We present a method that deploys clickable probes to directly quantify protein-RNA interactions on proteins. Our method facilitated global detection of RNA-interaction sites on RBPs that mediate recognition of coding and noncoding RNA. We performed functional profiling of known RNA-binding interfaces and discovery of RNA binding activity on proteins without prior RBP annotation. Collectively, we present a chemoproteomic method for systematic quantification of protein-RNA binding activity in living cells.

Especially in mammalian systems, post-transcriptional gene expression regulatory processes at the RNA level are often a key determinant of genetic information flow. From an engineering and therapeutic perspective, these RNA regulatory processes represent new ways to control or retune gene expression at the RNA level, if they can be harnessed. I will present technologies our group has developed to measure and control protein-RNA interactions with an eye toward therapeutic development.

Targeting RNA with small molecules is an emerging field. However, screening strategies usually rely on HTS, FBDD or derivatization of a known ligand while structure-based design is still the exception. In our work we demonstrate the suitability of (modified) protein-based docking tools for RNA targets and highlight special caveats. In a subsequent prospective virtual screening against the preQ1-riboswitch, 6 novel ligands with affinities between 30nM and 11µM were identified.

Modulating RNA with small molecules could dramatically expand the universe of druggable targets. However, properties such as low diversity, a charged backbone, and dynamic structure make targeting RNAs challenging compared to protein targets. Moreover, standard biophysical and structural biology techniques have not directly translated from the protein world. Here, I present several Artificial Intelligence (AI) tools that we are utilizing to overcome some of these challenges in RNA drug discovery.

An estimated 85% of the ~3 billion base pairs in the human genome is transcribed into RNA, but only ~1.5% of these code for proteins. While the chemical properties of protein binders are increasingly understood and interrogated, the field of RNA-targeted drugs is relatively new and the properties of small molecules that drive specific and selective targeting of RNA, and associated assays, are yet to be developed. At Serna Bio (previously Ladder Therapeutics) we are using an AI enabled, data-first approach to write the rules that define RNA-small molecule interactions.

Atomic AI has developed PARSE, the Platform for AI-driven RNA Structure Exploration, which can locate 3D structures at unprecedented speed and accuracy in disease-relevant RNA targets. PARSE builds on our work featured on the cover of Science, and involves a tight integration of high-throughput wet-lab experiments and cutting-edge artificial intelligence capabilities. Through this data-driven approach, Atomic AI is enabling and pursuing drug discovery against undruggable targets.

Here I will report on the chemoproteomics-driven discovery of small-molecule modulators targeting the RNA-binding protein NONO with a functional consequence. By engaging Cys145 of NONO, these electrophilic compounds decrease transcripts encoding the androgen receptor and its V7 splice variant in human prostate cancer cells in a stereoselective fashion. Our findings demonstrate that NONO can be co-opted by covalent small molecules to rewire the transcriptional state of cancer cells.

This talk illustrates the adaptability of small molecule recognition by RNA using ykkC riboswitches that use a similar mRNA structure to recognize very different small molecules. I will discuss our current efforts of structure-based design of synthetic small molecules to bind to thiamine pyrophosphate riboswitch and serve as potential anti-bacterials.

Anima's mRNA Lightning platform combines high-scale phenotypic screening with AI-driven MOA elucidation to discover small molecule mRNA drugs and their mechanisms of action. With unparalleled automation of phenotypic screening tailored to mRNA biology, we explore the vast target space of mRNA regulation from transcription to translation. Our MOAi technology streamlines the identification of active molecules' mechanism of action by integrating AI-driven MOA elucidation. Our pipeline includes lead programs in immunology and oncology, demonstrating efficacy in animal and patient-derived models. With deep expertise in mRNA biology, Anima is advancing drug discovery and development at an unprecedented rate of success.

 Familial dysautonomia (FD) is a rare neurodevelopmental and neurodegenerative disease associated with autonomic and sensory neuropathy. Over 99% of FD patients are homozygous for the “founder" mutation in intron 20 of the ELP1 gene (c.2204 + 6T > C) causing the “skipping” of exon 20 from the mature mRNA coding sequence. Here, we describe the discovery and development of small molecules that modulate the splicing of the mutant ELP1 RNA.  We utilize mouse models of FD to correlate drug levels to corrected splicing, increased protein levels, and improved phenotype. We discuss how these data can be used to better design clinical trials testing splicing modulators.

Familial dysautonomia is a debilitating neurodegenerative disorder arising from aberrant mutation-linked alternative splicing of ELP1 pre-mRNA. Treatment is limited to palliative care, however, due to the absence of effective etiology-targeted therapies. Addressing this issue, PTC Therapeutics has developed potent small molecule splicing modulators based on hits originally identified by collaborative partners at Massachusetts General Hospital. Their discovery and putative pharmacological mechanism of action is described.