Adverse Outcome Pathways for Developmental Toxicity Gleaned from the Embryological Literature: A Rich and Rutted Road
On this page:
An Adverse Outcome Pathway (AOP) is a framework for organizing knowledge about the etiology of an adverse phenotypic outcome. AOPs include a Molecular Initiating Event (MIE) and downstream Key Events (KEs) leading to an adverse outcome. Completion of the sequence depends on Key Event Relationships (KERs) between KEs. The purpose for which the AOP is fit (extrapolation, prioritization, risk assessment) depends on data richness. The AOP Knowledgebase (AOP-KB), including the AOP-Wiki, allows AOPs to be shared and developed. The AOP-KB contains few AOPs for adverse developmental outcomes in mammals. Yet, the embryological literature is a trove of knowledge for constructing AOPs for developmental toxicity. Putative AOPS can be derived from the etiology of common malformations including cardiac, craniofacial and axial skeletal defects, and others. Common MIEs include receptor agonism/antagonism, enzyme inhibition, ion channel blocking, oxidative stress; early KEs include effects on signaling pathways, cell migration, proliferation, differentiation and interaction; downstream KEs include dysmorphogenesis, altered pattern formation, cell death. Neural crest cells (NCC) are precursors for many structures, and NCC proliferation, migration, differentiation and apoptosis are common KEs. BMP and Wnt/β-catenin deficiencies are MIEs for heart malformations including atrioventricular septal defects (AVSD); AVSD KEs include decreased downstream signaling and proliferation in the secondary heart field and deficient dorsal mesenchymal protrusion. Neural tube closure involves BMP, Wnt, SHH and PCP signaling, blocking these could be MIEs; inhibited ectoderm expansion, NCC emigration or apoptosis are early KEs, followed by dysmorphogenesis and failure to fuse. Hox genes regulate development of many structures including the axial skeleton and limbs; somite differentiation involves Wnt, BMP, SHH and FGF signaling with temporal expression. Knowledge of molecular and cellular mechanisms linking common genomic/transcriptomic changes to diverse pathogenesis is sparse, but advances are being made using cell-based in silico models. Adverse outcomes like IUGR will likely have many AOPs and modifying factors (e.g. maternal nutrition, obesity, age, other exposures, etc.) will make determining inclusive KERs difficult. Example AOPs will be presented and the outlook for further development and application will be discussed.