Résumé :
|
Most current gene therapy strategies for inherited diseases are based on a complementation approach: a virus-borne functional copy of the mutant gene is randomly inserted into the genome, resulting in a phenotypic correction of the genetic defect. In contrast, targeted approaches, including targeted insertion, and in their more elaborate form, gene correction, should alleviate the odds of random insertion, such as gene extinction and activation of proto-oncogenes. The recent development of artificial endonucleases with tailored specificities has provided tools for such targeted strategies: redesigned endonucleases cleaving chosen sequences may be used in genome surgery to correct mutated genes or introduce transgenes in chosen loci. Thus, precise genome manipulations such as gene repair, gene replacement or gene inactivation can be realistically envisioned for treatment of genetic diseases. Artificial fusion proteins including Zinc-Finger binding domains have provided important proofs of concept. However, the toxicity of these proteins, which might stem from off site cleavage, is still an issue. Custom-designed meganucleases could represent an efficient alternative. Natural meganucleases belong to a widespread family of proteins encoded by mobile genetic elements. Their function is to trigger targeted recombination. We have designed several meganucleases targeting genes involved in Xeroderma Pigmentosum, SCID, thalassemia, and other genetic diseases that could be treated by cell therapy. These highly specific engineered endonucleases allow for up to 1% of gene correction in cells, and no toxic effect has been detected so far upon cell treatment. Although further characterization will be required for therapeutic use, the combined properties of these proteins (activity and specificity) may qualify them as ideal tools for genome surgery.
|