Estrogens exert their regulatory transcriptional effects, which can be stimulatory or repressive, at diverse gene sites via two estrogen receptors, ERalpha and ERbeta. Since these two ERs have different tissue distributions, ligands that have the capacity to selectively activate or inhibit these two ERs would be useful in elucidating the biology of these two receptors and might assist in the development of estrogen pharmaceuticals with improved tissue selectivity. We have developed several ligands that showed ERalpha or ERbeta selectivity at promoter-gene sites containing consensus estrogen response elements (EREs): ERalpha-selective agonist (propyl-pyrazole-triol (PPT)), ERalpha-selective antagonist (methyl-piperidino-pyrazole (MPP)), ERbeta-potency selective agonist (diarylpropionitrile (DPN)) and ERbeta-selective antagonist/ERalpha-agonist (R,R-tetrahydrochrysene (R,R-THC)). In this study, we have examined the activity of these compounds at a range of gene sites where ER stimulates gene expression through non-consensus EREs (complement C3), or multiple half-EREs (NHE-RF/EBP50), or by tethering to DNA via other proteins (TGF beta3 and progesterone receptor A/AP-1), and at gene sites where ER represses gene transcription (interleukin-6). At all of these genes, PPT showed full stimulation through ERalpha while displaying no agonism through ERbeta. MPP antagonized estradiol actions on gene transactivation and transrepression through ERalpha, with little or no effect on transcription mediated through ERbeta. DPN displayed subtype-selective agonism, being ca. 30-fold more potent through ERbeta. R,R-THC was a complete antagonist through ERbeta and displayed agonism through ERalpha, the level of which was promoter dependent. Because these ligands maintain their agonist or antagonist character and ER subtype-selectivity at gene sites of diverse nature, where estradiol is either stimulatory or inhibitory, these compounds should prove useful in elucidating the biological functions of ERalpha and ERbeta.