Potent vitamin D 3 analogs: Their abilities to enhance transactivation and to bind to the vitamin D 3 response element

Abstract

1,25 dihydroxyvitamin D 3 [1,25(OH) 2D 3] mediates its biological activities through specific binding to the vitamin D 3 receptor (VDR) and subsequent association with vitamin D 3 responsive elements (VDRE) in genes modulated by 1,25(OH) 2D 3. Several novel vitamin D 3 compounds (Cmpds) have recently been identified which have 5- to 1000-fold greater abilities to induce differentiation and to inhibit proliferation of HL-60 leukemic blast cells as compared to the parental 1,25(OH) 2D 3 (code name, Cmpd C). To clarify the mechanism by which five of these vitamin D 3 analogs [1,25(OH) 2-16ene-D 3, (Cmpd HM); 1,25(OH) 2-16ene-23yne-D 3, (Cmpd V); 1,25(OH) 2-16ene-23yne-26,27 F 6-D 3; 22-Oxa-1,25(OH) 2D 3; 1,25(OH) 2-23yne-D 3] mediate their remarkably potent biological activities, we have investigated their abilities in HL-60 cells to transactivate a chloramphenicol acetyl transferase (CAT) reporter gene containing a VDRE from the human osteocalcin gene attached to a thymidine kinase minimal promoter. Also, the abilities of the analogs to enhance the binding of the human recombinant VDR/retinoic X receptorα (RXRα) heterodimer to the VDRE were examined in gel mobility shift assays. In serumless cultures, a series of potent vitamin D 3 analogs had comparable abilities to transactivate the reporter gene as did the biologically less potent 1,25(OH) 2D 3 (≈ 15–20-fold stimulation in cultures containing 2 × 10 −8M of vitamin D 3 cmpds). Biologically very weak inducers of differentiation of HL-60 [24R,25(OH) 2D 3; 25(OH)-16ene-23yne-D 3] had markedly diminished abilities to induce transactivation. Dose-response studies of Cmpds C, V, HM (10 −7–10 −11M) showed that in serumless culture conditions, transactivation of the VDRE-CAT was similar; however, in the presence of serum, Cmpd C at 10 −9 M had 20-fold less activity than analogs V and HM. These results may reflect increased binding of Cmpd C to the D binding protein (DBP) in serum as compared to the lower binding affinities for DBP by Cmpds HM and V. Affinities of the biologically potent analogs for VDR did not parallel their abilities either to transactivate VDRE-CAT or to mediate a biological affect on HL-60 cells. In further studies, gel mobility shift assays showed that VDR alone did not have detectable binding to VDRE; likewise, VDR plus RXR had little binding to VDRE in the absence of ligand. In contrast, biologically active vitamin D 3 compounds (Cmpds HM, C, V) in a dose-dependent fashion enhanced the VDR/RXR (retinoid X receptor)-VDRE retarded band. Cmpds HM and C(10 −9–10 −7M) produced a greater enhancement of the retarded band than did Cmpd V, perhaps reflecting the lower binding affinity of the latter Cmpd for VDR. In summary, the transactivational studies suggest that the differential potencies of the new vitamin D 3 analogs may relate in part to their binding affinities to DBP. This is unlikely to be the entire explanation because we find a lack of parallelism between the rank order of the potent vitamin D 3 analogs to mediate their biological activities and their abilities to bind VDR, to transactivate a VDRE and to enhance the magnitude of the VDR/RXR-VDRE retarded band. Further studies should focus on differential metabolism of the analogs as well as potential differences in conformational changes of the VDR/RXR when bound to the vitamin D 3 ligands.