AML3/CBFα1 is required for androgen-specific activation of the enhancer of the mouse sex-limited protein (Slp) gene

A complex 120-base pair enhancer, derived from the mouse sex-limited protein (Slp) gene, is activated solely by the androgen receptor (AR) in specific tissues, although it contains a hormone response element recognized by several steroid receptors. The generation of this transcriptional specificity has been ascribed to the interactions of the receptor with tissue-specific nonreceptor factors bound to accessory sites within the enhancer. Protein-DNA interaction assays revealed two factors binding the 5' part of the enhancer that differ widely in abundance between cells showing AR-specific activation of the Sip element compared with those that also permit activation by glucocorticoid receptor (GR). The factor designated B formed a complex centered on the sequence TGTGGT, a core motif recognized by members of the AML/CBF alpha transcription factor family. This complex was competed by a high affinity binding site specific for AML/CBF alpha and was specifically supershifted by an antibody to AML3/CBF alpha 1, placing factor B within the AML3/CBF alpha 1 subclass. Interestingly, this factor was shown to bind to a second site in the 3' part of the enhancer, positioned between the two critical AR binding sites. Transfection studies revealed that AML1-ETO, a dominant-negative AML/CBF alpha construct, abrogated AR induction of the enhancer, but not of simple hormone response elements. Furthermore, overexpression of AML3/CBF alpha 1 could rescue the AML1-ETO repression. Finally, glutathione S-transferase-AML/CBF alpha fusion proteins demonstrated direct interaction between AML/ CBF alpha and steroid receptors. Although this interaction was equivalent between AML1/CBF alpha 2 and AR or GR, AML3/CBF alpha 1 showed stronger interaction with AR than with GR. These data demonstrate that AML3/CBF alpha 1 is functionally required for hormonal induction of the Sip enhancer and that direct, preferential protein-protein interactions may contribute to AR-specific activation. These results demonstrate an intriguing role of AML3/ CBF alpha 1 in steroid- as well as tissue-specific activation of target genes.