A negative tyrosine aminotransferase gene element that blocks glucocorticoid modulatory element-regulated modulation of glucocorticoid-induced gene expression.

Tyrosine aminotransferase (TAT) is the prototypic steroid-inducible gene. Recently, we have found that the modulation of TAT induction properties is reproduced by a novel cis-acting TAT gene element, the glucocorticoid modulatory element (GME). This GME lies about 1 kb upstream of the glucocorticoid response elements (GREs) of the TAT gene and binds a heterooligomer of two recently defined proteins. We now report the existence of an additional TAT gene element between the GME and the GREs that blocks the action of the GME and thus prevents the left shift in the glucocorticoid dose-response curve caused by the GME. This negative element has the properties of a silencer because its activity is relatively position- and orientation-independent. The interaction appears to be stoichiometric in that the effects of a single negative element can be overcome by a second GME. This negative element also has an intrinsic inhibitory activity in the absence of the GME. The majority of the negative element activity could be elicited by a 56-bp sequence between -3105 and -3050 bp of the TAT gene. Multiple, clustered mutations of this sequence reduced, but did not eliminate, the negative activity. Further efforts to restrict the negative element were unsuccessful, suggesting that multiple sequences are required for full activity. High affinity, sequence-specific binding of a trans-acting factor(s) was observed in gel shift assays. This binding was half-maximally competed by a 4.4-fold excess of nonradioactive probe and was very stable once formed (delta H [symbol: see text] dissoc. = 32 kcal/mol), suggesting that low concentrations of a high affinity binding protein(s) exist in nuclear extracts. Further support for this conclusion came from the observation that cotransfection of a plasmid containing multiple copies of the 56-bp negative element was able to relieve the negation of GME activity in a GME-56-bp-GRE reporter construct. These data directly support the role of a trans-acting factor(s) in binding to the 56-bp negative element and blocking GME activity. Collectively, these data suggest that glucocorticoid induction of TAT gene expression is subject to multiple levels of control by several new cis-acting elements and thus is much more complex than previously appreciated.