Targeting RNA G-Quadruplex Structures to Restore TDP-43 Cryptic Exon Repression in ALS/FTD

Aberrant RNA processing driven by nuclear depletion and cytoplasmic aggregation of the RNA-binding protein TDP-43 is a unifying feature of amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD). A critical consequence of TDP-43 loss is the pathological inclusion of normally silent “cryptic” exons into mature transcripts, leading to premature stop codons, transcript degradation, or production of toxic protein fragments. Although hundreds of cryptic exons have been catalogued, the structural cues that recruit TDP-43 and suppress these aberrant splice events remain poorly defined. Our preliminary data and recent studies identify RNA G- quadruplexes (GQs), which are stable, four-stranded guanine-rich motifs, proximal to or overlapping cryptic exons as essential regulators of TDP-43–dependent splicing repression. We hypothesize that GQs adjacent to cryptic splice sites serve as structural scaffolds for TDP-43, and that disruption of these motifs, whether by TDP- 43 depletion or ALS/FTD-linked sequence variants, undermines splicing fidelity. Moreover, pharmacological stabilization of GQs may compensate for TDP-43 loss and restore normal exon exclusion. In Aim 1, we will combine computational prediction, transcriptome-wide rG4-seq, and biophysical assays (circular dichroism, FRET, fluorescence anisotropy) to map and validate GQ structures at TDP-43-regulated cryptic exons in patient- derived neurons. We will then prioritize the most disease-relevant GQ targets based on in vitro stability and altered splicing in ALS/FTD tissue. In Aim 2, we will deploy an iterative, deep-learning–guided discovery platform, which integrates a junction tree variational autoencoder with random forest classifiers, to identify and optimize small molecules that selectively bind and stabilize GQs. Lead compounds will be screened using high- throughput FRET and cryptic-exon reporter assays, and the top candidates will be validated for splicing rescue, neuroprotection, and safety in ALS patient iPSC-derived neuronal models. By defining GQ-mediated mechanisms of cryptic exon control and delivering novel GQ-targeting ligands, this project establishes an innovative, structure-based therapeutic strategy to correct RNA misprocessing in ALS and FTD. Project Number: 1R21NS149721-01 | Fiscal Year: 2026 | NIH Institute/Center: National Institute of Neurological Disorders and Stroke (NINDS) | Principal Investigator: Aaron Haeusler (+1 co-PI) | Institution: THOMAS JEFFERSON UNIVERSITY, PHILADELPHIA, PA | Award Amount: $195,000 | Activity Code: R21 | Study Section: Special Emphasis Panel[ZRG1 AN-G (55)] View on NIH RePORTER: https://reporter.nih.gov/project-details/11391556