Use of short-term transcriptional profiles to assess the long-term cancer-related safety of environmental and industrial chemicals.
ABSTRACT The process for evaluating chemical safety is inefficient, costly, and animal intensive. There is growing consensus that the current process of safety testing needs to be significantly altered to improve efficiency and reduce the number of untested chemicals. In this study, the use of short-term gene expression profiles was evaluated for predicting the increased incidence of mouse lung tumors. Animals were exposed to a total of 26 diverse chemicals with matched vehicle controls over a period of 3 years. Upon completion, significant batch-related effects were observed. Adjustment for batch effects significantly improved the ability to predict increased lung tumor incidence. For the best statistical model, the estimated predictive accuracy under honest fivefold cross-validation was 79.3% with a sensitivity and specificity of 71.4 and 86.3%, respectively. A learning curve analysis demonstrated that gains in model performance reached a plateau at 25 chemicals, indicating that the size of current data set was sufficient to provide a robust classifier. The classification results showed that a small subset of chemicals contributed disproportionately to the misclassification rate. For these chemicals, the misclassification was more closely associated with genotoxicity status than with efficacy in the original bioassay. Statistical models were also used to predict dose-response increases in tumor incidence for methylene chloride and naphthalene. The average posterior probabilities for the top models matched the results from the bioassay for methylene chloride. For naphthalene, the average posterior probabilities for the top models overpredicted the tumor response, but the variability in predictions was significantly higher. The study provides both a set of gene expression biomarkers for predicting chemically induced mouse lung tumors and a broad assessment of important experimental and analysis criteria for developing microarray-based predictors of safety-related end points.
- SourceAvailable from: PubMed Central[Show abstract] [Hide abstract]
ABSTRACT: Translation elongation factor-1d (TEF-1δ) has been identified as a novel cadmium-responsive proto-oncogene. However, it is still unclear whether TEF-1δ could be a potential biomarker of cadmium exposure. Rats were treated with CdCl at different concentrations (high dose 1.225, mid-dose 0.612 and low dose 0.306 mg/kg body weight, respectively) for 14 weeks, and the cadmium levels, weight coefficients, serum alanine aminotransferase (ALT), aspartate aminotransferase (AST), blood urea nitrogen (BUN), serum creatinine (SCR), 24-h urine protein (24hPro), urinary creatinine (Cr) and pathological features were determined. The TEF-1δ expression in white blood cells and multiple organs were examined by reverse transcription polymerase chain reaction (PCR) and were also confirmed with fluorescence quantitative PCR. A cadmium dose-dependent increase (p < 0.05) of cadmium levels in blood, urine, liver, kidney, heart and lung, and the weight coefficients was observed. The liver and renal function indictors including AST, ALT, SCR, BUN and 24hPro, were elevated in a cadmium dose-dependent manner (p < 0.05). Significant pathological changes in liver, kidney, heart and lung were indicated. The TEF-1δ expression was up-regulated in both blood and organs (p < 0.05). Moreover, the expression level of blood TEF-1δ was positively correlated to TEF-1δ expression level, cadmium level and toxicity in the organs (p < 0.01). This study indicates that blood TEF-1δ is a novel valuable biomarker for cadmium exposure and its organ toxicity.International Journal of Molecular Sciences 01/2013; 14(3):5182-97. · 2.46 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: High throughput transcriptomics profiles such as those generated using microarrays have been useful in identifying biomarkers for different classification and toxicity prediction purposes. Here, we investigated the use of microarrays to predict chemical toxicants and their possible mechanisms of action. In this study, in vitro cultures of primary rat hepatocytes were exposed to 105 chemicals and vehicle controls, representing 14 compound classes. We comprehensively compared various normalization of gene expression profiles, feature selection and classification algorithms for the classification of these 105 chemicals into14 compound classes. We found that normalization had little effect on the averaged classification accuracy. Two support vector machine (SVM) methods, LibSVM and sequential minimal optimization, had better classification performance than other methods. SVM recursive feature selection (SVM-RFE) had the highest overfitting rate when an independent dataset was used for a prediction. Therefore, we developed a new feature selection algorithm called gradient method that had a relatively high training classification as well as prediction accuracy with the lowest overfitting rate of the methods tested. Analysis of biomarkers that distinguished the 14 classes of compounds identified a group of genes principally involved in cell cycle function that were significantly downregulated by metal and inflammatory compounds, but were induced by anti-microbial, cancer related drugs, pesticides, and PXR mediators. Our results indicate that using microarrays and a supervised machine learning approach to predict chemical toxicants, their potential toxicity and mechanisms of action is practical and efficient. Choosing the right feature and classification algorithms for this multiple category classification and prediction is critical.BMC Genomics 03/2014; 15(1):248. · 4.40 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: There is a high need to improve the assessment of, especially non-genotoxic, carcinogenic features of chemicals. We therefore explored a toxicogenomics-based approach using genome-wide microRNA and mRNA expression profiles upon short-term exposure in mice. For this, wild-type mice were exposed for seven days to three different classes of chemicals, i.e., four genotoxic carcinogens (GTXC), seven non-genotoxic carcinogens (NGTXC), and five toxic non-carcinogens. Hepatic expression patterns of mRNA and microRNA transcripts were determined after exposure and used to assess the discriminative power of the in vivo transcriptome for GTXC and NGTXC. A final classifier set, discriminative for GTXC and NGTXC, was generated from the transcriptomic data using a tiered approach. This appeared to be a valid approach, since the predictive power of the final classifier set in three different classifier algorithms was very high for the original training set of chemicals. Subsequent validation in an additional set of chemicals revealed that the predictive power for GTXC remained high, in contrast to NGTXC, which appeared to be more troublesome. Our study demonstrated that the in vivo microRNA-ome has less discriminative power to correctly identify (non-)genotoxic carcinogen classes. The results generally indicate that single mRNA transcripts do have the potential to be applied in risk assessment, but that additional (genomic) strategies are necessary to correctly predict the non-genotoxic carcinogenic potential of a chemical.Archives of Toxicology 01/2014; · 5.22 Impact Factor