Genetics and pathophysiology of systemic juvenile idiopathic arthritis and other complex autoinflammatory diseases

National Institute Of Arthritis And Musculoskeletal And Skin Diseases

Investigators

Linked publications, trials & patents

Abstract

We have continued to apply genomics-forward approaches to drive our translational investigations of Still’s disease in children (systemic juvenile idiopathic arthritis or sJIA) and adults (adult-onset Still’s disease or AOSD) during the current reporting period. We are undertaking a large, collaborative case-control association study to identify genetic risk factors for Still’s disease at the population level. Using family-based sequencing, we are exploring the role of high impact genetic variants in severely affected individuals. We are leading a collaborative genetic study of treatment-refractory forms of Still’s disease. We are exploring genetic and transcriptomic biomarkers that are predictive of disease course and therapeutic outcomes in Still's disease. All of these endeavours are designed to identify novel therapeutic targets or approaches that we hope to apply to the treatment of patients at the NIH Clinical Center. In addition to our studies of Still’s disease, we have continued to explore variation of phospholipase C gamma 2 (PLCG2) in patients with immune dysregulation. The following are 2 examples of work that we have undertaken in the current reporting period. 1. Interferon stimulated gene signature in Still’s disease associates with lung disease, drug reactions and rare variation in interferon-linked pathways. A subset of patients with Still’s disease can develop a severe form of lung disease (LD) associated with drug-associated immune reactions (DAIR) to anakinra, canakinumab and tocilizumab. Although the underlying cause of this syndrome remains unknown, our research has demonstrated that stopping these medications can lead to significant clinical improvement in affected individuals. Therefore, identifying clinical or laboratory biomarkers that predict which patients are at greatest risk for this emerging complication is critically important. We have previously described a strong association between Still’s LD and DAIR and the HLA-DRB1*15 allele, which is present in approximately 80% of LD-DAIR patients. However, since not all Still’s disease patients with HLA-DRB1*15 develop LD-DAIR, this allele's predictive value as a therapeutic biomarker is limited. Several studies have described enhanced expression of interferon (IFN) stimulated genes in people with severe Still’s disease, leading us to consider whether high ISG expression (an IFN signature) could hold value as a biomarker. To evaluate the relationship between Still’s disease and IFN signaling, we measured whole blood IFN stimulated gene (ISG) expression in 57 consecutive participants in our natural history study of Still’s disease at the time of enrollment. This revealed a group of 16 patients with an IFN signature that set them apart from the rest of the cohort. When the clinical characteristics of IFN signature patients were compared to the rest of the group, DAIR (p = 0.003, odds ratio 7.3) and LD (p = 0.007, odds ratio 6.9) were the features that most strongly associated with the IFN signature. The IFN signature was not associated with either HLA-DRB1*15 status or biomarkers indicative of Still’s disease activity (interleukin-18, C-reactive protein, ferritin). Importantly, when the IFN signature was combined with the presence of HLA-DRB1*15 alleles, the presence of both variables yielded post-hoc identification of DAIR and LD in our cohort with high sensitivity (0.88 and 1.0, respectively), while their combined absence displayed high negative predictive values (0.93 and 1.0, respectively). To evaluate whether genetic variation contributed to the IFN signatures, exome sequencing was performed with the 16 IFN signature patients and their parents. The results were subjected to family-based variant prioritization, which produced a list of 89 candidate genes that harbored de novo or recessive variants among the 16 participants. Pathway-based over-representation analysis (ORA) of this candidate gene list revealed enrichment of genes in IFN-linked pathways, including toll-like receptor signaling (34.7-fold enrichment), chaperone-mediated autophagy (29.8-fold enrichment), macrophage activation (10.4-fold enrichment) and type I IFN biosynthesis (9.0-fold enrichment). ORA of the candidate gene list from individuals without an IFN signature revealed no enriched pathways. This study nominates the combination of HLA-DRB1*15 and an IFN signature as a candidate biomarker with the potential to predict LD and DAIR in Still’s disease. Moreover, this suggests that rare de novo and recessive variation could promote enhanced IFN signaling and contribute to LD and DAIR. A manuscript describing these findings is in revision. 2. A novel multiplex cDNA sequencing assay reveals de novo and splice-site variants causing PLCG2-associated immune dysregulation with cold urticaria (PLAID-CU). In 2012, we reported the first human cases of immune dysregulation associated with variation in PLCG2. These included three families with fully penetrant, autosomal dominant cold urticaria and immune dysregulation caused by dominant-negative, in-frame deletions of PLCG2. These deletions removed a critical regulatory region while leaving the remainder of the protein intact. Over the subsequent 13 years, our ongoing studies have examined the relationship between PLCG2 variation and immune-mediated phenotypes in more than 100 probands. This work has defined a spectrum of PLCG2-associated immune dysregulation (PLAID) that includes mixed dominant-negative variants with spontaneous cold-induced activation (PLAID with cold urticaria, or PLAID-CU); gain-of-function variants (PLAID-GOF); and loss-of-function variants (PLAID-LOF).In our original report, each of the three families with PLAID-CU had deletions arising from distinct historical Alu-mediated recombination events. Within our larger cohort, we identified nine individuals or families presenting with cold urticaria and immune dysregulation clinically indistinguishable from PLAID-CU, but lacking PLCG2 genomic deletions detectable by standard methods. We hypothesized that non-deletional lesions, such as splice-site variants, could produce PLCG2 transcripts with in-frame deletions similar to those observed in PLAID-CU. To test this hypothesis, we developed a PLCG2-specific complementary DNA (cDNA) sequencing assay to detect PLAID-CU–associated transcripts independent of the underlying genetic variant. Using PLCG2-specific reverse transcription, we generated full-length cDNA from PLCG2 mRNA in primary leukocytes, followed by high-throughput Illumina sequencing. This approach successfully identified expected transcripts in patients with known deletions of exon 19 or exons 20–22. By incorporating unique molecular barcodes, we adapted the assay to evaluate multiple mutation-negative probands simultaneously in a multiplexed format. This revealed two novel disease-causing PLCG2 mutations, representing the fourth and fifth families with PLAID-CU. The first was a splice-site variant that generated a transcript lacking exon 19, recapitulating the effect of known exon 19 deletions. The second was a novel de novo deletion in a young boy, removing exons 19–22 — a pattern not previously reported. This novel multiplex PLCG2-specific cDNA sequencing assay provides an unbiased, accurate, sensitive, and efficient method to detect the full range of PLCG2 transcript variants in primary leukocytes. These findings were published in the Journal of Allergy and Clinical Immunology.