Fig. 4. Validation of causal variant for Late-Spike cis regulatory program.

(a) Location of 6 fine-mapped non-coding SNPs around HLA-DQB1. Tracks showing open chromatin regions (ATAC-seq) or regions marked by histone modifications (ChIP-seq). (b) CRISPR/Cas9 cuts at or near six fine-mapped SNPs in HH T cell lines. Left, experiment scheme. Middle, representative example of HLA-DQ expression levels. Right, HLA-DQ median fluorescence intensity relative to control, for each of the 6 SNPs, in triplicate. (c) Validation of causal SNP rs71542466 with CRISPR/Ca9 and ssDNA oligo HDR template. Left, experiment scheme. Middle, mRNA HLA-DQB1 quantification with qPCR Taqman assay for 7 wild type (WT, homozygous for C allele), and 7 SNP edited (ALT, homozygous for G allele) expanded clones for rs71542466, as well as a cell line clone with an indel at the same target position. Right, HLA-DQ protein levels measured with flow cytometry. N = 5 WT and 4 ALT clones. (d) Electrophoretic Mobility Shift Assay using nuclear extract of three cell lines, with biotin labeled probes with reference (REF) or alternative (ALT) alleles for rs71542466. Blots were cropped from original shown in Supplementary Figures. (e) Luciferase assay in three cell lines. Cntrl, control. REF, reference allele. ALT, alternative allele. All P-values from Mann-Whitney one-tailed test. Error bars are S.E.M.