UBE3A reinstatement restores behaviorand proteome in an Angelman syndrome mouse model of imprinting defects

Angelman Syndrome (AS) is a severe neurodevelopmental disorder with only symptomatic treatment currently available. The primary cause of AS is loss of functional UBE3A protein. This can be caused by deletions in the maternal 15q11-q13 region, maternal AS-imprinting center defects (mICD), paternal uniparental disomy of chromosome 15 (UPD) or mutations within the UBE3A gene. Current mouse models are Ube3a-centric and do not address expression changes of other genes in the 15q11-q13 locus on the pathophysiology of AS. This limits the ability to discern differences in therapeutic responses to current UBE3A-targeting strategies and hampers the identification of novel therapeutics/co-therapeutics.

Using a mouse line that harbors a maternally inherited mutation affecting the AS-PWS imprinting center ('mICD mice'), we studied the impact of the mICD or UPD AS subtype on behavior, seizure susceptibility and proteome. Additionally, by using mice overexpressing two copies of Ube3a or antisense oligonucleotide (ASO) targeting Ube3a-ATS, we analyzed the impact of bi-allelic Ube3a activation on behavior and proteome.

mICD mice showed 80% reduction in UBE3A protein, bi-allelic expression of Ube3a-ATS and Mkrn3-Snord115 gene cluster, leading to robust AS behavioral deficits and proteome alterations similar to Ube3a^m-/p+ mice. Genetic UBE3A overexpression in mICD mice, mimicking therapeutic strategies that effectively activate the biallelic silenced Ube3a gene, resulted in a complete rescue of all behavioral phenotypes, seizure susceptibility and proteome alterations. Subsequently, treatment with an antisense oligonucleotide (ASO) to directly activate the biallelic silenced Ube3a gene in mICD mice also resulted in efficient reinstatement of UBE3A, 30% higher relative to WT, alongside a partial rescue of behavioral phenotypes.

Despite using a highly robust AS-specific behavioral battery, we did not investigate readouts such as neuronal activity and sleep, for which impairments in Ube3a^m-/p+ mice were described.

Taken together, these findings demonstrate that the loss of UBE3A protein is the primary factor underlying AS phenotypes in this mICD/UPD mouse model of AS, while the biallelic expressed genes in this locus play either a marginal or yet unidentified role. These findings also corroborate UBE3A reinstatement as an attractive therapeutic strategy for AS individuals carrying an mICD or UPD mutation.