In this study, the expression of several genes involved in cortisol synthesis in head kidneys, the site of cortisol production, and in the rainbow trout ( Oncorhynchus mykiss) was examined in response to two different acute stressors and an acute ACTH treatment. mRNAs levels of the “steroidogenic acute regulatory” (StAR) sterol transport protein, which transports cholesterol to the inner mitochondrial membrane as well as cytochrome P450 cholesterol side chain cleavage (P450 SCC) were determined in head kidney (containing the interrenal tissue). In one experiment, we also quantified 3-β-hydroxysteroid dehydrogenase (3B-HSD) and cytochrome P450 11β (11B-H) mRNAs. The presence of these four transcripts in the head kidney was confirmed by Northern blot analysis. For each stress condition, mRNA levels were quantified by quantitative or real-time RT-PCR. The results of these two methods were highly correlated. An acute stress induced by capture, short confinement (2 min), and anesthesia (3 min) resulted in significant elevation of plasma cortisol (30-fold higher than controls) and an increase in levels of StAR and P450 SCC mRNAs 3 h post-stress. When fish were submitted to an acute stress caused by 5 min of chase with a net in a tank, plasma cortisol reached a peak within 1 h, but after 3 h, levels were only 5-fold higher in stressed trout than in controls and no variations in the expression of StAR, P450 SCC, 3B-HSD, and 11B-H were observed whatever the time post-stress. One hour after acute ACTH stimulation (5 IU/kg), plasma cortisol level was 4-fold higher than in control trout and no changes in StAR and P450 SCC mRNAs levels were detected. The data suggest that the high levels of cortisol after stress need an activation of genes involved in cortisol synthesis, but lower levels do not. Futhermore, under these three test conditions, we always found a strong positive correlation between mRNA levels of StAR and P450 SCC, in contrast to what has been described in mammals. Consequently, the absence of transcription activation with low increase in cortisol levels suggests that other levels of regulation, particularly activation of pre-existing proteins, govern cortisol production.