Discriminating Different Classes of Toxicants by Transcript Profiling

Abstract

Male rats were treated with various model compounds or the appropriate vehicle controls. Most substances were either well-known hepatotoxicants or showed hepatotoxicity during preclinical testing. The aim of the present study was to determine if biological samples from rats treated with various compounds can be classified based on gene expression profiles. In addition to gene expression analysis using microarrays, a complete serum chemistry profile and liver and kidney histopathology were performed. We analyzed hepatic gene expression profiles using a supervised learning method (support vector machines; SVMs) to generate classification rules and combined this with recursive feature elimination to improve classification performance and to identify a compact subset of probe sets with potential use as biomarkers. Two different SVM algorithms were tested, and the models obtained were validated with a compound-based external cross-validation approach. Our predictive models were able to discriminate between hepatotoxic and nonhepatotoxic compounds. Furthermore, they predicted the correct class of hepatotoxicant in most cases. We provide an example showing that a predictive model built on transcript profiles from one rat strain can successfully classify profiles from another rat strain. In addition, we demonstrate that the predictive models identify nonresponders and are able to discriminate between gene changes related to pharmacology and toxicity. This work confirms the hypothesis that compound classification based on gene expression data is feasible.

Meeting Report: Validation of Toxicogenomics-Based Test Systems: ECVAM–ICCVAM/NICEATM Considerations for Regulatory Use

“…Gene expression profiling has been used to classify compounds by chemical class and mechanism (Hughes et al 2000; Scherf et al 2000; Steiner et al 2004; Thomas et al 2001), tumors by origin and type (Chung et al 2002), and breast cancer patients for follow-up chemotherapy (van ‘t Veer et al 2002). In all cases, classification was based on a set of discriminatory gene elements, between 10 and several hundred, identified from a larger pool of genes on a microarray.…”

Data mining reveals a network of early-response genes as a consensus signature of drug-induced in vitro and in vivo toxicity

“…1, 2, 3 Over the past 10 years, gene expression profiling has been introduced into drug development to predict and understand toxicity in pre-clinical settings, either as a stand-alone method 4, 5, 6, 7 or integrated within systematic approaches. 8, 9, 10, 11, 12, 13 …”

Identification of biomarkers that distinguish chemical contaminants based on gene expression profiles

“…For instance, gene expression profiles have been successfully applied to the classification of toxicants in rodents [7-11], and discriminate between hepatotoxic and nonhepatotoxic chemical compounds [12,13]. Similarly, this method was used to successfully distinguish genotoxic from nongenotoxic carcinogenetic chemicals by gene expression profiles in primary mouse hepatocyes [14].…”

“…The liver is the major site of chemical metabolism and a principle organ affected by the toxicity of chemical compounds [12,25,26]. Primary cultured cells such as hepatocytes offer a convenient in vitro system that can easily be manipulated and used to screen chemicals for toxicity using different molecular and biochemical methods.…”