Ingestion of oats and barley in patients with celiac disease mobilizes cross-reactive T cells activated by avenin peptides and immuno-dominant hordein peptides

Highlights

• Oats ingestion mobilizes avenin-specific T cells in 8% of CD patients in vivo.

• Barley, but not wheat or rye, mobilizes T cells that cross-react with avenin.

• Barley stimulates an avenin-specific T cell response in most CD patients.

• Protease sensitivity and lower HLA binding reduce the immunoreactivity of avenin.

• Low-level oats consumption may be insufficient for clinical relapse in CD patients.

Abstract

Celiac disease (CD) is a common CD4⁺ T cell mediated enteropathy driven by gluten in wheat, rye, and barley. Whilst clinical feeding studies generally support the safety of oats ingestion in CD, the avenin protein from oats can stimulate intestinal gluten-reactive T cells isolated from some CD patients in vitro. Our objective was to establish whether ingestion of oats or other grains toxic in CD stimulate an avenin-specific T cell response in vivo.

We fed participants a meal of oats (100 g/day over 3 days) to measure the in vivo polyclonal avenin-specific T cell responses to peptides contained within comprehensive avenin peptide libraries in 73 HLA-DQ2.5⁺ CD patients. Grain cross-reactivity was investigated using oral challenge with wheat, barley, and rye.

Avenin-specific responses were observed in 6/73 HLA-DQ2.5⁺ CD patients (8%), against four closely related peptides. Oral barley challenge efficiently induced cross-reactive avenin/hordein-specific T cells in most CD patients, whereas wheat or rye challenge did not. In vitro, immunogenic avenin peptides were susceptible to digestive endopeptidases and showed weak HLA-DQ2.5 binding stability.

Our findings indicate that CD patients possess T cells capable of responding to immuno-dominant hordein epitopes and homologous avenin peptides ex vivo, but the frequency and consistency of these T cells in blood is substantially higher after oral challenge with barley compared to oats. The low rates of T cell activation after a substantial oats challenge (100 g/d) suggests that doses of oats commonly consumed are insufficient to cause clinical relapse, and supports the safety of oats demonstrated in long-term feeding studies.

Introduction

Celiac disease (CD) is a common CD4⁺ T cell-mediated disease recently described by Anderson and Mackay as a “singularly compelling model” of autoimmunity with a particularly strong association to specific human leukocyte antigens (HLA) [1]. Activation of gluten-reactive CD4⁺ T cells, restricted by the HLA heterodimers HLA-DQ2.5, HLA-DQ2.2, or HLA-DQ8, is the central event in the immuno-pathogenesis of CD. The illness is characterized by a variable combination of gluten-dependent clinical manifestations, CD-specific antibodies, and enteropathy [2]. Currently, the only treatment for CD patients is the adoption of a strict gluten-free diet (GFD), which excludes wheat, barley, and rye. Occasional CD patients with persistent mucosal damage or digestive symptoms on GFD are also recommended to exclude oats, and in some countries such as Australia, regulatory authorities require products derived from oats be excluded in food labeled “gluten-free” [3]. The disparity between the safety of oats established in feeding studies with the limited evidence supporting in vitro immunotoxicity of oats avenin has meant caution is typically advised when oats are included in the GFD [4].

Dicke's seminal food challenge studies first proposed oats as potentially toxic in patients with CD [5]. However subsequent feeding studies have shown that whilst digestive symptoms such as bloating and diarrhea are not uncommon and can lead to discontinuation of oat challenges [6], [7], [8], [9], the risk of histological relapse is low if oats are confirmed wheat contamination-free. Although there is increasing evidence collected from a number of oats feeding studies suggesting that the consumption of oats is safe for the majority of CD patients [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], the data is still conflicting. There are three published cases of disease relapse in CD patients after the consumption of contamination-free oats, confirmed by intestinal histology and later re-challenge with oats [8], [10]. However, the reappearance of tissue transglutaminase (tTG)-IgA or other CD-specific serology was never detected following long-term oats consumption [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23]. Moreover, two recent studies cast further doubt on the safety of oats, describing that long-term oats consumption in CD patients can be associated with persistent intestinal intra-epithelial lymphocytosis [24] and other sensitive markers, such as intestinal mRNA expression of inflammatory cytokines, are elevated [25].

In vitro studies by Arentz-Hansen et al. were successful in raising polyclonal intestinal T cell lines (TCLs) to avenin protein in five of nine HLA-DQ2.5⁺ CD donors, two of whom did not manifest histological relapse when consuming oats [10]. Three of these five avenin-reactive TCL also responded to deamidated wheat gluten and one of two avenin peptides. Arentz-Hansen et al. derived a single avenin-peptide-specific T cell clone (TCC), but it failed to respond to wheat gluten. In epitope prediction studies supported by in vitro findings with intestinal TCLs raised to wheat gluten, Vader et al. observed that avenin homologs of immuno-dominant α-gliadin T cell epitopes were recognized by gluten-reactive TCLs, but neither peptide was recognized by TCCs specific for the α-gliadin epitopes [26]. These in vitro studies have led some to conclude that activation of T cells specific for wheat gluten by avenin peptides generated during digestion of oats could cause symptoms of oats intolerance and mucosal inflammation in vivo. However, presently there is no direct (in vivo) evidence of gluten-reactive T cells being activated following ingestion of oats.

Prior to the appearance of histological damage and reappearance of CD-specific antibodies, oral gluten challenge in CD volunteers for three days with four slices of bread (approximately 10 g/day gluten) leads to the appearance of circulating gluten-specific CD4⁺ T cells, typical of a recall immune response [27], [28]. Such cereal challenges with wheat, barley, and rye have provided a detailed understanding of the specificity and hierarchy of peptides that activate gluten-specific CD4⁺ T cells in vivo [29]. T cells specific for certain post-translationally modified (deamidated) peptides consistently make a substantial contribution to the total gluten-reactive T cell population mobilized in blood by oral cereal challenge in vivo, or expanded from intestinal tissue in vitro. Many gluten peptides derived from wheat, barley, and rye prolamins are weak agonists in vitro for cross-reactive T cells specific for “dominant” epitopes [29]. In the current study our aim was to determine in vivo the effects of ingesting oats and other cereals on T cells recognizing avenin peptides, and determine if a recall immune response caused by oats is actually because of cross-reactivity of ubiquitous T cells specific for other gluten-containing grains or if it is an idiosyncratic phenomenon occurring only rarely in CD patients.