals, followed by an expanded evaluation of 79 chemicals. In accordance with Document No. 219, KeraSkin™ demonstrated overall compliance with mandatory requirements. For binary classification (n = 24), the model showed within-laboratory reproducibility (WLR) ≥95.8% and between-laboratory reproducibility (BLR) of 100.0%, satisfying predictive benchmarks (n = 20) with 100% sensitivity and 85.6% accuracy. For UN GHS subcategorization (n = 30), the model met acceptance criteria with WLR ≥90.0% and BLR ≥80.0% using the original prediction model (PM), achieving 100% sensitivity for Category 1A. While one laboratory's three-category accuracy (65.6%) fell below the indicative target, the aggregated multi-laboratory results satisfied all mandatory requirements of Document No. 219. In the expanded dataset (n = 79), the optimized PMvar2 effectively addressed laboratory variability, increasing overall accuracy to 73.5% while maintaining strict control over Category 1A misclassification. Whereas the original PM ensures current regulatory compliance, PMvar2 is presented as a promising strategy for enhancing predictive stability, with its adoption intended for further discussion during the formal OECD evaluation. Together, KeraSkin™ is a robust, regulatory-compliant RhE model fulfilling all OECD TG 431 inclusion criteria for non-animal skin corrosion hazard classification.