A bipartite boundary element restricts UBE3A imprinting to mature neurons

Abstract

Angelman syndrome (AS) is a severe neurodevelopmental disorder caused by the loss of function from the maternal allele ofUBE3A, a gene encoding an E3 ubiquitin ligase.UBE3Ais only expressed from the maternally inherited allele in mature human neurons due to tissue-specific genomic imprinting. Imprinted expression ofUBE3Ais restricted to neurons by expression ofUBE3A antisense transcript(UBE3A-ATS) from the paternally inherited allele, which silences the paternal allele ofUBE3Aincis. However, the mechanism restrictingUBE3A-ATSexpression andUBE3Aimprinting to neurons is not understood. We used CRISPR/Cas9-mediated genome editing to functionally define a bipartite boundary element critical for neuron-specific expression ofUBE3A-ATSin humans. Removal of this element led to up-regulation ofUBE3A-ATSwithout repressing paternalUBE3A. However, increasing expression ofUBE3A-ATSin the absence of the boundary element resulted in full repression of paternalUBE3A, demonstrating thatUBE3Aimprinting requires both the loss of function from the boundary element as well as the up-regulation ofUBE3A-ATS. These results suggest that manipulation of the competition betweenUBE3A-ATSandUBE3Amay provide a potential therapeutic approach for AS.

Human cerebral organoids capture the spatiotemporal complexity and disease dynamics of UBE3A

“…As expected, UBE3A transcripts decreased only in AS hCOs, after 6 weeks ( Figure 3B). In addition, the ~3 week delay between when UBE3A-ATS began to increase and UBE3A decreased is similar to previous observations (Hsiao et al, 2019). We also observed an interesting transient increase in UBE3A in both neurotypical and AS hCOs, suggesting that a transient elevation in UBE3A expression may occur in neurodevelopment prior to imprinting.…”

“…The silencing of paternal UBE3A is also associated with a long, non-coding antisense RNA (UBE3A-ATS). As cells differentiate into neurons, UBE3A-ATS expression increases to a high enough level to silence paternal UBE3A, potentially through a hypothesized collision mechanism (Hsiao et al, 2019;Stanurova et al, 2016) between these two opposed and overlapping transcripts.…”

“…However, at 4 weeks, we observed only a reduction in UBE3A-ATS and did not detect an increase in UBE3A expression ( Figure 4D). This could be attributed to a delay between transcription of UBE3A-ATS and silencing of UBE3A (Hsiao et al, 2019;Stanurova et al, 2016) (Figures 3A and 3B) so that UBE3A expression levels are still high at that stage of hCO development.…”

“…These provide important biological context and molecular benchmarks for hCOs. There are also significant anatomical differences between mouse and human brains including in UBE3A isoforms (LaSalle et al, 2015) and cisregulatory elements (Chamberlain et al, 2010;Hsiao et al, 2019) that could result in substantially different UBE3A biology. Thus, hCOs also present an opportunity to identify key species-specific differences, with both similarities and differences potentially providing key insights into disease mechanisms and treatment strategies.…”

Small molecule inhibitors of G9a reactivate the maternal PWS genes in Prader-Willi-Syndrome patient derived neural stem cells and differentiated neurons

“…Increased UBE3A-ATS expression is consistent with activation of SNORD116 and SNORD115 , since they are part of the same transcriptional unit. However, UBE3A-ATS may not be sufficiently upregulated to repress UBE3A (Hsiao et al, 2019; Langouet et al, 2018). Although it is not clear why UBE3A expression is increased following G9a inhibition, this result may alleviate one potential safety concern related to gene activation specificity at this locus.…”