Quantitative Structure-Activity Relationship Models for Predicting Inflammatory Potential of Metal Oxide...

Area of Science:

• Nanomaterial toxicology
• Computational toxicology
• Nanosafety assessment

Background:

• Assessing the inflammatory effects of engineered nanomaterials (ENMs) is experimentally challenging and time-consuming.
• Quantitative structure-activity relationship (QSAR) models offer an alternative for predicting nanosafety.
• No prior QSAR models existed for predicting ENM inflammatory potential.

Purpose of the Study:

• To develop QSAR models for predicting the inflammatory potential of metal oxide nanoparticles (MeONPs) based on their physicochemical properties.
• To establish a computational method for assessing ENM inflammatory potential, overcoming experimental limitations.

Main Methods:

• A dataset of 30 MeONPs was used to screen for the release of the proinflammatory cytokine interleukin-1 beta (IL-1β) in THP-1 cells.
• QSAR models were developed using machine learning and validated against independent MeONP datasets.
• Density functional theory (DFT) computations were employed to elucidate underlying inflammatory mechanisms.

Main Results:

• Seventeen of 30 MeONPs induced significant IL-1β release in vitro, with high relevance to in vivo outcomes.
• QSAR models achieved over 90% predictive accuracy for inflammatory potential and were validated experimentally.
• DFT and experimental data identified metal electronegativity and positive surface charge as key factors in MeONP-induced inflammation.

Conclusions:

• Interleukin-1 beta (IL-1β) release serves as a reliable index for ranking MeONP inflammatory potential.
• Developed QSAR models accurately predict MeONP inflammatory potential using physicochemical properties.
• This computational approach significantly reduces the time and labor required for nanosafety assessment.