Authors :
Presenting Author: Shreeya Bakshi, PhD – University of Michigan Medical School
Chunling Chen, MD – University of Michigan
Chloe Moore, BS – University of Michigan Medical School
Isha Verma, PhD – University of Michigan Medical School
Joe Minton, MS – Northwestern University
Christelle Moufawad El Achkar, MD – Boston Children's Hospital
Annapurna Poduri, MD, MPH – Boston Children's Hospital
Jack Parent, MD – University of Michigan Medical School
Lori Isom, PhD – University of Michigan.
Rationale:
Inherited, biallelic variants in SCN1B, encoding the voltage-gated sodium channel β1 subunit , are linked to Dravet syndrome (DS) or the more severe Early Infantile Developmental and Epileptic Encephalopathy (EIDEE). SCN1B generates two splice isoforms including transmembrane b1 (exons 1-6) and SCN1B-IR transcript that results from the retention of intron 3 and contains an in-frame stop codon. β1 subunits are multi-functional proteins that participate in current modulation, regulation of channel cell-surface expression, cell adhesion, and transcriptional regulation. The function of SCN1B-IR is unknown, although several pathogenic variants are identified within SCN1B intron 3, suggesting that it is functionally important. Here, we investigate whether SCN1B-IR is a target of nonsense mediated decay (NMD) or functions as a nuclear detained intron that undergoes cue-dependent splicing. Understanding the function and mechanism of SCN1B mRNA splicing is critical to understanding the pathogenic mechanisms of disease variants identified in intron 3. Here, we investigate a pathogenic splice site variant in intron 3, SCN1B c.449-2A >G, linked to EIDEE in three pedigrees. Methods:
We studied Scn1b transcripts in mouse brain, mouse cerebellar granule neurons, SCN1B c.449-2A >G patient-derived iPSC neurons and nonepileptic controls, as well as three human cancer cell lines with high levels of SCN1B intron 3 retention: K562, MCF7, and SH-SY5Y. We conducted RT-PCRs using specific SCN1B primers. Finally, we developed a transgenic mouse model that expresses Scn1b-IR but lacks transmembrane β1.Results:
Using an NMD inhibitor, we show that SCN1B-IR undergoes NMD in some of the models examined, and not others. We show that SCN1B-IR transcript includes exons 1-6 and therefore could serve as a precursor transcript to generate mature SCN1B mRNA following the final splicing step. Most of SCN1B-IR is confined to the nuclear fraction of K562, MCF7, and SH-SY5Y cells, with a very low rate of decay, characteristic of nuclear detained intron transcripts. In SCN1B c.449-2A >G patient-derived iPSC neurons, three transcripts are expressed including SCN1B-IR transcript and two different mis-spliced transcripts that lack the sequence encoding the β1 transmembrane segment. Importantly, patient iPSC neurons do not express transmembrane β1 subunits. Transgenic mice that lack transmembrane β1 but maintain expression of Scn1b-IR phenocopy Scn1b null mice, a model that lack both transcripts.Conclusions:
Our work suggests that SCN1B-IR may serve multiple, cell type specific functions to modulate β1 transcript levels. Furthermore, SCN1B c.449-2A >G patient derived neurons improperly splice intron 3, resulting in the absence of transmembrane b1 expression and EIDEE. Taken together, we find that SCN1B-IR is an NMD-sensitive transcript that cannot compensate for the absence of transmembrane β1.Funding:
NIH R01-HL149363 and NIH R37- NS076752