"Random chimeragenesis" (Kim, J.-Y., and Devreotes, P. N. (1994) J. Biol. Chem. 269, 28724-28731; Levin, L. R., and Reed, R. R. (1995) J. Biol. Chem. 270, 7573-7579) has been used to generate chimeras between two distant P450s involved in catabolism of cholesterol in mammals, P450scc (product of CYP11A1 gene) and P450c27 (product of CYP27 gene). Both are mitochondrial P450s which hydroxylate the side chain of cholesterol. Even though these P450s are only about 25% identical, we wondered whether their similar substrate specificity might permit mapping of the active sites by this technique. Four chimeric DNAs encoding three different proteins have been obtained testifying that short stretches of nucleotide identity (six nucleotides preceding the crossover point) are sufficient for homologous recombination in Escherichia coli, the basis of the random chimeragenesis technique. The N-terminal part of the chimeras was formed by P450scc and the C-terminus was from P450c27. The chimeric P450s have been expressed in E. coli and partially purified. Though they displayed a peak at 453 nm in their CO-difference spectra, indicating that the proteins are properly folded and hemin is correctly incorporated, none of the chimeras had detectable catalytic activity with either cholesterol (substrate for P450scc) or 5beta-cholestane-3alpha,7alpha,12alpha-triol (substrate for P450c27). To investigate whether the chimeras bind substrate, titrations with 22R-hydroxycholesterol have been carried out. Addition of 22R-hydroxycholesterol to recombinant P450scc which is in the low spin form leads to conversion to the high spin form. Titration of chimeras with 22R-hydroxycholesterol did not result in their conversion to the high spin form. However, a shift to 426-430 nm was observed in the difference spectrum of two of the chimeras with a minimum around 406 nm. Thus, addition of 22R-hydroxycholesterol to these chimeras results in binding of the steroid in the enzyme active site and conversion of one low spin form to a different low spin form. Possible explanations for the absence of enzymatic activity and the potential advantages of the random chimeragenesis technique to generate chimeras between different P450s are discussed.