A chemoinformatics-guided platform for efficient discovery of RNA-binding small molecules: Proof-of-concept for...

Area of Science:

• Medicinal Chemistry
• RNA Biology
• Computational Chemistry

Background:

• Structured RNAs are implicated in human diseases but are difficult to target with small molecules.
• Developing selective small molecule ligands for disease-associated RNAs is a significant challenge in drug discovery.

Purpose of the Study:

• To establish a chemoinformatics-guided discovery framework for identifying small molecules that bind selectively to structured, disease-associated RNAs.
• To discover and characterize small molecule ligands for the pathogenic r(CUG)exp repeat RNA in myotonic dystrophy type 1 (DM1).

Main Methods:

• Utilized a fingerprint-based molecular design and similarity screening of over 8 million compounds.
• Employed experimental validation including target engagement, cell-based assays, biophysical, and single-molecule analyses.
• Integrated mechanistic profiling to understand small molecule-RNA interactions and their functional consequences.

Main Results:

• Identified a 150-member library enriched in RNA-active scaffolds.
• Discovered selective small molecule ligands for the r(CUG)exp repeat RNA.
• Demonstrated that identified molecules bind to internal loops in r(CUG)exp, partially inhibit muscleblind-like 1 (MBNL1) binding, and modulate RNA folding.
• Optimized scaffolds rescued MBNL1-dependent splicing defects in patient-derived cells with micromolar potency and low cytotoxicity.

Conclusions:

• Established a generalizable, data-driven platform for discovering drug-like small molecules that bind to RNA.
• Successfully applied the platform to identify lead compounds for the toxic repeat expansion RNA in DM1.
• The findings provide a promising strategy for developing novel therapeutics for RNA-mediated diseases.