Abstract
Estrogen receptor beta (ERbeta) has been reported to have lower estradiol (E2)-induced transcriptional activity than human (h)ERalpha from estrogen response element (ERE)-driven reporters in transiently transfected cells. Conflicting data for activities of full-length and short hERbeta [hERbeta1, 530 amino acids (aa); and hERbeta1s, 477aa] have been reported. To test the hypothesis that hERbeta1 has higher transcriptional activity than hERbeta1s, we compared E2, 2,3-bis(4-hydroxyphenyl)propionitrile (a selective ERbeta agonist), and resveratrol-induced transcription by hERbeta1, hERbeta1s, and rat (r) ERbeta with hERalpha on different EREs in transiently transfected CHO-K1 and HEC-1A cells. Our results demonstrate for the first time that hERbeta1 has similar E2-induced activity to hERalpha and greater activity than rERbeta or hERbeta1s on a consensus palindromic ERE, either as a single or double copy; a minimal ERE; and the nonpalindromic pS2 ERE. 2,3-Bis(4-hydroxyphenyl)propionitrile showed greater efficacy with hERbeta1 and hERbeta1s than for rERbeta or hERalpha. We found that transcriptional differences between the ERbeta isoforms and ERalpha depend on the ERE sequence, confirming that the DNA sequence bound by ER is an allosteric effector of ER action. For the minimal 13-bp ERE and the pS2 ERE, the increase in transcriptional activity with hERbeta1 correlated with increased binding affinity. Coactivators steroid receptor coactivator-1 and cAMP response element binding protein-binding protein synergistically activated hERalpha and ERbeta transcription and showed reduced efficacy with rERbeta and hERbeta1s, suggesting a role for the N terminus of ERbeta1 in coactivator interaction. Collectively, these data indicate that the cellular expression of ERbeta isoforms may differentially impact ERE-regulated target gene expression in a ligand-dependent manner.
