Genetic basis of autoimmunity

Abstract

Autoimmune diseases affect up to approximately 10% of the population. While rare Mendelian autoimmunity syndromes can result from monogenic mutations disrupting essential mechanisms of central and peripheral tolerance, more common human autoimmune diseases are complex disorders that arise from the interaction between polygenic risk factors and environmental factors. Although the risk attributable to most individual nucleotide variants is modest, genome-wide association studies (GWAS) have the potential to provide an unbiased view of biological pathways that drive human autoimmune diseases. Interpretation of GWAS requires integration of multiple genomic datasets including dense genotyping, cis-regulatory maps of primary immune cells, and genotyped studies of gene expression in relevant cell types and cellular conditions. Improved understanding of the genetic basis of autoimmunity may lead to a more sophisticated understanding of underlying cellular phenotypes and, eventually, novel diagnostics and targeted therapies.

Figure 2. Integration of genetic and epigenetic fine mapping reveals candidate causal SNPs and pathogenic cell circuits.

(A) The goal of genetic fine mapping is to progress from multiple linked SNPs to one (or very few) candidate causal SNPs. Bayesian algorithms have been developed to fine map credible sets of candidate causal SNPs (37, 86) on the basis of dense genotyping data (e.g., Immunochip data [ref. 76]) or on the basis of imputation of sparser genotyping data to the 1000 Genomes Project (87). (B) Approximately 90% of causal variants associated with autoimmune diseases are noncoding. Genome-wide chromatin maps of active regulatory elements in primary immune cells in both resting and stimulated conditions serve as a powerful resource to identify functional noncoding elements that can be disrupted by disease-associated variants. Candidate causal disease variants (red box) map to regulatory elements (notably enhancers) active in specialized cell types, especially stimulated CD4⁺ T cell subsets. Enhancers contain binding sites for transcription factors and regulate transcription through long-range interactions with RNA polymerase (RNA Pol) machinery (88). The locations of disease-associated SNPs highlight pathogenic cell types and cell conditions based on the activity patterns of affected enhancers. Quantitative trait studies are beginning to characterize the functional effects of causal autoimmune disease variants in modulating transcription factor (TF) binding (circles), chromatin state, target gene regulation, and cellular phenotype. Detailed studies might eventually diagnose specific gene regulatory defects caused by autoimmune disease variants and provide novel targets for therapeutic intervention. Adapted with permission from Immunity (116).

Figure 1. Moving from GWAS loci to cellular pathways.

The causal SNPs that contribute to autoimmune disease risk are often inherited along with neighboring neutral SNPs as a result of linkage disequilibrium. The index SNPs that are genotyped and associated with disease risk in GWAS implicate genomic loci — linkage disequilibrium blocks — composed of multiple linked SNPs (gray boxes). GWAS loci associated with autoimmune disease are enriched in genes (rectangles) that are preferentially expressed in particular immune cell subsets (autoimmune disease cell signatures; bottom left) (38) and encode proteins (circles) that participate in a disproportionate number of direct and indirect physical interactions to form biological pathways (autoimmune disease pathways; bottom right) (41). Expression patterns and protein interaction network analysis have been used to triage candidate genes within linkage disequilibrium blocks. These analyses also suggest pathogenic cell types, protein complexes, and pathways that are affected by disease variants, which begins to elucidate the biology underlying complex autoimmune diseases and could direct drug discovery efforts. Adapted with permission from American Journal of Human Genetics (38) and PLoS Genetics (41).