Dual-AAV Gene Therapy Corrects Dravet Syndrome in Mice

The Breakthrough

Scientists have made a groundbreaking advancement in the treatment of Dravet syndrome (DS), a rare and severe form of epilepsy, by developing a novel gene replacement therapy in mice. The study, published in Science Translational Medicine, was a collaborative effort between the Allen Institute and Seattle Children’s Research Institute. The therapy successfully alleviated symptoms in mouse models, promoting long-term recovery without causing toxicity, side effects, or increased mortality.

“While current epilepsy drugs can reduce seizures, they often broadly impact brain function, leading to significant side effects,” said Boaz Levi, Ph.D., associate investigator at the Allen Institute and co-lead of the study alongside Franck Kalume, Ph.D., an associate professor at the University of Washington and principal investigator at Seattle Children’s Research Institute.

DS is a debilitating form of epilepsy that typically emerges in the first year of life, severely impairing a child’s development and posing life-threatening risks, including sudden death during sleep. John Mich, the paper’s first author and a senior scientist at the Allen Institute, describes the condition as “severe, long-term, and debilitating,” noting that “up to 10–20% of affected children do not survive due to its severity.”

The research team aimed for precision by delivering the missing SCN1A gene directly to the affected neural circuits. According to Levi, this targeted approach could be significantly safer, more effective, and capable of minimizing side effects compared to existing treatments.

Dravet Syndrome (DS)

DS is a severe, early-onset developmental epileptic encephalopathy that profoundly impacts patients and their families. It is characterized by treatment-resistant seizures, cognitive and motor impairments, developmental delays, and a high risk of premature death due to sudden unexpected death in epilepsy. Affecting approximately 1 in 16,000 births, DS typically manifests in the first year of life and requires lifelong care.

The primary genetic cause of DS is a monoallelic loss-of-function mutation in the SCN1A gene, which encodes the NaV1.1 voltage-gated sodium channel. This mutation predominantly affects GABAergic inhibitory interneurons, leading to an imbalance between excitation and inhibition in the brain. Current treatments focus on symptomatic management using antiepileptic drugs (AED), but these therapies are often ineffective, and no disease-modifying treatments are currently approved. This highlights the urgent need for novel therapeutic strategies that address the root cause of DS.

Current Therapy Landscape

The current treatment options for DS include:

• Symptomatic Management: First-line AEDs like valproate and clobazam are used to reduce seizure frequency but often fail to provide complete control. Sodium channel blockers are contraindicated as they exacerbate symptoms.
• Emerging Therapies: Recently approved drugs such as fenfluramine and cannabidiol have shown promise in reducing seizure burden but do not address the underlying genetic defect.
• Zorevunersen: More recently, Biogen has partnered with Stoke Therapeutics to develop and commercialize zorevunersen (1), an investigational antisense oligonucleotide therapy for DS. Zorevunersen is designed to upregulate NaV1.1 protein expression by leveraging the non-mutant (wild-type) copy of the SCN1A gene, thereby reducing seizures and associated comorbidities. Zorevunersen has received orphan drug designation from the FDA and the EMA and is currently being evaluated in clinical trials.
• Unmet Needs: Despite advancements, there remains a critical need for disease-modifying therapies that restore NaV1.1 function specifically in inhibitory interneurons.

Research Efforts: Dual-AAV SCN1A Gene Replacement for DS

Mich et al.’s publication, “Interneuron-specific dual-AAV SCN1A gene replacement corrects epileptic phenotypes in mouse models of Dravet syndrome (2),”presents a novel gene therapy approach to restore SCN1A function specifically in GABAergic interneurons using dual AAV vectors. Key innovations include:

1. Split-Intein Technology and Dual AAV Delivery:

• • The SCN1Agene is too large to fit within the packaging limit of AAV vectors (~4.7 kb). To overcome this limitation, the researchers divided SCN1A into two halves and used intein-mediated ligation technology to reconstitute a full-length functional NaV1.1 protein after delivery.
  • They used the Cfa split-intein, which was engineered for rapid activity and chemical stability, placing it directly upstream of the cysteine at position 1050 (Cys1050) in the SCN1A sequence. This strategic placement was necessary because intein-mediated protein ligation requires a cysteine residue adjacent to the split site. The N and C termini of the Cfa split-intein were fused to the N and C termini of the halves of the split SCN1A gene, respectively. The intein-mediated ligation ensures scarless reconstitution of the protein (Figure 1).
  • Two AAV vectors (AAV2/PHP.eB) were used to deliver each half of SCN1A. This approach ensures efficient co-expression and reconstitution of NaV1.1. These constructs were packaged into viral vectors at PackGene (Figure 2).

Figure 1. Design of the split-intein halves of SCN1A, leading to reconstitution of functional NaV1.1.

2. Interneuron-Specific Targeting:

• • The class-specific DLX2.0 enhancer was employed to drive SCN1Aexpression specifically in telencephalic GABAergic interneurons. DLX2.0 is derived from concatenated core sequence elements of the hDLXI5/6i enhancer (Figure 2).
  • This targeting strategy was based on the understanding that DS is primarily associated with dysfunction in inhibitory interneurons, particularly telencephalic fast-spiking interneurons. By directing the therapeutic gene specifically to these critical cell populations, the researchers aimed to address the fundamental pathophysiology of DS while minimizing potential off-target effects and adverse outcomes associated with pan-neuronal expression.

Figure 2. Delivery of SCN1A using the optimized enhancer DLX2.0 with dual AAV2/PHP.eB vectors.

Interpretation of Key Figures/Data

1. Functional Reconstitution of NaV1.1

• Western blot analysis confirmed successful reconstitution of full-length NaV1.1 protein from split-intein constructs in HEK293 cells.
• Patch-clamp electrophysiology demonstrated that reconstituted NaV1.1 channels exhibited normal sodium current properties and voltage-dependent gating similar to wild-type channels.
• Quantitative analysis revealed that cells expressing both halves of SCN1A showed significantly higher peak current densities compared to cells expressing either half alone or control constructs, suggesting that functional NaV1.1 sodium channels with near-native properties could be reconstituted from the two SCN1A split-intein constructs (Figure 3).

Figure 3. Functional NaV1.1 sodium channels with near-native properties could be reconstituted from the two SCN1A split-intein constructs.

2. Specificity of In Vivo Interneuron Targeting

For in vivo expression, the dual vectors were delivered via bilateral intracerebroventricular (ICV) injection in neonatal mice at postnatal day 2. This early intervention was strategically timed to address the developmental nature of  DS, which typically manifests in the first year of life in humans.

• Mice injected with both vectors showed strong bands near the expected apparent size of intact NaV1.1 (250 kDa). Immunohistochemistry revealed that the expression was highly specific to GABAergic interneurons. Quantitative analysis indicated that 98–99% of transduced cells were GABAergic interneurons, confirming high specificity.
• The expression was observed in a substantial proportion of the telencephalic Gad67+ GABAergic neuron population in different brain regions, including those known to be involved in seizure generation such as hippocampus and cortex.

Figure 4. DLX2.0-SCN1A–N + C AAVs specifically transduce telencephalic GABAergic interneurons in neonatal mice.

3. Efficacy in Preventing Mortality and Seizure Protection

One of the most striking findings was the complete prevention of mortality in Drave syndrome mouse models treated with DLX2.0-SCN1A–N+C AAVs.

• Dual AAV delivery significantly reduced seizure frequency and improved survival rates in two DS mouse models (Scn1afl/+;Meox2-Creand Scn1a+/R613X). Some of these mice were monitored for long-term survival and exhibited 100% survival beyond 1 year.
• The DLX2.0-SCN1A–N+C AAVs provided substantial protection against both thermally induced and spontaneous seizures (Figure 5).
• Dose-dependent protection against postnatal mortality was observed without adverse effects such as gliosis or weight loss.

Figure 5. Therapeutic efficacy in preventing mortality and seizure protection in mouse models.

4. Comparison of Interneuron-Specific vs. Non-Selective Expression

Another key finding of the study was the comparison between interneuron-specific (DLX2.0- driven) and non-selective neuronal (hSyn1-driven) expression of SCN1A. SCN1A expression was driven in all neurons using the human hSyn1promoter.

• Mice treated with hSyn1-SCN1A–N+C AAVs exhibited increased mortality during the preweaning period. The extent of preweaning mortality was dose-dependent and significantly greater than that observed under control conditions.
• In surviving mice treated with the high dose of hSyn1-SCN1A–N+C AAVs, there was significant protection from thermally induced MC and GTC seizures (Figure 6). However, the low dose did not protect from thermally induced seizures. Importantly, the non-selective approach did not improve overall mortality in the DS mouse model, despite providing some seizure protection in surviving animals.

In summary, while the interneuron-specific approach provided complete protection against mortality and seizures, the non-selective approach had mixed results and significant adverse effects. These comparative results highlight the importance of targeting SCN1A expression to specific cell types, such as interneurons, to avoid potential toxicity and maximize therapeutic benefit.

Figure 6. Early preweaning toxicity and weak protection from epileptic symptoms with hSyn1-SCN1A–N + C AAVs.

Conclusion

The research presented in this article represents a significant advancement in the development of potential gene therapies for DS, a severe developmental epileptic encephalopathy with limited treatment options. The study demonstrates proof of concept that interneuron-specific AAV-mediated SCN1A gene replacement can effectively rescue DS phenotypes in mouse models, suggesting a promising therapeutic approach for patients with this devastating condition.

The key innovation of this work lies in the combination of three critical elements: the split-intein approach to overcome AAV packaging limitations, the dual-vector delivery system to enable full-length SCN1A expression, and the interneuron-specific targeting using the DLX2.0 enhancer. Together, these elements allowed for the precise delivery of functional NaV1.1 channels to the cell population most critically affected in DS—telencephalic GABAergic interneurons.

The study’s findings have profound implications for the treatment of DS. The complete prevention of mortality and substantial reduction in seizure activity across multiple mouse models suggests that this approach could potentially address the most severe aspects of the condition, including the high risk of sudden unexpected death in epilepsy (SUDEP). The long-term survival (up to 1 year) observed in treated mice further supports the durability of the therapeutic effect, a crucial consideration for a chronic condition like DS.

Perhaps most importantly, the comparison between interneuron-specific and nonselective neuronal expression of SCN1A highlights the critical importance of cell type specific targeting in gene therapy for neurological disorders. The adverse effects observed with non-selective expression, particularly increased preweaning mortality, underscore the potential risks of imprecise gene delivery approaches. This finding aligns with the current understanding of DS pathophysiology, which suggests that the condition results from an imbalance in excitatory and inhibitory neuronal activity.

In conclusion, this study provides compelling evidence that interneuron-specific dual-AAV SCN1A gene replacement can correct epileptic phenotypes in mouse models of DS. By targeting the fundamental genetic deficit in the specific cell population driving the disease, this approach represents a potentially transformative therapeutic strategy for a condition with significant unmet medical needs. When considered alongside other emerging approaches like Biogen and Stoke Therapeutics’ zorevunersen, these developments reflect the growing momentum toward disease- modifying treatments that address the underlying genetic basis of DS.

References

1. https://www.biopharmadive.com/news/biogen-stoke-deal-license-dravet-syndrome-rare-neuroscience/740197/
2. https://www.science.org/doi/10.1126/scitranslmed.adn560