were linked either to a minimal promoter sequence derived from the human C-Reactive Protein (CRP) gene (7 x H1/ CRP promoter) or to the TATA box sequence from the adenovirus E1b gene (7 x H1/E1b promoter) (149).
Artificial promoters containing multimeric HNF1 binding sites were silent in muscles and in cell lines that lacked endogenous HNF1a but were strongly activated by HEA-3 in the presence of 4-OHT. The system was half-maximally stimulated at a 4-OHT concentration of about 40 nM. In transiently transfected cells, HEA-3 induced thousands-fold induction of reporter genes in a 4-OHT-dependent manner. Importantly, E2, progesterone, and RU486 did not stimulate the system activators (149).
The system was delivered in mice as plasmid DNA by using muscle electroinjection and induced by oral doses of Tamoxifen (TAM, Fig. 15), which is converted to the active metabolite 4-OHT in vivo. In this experimental setting, HEA-3 upregulated by about 100-fold SEAP gene expression in a strictly TAM-dependent manner. Dose response experiments demonstrated that the system was significantly induced by as low as 1 mg/kg of TAM. HEA-3 also enabled long-term and fully reversible regulation of mouse erythropoietin gene over a 10-month period (149).
5. Inducer Drug: Tamoxifen
Tamoxifen is a clinically validated drug that is currently used to prevent recurrence of breast cancer (155). However, several lines of evidence suggest that the system should be significantly improved in order to meet the safety requirements required for applications in humans.
An oral dose of 1 mg/kg/day of TAM significantly (but not maximally) stimulates HEA-3 in mice (149). This dosage regimen ensures a steady-state 4-OHT plasma concentration in mice of about 30 nM, a value close to the EC50 of the drug for HEA-3 (see above). The plasma concentration of 4-OHT in patients undergoing long-term TAM therapy (20 mg/kg, daily) is about 10 nM: this indicates that TAM dosages higher than those normally used in clinical practice would be required in humans to exploit the full dynamic range of the HEA-3-based system (155-158). Therefore, the sensitivity of this system should be increased by at least one order of magnitude for gene therapy applications in humans. This is of particular relevance when considering that TAM is extremely well tolerated at the usual dosage, but that prolonged treatment (i.e., years) is associated with a low, but well-defined, increased risk of endometrial cancer (157,159).
HEA-3 was specifically designed to work in muscle, as the use of this tissue as a bioreactor for producing therapeutic proteins is an attractive possibility for gene therapy (160). However, HEA-3 can be considered a prototype of a series of humanized regulatory switches. DBDs of transcription factors other than that of HNF1 can thus be selected to construct additional humanized chimeras specifically designed to work in tissues other than muscle. It is tempting to speculate that the combined use of tissue-specific DBDs and of mutant HBDs recognizing only synthetic steroids would generate a whole series of ''customized,'' ligand-dependent, humanized chimeras, specifically designed to work in a restricted range of tissues and cells.
Such customized molecular switches would recognize DNA sites within the chromosomes and might therefore affect endogenous gene expression. This may or may not be a concern, depending of the affected genes. However, it is worth considering that gene transcription in higher eukaryotes is governed by a complex network of interacting activating and repressing factors, whose equilibrium in a specific cell cannot be easily modified by overexpressing a single transcription factor not normally present (161). Therefore, we can hypothesize that molecular switches such as HEA-3 are unlikely to profoundly modify the expression profile of a target cell. This is partially confirmed by the observation that both albumin and a 1-AT genes, whose expression in hepatocytes is regulated by HNF1, are not turned on in HEA-3-transduced muscles (149). Nonetheless, it will be necessary to study this safety issue in greater detail with the help of new sensitive technologies such as DNA microarray or whole-cell proteomics.
HEA-3 uses tamoxifen, a clinically accepted drug that acts as a steroid receptor antagonist. For gene therapy applications, inducer drugs that do not have any effect on endogenous proteins would be more appropriate. Future work should focus on redesigning the HBD-hormone interface in order to differentiate it from endogenous receptors, and to identify novel ligands that only bind to those mutated HBDs, an approach that has been successfully used to generate nonimmunosup-pressive analogs of rapamycin or FK506.
Was this article helpful?