Abstract
Jones DJ, Soundararajan D, Taylor NK, Aimiuwu OV, Mathkar P, Shore A, Teoh JJ, Wang W, Sands TT, Weston MC, Harper SQ, Frankel WN. Mol Ther. 2024 Aug 10:S1525-0016(24)00531-8. doi: 10.1016/j.ymthe.2024.08.009. Epub ahead of print. PMID: 39127888. Effective gene therapy for gain-of-function or dominant-negative disease mutations may require eliminating expression of the mutant copy together with wild-type replacement. We evaluated such a knockdown-replace strategy in a mouse model of DNM1 disease, a debilitating and intractable neurodevelopmental epilepsy. To challenge the approach robustly, we expressed a patient-based variant in GABAergic neurons—which resulted in growth delay and lethal seizures evident by postnatal week three—and delivered to newborn pups an AAV9-based vector encoding a ubiquitously expressed, Dnm1-specific interfering RNA (RNAi) bivalently in tail-to-tail configuration with a neuron-specific, RNAi-resistant, codon-optimized Dnm1 cDNA. Pups receiving RNAi or cDNA alone fared no better than untreated pups, whereas the vast majority of mutants receiving modest doses survived with almost full growth recovery. Synaptic recordings of cortical neurons derived from treated pups revealed that significant alterations in transmission from inhibitory to excitatory neurons were rectified by bivalent vector application. To examine the mutant transcriptome and impact of treatment, we used RNA sequencing and functional annotation clustering. Mutants displayed abnormal expression of more than 1000 genes in highly significant and relevant functional clusters, clusters that were abrogated by treatment. Together these results suggest knockdown-replace as a potentially effective strategy for treating DNM1 and related genetic neurodevelopmental disease.
Commentary
Over the past decade, hundreds of genes have been identified as monogenic causes of epilepsy. 1 These genetic epilepsies have predominantly been associated with developmental and epileptic encephalopathies (DEEs), a heterogeneous group of severe, early-onset epilepsy disorders that are largely treatment-resistant. Gene therapies that target the underlying genetic cause of a given DEE likely represent the ideal treatment. Potential genetic mechanisms underlying DEE include loss-of-function (LOF), gain-of-function (GOF), or dominant-negative (DN). Each of these scenarios requires a different gene therapy strategy. In the case of LOF, boosting gene expression of an intact allele of the gene, if available, could provide enough wild-type protein to overcome haploinsufficiency. Gene replacement therapy, whereby a copy of the gene is delivered using a viral vector, would also be a viable option, provided the genetic material did not exceed the packaging capacity of the virus. GOF or DN mechanisms require an alternative strategy that would knock down expression of the allele containing the pathogenic variant. However, many pathogenic variants are single-nucleotide changes that are extremely challenging to selectively target. In this case, it can be effective to target the gene for knockdown while also providing additional wild-type copies of the gene. This strategy is termed knockdown-replace.
In the current study, Jones et al 2 demonstrate that knockdown-replace is a viable strategy for treatment in a mouse model of DN DEE caused by variants in DNM1. DNM1 encodes dynamin-1, a neuronal GTPase that catalyzes endocytosis and synaptic vesicle recycling at the presynaptic terminal. Individuals with DNM1 DEE have treatment-resistant seizures, cognitive impairment, developmental delay, and hypotonia. 3 The investigators developed a conditional mouse model of DNM1 DEE based on the patient variant p.G359A such that the G359A mutant allele would be expressed upon exposure to Cre recombinase. Pan-neuronal expression of G359A resulted in seizures, growth deficits, and some lethality in adulthood. To increase the penetrance of the phenotype for efficient testing of knockdown-replace gene therapy, they restricted expression of G359A to GABAergic neurons based on prior studies indicating that GABAergic neurons play a major role in DNM1 DEE. 4 G359A expression in GABAergic neurons resulted in a growth deficit and 100% lethal seizures between 2 and 3 weeks of age.
The gene therapy consisted of a bivalent adeno-associated virus 9 (AAV9) vector that expressed both a miRNA against Dnm1 and a Dnm1 cDNA. The miRNA would target Dnm1 transcripts for interfering RNA (RNAi)-induced degradation, while the RNAi-resistant Dnm1 cDNA would encode wild-type DNM1 protein. Neonatal G359A/+ pups received intracerebroventricular injections of AAV. The AAV treatment prevented severe seizures, prolonged survival, and improved the growth rate of G359A pups. Importantly, the degree of efficacy depended on the dose of the AAV, with lower doses performing better than the maximal deliverable dose. The bivalent AAV rescued G359A/+ mice much better than either the miRNA or the cDNA treatment alone, indicating that knockdown-replace worked better than either strategy on their own.
The team assessed synaptic transmission in treated and untreated G359A/+ mice using electrophysiological recordings from cultured neuron pairs. Electrophysiological changes were detected in G359A/+ mice, including decreased evoked inhibitory responses onto excitatory neurons and impaired presynaptic GABA release. All of the electrophysiological changes were reversed by bivalent AAV **knockdown-replace treatment.
This study demonstrates proof-of-principle that knockdown-replace may be a viable treatment option for genetic GOF and DN epilepsies. The knockdown-replace strategy does not rely on targeting only a specific pathogenic allele, which would require generation of a new AAV construct for each individual patient mutation. Thus, knockdown-replace has broad applicability to any pathogenic allele and could be used in a wide variety of patients with a given genetic epilepsy. An alternative means to knock down GOF variants in genetic epilepsies involves the use of antisense oligonucleotides (ASOs). ASOs have recently been used to reduce seizures and prolong survival in several mouse models of sodium and potassium channel-related epilepsies.5–7 However, the ASOs eventually degrade and lose efficacy. The miRNA utilized in the bivalent AAV in this study remains active indefinitely, potentially providing a longer-lasting rescue. The ASO could also potentially knock gene expression down to levels that are problematic, such as the case with SCN8A in which both GOF and LOF are associated with neurological disorders. To provide replacement gene expression would require additional genetic engineering. The bivalent AAV approach is more efficient in this regard, as it provides the miRNA knockdown and the gene replacement in 1 vector. In this study, higher doses of the AAV were somewhat toxic, leading to lethality in the heterozygous littermates. It will be important to understand why this lethality occurred. Because the DEE model had the mutation selectively expressed in interneurons, the investigators could consider using a promoter to drive AAV expression only in interneurons to reduce widespread toxicity. Additional studies will be required to determine the efficacy and limitations of the knockdown-replace approach in models of DEE that do not have the limitations of the Dnm1 DEE model (eg, the lack of appropriate phenotype when expressed pan-cellularly or pan-neuronally).
The bivalent AAV approach may have advantages over another leading gene therapy strategy that utilizes the CRISPR/Cas system. AAV vectors can safely deliver components of CRISPR; however, this does not limit the risk of off-target effects and DNA-damage toxicity caused by CRISPR-induced double-strand DNA breaks (DSBs). The use of catalytically inactive Cas9 to manipulate expression of target genes mitigates the risk of DSBs and has been shown to be effective in preventing seizures in mouse models of epilepsy.8,9 However, the knockdown-replace strategy in this scenario would require multiple AAV vectors to be used to provide the dCas9 and the replacement gene. Finally, very large genes will exceed the packaging capacity of AAV vectors, again requiring additional genetic engineering, such as the split intein approach recently utilized to deliver dystrophin in a mouse model of Duchenne muscular dystrophy.10,11 Overall, there are many strategies being developed to improve the treatment of genetic neurological disease. It will be interesting to see which of these gene therapy approaches will result in the most efficient and effective treatments for LOF, GOF, and DN genetic epilepsies.
