Differential eigengene network analysis reveals benzo[a]pyrene and 2,3,7,8-tetrachlorodibenzo-p-dioxin consensus regulatory network in human liver cell line HepG2.

Aryl hydrocarbon receptor (AHR) is one of the main mediators of the toxic effects of benzo[a]pyrene (BaP) and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). However, a vast number of BaP- and TCDD-affected genes may suggest a more complex transcriptional regulatory network driving common adverse effects of these two chemicals. Unlike TCDD, BaP is rapidly metabolized in the liver, yielding products with a questionable ability to bind and activate AHR. In this study, we used transcriptomics data from the BaP- and TCCD-exposed human liver cell line HepG2, and performed differential eigengene network analysis to understand the correlation among genes and to untangle the common regulatory mechanism in the action of BaP and TCDD. The genes were grouped into 11 meta-modules with an overall preservation of 0.72 and were also segregated into three consensus time clusters: 12, 24, and 48 h. The analysis showed that the consensus genes in each time cluster were either directly regulated by the AHR or the AHR-TF interactions. Some TFs form a direct physical interaction with AHR such as ESR1, FOXA1, and E2F1, whereas others, including CTCF, RXRA, FOXO1, CEBPA, CEBPB, and TP53 show an indirect interaction with AHR. The analysis of biological processes (BPs) identified unique and common BPs in BaP and TCDD samples, with DNA damage response detected in all three time points. In summary, we identified a consensus transcriptional regulatory network common for BaP and TCDD consisting of direct AHR targets and AHR-TF targets. This analysis sheds new light on the common mechanism of action of a genotoxic (BaP) and non-genotoxic (TCDD) chemical in liver cells.

An in silico toxicogenomic approach in constructing the aflatoxin B1-mediated regulatory network of hub genes in hepatocellular carcinoma.

Aflatoxin B1 (AFB1) can cause hepatocellular carcinoma (HCC) through a mutagenic mode of action but can also lead to global changes in gene expression; however, the AFB1 network of molecular pathways involved in HCC is not known. Here, we used toxicogenomic data from human liver cells exposed to AFB1 to infer the network of AFB1-responsive molecular pathways involved in HCC. The following computational tools: STRING, MCODE, cytoHubba, iRegulon, kinase enrichment tool KEA3, and DAVID were used to identify protein-protein interaction network, hub genes, transcription factors (TFs), upstream kinases, and biological processes (BPs). Predicted molecular events were validated with an external dataset, whereas the hub genes in HCC were validated using the UALCAN database. The results revealed an association between AFB1 and the hub genes involved in the cell cycle. We identified TFs that regulate the hub genes and linked them with upstream kinases including cyclin-dependent kinases, mitogen-activated protein kinase 1, and AKT. This approach enabled the construction of the AFB1-mediated regulatory network consisting of upstream kinases, TFs, hub genes, and BPs, thus revealing the signaling hierarchy and information flow that may contribute to AFB1-induced HCC. This could be a useful tool in predicting the molecular mechanisms involved in chemical-induced diseases when available toxicogenomic data exist.