Address reprint requests to: Miguel Beato, MD, Institut fur Molekularbiologie und Tumorforschung, Philipps Universitat, Emil-Mannkopff Strasse 2, 35037 Marburg, Germany.* Supported by the Human Capital and Mobility Program of the European Union (AS), and by grants from the Deutsche Forschungsgemeinschat and der Fond der chemischen Industrie. The experimental work summarized in this paper has been supported by a fellowship of the European Union to AS, and by grants from the European Union, the Deutsche Forschungsgemeinschaft and the Fond der chemischen Industrie. of their target sequences on DNA at the end of the 1980s, the question of how steroid hormones modulate the activity of various genes in target cells seemed to be solved. It was clear that steroid hormone receptors are ligand-activated transcriptional modulators, which in most cases act through binding to specific sequences on DNA called hormone-responsive elements (HREs)(1). Careful mutational analysis identified a modular structure of the receptors composed of a DNA-binding domain, nuclear localization signals, a ligand-binding domain, and several transactivation domains (2). Interactions of the receptors with components of the basal transcriptional machinery and with sequence-specific transcription factors were assumed to mediate their transcriptional effects. The prevalent opinion was that the mechanism of action of steroid hormones was a closed chapter ready for the textbooks and what remained to be done to complete the picture was merely the collection of details. Many of the leading laboratories in the field moved on to the study of what seemed to be more relevant questions, such as the role of the receptors in mouse physiology and development, as deduced from the targeted disruption of the receptor genes. However, the initial burst of data represented just a catalogue of some of the components of the system, and the question of how the hormone-receptor complex really does regulate transcriptional efficiency in the intact cells remained wide open. Hence, although the steroid receptors were among the first transcription factors from higher eukaryotes to be identified and studied in detail, the first round of results did not add essential new insights into eukaryotic-specific regulatory mechanisms.

On the other hand, in the last 5 yr our knowledge of transcriptional control in animal cells has been vastly increased with the revelation of eukaryote-specific mechanisms. Indeed more polypeptides involved in eukaryotic transcription initiation have been identified in the past 5 yr than in all the previous history of the field. This allows a reformulation of the mechanisms of transcriptional control by steroid hormone receptors on a much more solid basis. In revising the wealth of recent literature we will focus on transcription initiation and will deliberately omit many aspects of receptor biology not directly involved in transcriptional control, eg their interactions with heat shock proteins and the cross-talk with other signaling pathways acting at the level of the cell membrane. Although these interactions may ultimately impinge on the transcriptional properties of the receptors, these effects are indirect, and a detailed description would deviate from the main focus of this review. Given