Analysis of gene regulation after bisphenol-A (BPA) challenge has become a major focus of toxicologic research, and gene-expression analysis through real-time quantitative RT-PCR (qPCR) requires appropriate normalization. Since activity of most genes tends to vary depending upon physiologic state and under different conditions, it is important to identify and validate stable reference genes for data normalization and analysis. The expression of five candidate reference genes was studied in two tissues (brain and liver) from Catla catla after exposure to graded concentrations of BPA (10, 100, and 1000 µg/l) for 14 days. Expression of each gene was plotted relative to control, and stability of candidate reference genes was determined using the algorithmic models NormFinder and Bestkeeper. Expression of two biomarker genes was studied, i.e., vitellogenin (vtg) in liver and aromatase (cyp19b) in brain. In liver, expression of gapdh, eef1a, and actb was strongly regulated by BPA treatment. However, in brain, actb, 18S, and of the predicted product size were the most affected genes. Moreover, the magnitude of vtg expression in liver varied when normalized to different reference genes. In brain, cyp19b expression showed an inverted U-shaped curved when normalized to gapdh, eef1a, and 18S, but an increasing trend was observed when normalized to actb and tbp. Our study shows that the abundance and expression of most genes were treatment and tissue dependent and pre-validation of internal control reference genes is very important for toxicologic studies that entail gene expression.