27 Toxicity of an Al₂O₃ nanoparticle and HCl mixture: inhalations in vivo versus <i>in vitro</i> at the air–liquid interface

Abstract Combustion of composite solid propellants in rockets and missiles releases high concentrations of alumina nanoparticles (Al₂O₃ NPs) and gaseous hydrogen chloride (HCl), raising concerns about pulmonary toxicity from repeated inhalation, especially in military contexts. Advanced air–liquid interface (ALI) exposure systems offer promising non-animal alternatives for aerosol toxicity assessment. However, their predictive value for repeated inhalation toxicity of mixtures still requires evaluation against animal models, the current gold standard. This study compared in vivo and in vitro inhalation approaches to assess respiratory toxicity of Al₂O₃ NPs and HCl mixtures. Wistar rats were exposed nose-only to low, medium, or high concentrations once daily for 4 d. In parallel, an alveolar barrier co-culture (hAELVI/HPMEC-ST1.6R) was exposed at the ALI using a Vitrocell® Cloud system, following either a single exposure or a repeated 4-d exposure. Biological responses were analyzed 24 h after the last exposure, focusing on cytotoxicity, inflammation, and genotoxicity. Dose normalization (µg/cm²) enabled quantitative extrapolation between models. In vitro, only the highest dose (2.64 µg/cm²) induced toxicity, with inflammation after single exposure and cytotoxicity after repeated exposure. In vivo, the highest concentration (20 mg/m³) also induced inflammation. Notably, the dose causing toxicity in vivo (0.12 µg/cm²) was ∼20-fold lower than in vitro. Genotoxicity evaluation is ongoing, both in vitro and in vivo. Overall, ALI models predicted key pulmonary effects after single and repeated exposures, supporting their value as complementary or alternative methods. Yet, sensitivity differences highlight the need for further refinement before regulatory or risk-assessment use.