Les surprises du décodage de l'information génétique.

Abstract

La conservation du code genetique et de son decryptage par la machinerie traductionnelle est une des universalites du monde vivant. Le genie genetique en fournit une illustration quotidienne en permettant de produire des proteines d' origine humaine chez une bacterie ou chez une plante. Le biologiste moleculaire sait traduire facilement une sequence nucleotidique codante en une sequence proteique et des programmes informatiques simples ont ete developpes pour automatiser cette tache. Cependant, pour lire le message genetique, la cellule ne depend pas d' un logiciel mais d' un mecanisme biologique complexe qui est parfois sujet a des deviations. Il ne s' agit pas ici des exceptions au code universel que l' on observe pour tous les genes de certains organismes comme les paramecies, mais de modifications qui affectent des genes specifiques et qui dependent de sequences precises de l' ARN messager. L' importance biologique de ces deviations ne doit pas etre negligee : ce sont elles qui sont responsables, par exemple, de la synthese de la transcriptase inverse des retrovirus. A travers les elements accumules ces dernieres annees, sur les mecanismes moleculaires de ces nouveaux modes de codage chez les eucaryotes, apparaissent les enjeux de ces travaux en recherche fondamentale et appliquee.

Recoding refers to various translational phenomena in which expression of viral and cellular functions uses programmed modifications of canonical decoding rules. It includes ribosome slipping, either forward or backward (frameshifting), incorporation of amino acids instead of terminating translation (readthrough), and even skipping long segments of a message (ribosome hopping). The role of recoding is crucial for the expression of some important biological functions, like the synthesis of protease, reverse transcriptase, RNase H and integrase activities encoded by the pol gene of retroviruses. In this case, pol translation is dependent on a frameshifting or a readthrough event, during upstream gag translation. Specific sequence contexts are involved in recoding: slippery sequences and stem-loop/pseudoknot RNA secondary structures for -1 frameshifting, special tRNAs for +1 frameshifting, poor termination context for readthrough. In many cases, a translational pause is needed to stimulate recoding efficiency. Recoding is widely conserved, especially in retrotransposons, suggesting that is appeared early during evolution. Since similar mechanisms are often involved for the same event in different species, recoding should lie on very general properties of the translational apparatus. A better understanding of recoding has two important goals. First, in the genomic field, it will help to identify recoded genes which would otherwise escape the routine procedures developped to identify putative protein coding regions. Second, recoding can potentially be used as a therapeutic target since it is essential for the life cycle of numerous pathogenic viruses, including HIV.