The stringency and magnitude of androgen-specific gene activation are combinatorial functions of receptor and nonreceptor binding site sequences.

The mechanism by which specific hormonal regulation of gene expression is attained in vivo is a paradox in that several of the steroid receptors recognize the same DNA element in vitro. We have characterized a complex enhancer of the mouse sex-limited protein (Slp) gene that is activated exclusively by androgens but not by glucocorticoids in transfection. Potent androgen induction requires both the consensus hormone response element (HRE) and auxiliary elements residing within the 120-bp DNA fragment C' delta 9. Multiple nonreceptor factors are involved in androgen specificity, with respect to both the elevation of androgen receptor activity and the inactivity of glucocorticoid receptor (GR), since clustered base changes at any of several sites reduce or abolish androgen induction and do not increase glucocorticoid response. However, moving the HRE as little as 10 bases away from the rest of the enhancer allows GR to function, suggesting that GR is repressed by juxtaposition to particular factors within the androgen-specific complex. Surprisingly, some sequence variations of the HRE itself, within the context of C' delta 9, alter the stringency of specificity, as well as the magnitude, of hormonal response. These HRE sequence effects on expression correspond in a qualitative manner with receptor binding, i.e., GR shows a threefold difference in affinities for HREs amongst which androgen receptor does not discriminate. Altering the HRE orientation within the enhancer also affects hormonal stringency, increasing glucocorticoid but not androgen response. The effect of these subtle variations suggests that they alter receptor position with respect to other factors. Thus, protein-protein interactions that elicit specific gene regulation are established by the array of DNA elements in a complex enhancer and can be modulated by sequence variations within these elements that may influence selection of precise protein contacts.     

Multiple components of a complex androgen-dependent enhancer.

Sex-limited protein (Slp) is expressed in adult male mice. A 160-basepair fragment 2 kilobases upstream of the gene serves as an androgen-dependent enhancer of chloramphenicol acetyltransferase expression in transient transfection assays in cells with endogenous or cotransfected androgen receptor. One element that is necessary, but not sufficient, for induction is a consensus glucocorticoid (or hormone) response element (HRE). This element binds to the mouse androgen receptor in vitro, but with apparent weak affinity. Induction by the HRE is greatly augmented by an accessory sequence within the 160 basepairs, suggesting that cooperative interactions confer strong response to androgen. Additional elements within the enhancer modulate induction, positively or negatively, and exhibit cell-specific behavior. Of particular interest are two degenerate HREs that are adjacent to the consensus sequence; they show no independent activity, but are functionally significant in conjunction with other elements. The complexity of this enhancer may reflect biological mechanisms that ensure specificity of hormonal response and allow gene expression to respond to changes in hormone concentration.     

Determinants of target gene specificity for steroid/thyroid hormone receptors

The molecular specificity of the receptors for steroid and thyroid hormones is achieved by their selective interaction with DNA binding sites referred to as hormone response elements (HREs). HREs can differ in primary nucleotide sequence as well as in the spacing of their dyadic half-sites. The target gene specificity of the glucocorticoid receptor can be converted to that of the estrogen receptor by changing three amino acids clustered in the first zinc finger. Remarkably, a single Gly to Glu change in this region produces a receptor that recognizes both glucocorticoid and estrogen response elements. Further replacement of five amino acids in the stem of the second zinc finger transforms the specificity to that of the thyroid hormone receptor. These findings localize structural determinants required for discrimination of HRE sequence and half-site spacing, respectively, and suggest a simple pathway for the coevolution of receptor DNA binding domains and hormone-responsive gene networks.     

The ligand insertion hypothesis in the genomic action of steroid hormones

Gene regulation by steroids is tightly coupled to hormone concentration and stereochemistry. A key step is binding of hormones to receptors which interact with consensus DNA sequences known as hormone response elements (HREs). The specificity and strength of hormone binding do not correlate well with hormonal activity suggesting an additional step involving recognition of ligand by the gene. Stereospecific fit of hormones between base pairs and correlation of fit with hormonal activity led to the proposal that such recognition involves insertion of hormone into DNA. Here, the feasibility of insertion was investigated using computer models of the glucocorticoid receptor DNA binding domain bound to its HRE. The site reported to accommodate glucocorticoids was found in the HRE and was exposed to permit unwinding at this locus. The resulting cavity in the unwound DNA/receptor interface fit cortisol remarkably well; cortisol formed hydrogen bonds to both the receptor and DNA. Current experimental evidence is generally consistent with ligand binding domains of receptors undergoing a conformational change which facilitates transfer of the ligand into the unwound DNA/receptor interface. We propose this step is rate limiting and alterations in receptor, DNA or hormone which attenuate insertion impair hormonal regulation of gene function.     

Differential gene activation by glucocorticoids and progestins through the hormone regulatory element of mouse mammary tumor virus

The hormone regulatory element (HRE) of mouse mammary tumor virus can mediate activation of an adjacent promoter by glucocorticoids and progestins. A detailed comparison of the DNA binding of receptors for both hormones using DNAase I footprinting and methylation protection detects clear differences in their interactions with the HRE region between positions -130 and -100. Binding studies and gene transfer experiments with a variety of mutants covering the entire HRE demonstrate differences in the relevance of the individual sequence motifs for induction by each hormone. The influence of changes in the angular orientation of receptor binding sites is also different for glucocorticoid and progesterone induction. In transfection experiments with mutated HREs, we find a functional cooperation between the receptor binding sites that does not correlate with variations in the in vitro affinity of the receptors for the corresponding DNA fragment.     

Potential interactions between estrogen receptor and thyroid receptors relevant for neuroendocrine systems

Environmental signals can profoundly affect reproductive behavior, physiology and responses to steroids. One consequence of nutritional or temperature stress is altered plasma concentrations of thyroid hormone. Recent in vivo and in vitro data indicate that manipulations of estrogen and thyroid hormone levels can alter each other's functions. One possible mechanism for interaction may be that thyroid and estrogen receptors bind to parts of the same hormone response elements of target genes and compete with each other, thus serving to integrate environmental signals with neuroendocrine responses.     

The Mechanism of σ-Crystallin Gene Regulation: Cooperation of Lens-Specific and Non-Specific Elements1

Employing δ-crystallin gene as a model, we have investigated tissue-specific gene regulation. Our approach was to analyze regulatory elements associated with the gene utilizing gene transfer techniques. Introduction of the chicken δ1-crystallin gene into the genome of developing mouse embryos resulted in lens-specific expression, indicating that the elements governing the tissue specificity are located in the DNA sequence introduced. Through analysis of various regions of the δ1-crystallin gene and the associated DNA sequences, we identified a lens-specific enhancer in the third intron of the gene. It was demonstrated that this enhancer alone is sufficient to account for lens specificity of the δ1-crystallin gene. Dissection of the δ1-crystallin enhancer and functional assessment by multiplication of enhancer fragments demonstrated the cooperative interaction of lens-specific and nonspecific elements in the enhancer. The mechanism by which heterologous elements cooperate in generating enhancer activity unquestionably provides great flexibility to the regulatory system, and may account for developmental modulation of gene activity superimposed on tissue specificity.