Celiac disease biomarkers identified by transcriptome analysis of small intestinal biopsies

Hanna Bragde,Mats Fredrikson,Ulf Jansson,Jan Söderman,Ewa Grodzinsky

doi:10.1007/s00018-018-2898-5

Abstract

Establishing a celiac disease (CD) diagnosis can be difficult, such as when CD-specific antibody levels are just above cutoff or when small intestinal biopsies show low-grade injuries. To investigate the biological pathways involved in CD and select potential biomarkers to aid in CD diagnosis, RNA sequencing of duodenal biopsies from subjects with either confirmed Active CD (n = 20) or without any signs of CD (n = 20) was performed. Gene enrichment and pathway analysis highlighted contexts, such as immune response, microbial infection, phagocytosis, intestinal barrier function, metabolism, and transportation. Twenty-nine potential CD biomarkers were selected based on differential expression and biological context. The biomarkers were validated by real-time polymerase chain reaction of eight RNA sequencing study subjects, and further investigated using an independent study group (n = 43) consisting of subjects not affected by CD, with a clear diagnosis of CD on either a gluten-containing or a gluten-free diet, or with low-grade intestinal injury. Selected biomarkers were able to classify subjects with clear CD/non-CD status, and a subset of the biomarkers (CXCL10, GBP5, IFI27, IFNG, and UBD) showed differential expression in biopsies from subjects with no or low-grade intestinal injury that received a CD diagnosis based on biopsies taken at a later time point. A large number of pathways are involved in CD pathogenesis, and gene expression is affected in CD mucosa already in low-grade intestinal injuries. RNA sequencing of low-grade intestinal injuries might discover pathways and biomarkers involved in early stages of CD pathogenesis.

Highlights

Celiac disease (CD) is an immune-mediated systemic disorder elicited by gluten and related prolamines in genetically susceptible individuals, with the presence of a variable combination of gluten-dependent clinical manifestations, celiac disease (CD)-specific antibodies, such as IgA autoantibodies against tissue transglutaminase, human leukocyte antigen (HLA)-DQ2 or HLA-DQ8 haplotypes, and enteropathy [1]
In a principal component analysis (PCA) based on all of the genes with a mean expression > 0.3 RPKM, three principal components (PCs) each accounted for more than 10% of the total variation (PC1 = 27%, PC2 = 22%, and PC3 = 11%; visualized in Online Resources 2 and 3). Sample coordinates along these three PCs were analyzed using one-way analysis of variance (ANOVA) with respect to categorical variables and using Spearman rank correlation with respect to continuous variables to identify relationships between these variables and the PC coordinates
Marsh grade was found to be associated with the coordinates along all three PCs (PC1, PC2, and PC3: false discovery rate (FDR)-adjusted p values = 0.024, 0.0018, and 5.9E−07, respectively), but no significant associations were found between PCA coordinates and gender

Summary

Introduction

Celiac disease (CD) is an immune-mediated systemic disorder elicited by gluten and related prolamines in genetically susceptible individuals, with the presence of a variable combination of gluten-dependent clinical manifestations, CD-specific antibodies, such as IgA autoantibodies against tissue transglutaminase (anti-TG2), human leukocyte antigen (HLA)-DQ2 or HLA-DQ8 haplotypes, and enteropathy [1]. The histological alterations in the small intestine can be graded according to the modified Marsh scale [2, 3], and age has been shown to correlate inversely with intestinal lesion severity and anti-TG2 levels [4]. For children under 2 years of age, anti-TG2 levels can be below cutoff despite presence of Marsh grade 3 intestinal lesions [5]. The presence of anti-TG2 in the blood combined with Marsh grade 3 intestinal lesions is a strong indicator of CD, but diagnosis is less clear in cases with histopathology of Marsh grade 1–2 or in cases with only slightly elevated anti-TG2 levels [6]. The amount of gluten consumed by an individual with CD and the amount of gluten that they can tolerate [10] affect enteropathy and CD-specific antibody levels

Objectives

Methods

Results

Discussion

Conclusion