The role of flexibility in the rational design of modularly assembled ligands targeting the RNAs that cause the myotonic dystrophies.

Abstract

Modularly assembled ligands were designed to target the RNAs that cause two currently untreatable neuromuscular disorders, myotonic dystrophy types 1 (DM1) and 2 (DM2). DM1 is caused by an expanded repeating sequence of CUG, and DM2 is caused by expanded CCUG repeats. Both are present in noncoding regions and fold into hairpins with either repeating 1x1 nucleotide UU (DM1) or 2x2 nucleotide 5'-CU/3'-UC (DM2) internal loops separated by two GC pairs. The repeats are toxic because they sequester the RNA splicing regulator muscleblind-like 1 protein (MBNL1). Rational design of ligands targeting these RNAs was enabled by a database of RNA motif-ligand partners compiled by using two-dimensional combinatorial screening (2DCS). One 2DCS study found that the 6''-azido-kanamycin A module binds internal loops similar to those found in DM1 and DM2. In order to further enhance affinity and specificity, the ligand was assembled on a peptoid backbone to precisely control valency and the distance between ligand modules. Designed compounds are more potent and specific binders to the toxic RNAs than MBNL1 and inhibit the formation of the RNA-protein complexes with nanomolar IC(50) values. This study shows that three important factors govern potent inhibition: 1) the surface area sequestered by the assembled ligands; 2) the spacing between ligand modules since a longer distance is required to target DM2 RNAs than DM1 RNAs; and 3) flexibility in the modular assembly scaffold used to display the RNA-binding module. These results have impacts on the general design of assembled ligands targeting RNAs present in genomic sequence.