Expression QTL analysis from primary immune cells of a multi-ethnic cohort identifies novel disease-causing regulatory effects. B. E. Stranger1,2,3,4,5, T. Raj3,4,5,6, C. Ye3, S. Mostafavi7, K. L. Rothamel8, M. Lee3,5, J. M. Replogle3,4,6, T. Feng8, S. H. Imboywa6, M. Lee6, C. McCabe3,6, D. Koller7, A. Regev3, N. Hacohen3,9, C. O. Benoist3,5,8,, P. L. De Jager3,4,5,6, The Immunological Variation Consortium 1) Section of Genetic Medicine, University of Chicago, Chicago, IL; 2) Institute for Genomics and Systems Biology, University of Chicago, Chicago, IL; 3) The Broad Institute of Harvard and MIT, Cambridge, MA; 4) Division of Genetics, Brigham and Women's Hospital, Boston, MA; 5) Harvard Medical School, Boston, MA; 6) Department of Neurology, Brigham and Womens Hospital, Boston, MA; 7) Department of Computer Science, Stanford University, Stanford, CA; 8) Department of Microbiology and Immunobiology, Harvard Medical School, Boston, MA; 9) Department of Medicine, Massachusetts General Hospital, Boston, MA.
The extent to which human genetic variation drives variation in immune function is critical to the systematic dissection of altered immune function underlying complex immune-mediated disease. As part of The Immunological Variation Consortium, we have performed an expression quantitative trait locus (eQTL) study in two primary human immune cell-types representing adaptive and innate immune function. Gene expression levels were quantified in highly purified CD4+ T cells and CD14+CD16- monocytes from 162 individuals of African American, 155 East Asian and 377 European ancestry, that were genotyped and imputed to approximately 10M SNPs.
We show that the small differences (6-10%) in cis-eQTL found among populations are driven primarily by population divergence in allele frequencies. Similarly, after conducting a multi-ethnic meta-analysis for both T cells and monocytes, we find that most (63%) cis-eQTLs are shared between the two cell types. Combining data across populations and cell types increases the total number of cis-associations (6,546 and 5,632 genes at FDR 0.05 in monocytes and T cells, respectively), as well as enhances our ability to refine the location of the associations due to differences in LD among populations.
We identified 892 trait-associated SNPs that are also significant eQTLs, where 33% of the eQTLs have not been previously reported. We find that susceptibility alleles for inflammatory diseases display a preponderance of regulatory effects in T cells, the adaptive arm of the immune system. In exception to this trend, Alzheimers and Parkinsons disease (PD) susceptibility alleles show a striking enrichment of effects in monocytes, the innate arm of the immune system. We report evidence for several disease-associated variants affecting RNA expression in trans: for example, the PD-associated MAPT H1 haplotype drives the expression of MAPK8IP1 (p=3.48 x 10-32), and a shRNA knockdown experiment suggests the effect is mediated by MAPT expression and not one of the other 5 genes found in this haplotype. Our multi-ethnic exploration of both arms of the immune system highlights a remarkable level of cis-eQTL sharing across populations and cell types, and provides important new insights into the role of immune cell populations in the onset of neurodegenerative, inflammatory, and other diseases.
You may contact the first author (during and after the meeting) at