Artificial regulatory networks and cascades for discrete multilevel transgene control in mammalian cells

'Prototype drug-adjustable heterologous transcription control systems designed for gene therapy applications typically show sigmoid dose-response characteristics and enable fine-tuning of therapeutic transgenes only within a narrow inducer concentration range of a few nanograms. However, the design of clinical dosing regimes which achieve tissue-specific concentrations with nanogram precision is yet a "mission impossible." Therefore, most of today's transcription control systems operate as ON/OFF switches and not in a true adjustable mode. The availability of robust transcription control configurations which lock expression of a single therapeutic transgene at desired levels in response to fixed clinical doses of different inducers rather than minute concentration changes of a single inducer would be highly desirable. Based on in silico predictions, we have constructed a variety of mammalian artificial regulatory networks by interconnecting the tetracycline- (TETOFF) streptogramin- (PIPOFF), and macrolide- (E-OFF) repressible gene regulation systems as linear (auto)regulatory cascades. These networks enable multilevel expression control of several transgenes in response to different antibiotics or allow titration of a single transgene to four discrete expression levels by clinical dosing of a single antibiotic: 1) high expression in the absence of any antibiotic (+++), 2) medium level expression following addition of tetracycline (++), 3) low level expression in response to the macrolicle erythromycin (+), and 4) complete repression by streptogramins such as pristinamycin (-). The first-generation artificial regulatory networks exemplify modular interconnections of different heterologous gene regulations systems to achieve multigene expression, fine-tuning, or to design novel control networks with unprecedented transgene regulation properties. Such higher-level transcription control modalities will lead the way towards composite artificial regulatory networks able to effect complex therapeutic interventions in future gene therapy and tissue engineering scenarios. (C) 2003 Wiley Periodicals.