Retroviruses and molecular evolution

This project concerns the development of a new system of randomised evolution exploiting the ability of HIV replication machinery to introduce mutations and generate recombination to drive evolution of non-viral genes of interest in cancer research.

Randomised mutagenesis of genes and directed evolution strategies (repeated rounds of mutation and selection in order to isolate proteins with improved or new function) have deserved increasing interest both for purposes of applied research as well as for developing tools for basic research. Classically, a library of mutated genes is first generated and then screened for the presence of mutants possessing a given property either in vitro or in bacterial cells. When eukaryotic genes are targeted, the properties observed in vitro for a given mutant often do not result in the desired phenotype when introduced in the human cell, though (figure 4).

This is essentially due to the complexity of these cells for which a phenotype is the resultant of concomitant effects of different sets of genes that ensure the robustness of the expression of the phenotype and allow its subtle regulation.

We developed a new procedure aimed at bypassing these problems by allowing the generation and the screening of libraries of mutants of cellular genes directly in the human cell. The rationale is to couple the ability of Human Immunodeficiency Virus (HIV-1) derived vectors to foster genetic variability to that of delivering exogenous genes to human cells. HIV-1 reverse transcriptase, the enzyme responsible for the reverse transcription of the viral RNA into DNA, is an error prone enzyme and introduces 3×10-5 mutations in the viral genetic material at every infectious cycle. By inserting cellular genes in HIV-1 derived vectors and performing repeated cycles of infection, the viral replication machinery will drive their evolution, generating libraries of mutants directly in the human cell where they can be screened. This original system (that makes the object of two international patent applications) allows the generation of a library of variants of the gene one wants to make evolve, directly in the vector to use for gene delivery. Using this system, we are currently targeting some genes of interest for cancer research in relation to gene therapy approaches.