Micro- and nanoplastic pollution poses a global threat due to environmental accumulation, poor reversibility, and adverse effects on humans and wildlife. Nanoplastics can cross the blood-brain barrier (BBB) and induce neurotoxicity, yet studies investigating the developmental effects of nanoplastic exposure during early life (a period vulnerable to environmental stressors) and their influence on later-life outcomes remain limited. This study examined whether early-life exposure to polystyrene nanoplastics (PS-NPs) during embryonic and larval stages has lasting effects and alters susceptibility to hexabromocyclododecane (HBCD), a brominated flame retardant globally detected in aquatic environment, in zebrafish. Developmental exposure to PS-NPs led to reduced locomotor activity in both larvae (5 dpf) and adults (157 dpf). Furthermore, early PS-NP exposure modified behavioral response to HBCD in adulthood: adult fish exposed to HBCD for the first time exhibited hyperactivity, whereas those previously exposed to PS-NPs showed no significant change. These differential responses were strongly correlated with altered expressions of neurological and neurodevelopmental genes (e.g., drd2a , glud1a , fezf2 ) and global DNA methylation levels. Our findings suggest that epigenetic processes may contribute to the observed differences in adult locomotor behavior and susceptibility to subsequent HBCD exposure. This study highlights the need for further research into locus-specific epigenetic regulation of neural genes during development and their quantitative relationship to adverse outcomes. In light of the conservation of epigenetic mechanisms across species and the widespread presence of environmental substances from industrial and consumer products that can disrupt these pathways, our study underscores the broader implications for environmental health. • Early exposure to PS-NPs reduced locomotor activity in larvae and adult zebrafish. • Developmental PS-NPs can influence adult response to HBCD-induced hyperactivity. • PS-NPs disrupted expression of neurodevelopmental and neurotransmission genes. • Distinct global DNA methylation patterns were observed across exposure scenarios. • Epigenetic regulation may contribute to persistent neurobehavioral outcomes.

The rapid transition toward animal-free chemical safety evaluation has positioned in vitro new approach methodologies (NAMs) as central components of next-generation risk assessment (NGRA). Advances in complex in vitro systems, high-content phenotypic profiling, multi-omics technologies, and AI-assisted analytics have greatly expanded the capacity to characterize human-relevant biological responses. However, despite their scientific promise, the translation of NAM-derived information into regulatory decision-making remains challenging in general. Key bottlenecks include incomplete alignment with apical regulatory endpoints, limited toxicokinetic context in conventional in vitro systems, and substantial variability across assays, data structures, and analytical pipelines. This review aims to summarize the current state of in vitro NAM technologies, evaluate the major barriers limiting their regulatory application, and discuss emerging frameworks that enable their integration into NGRA. To strengthen regulatory relevance, increasing efforts focus on integrating mechanistic NAM outputs into adverse outcome pathway (AOP) frameworks and applying high-throughput toxicokinetic (HTTK) modeling to support in vitro –to– in vivo extrapolation (IVIVE). Early NGRA case studies show that NAM-based points of departure can, in some instances, approximate or bracket traditional in vivo thresholds, although results remain heterogeneous across chemical classes and endpoint domains. Going forward, progress in NAM-based risk assessment will depend not only on advancements in assay technologies but also on decision frameworks capable of effectively incorporating existing NAM evidence. Tiered and evidence-integrated approaches will be essential, particularly in light of the varied NAM data availability across chemicals. Strengthening the iterative exchange between NAM application and method development will help guide future improvements and support a more transparent, adaptive, and human-relevant assessment paradigm.

Developmental and reproductive toxicity (DART) testing has traditionally relied on animal studies, which are costly, time-consuming, and ethically constrained. To advance new approach methodologies (NAMs), we developed a mechanism-informed deep learning framework for predicting DART using in vitro bioactivity data from 23 ToxCast assays mechanistically linked to key developmental and reproductive pathways. Four state-of-the-art (SOTA) deep learning architectures (DGCL, TransFoxMol, MolPath, and MolFormer) were evaluated to address performance limitations commonly observed in traditional supervised learning approaches. Each model was fine-tuned using the curated ToxCast dataset, with the F1 score serving as the primary evaluation metric. Among these, the DGCL model consistently outperformed baseline machine learning algorithms, including random forest, XGB, GBT, decision tree, and logistic regression. Extending DGCL to a multi-task learning framework further improved model stability and performance for endpoints with limited active data. External validation with 91 reference chemicals curated and verified by the ECVAM ReProTect program demonstrated balanced predictive performance (F1 = 0.68), confirming the reliability and generalizability of the fine-tuned DGCL model. By leveraging advanced deep learning architectures, the model effectively handles mechanistically diverse and imbalanced assay data with limited active samples, resulting in improved predictive performance across DART-related effects. Overall, this study demonstrates the potential of integrating mechanistic bioassay information with deep learning to develop reliable, mechanism-based, and non-animal methods for DART prediction and potential regulatory application.

Micro- and nanoplastic pollution poses a global threat due to environmental accumulation, poor reversibility, and adverse effects on humans and wildlife. Nanoplastics can cross the blood-brain barrier (BBB) and induce neurotoxicity, yet studies investigating the developmental effects of nanoplastic exposure during early life (a period vulnerable to environmental stressors) and their influence on later-life outcomes remain limited. This study examined whether early-life exposure to polystyrene nanoplastics (PS-NPs) during embryonic and larval stages has lasting effects and alters susceptibility to hexabromocyclododecane (HBCD), a brominated flame retardant globally detected in aquatic environment, in zebrafish. Developmental exposure to PS-NPs led to reduced locomotor activity in both larvae (5 dpf) and adults (157 dpf). Furthermore, early PS-NP exposure modified behavioral response to HBCD in adulthood: adult fish exposed to HBCD for the first time exhibited hyperactivity, whereas those previously exposed to PS-NPs showed no significant change. These differential responses were strongly correlated with altered expressions of neurological and neurodevelopmental genes (e.g., drd2a , glud1a , fezf2 ) and global DNA methylation levels. Our findings suggest that epigenetic processes may contribute to the observed differences in adult locomotor behavior and susceptibility to subsequent HBCD exposure. This study highlights the need for further research into locus-specific epigenetic regulation of neural genes during development and their quantitative relationship to adverse outcomes. In light of the conservation of epigenetic mechanisms across species and the widespread presence of environmental substances from industrial and consumer products that can disrupt these pathways, our study underscores the broader implications for environmental health. • Early exposure to PS-NPs reduced locomotor activity in larvae and adult zebrafish. • Developmental PS-NPs can influence adult response to HBCD-induced hyperactivity. • PS-NPs disrupted expression of neurodevelopmental and neurotransmission genes. • Distinct global DNA methylation patterns were observed across exposure scenarios. • Epigenetic regulation may contribute to persistent neurobehavioral outcomes.

The rapid transition toward animal-free chemical safety evaluation has positioned in vitro new approach methodologies (NAMs) as central components of next-generation risk assessment (NGRA). Advances in complex in vitro systems, high-content phenotypic profiling, multi-omics technologies, and AI-assisted analytics have greatly expanded the capacity to characterize human-relevant biological responses. However, despite their scientific promise, the translation of NAM-derived information into regulatory decision-making remains challenging in general. Key bottlenecks include incomplete alignment with apical regulatory endpoints, limited toxicokinetic context in conventional in vitro systems, and substantial variability across assays, data structures, and analytical pipelines. This review aims to summarize the current state of in vitro NAM technologies, evaluate the major barriers limiting their regulatory application, and discuss emerging frameworks that enable their integration into NGRA. To strengthen regulatory relevance, increasing efforts focus on integrating mechanistic NAM outputs into adverse outcome pathway (AOP) frameworks and applying high-throughput toxicokinetic (HTTK) modeling to support in vitro –to– in vivo extrapolation (IVIVE). Early NGRA case studies show that NAM-based points of departure can, in some instances, approximate or bracket traditional in vivo thresholds, although results remain heterogeneous across chemical classes and endpoint domains. Going forward, progress in NAM-based risk assessment will depend not only on advancements in assay technologies but also on decision frameworks capable of effectively incorporating existing NAM evidence. Tiered and evidence-integrated approaches will be essential, particularly in light of the varied NAM data availability across chemicals. Strengthening the iterative exchange between NAM application and method development will help guide future improvements and support a more transparent, adaptive, and human-relevant assessment paradigm.

Developmental and reproductive toxicity (DART) testing has traditionally relied on animal studies, which are costly, time-consuming, and ethically constrained. To advance new approach methodologies (NAMs), we developed a mechanism-informed deep learning framework for predicting DART using in vitro bioactivity data from 23 ToxCast assays mechanistically linked to key developmental and reproductive pathways. Four state-of-the-art (SOTA) deep learning architectures (DGCL, TransFoxMol, MolPath, and MolFormer) were evaluated to address performance limitations commonly observed in traditional supervised learning approaches. Each model was fine-tuned using the curated ToxCast dataset, with the F1 score serving as the primary evaluation metric. Among these, the DGCL model consistently outperformed baseline machine learning algorithms, including random forest, XGB, GBT, decision tree, and logistic regression. Extending DGCL to a multi-task learning framework further improved model stability and performance for endpoints with limited active data. External validation with 91 reference chemicals curated and verified by the ECVAM ReProTect program demonstrated balanced predictive performance (F1 = 0.68), confirming the reliability and generalizability of the fine-tuned DGCL model. By leveraging advanced deep learning architectures, the model effectively handles mechanistically diverse and imbalanced assay data with limited active samples, resulting in improved predictive performance across DART-related effects. Overall, this study demonstrates the potential of integrating mechanistic bioassay information with deep learning to develop reliable, mechanism-based, and non-animal methods for DART prediction and potential regulatory application.

전통적인 화학물질 위해성평가의 한계가 명확하게 드러나면서, 독성학 분야의 패러다임이 전환되고 있다. 그 중심에, 다양한 최첨단 기술을 이용해 화학물질의 독성을 평가하는 신규접근법 (New approach methodologies, NAMs)이 있으며, 이는 곧 차세대 위해성평가 (Next generation risk assessment, NGRA) 프레임워크로 구현되고 있다. 이에 미국 환경청에서는 EPA CompTox Chemical Dashboard라는 웹 기반 어플리케이션을 개발해 다양한 데이터를 통합하여 예측기반 독성평가를 지원하고 있다. 이에 본 총설에서는 EPA CompTox Chemical Dashboard에서 제공하는 데이터와 계산독성학 기능에 대한 총체적 분석과 이를 활용한 개별 사례연구를 통해 화학물질 독성예측과 차세대 위해성평가 구현을 위한 EPA CompTox Chemical Dashboard의 활용에 대한 전망을 제시하였다.

동물실험에 의존하는 전통적인 위해성평가 방법의 한계가 명확하게 드러나면서 , 동물실험이 아닌 최첨단 기술을 활용한 화학물질 위해성평가 방법론들이 연구되고 있다 . 이는 곧 신규접근법 (New Approach Methodologies, NAMs)이라는 개념으로구체화되었으며 차세대 위해성평가 (Next generation risk assessment, NGRA) 프레임워크로 구현되었다 . 하지만 신규접근법 데이터는 신뢰도 문제로 인해 극히 제한적으로 화학물질의 규제 의사결정과정에 활용되고있다 . 이에 규제 공동협력기관 APCRA 이니셔티브는 화학물질 위해성평가에 신규접근법 데이터의 활용을 증진시키기 위해 여러 사례연구를 수행하였으며 , 이를 바탕으로 차세대 위해성평가의 전망에 대해 논의한 바 있다 . 따라서 본 총설에는 APCRA 이니셔티브가 수행한 사례연구 두 편을 중심으로 신규접근법을 활용한 차세대 위해성평가의 개념과 전망을 제시하였다.

As the number of new chemicals increases, traditional animal testinghas limitations in evaluating the toxicity of chemicals. In Europe and the United States, the use of non-animal alternative toxicity test methods for chemical regulation is encouraged and related research is actively carried out. In the United States, the ToxCast program is in progress using high-throughput screening (HTS) techniques. The program produced HTS in vitrotest results for 1200 assays and 9000 chemicals to develop chemical toxicity prediction models and set the priorities for toxicity assessment. In this review, we explored the outline of ToxCast, the types of assays used, and the progress of each phase. We also explored how to analyze the numerous toxicity data generated through the ToxCast program, including pre-processing, dose-response model fit, conclusion and categorization. Finally, implications for domestic utilization ofthe ToxCast database are derived.

Objectives The widely promising applications of graphene nanomaterials raise considerable concerns regarding their environmental and human health risk assessment. The aim of the current study was to evaluate the toxicity profiling of graphene family nananomaterials (GFNs) in alternative in vitro and in vivo toxicity testing models. Methods The GFNs used in this study are graphene nanoplatelets ([GNPs]–pristine, carboxylate [COOH] and amide [NH2]) and graphene oxides (single layer [SLGO] and few layers [FLGO]). The human bronchial epithelial cells (Beas2B cells) as in vitro system and the nematode Caenorhabditis elegans as in vivo system were used to profile the toxicity response of GFNs. Cytotoxicity assays, colony formation assay for cellular toxicity and reproduction potentiality in C. elegans were used as end points to evaluate the GFNs’ toxicity. Results In general, GNPs exhibited higher toxicity than GOs in Beas2B cells, and among the GNPs the order of toxicity was pristine>NH2>COOH. Although the order of toxicity of the GNPs was maintained in C. elegans reproductive toxicity, but GOs were found to be more toxic in the worms than GNPs. In both systems, SLGO exhibited profoundly greater dose dependency than FLGO. The possible reason of their differential toxicity lay in their distinctive physicochemical characteristics and agglomeration behavior in the exposure media. Conclusions The present study revealed that the toxicity of GFNs is dependent on the graphene nanomaterial’s physical forms, surface functionalizations, number of layers, dose, time of exposure and obviously, on the alternative model systems used for toxicity assessment.

Objectives In this study, the effect of tube length and outer diameter (OD) size of hydroxylated- multi walled carbon nanotubes (OH-MWCNTs) on their uptake and toxicity was investigated in the nematode Caenorhabditis elegans using a functional mutant analysis. Methods The physicochemical properties of three different OH-MWCNTs were characterized. Uptake and toxicity were subsequently investigated on C. elegans exposed to MWCNTs with different ODs and tube lengths. Results The results of mutant analysis suggest that ingestion is the main route of MWCNTs uptake. We found that OH-MWCNTs with smaller ODs were more toxic than those with larger ODs, and OH-MWCNTs with shorter tube lengths were more toxic than longer counterparts to C. elegans. Conclusions Overall the results suggest the aspect ratio affects the toxicity of MWCNTs in C. elegans. Further thorough study on the relationship between physicochemical properties and toxicity needs to be conducted for more comprehensive understanding of the uptake and toxicity of MWCNTs.