Gluten and the Gut--Lessons for Immune Regulation

See allHide authors and affiliations

Science  27 Sep 2002:
Vol. 297, Issue 5590, pp. 2218-2220
DOI: 10.1126/science.1077572

Celiac sprue is an inflammatory disease that leads to destruction of the microscopic fingerlike projections of the small intestine called villi. The disease is triggered by ingestion of the gluten proteins contained in wheat, barley, and rye, and symptoms range from minor complaints to severe nutrient malabsorption (1). The report by Shan et al. on page 2275 of this issue (2) significantly enhances our understanding of celiac sprue pathogenesis and hints at potential intervention strategies to treat this common disease.

There are three remarkable features of celiac sprue: (i) It usually remits upon strict dietary avoidance of gluten; (ii) it requires a unique genetic background for antigen presentation—expression of the human leukocyte antigen (HLA) class II molecules DQ2 or DQ8; and (iii) patients have characteristic circulating mucosal (immunoglobulin A) autoantibodies to the ubiquitous enzyme tissue transglutaminase (tTGase) (3, 4). What is intriguing is the connection between these three features. Gluten peptides presented in the context of HLA-DQ2 or HLA-DQ8 molecules elicit proliferation of intestinal T cells from celiac sprue patients and induce these cells to release inflammatory cytokines (5). The autoantigen, tTGase, catalyzes transamidation between a glutamine residue of peptide 1 (glutamine donor) and a lysine residue of peptide 2 (glutamine acceptor), creating stable covalent complexes among a limited set of mostly extracellular matrix proteins (6). This enzyme is highly expressed in the subepithelial cells of the gut, where it is stored in an intracellular inactive form. It is released in response to mechanical or inflammatory stress and is activated by high extracellular calcium levels. The strong affinity of tTGase for gluten reflects the fact that 30 to 50% of the amino acids in gluten are glutamine. This enzyme induces formation of aggregates of gluten and other antigens, which seems to be important for efficient antigen uptake by antigen-presenting cells of the immune system (7). At low pH or in the absence of glutamine acceptors, tTGase deamidates certain glutamine residues of gluten to glutamic acid. This posttranslational modification enhances binding of gluten epitopes to HLA-DQ2 or HLA-DQ8 and potentiates their ability to stimulate T cells (8). Several immunodominant gluten peptides (all substrates of tTGase) have been identified, but it is unclear to what extent these peptides reach the small intestinal mucosa after exposure to gastric and duodenal proteases.

New ways to treat celiac sprue.

Molecular pathogenesis of celiac sprue and possible therapeutic interventions. Dietary gluten peptides reach the subepithelial connective tissue (lamina propria), particularly once intestinal mucosal integrity is impaired during infection or during mechanical or chemical injury. In genetically susceptible (HLA-DQ2/HLA-DQ8-positive) individuals, certain gluten peptides are displayed on professional antigen-presenting cells, in particular on dendritic cells, but also on B lymphocytes and activated intestinal epithelia. This results in driving of the CD4+ T cell response toward either inflammation and tissue remodeling (TH1 response) or antibody production (TH2 response). The enzyme tTGase is released from endothelia, fibroblasts, and inflammatory cells residing in the lamina propria, where it encounters its prime external substrate, gluten. Cross-linking of gluten by tTGase potentiates its uptake and presentation by antigen-presenting cells, and its deamidation enhances its binding to HLA-DQ2/HLA-DQ8, resulting in the triggering of a vigorous T cell response. Possible intervention strategies (numbered boxes) include (1) addition of bacterial endopeptidase to the diet to destroy antigenic gluten peptides, such as the 33-amino acid peptide, that are resistant to gastrointestinal proteases; (2) treatment with tTGase inhibitors that block potentiation of gluten antigenicity; (3) directing dendritic cell differentiation toward promotion of CD4+ T cell anergy or induction of tolerogenic (Tr1 and TH3) T cells by, for example, early exposure to lipopolysaccharide (LPS) or the cytokines IL-10 and transforming growth factor-β1 (TGFβ1). (Not shown is the fact that dendritic cells and T cells circulate to mesenterial lymph nodes where important encounters with antigen take place, before returning to the lamina propria.)


In the new work, Shan and colleagues (2) isolated a unique 33-amino acid peptide from the 266-amino acid α2-gliadin (the homologous gliadins represent the major storage proteins of wheat and harbor most of gluten's antigenic epitopes). They tested the resulting peptide fractions against different gliadin-reactive T cell lines in the context of HLA-DQ2. Currently, more than 10 antigenic epitopes among the more than 50 gluten proteins have been described, and as many as 50 additional epitopes are thought to exist (9). Remarkably, the 33-amino acid peptide (which contained six partially overlapping antigenic hot spots) stimulated all of the T cell lines that the authors tested. Moreover, deamidation by tTGase potentiated this T cell stimulation. Addition of a bacterial prolyl endopeptidase caused degradation and loss of the antigenicity of this peptide, which is normally resistant to breakdown by endogenous proteases. This finding opens up the possibility that bacterial endopeptidases could be used to “detoxify” this and other immunodominant peptides of gluten.

Celiac sprue is perhaps the most common human genetic disorder, with a prevalence of 0.5 to 1% in Western, Arabian, and Indian populations. Patients with celiac sprue have to maintain a life-long, strictly gluten-free diet. Thus, there is a strong need to find ways to detoxify wheat, so that it can be eaten by these individuals. Current approaches include creation of a genetically modified wheat that is devoid of antigenic gliadin sequences. This assumes that the glutenins, which are structurally unrelated to gliadins and which are needed for their baking properties, do not harbor T cell-stimulatory sequences. However, two “toxic” glutenin epitopes have been described, and many more are thought to exist (10). This would necessitate the “knockout” of essentially all genes encoding wheat-storage proteins. Apart from general reservations about genetically modified foods, the product would probably have lost all of the properties of wheat. As Shan et al. suggest, supplementing the diet with a bacterial endopeptidase that destroys major immunodominant gliadin epitopes—and probably glutenin peptides as well (10)—is attractive and appears practical, because the protease could be ingested along with a diet containing gluten.

Research into celiac sprue pathogenesis has focused on gluten-reactive T cell lines or clones and the destructive T helper cell 1 (TH1) reaction induced by immunodominant gluten peptides. However, intestinal immune regulation is complex, and we poorly understand mechanisms of T cell anergy or active suppression that are central to distinguishing between beneficial nutritional (or microbial) antigens and detrimental antigens in the gut. HLA-DQ2 or HLA-DQ8 is found in roughly 30% of Western populations, but celiac sprue is encountered in only 1 out of 50 carriers. Thus, most carriers must harbor some form of immune protection. Key immune regulators are dendritic cells—potent antigen-presenting cells that can prime T cells not only for destruction, but also for tolerance or anergy (11, 12). In addition, the route and sequence of antigen supply determine whether the T cell response will be destructive or suppressive, as illustrated by experimental autoimmune neuritis. This autoimmune disease can be induced in mice by subcutaneous vaccination with an oligomeric, immunodominant myelin P2 peptide; in contrast, the same peptide after intravenous or oral application prevents or even reverses established disease (13).

But dendritic cell and suppressor T cell biology is ill-defined in celiac sprue. In addition, numerous external factors are involved in gastrointestinal nutrient and gluten processing. These include (i) gastric acid, which regulates pepsin activity and can be modified by nutrient supply or anti-acid medications like proton-pump inhibitors; (ii) the activity of pancreatic and small intestinal proteases, which relies on pancreatic function; and (iii) the small intestinal microbial milieu, which may alter epithelial permeability, admitting an influx of gluten and other antigens.

The multiple levels of immune regulation explain the observed broad spectrum of gluten sensitivity in patients with celiac sprue, and will allow us to test several intervention strategies. With the ease of obtaining duodenal biopsies, immunomodulatory therapies can be verified in celiac sprue patients, an approach that should benefit all individuals who suffer from immune diseases of the intestinal system.

References and Notes

View Abstract

Stay Connected to Science

Navigate This Article