Metabolism of the vitamin D analog EB1089 by cultured human cells: redirection of hydroxylation site to distal carbons of the side-chain.

1(S),3(R)-dihydroxy-20(R)-(5'-ethyl-5'-hydroxy-hepta-1'(E),3' (E)-dien-1'-yl)-9,10-secopregna-5(Z),7(E),10(19)-triene (EB1089) is a novel synthetic analog of 1 alpha,25-dihydroxyvitamin D [1,25-(OH)2D3] with potential for use in the treatment of hyperproliferative disorders. It has an altered side-chain structure compared to 1,25-(OH)2D3, featuring 26,27 dimethyl groups, insertion of an extra carbon atom (24a) at C-24, and two double bonds at C-22,23 and C-24,24a. In vitro metabolism of EB1089 was studied in a human keratinocyte cell model, HPK1A-ras, previously shown to metabolize 1,25-(OH)2D3. Four metabolites were formed, all of which possessed the same UV chromophore as EB1089, indicating the retention of the side-chain conjugated double bond system. Two metabolites were present in sufficient quantities to identify them as 26-hydroxy EB1089 (major product) and 26a-hydroxy EB1089 (minor product), based on mass spectral analysis and cochromatography with synthetic standards. Similar metabolites were generated in vivo and using a liver postmitochondrial fraction in vitro (Kissmeyer et al., companion paper). Studies with the human hepatoma Hep G2 gave rise to 2 isomers of 26-hydroxy EB1089. Studies using ketoconazole, a general cytochrome P450 inhibitor, implicated cytochrome P450s in the formation of the EB1089 metabolites. COS-1 transfection cell experiments using vectors containing CYP27 and CYP24 suggest that these cytochrome P450s are probably not involved in 26- or 26a-hydroxylation of EB1089. Other experiments that examined the HPK1A-ras metabolism of related analogs containing only a single side-chain double bond: 1(S),3(R)-dihydroxy-20(R)-(5'-ethyl-5'-hydroxy-hepta-1' (E)-en-1'-yl)-9,10-secopregna-5(Z),7(E),10(19)-triene (MC1473; double bond at C-22,23) and 1(S),3(R)-dihydroxy-20(R)-(5'-ethyl-5'-hydroxy-hepta-3'(E)-en-1'-yl)-9, 10-secopregna-5(Z),7(E),10(19)-triene (MC1611; double bond at C-24,24a) revealed that the former compound was subject to 24-hydroxylation and the latter compound was mainly 23-hydroxylated. Metabolism experiments involving EB1089, MC1473, and MC1611 in competition with [1 beta-3H]1,25-(OH)2D3 in HPK1A-ras confirmed that CYP24 is probably not involved in the metabolism of EB1089 whereas, in the case of MC1473 and MC1611, it does appear to carry out side-chain hydroxylation. Our interpretation is that the conjugated double bond system in the side-chain of EB1089 is responsible for directing the target cell hydroxylation to the distal positions, C-26 and C-26a. We conclude that EB1089 is slowly metabolized via unique in vitro metabolic pathways, and that these features may explain the relative stability of EB1089 compared to other analogs in vivo.