- Email: [email protected]
Current concepts of celiac disease pathogenesis
GASTROENTEROLOGY 2000;119:234–242
SPECIAL REPORTS AND REVIEWS Current Concepts of Celiac Disease Pathogenesis DETLEF SCHUPPAN First Department of Medicine, University of Erlangen-Nuernberg, Erlangen, Germany
Our knowledge of celiac disease pathogenesis has recently made rapid progress. The disorder is now considered the result of a complex interplay of intrinsic (genetic) and variable extrinsic (environmental) factors that explain the wide spectrum of clinical manifestations ranging from asymptomatic to severe malabsorption. Gluten peptides are efficiently presented by celiac disease–specific HLA-DQ2– and HLA-DQ8–positive antigen-presenting cells, and thus drive the immune response, predominantly in the connective tissue of the lamina propria. Tissue transglutaminase, which has been identified as the highly specific endomysial autoantigen, is released from cells during inflammation. It may potentiate antigen presentation by HLA-DQ2 and HLA-DQ8 by deamidating or cross-linking gluten peptides. The result is lamina propria T-cell activation and mucosal transformation by activated intestinal fibroblasts. In the future, manipulation of the gut-associated lymphoid tissue may allow reduced sensitivity or even generate oral tolerance to gluten. Long-standing untreated celiac disease, even if clinically silent, predisposes for other autoimmune diseases. Therefore, population screening for immunoglobulin A antibodies to tissue transglutaminase seems justified.
he realization of a close link between the ingestion of wheat protein and the clinical manifestation of celiac sprue (celiac disease), previously a severe malabsorptive disorder of childhood, can be considered a milestone of modern medicine. Table 1 summarizes the history of celiac disease and major advances in its understanding.1–22
T
Gluten as a Trigger of Celiac Disease Gluten is the protein fraction of wheat, rye, and barley that confers the properties of stickiness and thus allows the baking of bread. Gluten can be fractioned into the ethanol-soluble prolamines and ethanol-insoluble glutenins. Studies have been performed with more soluble prolamines, but recent data suggest that glutenins can also damage the intestinal mucosa.23–25 A common
feature of the prolamines of wheat is a high content of glutamine (.30%) and proline (.15%), whereas the nontoxic prolamines of rice and corn have a lower glutamine and proline content (Table 2).23–26 In this line, the toxicity of the prolamines of oats (avenins), which have an intermediate amino acid composition, is disputed, and only excessive ingestion of this cereal may be detrimental.27 The wheat prolamines are subdivided into a, b, g, and v gliadins, displaying molecular weights between 20,000 and 75,000 daltons, and containing similar or repetitive glutamine- and proline-rich peptide sequences such as Pro-Ser-Gln-Gln and Gln-Gln-Gln-Pro that appear to be responsible for the observed ‘‘toxicity’’ of gluten in celiac disease. This causal relation has been shown in several studies, by gluten challenge either of intestinal biopsy specimens in vitro or of the proximal and distal intestine in vivo.13,17 Glutamine- and proline-rich peptide sequences are also found in the genetically related prolamines of rye (secalins) and barley (hordeins). The highest allowable daily intake of gluten by celiac patients is a matter of debate. Small intestinal damage is clearly related to the amount of gluten ingested by susceptible individuals. A study from Australia showed that 15 of 22 patients with celiac disease who were still symptomatic while on a ‘‘codex-alimentarius gluten-free diet’’ (defined as ,0.3% gluten from grains) became asymptomatic or improved significantly when adhering to a more rigid ‘‘no detectable gluten diet.’’28 A daily gluten consumption of ,50 mg, compared with an average of about 13 g in most Western countries,29 is considered safe by many experts. However, gluten sensitivity differs among individuals, and there are other factors that trigger or maintain the disease (see below). Matters are further complicated because, despite the recent development of assays for the measurement of toxic Abbreviations used in this paper: APC, antigen-presenting cell; EMA, endomysial autoantibody(ies); HSP, heat-shock protein; MMP, matrix metalloproteinase; TGF, transforming growth factor; TNF, tumor necrosis factor; tTG, tissue transglutaminase. r 2000 by the American Gastroenterological Association 0016-5085/00/$10.00 doi:10.1053/gast.2000.8521
July 2000
CELIAC DISEASE PATHOGENESIS
(a)-gliadins in cereals,30 a precise quantification of potentially toxic cereal proteins that may be present in foods is presently impossible.
Histopathologic Characteristics of the Celiac Lesion On the basis of histologic follow-up of celiac patients who were in remission after a gluten-free diet Table 1. Historical Milestones in Celiac Disease Pathogenesis First report on a condition resembling celiac disease First detailed clinical description of the ‘‘coeliac affection’’ First description of the mucosal transformation characteristic of celiac disease Successful diet consisting of fruit, vegetables, and milk powder Dietary wheat and related cereals as trigger of celiac disease Identification of gluten as toxic agent Villous atrophy and crypt hyperplasia as pathognomonic lesions of celiac disease Celiac disease of childhood and adult nontropical sprue share the same pathogenesis; use of the Watson capsule for biopsy Genetic association of celiac disease Association of dermatitis herpetiformis and celiac disease Untreated celiac disease as a predisposition for malignancy Association of selective IgA deficiency with celiac disease Gliadin triggers cytokine release in celiac disease intestinal biopsy specimens Close association of celiac disease with HLA-D Celiac disease–specific HLA-DQ2 is encoded in cis or trans Intraepithelial g/d T cells as characteristic feature of celiac disease Postulation of 5 phases of mucosal progression Isolation of gliadin-specific, HLADQ2–restricted T cells from the intestinal mucosa of celiac disease patients Autoimmune features of celiac disease IgA antireticulin IgA antiendomysium IgA anti–umbilical cord tTG As EMA autoantigen
Aretaeus from Cappadochia, 2nd century AD1 Gee,2 1888 Benecke,3 1910
and who were rechallenged with a peptic-tryptic gliadin digest (Frazer fraction III), Marsh17 was the first to suggest a sequence of progression of the celiac lesion. The initial event observed is an increase in intraepithelial lymphocyte count, followed by infiltration of the lamina propria with lymphocytes (stage 1). Crypt hyperplasia (stage 2) precedes villous atrophy (stage 3) and is only observed in the presence of lamina propria lymphocytosis, suggesting that mere intraepithelial lymphocytosis is not sufficient for intestinal transformation in celiac disease. Although clinical symptoms grossly correlate with the stage of mucosal transformation, patients with no or only few complaints can present with a stage 2 or 3 lesion. This may represent the limitations of localized proximal histology in a disease that can involve the whole small intestine.
Haas,4 1924–1932 Dicke,5
Genetic Association of Celiac Disease
1950
van de Kamer et al.,6 1953 Paulley,7 1954
Rubin et al.,8 1960
McDonald et al.,9 1965 Marks et al.,10 1966 Harris et al.,11 1967 Mawhinney and Tomkin,12 1971 Ferguson et al.,13 1975
Howell et al.,14 1986 Sollid et al.,15 1989 Spencer et al.,16 1989
Marsh,17 1992 Lundin et al.,18 1993
Seah et al.,19 1971 Chorzelski et al.,20 1983 Ladinser et al.,21 1994 Dieterich et al.,22 1997
NOTE. Table is based on a personal selection. Other important discoveries are discussed in the text.
235
Concordance for celiac disease in first-degree relatives ranges between 8% and 18% and reaches 70% in monozygotic twins. Based on family studies, McDonald et al.9 were the first to suggest an autosomal dominant inheritance of celiac disease with incomplete penetrance. Later Greenberg and Lange31 found evidence for 2 unlinked recessive celiac disease genes, 1 associated with an HLA-locus. HLA-DQ2 is found in 95% and the related HLA-DQ8 in most of the remaining patients with celiac disease.14,15 A further advance was the finding that the dimeric HLA-DQ2 is encoded either in cis or in trans, explaining its association with HLA-D3 or HLA-DR5/7, respectively (Figure 1).15 Early studies showed that gliadin elicits an inflammatory T-cell reaction when added to intestinal biopsy specimens of celiac patients in vitro,13 and a link to the genetic predisposition was provided by the isolation of gliadin-specific HLA-DQ2– restricted T-cell clones from celiac disease mucosa.18 However, the prevalence of HLA-DQ2 is high in the normal population (25%–30%), suggesting the involveTable 2. Toxic Cereals in Celiac Disease Cereal
Prolamine
Composition
Toxicity
Wheat Barley Rye Oats Maize Millet Rice
a-Gliadin Hordeins Secalins Avenins Zeins ? ?
36% Q, 17%–23% P 36% Q, 17%–23% P 36% Q, 17%–23% P High Q, low P Low Q, high A, L Low Q, high A, L Low Q, high A, L
111 11 11 1 – – –
NOTE. The major prolamines that drive the immune response in celiac disease are rich in glutamine and proline. A, alanine; L, leucine; P, proline; Q, glutamine. (Data from references 17, 23, 24, 26.)
236
DETLEF SCHUPPAN
ment of additional, probably non–HLA-linked genes in celiac disease pathogenesis. Recent studies found associations with chromosome 15q26, which harbors an insulindependent diabetes mellitus susceptibility locus,32 and with chromosome 5q and possibly 11q.33 How far these genes are ubiquitous or related to particular racial or geographic groups is the matter of current multinational investigations. Despite the close association of predisposing genes such as HLA-DQ2 and -DQ8 with silent or overt celiac disease, the amount of gluten consumed and, particularly, the time point of first gluten exposure play a prominent role in disease manifestation. Thus, the 5–10-fold higher incidence of clinically overt celiac disease in children from Sweden compared with Denmark, 2 populations with similar genetic and cultural backgrounds, was unexplained. More recent analyses found that in Sweden compared with Denmark, infant formulas contained 40 and 4 times more gliadin at the age of 8 and 12 months, respectively.34 This suggests that early exposure of the immature immune system to gliadin is a prominent cofactor for manifestation of clinically overt celiac disease, probably by skewing the immune system toward a T helper (Th) 1 T-cell response (see below). However, early patterns of gluten consumption or the duration of breast-feeding do not change the overall prevalence of celiac disease in the population, when also the oligosymptomatic or asymptomatic cases of adolescents and adults are taken into account (see below).35
Autoimmunity in Celiac Disease Gliadin and related cereal proteins are the undisputed triggers of celiac disease. Their elimination from the diet leads to histologic and clinical improvement in most patients. Active celiac disease is accompanied by mucosal (especially immunoglobulin [Ig] A) autoantibodies to reticulin, a common constituent of the extracellular matrix. Antireticulin autoantibodies are identical to antiendomysial, antijejunal or anti–umbilical cord antibodies.19–21 IgA antiendomysial autoantibodies (EMA) allow a screening for biopsy-proven celiac disease with an almost 100% sensitivity and specificity.36 This contrasts with serologic diagnosis in other, classical autoimmune diseases for which antibody tests show a much lower positive predictive value. Thus, a large field study in 17,200 north Italian school children used EMA positivity as a final screening tool before performing small intestinal biopsy. Biopsy confirmed previously undiagnosed villous atrophy in 75 of 98 EMA-positive children,37 yielding a prevalence of (clinically silent) celiac disease of 0.44% in Italy.
GASTROENTEROLOGY Vol. 119, No. 1
Another feature of celiac disease is the high prevalence of various autoimmune disorders, especially type I diabetes, dermatitis herpetiformis, autoimmune thyroiditis, collagen diseases, autoimmune alopecia, and autoimmune hepatitis. Ventura et al.38 screened 929 adolescents with celiac disease (mean age, 16.1 6 3.8 years). They found that 35% of patients in whom celiac disease was diagnosed at age 20 years or older, including many with clinically silent celiac disease, had associated autoimmune disorders, compared with only 5% who were diagnosed and properly treated with a gluten-free diet since infancy (age , 2 years). This indicates that, by unknown mechanisms, long-term undiagnosed and untreated celiac disease predisposes to autoimmunity to other organs.
Tissue Transglutaminase as Autoantigen Immunoprecipitation of proteins extracted from metabolically labeled fibrosarcoma cells with IgA from celiac disease patients led to the identification of tissue transglutaminase (tTG) as the prominent, if not sole, endomysial autoantigen.22 tTG is a calcium-dependent ubiquitous intracellular enzyme that belongs to a family with 3 epidermal and 2 extracellular transglutaminases (prostate transglutaminase and factor XIII).39,40 The transglutaminases catalyse the covalent and irreversible cross-linking of proteins resulting in the formation of an e-(g-glutamyl)-lysine (isopeptidyl) bond (Figure 2). tTG shows a high substrate specificity for an increasing spectrum of glutamyl-donor proteins (Table 3), whereas selectivity is lower for glutamyl acceptor proteins. The enzyme is normally stored intracellularly but can be released during cellular wounding brought about by mechanical stress, inflammation, infection, or during apoptosis. tTG cross-links several matrix proteins, stabi-
Figure 1. Celiac disease–associated HLA-DQ2 locus. More than 95% of patients with celiac disease are either heterozygous for DR3 or for DR5 and DR7, encoding HLA-DQ2 in cis or trans, respectively. (Data from Sollid et al.15)
July 2000
CELIAC DISEASE PATHOGENESIS
237
Figure 2. Protein cross-linking by tTG. tTG catalyzes a calciumdependent glutamyl-lysine transfer that results in irreversible protein cross-linking by an e-(gglutamyl)-lysine bond. Although glutamyl acceptors are numerous, the number of glutamyl donors is limited.
lizing the connective tissue scaffold on which the cells rest. However, when activated intracellularly, as in severe damage, the enzyme reacts with several structural and functional proteins setting the stage for apoptosis, preventing leakage of potentially harmful or antigenic proteins from the cell.40 Activated endothelia, fibroblasts, and mononuclear cells are particularly rich sources of tTG.
Gliadins as Preferred Substrates for tTG Gliadins that are glutamine- and proline-rich proteins (Table 2) are excellent glutamyl donor substrates for tTG, giving rise to gliadin-gliadin cross-links and even the covalent incorporation of tTG itself into highmolecular-weight complexes.22,41 These observations were already made several years ago, yet without having identified tTG as the celiac disease autoantigen.42,43 Furthermore, IgA antibodies of patients with celiac disease are also directed to cross-linked neoepitopes and their titer may occasionally exceed that of IgA anti-tTG antibodies.41 Because of numerous potential lysinecontaining acceptor proteins, with gliadin serving as glutamine-donor, tTG may generate additional antigenic neoepitopes, e.g., by cross-linking molecules of the extracellular matrix with gliadin or with tTG-gliadin complexes.
Deamidation by tTG Can Create More Potent T-Cell Epitopes Apart from cross-linking a variety of proteins, tTG can also deamidate glutamyl donor substrates, especially when no acceptor protein is available. This deamidation converts certain peptide-bound glutamine residues into a negatively charged glutamic acid. Microsequencing of peptides eluted from HLA-DQ2 has revealed so-called anchor amino acid residues that are required for optimal binding to the HLA, including a negative charge at positions 4, 6, and 9 of the nonapeptide recognition groove.44 Thus certain gliadin peptides that are modified by deamidation after in vitro incubation with tTG bind much better to the celiac-specific HLA-DQ2 than their nondeamidated counterparts, and can elicit a stronger proliferative response of gliadin specific T-cell clones (Figure 3).45,46
Gliadin-Reactive T Cells and Mucosal Destruction The isolation of T-cell clones from the intestinal mucosa of patients with celiac disease that can be stimulated with gliadin peptides presented in the context of HLA-DQ2 and -DQ8 supports a central role of gliadin
Table 3. Amine Acceptor Substrates for tTG Matrix proteins
Antiproteases
Intracellular proteins
Fibronectin Fibrillin SPARC (BM 40) Osteonectin Nidogen Collagen II Collagen V Procollagen III N propeptide
a1-Proteinase inhibitor a2-Plasminogen inhibitor PAI-2 Elafin (SKALP) CNS proteins b-Amyloid Huntingtin Tau-protein
Phospholipase A2 Lipocortin a-Actin rho-a bA3 crystallin pRB Importin a3 HSP23 eIF4F/elF5A histones 2A/B, 3, 4
Plasma proteins Fibrinogen Plasminogen IgG Cytokines Midkine IL-2 Gliadins
NOTE. tTG activity is highly restricted to a limited but expanding number of glutamine donor (amine acceptor) substrates, whereas many proteins can serve as amine donors (glutamine acceptors) using the e-amino group of a lysine residue (see also Figure 2). CNS, central nervous system; elF, elongation factor; HSP, heat shock protein; PAI, plasminogen-activator inhibitor; pRB, retinoblastoma gene product; SPARC, secreted protein acidic and rich in cysteine.
238
DETLEF SCHUPPAN
GASTROENTEROLOGY Vol. 119, No. 1
in vivo.49 In addition, activated lamina propria fibroblasts are the major source of keratinocyte growth factor, an important epithelial mitogen,50 which correlates well with the observed hyperplasia of crypt epithelial cells in celiac disease.
Autoantibodies Block tTG Bioactivity
Figure 3. Deamidation can enhance binding of gliadin peptides to HLA-DQ2. tTG can deamidate gliadin peptides that generate acidic, negatively charged glutamic acid residues from neutral glutamines. Because negatively charged residues are preferred in positions 4, 6, and 7 of the antigen-binding grove of HLA-DQ2,44 some of the deamidated gliadin variants may elicit stronger T-cell responses. The underlined E denominates a glutamic acid residue generated by tTG from a deamidated glutamine in vitro.45 H, hydrophobic pocket; 2, negatively charged region on HLA-DQ2. Amino acids are depicted in the one-letter code: D, aspartic acid; F, phenylalanine; I, isoleucine; L, leucine; M, methionine; P, proline, Q, glutamine; R, arginine; V, valine; Y, tyrosine; W, tryptophane.
in the initiation and maintenance of the celiac lesion. However, it does not explain the highly specific antibody response to tTG as the celiac disease autoantigen. A model was recently proposed based on the ability of B cells to present antigen in the context of HLA-DQ2.47 In this model, autoreactive B-cell clones that bind tTG or tTG-gliadin complexes via their B-cell receptors either present gliadin sequences or (less likely) transglutaminase sequences or cross-linked peptides to CD41 T cells via their HLA-DQ2 (Figure 4). The T cells thus stimulated by gliadin (by tTG or by cross-linked epitopes) will then secrete Th2 cytokines such as interleukin (IL)-4 to allow for expansion of the autoreactive B-cell clones and for subsequent autoantibody production. An important link between T-cell activation and mucosal transformation was provided by Pender et al.48 who used human fetal intestinal organ culture to show that T cells with Th1 features release tumor necrosis factor (TNF). TNF triggered intestinal fibroblasts to secrete matrix metalloproteinases (MMPs) that caused muscosal destruction by dissolution of connective tissue (Figure 5). In this model, either inhibition of TNF or of MMP-3, but not of interferon gamma, prevented mucosal damage. Because MMPs set the stage for mesenchymal migration and proliferation, their enhanced release does not contradict the observed mucosal transformation in active celiac disease, which finally leads to a 2–3-fold increase in the volume of the lamina propria. Accordingly, a recent study found a highly increased focal expression of MMP-1 and MMP-3 messenger RNA in fibroblastic cells of celiac disease small intestinal mucosa
The autoantibodies to tTG circulating in patients with celiac disease may have a biological role. This was suggested by an in vitro study that used T84 crypt epithelial cell differentiation in a fibroblast coculture model.51 This differentiation, which is dependent on transforming growth factor (TGF)-b,52 can be prevented either by addition of a blocking antibody to TGF-b or of IgA autoantibodies to tTG. tTG and plasmin are required to generate active TGF-b from the inactive (latent) TGF-b precursor,53 suggesting that local overproduction of the autoantibodies may contribute to the mucosal transformation observed in the active celiac lesion (Figure 6). This applies equally well to autoantibodies of the IgG class, particularly in IgA deficiency that is frequent in celiac disease,12 because IgG anti-tTG titers in celiac patients are high.22 Nonetheless, the lack of epithelial differentiation observed in celiac disease may rather result from villous atrophy that deprives the migrating crypt epithelia from important matrix contacts.
Figure 4. Antigen presentation and generation of IgA antibodies to gliadin and tTG. A B cell serves as an HLA-DQ–positive APC. Gliadin, deamidated gliadin, or a gliadin-tTG complex is swallowed (the monoclonal reactivity of different B-cell clones has been contracted to a single cell, to simplify the illustration), processed intracellularly, and presented to CD41 helper T cells (pathway a-1, b-1, a-2, b-2, or c-2). The T cells then provide help for production of antibodies to gliadin. Autoantibodies to tTG can be generated when spurious B-cell clones that produce antibodies to tTG present gliadin peptides via HLA-DQ to gliadin-specific CD41 T cells. The presented gliadin peptides may be generated from the processing of tTG-gliadin complexes, which serve as hapten-carrier complexes. These T cells then provide help for tTG-antibody production (pathway d-1). Alternatively, but still unproven, there may be T cells that recognize tTG or tTG-gliadin cross-links in the context of HLA-DQ, triggering production of the respective antibodies by their B-cell counterparts (pathway d-3 or e-4).
July 2000
CELIAC DISEASE PATHOGENESIS
239
Figure 5. Gliadin as trigger for antibody production and mucosal remodeling. Traces of dietary gliadin reach the lamina propria, especially when mucosal integrity is compromised, as in small intestinal infections or after chemical injury. In genetically predisposed (HLA-DQ2– or HLA-DQ8–positive) individuals, gliadin is presented on professional APCs such as B cells (see Figure 4) macrophages (MF) and dendritic cells (DC), which drive T-cell responses toward antibody production (Th2) or toward inflammation and tissue remodeling (Th1). The monoclonal reactivity of different T- and B-cell clones has been contracted to single cells, to simplify the illustration. Th1 cells release TNF, which causes release of MMPs from intestinal fibroblasts. MMP-3 appears to play a central role in tissue remodeling, because it degrades various noncollagenous matrix components and activates MMP-1, which degrades fibrillar collagens. In the inflammatory environment, mononuclear cells and fibroblasts are also the major source of tTG, which further enhances immune activation by gliadin deamidation or cross-linking. Tcyt, cytotoxic T cell.
g/d d T Cells as Mucosal Guardians g/d T cells are considered a hallmark of celiac disease. They populate mucosal surfaces, particularly in the airways and the gut, residing within the epithelial lining. Recent data have advanced our understanding of the role of g/d T cells (Figure 7).16,54–56 These primitive lymphocytes recognize bacterial nonpeptide antigens and unprocessed stress-related proteins. Two important stress-
Figure 6. Potential pathogenic role of tTG autoantibodies. The T84 crypt epithelial cell line spontaneously forms differentiated acinar structures in a fibroblast coculture model. Differentiation is inhibited by addition of blocking antibodies to TGF-b or of IgA purified from the serum of celiac patients. TGF-b can promote epithelial differentiation, and both tTG and plasmin are required to generate active TGF-b from the inactive (latent) TGF-b precursor.
induced proteins that are increased on intestinal epithelial cells by interferon gamma are MICA and MICB, which resemble major histocompatibility class I genes. MICA and MICB gene expression is regulated by promoter heat-shock elements similar to heat-shock protein (HSP) 70.56 The receptor (NKG2D) for MICA on natural killer and g/d T cells has recently been identified.57 Once activated, g/d T cells secrete chemokines that attract and stimulate cells of the unspecific (innate) immune response (monocytes/macrophages, neutrophils, and eosinophils). However, they modulate the antigen-specific immune response by secreting IL-4, which dampens the Th1 in favor of Th2 reactivity. Therefore, g/d T cells appear to protect the intestinal mucosa from chronic exposure to damaging agents such as dietary gluten in gluten-intolerant individuals, and their continuous presence in patients undergoing long-term gluten withdrawal may be caused by continuous inadvertent gluten ingestion.
Screening for Detection of Oligosymptomatic or Silent Celiac Disease Given the high positive predictive value of EMA testing by an experienced laboratory,36,37 the availability of an observer-independent test system based on guinea
240
DETLEF SCHUPPAN
GASTROENTEROLOGY Vol. 119, No. 1
Figure 7. Potential role of g/d T cells. Residing in the epithelium, g/d T cells are pathognomonic for celiac disease. They recognize bacterial nonpeptide antigens and stress-related proteins. Localized at the interface between the external and internal environment, g/d T cells serve as a link between the innate (unspecific) immune system involving monocytes/macrophages, neutrophils, and eosinophils, aimed at limiting the spread of infection, and the acquired immune system involving T and B cells, which target specific antigenic structures. g/d T cells secrete chemokines that activate the innate immune system and IL-4, which supports the maturation of Th2 cells.
pig tTG,58,59 and recently human recombinant tTG,60 makes population screening for clinically silent celiac disease possible. The usefulness of such an approach that might create unnecessary morbidity in previously healthy individuals has been questioned, but recent evidence is clearly in favor of identifying celiac disease patients with minor or no symptoms. Examples are the previously mentioned high prevalence of autoimmune disorders when celiac disease remains undetected for many years,38 the significant increase in bone mineral density in screening-detected asymptomatic adult celiac patients who were subject to a gluten-free diet for only 1 year,61 or the identification of celiac patients among patients with type I diabetes mellitus (who have a prevalence of celiac disease of 3%–7%), which can dramatically improve insulin therapy in previously brittle diabetics.62 Early detection may be particularly relevant in the United States where overt celiac disease is rare, but where screening studies using EMA have identified prevalences of at least 1:250 of EMA-positive subjects, comparable to the prevalences in Europe, which indicates a high proportion of subclinical disease.63
factors. Given the undisputable role of gliadin in driving inflammation and autoimmunity, celiac disease can serve as a model disease with autoimmune features for which, in contrast to most other autoimmune diseases, the trigger (gliadin), a close genetic association (with HLADQ2 or -DQ8), and a highly specific humoral autoimmune response (autoantibodies to tTG) are known. Yet, full manifestation of the celiac lesion appears to depend on additional factors, such as (1) bacterial or viral infection that can favor a destructive, Th1-like immune response and (2) the dose and earliest time point when the intestinal immune system is confronted with gluten dose, particularly in infants (Figure 8). Despite great advances in understanding the pathogenesis of sprue, many questions of immediate practical
Future Directions Celiac disease is now considered the result of a complex interplay of intrinsic (genetic) and extrinsic
Figure 8. Triggers and predisposing factors in celiac disease.
July 2000
significance remain. What are the other genes predisposing to celiac disease? Does tTG play a pathogenetic role beyond changing gluten antigenicity? Can the immune system be manipulated to generate tolerance to gluten? What is the relation of atopy and clinical manifestation of celiac disease in genetically predisposed individuals?64 Will screening for silent or subclinical celiac disease and institution of a gluten-free diet in asymptomatic patients prevent autoimmune or chronic malabsorptive disorders? These questions are currently the focus of many research groups and a large-scale European-based research effort.
CELIAC DISEASE PATHOGENESIS
19.
20.
21. 22.
References 1. Thomas C. On the coeliac affection. In: Major RH, ed. Classic descriptions of disease. Springfield, IL: Charles C. Thomas, 1945:600–601. 2. Gee S. On the coeliac affection. St Bartholomews Hosp Rep 1888;24:17–20. 3. Benecke R. Ueber die Spruekrankheit (Aphthae tropicae). Verh Dtsch Ges Pathol 1910;14:132. 4. Haas SV. Celiac disease, its specific treatment and cure without nutritional relapse. JAMA 1932;99:448–452. 5. Dicke WK. Investigation of the harmful effects of certain types of cereal on patients with coeliac disease [doctoral thesis]. University of Utrecht, the Netherlands, 1950. 6. Van de Kamer JH, Weyers HA, Dicke WK. Coeliac disease IV. An investigation into the injurious constituents of wheat in connection with their action on patients with coeliac disease. Acta Paediatr Scand 1953;42:223–231. 7. Paulley LW. Observations on the aetiology of idiopathic steatorrhoea. Br Med J 1954;2:1318–1321. 8. Rubin C, Brandborg LL, Phelps PC, Taylor HC. Studies of celiac disease. Part I. The apparent identical and specific nature of the duodenal and proximal jejunal lesion in celiac disease and idiopathic sprue. Gastroenterology 1960;38:28–38. 9. McDonald WC, Dobbins WO III, Rubin CE. Studies of the familial nature of celiac sprue using biopsy of the small intestine. N Engl J Med 1965;272:448–456. 10. Marks J, Schuster S, Watson AJ. Small bowel changes in dermatitis herpetiformis. Lancet 1966;2:1280–1282. 11. Harris OD, Cooke WT, Thompson JA, Waterhouse A. Malignancy in adult coeliac disease and idiopathic steatorrhea. Am J Med 1967;42:899–912. 12. Mawhinney H, Tomkin GH. Gluten enteropathy associated with selective IgA deficiency. Lancet 1971;2:121–124. 13. Ferguson A, MacDonald TT, McClure JP, Holden RJ. Cell-mediated immunity to gliadin within the small-intestinal mucosa in coeliac disease. Lancet 1975;1:895–897. 14. Howell MD, Austin RK, Kelleher D, Nepom GT, Kagnoff MF. An HLA-D region restriction length polymorphism associated with celiac disease. J Exp Med 1986;164:333–339. 15. Sollid L, Markussen G, Ek J, Gjerde H, Vartdal F, Thorsby E. Evidence for a primary association of celiac disease to a particular HLA-DQ-a/b heterodimer. J Exp Med 1989;169:345– 350. 16. Spencer J, Issacson PJ, Diss TC, MacDonald TT. Expression of disulfide-linked and non–disulfide-linked forms of the T cell receptor gamma/delta heterodimer in human intestinal intraepithelial lymphocytes. Eur J Immunol 1989;19:1335–1338. 17. Marsh MN. Gluten, major histocompatibility complex, and the small intestine: a molecular and immunobiologic approach to the spectrum of gluten sensitivity (‘celiac sprue’). Gastroenterology 1992;102:330–354. 18. Lundin KE, Scott H, Hansen T, Paulsen G, Halstensen TS, Fausa
23.
24. 25.
26.
27.
28.
29.
30.
31. 32.
33.
34.
35.
36.
241
O, Thorsby E, Sollid LM. Gliadin-specific, HLA-DQ (alpha*0501, beta*0201) restricted T cells isolated from the small intestinal mucosa of celiac disease patients. J Exp Med 1993;178:187– 196. Seah PP, Fry LL, Rossiter MA, Hoffbrand AV, Holborow EJ. Anti-reticulin antibodies in childhood celiac disease. Lancet 1971;2:681–682. Chorzelski TP, Sulej J, Tchorzewska H, Jablonska S, Beutner EH, Kumar V. IgA class endomysium antibodies in dermatitis herpetiformis and coeliac disease. Ann N Y Acad Sci 1983;420:325– 334. Ladinser B, Rossipal E, Pittschieler K. Endomysium antibodies in coeliac disease: an improved method. Gut 1994;35:776–778. Dieterich W, Ehnis T, Bauer M, Donner P, Volta U, Riecken EO, Schuppan D. Identification of tissue transglutaminase as the autoantigen of celiac disease. Nat Med 1997;3:797–801. Cornell H, Wieser H, Belitz HD. Characterization of the gliadinderived peptides which are biologically active in coeliac disease. Clin Chim Acta 1992;213:37–50. Wieser H. The precipitating factor in coeliac disease. Baillieres Clin Gastroenterol 1995;9:191–207. De Vincenzi M, Luchetti R, Peruffo AD, Curioni A, Pogna NE, Gasbarrini G. In vitro assessment of acetic acid-soluble proteins (glutenin) in celiac disease. J Biochem Toxicol 1996;11:205– 210. Field JM, Shewry PR, Miflin BJ. The purification and characterization of homologous high molecular weight storage proteins from grains of wheat, rye and barley. Theor Appl Genet 1982;62:329– 336. Janatuinen EK, Pikkarainen PH, Kemppainen TA, Kosma VM, Ja¨rvinen RM, Uusitupa MI, Julkunen RJ. A comparison of diets with and without oats in adults with celiac disease. N Engl J Med 1995;333:1033–1037. Ellis HJ, Rosen-Bronson S, O’Reilly N, Ciclitira PJ. Measurement of gluten using a monoclonal antibody to a coeliac toxic peptide of A-gliadin. Gut 1998;43:190–195. van Overbeek FM, Uil-Dieterman IG, Mol IW, Kohler-Brands L, Heymans HS, Mulder CJ. The daily gluten intake in relatives of patients with coeliac disease compared with that of the general Dutch population. Eur J Gastroenterol Hepatol 1997;9:1097– 1099. Faulkner-Hogg KB, Selby WS, Loblay RH. Dietary analysis in symptomatic patients with coeliac disease on a gluten-free diet: the role of trace amounts of gluten and non-gluten food intolerances. Scand J Gastroenterol 1999; 34:784–789. Greenberg DA, Lange KL. A maximum likelihood test of the two locus model for celiac disease. Am J Med Genet 1982;12:75–82. Houlston RS, Tomlinson IPM, Ford D, Seal S, Marossy AM, Ferguson A, Holmes GKT, Hosie KB, Howdle PD, Jewell DP, Godkin A, Kerr GD, Kumar P, Logan RFA, Love AHG, Johnston S, Marsh MN, Mitton S, O’Donoghue D, Roberts A, Walker-Smith JA, Stratton MF. Linkage analysis of candidate regions for celiac disease genes. Hum Mol Genet 1997;6:1335–1339. Greco L, Corazza G, Babron MC, Clot F, Fulchignoni-Lataud MC, Percopo S, Zavattari P, Bouguerra F, Dib C, Tosi R, Troncone R, Ventura A. Genome search in celiac disease. Am J Hum Genet 1998;62:669–675. Weile B, Cavell B, Nivenius K, Krasilnikoff PR. Striking difference in the incidence of childhood celiac disease between Denmark and Sweden: a plausible explanation. J Pediatr Gastroenterol 1995;21:64–68. Ascher H. The role of quantity and quality of gluten-containing cereals in the epidemiology of coeliac disease. In: Ma¨ki M, Collin P, Visakorpi JK, eds. Coeliac disease. Vammala, Finland: Vammalan Kirjapaino, 1997:15–22. Corrao G, Corazza GR, Andreani ML, Torchio P, Valentini RA, Galatola G, Quaglino D, Gasbarrini G, di Orio F. Serological
242
37.
38.
39.
40. 41.
42.
43.
44.
45.
46.
47. 48.
49.
50.
51.
52.
DETLEF SCHUPPAN
screening of coeliac disease: choosing the optimal procedure according to various prevalence values. Gut 1994;35:771–775. Catassi C, Ratsch IM, Fabiani E, Rossini M, Bordicchia F, Candela F, Coppa GV, Giorgi PL. Coeliac disease in the year 2000: exploring the iceberg. Lancet 1994;343:200–203. Ventura A, Magazzu G, Greco L, and the SIGEP study group. Duration of exposure to gluten and risk for autoimmune disorders in patients with celiac disease. Gastroenterology 1999;117:297– 303. Greenberg CS, Birkbichler PJ, Rice RH. Transglutaminases: multifunctional cross-linking enzymes that stabilize tissues. FASEB J 1991;5:3071–3077. Piacentini M, Colizzi V. Tissue transglutaminase: apoptosis versus autoimmunity. Immunol Today 1999;3:130–134. Schuppan D, Ehnis T, Bauer M, Volta A, Donner P, Riecken EO, Dieterich W. Identification of the autoantigen of coeliac disease. Ann N Y Acad Sci 1998;859:121–126. Bruce SE, Bjarnason I, Peters TJ. Human jejunal transglutaminase: demonstration of activity, enzyme kinetics and substrate specificity with special relation to gliadin and coeliac disease. Clin Sci 1985;68:S73–S79. Szabolcs M, Sipka S, Csorba S. In vitro cross-linking of gluten into high molecular weight polymers with transglutaminase. Acta Paediatr Hung 1987;28:215–227. Johannsen BH, Vartdal F, Eriksen JA, Thorsby E, Sollid LM. Identification of a putative motif for binding of peptides to HLA-DQ2. Int Immunol 1996;8:177–182. Molberg Ø, MacAdam SN, Korner R, Quarsten H, Christiansen C, Madsen L, Fugger L, Scott H, Noren O, Roepstorff P, Lundin KEA, Sjøstrom H, Sollid L. Tissue transglutaminase selectively modifies gliadin peptides that are recognized by gut-derived T cells in celiac disease. Nat Med 1998;4:713–717. Van de Wal Y, Kooy Y, van Veelen P, Pena S, Mearin L, Papadopoulos G, Koning F. Selective deamidation by tissue transglutaminase strongly enhances gliading-specific T cell reactivity. J Immunol 1998;161:1585–1588. Sollid LM. Autoantibodies in coeliac disease: tissue transglutaminase-guilt by association? Gut 1997;47:851–852. Pender SL, Tickle SP, Docherty AJ, Howie D, Wathen NC, McDonald TT. A major role of matrix metalloproteinases in T cell injury in the gut. J Immunol 1997;158:1582–1590. Daum S, Bauer U, Foss HD, Schuppan D, Stein H, Riecken EO, Ullrich R. Increased expression of mRNA for matrix metalloproteinase-1 and -3 and tissue inhibitor of metalloproteinases-1 in intestinal biopsy specimens from patients with coeliac disease. Gut 1999;44:17–25. Bajaj-Elliott M, Poulsom R, Pender SL, Wathen NC, McDonald TT. Interactions between stromal cell-derived keratinocyte growth factor and epithelial transforming growth factor in immunemediated crypt cell hyperplasia. J Clin Invest 1999;102:1473– 1480. Halttunen T, Ma¨ki M. Serum IgA from patients with celiac disease inhibits human T84 intestinal crypt epithelial cell differentiation. Gastroenterology 1999;116:566–572. Dignass AU, Podolsky DK. Cytokine modulation of intestinal
GASTROENTEROLOGY Vol. 119, No. 1
53.
54.
55. 56.
57.
58.
59.
60.
61.
62.
63.
64.
epithelial cell restitution: central role of transforming growth factor beta. Gastroenterology 1993;105:1323–1332. Nunes I, Gleizes PE, Metz CN, Rifkin DB. Latent transforming growth factor binding protein domains involved in activation and transglutaminase-dependent crosslinking of latent transforming growth factor-beta. J Cell Biol 1997;136:1151–1163. King DP, Hyde DM, Jackson KA, Novosad DM, Ellis TN, Putney L, Stovall MY, Van Winkle LS, Beaman BL, Ferrick DA. Protective response to pulmonary injury requires gamma delta T lymphocytes. J Immunol 1999;162:5033–5036. Mak TW, Ferrick DA. The g/dT-cell bridge: linking innate and acquired immunity. Nat Med 1998;7:764–765. Groh V, Steinle A, Bauer S, Spies T. Recognition of stress-induced MHC molecules by intestinal epithelial g/d T cells. Science 1998;279:1737–1740. Bauer S, Groh V, Wu J, Steinle A, Philipps JH, Lanier LL, Spies T. Activation of NK cells by NKG2D, a receptor for stress-induced MICA. Science 1999;285:727–729. Dieterich W, Laag E, Scho¨pper H, Volta U, Ferguson A, Gillet H, Riecken EO, Schuppan D. Autoantibodies to tissue transglutaminase as predictors of coeliac disease. Gastroenterology 1998; 115:1317–1321. Sulkanen S, Halttunen T, Laurila K, Kolho KL, Korponay-Szabo IR, Sarnesto A, Savilahti E, Collin P, Ma¨ki M. Tissue transglutaminase autoantibody enzyme-linked immunosorbent assay in detecting celiac disease. Gastroenterology 1998;115:1322–1328. Seissler J, Borns S, Wohlrab U, Morgenthaler NG, Mothes T, Boehm BO, Scherbaum WA. Antibodies to human tissue transglutaminase measured by radioligand assay: evidence for high diagnostic sensitivity for celiac disease. Horm Metab Res 1999; 31:375–379. Mustalahti K, Collin P, Sievanen H, Ma¨ki M. Osteopenia in patients with clinically silent celiac disease. Lancet 1999;354: 744–745. Kordonouri O, Dieterich W, Schuppan D, Weber G, Mu¨ller C, Sarioglu N, Becker M, Danne T. Autoantibodies to tissue transglutaminase are sensitive serological parameters for detecting silent celiac disease in patients with type 1 diabetes. Diabetes Med (in press). Not T, Horvath K, Hill ID, Partanen J, Mammed A, Magazzu G, Fasano A. Celiac disease in the USA: high prevalence of antiendomysium antibodies in healthy blood donors. Scand J Gastroenterol 1998;33:494–498. Erb K. Atopic disorders: a default pathway in the absence of infection? Immunol Today 1999;7:317–322.
Received December 7, 1999. Accepted February 23, 2000. Address requests for reprints to: Detlef Schuppan, M.D., Ph.D., First Department of Medicine, University of Erlangen-Nuernberg, Krankenhausstrasse 12, 91054 Erlangen, Germany. e-mail: [email protected]; fax: (49) 9131-8536003. Supported by Deutsche Forschungsgemeinschaft (grant Schu 646/11-1) and a grant from the German Celiac Foundation.
SPECIAL REPORTS AND REVIEWS Current Concepts of Celiac Disease Pathogenesis DETLEF SCHUPPAN First Department of Medicine, University of Erlangen-Nuernberg, Erlangen, Germany
Our knowledge of celiac disease pathogenesis has recently made rapid progress. The disorder is now considered the result of a complex interplay of intrinsic (genetic) and variable extrinsic (environmental) factors that explain the wide spectrum of clinical manifestations ranging from asymptomatic to severe malabsorption. Gluten peptides are efficiently presented by celiac disease–specific HLA-DQ2– and HLA-DQ8–positive antigen-presenting cells, and thus drive the immune response, predominantly in the connective tissue of the lamina propria. Tissue transglutaminase, which has been identified as the highly specific endomysial autoantigen, is released from cells during inflammation. It may potentiate antigen presentation by HLA-DQ2 and HLA-DQ8 by deamidating or cross-linking gluten peptides. The result is lamina propria T-cell activation and mucosal transformation by activated intestinal fibroblasts. In the future, manipulation of the gut-associated lymphoid tissue may allow reduced sensitivity or even generate oral tolerance to gluten. Long-standing untreated celiac disease, even if clinically silent, predisposes for other autoimmune diseases. Therefore, population screening for immunoglobulin A antibodies to tissue transglutaminase seems justified.
he realization of a close link between the ingestion of wheat protein and the clinical manifestation of celiac sprue (celiac disease), previously a severe malabsorptive disorder of childhood, can be considered a milestone of modern medicine. Table 1 summarizes the history of celiac disease and major advances in its understanding.1–22
T
Gluten as a Trigger of Celiac Disease Gluten is the protein fraction of wheat, rye, and barley that confers the properties of stickiness and thus allows the baking of bread. Gluten can be fractioned into the ethanol-soluble prolamines and ethanol-insoluble glutenins. Studies have been performed with more soluble prolamines, but recent data suggest that glutenins can also damage the intestinal mucosa.23–25 A common
feature of the prolamines of wheat is a high content of glutamine (.30%) and proline (.15%), whereas the nontoxic prolamines of rice and corn have a lower glutamine and proline content (Table 2).23–26 In this line, the toxicity of the prolamines of oats (avenins), which have an intermediate amino acid composition, is disputed, and only excessive ingestion of this cereal may be detrimental.27 The wheat prolamines are subdivided into a, b, g, and v gliadins, displaying molecular weights between 20,000 and 75,000 daltons, and containing similar or repetitive glutamine- and proline-rich peptide sequences such as Pro-Ser-Gln-Gln and Gln-Gln-Gln-Pro that appear to be responsible for the observed ‘‘toxicity’’ of gluten in celiac disease. This causal relation has been shown in several studies, by gluten challenge either of intestinal biopsy specimens in vitro or of the proximal and distal intestine in vivo.13,17 Glutamine- and proline-rich peptide sequences are also found in the genetically related prolamines of rye (secalins) and barley (hordeins). The highest allowable daily intake of gluten by celiac patients is a matter of debate. Small intestinal damage is clearly related to the amount of gluten ingested by susceptible individuals. A study from Australia showed that 15 of 22 patients with celiac disease who were still symptomatic while on a ‘‘codex-alimentarius gluten-free diet’’ (defined as ,0.3% gluten from grains) became asymptomatic or improved significantly when adhering to a more rigid ‘‘no detectable gluten diet.’’28 A daily gluten consumption of ,50 mg, compared with an average of about 13 g in most Western countries,29 is considered safe by many experts. However, gluten sensitivity differs among individuals, and there are other factors that trigger or maintain the disease (see below). Matters are further complicated because, despite the recent development of assays for the measurement of toxic Abbreviations used in this paper: APC, antigen-presenting cell; EMA, endomysial autoantibody(ies); HSP, heat-shock protein; MMP, matrix metalloproteinase; TGF, transforming growth factor; TNF, tumor necrosis factor; tTG, tissue transglutaminase. r 2000 by the American Gastroenterological Association 0016-5085/00/$10.00 doi:10.1053/gast.2000.8521
July 2000
CELIAC DISEASE PATHOGENESIS
(a)-gliadins in cereals,30 a precise quantification of potentially toxic cereal proteins that may be present in foods is presently impossible.
Histopathologic Characteristics of the Celiac Lesion On the basis of histologic follow-up of celiac patients who were in remission after a gluten-free diet Table 1. Historical Milestones in Celiac Disease Pathogenesis First report on a condition resembling celiac disease First detailed clinical description of the ‘‘coeliac affection’’ First description of the mucosal transformation characteristic of celiac disease Successful diet consisting of fruit, vegetables, and milk powder Dietary wheat and related cereals as trigger of celiac disease Identification of gluten as toxic agent Villous atrophy and crypt hyperplasia as pathognomonic lesions of celiac disease Celiac disease of childhood and adult nontropical sprue share the same pathogenesis; use of the Watson capsule for biopsy Genetic association of celiac disease Association of dermatitis herpetiformis and celiac disease Untreated celiac disease as a predisposition for malignancy Association of selective IgA deficiency with celiac disease Gliadin triggers cytokine release in celiac disease intestinal biopsy specimens Close association of celiac disease with HLA-D Celiac disease–specific HLA-DQ2 is encoded in cis or trans Intraepithelial g/d T cells as characteristic feature of celiac disease Postulation of 5 phases of mucosal progression Isolation of gliadin-specific, HLADQ2–restricted T cells from the intestinal mucosa of celiac disease patients Autoimmune features of celiac disease IgA antireticulin IgA antiendomysium IgA anti–umbilical cord tTG As EMA autoantigen
Aretaeus from Cappadochia, 2nd century AD1 Gee,2 1888 Benecke,3 1910
and who were rechallenged with a peptic-tryptic gliadin digest (Frazer fraction III), Marsh17 was the first to suggest a sequence of progression of the celiac lesion. The initial event observed is an increase in intraepithelial lymphocyte count, followed by infiltration of the lamina propria with lymphocytes (stage 1). Crypt hyperplasia (stage 2) precedes villous atrophy (stage 3) and is only observed in the presence of lamina propria lymphocytosis, suggesting that mere intraepithelial lymphocytosis is not sufficient for intestinal transformation in celiac disease. Although clinical symptoms grossly correlate with the stage of mucosal transformation, patients with no or only few complaints can present with a stage 2 or 3 lesion. This may represent the limitations of localized proximal histology in a disease that can involve the whole small intestine.
Haas,4 1924–1932 Dicke,5
Genetic Association of Celiac Disease
1950
van de Kamer et al.,6 1953 Paulley,7 1954
Rubin et al.,8 1960
McDonald et al.,9 1965 Marks et al.,10 1966 Harris et al.,11 1967 Mawhinney and Tomkin,12 1971 Ferguson et al.,13 1975
Howell et al.,14 1986 Sollid et al.,15 1989 Spencer et al.,16 1989
Marsh,17 1992 Lundin et al.,18 1993
Seah et al.,19 1971 Chorzelski et al.,20 1983 Ladinser et al.,21 1994 Dieterich et al.,22 1997
NOTE. Table is based on a personal selection. Other important discoveries are discussed in the text.
235
Concordance for celiac disease in first-degree relatives ranges between 8% and 18% and reaches 70% in monozygotic twins. Based on family studies, McDonald et al.9 were the first to suggest an autosomal dominant inheritance of celiac disease with incomplete penetrance. Later Greenberg and Lange31 found evidence for 2 unlinked recessive celiac disease genes, 1 associated with an HLA-locus. HLA-DQ2 is found in 95% and the related HLA-DQ8 in most of the remaining patients with celiac disease.14,15 A further advance was the finding that the dimeric HLA-DQ2 is encoded either in cis or in trans, explaining its association with HLA-D3 or HLA-DR5/7, respectively (Figure 1).15 Early studies showed that gliadin elicits an inflammatory T-cell reaction when added to intestinal biopsy specimens of celiac patients in vitro,13 and a link to the genetic predisposition was provided by the isolation of gliadin-specific HLA-DQ2– restricted T-cell clones from celiac disease mucosa.18 However, the prevalence of HLA-DQ2 is high in the normal population (25%–30%), suggesting the involveTable 2. Toxic Cereals in Celiac Disease Cereal
Prolamine
Composition
Toxicity
Wheat Barley Rye Oats Maize Millet Rice
a-Gliadin Hordeins Secalins Avenins Zeins ? ?
36% Q, 17%–23% P 36% Q, 17%–23% P 36% Q, 17%–23% P High Q, low P Low Q, high A, L Low Q, high A, L Low Q, high A, L
111 11 11 1 – – –
NOTE. The major prolamines that drive the immune response in celiac disease are rich in glutamine and proline. A, alanine; L, leucine; P, proline; Q, glutamine. (Data from references 17, 23, 24, 26.)
236
DETLEF SCHUPPAN
ment of additional, probably non–HLA-linked genes in celiac disease pathogenesis. Recent studies found associations with chromosome 15q26, which harbors an insulindependent diabetes mellitus susceptibility locus,32 and with chromosome 5q and possibly 11q.33 How far these genes are ubiquitous or related to particular racial or geographic groups is the matter of current multinational investigations. Despite the close association of predisposing genes such as HLA-DQ2 and -DQ8 with silent or overt celiac disease, the amount of gluten consumed and, particularly, the time point of first gluten exposure play a prominent role in disease manifestation. Thus, the 5–10-fold higher incidence of clinically overt celiac disease in children from Sweden compared with Denmark, 2 populations with similar genetic and cultural backgrounds, was unexplained. More recent analyses found that in Sweden compared with Denmark, infant formulas contained 40 and 4 times more gliadin at the age of 8 and 12 months, respectively.34 This suggests that early exposure of the immature immune system to gliadin is a prominent cofactor for manifestation of clinically overt celiac disease, probably by skewing the immune system toward a T helper (Th) 1 T-cell response (see below). However, early patterns of gluten consumption or the duration of breast-feeding do not change the overall prevalence of celiac disease in the population, when also the oligosymptomatic or asymptomatic cases of adolescents and adults are taken into account (see below).35
Autoimmunity in Celiac Disease Gliadin and related cereal proteins are the undisputed triggers of celiac disease. Their elimination from the diet leads to histologic and clinical improvement in most patients. Active celiac disease is accompanied by mucosal (especially immunoglobulin [Ig] A) autoantibodies to reticulin, a common constituent of the extracellular matrix. Antireticulin autoantibodies are identical to antiendomysial, antijejunal or anti–umbilical cord antibodies.19–21 IgA antiendomysial autoantibodies (EMA) allow a screening for biopsy-proven celiac disease with an almost 100% sensitivity and specificity.36 This contrasts with serologic diagnosis in other, classical autoimmune diseases for which antibody tests show a much lower positive predictive value. Thus, a large field study in 17,200 north Italian school children used EMA positivity as a final screening tool before performing small intestinal biopsy. Biopsy confirmed previously undiagnosed villous atrophy in 75 of 98 EMA-positive children,37 yielding a prevalence of (clinically silent) celiac disease of 0.44% in Italy.
GASTROENTEROLOGY Vol. 119, No. 1
Another feature of celiac disease is the high prevalence of various autoimmune disorders, especially type I diabetes, dermatitis herpetiformis, autoimmune thyroiditis, collagen diseases, autoimmune alopecia, and autoimmune hepatitis. Ventura et al.38 screened 929 adolescents with celiac disease (mean age, 16.1 6 3.8 years). They found that 35% of patients in whom celiac disease was diagnosed at age 20 years or older, including many with clinically silent celiac disease, had associated autoimmune disorders, compared with only 5% who were diagnosed and properly treated with a gluten-free diet since infancy (age , 2 years). This indicates that, by unknown mechanisms, long-term undiagnosed and untreated celiac disease predisposes to autoimmunity to other organs.
Tissue Transglutaminase as Autoantigen Immunoprecipitation of proteins extracted from metabolically labeled fibrosarcoma cells with IgA from celiac disease patients led to the identification of tissue transglutaminase (tTG) as the prominent, if not sole, endomysial autoantigen.22 tTG is a calcium-dependent ubiquitous intracellular enzyme that belongs to a family with 3 epidermal and 2 extracellular transglutaminases (prostate transglutaminase and factor XIII).39,40 The transglutaminases catalyse the covalent and irreversible cross-linking of proteins resulting in the formation of an e-(g-glutamyl)-lysine (isopeptidyl) bond (Figure 2). tTG shows a high substrate specificity for an increasing spectrum of glutamyl-donor proteins (Table 3), whereas selectivity is lower for glutamyl acceptor proteins. The enzyme is normally stored intracellularly but can be released during cellular wounding brought about by mechanical stress, inflammation, infection, or during apoptosis. tTG cross-links several matrix proteins, stabi-
Figure 1. Celiac disease–associated HLA-DQ2 locus. More than 95% of patients with celiac disease are either heterozygous for DR3 or for DR5 and DR7, encoding HLA-DQ2 in cis or trans, respectively. (Data from Sollid et al.15)
July 2000
CELIAC DISEASE PATHOGENESIS
237
Figure 2. Protein cross-linking by tTG. tTG catalyzes a calciumdependent glutamyl-lysine transfer that results in irreversible protein cross-linking by an e-(gglutamyl)-lysine bond. Although glutamyl acceptors are numerous, the number of glutamyl donors is limited.
lizing the connective tissue scaffold on which the cells rest. However, when activated intracellularly, as in severe damage, the enzyme reacts with several structural and functional proteins setting the stage for apoptosis, preventing leakage of potentially harmful or antigenic proteins from the cell.40 Activated endothelia, fibroblasts, and mononuclear cells are particularly rich sources of tTG.
Gliadins as Preferred Substrates for tTG Gliadins that are glutamine- and proline-rich proteins (Table 2) are excellent glutamyl donor substrates for tTG, giving rise to gliadin-gliadin cross-links and even the covalent incorporation of tTG itself into highmolecular-weight complexes.22,41 These observations were already made several years ago, yet without having identified tTG as the celiac disease autoantigen.42,43 Furthermore, IgA antibodies of patients with celiac disease are also directed to cross-linked neoepitopes and their titer may occasionally exceed that of IgA anti-tTG antibodies.41 Because of numerous potential lysinecontaining acceptor proteins, with gliadin serving as glutamine-donor, tTG may generate additional antigenic neoepitopes, e.g., by cross-linking molecules of the extracellular matrix with gliadin or with tTG-gliadin complexes.
Deamidation by tTG Can Create More Potent T-Cell Epitopes Apart from cross-linking a variety of proteins, tTG can also deamidate glutamyl donor substrates, especially when no acceptor protein is available. This deamidation converts certain peptide-bound glutamine residues into a negatively charged glutamic acid. Microsequencing of peptides eluted from HLA-DQ2 has revealed so-called anchor amino acid residues that are required for optimal binding to the HLA, including a negative charge at positions 4, 6, and 9 of the nonapeptide recognition groove.44 Thus certain gliadin peptides that are modified by deamidation after in vitro incubation with tTG bind much better to the celiac-specific HLA-DQ2 than their nondeamidated counterparts, and can elicit a stronger proliferative response of gliadin specific T-cell clones (Figure 3).45,46
Gliadin-Reactive T Cells and Mucosal Destruction The isolation of T-cell clones from the intestinal mucosa of patients with celiac disease that can be stimulated with gliadin peptides presented in the context of HLA-DQ2 and -DQ8 supports a central role of gliadin
Table 3. Amine Acceptor Substrates for tTG Matrix proteins
Antiproteases
Intracellular proteins
Fibronectin Fibrillin SPARC (BM 40) Osteonectin Nidogen Collagen II Collagen V Procollagen III N propeptide
a1-Proteinase inhibitor a2-Plasminogen inhibitor PAI-2 Elafin (SKALP) CNS proteins b-Amyloid Huntingtin Tau-protein
Phospholipase A2 Lipocortin a-Actin rho-a bA3 crystallin pRB Importin a3 HSP23 eIF4F/elF5A histones 2A/B, 3, 4
Plasma proteins Fibrinogen Plasminogen IgG Cytokines Midkine IL-2 Gliadins
NOTE. tTG activity is highly restricted to a limited but expanding number of glutamine donor (amine acceptor) substrates, whereas many proteins can serve as amine donors (glutamine acceptors) using the e-amino group of a lysine residue (see also Figure 2). CNS, central nervous system; elF, elongation factor; HSP, heat shock protein; PAI, plasminogen-activator inhibitor; pRB, retinoblastoma gene product; SPARC, secreted protein acidic and rich in cysteine.
238
DETLEF SCHUPPAN
GASTROENTEROLOGY Vol. 119, No. 1
in vivo.49 In addition, activated lamina propria fibroblasts are the major source of keratinocyte growth factor, an important epithelial mitogen,50 which correlates well with the observed hyperplasia of crypt epithelial cells in celiac disease.
Autoantibodies Block tTG Bioactivity
Figure 3. Deamidation can enhance binding of gliadin peptides to HLA-DQ2. tTG can deamidate gliadin peptides that generate acidic, negatively charged glutamic acid residues from neutral glutamines. Because negatively charged residues are preferred in positions 4, 6, and 7 of the antigen-binding grove of HLA-DQ2,44 some of the deamidated gliadin variants may elicit stronger T-cell responses. The underlined E denominates a glutamic acid residue generated by tTG from a deamidated glutamine in vitro.45 H, hydrophobic pocket; 2, negatively charged region on HLA-DQ2. Amino acids are depicted in the one-letter code: D, aspartic acid; F, phenylalanine; I, isoleucine; L, leucine; M, methionine; P, proline, Q, glutamine; R, arginine; V, valine; Y, tyrosine; W, tryptophane.
in the initiation and maintenance of the celiac lesion. However, it does not explain the highly specific antibody response to tTG as the celiac disease autoantigen. A model was recently proposed based on the ability of B cells to present antigen in the context of HLA-DQ2.47 In this model, autoreactive B-cell clones that bind tTG or tTG-gliadin complexes via their B-cell receptors either present gliadin sequences or (less likely) transglutaminase sequences or cross-linked peptides to CD41 T cells via their HLA-DQ2 (Figure 4). The T cells thus stimulated by gliadin (by tTG or by cross-linked epitopes) will then secrete Th2 cytokines such as interleukin (IL)-4 to allow for expansion of the autoreactive B-cell clones and for subsequent autoantibody production. An important link between T-cell activation and mucosal transformation was provided by Pender et al.48 who used human fetal intestinal organ culture to show that T cells with Th1 features release tumor necrosis factor (TNF). TNF triggered intestinal fibroblasts to secrete matrix metalloproteinases (MMPs) that caused muscosal destruction by dissolution of connective tissue (Figure 5). In this model, either inhibition of TNF or of MMP-3, but not of interferon gamma, prevented mucosal damage. Because MMPs set the stage for mesenchymal migration and proliferation, their enhanced release does not contradict the observed mucosal transformation in active celiac disease, which finally leads to a 2–3-fold increase in the volume of the lamina propria. Accordingly, a recent study found a highly increased focal expression of MMP-1 and MMP-3 messenger RNA in fibroblastic cells of celiac disease small intestinal mucosa
The autoantibodies to tTG circulating in patients with celiac disease may have a biological role. This was suggested by an in vitro study that used T84 crypt epithelial cell differentiation in a fibroblast coculture model.51 This differentiation, which is dependent on transforming growth factor (TGF)-b,52 can be prevented either by addition of a blocking antibody to TGF-b or of IgA autoantibodies to tTG. tTG and plasmin are required to generate active TGF-b from the inactive (latent) TGF-b precursor,53 suggesting that local overproduction of the autoantibodies may contribute to the mucosal transformation observed in the active celiac lesion (Figure 6). This applies equally well to autoantibodies of the IgG class, particularly in IgA deficiency that is frequent in celiac disease,12 because IgG anti-tTG titers in celiac patients are high.22 Nonetheless, the lack of epithelial differentiation observed in celiac disease may rather result from villous atrophy that deprives the migrating crypt epithelia from important matrix contacts.
Figure 4. Antigen presentation and generation of IgA antibodies to gliadin and tTG. A B cell serves as an HLA-DQ–positive APC. Gliadin, deamidated gliadin, or a gliadin-tTG complex is swallowed (the monoclonal reactivity of different B-cell clones has been contracted to a single cell, to simplify the illustration), processed intracellularly, and presented to CD41 helper T cells (pathway a-1, b-1, a-2, b-2, or c-2). The T cells then provide help for production of antibodies to gliadin. Autoantibodies to tTG can be generated when spurious B-cell clones that produce antibodies to tTG present gliadin peptides via HLA-DQ to gliadin-specific CD41 T cells. The presented gliadin peptides may be generated from the processing of tTG-gliadin complexes, which serve as hapten-carrier complexes. These T cells then provide help for tTG-antibody production (pathway d-1). Alternatively, but still unproven, there may be T cells that recognize tTG or tTG-gliadin cross-links in the context of HLA-DQ, triggering production of the respective antibodies by their B-cell counterparts (pathway d-3 or e-4).
July 2000
CELIAC DISEASE PATHOGENESIS
239
Figure 5. Gliadin as trigger for antibody production and mucosal remodeling. Traces of dietary gliadin reach the lamina propria, especially when mucosal integrity is compromised, as in small intestinal infections or after chemical injury. In genetically predisposed (HLA-DQ2– or HLA-DQ8–positive) individuals, gliadin is presented on professional APCs such as B cells (see Figure 4) macrophages (MF) and dendritic cells (DC), which drive T-cell responses toward antibody production (Th2) or toward inflammation and tissue remodeling (Th1). The monoclonal reactivity of different T- and B-cell clones has been contracted to single cells, to simplify the illustration. Th1 cells release TNF, which causes release of MMPs from intestinal fibroblasts. MMP-3 appears to play a central role in tissue remodeling, because it degrades various noncollagenous matrix components and activates MMP-1, which degrades fibrillar collagens. In the inflammatory environment, mononuclear cells and fibroblasts are also the major source of tTG, which further enhances immune activation by gliadin deamidation or cross-linking. Tcyt, cytotoxic T cell.
g/d d T Cells as Mucosal Guardians g/d T cells are considered a hallmark of celiac disease. They populate mucosal surfaces, particularly in the airways and the gut, residing within the epithelial lining. Recent data have advanced our understanding of the role of g/d T cells (Figure 7).16,54–56 These primitive lymphocytes recognize bacterial nonpeptide antigens and unprocessed stress-related proteins. Two important stress-
Figure 6. Potential pathogenic role of tTG autoantibodies. The T84 crypt epithelial cell line spontaneously forms differentiated acinar structures in a fibroblast coculture model. Differentiation is inhibited by addition of blocking antibodies to TGF-b or of IgA purified from the serum of celiac patients. TGF-b can promote epithelial differentiation, and both tTG and plasmin are required to generate active TGF-b from the inactive (latent) TGF-b precursor.
induced proteins that are increased on intestinal epithelial cells by interferon gamma are MICA and MICB, which resemble major histocompatibility class I genes. MICA and MICB gene expression is regulated by promoter heat-shock elements similar to heat-shock protein (HSP) 70.56 The receptor (NKG2D) for MICA on natural killer and g/d T cells has recently been identified.57 Once activated, g/d T cells secrete chemokines that attract and stimulate cells of the unspecific (innate) immune response (monocytes/macrophages, neutrophils, and eosinophils). However, they modulate the antigen-specific immune response by secreting IL-4, which dampens the Th1 in favor of Th2 reactivity. Therefore, g/d T cells appear to protect the intestinal mucosa from chronic exposure to damaging agents such as dietary gluten in gluten-intolerant individuals, and their continuous presence in patients undergoing long-term gluten withdrawal may be caused by continuous inadvertent gluten ingestion.
Screening for Detection of Oligosymptomatic or Silent Celiac Disease Given the high positive predictive value of EMA testing by an experienced laboratory,36,37 the availability of an observer-independent test system based on guinea
240
DETLEF SCHUPPAN
GASTROENTEROLOGY Vol. 119, No. 1
Figure 7. Potential role of g/d T cells. Residing in the epithelium, g/d T cells are pathognomonic for celiac disease. They recognize bacterial nonpeptide antigens and stress-related proteins. Localized at the interface between the external and internal environment, g/d T cells serve as a link between the innate (unspecific) immune system involving monocytes/macrophages, neutrophils, and eosinophils, aimed at limiting the spread of infection, and the acquired immune system involving T and B cells, which target specific antigenic structures. g/d T cells secrete chemokines that activate the innate immune system and IL-4, which supports the maturation of Th2 cells.
pig tTG,58,59 and recently human recombinant tTG,60 makes population screening for clinically silent celiac disease possible. The usefulness of such an approach that might create unnecessary morbidity in previously healthy individuals has been questioned, but recent evidence is clearly in favor of identifying celiac disease patients with minor or no symptoms. Examples are the previously mentioned high prevalence of autoimmune disorders when celiac disease remains undetected for many years,38 the significant increase in bone mineral density in screening-detected asymptomatic adult celiac patients who were subject to a gluten-free diet for only 1 year,61 or the identification of celiac patients among patients with type I diabetes mellitus (who have a prevalence of celiac disease of 3%–7%), which can dramatically improve insulin therapy in previously brittle diabetics.62 Early detection may be particularly relevant in the United States where overt celiac disease is rare, but where screening studies using EMA have identified prevalences of at least 1:250 of EMA-positive subjects, comparable to the prevalences in Europe, which indicates a high proportion of subclinical disease.63
factors. Given the undisputable role of gliadin in driving inflammation and autoimmunity, celiac disease can serve as a model disease with autoimmune features for which, in contrast to most other autoimmune diseases, the trigger (gliadin), a close genetic association (with HLADQ2 or -DQ8), and a highly specific humoral autoimmune response (autoantibodies to tTG) are known. Yet, full manifestation of the celiac lesion appears to depend on additional factors, such as (1) bacterial or viral infection that can favor a destructive, Th1-like immune response and (2) the dose and earliest time point when the intestinal immune system is confronted with gluten dose, particularly in infants (Figure 8). Despite great advances in understanding the pathogenesis of sprue, many questions of immediate practical
Future Directions Celiac disease is now considered the result of a complex interplay of intrinsic (genetic) and extrinsic
Figure 8. Triggers and predisposing factors in celiac disease.
July 2000
significance remain. What are the other genes predisposing to celiac disease? Does tTG play a pathogenetic role beyond changing gluten antigenicity? Can the immune system be manipulated to generate tolerance to gluten? What is the relation of atopy and clinical manifestation of celiac disease in genetically predisposed individuals?64 Will screening for silent or subclinical celiac disease and institution of a gluten-free diet in asymptomatic patients prevent autoimmune or chronic malabsorptive disorders? These questions are currently the focus of many research groups and a large-scale European-based research effort.
CELIAC DISEASE PATHOGENESIS
19.
20.
21. 22.
References 1. Thomas C. On the coeliac affection. In: Major RH, ed. Classic descriptions of disease. Springfield, IL: Charles C. Thomas, 1945:600–601. 2. Gee S. On the coeliac affection. St Bartholomews Hosp Rep 1888;24:17–20. 3. Benecke R. Ueber die Spruekrankheit (Aphthae tropicae). Verh Dtsch Ges Pathol 1910;14:132. 4. Haas SV. Celiac disease, its specific treatment and cure without nutritional relapse. JAMA 1932;99:448–452. 5. Dicke WK. Investigation of the harmful effects of certain types of cereal on patients with coeliac disease [doctoral thesis]. University of Utrecht, the Netherlands, 1950. 6. Van de Kamer JH, Weyers HA, Dicke WK. Coeliac disease IV. An investigation into the injurious constituents of wheat in connection with their action on patients with coeliac disease. Acta Paediatr Scand 1953;42:223–231. 7. Paulley LW. Observations on the aetiology of idiopathic steatorrhoea. Br Med J 1954;2:1318–1321. 8. Rubin C, Brandborg LL, Phelps PC, Taylor HC. Studies of celiac disease. Part I. The apparent identical and specific nature of the duodenal and proximal jejunal lesion in celiac disease and idiopathic sprue. Gastroenterology 1960;38:28–38. 9. McDonald WC, Dobbins WO III, Rubin CE. Studies of the familial nature of celiac sprue using biopsy of the small intestine. N Engl J Med 1965;272:448–456. 10. Marks J, Schuster S, Watson AJ. Small bowel changes in dermatitis herpetiformis. Lancet 1966;2:1280–1282. 11. Harris OD, Cooke WT, Thompson JA, Waterhouse A. Malignancy in adult coeliac disease and idiopathic steatorrhea. Am J Med 1967;42:899–912. 12. Mawhinney H, Tomkin GH. Gluten enteropathy associated with selective IgA deficiency. Lancet 1971;2:121–124. 13. Ferguson A, MacDonald TT, McClure JP, Holden RJ. Cell-mediated immunity to gliadin within the small-intestinal mucosa in coeliac disease. Lancet 1975;1:895–897. 14. Howell MD, Austin RK, Kelleher D, Nepom GT, Kagnoff MF. An HLA-D region restriction length polymorphism associated with celiac disease. J Exp Med 1986;164:333–339. 15. Sollid L, Markussen G, Ek J, Gjerde H, Vartdal F, Thorsby E. Evidence for a primary association of celiac disease to a particular HLA-DQ-a/b heterodimer. J Exp Med 1989;169:345– 350. 16. Spencer J, Issacson PJ, Diss TC, MacDonald TT. Expression of disulfide-linked and non–disulfide-linked forms of the T cell receptor gamma/delta heterodimer in human intestinal intraepithelial lymphocytes. Eur J Immunol 1989;19:1335–1338. 17. Marsh MN. Gluten, major histocompatibility complex, and the small intestine: a molecular and immunobiologic approach to the spectrum of gluten sensitivity (‘celiac sprue’). Gastroenterology 1992;102:330–354. 18. Lundin KE, Scott H, Hansen T, Paulsen G, Halstensen TS, Fausa
23.
24. 25.
26.
27.
28.
29.
30.
31. 32.
33.
34.
35.
36.
241
O, Thorsby E, Sollid LM. Gliadin-specific, HLA-DQ (alpha*0501, beta*0201) restricted T cells isolated from the small intestinal mucosa of celiac disease patients. J Exp Med 1993;178:187– 196. Seah PP, Fry LL, Rossiter MA, Hoffbrand AV, Holborow EJ. Anti-reticulin antibodies in childhood celiac disease. Lancet 1971;2:681–682. Chorzelski TP, Sulej J, Tchorzewska H, Jablonska S, Beutner EH, Kumar V. IgA class endomysium antibodies in dermatitis herpetiformis and coeliac disease. Ann N Y Acad Sci 1983;420:325– 334. Ladinser B, Rossipal E, Pittschieler K. Endomysium antibodies in coeliac disease: an improved method. Gut 1994;35:776–778. Dieterich W, Ehnis T, Bauer M, Donner P, Volta U, Riecken EO, Schuppan D. Identification of tissue transglutaminase as the autoantigen of celiac disease. Nat Med 1997;3:797–801. Cornell H, Wieser H, Belitz HD. Characterization of the gliadinderived peptides which are biologically active in coeliac disease. Clin Chim Acta 1992;213:37–50. Wieser H. The precipitating factor in coeliac disease. Baillieres Clin Gastroenterol 1995;9:191–207. De Vincenzi M, Luchetti R, Peruffo AD, Curioni A, Pogna NE, Gasbarrini G. In vitro assessment of acetic acid-soluble proteins (glutenin) in celiac disease. J Biochem Toxicol 1996;11:205– 210. Field JM, Shewry PR, Miflin BJ. The purification and characterization of homologous high molecular weight storage proteins from grains of wheat, rye and barley. Theor Appl Genet 1982;62:329– 336. Janatuinen EK, Pikkarainen PH, Kemppainen TA, Kosma VM, Ja¨rvinen RM, Uusitupa MI, Julkunen RJ. A comparison of diets with and without oats in adults with celiac disease. N Engl J Med 1995;333:1033–1037. Ellis HJ, Rosen-Bronson S, O’Reilly N, Ciclitira PJ. Measurement of gluten using a monoclonal antibody to a coeliac toxic peptide of A-gliadin. Gut 1998;43:190–195. van Overbeek FM, Uil-Dieterman IG, Mol IW, Kohler-Brands L, Heymans HS, Mulder CJ. The daily gluten intake in relatives of patients with coeliac disease compared with that of the general Dutch population. Eur J Gastroenterol Hepatol 1997;9:1097– 1099. Faulkner-Hogg KB, Selby WS, Loblay RH. Dietary analysis in symptomatic patients with coeliac disease on a gluten-free diet: the role of trace amounts of gluten and non-gluten food intolerances. Scand J Gastroenterol 1999; 34:784–789. Greenberg DA, Lange KL. A maximum likelihood test of the two locus model for celiac disease. Am J Med Genet 1982;12:75–82. Houlston RS, Tomlinson IPM, Ford D, Seal S, Marossy AM, Ferguson A, Holmes GKT, Hosie KB, Howdle PD, Jewell DP, Godkin A, Kerr GD, Kumar P, Logan RFA, Love AHG, Johnston S, Marsh MN, Mitton S, O’Donoghue D, Roberts A, Walker-Smith JA, Stratton MF. Linkage analysis of candidate regions for celiac disease genes. Hum Mol Genet 1997;6:1335–1339. Greco L, Corazza G, Babron MC, Clot F, Fulchignoni-Lataud MC, Percopo S, Zavattari P, Bouguerra F, Dib C, Tosi R, Troncone R, Ventura A. Genome search in celiac disease. Am J Hum Genet 1998;62:669–675. Weile B, Cavell B, Nivenius K, Krasilnikoff PR. Striking difference in the incidence of childhood celiac disease between Denmark and Sweden: a plausible explanation. J Pediatr Gastroenterol 1995;21:64–68. Ascher H. The role of quantity and quality of gluten-containing cereals in the epidemiology of coeliac disease. In: Ma¨ki M, Collin P, Visakorpi JK, eds. Coeliac disease. Vammala, Finland: Vammalan Kirjapaino, 1997:15–22. Corrao G, Corazza GR, Andreani ML, Torchio P, Valentini RA, Galatola G, Quaglino D, Gasbarrini G, di Orio F. Serological
242
37.
38.
39.
40. 41.
42.
43.
44.
45.
46.
47. 48.
49.
50.
51.
52.
DETLEF SCHUPPAN
screening of coeliac disease: choosing the optimal procedure according to various prevalence values. Gut 1994;35:771–775. Catassi C, Ratsch IM, Fabiani E, Rossini M, Bordicchia F, Candela F, Coppa GV, Giorgi PL. Coeliac disease in the year 2000: exploring the iceberg. Lancet 1994;343:200–203. Ventura A, Magazzu G, Greco L, and the SIGEP study group. Duration of exposure to gluten and risk for autoimmune disorders in patients with celiac disease. Gastroenterology 1999;117:297– 303. Greenberg CS, Birkbichler PJ, Rice RH. Transglutaminases: multifunctional cross-linking enzymes that stabilize tissues. FASEB J 1991;5:3071–3077. Piacentini M, Colizzi V. Tissue transglutaminase: apoptosis versus autoimmunity. Immunol Today 1999;3:130–134. Schuppan D, Ehnis T, Bauer M, Volta A, Donner P, Riecken EO, Dieterich W. Identification of the autoantigen of coeliac disease. Ann N Y Acad Sci 1998;859:121–126. Bruce SE, Bjarnason I, Peters TJ. Human jejunal transglutaminase: demonstration of activity, enzyme kinetics and substrate specificity with special relation to gliadin and coeliac disease. Clin Sci 1985;68:S73–S79. Szabolcs M, Sipka S, Csorba S. In vitro cross-linking of gluten into high molecular weight polymers with transglutaminase. Acta Paediatr Hung 1987;28:215–227. Johannsen BH, Vartdal F, Eriksen JA, Thorsby E, Sollid LM. Identification of a putative motif for binding of peptides to HLA-DQ2. Int Immunol 1996;8:177–182. Molberg Ø, MacAdam SN, Korner R, Quarsten H, Christiansen C, Madsen L, Fugger L, Scott H, Noren O, Roepstorff P, Lundin KEA, Sjøstrom H, Sollid L. Tissue transglutaminase selectively modifies gliadin peptides that are recognized by gut-derived T cells in celiac disease. Nat Med 1998;4:713–717. Van de Wal Y, Kooy Y, van Veelen P, Pena S, Mearin L, Papadopoulos G, Koning F. Selective deamidation by tissue transglutaminase strongly enhances gliading-specific T cell reactivity. J Immunol 1998;161:1585–1588. Sollid LM. Autoantibodies in coeliac disease: tissue transglutaminase-guilt by association? Gut 1997;47:851–852. Pender SL, Tickle SP, Docherty AJ, Howie D, Wathen NC, McDonald TT. A major role of matrix metalloproteinases in T cell injury in the gut. J Immunol 1997;158:1582–1590. Daum S, Bauer U, Foss HD, Schuppan D, Stein H, Riecken EO, Ullrich R. Increased expression of mRNA for matrix metalloproteinase-1 and -3 and tissue inhibitor of metalloproteinases-1 in intestinal biopsy specimens from patients with coeliac disease. Gut 1999;44:17–25. Bajaj-Elliott M, Poulsom R, Pender SL, Wathen NC, McDonald TT. Interactions between stromal cell-derived keratinocyte growth factor and epithelial transforming growth factor in immunemediated crypt cell hyperplasia. J Clin Invest 1999;102:1473– 1480. Halttunen T, Ma¨ki M. Serum IgA from patients with celiac disease inhibits human T84 intestinal crypt epithelial cell differentiation. Gastroenterology 1999;116:566–572. Dignass AU, Podolsky DK. Cytokine modulation of intestinal
GASTROENTEROLOGY Vol. 119, No. 1
53.
54.
55. 56.
57.
58.
59.
60.
61.
62.
63.
64.
epithelial cell restitution: central role of transforming growth factor beta. Gastroenterology 1993;105:1323–1332. Nunes I, Gleizes PE, Metz CN, Rifkin DB. Latent transforming growth factor binding protein domains involved in activation and transglutaminase-dependent crosslinking of latent transforming growth factor-beta. J Cell Biol 1997;136:1151–1163. King DP, Hyde DM, Jackson KA, Novosad DM, Ellis TN, Putney L, Stovall MY, Van Winkle LS, Beaman BL, Ferrick DA. Protective response to pulmonary injury requires gamma delta T lymphocytes. J Immunol 1999;162:5033–5036. Mak TW, Ferrick DA. The g/dT-cell bridge: linking innate and acquired immunity. Nat Med 1998;7:764–765. Groh V, Steinle A, Bauer S, Spies T. Recognition of stress-induced MHC molecules by intestinal epithelial g/d T cells. Science 1998;279:1737–1740. Bauer S, Groh V, Wu J, Steinle A, Philipps JH, Lanier LL, Spies T. Activation of NK cells by NKG2D, a receptor for stress-induced MICA. Science 1999;285:727–729. Dieterich W, Laag E, Scho¨pper H, Volta U, Ferguson A, Gillet H, Riecken EO, Schuppan D. Autoantibodies to tissue transglutaminase as predictors of coeliac disease. Gastroenterology 1998; 115:1317–1321. Sulkanen S, Halttunen T, Laurila K, Kolho KL, Korponay-Szabo IR, Sarnesto A, Savilahti E, Collin P, Ma¨ki M. Tissue transglutaminase autoantibody enzyme-linked immunosorbent assay in detecting celiac disease. Gastroenterology 1998;115:1322–1328. Seissler J, Borns S, Wohlrab U, Morgenthaler NG, Mothes T, Boehm BO, Scherbaum WA. Antibodies to human tissue transglutaminase measured by radioligand assay: evidence for high diagnostic sensitivity for celiac disease. Horm Metab Res 1999; 31:375–379. Mustalahti K, Collin P, Sievanen H, Ma¨ki M. Osteopenia in patients with clinically silent celiac disease. Lancet 1999;354: 744–745. Kordonouri O, Dieterich W, Schuppan D, Weber G, Mu¨ller C, Sarioglu N, Becker M, Danne T. Autoantibodies to tissue transglutaminase are sensitive serological parameters for detecting silent celiac disease in patients with type 1 diabetes. Diabetes Med (in press). Not T, Horvath K, Hill ID, Partanen J, Mammed A, Magazzu G, Fasano A. Celiac disease in the USA: high prevalence of antiendomysium antibodies in healthy blood donors. Scand J Gastroenterol 1998;33:494–498. Erb K. Atopic disorders: a default pathway in the absence of infection? Immunol Today 1999;7:317–322.
Received December 7, 1999. Accepted February 23, 2000. Address requests for reprints to: Detlef Schuppan, M.D., Ph.D., First Department of Medicine, University of Erlangen-Nuernberg, Krankenhausstrasse 12, 91054 Erlangen, Germany. e-mail: [email protected]; fax: (49) 9131-8536003. Supported by Deutsche Forschungsgemeinschaft (grant Schu 646/11-1) and a grant from the German Celiac Foundation.