Secondary Logo

Journal Logo

Invited Reviews

Role of Tissue Transglutaminase in Celiac Disease

Molberg, Øyvind; McAdam, Stephen N.; Sollid, Ludvig M.

Author Information
Journal of Pediatric Gastroenterology and Nutrition: March 2000 - Volume 30 - Issue 3 - p 232-240
  • Free

Celiac disease (CD), a gluten-sensitive enteropathy, is a multifactorial disease affecting 1:200 to 1:400 persons within the European populations (1). The chronic small intestinal inflammation of CD is the result of an inappropriate T-cell–mediated immune response against ingested gluten proteins from wheat and similar proteins in barley and rye (2,3). Normally, our immune system displays tolerance to edible proteins. To understand how and why CD develops, it is therefore critical to identify the genetic and environmental factors involved in the failure of tolerance to gluten proteins. There is a strong genetic influence on the susceptibility to CD and the predisposing genes involve both HLA and non-HLA genes (4,5). It is notable that, in common with most other HLA-associated diseases, CD is not a result of a gene defect, but rather the combined effect of normally functioning variant gene products (6). The primary HLA-association in CD is to the HLA-DQA1*0501, DQB1*0201 genes encoding the DQ2 molecule and to a lesser extent to the DQA1*0301, DQB*0302 genes encoding the DQ8 molecule (7). Interestingly, it seems that non-HLA genes together contribute more to genetic susceptibility than do the HLA-genes, but the contribution from each single, predisposing non-HLA gene appears to be modest (1).

Disease-specific CD4+ memory T cells that recognize gluten proteins can be isolated from biopsy specimens of affected intestine (8–12). A striking feature of these gluten-specific T cells is that most of them use the disease-associated DQ2 molecule (or the DQ8 molecule in the few DQ8+ patients) as their restriction elements (8–10). Moreover, CD has an autoimmune component characterized by the occurrence of disease-specific autoreactive antibodies (Abs), whose serum titers fluctuate with exposure to gluten in the diet. In 1996, Dieterich et al. (13) demonstrated that these autoreactive serum IgA Abs targeted the enzyme tissue transglutaminase (tTG). The diagnostic implications of this seminal study are currently under intense investigation (14,15). The finding has also renewed the interest in defining the roles of this enzyme in CD.

This review is primarily focused on recent advances in our understanding of how tTG may be involved in disease pathogenesis. First, however, we provide an overview of the reactions catalyzed by tTG and the various functions of the enzyme, along with a brief description of the potential roles of tTG in human diseases other than CD.


The most established property of tTG and the other transglutaminases is to catalyze a Ca2+-dependent cross-linking of a specific glutamine residue in the substrate protein to a primary amine (Fig. 1) (16,17). The acceptor amine can be either a protein-bound lysine (Lys) or a polyamine. Notably, the reaction between substrate and tTG occurs in two distinct steps (16). The first step is binding of a Gln side-chain amino group from the substrate to the catalytic site of tTG, which is followed by release of NH3 and formation of a substrate–enzyme intermediate. The second step is the reaction between substrate and an available amine, or in the absence of amines, the reaction with H2O. The substrate can become cross-linked to Lys in other proteins, to polyamines, or if the substrate itself contains available Lys residues to itself through intra-or intersubstrate cross-linking. If the substrate reacts with H2O, deamidation of the reactive Gln to yield glutamic acid (Glu) will occur. It should be noted that the reaction with H2O occurs with a slower rate than the reaction with acceptor amines (18), for this reason the deamidation of substrate by tTG has often been disregarded (19). Importantly, the result of the two-step reaction between tTG and substrate is always a posttranslational modification of a specific Gln residue in the substrate.

FIG. 1.
FIG. 1.:
Tissue transglutaminase modifies its substrates by cross-linking to primary amines or by deamidation of specific glutamine residues. (A) Tissue transglutaminase catalyzed formation of a ε-(γ-glutaminyl)-lysine (amine) bond that cross-links the substrate to a protein-bound lysine or a primary amine. (B) Tissue transglutaminase catalyzed deamidation of substrate.

All the transglutaminases display a high degree of specificity such that only a limited number of proteins can act as substrate. The rules governing the binding of protein-bound Gln residues to tTG are complex and far from clear (20–22). In contrast, it appears that most protein-bound Lys and polyamines can be accommodated in the catalytic site to cross-link the substrate (23). In addition to the catalytic site and a minimum of two Ca2+-binding sites, tTG has an N-terminal fibronectin association site (24) and a site that binds and hydrolyses guanosine triphosphate (GTP) (25). This latter GTPase activity of tTG is independent of the transglutaminating activity, but binding of GTP inhibits substrate binding to the catalytic site. In tTG-transfected cells, the intracellular GTPase activity of tTG mediates α1-adrenergic receptor–mediated stimulation of phospholipase C (25). Notably, whereas the catalysis of substrate modification by tTG takes place both in intracellular and extracellular compartments, the GTP hydrolysis by tTG is strictly an intracellular function (19).


The transglutaminases are a large family of related enzymes that are distributed widely in tissues and body fluids of all mammals (19,23). The genes encoding transglutaminases are well conserved between species (for instance 80% homology between guinea pig tTG and human tTG) (19), reinforcing the notion that these enzymes have important biological functions. In humans the transglutaminase family includes tTG, blood coagulation factor XIII, keratinocyte TG, epidermal TG, prostate TG, erythrocyte band 4.2, and a novel transglutaminase from the skin (26). The different human transglutaminases share a common motif in the catalytic site and have similar gene structures, indicating that they stem from the same ancestral gene. They are, however, encoded by separate genes and display unique expression patterns as well as differences in substrate specificity. The human tTG gene encodes a 684-amino-acid nonglycosylated single-polypeptide chain (27). It is not clear whether the gene contains any polymorphisms. The tTG gene is expressed in a broad range of tissues, under complex but tight regulation that is influenced by many factors including retinoids, transforming growth factor (TGF)-β1, interleukin-6, and DNA methylation (27–31). Constitutive expression of tTG is found in endothelial cells in blood vessels, in smooth muscle cells, and in some other organ-restricted cells (32). In CD, it is interesting that tTG has been detected in all layers of the small intestinal wall. The enzyme is most heavily expressed in the submucosa, and only 1% of the small intestinal tTG is located within the epithelium. It is noteworthy that crypt epithelial cells do not express tTG, but the intracellular expression increases as the cells maturate and migrate toward the small intestinal villi (32).

The physiological roles of tTG are not fully characterized. In fact, many basic questions remain unresolved. Tissue transglutaminase does not have a typical signal sequence for routing through the normal secretory pathway of cells, but despite this, tTG appears to be actively transported to the extracellular space through an unknown mechanism and not simply shed from dead cells (19). After its synthesis, tTG is found in the cytosol as an inactive enzyme. The effect of Ca2+ ions and GTP on enzyme activation is well described (33). Otherwise, the requirements for intra-and extracellular activation of tTG are largely unknown. An exciting observation is that binding of tTG to specific membrane lipids is involved in the regulation by decreasing the dependence of Ca2+(34). Taken together, available data suggest that the biosynthesis, secretion, and activation of tTG are strictly but independently regulated processes.

A number of in vitro studies implicate tTG in various, often critical biological processes as diverse as cell growth and differentiation (35), formation of extracellular matrix (36), tissue repair (37), apoptosis (38), and, independently, signal transduction through the α1-adrenergic receptor (25). Because all these processes appear to be critical for homeostasis, it could be envisaged that one or more of them could be involved in CD.

While bearing in mind that CD is an immune-mediated disorder, it should also be noted that the activity of tTG has been associated with receptor-mediated endocytosis (39), phagocytosis (40), antigen presentation (41,42), and differentiation of lymphocytes, monocytes, and B cells (43–45). It is particularly interesting that macrophages display increased tTG activity after inflammatory stimuli and that this increase in enzyme activity leads to enhanced Fc receptor-mediated endocytosis in the macrophages (40,46).


Tissue TG has been linked to a number of diseases, including neurologic disorders (47–50), cancer (51,52), human immunodeficiency virus (HIV) infection (38), inflammatory bowel disease (53), liver cirrhosis (29), cataract of the eye (30), and various autoimmune disorders (54). The role of tTG in certain neurologic disorders, including Huntington's disease has been extensively studied. The huntingtin proteins associated with the central nervous system involvement of this disorder are good substrates for tTG (49,50). Interestingly, huntingtin proteins from patients with Huntington's disease have a polyglutamine stretch above a critical length, a feature that enhances their activity as substrates for tTG (50). Moreover, brains from patients with Huntington's disease display higher tTG activity than control brains (49). Another example of pathologic protein cross-linking in the central nervous system is found in Alzheimer's disease. In this disorder it is the disease-associated tau proteins that become cross-linked by tTG (47).

A close correlation between the expression of tTG and the rate of apoptosis have been reported in HIV-infected CD4+ T cells (38), this supports the hypothesis that tTG-mediated cross-linking is a critical event during induction of apoptosis. It has been proposed that the tTG-mediated modification of intracellular proteins during apoptosis can lead to formation of neoantigens that may trigger autoimmunity (54,55). This possibility has hitherto been supported only by circumstantial evidence.

The deamidating activity of tTG implicated in CD (see later description) has so far not been associated with other diseases. However, the lenses of the eye have high TG activity, and it has been shown that specific Gln residues in water-soluble eye lens proteins are deamidated (56). Whether these deamidations, which are associated with cataract development, are catalyzed by tTG is presently unknown (57). In addition, recent studies have described deamidation of specific Glns in human proteins by a bacterial TG. Notably, these deamidations dramatically change the functional properties of the human proteins (58).


In a pioneering study in 1985, Bruce et al. (59) demonstrated that homogenates of intestinal biopsy specimens contain an enzyme that has the ability to cross-link labeled polyamines to a defined transglutaminase substrate. They also showed that specimens from treated and untreated patients with CD had higher transglutaminase activity than control specimens. Notably, the homogenates also catalyzed cross-linking of the labeled polyamines to gliadin proteins, indicating that gliadin could also act as a substrate for transglutaminase. An Italian study confirmed the finding of transglutaminase activity in biopsy specimens from patients with CD and further reported decreased serum transglutaminase activity in the same patients (60). However, it should be noted that serum transglutaminase activity is mainly mediated by factor XIII and not by tTG.

Later, gliadin proteins were identified as good substrates for tTG. In fact, up to 36% of the Gln residues in gliadin were accessible to modification by tTG, when evaluated either by polyamine incorporation or by liberation of NH3 during deamidation of the gliadin (61). In both cases, large polymers, probably gliadin-gliadin complexes (formed by cross-linking of Gln to one of the few gliadin Lys residues) and/or tTG-gliadin complexes were detected.

The identification of tTG as the target of CD-specific IgA anti-endomysial Abs (EMAs) revived the idea that these large complexes may play a role in the pathogenesis of CD (13). Dieterich et al. (62) demonstrated that tTG-gliadin complexes could be formed in vitro and argued that these complexes may contain neoepitopes that trigger mucosal T cells and degrade tolerance. The production of the anti-tTG IgA Abs must be dependent on cognate T-cell help to facilitate the isotype switching of the autoreactive B cells. Dieterich et al. suggested that the necessary help for the B-cell production of anti-tTG IgA could be provided by autoreactive T cells specific for tTG. We, however, doubt the existence of autoreactive T cells. Because tTG is expressed in the thymic epithelium, it is likely that tTG-reactive T cells are deleted during negative selection in the thymus (32). Moreover, the consequence of having autoreactive T cells specific for such a ubiquitously expressed enzyme as tTG, would most likely be systemic autoimmunity (63). As an alternative, we have suggested a model that is not dependent on the presence of tTG-reactive T cells. Instead the production of anti-tTG antibodies is driven entirely by intestinal gliadin-specific T cells (63). Our hypothesis is based on the premise that the tTG-gliadin complexes form in vivo function analogous to the hapten-carrier complexes described in mouse immunology (64). The principle is simple. The tTG-gliadin complexes must be bound to the tTG-specific B cells to be endocytosed and processed. This processing allows both tTG and gliadin fragments to be presented by DQ2 and the other class II molecules expressed by the tTG-specific B cells. The gliadin-DQ2 complexes presented by the tTG-specific B cells are recognized by the gliadin-specific T cells present in the celiac mucosa. The activation of these T cells results in proliferation and provides in return the necessary help for the tTG-specific B cell to produce anti-tTG Abs (63). Recently, two reports have supported the notion that anti-tTG Abs can be produced by B cells present in the celiac lesions. Both the studies demonstrated that anti-tTG Abs were detectable in supernatants from intestinal CD biopsy specimens cultured ex vivo (65,66). In one of these studies the production of anti-tTG was independent of biopsy challenge (66), whereas in the other the production of anti-tTG was found to be dependent on gliadin challenge (65). The possibility of a functional role of the anti-tTG Abs in disease pathogenesis will be discussed later.

Although the studies mentioned indicate a role of tTG in disease pathogenesis, their results were not easily reconciled with the prevailing view of CD as a T-cell–mediated disorder. We wanted to examine whether such a link existed, particularly because we knew at that time that intestinal T cells of patients with CD recognize deamidated gliadins and because tTG was capable of deamidating these proteins. Intriguingly, tTG had a potent and specific effect on the intestinal T-cell recognition of gliadin. We demonstrated that tTG in vitro catalyzed an ordered and specific deamidation of certain glutamine (Gln) residues in wheat gliadin peptides. These specific modifications generated an antigen that was optimal for gut-derived, DQ-restricted T cells (67). Fine-specificity analysis of an immunorelevant γ-gliadin epitope (DQ2-γ-I) recognized by a DQ2-restricted intestinal T cell (68) unveiled that the tTG-mediated deamidations occur in positions that were beneficial for the binding of the peptide to DQ2 and critical for recognition by the intestinal T cell (67). Independently, van de Wal et al. demonstrated that tTG-modification of a gliadin epitope recognized by a DQ8-restricted intestinal T-cell clone greatly enhanced the reactivity of the T cell (69).

Recently, two different intestinal T-cell epitopes (DQ2-α-I and DQ2-α-II) located within two recombinant α-gliadins (α-9 and α-2, respectively) were identified (70). The α-gliadin peptides containing the epitopes were distinct but partially overlapping and shared a common seven-amino-acid motif—α-2 (62–75) and α-9 (57–68). In their native form, these peptides were not recognized by intestinal T cells, but after deamidation by tTG both became very efficient T-cell antigens. Intriguingly, tTG was found to catalyze deamidation of the same single Gln residue (Gln65) in epitope DQ2-α-I (α-9[57-68]E65) and DQ2-α-II (α-2[62-75]E65). Moreover, it appeared that the DQ2-α-I and DQ2-α-II epitopes were very frequently recognized by intestinal T cells from adult DQ2+ Norwegian patients with CD. Combined analysis of T-cell recognition and DQ2 binding assays indicated that the Glu65 residue of the DQ2-α-I epitope was accommodated in relative position 6 in the DQ2 peptide-binding groove and the Glu65 residue of DQ2-α-II in relative position 4. The DQ2-binding assays demonstrated that the native α-2 and α-9 gliadin peptides bound poorly to DQ2, whereas the deamidated epitopes bound with reasonable, but by no means exceptional, binding affinity (70).

The data from the mapping of three distinct intestinal T-cell epitopes allows the following conclusions. In their native form, the three peptides bind poorly to DQ2 and are at best only barely recognized by intestinal T cells. After in vitro tTG treatment, they are, however, transformed into epitopes that bind reasonably well to DQ2 and are recognized by DQ2-restricted T cells isolated from biopsy specimens of patients with CD. It appears that the deamidation of the gliadin peptides have a dual effect. The introduction of negatively charged Glu residues at defined positions in the gliadin peptides markedly increases their binding affinity to the disease-associated DQ2 molecule (71). This is in line with the proposed peptide-binding motif for DQ2 that predicts a preference for negatively charged residues in certain positions (72–75). Notably, the deamidated residues in the three defined epitopes are located in different relative positions. This indicates that the immune response against gliadin in CD is not directed against one pathogenic gliadin peptide or a single “pathogenic motif.” The tTG-mediated deamidation of the three different gliadin peptides is also critical for recognition by intestinal T cells. Collectively, these data suggest that tTG, by deamidating certain residues in ingested gliadin proteins, could be a key factor in the breakdown of tolerance in CD.

The data above demonstrated the in vitro effect of tTG-catalyzed deamidations of gliadin. A critical question, however, is whether tTG can also deamidate gliadin proteins in vivo. Immunohistochemical staining of intestinal biopsy specimens from patients with CD demonstrated a spatial relationship between extracellular tTG, DQ-expressing cells, and T cells in the subepithelial region (67). Moreover, we have preliminary data to suggest that T cells from CD lesions recognize gliadin epitopes deamidated in situ by endogenous tTG (Molberg et al., submitted, 2000). Taken together, these observations provide persuasive evidence that the deamidation of gliadin peptides in vivo is mediated by endogenous tTG in a microenvironment containing the necessary factors to initiate an immune response.

Gliadin-specific T cells isolated from the peripheral blood of patients with CD are neither predominantly DQ restricted nor sensitive to tTG treatment of the gliadin (67,76). This strongly argues that they recognize gliadin epitopes different from the epitopes recognized by intestinal T cells. A possible explanation for this is that the gliadin-specific T cells isolated from the celiac lesions represent a distinct subpopulation of memory T cells that selectively home to the intestinal microenvironment from whence they were initially triggered. Gliadin-specific T cells can be isolated from the mucosa of patients with CD who have consumed a gluten-free diet for 10 to 20 years, indicating that at least some of them are localized in the lamina propria throughout the patients' lifespans. It is tempting to speculate that tTG, by interacting with extracellular proteins and perhaps plasma membrane proteins can be involved in the formation of a reticular network that colocalizes memory T cells and DQ-expressing cells in the celiac lesions.

In the introduction, we stated that an understanding of the pathogenesis of CD was dependent on the identification of the involved genetic and environmental factors. From such a perspective, the demonstration of the connection between intestinal T-cell reactivity and tTG-mediated deamidation of gliadin generates as many new questions as answers.

An attractive but speculative hypothesis is that tolerance to gluten is broken in patients with CD when endogenous tTG starts to form deamidated gluten peptides that bind to DQ2 molecules in the mucosa and thereby trigger local CD4+ T cells (Fig. 2). Based on this hypothesis, the chronic intestinal inflammation can be viewed as the final consequence of complex interactions between the products of disease predisposing genes and environmental factors. That biopsy specimens from both untreated patients with CD and patients consuming a gluten-free diet display higher tTG activity than those of control subjects indicates that some of the genes predisposing to CD may be identified among the factors that affect the transcription regulation or functional activity of the enzyme.

FIG. 2.
FIG. 2.:
The specific deamidation of gluten peptides by intestinal tissue transglutaminase (tTG) has a dual effect. First, it increases the binding affinity of the peptides to DQ2-molecules by introducing negatively charged glutamic acid residues. Second, it transforms the peptides from nonstimulatory to efficient T-cell antigens.

The triggering of pathogenic immune responses may be related to induction of the deamidating activity of tTG. Feasibly, deamidation only happens if tTG activity is high, substrate is in excess, and the level of acceptor amines is low. During episodes of intestinal inflammation, such situations could occur. Inflammation induces tTG activity, disturbs epithelial integrity, and, notably, increases the need for exogenous and endogenous polyamines, in that cross-linking of polyamines by tTG is critical for normal repair functions of the intestinal mucosa (37). The polyamine content in a normal diet is not high, and a rate-limiting enzyme restricts synthesis of polyamines (77). An inflammation may thus cause a local depletion of polyamines concurrently with an increased leakage of gluten proteins across the epithelial barrier. Notably, the possibility that the deamidating activity identified in celiac lesions is not mediated by tTG but by some other as yet unidentified enzyme should not be discarded. No other human enzyme that catalyzes deamidation of specific protein-bound Gln residues has been identified, but it should be remembered that bacterial TG, which could be present in the gut lumen, preferentially deamidates the certain subtrates, instead of cross-linking them (58).

The activity of tTG is implied in critical biological functions that, if dysfunctional, may disrupt homeostasis in the intestinal mucosa. Although there are currently few data to support these ideas, it is worth mentioning a few examples. Increased apoptosis of epithelial cells may destroy the integrity of the epithelium and cause increased leakage of gluten proteins. Moreover, tTG-associated changes in the apoptotic rate of local immune-competent cells may skew the balance from tolerance to aggressive immune responses. The activation of TGF-β, a pleiotropic growth factor, is partly driven by tTG (78), and alterations in the level of active TGF-β have specific effects on T cells, B cells, and antigen-presenting cells (79). Finally, it has been demonstrated that binding of calreticulin to tTG in vitro inhibits both the TG and the GTPase activity of the enzyme (80). Notably, calreticulin has been described as an additional target for Abs in CD (81). The binding of calreticulin to tTG, in the presence of gliadin, enables formation of hapten-carrier complexes comparable to the previously described tTG-gliadin complexes. It may thus be envisaged that gliadin-specific T cells provide cognate help to production of anti-calreticulin Abs by calreticulin-specific B cells in a fashion similar to the help provided to the production of anti-tTG Abs.

The anti-tTG IgA Abs from patient sera preferentially bind Ca2+-activated tTG (14,15). This strongly indicates that they recognize conformation-dependent epitopes in the enzyme. In certain autoimmune diseases, including Mb Addison, autoimmune thyroiditis, and type I diabetes (82–84), similar conformation-dependent serum auto-Abs that target disease-relevant enzymes have been identified. A subset of these conformation-dependent auto-Abs have been shown to interfere with the catalytic activity of the enzyme (82,83). Thus, a critical question that would have a profound impact on our understanding of the pathogenesis of CD is whether anti-tTG Abs modulate the activity of tTG in any way (85).

Given that anti-tTG Abs affect tTG, many different outcomes are possible. The antibodies could inhibit or enhance the cross-linking, deamidating and/or GTPase activity of the enzyme. In any case, a feedback loop between anti-tTG Abs and the activity of tTG could be envisaged. In CD the most obvious feedback system is connected to the tTG-mediated deamidation of gluten peptides. As outlined earlier, this is followed in vivo by binding of the deamidated peptides to DQ2 and specific activation of intestinal T cells that provides help for the tTG-specific B cells to produce the anti-tTG Abs that finally bind to activated tTG. The effect of this feedback system is dependent on how the Abs influence the rate of deamidation of the gliadin peptides. If antibody binding inhibits the catalytic activity of tTG, the effect should be beneficial, because fewer gliadin peptides become deamidated. In contrast, if binding enhances tTG activity, a vicious circle perpetuated by gliadin exposure is formed.

Indications of an inhibiting effect on the cross-linking activity of tTG by serum anti-tTG IgA Abs from patients with CD have been reported in a recent in vitro study (86). It should be noted, however, that this study did not provide formal evidence that the observed effects were due to interactions between anti-tTG and tTG (86). Many relevant arguments against a role of anti-tTG IgA Abs in CD can also be put forth. One is that CD occurs with an increased frequency among patients with primary IgA deficiency (1). In this subgroup of patients, intestinal lesions develop that are indistinguishable from those in other patients, a fact that argues against a critical role of anti-tTG IgA in disease pathogenesis. It should be noted, however, that it is not clear whether the anti-tTG IgG found in the IgA-deficient patients with CD may substitute for the anti-tTG IgA produced by the other patients. Another concern is the possible effects of binding of serum anti-tTG Abs to tTG expressed outside the small intestinal mucosa. It could be speculated that if this binding of serum Abs to enzyme influences tTG activity, it would cause systemic responses. Along the same lines is the suggestion that serum anti-tTG Abs could cross-react with the other transglutaminases. This possibility is particularly interesting with respect to dermatitis herpetiformis. Patients with this CD-associated skin disorder have levels of serum anti-tTG IgA comparable with other patients with CD (87). Therefore, it can not be excluded that the enigmatic (and diagnostic) IgA deposits in the skin of patients with dermatitis herpetiformis are in fact anti-tTG Abs cross-reacting with a transglutaminase expressed in skin (87).

The elucidation of the role of tTG in CD will be facilitated by the functional characterization of the anti-tTG Abs. If they affect enzymatic activity, then CD must be redefined as a T-cell–mediated disorder modulated by autoimmune B cells. Ongoing research should clarify these critical issues.


We have focused on a discussion of tTG and its potential role in CD. The data obtained so far demonstrate at least two different, although not necessarily independent, roles of this enzyme in disease pathogenesis. First, activated tTG is the main target of disease-specific autoantibodies. Second, the deamidating activity of the enzyme has been shown to create gliadin peptides that bind DQ2 to be recognized by disease-specific intestinal T cells. Future work should establish the connection between these two immune-related functions of tTG and, as discussed, should search for further roles of the enzyme. Notably, the phenomenon of tTG-mediated modification of substrate has been linked to several diseases, but it is only in CD that the role of the deamidating activity of the enzyme has been defined.

On a more general level, it is interesting to note that an increasing number of reports identify strong, specific immune responses against posttranslationally modified proteins (88–91). The modification of proteins could occur spontaneously, or as is the case in CD, could be catalyzed by enzymes. A multitude of enzymes capable of catalyzing specific protein modifications exists, and abnormal activation of any of these in microenvironments with ongoing immune activation could modify self or non–self proteins to create neoepitopes. Because they by definition are novel to the immune system, all such neoepitopes could be involved in the breaking of tolerance and precipitation of disease.


1. Sollid LM. Molecular basis of celiac disease. Ann Rev Immunol (in press).
2. Marsh MN, Ensari A, Morgan S. Evidence that gluten sensitivity is an immunologic disease. Curr Opin Gastroenterol 1993; 9:994–1000.
3. Trier JS. Celiac sprue. N Engl J Med 1991; 325:1709–19.
4. Petronzelli F, Bonamico M, Ferrante P, et al. Genetic contribution of the HLA region to the familial clustering of coeliac disease. Ann Hum Genet 1997; 6:307–17.
5. Greco L, Corazza G, Babron MC, et al. Genome search in celiac disease. Am J Hum Genet 1998; 62:669–75.
6. Thorsby E. Invited anniversary review: HLA associated diseases. Hum Immunol 1997; 53:1–11.
7. Sollid LM, Thorsby E. HLA susceptibility genes in celiac disease: Genetic mapping and role in pathogenesis. Gastroenterology 1993; 105:910–22.
8. Lundin KEA, Scott H, Hansen T et al. Gliadin-specific, HLA-DQ(1*0501,1*0201) restricted T cells isolated from the small intestinal mucosa of celiac disease patients. J Exp Med 1993; 178:187–96.
9. Lundin KEA, Scott H, Fausa O, Thorsby E, Sollid LM. T cells from the small intestinal mucosa of a DR4, DQ7/ DQ8 celiac disease patient preferentially recognize gliadin when presented by DQ8. Human Immunology 1994; 41:285–91.
10. Molberg Ø, Lundin KEA, Nilsen EM, et al. HLA restriction patterns of gliadin-and astrovirus-specific CD4+ T cells isolated in parallel from the small intestine of celiac disease patients. Tissue Antigens 1998; 52:407–15.
11. Troncone R, Gianfrani C, Mazzarella G, et al. Majority of gliadin-specific T-cell clones from celiac small intestinal mucosa produce interferon-and interleukin-4. Dig Dis Sci 1998; 43:156–61.
12. van de Wal Y, Kooy YC, van Veelen PA, et al. Small intestinal T cells of celiac disease patients recognize a natural pepsin fragment of gliadin. Proc Natl Acad Sci USA 1998; 95:10050–4.
13. Dieterich W, Ehnis T, Bauer M, et al. Identification of tissue transglutaminase as the autoantigen of celiac disease. Nat Med 1997; 3:797–801.
14. Dieterich W, Laag E, Schopper H, et al. Autoantibodies to tissue transglutaminase as predictors of celiac disease. Gastroenterology 1998; 115:1317–21.
15. Sulkanen S, Halttunen T, Laurila K, et al. Tissue transglutaminase autoantibody enzyme-linked immunosorbent assay in detecting celiac disease. Gastroenterology 1998; 115:1322–8.
16. Folk JE. Mechanism and basis for specificity of transglutaminase-catalyzed ε-γ-glutamyl) lysine bond formation. Adv Enzymol Relat Areas Mol Biol 1983; 54:1–56.
17. Lorand L, Conrad SM. Transglutaminases. Mol Cell Biochem 1984; 58:9–35.
18. Folk JE, Cole PW, Mullooly JP. Mechanism of action of guinea pig liver transglutaminase. V: The hydrolysis reaction. J Biol Chem 1968; 243:418–27.
19. Aeschlimann D, Paulsson M. Transglutaminases: Protein cross-linking enzymes in tissues and body fluids. Thromb Haemost 1994; 71:402–15.
20. Aeschlimann D, Paulsson M, Mann K. Identification of Gln726 in nidogen as the amine acceptor in transglutaminase-catalyzed cross-linking of laminin-nidogen complexes. J Biol Chem 1992; 267:11316–21.
21. Gorman JJ, Folk JE. Structural features of glutamine substrates for transglutaminases: Role of extended interactions in the specificity of human plasma factor XIIIa and of the guinea pig liver enzyme. J Biol Chem 1984; 259:9007–10.
22. Kahlem P, Terre C, Green H, Djian P. Peptides containing glutamine repeats as substrates for transglutaminase-catalyzed cross-linking: Relevance to diseases of the nervous system. Proc Natl Acad Sci USA 1996; 93:14580–5.
23. Greenberg CS, Birckbichler PJ, Rice RH. Transglutaminases: Multifunctional cross-linking enzymes that stabilize tissues. FASEB J 1991; 5:3071–7.
24. Jeong JM, Murthy SN, Radek JT, Lorand L. The fibronectin-binding domain of transglutaminase. J Biol Chem 1995; 10:5654–8.
25. Nakaoka H, Perez DM, Baek KJ, et al. Gh: A GTP-binding protein with transglutaminase activity and receptor signaling function. Science 1994; 264:1593–6.
26. Aeschlimann D, Koeller MK, Allen-Hoffmann BL, Mosher DF. Isolation of a cDNA encoding a novel member of the transglutaminase gene family from human keratinocytes: Detection and identification of transglutaminase gene products based on reverse transcription-polymerase chain reaction with degenerate primers. J Biol Chem 1998; 273:3452–60.
27. Lu S, Saydak M, Gentile V, Stein JP, Davies PJ. Isolation and characterization of the human tissue transglutaminase gene promoter. J Biol Chem 1995; 270:9748–56.
28. Lu S, Davies PA. Regulation of the expression of the tissue transglutaminase gene by DNA methylation. Proc Natl Acad Sci USA 1997; 94:4692–7.
29. Kuncio GS, Tsyganskaya M, Zhu J, et al. TNF-α modulates expression of the tissue transglutaminase gene in liver cells. Am J Physiol 1998; 274:1–5.
30. Murthy SN, Velasco PT, Lorand L. Properties of purified lens transglutaminase and regulation of its transamidase/crosslinking activity by GTP. Exp Eye Res 1998; 67:273–81.
31. Suto N, Ikura K, Sasaki R. Expression induced by interleukin-6 of tissue-type transglutaminase in human hepatoblastoma HepG2 cells. J Biol Chem 1993; 268:7469–73.
32. Thomazy V, Fesus L. Differential expression of tissue transglutaminase in human cells: An immunohistochemical study. Cell Tissue Res 1989; 255:215–24.
33. Zhang J, Lesort M, Guttmann RP, Johnson GV. Modulation of the in situ activity of tissue transglutaminase by calcium and GTP. J Biol Chem 1998; 273:2288–95.
34. Lai TS, Bielawska A, Peoples KA, Hannun YA, Greenberg CS. Sphingosylphosphocholine reduces the calcium ion requirement for activating tissue transglutaminase. J Biol Chem 1997; 272:16295–300.
35. Aeschlimann D, Mosher D, Paulsson M. Tissue transglutaminase and factor XIII in cartilage and bone remodeling. Semin Thromb Hemost 1996; 22:437–43.
36. Schittny JC, Paulsson M, Vallan C, et al. Protein cross-linking mediated by tissue transglutaminase correlates with the maturation of extracellular matrices during lung development. Am J Respir Cell Mol Biol 1997; 17:334–43.
37. Wang JY, Johnson LR. Role of transglutaminase and protein cross-linking in the repair of mucosal stress erosions. Am J Physiol 1992; 262:19–25.
38. Amendola A, Gougeon ML, Poccia F, et al. Induction of “tissue” transglutaminase in HIV pathogenesis: Evidence for high rate of apoptosis of CD4+ T lymphocytes and accessory cells in lymphoid tissues. Proc Natl Acad Sci USA 1996; 93:11057–62.
39. Davies PJ, Davies DR, Levitzki A, et al. Transglutaminase is essential in receptor-mediated endocytosis of 2-macroglobulin and polypeptide hormones. Nature 1980; 283:162–7.
40. Davies PJ, Murtaugh MP. Transglutaminase and receptor-mediated endocytosis in macrophages and cultured fibroblasts. Mol Cell Biochem 1984; 58:69–77.
41. Teshigawara K, Kannagi R, Noro N, Masuda T. Possible involvement of transglutaminase in endocytosis and antigen presentation. Microbiol Immunol 1985; 29:737–50.
42. Pober JS, Strominger JL. Transglutaminase modifies the carboxy-terminal intracellular region of HLA-A and -B antigens. Nature 1981; 289:819–21.
43. Murtaugh MP, Arend WP, Davies PJ. Induction of tissue transglutaminase in human peripheral blood monocytes. J Exp Med 1984; 159:114–25.
44. Julian C, Speck NA, Pierce SK. Primary amines inhibit the triggering of B lymphocytes to antibody synthesis. J Immunol 1983; 130:91–6.
45. Metafora S, Peluso G, Ravagnan G, et al. Implication of transglutaminase in mitogen-induced human lymphocyte blast transformation. Adv Exp Med Biol 1988;231–84.
46. Leu RW, Herriott MJ, Moore PE, Orr GR, Birckbichler PJ. Enhanced transglutaminase activity associated with macrophage activation. Possible role in Fc-mediated phagocytosis. Exp Cell Res 1982; 141:191–9.
47. Murthy SN, Wilson JH, Lukas TJ, Kuret J, Lorand L. Cross-linking sites of the human tau protein, probed by reactions with human transglutaminase. J Neurochem 1998; 71:2607–14.
48. Selkoe DJ, Abraham C, Ihara Y. Brain transglutaminase: In vitro crosslinking of human neurofilament proteins into insoluble polymers. Proc Natl Acad Sci USA 1982; 79:6070–4.
49. Karpuj MV, Garren H, Slunt H, et al. Transglutaminase aggregates huntingtin into nonamyloidogenic polymers, and its enzymatic activity increases in Huntington's disease brain nuclei. Proc Natl Acad Sci USA 1999; 46:7388–93.
50. Kahlem P, Green H, Djian P. Transglutaminase action imitates Huntington's disease: Selective polymerization of Huntingtin containing expanded polyglutamine. Mol Cell 1998; 1:595–601.
51. Darro F, Cahen P, Vianna A, et al. Growth inhibition of human in vitro and mouse in vitro and in vivo mammary tumor models by retinoids in comparison with tamoxifen and the RU-486 anti-progestagen. Breast Cancer Res Treat 1998; 51:39–55.
52. Walker AM, Montgomery DW, Saraiya S, et al. Prolactin-immunoglobulin G complexes from human serum act as costimulatory ligands causing proliferation of malignant B lymphocytes. Proc Natl Acad Sci USA 1995; 92:3278–82.
53. D'Argenio G, Biancone L, Cosenza V, et al. Transglutaminases in Crohn's disease. Gut 1995; 37:690–5.
54. Piacentini M, Colizzi V. Tissue transglutaminase: Apoptosis versus autoimmunity. Immunol Today 1999; 20:130–4.
55. Utz PJ, Anderson P. Posttranslational protein modifications, apoptosis, and the bypass of tolerance to autoantigens. Arthritis Rheum 1998; 41:1152–60.
56. Hanson SR, Smith DL, Smith JB. Deamidation and disulfide bonding in human lens-crystallins. Exp Eye Res 1998; 67:301–12.
57. Lampi KJ, Ma Z, Hanson SR, et al. Age-related changes in human lens crystallins identified by two-dimensional electrophoresis and mass spectrometry. Exp Eye Res 1998; 67:31–43.
58. Schmidt G, Selzer J, Lerm M, Aktories K. The rho-deamidating cytotoxic necrotizing factor 1 from escherichia coli possesses transglutaminase activity: Cysteine 866 and histidine 881 are essential for enzyme activity. J Biol Chem 1998; 273:13669–74.
59. 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 (Colch) 1985; 68:573–9.
60. D'Argenio G, Sorrentini I, Ciacci C, et al. Human serum transglutaminase and coeliac disease: Correlation between serum and mucosal activity in an experimental model of rat small bowel enteropathy. Gut 1989; 30:950–4.
61. Larre C, Chiarello M, Blanloeil Y, Chenu M, Gueguen J. Gliadin modifications catalysed by guinea pig liver transglutaminase. J Food Biochem 1993; 112:267–82.
62. Dietrich W, Ehnis T, Bauer M, Riecken EO, Schuppan D. Gliadin is a preferred substrate for tissue transglutaminase, the autoantigen in coeliac disease (abstract). Gastroenterology 1997;A359.
63. Sollid LM, Molberg Ø, McAdam S, Lundin KEA. Autoantibodies in coeliac disease: Tissue transglutaminase—guilt by association. Gut 1997; 41:851–2.
64. Paul WE, Katz DH, Goidl EA, Benacerraf B. Carrier function in anti-hapten immune responses. II: Specific properties of carrier cells capable of enhancing anti-hapten antibody responses. J Exp Med 1970; 132:283–99.
65. Picarelli A, Maiuri L, Frate A, et al. Production of antiendomysial antibodies after in-vitro gliadin challenge of small intestine biopsy samples from patients with coeliac disease. Lancet 1996; 348:1065–7.
66. Vogelsang H, Schwarzenhofer M, Granditsch G, Oberhuber G. In vitro production of endomysial antibodies in cultured duodenal mucosa from patients with celiac disease. Am J Gastroenterol 1999; 94:1057–61.
67. Molberg Ø, Mcadam SN, Körner R, et al. Tissue transglutaminase selectively modifies gliadin peptides that are recognized by gut-derived T cells in celiac disease. Nat Med 1998; 4:713–7.
68. Sjöström H, Lundin KEA, Molberg Ø, et al. Identification of a gliadin T-cell epitope in coeliac disease: General importance of gliadin deamidation for intestinal T-cell recognition. Scand J Immunol 1998; 48:111–5.
69. van de Wal Y, Kooy Y, van Veelen P, et al. Selective deamidation by tissue transglutaminase strongly enhances gliadin-specific T cell reactivity. J Immunol 1998; 161:1585–8.
70. Arentz-Hansen EH, Körner R, Molberg Ø, et al. The intestinal T cell response to -gliadin in adult celiac disease is focused on a single deamidated glutamine targeted by tissue transglutaminase. J Exp Med (in press).
71. Quarsten H, Molberg Ø, Fugger L, Mcadam SN, Sollid LM. HLA binding and T cell recognition of a tissue transglutaminase-modified gliadin epitope. Eur J Immunol 1999; 99:2506–14.
72. Kwok WW, Domeier ME, Johnson ML, Nepom GT, Koelle DM. HLA-DQB1 codon 57 is critical for peptide binding and recognition. J Exp Med 1996; 183:1253–8.
73. Johansen BH, Vartdal F, Eriksen JA, Thorsby E, Sollid LM. Identification of a putative motif for binding of peptides to HLA-DQ2. Int Immunol 1996; 79:177–82.
74. Vartdal F, Johansen BH, Friede T, et al. The peptide binding motif of the disease associated HLA-DQ (1* 0501, 1* 0201) molecule. Eur J Immunol 1996; 26:2764–72.
75. van de Wal Y, Kooy YC, Drijfhout JW, Amons R, Koning F. Peptide binding characteristics of the coeliac disease-associated DQ(1*0501, 1*0201) molecule. Immunogenetics 1996; 44:246–53.
76. Gjertsen HA, Sollid LM, Ek J, Thorsby E, Lundin KE. T cells from the peripheral blood of coeliac disease patients recognize gluten antigens when presented by HLA-DR, -DQ, or -DP molecules. Scand J Immunol 1994; 39:567–74.
77. Tabor CW, Tabor H. Polyamines. Annu Rev Biochem 1984; 53: 749–90.
78. Nunes I, Gleizes PE, Metz CN, Rifkin DB. Latent transforming growth factor-beta binding protein domains involved in activation and transglutaminase-dependent cross-linking of latent transforming growth factor-β. J Cell Biol 1997; 136:1151–63.
79. Letterio JJ, Roberts AB. Regulation of immune responses by TGF-β. Annu Rev Immunol 1998; 16:137–61.
80. Feng JF, Readon M, Yadav SP, Im MJ. Calreticulin down-regulates both GTP binding and transglutaminase activities of transglutaminase II. Biochemistry 1999; 38:10743–9.
81. Krupickova S, Tuckova L, Flegelova Z, et al. Identification of common epitopes on gliadin, enterocytes, and calreticulin recognised by antigliadin antibodies of patients with coeliac disease. Gut 1999; 44:168–73.
82. Nikoshkov A, Falorni A, Lajic S, et al. A conformation-dependent epitope in Addison's disease and other endocrinological autoimmune diseases maps to a carboxyl-terminal functional domain of human steroid 21-hydroxylase. J Immunol 1999; 162:2422–6.
83. Kohno Y, Yamaguchi F, Saito K, et al. Anti-thyroid peroxidase antibodies in sera from healthy subjects and from patients with chronic thyroiditis: Differences in the ability to inhibit thyroid peroxidase activities. Clin Exp Immunol 1991; 85:459–63.
84. Schwartz HL, Chandonia JM, Kash SF, et al. High-resolution autoreactive epitope mapping and structural modeling of the 65 kDa form of human glutamic acid decarboxylase. J Mol Biol 1999; 287:983–99.
85. Sollid LM, Scott H. New tool to predict celiac disease on its way to the clinics. Gastroenterology 1998; 115:1584–6.
86. Halttunen T, Maki M. Serum immunoglobulin A from patients with celiac disease inhibits human T84 intestinal crypt epithelial cell differentiation. Gastroenterology 1999; 116:566–72.
87. Dieterich W, Laag E, Bruckner-Tuderman L, et al. Antibodies to tissue transglutaminase as serologic markers in patients with dermatitis herpetiformis. J Invest Dermatol 1999; 113:133–6.
88. Meadows L, Wang W, Den HJ, et al. The HLA-A*0201-restricted H-Y antigen contains a posttranslationally modified cysteine that significantly affects T cell recognition. Immunity 1997; 6:273–81.
89. Skipper JC, Hendrickson RC, Gulden PH, et al. An HLA-A2-restricted tyrosinase antigen on melanoma cells results from posttranslational modification and suggests a novel pathway for processing of membrane proteins. J Exp Med 1996; 183:527–34.
90. Selby M, Erickson A, Dong C, et al. Hepatitis C virus envelope glycoprotein E1 originates in the endoplasmic reticulum and requires cytoplasmic processing for presentation by class I MHC molecules. J Immunol 1999; 162:669–76.
91. Mamula MJ, Gee RJ, Elliott JI, et al. Isoaspartyl post-translational modification triggers autoimmune responses to self-proteins. J Biol Chem 1999; 274:22321–7.
© 2000 Lippincott Williams & Wilkins, Inc.