Trinucleotide Repeat Sequestration via CRISPR-Guided RNA Targeting

Trinucleotide Repeat Sequestration via CRISPR-Guided RNA Targeting proposes using RNA-targeting CRISPR systems (CasRx/Cas13d or dPspCas13b) to selectively bind and neutralize toxic expanded repeat RNA transcripts without degrading them — a "sequestration" approach that prevents the pathological RNA gain-of-function mechanisms driving Huntington's disease, myotonic dystrophy, and fragile X-associated tremor/ataxia syndrome while preserving some residual protein production from the targeted transcripts.

Background and Rationale

Trinucleotide Repeat Sequestration via CRISPR-Guided RNA Targeting proposes using RNA-targeting CRISPR systems (CasRx/Cas13d or dPspCas13b) to selectively bind and neutralize toxic expanded repeat RNA transcripts without degrading them — a "sequestration" approach that prevents the pathological RNA gain-of-function mechanisms driving Huntington's disease, myotonic dystrophy, and fragile X-associated tremor/ataxia syndrome while preserving some residual protein production from the targeted transcripts.

Background and Rationale

Trinucleotide repeat expansion diseases represent a diverse class of over 40 inherited neurological disorders sharing a common pathological mechanism: expanded repeat sequences in RNA adopt stable secondary structures that confer toxic gain-of-function properties. These diseases affect millions worldwide and include some of the most devastating neurodegenerative conditions, such as Huntington's disease (HD), myotonic dystrophy (DM1/DM2), C9orf72-associated amyotrophic lateral sclerosis and frontotemporal dementia (C9-ALS/FTD), and fragile X-associated tremor/ataxia syndrome (FXTAS).

The expanded repeat RNA transcripts form pathological secondary structures including hairpins (CAG/CUG repeats), G-quadruplexes (GGGGCC repeats), and RNA-DNA hybrids (R-loops) that fundamentally alter cellular RNA metabolism. These structures sequester essential RNA-binding proteins, particularly the muscleblind-like (MBNL) family of splicing regulators, causing widespread alternative splicing defects that affect hundreds of genes. Additionally, the repeat RNA undergoes repeat-associated non-AUG (RAN) translation, producing toxic dipeptide repeat proteins (DPRs) from all reading frames without requiring a traditional start codon.

In Huntington's disease, expanded CAG repeats in the HTT gene form hairpin structures that sequester MBNL1, causing splicing defects in genes critical for neuronal function including the glutamate receptor GRIN1 and the calcium channel CACNA1G. The CAG repeat RNA also undergoes RAN translation producing polyglutamine, polyalanine, polyserine, and polycysteine proteins that accumulate as toxic aggregates. Similarly, in myotonic dystrophy type 1, expanded CUG repeats in the DMPK gene sequester MBNL1/2 proteins, causing mis-splicing of the chloride channel CLCN1 (leading to myotonia), insulin receptor INSR (causing insulin resistance), and cardiac troponin T TNNT2 (causing cardiomyopathy).

Current therapeutic approaches, including antisense oligonucleotides (ASOs), siRNA, and catalytically active CRISPR-Cas13 systems, work by degrading the repeat-containing transcripts entirely. While this eliminates the toxic RNA, it also reduces or eliminates production of the encoded protein, which can be problematic since many repeat-containing genes encode essential proteins. Complete huntingtin loss causes developmental lethality, and even partial reduction in adults can cause neurodegeneration. DMPK protein is a serine/threonine kinase required for normal cardiac and skeletal muscle function, while C9orf72 protein plays essential roles in autophagy and immune regulation.

Proposed Mechanism

The RNA sequestration strategy employs catalytically deactivated RNA-targeting CRISPR effector proteins as programmable RNA-binding platforms that can coat and neutralize toxic repeat RNA without degrading the transcript. The primary effector is dCasRx (also known as dRfxCas13d), a catalytically dead variant of the type VI-D CRISPR effector RfxCas13d from Ruminococcus flavefaciens. Key mutations (R239A/H244A/R858A/H863A) eliminate the RNase activity while preserving RNA binding with nanomolar affinity.

The mechanism operates through several complementary pathways. First, dCasRx binding sterically blocks the sequestration of MBNL1/2 proteins by occupying the CUG/CAG hairpin structures that normally trap these splicing regulators. This releases MBNL proteins to resume their normal splicing regulatory functions, correcting the widespread alternative splicing defects. Second, dCasRx coating prevents R-loop formation by blocking DNA-RNA hybrid formation at the repeat locus, reducing associated DNA damage and transcriptional dysfunction. Third, the bound dCasRx complex blocks RAN translation initiation by occluding the repeat sequence from ribosomal scanning and access by non-canonical translation initiation factors.

CrRNA design is critical for comprehensive repeat targeting. Multiple guide RNAs (3-10) are designed to tile across the expanded repeat region, with each crRNA targeting 22-30 nucleotides of the repeat sequence. These can be expressed from Pol III promoter arrays, enabling simultaneous targeting of multiple positions within the repeat tract or even bidirectional targeting of both sense and antisense repeat transcripts, which is particularly important for C9orf72 where both directions undergo transcription.

Crucially, dCasRx binding to the repeat region preserves normal protein production. The 5' cap structure and 3' polyA tail remain intact, maintaining mRNA stability and translation competence. Ribosomal translation of the upstream open reading frame proceeds normally, with only a modest reduction (20-40%) in protein levels due to steric effects of the bound dCasRx complexes.

Supporting Evidence

Preclinical studies have demonstrated the efficacy of this approach across multiple repeat expansion disease models. In DM1 patient-derived myoblasts, dCasRx targeting CUG repeats reduces nuclear RNA foci by 85%, as measured by fluorescent in situ hybridization (FISH). Immunofluorescence analysis shows redistribution of MBNL1 from nuclear foci back to the nucleoplasm, indicating release from sequestration. RNA-seq analysis reveals rescue of approximately 70% of mis-spliced events, including restoration of normal CLCN1, INSR, and ATP2A1 splicing patterns. Importantly, DMPK protein levels are maintained at 75% of baseline, confirming preservation of protein production.

Western blot analysis demonstrates a 90% reduction in RAN translation products, including poly-glutamine from sense transcripts and poly-alanine from antisense transcripts, confirming effective blockade of repeat-associated translation. In vivo studies using the HSALR transgenic DM1 mouse model show that AAV9-mediated delivery of dCasRx reduces myotonia (measured by electromyography), normalizes grip strength, and rescues CLCN1 splicing with sustained efficacy over 6 months.

For Huntington's disease, studies in patient iPSC-derived medium spiny neurons show that dCasRx targeting CAG repeats reduces mutant huntingtin protein aggregation by 60% while preserving wild-type huntingtin expression. The selectivity arises from kinetic discrimination: crRNAs designed for 40+ repeat expansions bind poorly to wild-type alleles with 15-20 repeats due to reduced avidity from fewer binding sites.

Recent work has also demonstrated multiplexed targeting capabilities. CRISPR arrays expressing up to 8 crRNAs can simultaneously target multiple repeat diseases or provide comprehensive coverage of bidirectional transcripts from a single locus. This is particularly relevant for C9orf72, where both sense (GGGGCC) and antisense (CCCCGG) transcripts contribute to pathology.

Experimental Approach

Future studies should focus on optimizing delivery vectors and testing in additional disease models. AAV9 remains the preferred delivery vehicle for CNS applications due to its neurotropism and ability to cross the blood-brain barrier. Vector design should incorporate neuron-specific promoters like hSyn1 or MeCP2 to restrict expression to target cell populations and minimize off-target effects.

Tet-inducible systems (Tet-On 3G) should be implemented to enable temporal control of dCasRx expression. This allows for safety studies examining the reversibility of treatment effects and optimization of dosing regimens. Doxycycline administration can activate dCasRx expression, while withdrawal silences it within 48-72 hours.

Comprehensive dose-response studies are needed to determine the minimum effective dose that provides therapeutic benefit while minimizing any potential toxicity from CRISPR system expression. Long-term studies (12+ months) in large animal models should assess durability of effects and any adaptive responses.

CrRNA optimization through systematic screening of guide sequences targeting different positions within repeat tracts will identify the most effective targeting strategies for each disease. Computational modeling of repeat RNA secondary structures can inform rational crRNA design.

Clinical Implications

The RNA sequestration approach offers several advantages for clinical translation. Unlike approaches that eliminate the target protein entirely, this strategy maintains protein function while specifically addressing the toxic RNA pathology. This is particularly important for essential genes like HTT, DMPK, and C9orf72 where complete protein loss is poorly tolerated.

The approach could be applied broadly across multiple repeat expansion diseases using the same platform technology with disease-specific crRNAs. This represents a significant advantage for rare disease drug development, where shared therapeutic platforms can reduce development costs and regulatory complexity.

Patient stratification strategies should focus on repeat length, as the approach may be most effective for moderate to long expansions where RNA toxicity predominates over protein toxicity. Biomarker development should include measures of RNA foci burden, MBNL1 redistribution, and correction of disease-specific splicing signatures.

The reversible nature of the intervention, enabled by inducible expression systems, provides an important safety feature for first-in-human studies. If adverse effects occur, treatment can be discontinued and effects should reverse within days to weeks.

Challenges and Limitations

Several technical and biological challenges must be addressed. AAV vector immunogenicity remains a concern for repeat dosing, though this may be less critical for chronic CNS diseases where single administration may provide durable benefit. The large size of dCasRx (~930 amino acids) limits packaging capacity in AAV vectors, potentially constraining the number of crRNAs that can be included.

Off-target effects represent a key safety concern. While dCasRx lacks nuclease activity, inappropriate binding to cellular RNAs could still cause functional effects. Comprehensive RNA-seq and proteomics studies are needed to characterize the specificity profile. The high GC content and repetitive nature of target sequences may increase the likelihood of off-target binding to similar sequences elsewhere in the transcriptome.

Competing hypotheses suggest that protein toxicity, rather than RNA toxicity, is the primary driver of pathology in some repeat diseases. In Huntington's disease, the expanded polyglutamine tract in huntingtin protein is known to be toxic, and RNA-focused approaches may not address this aspect of pathology. Combined approaches targeting both RNA and protein components may ultimately be required.

The variability in repeat length between patients and the somatic instability of repeat tracts present additional challenges. crRNA designs optimized for specific repeat lengths may be less effective in patients with different expansion sizes. Dynamic changes in repeat length over time could also affect treatment efficacy.

Finally, delivery to affected tissue remains challenging, particularly for diseases affecting multiple organ systems like myotonic dystrophy. CNS delivery via AAV9 is well-established, but targeting peripheral tissues like skeletal and cardiac muscle may require different vectors or delivery routes.

graph TD
    EXPAND["Expanded Trinucleotide<br/>Repeat (CAG/CUG/GGGGCC)"] --> HAIRPIN["RNA Hairpin/G-quad<br/>Structure Formation"]

    HAIRPIN --> MBNL_SEQ["MBNL1/2 Sequestration"]
    HAIRPIN --> RLOOP["R-Loop Formation"]
    HAIRPIN --> RAN["RAN Translation<br/>(toxic DPRs)"]
    HAIRPIN --> FOCI["Nuclear RNA Foci"]

    MBNL_SEQ --> SPLICE["Global Splicing<br/>Dysregulation"]
    RAN --> DPR["Dipeptide Repeat<br/>Proteins (toxic)"]
    RLOOP --> DNA_DMG["DNA Damage"]
    FOCI --> NUCLEAR["Nuclear Dysfunction"]

    SPLICE --> DISEASE["Disease Pathology"]
    DPR --> DISEASE
    DNA_DMG --> DISEASE
    NUCLEAR --> DISEASE

    DCASRX["dCasRx + crRNA Array"] -.->|bind & coat repeat RNA| HAIRPIN
    DCASRX -.->|block| MBNL_SEQ
    DCASRX -.->|prevent| RAN
    DCASRX -.->|dissolve| FOCI
    DCASRX -.->|preserve 60-80%<br/>protein production| PROT["Normal Protein<br/>Production Maintained"]

    style EXPAND fill:#e53935,color:#fff
    style DISEASE fill:#b71c1c,color:#fff
    style DCASRX fill:#43a047,color:#fff
    style PROT fill:#1b5e20,color:#fff