Expression analysis in intestinal mucosa reveals complex relations among genes under the association peaks in celiac disease

• Published in: European journal of human genetics : EJHG 2015-08, Vol.23 (8), p.1100-1105
• Main Authors: Plaza-Izurieta, Leticia, Fernandez-Jimenez, Nora, Irastorza, Iñaki, Jauregi-Miguel, Amaia, Romero-Garmendia, Irati, Vitoria, Juan Carlos, Bilbao, Jose Ramon
• Format: Article
• Language: English
• Subjects: Biopsy; Celiac disease; Celiac Disease - genetics; Celiac Disease - metabolism; Celiac Disease - pathology; Diabetes; Diagnosis; Disease; Female; Gene expression; Gene Expression Regulation; Genetic Association Studies; Genetic Predisposition to Disease; Genetics; Genomes; Genotypes; Histocompatibility antigen HLA; Humans; Immune response; Intestinal Mucosa - metabolism; Intestinal Mucosa - pathology; Intestine; Laboratories; Male; Mucosa; Pathogenesis; Patients; Physical anthropology; Physiology; Polymorphism, Single Nucleotide; Quantitative Trait Loci - genetics; Rheumatoid arthritis; RNA, Long Noncoding - genetics; Single-nucleotide polymorphism; Small intestine; Studies; Transcription
• ISSN: 1018-4813; 1476-5438
• DOI: 10.1038/ejhg.2014.244

Celiac disease is a chronic immune-mediated disorder with an important genetic component. To date, there are 57 independent association signals from 39 non-HLA loci, and a total of 66 candidate genes have been proposed. We aimed to scrutinize the functional implication of 45 of those genes by analyzing their expression in the disease tissue of celiac patients (at diagnosis/treatment) compared with non-celiac controls. Moreover, we investigated the SNP genotype effect in gene expression and performed coexpression analyses. Several genes showed differential expression among disease groups, most of them related to immune response. Multiple trans-eQTLs but only four cis-eQTLs were found, and surprisingly the genotype effect seems to be stimulus dependent as it differs among groups. Coexpression levels vary from higher to lower levels in active patients at diagnosis, treated patients and non-celiac controls respectively. A subset of 18 genes tightly correlated in both groups of patients but not in controls was identified. Interestingly, this subset of genes was influenced by the genotype of three SNPs. One of the SNPs, rs1018326 on chromosome two is on top of a known lincRNA whose function is not yet described, and whose expression seems to be upregulated in active disease when comparing biopsy pairs from the same individuals. Our results strongly suggest that the effects of disease-associated SNPs go far beyond the oversimplistic idea of transcriptional control at a nearby locus. Further investigations are needed to determine how each variant disrupts fine-tuning mechanisms in the genome that eventually lead to disease.