Poster Presentation 47th Lorne Genome Conference 2026

Dissecting the genetic consequences of non-coding autoimmune risk variants in primary B cells (132841)

Jeralyn Wen Hui Ching 1 , Viacheslav Kriachkov 1 , James Lancaster 1 , Esther Bandala-Sanchez 1 , Melaine Neeland 2 , Shivanthan Shanthikumar 2 , Liam Gubbels 2 , Stephen Nutt 1 , Hamish King 1
  1. Genetics and Gene regulation, Walter and Eliza Hall Institute of Medial Research, Melbourne, 3, Australia
  2. Respiratory Group, Murdoch Children's Research Institute, Parkville, VIC, Australia

Over 90% of disease-risk variants identified by genome-wide association studies reside in the non-coding genome, with many located within open chromatin that are presumed to contain distal regulatory elements (DRE). Non-coding variants can potentially influence phenotypic outcomes by disrupting DREs thereby altering gene expression and downstream pathways. However, deciphering how these variants contribute to disease risk remains a significant challenge, particularly in primary human B cells, which are traditionally resistant to genetic manipulation. Here, we selected an autoimmune risk locus, CD83, a gene important for B cell immune response, and carries many GWAS SNPs in proximity. We performed a CRISPR activation (CRISPRa) screen in primary B cells by tiling a 300 kb region of the CD83 locus and sorted for the top 20% of CD83-expressing cells. Regions enriched for gRNAs corresponded to DREs that regulate CD83 expression and typically overlaps with open chromatin within the locus. Three statistically fine-mapped SNPs (rs74405933, rs12529514, rs12530098) linked with rheumatoid arthritis were then prioritized to test for the impact of these risk variants. We used a CRISPR prime targeting approach (Martyn et al., Cell, 2025) to introduce risk alleles into primary B cells, followed by flow-sorting based on CD83 expression. Edited cells carrying the risk allele were enriched in the low CD83-expressing cell population, consistent with a loss-of-function effect for each SNP. We have subsequently applied this method to determine the consequence of other autoimmune risk variants on their candidate target genes (identified by single-cell CRISPRa screen) including rs1432296 (SLE) of which we observed a gain-of-function effect for REL. Here, we present an approach to determine the consequence of non-coding disease-associated variants in primary human B cells. This approach bridges the gap between statistical association and functional consequence of GWAS-predicted variants, providing a foundation for uncovering causal regulatory mechanisms underlying disease.