Developing adverse outcome pathways on silver nanoparticle-induced reproductive toxicity via oxidative stress in the nematode Caenorhabditis elegans using a Bayesian network model

Abstract

An adverse outcome pathway (AOP) is a framework that organizes the mechanistic or predictive relationships between molecular initiating events (MIEs), key events (KEs), and adverse outcomes (AOs). Previously, we intensively investigated the molecular mechanism that underlies toxicity caused by AgNPs in the nematode Caenorhabditis elegans. Using transcriptomics, functional genetics, and various molecular/biochemical tools, we identified oxidative stress as the major mechanism underlying toxicity and reproduction failure as the outcome. With this information, here we conducted a case study of building an AOP to link oxidative stress with reproductive toxicity. To validate this AOP, we filled the gaps by conducting further experiments on its elements, such as NADPH oxidase, ROS formation, PMK-1 P38 MAPK activation, HIF-1 activation, mitochondrial damage, DNA damage, and apoptosis. The establishment of a causal link between the MIE and AO is critical for the construction of an AOP. Therefore, causal relationships between each KE and AO were verified by using functional genetic mutants of each KE. By combining these experimental data with our previously published results, we established causal relationships between the MIE, KEs, and AO using a Bayesian network (BN) model, culminating in an AOP entitled ‘NADPH oxidase and P38 MAPK activation leading to reproductive failure in C. elegans (https://aopwiki.org/aops/207)’. Overall, our approach shows that an AOP can be developed using existing data and further experiments can be conducted to fill the gaps between the MIE, KEs, and the AO. This study also shows that BN modeling has the potential to identify causal relationships in an AOP.

Keywords:

• Adverse Outcome Pathway
• Bayesian network; C. elegans
• silver nanoparticles
• nanotoxicity

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the Korean Ministry of Environment under the ‘Environmental Health R&D Program’ (2017001370001).