OA30 Identification of causal genes and mechanisms by which genetic variation mediates juvenile idiopathic arthritis susceptibility using functional genomics and CRISPR-Cas9

Abstract

Abstract Background/Aims We recently performed the largest juvenile idiopathic arthritis (JIA) genome-wide association study (GWAS) to date. Disease-associated loci contain multiple single nucleotide polymorphism (SNPs), and the majority map to non-coding enhancers, making it challenging to define causal variants and genes. Functional genomics datasets in disease-relevant tissues have been shown to be essential for the functional interpretation of GWAS loci. In particular, capture Hi-C (CHi-C) has been successful in detecting chromosomal interactions linking GWAS loci to their target genes. However, such datasets are lacking in JIA. The aim of this study is to bridge this gap and advance the knowledge of the biological mechanisms that underpin susceptibility to JIA, by integrating GWAS with public epigenomics datasets and in-house generated CHi-C from JIA patients. We focus on CD4+ T-cells, which have been shown to be one of the most relevant cell types in JIA. In addition, we use CRISPR-Cas9 to validate the regulatory effect of prioritised variants on their predicted target genes. Methods Credible SNP sets for the top JIA risk loci (P < 5x10-6) were annotated using EpiMap data. Low input whole genome promoter CHi-C (PCHi-C) was performed on CD4+ T-cells isolated from blood from 3 JIA oligoarthritis patients, and data was analysed using CHiCAGO. We employed CRISPR activation (CRISPRa) and CRISPR interference (CRISPRi) in Jurkats to assess whether prioritized JIA variants are capable of regulating the expression of the interacting genes. Results 614 SNPs (out of 735) were found to overlap active enhancers in CD4+ T-cells, and were prioritized for further analysis. We identified numerous significant chromatin interactions in 19 out of 44 non-MHC JIA associated loci, linking JIA SNPs mapping to T-cell enhancers to a total of 61 target genes and revealing potential novel disease pathways. A JIA-associated locus on chromosome 3 contains 39 SNPs. It maps to an intergenic region and the causal gene/s are unclear. Our PCHi-C data revealed that this JIA locus presents chromatin interactions with the promoters of several genes, such as CCRL2, CCR2, CCR3 and CCR5. Two variants were selected for further analysis: rs79815064, which had the highest posterior probability, and rs8005404, the only variant within a CD4+ T-cell enhancer linked to surrounding gene activity. When both SNPs were targeted with CRISPRa and CRISPRi, we observed an increased and decreased expression, respectively, of CCRL2, CCR2, CCR3 and CCR5, confirming their role in disease. These genes belong to the chemokine receptor family and are important regulators of the inflammatory response. Conclusion Our work shows how functional genomics can help identify biological mechanisms by which GWAS variants increase risk of JIA, which in turn will benefit patients through personalised medicine and the identification of therapeutic targets. Disclosure A. Frantzeskos: None. V. Malysheva: None. C. Shi: None. J. Ding: None. J. Bowes: None. W. Thomson: None. S. Eyre: None. M. Spivakov: Shareholder/stock ownership; M.S. is co-founder and shareholder of Enhanc3D Genomics Ltd. G. Orozco: None.