Genetic and Epigenetic Fine-Mapping of Causal Autoimmune Disease Variants. Genome wide association studies (GWAS) have identified loci underlying human diseases, but the causal nucleotide changes and mechanisms remain largely unknown. In collaboration with Brad Bernstein, David Hafler and Mark Daly, we leveraged new dense genotyping data to develop PICS (Probabilistic Identification of Causal SNPs), an algorithm that explicitly predicts the probability that each individual SNP represents a causal variant for 21 autoimmune diseases (Farh and Marson et al., Nature 2015). To investigate the functions of the candidate causal non-coding variants, we generated a large resource of epigenomic and transcriptomic maps for highly specific human immune cell subsets, including Tregs, naïve CD4+ T cells, and ex vivo stimulated T cell subsets.Integrating fine-mapped genetic data with orthogonal epigenomic maps for 33 cell types, including T cell subsets, revealed the activity patterns of cis-regulatory elements that coincide with PICS SNPs, predicting cell types contributing to each phenotype. Causal autoimmune disease SNPs strongly mapped to regulatory elements active in CD4+ T cell subsets. This study begins to parse human diseases by regulatory elements, target genes and cell types affected by candidate causal disease variants. The data highlight crucial regulatory elements in T cell circuitry.

Autoimmune disease risk SNPs (single nucleotide polymorphisms) map to T cell enhancers. A) Heatmap displays the correlation between fine-mapped SNPs associated with diseases and active enhancers/promoters in specific cell types. Red bars highlight CD4+ T cells subsets (Y-axis) and autoimmune disease (X-axis). B) Enhancers (identified based on H3K27ac patterns) with PICS fine-mapped autoimmunity SNPs (shaded) are shown at the PTGER4 locus (adapted from Farh and Marson et al., Nature 2015).


Genome Engineering Primary Human T Cells. Functional testing of human genome sequences in primary immune cells has been largely impossible until recently, but our recent advances in genome engineering methods offer new opportunities. We developed ­– in collaboration with Jennifer Doudna – a robust CRISPR/Cas9-based technology that enables both “knock-out” and “knock-in” genome editing in primary human T cells (Schumann and Lin et al., PNAS 2015). These advances now allow us to pursue a functional genomic strategy to test how coding and non-coding genetic variation controls essential programs for T cell differentiation and function. Human T cell genome engineering also holds great promise for cell-based therapies for cancer, HIV, primary immune deficiencies, and autoimmune diseases.


Cas9 RNPs provide robust ‘knock-in’ technology to replace nucleotides in primary human T cells.A) Experimental scheme of Cas9:single-guide RNA ribonucleoprotein (Cas9 RNP) and HDR template delivery to primary human T cells for genome editing. B) ‘Knock-in’ of HindIII site in the CXCR4 gene was confirmed with restriction enzyme digestion. C) Cas9 RNPs successfully ablated CXCR4 surface expression.


Chemical Inhibition of Pathogenic T Cell Circuitry. We applied genomic analyses to develop a strategy for pharmacological inhibition of the core transcriptional circuitry of Th17 cells, a pro-inflammatory CD4+ T cell subset that plays a major role in autoimmune diseases (Xiao and Yosef et al., Immunity 2014). In collaboration with Aviv Regev and Vijay Kuchroo, we discovered that RORγt, the major transcriptional regulator of Th17 cells, directly controls the expression of a set of genes at the core of Th17 cell identity and also contributes to the repression of signature genes of other CD4+ T cell lineages, including Tregs. Through a library screen, we identified several novel small molecules that bind to RORγt. We demonstrated in vitro and in vivo that these compounds repress the development of Th17 cells and ameliorate murine models of autoimmunity. We used ChIP-Seq and RNA-Seq to dissect the molecular effects of these potential pharmaceutical agents. This integrative genomic approach is generally applicable to decipher gene regulatory circuitry and aid in drug discovery efforts. Our demonstrated ability to ‘drug’ transcriptional programs that differentiate Tregs and pro-inflammatory T cells motivates our ongoing systematic characterization of the human CD4+ T cell circuitry.

 TMP778 (a) is a novel small molecule that inhibits Th17 differentiation (b) and ameliorates murine model of multiple sclerosis (c) by inhibiting the transcriptional circuitry controlled by RORγt (RORC) (d).