Immunogenomic approaches to understand the function of immune disease variants

Abstract

Mapping hundreds of genetic variants through genome wide association studies provided an opportunity to gain insights into the pathobiology of immune-mediated diseases. However, as most of the disease variants fall outside the gene coding sequences the functional interpretation of the exact role of the associated variants remains to be determined. The integration of disease-associated variants with large scale genomic maps of cell-type-specific gene regulation at both chromatin and transcript levels deliver examples of functionally prioritized causal variants and genes. In particular, the enrichment of disease variants with histone marks can point towards the cell types most relevant to disease development. Furthermore, chromatin contact maps that link enhancers to promoter regions in a direct way allow the identification of genes that can be regulated by the disease variants. Candidate genes implicated with such approaches can be further examined through the correlation of gene expression with genotypes. Additionally, in the context of immune-mediated diseases it is important to combine genomics with immunology approaches. Genotype correlations with the immune system as a whole, as well as with cellular responses to different stimuli, provide a valuable platform for understanding the functional impact of disease-associated variants. The intersection of immunogenomic resources with disease-associated variants paints a detailed picture of disease causal mechanisms. Here, we provide an overview of recent studies that combine these approaches to identify disease vulnerable pathways.

Keywords: activation; autoimmunity; cell activation; genomics.

Figure 1

Immunogenomic approaches to infer the role of disease variants. Variants associated to immune diseases can be functionally annotated using immunological or a genomic approaches. Immunological approaches include the (a) systemic analysis of immune function, such as measuring cell type ratios and cytokine levels, and (b) cellular analysis, such as proliferation assays of a specific cell type in response to a stimulus, the lineage specification markers or protein expression levels. Genomic assays, include (c) gene expression and (d) epigenetics; transcription factor binding sites, chromatin accessibility, histone modifications and DNA methylation.

Figure 2

Cell‐type‐specific and cell‐state‐specific expression quantitative trait loci. Cellular phenotypes, such as gene expression, cytokine secretion or chromatin accessibility, might be affected by a genetic variant only in a specific cell type and under specific conditions, e.g. at a specific time‐point following a stimulation. Here, a bulk of cell types (upper panel), as well as each cell type individually, were stimulated for 2 and 24 hr. In all scenarios, the measured phenotype, e.g. gene expression, increased upon stimulation, but only in the green cell type (middle panel) was the effect correlated with the genotypes. The effect was missed when measured in the sample containing the mixed cell population (upper panel). The blue cell type (bottom panel) expresses the strongest up‐regulation in expression upon stimulation and largely drives the observed increase in expression in the bulk sample.

Figure 3

Causal disease variants overlap cell‐type‐specific chromatin marks. Chromatin immunoprecipitation followed by sequencing and chromatin accessibility assays produce pile ups of reads that form ‘peaks’. Such genomic annotations generated from different tissues [lymph nodes (green), lungs (blue), femur (pink)] provide a valuable roadmap of cell‐type‐specific genome activity. The genome annotations can be overlapped with genetic variants associated with a phenotype of interest, such as an autoimmune disease, represented as grey circles. If a statistically significant proportion of associated variants overlaps with peaks specific to a cell type it can point towards disease‐relevant tissue and prioritize the causal variants. Here, we illustrate a single‐associated locus where only one single nucleotide polymorphism (red circle) overlaps with a peak specific to the lymph nodes and absent from the other two tissues.