AUTHOR=AbdulHameed Mohamed Diwan M. , Pannala Venkat R. , Wallqvist Anders 

TITLE=Mining Public Toxicogenomic Data Reveals Insights and Challenges in Delineating Liver Steatosis Adverse Outcome Pathways

JOURNAL=Frontiers in Genetics

VOLUME=10

YEAR=2019

URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2019.01007

DOI=10.3389/fgene.2019.01007

ISSN=1664-8021

ABSTRACT=<p>Exposure to chemicals contributes to the development and progression of fatty liver, or steatosis, a process characterized by abnormal accumulation of lipids within liver cells. However, lack of knowledge on how chemicals cause steatosis has prevented any large-scale assessment of the 80,000+ chemicals in current use. To address this gap, we mined a large, publicly available toxicogenomic dataset associated with 18 known steatogenic chemicals to assess responses across assays (<italic>in vitro</italic> and <italic>in vivo</italic>) and species (i.e., rats and humans). We identified genes that were differentially expressed (DEGs) in rat <italic>in vivo</italic>, rat <italic>in vitro</italic>, and human <italic>in vitro</italic> studies in which rats or <italic>in vitro</italic> primary cell lines were exposed to the chemicals at different doses and durations. Using these DEGs, we performed pathway enrichment analysis, analyzed the molecular initiating events (MIEs) of the steatosis adverse outcome pathway (AOP), and predicted metabolite changes using metabolic network analysis. Genes indicative of oxidative stress were among the DEGs most frequently observed in the rat <italic>in vivo</italic> studies. <italic>Nox4</italic>, a pro-fibrotic gene, was down-regulated across these chemical exposure conditions. We identified eight genes (<italic>Cyp1a1</italic>, <italic>Egr1</italic>, <italic>Ccnb1</italic>, <italic>Gdf15</italic>, <italic>Cdk1</italic>, <italic>Pdk4</italic>, <italic>Ccna2</italic>, and <italic>Ns5atp9</italic>) and one pathway (retinol metabolism), associated with steatogenic chemicals and whose response was conserved across the three <italic>in vitro</italic> and <italic>in vivo</italic> systems. Similarly, we found the predicted metabolite changes, such as increases of saturated and unsaturated fatty acids, conserved across the three systems. Analysis of the target genes associated with the MIEs of the current steatosis AOP did not provide a clear association between these 18 chemicals and the MIEs, underlining the multi-factorial nature of this disease. Notably, our overall analysis implicated mitochondrial toxicity as an important and overlooked MIE for chemical-induced steatosis. The integrated toxicogenomics approach to identify genes, pathways, and metabolites based on known steatogenic chemicals, provide an important mean to assess development of AOPs and gauging the relevance of new testing strategies.</p>