Background

The aim of this work is to use  a genome-wide transcriptomics analysis, to identify gene signatures that could predict the carcinogenic potential of compounds to the kidney and help to discriminate between genotoxic and non-genotoxic carcinogens. 

Method

Normal Rat Kideny cell line (NRK-52E) was exposed individually to 5 genotoxic (GTX) or 5 non-genotoxic (NGTX) rat renal carcinogens, and to 5 non-carcinogenic compounds at the IC₁₀ concentration for 72h. Controls were exposed 0.1% DMSO. Total RNA was isolated from cells exposed for 6h, 24h and 72h,  purified and quality checked. Microarray analysis was performed using Affymetrix rat GeneChip 230 2.0. Analysis of gene changes was performed using GeneSpring GX11 software from Agilent. The data was normalised using a GC-RMA algorithm. Statistical analysis was performed between control and treated cells on genes with a 2-fold threshold change using a p-value of <0.05 with Benjamin-Hochberg correction.

Results

In all three groups more genes were up-regulated or down-regulated after 72h exposure. However, some compounds induced a low number of gene changes e.g. bromodichloromethane -NGTX, while others produced a very large number e.g. ochratoxin A-NGTX. Principal Components Analysis of total gene showed that the chemicals did not cluster into clearly defined groups.

Conclusion

The time course of gene changes was variable across all compounds, some have a marked effect within 6h while others took up to 72h. NGTX carcinogens induced more gene changes and deregulated more pathways than GTX carcinogens, implicated a more diverse mechanism of action.