Dataset Information

Identification of consensus biomarkers for predicting non-genotoxic hepatocarcinogens.

ABSTRACT: The assessment of non-genotoxic hepatocarcinogens (NGHCs) is currently relying on two-year rodent bioassays. Toxicogenomics biomarkers provide a potential alternative method for the prioritization of NGHCs that could be useful for risk assessment. However, previous studies using inconsistently classified chemicals as the training set and a single microarray dataset concluded no consensus biomarkers. In this study, 4 consensus biomarkers of A2m, Ca3, Cxcl1, and Cyp8b1 were identified from four large-scale microarray datasets of the one-day single maximum tolerated dose and a large set of chemicals without inconsistent classifications. Machine learning techniques were subsequently applied to develop prediction models for NGHCs. The final bagging decision tree models were constructed with an average AUC performance of 0.803 for an independent test. A set of 16 chemicals with controversial classifications were reclassified according to the consensus biomarkers. The developed prediction models and identified consensus biomarkers are expected to be potential alternative methods for prioritization of NGHCs for further experimental validation.

SUBMITTER: Huang SH

PROVIDER: S-EPMC5259716 | biostudies-literature | 2017 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Identification of consensus biomarkers for predicting non-genotoxic hepatocarcinogens.

Huang Shan-Han SH Tung Chun-Wei CW

Scientific reports 20170124

The assessment of non-genotoxic hepatocarcinogens (NGHCs) is currently relying on two-year rodent bioassays. Toxicogenomics biomarkers provide a potential alternative method for the prioritization of NGHCs that could be useful for risk assessment. However, previous studies using inconsistently classified chemicals as the training set and a single microarray dataset concluded no consensus biomarkers. In this study, 4 consensus biomarkers of A2m, Ca3, Cxcl1, and Cyp8b1 were identified from four la ...[more]

PMID: 28117354

Similar Datasets

Project description:Introduction:Various challenges have been overcome with regard to applying 'omics technologies for chemical risk assessments. Previously we published results detailing targeted mRNA sequencing (RNA-Seq) on a next generation sequencer using intact RNA derived from freshly frozen rat liver tissues. We successfully discriminated genotoxic hepatocarcinogens (GTHCs) from non-genotoxic hepatocarcinogens (NGTHCs) using 11 selected marker genes. Based on this, we next attempted to use formalin-fixed paraffin-embedded (FFPE) pathology specimens for RNA-Seq analyses. Findings:In this study we performed FFPE RNA-Seq to compare a typical GTHC, 2-acetylaminofluorene (AAF) to genotoxicity equivocal p-cresidine (CRE). CRE is used as a synthetic chemical intermediate, and this compound is classified as an IARC 2B carcinogen and is mutagenic in S. typhimurium, which is non-genotoxic to rat livers as assessed by single strand DNA damage analysis. RNA-Seq was used to examine liver FFPE samples obtained from groups of five 10-week-old male F344 rats that were fed with chemicals (AAF: 0.025% and CRE: 1% in food) for 4?weeks or from controls that were fed a basal diet. We extracted RNAs from FFPE samples and RNA-Seq was performed on a MiniSeq (Illumina) using the TruSeq custom RNA panel. AAF induced remarkable differences in the expression of eight genes (Aen, Bax, Btg2, Ccng1, Gdf15, Mbd1, Phlda3 and Tubb4b) from that in the control group, while CRE only induced expression changes in Gdf15, as shown using Tukey's test. Gene expression profiles for nine genes (Aen, Bax, Btg2, Ccng1, Cdkn1a, Gdf15, Mbd1, Phlda3, and Plk2) differed.between samples treated with AAF and CRE. Finally, principal component analysis (PCA) of 12 genes (Aen, Bax, Btg2, Ccnf, Ccng1, Cdkn1a, Gdf15, Lrp1, Mbd1, Phlda3, Plk2, and Tubb4b) using our previous Open TG-GATE data plus FFPE-AAF and FFPE-CRE successfully differentiated FFPE-AAF, as GTHC, from FFPE-CRE, as NGHTC. Conclusion:Our results suggest that FFPE RNA-Seq and PCA are useful for evaluating typical rat GTHCs and NGTHCs.

Project description:To clarify difference in the responses on the reprogramming of metabolism toward carcinogenesis between genotoxic and non-genotoxic hepatocarcinogens in the liver, rats were repeatedly administered genotoxic hepatocarcinogens (N-nitrosodiethylamine, aflatoxin B1, N-nitrosopyrrolidine, or carbadox) or non-genotoxic hepatocarcinogens (carbon tetrachloride, thioacetamide, or methapyrilene hydrochloride) for 28, 84, or 90 days. Non-genotoxic hepatocarcinogens revealed transcript expression changes suggestive of suppressed mitochondrial oxidative phosphorylation (OXPHOS) after 28 days and increased glutathione S-transferase placental form-positive (GST-P+) foci downregulating adenosine triphosphate (ATP) synthase subunit beta, mitochondrial precursor (ATPB), compared with genotoxic hepatocarcinogens after 84 or 90 days, suggesting that non-genotoxic hepatocarcinogens are prone to suppress OXPHOS from the early stage of treatment, which is in contrast to genotoxic hepatocarcinogens. Both genotoxic and non-genotoxic hepatocarcinogens upregulated glycolytic enzyme genes and increased cellular membrane solute carrier family 2, facilitated glucose transporter member 1 (GLUT1) expression in GST-P+ foci for up to 90 days, suggesting induction of a metabolic shift from OXPHOS to glycolysis at early hepatocarcinogenesis by hepatocarcinogens unrelated to genotoxic potential. Non-genotoxic hepatocarcinogens increased c-MYC+ cells after 28 days and downregulated Tp53 after 84 or 90 days, suggesting a commitment to enhanced metabolic shift and cell proliferation. Genotoxic hepatocarcinogens also enhanced c-MYC activation-related metabolic shift until 84 or 90 days. In addition, both genotoxic and non-genotoxic hepatocarcinogens upregulated glutaminolysis-related Slc1a5 or Gls, or both, after 28 days and induced liver cell foci immunoreactive for neutral amino acid transporter B(0) (SLC1A5) in the subpopulation of GST-P+ foci after 84 or 90 days, suggesting glutaminolysis-mediated facilitation of cell proliferation toward hepatocarcinogenesis. These results suggest differential responses between genotoxic and non-genotoxic hepatocarcinogens on reprogramming of energy metabolic pathways toward carcinogenesis in liver cells from the early stage of hepatocarcinogen treatment.

Project description:Liver tumors in rodents are frequently induced by non-genotoxic carcinogens. These hepatocarcinogens generally activate hepatic nuclear receptors (e.g., CAR and PXR), resulting in a cascade of signals causing modifications in the expression of genes responsible for several processes involved in carcinogenesis. Evaluation of the carcinogenic potential of chemicals is a regulatory requirement but is time-consuming and expensive. Consequently, several short-term in vivo and in vitro approaches, using molecular tools, have been proposed as predictive models for non-genotoxic hepatocarcinogens. The objective of our study was to discriminate between chemicals that are either non-genotoxic hepatocarcinogens or merely hepatotoxicants and also between CAR and PXR modulators on the basis of their gene expression profiles. Thus, we treated rats for seven days with the hepatoxicants, diclofenac and diazepam, or with several CAR and PXR modulators, which were mainly hepatocarcinogens. Different hepatic gene expression profiles were obtained not only between the hepatotoxicants and the non-genotoxic hepatocarcinogens but also between the CAR activators phenobarbital, phenytoin and 1,1-bis-(4-chlorophenyl)-2,2-dichloroethene which were grouped together, and the two PXR activators pregnenolone 16Î±-carbonitrile and clotrimazole. Diethylstilbestrol had an expression profile that was quite distinct from the other PXR activators, suggesting that this compound is certainly not a classic PXR modulator. Moreover, some differences were observed between phenytoin (not considered as a hepatocarcinogen), and the other two CAR activators. Our data therefore indicate that discrimination is possible between hepatocarcinogens and hepatotoxicants, between CAR and PXR modulators and also between compounds within the same class of modulators using a short-term transcriptomic approach. CAR or PXR inducers were administered in suspension to rats (7 weeks old at start of treatment) by oral gavage at a daily dose for 7 consecutive days. Treatment-related changes in gene expression were determined in the liver using whole genome oligonucleotide microarrays.

Dataset Information

Identification of consensus biomarkers for predicting non-genotoxic hepatocarcinogens.

Publications

Identification of consensus biomarkers for predicting non-genotoxic hepatocarcinogens.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure