EFFECTS OF ESTRADIOL-17-BETA ANALOGS ON ACTIVATION OF ESTROGEN RESPONSE ELEMENT REGULATED CHLORAMPHENICOL ACETYLTRANSFERASE EXPRESSION

These experiments were designed to examine the effect of structural modifications to the estradiol-17beta (E2) molecule on the estrogen response element (ERE) dependent activation of the thymidine kinase (tk) promoter. Estrogen receptor (ER) positive MCF-7 cells were transfected with plasmids containing one or two vitellogenin EREs inserted upstream of the tk promoter in p(-37)tk. Transient expression of the CAT gene in these constructs was measured after cells had been maintained for 36-42 h in the presence of E2 or an E2 analogue. E2 induced CAT expression at levels as low as 10(-13) M, with maximum induction at 10(-11) M. CAT activity decreased at higher concentrations of E2. Estratriene, which has low affinity for ER, was active only at micromolar concentrations. 3-Hydroxyestratriene displayed maximal activity at 10(-9) M, with higher levels being less active. Still higher concentrations (10(-7) M) of estratrien-17beta-ol were required to induce maximum CAT activity. All positional and conformational alterations in the D-ring hydroxyl group of E2 yielded active ligands. Movement of the phenolic hydroxyl group of E2 to other positions on the A-ring produced dihydroxyestrogens with varied capacities to activate CAT (2-hydroxyestratrien-17beta-ol produced maximum CAT activation at 10(-11) M; 1-hydroxyestratrien-17beta-ol required a 10(-8) M concentration for maximum activity; 4-hydroxyestratrien-17beta-ol gave maximum CAT activation at 10(-6) M). Only those androstanediols or 5-androstenediols with a 3beta-hydroxyl group were capable of activating CAT expression. CAT constructs with two consensus EREs placed 6 or 19 bp apart were equally active and displayed a response to E2 or to the estrogen analogues similar to that of plasmids with a single consensus ERE. The concentrations of structurally modified estrogens which generated maximum CAT responses from these constructs were directly related to their affinity for ER. This contrasts with results obtained with the more complex regulatory regions of endogenous E2-responsive genes in MCF-7 cells. It is suggested that transcriptional activation following the binding of ER-ligand complexes to the ERE can be modulated by interactions with factors bound to other cis regulatory elements. Such protein-protein interactions appear to be influenced by the structure of the ligand.