Control elements of the tetracycline-resistance operon encoded in Tn10 of Escherichia coli have been utilized to establish a highly efficient regulatory system in mammalian cells. By fusing the tet repressor with the activating domain of virion protein 16 of herpes simplex virus, a tetracycline-controlled transactivator (tTA) was generated that is constitutively expressed in HeLa cells. This transactivator stimulates transcription from a minimal promoter sequence derived from the human cytomegalovirus promoter IE combined with tet operator sequences. Upon integration of a luciferase gene controlled by a tTA-dependent promoter into a tTA-producing HeLa cell line, high levels of luciferase expression were monitored. These activities are sensitive to tetracycline. Depending on the concentration of the antibiotic in the culture medium (0-1 microgram/ml), the luciferase activity can be regulated over up to five orders of magnitude. Thus, the system not only allows differential control of the activity of an individual gene in mammalian cells but also is suitable for creation of "on/off" situations for such genes in a reversible way.

A transcriptional transactivator was developed that fuses the VP16 activation domain with a mutant Tet repressor from Escherichia coli . This transactivator requires certain tetracycline (Tc) derivatives for specific DNA binding. Thus, addition of doxycycline to HeLa cells that constitutively synthesized the transactivator and that contained an appropriate, stably integrated reporter unit rapidly induced gene expression more than a thousandfold. The specificity of the Tet repressor-operator-effector interaction and the pharmacological characteristics of Tc's make this regulatory system well suited for the control of gene activities in vivo, such as in transgenic animals and possibly in gene therapy.

Conditional overexpression of human cyclins B1, D1, and E was accomplished by using a synthetic cDNA expression system based on the Escherichia coli tetracycline repressor. After induction of these cyclins in asynchronous Rat-1 fibroblasts, a decrease in the length of the G1 interval was observed for cyclins D1 and E, consistent with an acceleration of the G1/S phase transition. We observed, in addition, a compensatory lengthening of S phase and G2 so that the mean cell cycle length in populations constitutively expressing these cyclins was unchanged relative to those of their uninduced counterparts. We found that expression of cyclin B1 had no effect on cell cycle dynamics, despite elevated levels of cyclin B-associated histone H1 kinase activity. Induction of cyclins D1 and E also accelerated entry into S phase for synchronized cultures emerging from quiescence. However, whereas cyclin E exerted a greater effect than cyclin D1 in asynchronous cycling cells, cyclin D1 conferred a greater effect upon stimulation from quiescence, suggesting a specific role for cyclin D1 in the G0-to-G1 transition. Overexpression of cyclins did not prevent cells from entering into quiescence upon serum starvation, although a slight delay in attainment of quiescence was observed for cells expressing either cyclin D1 or cyclin E. These results suggest that cyclins D1 and E are rate-limiting activators of the G1-to-S phase transition and that cyclin D1 might play a specialized role in facilitating emergence from quiescence.

The tet regulatory system in which doxycycline (dox) acts as an inducer of specifically engineered RNA polymerase II promoters was transferred into transgenic mice. Tight control and a broad range of regulation spanning up to five orders of magnitude were monitored dependent on the dox concentration in the water supply of the animals. Administration of dox rapidly induces the synthesis of the indicator enzyme luciferase whose activity rises over several orders of magnitude within the first 4 h in some organs. Induction is complete after 24 h in most organs analyzed. A comparable regulatory potential was revealed with the tet regulatory system where dox prevents transcription activation. Directing the synthesis of the tetracycline-controlled transactivator (tTA) to the liver led to highly specific regulation in hepatocytes where, in presence of dox, less than one molecule of luciferase was detected per cell. By contrast, a more than 10(5)-fold activation of the luciferase gene was observed in the absence of the antibiotic. This regulation was homogeneous throughout but stringently restricted to hepatocytes. These results demonstrate that both tetracycline-controlled transcriptional activation systems provide genetic switches that permit the quantitative control of gene activities in transgenic mice in a tissue-specific manner and, thus, suggest possibilities for the generation of a novel type of conditional mutants.

Promoters whose temporal activity can be directly manipulated in transgenic animals provide a tool for the study of gene functions in vivo. We have evaluated a tetracycline-responsive binary system for its ability to temporally control gene expression in transgenic mice. In this system, a tetracycline-controlled trans-activator protein (tTA), composed of the repressor of the tetracycline-resistance operon (tet from Escherichia coli transposon Tn10) and the activating domain of viral protein VP16 of herpes simplex virus, induces transcription from a minimal promoter (PhCMV*-1; see below) fused to seven tet operator sequences in the absence of tetracycline but not in its presence. Transgenic mice were generated that carried either a luciferase or a beta-galactosidase reporter gene under the control of PhCMV*-1 or a transgene containing the tTA coding sequence under the control of the human cytomegalovirus immediate early gene 1 (hCMV IE1) promoter/enhancer. Whereas little luciferase or beta-galactosidase activity was observed in tissues of mice carrying only the reporter genes, the presence of tTA in double-transgenic mice induced expression of the reporter genes up to several thousand-fold. This induction was abrogated to basal levels upon administration of tetracycline. These findings can be used, for example, to design dominant gain-of-function experiments in which temporal control of transgene expression is required.