Mastering transgene expression

Spatiotemporal control of transgene expression is of paramount importance in animal engineering: the exact activation/inactivation of your favorite transgene is a powerful tool to trace the hidden rules of integrative biology, and also to strive gene therapy. In mice, gene targeting can be restricted in a tissue and/or temporal manner through generation of 'premutant' mice (e.g., "floxed" or flanked with loxP sites) which are then breed with transgenic mates that opportunely express the corresponding DNA recombinase (e.g., Cre) allowing the desired conditional gene deletion/expression with respect to space and/or time. The opportune design of the loxP flanking sites may be considered an art, as exemplified by the stochastic combinatorial approach by Livet and Lichtman, however in most instances, is the choice of the Cre expression system to be the fundamental step.

Hundreds of mice have been generated with Cre expressed under control of tissue or cell specific promoters, allowing space/anatomical control of the recombination. And other mice were generated to check the correct recombination. Conversely, relatively few systems allow temporal control. Current solutions include the "activation" of Cre by a ligand given opportunely at selected time. Particularly, I read about:

1. The exploitation of a tetracycline-responsive promoter (St Onge, 1996);
2. The use a fusion protein to obtain modulation of Cre activity by steroids like ecdysone (No et al., 1996) or tamoxifen (Danielian et al., 1998);
3. A dimerizable approach, in which two Cre mojeties complement upon binding to a rapamycin molecule (Jullien et al., 2007).

Even if each of these systems compete for low basal activity and high inducibility, unfortunately none of such ligands affects only Cre activity: tetracycline, steroids and the mTOR-inhibitor rapamycin may be considered somewhat like as potential "metabolic/endocrine distruptors". Hence, particularly in animals, the treatment with the ligand may introduce a bias in subsequent endpoint measurements.

Recently, I read about noninvasive in vivo local heating by means of high-intensity focused ultrasound (HIFU) . Deckers and colleagues, implemented such technique in combination with a heat-inducible promoter [heat shock protein 70 (HSP70)] driving luciferase expression. They monitored local hypertermia with MRI thermometry and evaluated gene induction by bioluminescence imaging. Nice. Really cool. Now I'm wondering how can I convince my boss to buy one HIFU, one MRI and one more BLI workstation instead of just another syringe to inject tetracycline?!


R. Deckers, B. Quesson, J. Arsaut, S. Eimer, F. Couillaud, C. T. W. Moonen (2009). Image-guided, noninvasive, spatiotemporal control of gene expression Proceedings of the National Academy of Sciences, 106 (4), 1175-1180 DOI: 10.1073/pnas.0806936106