European Journal of Internal Medicine
Volume 19, Issue 2 , Pages 83-91, March 2008

Transglutaminase and the pathogenesis of coeliac disease☆

  • Pål Stenberg

      Affiliations

    • Hospital Pharmacy, Malmö University Hospital, S-205 02 Malmö, Sweden
    • Corresponding Author InformationCorresponding author. Tel.: +46 40 33 37 85; fax: +46 40 33 62 32.
    • ,
  • E. Bodil Roth

      Affiliations

    • Hospital Pharmacy, Malmö University Hospital, S-205 02 Malmö, Sweden
    • Department of Medicine, Lund University, Malmö University Hospital, Malmö, Sweden
    • ,
  • Klas Sjöberg

      Affiliations

    • Department of Medicine, Lund University, Malmö University Hospital, Malmö, Sweden

Received 8 December 2006; received in revised form 9 May 2007; accepted 10 May 2007. published online 22 November 2007.

Article Outline

Abstract 

In 1997, a German group demonstrated that the antigen of the biomarker EMA (endomysial antibodies) in coeliac disease is a calcium-dependent thiol enzyme, transglutaminase type 2 (TG2). This most important discovery opened up an exciting field of research aimed at a better understanding of the pathogenesis of coeliac disease, a T-cell-driven autoimmune disorder with a prevalence of about 1%. The accidental activation of TG2, possibly caused by a stress-induced local deficiency of zinc in the intestinal wall, might play a key role where the enzyme catalyzes an atypical deamidation of specific glutamine residues of food gliadins. The genetic contribution is HLA DQ2 or DQ8, which can form a complex with the TG2-modified gliadin residues, resulting in an immune response with the formation of antibodies against both gliadin and the enzyme. Indeed, the immunopathogenesis of coeliac disease can now be recognized partly at the molecular level. Progress has already improved the opportunities for laboratory diagnostics and, hopefully, new ways of treating and preventing coeliac disease will become available. These exciting developments might stimulate research within other fields of autoimmune disorders. With its focus on TG2, this review highlights some of the intriguing mechanisms of the pathogenesis of coeliac disease, such as the structure of the neo-antigen, the involvement of calcium and zinc, and the effects of coeliac antibodies on TG2 activity. Moreover, the many pitfalls due to dubious laboratory practice are addressed, as is the potential when a fundamental biological mechanism is understood at the molecular level.

Keywords: Coeliac disease, Transglutaminase, Autoimmune, Pathogenesis, Good laboratory practice, Zinc

 

Back to Article Outline

1. Introduction 

In spite of major efforts, the pathogenesis of autoimmune diseases has remained a mystery to science. Presently, a T-cell-driven intestinal disorder with a specific HLA requirement, coeliac disease (CD) (gluten intolerance; coeliac sprue), and an enzyme, transglutaminase 2 (TG2), also called tissue transglutaminase, are attracting the attention of researchers within the field of autoimmunity. Interest was first piqued in 1997 when Dieterich et al. [1] reported that the antigen of the biomarker EMA (endomysial antibodies) in CD is TG2. Suddenly, gastroenterologists found themselves becoming acquainted with an enzyme that they had hardly heard of, while people involved in TG research turned their attention to CD.

Today, it is possible to understand parts of the pathogenesis of CD at the molecular level. Moreover, CD is becoming a model for elucidating the mechanisms involved in other autoimmune diseases. By focusing on TG2, this review highlights some of the intriguing mechanisms involved in the pathogenesis of CD.

Back to Article Outline

2. Transglutaminases 

At the genomic level, eight members of the transglutaminase family (E.C. 2.3.2.13) have been identified [2]. Of these, six isoenzymes have been isolated and characterized as calcium-dependent thiol enzymes. In addition, a TG-like protein is found in red blood cells [3]. Since the position of the active site cysteine is replaced by an alanine residue, this protein, called erythrocyte membrane protein band 4.2, has no enzyme activity. Instead, it constitutes a major component of the red blood cell membrane.

A marked homology exists between the various types of mammalian TGs, and the amino acids surrounding the active site cysteine are all the same (GQCWV). There is an 80% homology between human and guinea pig TG2 [4]. All TGs can catalyze the formation of intermolecular isopeptide bonds between γ-carbonyl groups of glutamine residues and ɛ-amino groups of lysine residues (Fig. 1). Plasma factor XIII-catalyzed cross-linking of fibrin [5] and terminal keratinocyte differentiation by keratinocyte and epidermal TGs [6] illustrate the physiological roles of TGs. Furthermore, TG2 can function as a cell signal transducer by binding GTP, which promotes the activation of the α1B-adrenoreceptor [7]. At present, however, the physiological consequences remain unclear.

  • View full-size image.
  • Fig. 1. 

    Transglutaminase-catalyzed reactions. During calcium-induced activation, the concealed thiol group of the active site cysteine of the zymogen is unmasked and becomes reactive. A thio-ester intermediate is formed between the activated thiol and a specific glutamine residue in, for example, gliadin. In the presence of a protein-bound lysine residue, transamidation takes place. In the absence of an amine, hydrolysis occurs. In transamidation, the formation of the thio-ester is rate-limiting. Conversely, in hydrolysis, the second step is rate-limiting, making the concentration of the intermediary thio-ester comparatively high.

Back to Article Outline

3. Transglutaminase 2 — biological aspects 

A catalytic triad of amino acids (Cys-277, His-335, Asp-358) has been suggested as constituting a major part of the active site of TG2 [2]. Moreover, Trp-241 seems essential [8]. This ubiquitous enzyme, believed to be involved in the pathogenesis of CD, has been detected intracellularly [9], on the cell surface [10], and in the extracellular matrix [11], indicating a wide panorama of possible biological activities. The introduction of TG2 knockout mice has improved the potential for mapping physiological functions [12]. Mice lacking TG2 reproduce and show no significant developmental or histological abnormalities of major organs, including the small intestine [12]. Initially, an impaired glucose-stimulated insulin secretion was reported in these mice [13]. Later studies have focused on the role of TG2 in apoptosis, where macrophages perform the recognition, binding, and internalization of apoptotic cells without causing an inflammatory response, thanks to the release of TGF-β1. Interestingly, TG2 seems to be involved in the activation of this cytokine. Specifically, a lack of TG2 seems to result in an impaired capacity of macrophages to engulf apoptotic cells, which leads to an abnormal inflammatory response [14]. Because other TGs are able to take over the role of the tissue enzyme in this animal model, one cannot exclude the possibility that the physiological significance of TG2 may be underestimated. For example, monocytes in a cell culture contain factor XIII exclusively while the transformed macrophage is crowded with TG2 but without factor XIII [15]. Indeed, in fresh human macrophages collected via lung lavage, only TG2 activity can be detected. With substrates such as fibronectin and collagens, TG2 has also been associated with the stabilization of extracellular matrix proteins. Synthetic, low-molecular weight primary amines such as monodansylthiacadaverine and 5-dibenzylaminopentylamine are efficient pseudosubstrates for TGs, but these cross-linking inhibitors are also fairly toxic, thus supporting a vital role for TGs in general.

Back to Article Outline

4. Transglutaminase 2 — pathological aspects 

TG2 has also been associated with a series of pathological conditions. As early as 1966, Laki et al. [16], using the plasma cell tumor YPC-1 in a mouse model, reported a correlation between the extent of metastasis and TG activity of the organ. Interestingly, decades later, similar concepts, although more sophisticated, are still under investigation [17]. Furthermore, TG2 inhibitors have been tested in animal models against CAPD-induced peritoneal sclerosis [18] and against fibrosis of the kidney [19] in progressive nephropathy. In addition, TG2 may be involved in cataractogenesis [20], [21] and in neurological disorders such as Huntington's disease [22]. However, to our knowledge, TG2 has not been implicated in autoimmune diseases before 1997.

Back to Article Outline

5. Transglutaminases — basic research 

Most of the basic research on TGs was performed in the USA during the 1960–70s. J. E. Folk and S. I. Chung at NIH focused on TG2 while Laszlo Lorand at Northwestern University initially studied factor XIII. In the early 1970s, Lorand's group designed synthetic, low-molecular weight substrates that enabled the development of simple, specific, and sensitive continuous assays for TGs. Using β-phenylpropionylthiocholine iodide as the substrate substituting for the glutamine residue, transamidation could be studied in the presence of the versatile fluorescent monodansylcadaverine [23] while hydrolysis was assayed by monitoring the thiol group of the released thiocholine [24]. Using these assays, the activation [24], [25] and inhibition of TGs were studied and the enzyme kinetics were mapped. Zinc, in particular, was shown to compete with calcium for the enzyme metal binding site and thus served as a most potent inhibitor of the activation of TGs [26]. Interestingly, more than 30 years later, these results may be fundamental for understanding the pathogenesis of CD.

Back to Article Outline

6. Coeliac disease 

If the basic scientific contribution to understanding TG was done by an American, the CD connection has thus far been mainly European. Indeed, only recently, the FDA recognized CD as a health problem and initiated a task force to study this disease. The clinical evolution of a successful treatment for CD has been reviewed by Charlotte Anderson [27]. In 1887 in London, Samuel Gee described the condition that we now refer to as CD. For many years after Gee's report, much attention was focused on the mechanism of fat malabsorption. Due to steatorrhea, removal of fat from the diet was recommended, although it obviously did not cure the disease. Not until the late 1940s did the Dutch pediatrician Wim Dicke recognize for the first time the deleterious effects of wheat flour. In 1950, Dicke and co-workers submitted a paper on this issue to a well-known American pediatric medical journal. The manuscript was rejected and returned without comments to the authors but later published in a Scandinavian journal [28]. In his review, Marsh [29] also acknowledges other early contributions to this field. For example, it has been known for a long time that deamidated food gliadin peptides do not trigger the immune system in CD [30], [31], [32], nor does glutamine as a free amino acid. In 1985, Bruce at al. reported increased TG activity in duodenal biopsies taken from patients with CD as compared with normal subjects [33]. Later studies of CD have confirmed that the expression of TG2 is upregulated in the subepithelial lamina propria [34], [35], the tissue believed to be the primary site of action for TG2 in CD.

Shortly after the significant achievement by Dieterich et al., several groups reported that TG2, in a given order, can catalyze the deamidation of specific glutamine residues in proline-rich areas of gliadins to glutamic acids [36], [37]. In particular, a 33-mer fragment of residues 57 to 89 of α2-gliadin has attracted much attention [38]. This peptide (LQLQPFPQPQLPYPQPQLPYPQPQLPYPQPQPF) is rich in glutamines and prolines and comparatively resistant to intestinal proteolytic degradation. Q denotes glutamine residues that are targets for TG2-catalyzed deamidation. After studies on a number of peptides containing glutamine, Vader et al. [39] derived an algorithm to predict the sequences with glutamine residues as targets for TG2-catalyzed deamidation. Sequence patterns such as QXP and QXXF especially seem to favor this deamidation.

Back to Article Outline

7. The effects of calcium and zinc on the affinity between TG2 and CD antibodies 

The discovery by Dieterich et al. launched the introduction of several ELISAs for the detection of IgA-anti-TG2. Most reports indicated that methods based on recombinant human TG2 (rh-TG2) performed more specifically than those using guinea pig liver TG as the antigen [40]. Unfortunately, information about the composition of the commercial kits is limited, for commercial reasons. Similarly, in too many cases, the descriptions of the in-house ELISAs do not cover essential aspects. Furthermore, information about the way the enzymes were dissolved, stored, etc. is sometimes omitted, which certainly complicates interpretation of the data [41]. In some way, there seems to be a cultural difference between immunology and biochemistry. While the immunologist, in order to increase antigenicity, does not hesitate to use conditions far from physiological, the biochemist is anxious to keep the protein as natural as possible. In many cases, the differences may not affect results. However, when the aim is to understand the role of an antigen in a process, it is important to establish the exact nature of that protein. When the antigen is an enzyme, the best way to validate the integrity of the structure is to ensure full enzymatic activity. However, degraded TGs sometimes display activities as well [41]. In such cases, activity staining can be combined with electrophoresis under non-denaturing conditions [42].

Since TGs are calcium-dependent, some laboratories added calcium to the antigen during the coating procedure of the ELISA plates. Interestingly, the results were conflicting. While some groups found an increased affinity between calcium-treated TG2 and CD antibodies, others did not. Guinea pig liver TG and rh-TG2 were used as antigens and, again, differences in species were claimed to explain the different results. A more appropriate explanation would seem to involve the way in which the enzyme was handled before and during the experiments. When dissolved and stored under suboptimal conditions, TG2 is extremely fragile and rapidly loses its potential enzyme activity [41].

With most of the commercial ELISA kits used for the diagnosis of CD, the addition of calcium would probably not improve the results simply because the TG2 is not in a native state when being coated onto the plates. However, when good laboratory practice is applied, the addition of calcium to freshly dissolved native enzyme during the coating procedure will increase the affinity between TG2 and the CD antibodies several-fold (Fig. 2).

  • View full-size image.
  • Fig. 2. 

    The effect of calcium and zinc ions on the affinity between TG2 and antibodies from healthy controls (C1–C3) and patients with coeliac disease (P1–P7) expressed as relative absorbance in an ELISA. Calcium chloride and zinc acetate were added during the coating of the antigen (guinea pig liver TG) onto the ELISA plates. As can be seen, increasing concentrations of zinc ions reduce the calcium-dependent affinity in a dose-dependent manner.

With radio-bound immunoassays, the situation turned out to be even more complex. In these experiments, radio-labeled rh-TG2 is expressed with a commercial kit based on a lysate of rabbit reticulocytes. Previous studies had indicated that calcium reduces the affinity between rh-TG2 and CD antibodies [43]. The explanation for this rather astonishing effect was the discovery of TG2 in the kit originating from the rabbit reticulocytes [44]. Apparently, this calcium-activated native TG2 competed successfully with the radio-labeled rh-TG2 in the assay. Obviously, when using kits based on a lysate from rabbit reticulocytes in order to express human proteins, native proteins originating from the host cell may jeopardize the results.

The human body contains 2–3 g of zinc with relatively high concentrations in skin, nails, hair, Langerhans' islets, and in specific parts of the prostate gland and the eye. The homeostasis of zinc is maintained mainly in the gastrointestinal tract. While hyperzincemia is a rare condition, zinc deficiency is a major problem in several developing countries [45]. Naturally, all trace metals are reversibly bound to proteins in the body and there are several carriers for zinc. Obviously, the metal will bind to the structure that offers the highest affinity at that moment. This is illustrated by the rapid rearrangement of zinc in animals during experimental conditions when an infection is induced [46]. About 300 enzymes, many of which are part of the immune system, require zinc for full activity. Conversely, reports on zinc-induced inhibition of physiological enzyme systems are rare. However, zinc is a potent inhibitor of the calcium-induced activation of TGs [26]. Indeed, zinc may be a physiological moderator of the calcium-induced activation of these enzymes. Interestingly, zinc also abolishes the calcium-induced increase in affinity between TG2 and CD antibodies (Fig. 2) [41]. Obviously, these findings make zinc a potential agent for treatment and prophylaxis of CD. One must, however, bear in mind that this type of result would probably not be possible to achieve with denatured proteins that lack potential TG2 activity.

Back to Article Outline

8. The structure of the neo-antigen giving rise to antibodies against both TG2 and gliadin 

In serum from patients with CD, antibodies can be detected against both TG2 and gliadins. A rational explanation is the formation of a complex between TG2 and gliadin, constituting the neo-antigen. Gliadins are rich in glutamine residues. Thus, in a TG2-catalyzed reaction, gliadins can serve as the first substrate but cannot be cross-linked due to the lack of lysine residues. On the other hand, when purified TG2 is incubated with calcium in vitro, a complex is formed with at least partly retained enzyme activity. This is true also in the presence of a reducing agent, clearly showing that the complex is not based on the formation of disulfide bonds. Evidently, the TG2 molecule contains glutamine as well as lysine residues that might serve as substrates in a self-catalyzed intermolecular cross-linking. In the same way, it is possible for TG2 to catalyze the formation of ε-(γ-glutamyl)lysine bridges between the enzyme and gliadin. In fact, this is the type of complex that has been proposed by most groups. Obviously, a key issue is whether these types of cross-linking, including the enzyme itself, can take place in vivo during physiological conditions. Recently, Dieterich et al. [47] have shown that specific gliadins can be cross-linked to certain extracellular matrix proteins such as α-chains of collagen type I or III. Moreover, it is generally accepted that TG2, in a given order, can deamidate specific glutamine residues in proline-rich areas of gliadins. The negatively charged carboxylate residues in the deamidated gliadins are then able to fit into the HLA DQ2 or DQ8 molecule, explaining the genetic specificity of CD.

Let us now scrutinize these arguments. Although it seems plausible that TG2 in vivo, depending on the conditions, can catalyze two types of reactions, deamidation and transamidation, the kinetic information available tells us that transamidation is by far the faster reaction. Consequently, we would expect the enzyme to be cross-linked to a gliadin residue, and the specific glutamines in the complex to then be deamidated in a new TG2-catalyzed cycle. Although not completely out of the question, we should bear in mind that studies on the specificity of deamidation of certain glutamine residues have used gliadin – and not a complex between gliadin and TG2 – as the substrate.

In 2003, we postulated another structure of the neo-antigen [41]. All TG-catalyzed reactions proceed via a thio-ester intermediate, formed between the active site thiol of the enzyme and a glutamine residue of the substrate (Fig. 1). In the presence of a primary amine, transamidation will take place, but in the absence of an amine, or at a lowered pH, such as in a site of inflammation, deamidation will be the result of the reaction. In transamidation, the first step, i.e., the formation of the thio-ester intermediate, is rate-limiting [23]. In contrast, in hydrolysis, the second step is rate-limiting, meaning that the life span of the intermediate might be rather long. Consequently, we believe that the neo-antigen is constituted by the intermediate thio-ester formed between the TG2 and a previously deamidated gliadin molecule.

In conclusion, from an enzyme kinetic point of view, it seems difficult to accept ɛ-(γ-glutamine)lysine as the bridge keeping TG2 and gliadin together in the neo-antigen. Instead, there are good reasons to believe that an intermediate in the reaction, the thio-ester between the enzyme and a previously deamidated gliadin molecule, constitutes the neo-antigen.

Back to Article Outline

9. The effect of CD antibodies on the activity of TG2 

The role of Ig-anti-TG2 in CD has yet to be elucidated. A tempting theory would be that the antibodies inhibit TG2 in an attempt to ameliorate the damage initiated by the enzyme. Another hypothesis would be that inhibition of TG2 by the antibodies would worsen the situation and thus contribute to the pathogenesis. Several groups have addressed this issue, interestingly with divergent results. Esposito et al. [48] utilized a lysate of canine kidney cells containing rh-TG2, but possibly also contaminating native TG2 from the canine host cells, and studied the effects of purified IgA and IgG from serum of CD patients as well as human monoclonal anti-TG2 antibodies derived from intestinal lymphocytes of CD patients. TG activity was assayed with N,N-dimethylated casein and tritiated spermidine as the substrate pair. After pre-incubation of the enzyme with the various immunoglobulins for 10 min at room temperature with or without calcium, the assay was performed under rather harsh conditions at pH 8.0 at 37° for 1 h. The results indicated a moderate inhibition of TG activity by purified CD IgA, IgG, and human anti-TG2 monoclonal antibodies. The presence of calcium during the pre-incubation did not affect the results.

In another approach, this time with purified rh-TG2, Dieterich et al. [49] used two different assays, the first based on the increased intensity of fluorescence when monodansylcadaverine is incorporated into casein, and the second based on the incorporation of biotinylated cadaverine into gliadin, followed by SDS-PAGE. Compared with controls, enriched IgA and IgG fractions from five CD patients resulted in insignificant differences of inhibition of TG2 activity while affinity-purified anti-TG2 antibodies from CD patients resulted in a significant, dose-dependent partial reduction in enzyme TG activity. Pre-incubation of TG2 with 5 mM calcium chloride had no influence on the enzyme inhibition.

Furthermore, Roth et al. [41], in a pilot experiment, used guinea pig liver TG as the source of enzyme. After one-way agarose gel electrophoresis of the enzyme, areas with TG2 activity were transformed to another agarose gel containing 10% human serum from a CD patient. After this crossed immuno-electrophoresis in the presence of EDTA, the gel was stained for TG activity using monodansylthiacadaverine and casein as substrates. The immunoprecipitate formed between TG2 and antibodies from the CD patient displayed full TG activity (Fig. 3).

  • View full-size image.
  • Fig. 3. 

    The effect of coeliac antibodies on TG2 activity illustrated by a crossed immunoelectrophoretogram stained for TG activity [41]. After one-way electrophoresis of commercial guinea pig liver TG, areas with TG activity were transferred to another gel containing 10% human serum and a new electrophoresis was performed. a: Serum from a patient with CD; b: normal serum. The immunoprecipitate from the coeliac patient displays full TG activity. (Printed with permission from Taylor & Francis.)

Finally, in a recent study, Byrne et al. [50], using an ELISA technique, measured the binding of IgA and IgG anti-TG2 from CD patients to a mutagenic variant of TG2 expressed in E. coli and lacking the catalytic triad. Interestingly, the affinity for IgA-anti-TG2 to the mutant TG2 was strongly reduced compared with the binding to the wild type TG2, while the affinity for IgG anti-TG2 was not affected by the mutation.

When conclusions differ between various groups that have addressed a specific issue, the laboratory details may offer an explanation. Clearly, there are numerous ways to perform these experiments and many pitfalls to consider. Nevertheless, it is easy to agree with Dieterich et al. when they conclude that the residual TG activity, after being exposed to CD-anti-TG2, remains sufficiently high to cast doubt on the in vivo relevance of these antibodies. The TG2 epitopes formed in CD are probably not located at the active site of the enzyme. One possible explanation is that the active site may be concealed from the immune system in vivo, for example, via the formation of an intermediary thio-ester with the substrate, as suggested by Roth et al. [41].

Back to Article Outline

10. The pathogenesis of coeliac disease — a hypothesis 

The prevalence of CD is approximately 1% in the western world [51]. We know that HLA DQ2 or, to some extent, HLA DQ8 is required but not sufficient for the development of this disease [52]. About 30% of Caucasians carry one of these haplotypes. Needless to say, gluten is also required. Moreover, the way food containing gluten is introduced to infants is of utmost importance. This is illustrated by an unexpected dramatic threefold increase in the incidence of CD among Swedish infants less than 2 years of age, starting in the mid-1980s, when recommendations for the amount of gluten in food and when it should be introduced were inappropriately changed. At that time, it was suggested that the amount of gluten in infant formula should be increased. Moreover, mothers were recommended to introduce gluten in large amounts when the child was 6 months old. When the recommendations were changed back 10 years later, the incidence normalized. Based on this unintentional, huge epidemiologic experiment, it can be concluded that terminated breast feeding and higher amounts of gluten, possibly combined with infection at introduction, drastically increased the risk of CD [53].

Based on the data available, we would suggest the pathways for the initiation of CD shown in Fig. 4.

Back to Article Outline

11. Other autoimmune diseases 

In dermatitis herpetiformis, IgA against epidermal TG has been detected [54]. The parallel with other autoimmune diseases is also clear. In these cases, antibodies can be detected both against a modified substrate and against the enzyme catalyzing the modification. In rheumatoid arthritis (RA), IgG against citrulline – a deiminated form of arginine – has been observed years before the clinical symptoms are obvious and is now becoming a specific marker for RA. In RA, antibodies against the calcium-dependent thiol enzyme that catalyzes the deimination, peptidylarginine deiminase (PAD,) can be demonstrated as well [55]. Treatment with methotrexate reduces the expression of antibodies against PAD. Interestingly, like TG2, PAD is also inhibited by zinc [56]. Similarly, in type 1 diabetes, antibodies can be detected against another calcium-dependent thiol enzyme, glutamic acid decarboxylase (GAD).

Back to Article Outline

12. Pharmacological interventions of transglutaminase-catalyzed processes 

Thus far, the only established treatment for CD is a lifelong, gluten-free diet, which certainly might be problematic for patients. Thanks to our increased knowledge of the pathogenesis of CD, pharmacological interventions are now being explored. Specifically, inhibition of TG2 has been focused upon. Previously, based on Lorand's original idea [57], efforts had been made to develop factor XIII inhibitors as potential anti-thrombotic agents. Irreversible inhibition of TG-catalyzed fibrin cross-linking is easily achieved in vitro by pseudosubstrates of the amine type, such as monodansylcadaverine [58]. Similarly, pseudosubstrates of the glutamine type, e.g., esters of thiocholine, also inhibit fibrin stabilization [59]. One possible pitfall with this strategy is that the pseudosubstrate would be incorporated covalently into the physiological substrates with unknown immunological consequences. Furthermore, the specificity is limited, and doubtless reactions catalyzed by TG2 are affected by these pseudosubstrates as well. Nonetheless, these efforts have provided excellent tools for characterization of TGs.

Disulfides are another type of TG inhibitor. The basic idea is that the reactive, active site thiol of the enzyme will undergo a thiol-disulfide interchange reaction whereby the enzyme will be inactivated and become part of a new disulfide. Interestingly, cystamine (H2N–CH2CH2S–SCH2CH2–NH2) has been tested clinically in Huntington's disease [22]. This amine is both a pseudosubstrate of TGs and a disulfide with potentially irreversible inhibition capacity. In order to increase specificity, monodansylated cystamine has also been designed.

In other attempts to design specific, irreversible TG2-inhibitors, a series of peptidic derivatives with a halogenated dihydroisoxazole group has been tested. The rationale is that the peptidic backbone will anchor the substance in the vicinity of the active site and that the dihydroisoxazole group will then alkylate the active site thiolate with the halo atom serving as a leaving group. The result will be an inactive TG2 molecule with a fairly bulky group attached to the active site cysteine thiol. Although interesting from a scientific point of view, this approach raise many questions. Will the human immune system recognize the alkylated enzyme? Will specificity be high enough to protect other TGs from inhibition? Will other TGs take over the function if TG2 is irreversibly inhibited? And, finally, when considering the extending panorama of physiological functions of TG2, what about long-term toxicity?

In CD, another way of achieving a specific effect with no adverse reactions is to prevent activation of TG2 in the intestinal tract. Since zinc is a very potent, competitive inhibitor of calcium-induced activation of TGs, zinc-fortified flour might be an exciting way to treat and even prevent CD.

Furthermore, clinical trials are in progress with low doses of GAD65 in type 1 diabetes in order to induce immune tolerance [60]. If this attempt is successful, the same principle can be applied to other autoimmune diseases. Obvious candidates to achieve tolerance are antigens such as TG2 in CD, TG3 in dermatitis herpetiformis, and PAD2 or PAD4 in RA.

Back to Article Outline

13. Conclusions 

Thanks to a very important discovery by Dieterich et al. in 1997 [1], the pathogenesis of CD has been studied extensively in recent years. Today, the mechanisms can be recognized partly at the molecular level. As always, good laboratory practice is essential for progress. The accidental activation of TG2, a post-translationally active calcium-dependent thiol enzyme that catalyzes an atypical deamidation of specific glutamine residues in food gliadins, seems to play a crucial role. The genetic contribution is HLA DQ2 or DQ8, which can form a complex with the TG2-modified gliadin residues, resulting in an immune response. Progress has already improved the opportunities for laboratory diagnostics and, hopefully, new ways of treating and preventing CD will become available. Moreover, these exciting development may stimulate progress in other fields of autoimmune diseases.

Back to Article Outline

14. Learning points 


•The major antigen of the biomarker endomysial antibodies (EMA) in coeliac disease (CD) is a ubiquitous, calcium-dependent thiol enzyme, tissue transglutaminase (TG2).

•The pathogenesis of CD is now recognized partly at the molecular level, and CD is becoming a model for understanding the pathogenesis of other autoimmune diseases.

•Physiological levels of zinc inhibit the calcium-induced activation of TG2 and reduce the affinity between TG2 and coeliac antibodies.

•Activated TG2 catalyzes an atypical deamidation of specific glutamine residues. An abnormally long-lived intermediate between TG2 and partially deamidated gliadin can activate T cells in persons with HLA DQ2 or DQ8.

•The immune reaction generates antibodies against both TG2 and gliadin.

•Good laboratory practice is essential for this type of study.

Back to Article Outline

References 

  1. Dieterich W, Ehnis T, Bauer M, et al. Identification of tissue transglutaminase as the autoantigen of coeliac disease. Nat Med. 1997;3:797–801
  2. Beninati S, Piacentini M. The transglutaminase family: an overview: minireview article. Amino Acids. 2004;26:367–372
  3. Korsgren C, Lawler J, Lambert S, et al. Complete amino acid sequence and homologies of human erythrocyte membrane protein band 4.2. Proc Natl Acad Sci U S A. 1990;87:613–617
  4. Aeschlimann D, Paulsson M. Transglutaminases: protein cross-linking enzymes in tissues and body fluids. Thromb Haemost. 1994;71:402–415
  5. Lorand L. Factor XIII and the clotting of fibrinogen: from basic research to medicine. J Thromb Haemost. 2005;3:1337–1348
  6. Hitomi K. Transglutaminases in skin epidermis. Eur J Dermatol. 2005;15:313–319
  7. Im MJ, Russell MA, Feng JF. Transglutaminase II: a new class of GTP-binding protein with new biological functions. Cell Signal. 1997;9:477–482
  8. Murthy SN, Iismaa S, Begg G, et al. Conserved tryptophan in the core domain of transglutaminase is essential for catalytic activity. Proc Natl Acad Sci U S A. 2002;99:2738–2742
  9. Lesort M, Attanavanich K, Zhang J, Johnson GV. Distinct nuclear localization and activity of tissue transglutaminase. J Biol Chem. 1998;273:11991–11994
  10. Martinez J, Chalupowicz DG, Roush RK, et al. Transglutaminase-mediated processing of fibronectin by endothelial cell monolayers. Biochemistry. 1994;33:2538–2545
  11. Upchurch HF, Conway E, Patterson MK, Maxwell MD. Localization of cellular transglutaminase on the extracellular matrix after wounding: characteristics of the matrix bound enzyme. J Cell Physiol. 1991;149:375–382
  12. De Laurenzi V, Melino G. Gene disruption of tissue transglutaminase. Mol Cell Biol. 2001;21:148–155
  13. Bernassola F, Federici M, Corazzari M, et al. Role of transglutaminase 2 in glucose tolerance: knockout mice studies and a putative mutation in a MODY patient. Faseb J. 2002;16:1371–1378
  14. Falasca L, Iadevaia V, Ciccosanti F, et al. Transglutaminase type II is a key element in the regulation of the anti-inflammatory response elicited by apoptotic cell engulfment. J Immunol. 2005;174:7330–7340
  15. Seiving B, Ohlsson K, Linder C, Stenberg P. Transglutaminase differentiation during maturation of human blood monocytes to macrophages. Eur J Haematol. 1991;46:263–271
  16. Laki K, Tyler HM, Yancey ST. Clot forming and clot stabilizing enzymes from the mouse tumor YPC-1. Biochem Biophys Res Commun. 1966;22:776–781
  17. Mann AP, Verma A, Sethi G, et al. Overexpression of tissue transglutaminase leads to constitutive activation of nuclear factor-{kappa}b in cancer cells: delineation of a novel pathway. Cancer Res. 2006;66:8788–8795
  18. Skill NJ, Griffin M, El Nahas AM, et al. Increases in renal epsilon-(gamma-glutamyl)-lysine crosslinks result from compartment-specific changes in tissue transglutaminase in early experimental diabetic nephropathy: pathologic implications. Lab Invest. 2001;81:705–716
  19. Skill NJ, Johnson TS, Coutts IG, et al. Inhibition of transglutaminase activity reduces extracellular matrix accumulation induced by high glucose levels in proximal tubular epithelial cells. J Biol Chem. 2004;279:47754–47762
  20. Lorand L, Hsu LK, Siefring GE, Rafferty NS. Lens transglutaminase and cataract formation. Proc Natl Acad Sci U S A. 1981;78:1356–1360
  21. Samuelsson B, Stenberg P, Pandolfi M. Localization and characteristics of transglutaminase in the rabbit and human eye. Graefes Arch Clin Exp Ophthalmol. 1982;218:233–236
  22. Karpuj MV, Becher MW, Springer JE, et al. Prolonged survival and decreased abnormal movements in transgenic model of Huntington disease, with administration of the transglutaminase inhibitor cystamine. Nat Med. 2002;8:143–149
  23. Stenberg P, Curtis CG, Wing D, et al. Transamidase kinetics. Amide formation in the enzymic reactions of thiol esters with amines. Biochem J. 1975;147:155–163
  24. Curtis CG, Stenberg P, Brown KL, et al. Kinetics of transamidating enzymes. Production of thiol in the reactions of thiol esters with fibrinoligase. Biochemistry. 1974;13:3257–3262
  25. Lorand L, Gray AJ, Brown K, et al. Dissociation of the subunit structure of fibrin stabilizing factor during activation of the zymogen. Biochem Biophys Res Commun. 1974;56:914–922
  26. Credo RB, Stenberg P, Tong YS, Lorand L. Inhibition of fibrinoligase and transglutaminase by zinc ions. In: Fed. Proc. San Fransisco. 1976;
  27. Marsh MN. Coeliac Disease. Oxford: Blackwell Scientific Publications; 1992;1–16 pp.
  28. Dicke WK, Weijers HA, Van De Kamer JH. Coeliac disease. II. The presence in wheat of a factor having a deleterious effect in cases of coeliac disease. Acta Paediatr. 1953;42:34–42
  29. Marsh MN. Gluten, major histocompatibility complex, and the small intestine. A molecular and immunobiologic approach to the spectrum of gluten sensitivity (‘coeliac sprue’). Gastroenterology. 1992;102:330–354
  30. Frazer AC. Discussion of some problems of steatorrhea and reduced stature. Proc R Soc Med. 1956;49:1009–1013
  31. Booth CC. The enterocyte in coeliac disease. Br Med J. 1970;4:14–17
  32. Van De Kamer JH, Weijers HA. Coeliac disease. V. Some experiments on the cause of the harmful effect of wheat gliadin. Acta Paediatr. 1955;44:465–469
  33. 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 (Lond). 1985;68:573–579
  34. Skovbjerg H, Hansen GH, Niels-Christiansen LL, et al. Intestinal tissue transglutaminase in coeliac disease of children and adults: ultrastructural localization and variation in expression. Scand J Gastroenterol. 2004;39:1219–1227
  35. Hansson T, Ulfgren AK, Lindroos E, et al. Transforming growth factor-beta (TGF-beta) and tissue transglutaminase expression in the small intestine in children with coeliac disease. Scand J Immunol. 2002;56:530–537
  36. Sjostrom H, Lundin KE, Molberg O, 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;48:111–115
  37. Molberg O, McAdam SN, Korner R, et al. Tissue transglutaminase selectively modifies gliadin peptides that are recognized by gut-derived T cells in coeliac disease. Nat Med. 1998;4:713–717
  38. Shan L, Molberg O, Parrot I, et al. Structural basis for gluten intolerance in coeliac sprue. Science. 2002;297:2275–2279
  39. Vader LW, de Ru A, van der Wal Y, et al. Specificity of tissue transglutaminase explains cereal toxicity in coeliac disease. J Exp Med. 2002;195:643–649
  40. Wong RC, Wilson RJ, Steele RH, et al. A comparison of 13 guinea pig and human anti-tissue transglutaminase antibody ELISA kits. J Clin Pathol. 2002;55:488–494
  41. Roth EB, Sjoberg K, Stenberg P. Biochemical and immuno-pathological aspects of tissue transglutaminase in coeliac disease. Autoimmunity. 2003;36:221–226
  42. Stenberg P, Stenflo J. A rapid and specific fluorescent activity staining procedure for transamidating enzymes. Anal Biochem. 1979;93:445–452
  43. Nakachi K, Swift G, Wilmot D, et al. Antibodies to tissue transglutaminase: comparison of ELISA and immunoprecipitation assay in the presence and in the absence of calcium ions. Clin Chim Acta. 2001;304:75–84
  44. Agardh D, Roth B, Lernmark A, Stenberg P. Calcium activation of tissue transglutaminase in radioligand binding and enzyme-linked autoantibody immunoassays in childhood coeliac disease. Clin Chim Acta. 2005;358:95–103
  45. Prasad AS. Zinc deficiency. BMJ. 2003;326:409–410
  46. Srinivas U, Ohlsson T, Hallstadius L, et al. Organ sequestration of (65)Zn during experimental sepsis. Clin Nutr. 1989;8:263–267
  47. Dieterich W, Esslinger B, Trapp D, et al. Cross linking to tissue transglutaminase and collagen favours gliadin toxicity in coeliac disease. Gut. 2006;55:478–484
  48. Esposito C, Paparo F, Caputo I, et al. Anti-tissue transglutaminase antibodies from coeliac patients inhibit transglutaminase activity both in vitro and in situ. Gut. 2002;51:177–181
  49. Dieterich W, Trapp D, Esslinger B, et al. Autoantibodies of patients with coeliac disease are insufficient to block tissue transglutaminase activity. Gut. 2003;52:1562–1566
  50. Byrne G, Ryan F, Jackson J, et al. Mutagenesis of the catalytic triad of tissue transglutaminase abrogates coeliac disease serum IgA autoantibody binding. Gut. 2006;
  51. van Heel DA, West J. Recent advances in coeliac disease. Gut. 2006;55:1037–1046
  52. Green PH, Jabri B. Coeliac disease. Lancet. 2003;362:383–391
  53. Ivarsson A. The Swedish epidemic of coeliac disease explored using an epidemiological approach—some lessons to be learnt. Best Pract Res Clin Gastroenterol. 2005;19:425–440
  54. Sardy M, Karpati S, Merkl B, et al. Epidermal transglutaminase (TGase 3) is the autoantigen of dermatitis herpetiformis. J Exp Med. 2002;195:747–757
  55. Roth EB, Stenberg P, Book C, Sjoberg K. Antibodies against transglutaminases, peptidylarginine deiminase and citrulline in rheumatoid arthritis—new pathways to epitope spreading. Clin Exp Rheumatol. 2006;24:12–18
  56. Kearney PL, Bhatia M, Jones NG, et al. Kinetic characterization of protein arginine deiminase 4: a transcriptional corepressor implicated in the onset and progression of rheumatoid arthritis. Biochemistry. 2005;44:10570–10582
  57. Lorand JB, Pilkington TR, Lorand L. Inhibitors of fibrin cross-linking: relevance for thrombolysis. Nature. 1966;210:1273–1274
  58. Lorand L, Parameswaran KN, Stenberg P, et al. Specificity of guinea pig liver transglutaminase for amine substrates. Biochemistry. 1979;18:1756–1765
  59. Ljunggren C, Hoffmann KJ, Stenberg P, Nilsson JL. Fibrin-stabilizing factor inhibitors. 10. Thiol esters as potent enzyme inhibitors. J Med Chem. 1973;16:1186–1188
  60. Agardh CD, Cilio CM, Lethagen A, et al. Clinical evidence for the safety of GAD65 immunomodulation in adult-onset autoimmune diabetes. J Diabetes Complications. 2005;19:238–246

☆ We hereby certify that there are no financial or other relationships that might lead to a conflict of interest. The manuscript has been read and approved by all authors.

PII: S0953-6205(07)00256-7

doi:10.1016/j.ejim.2007.05.012

European Journal of Internal Medicine
Volume 19, Issue 2 , Pages 83-91, March 2008

Article Tools

* Image Usage & Resolution