Abstract
Background
Nanoplastics are emerging pollutants that can cross biological barriers and perturb metabolic homeostasis. Polystyrene nanoplastics (PS-NPs) are frequently detected in food and drinking water; however, their size-dependent metabolic effects and the modifying role of pre-existing diabetes remain incompletely understood. This study aimed to investigate whether subchronic oral exposure to PS-NPs of different particle sizes altered hepatic insulin signaling and glucose-lipid homeostasis in normal and diabetic mice, and whether oxidative-inflammatory imbalance was associated with these responses.Results
We found that PS-NP exposure induced size-dependent disturbances in metabolic and hepatic signaling endpoints. Among the tested particle sizes, 25 nm PS-NPs generally showed the largest apparent changes across several inflammatory, oxidative-stress, glycemic and insulin-signaling outcomes, although direct between-size differences were not uniformly established for all endpoints. PS-NP exposure increased serum levels of inflammatory cytokines (IL-6, CRP and TNF-α) and altered hepatic oxidative-stress markers, including the lipid peroxidation marker malondialdehyde (MDA) and antioxidant enzyme activities (CAT, SOD and GSH-Px), suggesting systemic inflammatory activation accompanied by hepatic oxidative-stress imbalance. These changes were associated with suppression of the hepatic IRS1-AKT-PPARγ axis and downregulation of GLUT4 expression, consistent with impaired insulin signaling and metabolic regulation. Compared with normal mice, diabetic mice exhibited exacerbated oxidative stress, inflammatory activation, and metabolic disturbance upon PS-NP exposure, suggesting higher susceptibility of metabolically compromised hosts to PS-NPs. Exploratory SEM provided supportive evidence for an association between oxidative-inflammatory imbalance and hepatic insulin-signaling impairment, while downstream links to systemic metabolic disturbance require further validation. Exploratory multivariate analyses, including principal component analysis (PCA), linear discriminant analysis (LDA), and random forest (RF), suggested response patterns associated with metabolic background and particle size, but these analyses should be interpreted as supportive pattern-recognition tools rather than validated predictive models.Conclusions
This study suggests that PS-NPs may act as metabolic stressors under the present experimental conditions, with their effects influenced by particle size and host metabolic background. By integrating classical toxicological methods with exploratory computational modeling, our findings provide mechanistic clues and preliminary data-driven evidence that PS-NP exposure may be associated with hepatic insulin-signaling impairment and metabolic disturbance, underscoring the need to assess nanomaterial safety in metabolically vulnerable populations.