Combining genetic mapping with genome-wide expression in experimental autoimmune encephalomyelitis highlights a gene network enriched for T cell functions and candidate genes regulating autoimmunity
Melanie Thessen Hedreul
2
Steffen Mo ller
0
1
Pernilla Stridh
2
Yask Gupta
1
Alan Gillett
2
Amennai Daniel Beyeen
2
Johan O ckinger
2
Sevasti Flytzani
2
Margarita Diez
2
Tomas Olsson
2
Maja Jagodic
2
0
Institute for Neuro- and Bioinformatics, University of Lu beck
, Ratzeburger Allee 160, 23538 Lu beck,
Germany
1
Department of Dermatology
2
Department of Clinical Neuroscience, Neuroimmunology Unit, Center for Molecular Medicine L8:04, Karolinska Institutet
, L8:04,
17176 Stockholm, Sweden
The experimental autoimmune encephalomyelitis (EAE) is an autoimmune disease of the central nervous system commonly used to study multiple sclerosis (MS). We combined clinical EAE phenotypes with genome-wide expression profiling in spleens from 150 backcross rats between susceptible DA and resistant PVG rat strains during the chronic EAE phase. This enabled correlation of transcripts with genotypes, other transcripts and clinical EAE phenotypes and implicated potential genetic causes and pathways in EAE. We detected 2285 expression quantitative trait loci (eQTLs). Sixty out of 599 cis-eQTLs overlapped well-known EAE QTLs and constitute positional candidate genes, including Ifit1 (Eae7), Atg7 (Eae20-22), Klrc3 (eEae22) and Mfsd4 (Eae17). A trans-eQTL that overlaps Eae23a regulated a large number of small RNAs and implicates a master regulator of transcription. We defined several disease-correlated networks enriched for pathways involved in cell-mediated immunity. They include C-type lectins, G protein coupled receptors, mitogen-activated protein kinases, transmembrane proteins, suppressors of transcription (Jundp2 and Nr1d1) and STAT transcription factors (Stat4) involved in interferon signaling. The most significant network was enriched for T cell functions, similar to genetic findings in MS, and revealed both established and novel gene interactions. Transcripts in the network have been associated with T cell proliferation and differentiation, the TCR signaling and regulation of regulatory T cells. A number of network genes and their family members have been associated with MS and/or other autoimmune diseases. Combining disease and genome-wide expression phenotypes provides a link between disease risk genes and distinct molecular pathways that are dysregulated during chronic autoimmune inflammation.
-
INTRODUCTION
A predisposition to develop a complex disease such as multiple
sclerosis (MS) is regulated by numerous genetic variants that
each contribute small effects (1). Clinically, MS is characterized
by immune-mediated destruction of myelin sheaths and axons in
the central nervous system, leading to progressive disability.
Despite recent substantial progress in deciphering genetic
variants that contribute to susceptibility (2), little is known about
the functional outcomes of these risk-associated variants, in part
due to limitations in access of relevant human samples.
In addition, identified risk alleles together explain only a fraction
of disease heritability and variance (2). Additional risk variants
conferring small effects may contribute to heritability of
complex diseases, and the clustering of genes (below thresholds
for significant association with MS) into functional networks
supports this hypothesis (3). Unraveling the functions of
susceptibility genes through identification of pathways enriched with
risk genes can reveal mechanisms central in disease regulation.
An animal model widely utilized to characterize the genetic
basis and disease mechanisms of relevance for MS is
experimental autoimmune encephalomyelitis (EAE). Myelin
oligodendrocyte glycoprotein (MOG)-induced EAE in rats mimics many
features of MS (4), including inflammation and demyelination,
relapses and remissions and immune cell infiltration. Linkage
analysis in experimental animal crosses can readily detect
quantitative trait loci (QTLs) related to clinical traits of complex
diseases, and over 50 QTLs have been identified in EAE (5). Several
genes underlying QTL effects in rats were positionally cloned
and a number of them have been subsequently confirmed to
regulate human counterpart (6). However, it has been challenging to
define single quantitative trait genes (7).
Given the high heritability of variation in gene expression (8),
identifying determinants of gene expression may give insights
into pathogenic mechanisms of complex traits. The approach of
mapping quantitative variation in gene expression was introduced
in 2001 (9,10). This approach yields expression QTLs (eQTLs)
(11), which influence expression of transcripts either in cis or in
trans, where cis-acting eQTLs are located in close proximity of
the target gene itself, while trans-acting eQTLs are located in a
region distant from the gene it regulates. Technical artifacts
excluded (e.g. hybridization differences) (12); cis-regulatory
effects can usually be mapped with high statistical significance
and could be explained in most cases by a variation in DNA
sequence in the regula (...truncated)