Loss-of-function mutations in one copy of a gene can reduce levels of messenger RNA and protein, creating haploinsufficiency and leading to human disease. Of the more than 300 genes associated with neurodevelopmental disorders (NDDs)^1,2,3, more than 200 are thought to cause autism spectrum disorder (ASD) through mechanisms involving haploinsufficiency. Using traditional gene therapy to deliver extra copies of the gene is a promising strategy to treat these conditions. However, recombinant adeno-associated virus (rAAV), the standard vector for transgene delivery, has a 4,700-base-pair cargo limit, necessitating alternative approaches for the many ASD-associated genes that exceed this size⁵.

In a new SFARI-supported study appearing in Nature, Kevin Bender and colleagues evaluated CRISPR-based cis-regulation therapy (CRT) as an alternative means to overcoming haploinsufficiency in NDDs⁶. The CRT platform modulates gene expression without directly editing the genome⁷. Instead, endogenous regulatory elements controlling transcription are targeted using nuclease-deficient Cas9 fused to a transcriptional activator — in this case, through CRISPR activation (CRISPRa) — to boost expression of the functional allele. This approach sidesteps the parcel-size limit that scientists often encounter when trying to deliver additional gene copies to cells.

As a proof of concept, the scientists focused on sodium voltage-gated channel alpha subunit 2 (SCN2A) haploinsufficiency. Loss-of-function variants in the 6-kilobase coding region of this gene in humans cause a relatively prevalent subset of NDDs associated with severe intellectual disability, developmental delay, ASD and treatment-resistant seizures⁸. In mice, Scn2a haploinsufficiency alters excitatory function in pyramidal cortical neurons, as shown by electrophysiology^9,10. Using rAAV-based CRISPRa to target the Scn2a promoter and enhance expression, the team rescued neural excitability deficits in a genetic model of SCN2A haploinsufficiency (Scn2a^+/–), in which gene expression levels are around 50% of typical levels. This was achieved about equally effectively through both stereotaxic injection into the prefrontal cortex and tail-vein injection, the latter demonstrating the feasibility of systemic administration. Moreover, although Scn2a^+/– mice are more susceptible to chemoconvulsant-induced seizures, the CRISPRa editing conferred wild-type levels of protection against these agents, suggesting that CRT could be a therapeutic avenue for managing refractory seizures in people with SCNA2 haploinsufficiency.

When upregulating gene expression, attaining physiological expression levels while avoiding excessive expression is critical. Because gain-of-function SCN2A variants are known to cause epileptic encephalopathies⁸, the researchers selected a transcriptional activator with modest upregulating potential for their CRISPRa editing. They also examined the effects of Scn2a overexpression in wild-type mice and found no increase in excitability. The authors propose that there is a “ceiling effect” on voltage-gated channel membrane density, implying an intrinsic regulatory mechanism that limits excessive channel activity.

Lastly, the group tested CRT in neurons differentiated from SCN2A-haploinsufficient human embryonic stem cells and again rescued the aberrant electrophysiological phenotype.

This work shows that CRT is an encouraging approach to treat this and other haploinsufficiency-driven conditions. Although earlier intervention might yield greater benefit, the successful restoration of normal neural function in adolescence indicates that CRT could still confer therapeutic value later in life. The platform has already shown efficacy in other rodent models of both CNS and non-CNS haploinsufficiency-related disorders^7,11–18.

1. Kaplanis J. et al. Nature 605, 503–508 (2022) PubMed
2. Fu J.M. et al. Nat. Genet. 54, 1320–1331 (2022) PubMed
3. Epi25 Collaborative. Nat. Neurosci. 27, 1864–1879 (2024) PubMed
4. Karczewski K.J. et al. Nature 581, 434–443 (2020) PubMed
5. Howe K.L. et al. Ensembl 2021. Nucleic Acids Res. 249, D884–D891 (2021) PubMed
6. Tamura S. et al. Nature 646, 983–991 (2025) PubMed
7. Matharu N. et al. Science 363, eaau0629 (2019) PubMed
8. Sanders S.J. et al. Trends Neurosci. 41, 442–456 (2018) PubMed
9. Spratt P.W.E. et al. Neuron 103, 673–685.e5 (2019) PubMed
10. Nelson A.D. et al. Neuron 112, 1133–1149.e6 (2024) PubMed
11. Colasante G. et al. Mol. Ther. 28, 235–253 (2020) PubMed
12. Colasante G. et al. Brain 143, 891–905 (2020) PubMed
13. Yamagata T. et al. Neurobiol. Dis. 141, 104954 (2020) PubMed
14. Chang H.C. et al. J. Biol. Chem. 299, 102728 (2023) PubMed
15. Wang G. et al. Nat. Immunol. 20, 1494–1505 (2019) PubMed
16. Liao H.K. et al. Cell 171, 1495–1507.e15 (2017) PubMed
17. Böhm S. et al. Sci. Adv. 6, eaba5614 (2020) PubMed
18. Kemaladewi D.U. et al. Nature 572, 125–130 (2019) PubMed