Abstract 11527: Regional Variation in Cardiovascular Genes Enables a Tractable Genome Editing Strategy

• Published in: Circulation (New York, N.Y.) Vol. 146; no. Suppl_1
• Main Authors: Krysov, Vikki A, Wilson, Rachel H, Ten, Nicholas S, Youlton, Nathan, De Jong, Hannah N, Sutton, Shirley, Huang, Yong, Reuter, Chloe, Grove, Megan E, Wheeler, Matthew T, Ashley, Euan A, Parikh, Victoria N
• Format: Journal Article
• Language: English
• Published: 08.11.2022
• ISSN: 0009-7322; 1524-4539
• DOI: 10.1161/circ.146.suppl_1.11527

Abstract only Recent rise in the number and therapeutic potential of genome engineering technologies has generated excitement in cardiovascular genetics. One significant barrier to their implementation is costly and time consuming reagent development for novel unique variants. Prior data have illuminated the potential for regional clustering of disease-causing genetic variants in known and potential novel functional protein domains. We hypothesized that most cardiovascular disease-relevant genes in ClinVar would display regional variant clustering, and that multiple variants within a regional hotspot could be targeted with limited prime editing reagents. We collated 2,471 high confidence pathogenic or likely pathogenic (P/LP) missense and truncating variants from 111 cardiovascular disease genes. We then defined a regional clustering index (RCI), the percent of P/LP variants in a given gene located within a pre-specified window around each variant. At a window size commensurate with maximally reported prime editing extension length, 78bp, we found that missense variants displayed a higher mean RCI than truncating variants. We next identified genes particularly attractive for pathogenic hotspot-targeted prime editing with at least 20 P/LP variants and observed that the mean RCI in missense variants remained higher than for truncating variants and for rare variants observed in the same genes in gnomAD (Mean±SD RCI 78bp : P/LP ClinVar Missense=5.2±5.9%; P/LP ClinVar Truncating=2.1±3.2%, gnomAD Missense=2±3.2%, gnomAD Truncating=1.1±2.4%. p<2.2e-16; ANOVA with post hoc Bonferroni correction). Further, we tested the feasibility of prime editing for multiple variants in a single hotspot in KCNH2, a gene with a high mean missense variant RCI. Sixty variants were induced in this hotspot in HEK293 cells with CRISPR-X. Correction of these variants was attempted with prime editing using two overlapping prime editing guide RNA extensions. The mean prime editing efficiency across CRISPR-X-enriched variants within this hotspot was 57±27%, including 3 P/LP variants. These data underscore the utility of pathogenic variant hotspots in the diagnosis and treatment of inherited cardiovascular disease.