G-quadruplexes are four-stranded nucleic acid structures thought to play widespread biological roles [1]. The growing list of cellular processes thought to be regulated by DNA or RNA G-quadruplexes includes transcription, RNA processing, translation, and mRNA localization. More than 30 proteins have been identified that interact with G-quadruplexes in various ways, and handful of cellular cofactors that bind G-quadruplexes have also been identified. This diversity of biochemical function raises an important question: how does the cellular machinery distinguish the many G-quadruplexes in the genome? We are exploring the hypothesis that this specificity can be achieved by mutations in the primary sequence of the G-quadruplex itself. To test this idea, we generated a 496-member G-quadruplex library, and tested each member for five different biochemical activities associated with
G-quadruplexes: the ability to bind GTP, to promote peroxidase reactions, to form dimers, to form tetramers, and to generate fluorescence [2-4]. This revealed that mutations in both tetrads and loops can significantly alter the specificity of a G-quadruplex to favor a particular biochemical activity. In some cases, changes in specificity are correlated with changes in the multimeric state of the G-quadruplex. We also identified a small-molecule ligand that inhibits multimerization, raising the possibility that G-quadruplex specificity can be modulated by small molecules. We are currently using a combination of NMR and X-ray crystallography to better understand these mutations from a structural perspective, and preliminary results in this area will be discussed. Taken together, these experiments provide new information about the mechanisms of G-quadruplex specificity, and should facilitate analysis of the biochemical roles of these structures in cells.