Characterization of canine cytochromes P450 2B and 3A

Abstract

The objective of these studies was to functionally characterize canine cytochromes P450 from subfamilies 2B and 3A. Studies of the canine hepatic 2B form PBD-2 had previously revealed that this enzyme exhibits a species specific ability to metabolize 2,2',4,4',5,5'-hexachlorobiphenyl (245-HCB) and catalyze progesterone 21-hydroxylation. Expression of the putative PBD-2 cDNA, 2B11, in three different heterologous systems revealed that the recombinant enzymes' apparent substrate specificities varied based upon the expression system. 2B11 expressed in COS cells and yeast did not metabolize progesterone in a manner consistent with that of the purified hepatic enzyme, suggesting that 2B11 did not encode PBD-2. However, subsequent expression of 2B11 in E. coli confirmed that this recombinant enzyme did metabolize progesterone into 21- and 16α-hydroxyprogesterone. The structural determinants of 2B11-mediated progesterone 21-hydroxylation were then examined via site-directed mutagenesis of amino acid residues 114, 290, 363, and 365. Consistent with the importance of these residues in androstenedione and 245-HCB metabolism, amino acid alterations Val 114 $\to$ Ile, Asp-290 → Ile, and Ile-363 → Val each decreased 2B11 progesterone 21-hydroxylation. 2B11 mutants at position 365 differed in their regioselective metabolism of progesterone, whereas these mutations had little affect on the stereoselective metabolism of androstenedione. Cytochrome P450 3A forms are of intense interest due to the wide range of therapeutic compounds metabolized by these enzymes. To determine if canine liver contained multiple 3A forms which exhibit substrate specificities similar to those of other species, the canine 3A form PBD-1 (3A12) was expressed in E. coli. 3A12, rabbit 3A6 and human 3A4 metabolized steroids and macrolide antibiotics into the same products at comparable rates. A 3-fold difference in the rate of troleandomycin demethylation between liver microsomes and 3A12, the identification of a novel putative 3A hepatic protein via NH₂-terminal sequencing, and isolation of a distinct 3A cDNA provided evidence for 3A multiplicity in canine liver microsomes. The identification of functionally distinct canine 3A forms could provide the framework for structure-function analysis of these clinically important enzymes. Furthermore, 3A multiplicity and the conservation of substrate specificity between canine and human forms suggests that the canine may be an appropriate model of human 3A metabolism.