Presence of anti-‘tissue’ transglutaminase antibodies in inflammatory intestinal diseases: an apoptosis-associated event?

_Dear Editor_ Celiac disease (CD) is described as a permanent intolerance to gluten, characterised by a wide spectrum of lesions in the intestinal mucosa that can ultimately lead to the

atrophy of the villi.1 Celiac disease is brought about by ingestion of gliadin which triggers the production of IgA class serum tissue autoantibodies.1 Identification of the autoantigen(s)

has been the aim of researchers for many years since it has been hypothesised that autoimmune mechanisms are important in the pathogenesis of celiac disease.1,2,3 ‘Tissue’ transglutaminase

(tTG; E.C. 2.3.2.13) or type-II transglutaminase has been identified as the major auto-antigens of endomysial antibodies found in coeliac disease.3 tTG human genes belong to a family of

intracellular and extracellular enzymes which catalyse Ca2+-dependent reactions resulting in the post-translational modification of proteins by establishing ɛ(γ-glutamyl)lysine

cross-linkings and/or covalent incorporation of di- and polyamines into proteins.4,5,6 These covalent cross-links might determine the olygomerization of substrate protein(s) which acquire

the peculiar features of resistance to breakage and chemical attack.4,5,6 The tTG gene encodes a protein with a molecular weight of about 80 kDa4,5,6 which is normally expressed in cells

localised in peculiar sites within mammalian tissues (endothelial cells, smooth muscle cells and mesangial cells).5 tTG is a multifunctional protein which, in its protein cross-linking

configuration, plays a complex role in apoptosis.4,5,6,7 The enzyme has been shown to be induced and activated in apoptotic cells both _in vivo_ and _in vitro_ experimental models.4,5,6,7 It

has been proposed that tTG may play more than one function within the cascade of events that lead to the establishment of the apoptotic phenotype.4,8 Dieterich and colleagues3 have proposed

that tTG is the predominant, if not the only, endomysial autoantigen characteristic for celiac disease. However, as to whether the tTG-induction observed in CD patients represents a primary

pathogenic trigger in coeliac disease (CD) is still unknown. In humans, it has been shown that the CD antigens are localised in the extracellular matrix (ECM) and are expressed by

fibroblasts.1,9 Although tTG is an intracellular enzyme, previous studies have demonstrated the involvement of tTG in diseases characterised by the induction of apoptosis associated with a

drastic remodelling of the affected tissue, particularly in ECM.10,11 Considering that very little is known on role of cell death in the pathogenesis of degenerative diseases affecting the

intestinal mucosa, we have investigated the role of apoptosis in the pathogenesis of CD. To this aim we have studied tTG expression and anti-tTG IgA production in relationship with apoptosis

both in CD patients and in other intestinal pathologies, such as Crohn and Ulcerative rectocolitis (UC), showing various levels of mucosal damage and scarring. Table 1 gives the analysis of

anti-tTG IgA levels in sera obtained from patients affected by various intestinal ulcerative diseases as well as from other autoimmune pathologies. CD, Crohn and UC patients showed elevated

levels of circulating anti-tTG antibodies which were significantly higher in respect to both healthy individuals and other autoimmune diseases. However, both Crohn and UC patients presented

no detectable antiendomysial antibodies (EMA). The lower absolute levels of circulating anti-tTG antibodies detected in Crohn and UC patients compared with those found in CD patients could

be due to the minor extent of the lesions. To get direct information on the expression of tTG in CD pathogenesis, we investigated tTG protein localisation by a standard immunohistochemical

procedure on serial sections obtained from biopsy samples of the second portion of duodenum from the previously described group of patients. Considering that the enhanced apoptosis is

paralleled by the induction of tTG both in physiological and pathological settings,4,5,6,7 we have analysed whether the lesions occurring in CD, as well as in Crohn's diseases (data not

shown), are associated with an abnormal onset of apoptosis. In order to correlate the expression of tTG with the induction of apoptosis we have analysed DNA fragmentation _in situ_ by TUNEL

and the expression of apoptosis-related genes on consecutive sections from the same patient. An increased expression of the tTG gene as evidenced by an intense staining with the tTG

antibody was detected in all patients in the damaged areas compared to controls (Figure 1). In biopsies obtained from CD patients lesions where the architecture of the mucosa was still

preserved, the positivity to the tTG antibody was specifically confined to the enterocytes localised in the upper part of the villi and very limited staining was detected in the submucosa

(Figure 1, panel 1). Furthermore, tTG expression was also detected in fibroblasts lining the intestinal epithelium (Figure 1, panel 1). Comparative analysis of serial sections obtained from

the same patient (Figure 1, panels 1,3) showed that DNA fragmentation and tTG staining co-localised in post-mitotic enterocytes present in the upper part of the villi; by contrast, no DNA

fragmentation was observed in the lamina propria where the tTG-positive fibroblasts were detected. It is interesting to note that the enterocytes showing DNA fragmentation and tTG expression

also displayed a strong induction of CD95L (Figure 1, panel 5). CD95L immunostaining were also detected in activated lymphocytes localised in the lamina propria and in the submucosa (Figure

1, panel 5). The comparative analysis of cell morphology, tTG and CD95L expression and TUNEL staining demonstrates that apoptosis occurs in early stages of the diseases and primarily

affects the enterocytes. In fact, the tTG-positive fibroblasts localised in the submucosa do not show the biochemical and morphological features of cells undergoing apoptosis (Figure 1). In

the CD lesions with a complete flattening of the mucosa a different expression pattern was observed. The positivity to the tTG antibody was detected not only in cells, but predominantly in

the ECM (Figure 1, panel 2). In the same lesions, very few cells were found positive to both CD95L and TUNEL staining, thus indicating that the enterocyte undergoing apoptosis in the earlier

stages of the disease were probably already lost in the intestinal lumen. By contrast, the tTG-positive cells detected in the lesions appeared to be preferentially localised in the scarring

tissue in proximity of the gut lumen where these cells were releasing their contents into the ECM (Figure 1, panel 2). In the wound healing area close to the intestinal lumen, we also

detected an intense staining with a specific antibody recognising the ɛ(γ-glutamyl)lysine cross-link,11 thus indicating that the released enzyme was actively crosslinking proteins localised

in the ECM (Figure 1, panel 8). The data reported here indicate that the presence of anti-tTG antibodies is not a specific event occurring in CD patients, but is a general phenomenon related

to mucosal lesions rather than to the autoimmune nature of CD. In keeping with this hypothesis we detected both circulating anti-tTG and induction of apoptosis (data not shown) in other

pathologies leading to intestinal lesions such as Crohn and UC, but not in other specific and systemic autoimmune diseases (diabetes, thyroiditis, multiple sclerosis and LES). We would like

to hypothesise that the accumulation of tTG in the enterocytes as well as its release in the ECM are a consequences of the induction of apoptosis in those regions undergoing the destruction

typical of severe CD-associated lesions. In fact, the enterocytes become tTG-positive and show the morphological and biochemical features typical of apoptotic cells including the expression

of the CD95L in early stages of the diseases. Noteworthy, in CD patients showing a complete flattening of the mucosa, a large proportion of the tTG staining was detected in the ECM of the

lesion. Although defects in the CD95 pathway have been shown to play an important role in several autoimmune diseases,8,12 very little is known about the CD95/CD95L pathway in CD. Here we

present data suggesting the involvement of this receptor in the CD pathogenesis. Interestingly, a deregulated tTG is present in autoimmunity-prone MRL _lpr_/_lpr_ mice,13 where tTG is also

detected into the ECM similarly to the wound healing process and tissue repair occurring in CD lesions. Several studies have indicated that tTG, by crosslinking a number of proteins such as

collagens, fibronectin, laminin, nidogen and Transforming Growth Factor-1, might play an important role in the modification of the ECM occurring in degenerative diseases.11,14 The presence

of the tTG in the ECM in CD lesions might represent _per se_ an important pathogenic event. In fact, displaying of cryptic molecular determinants has been proposed to play an important role

in the induction of the pathogenic autoimmune response.8 The question arises as to how the activation of the tTG released in the ECM may lead to the formation of pathogenic

neo-determinants.8 As evidenced by the presence of high levels of ɛ(γ-glutamyl)lysine crosslinks compared to controls (data not shown), the extracellular tTG localised in CD lesions is

active. Hence, it is very likely that the tTG-mediated polymerisation of ECM proteins may generate new self-antigens and thus contributing to elicit the autoimmune responses. In keeping with

this assumption, the binding of gliadin to reticular matrix components is Ca2+-dependent, is inhibited by putrescine, and by preincubation with antibodies against tTG suggesting that

binding of gliadin might be a tTG-mediated event.15,16 Taken together these findings suggest an important complex role for tTG and apoptosis in the pathogenesis of CD REFERENCES * Maki M _et

al_. 1997 _Lancet_ 349: 1755–1759 * Ferreira M _et al_. 1992 _Gut_ 33: 1633–1637 Article  CAS  Google Scholar  * Dieterich W _et al_. 1997 _Nature Med_ 7: 797–801 * Melino G _et al_. 1998

_FEBS Lett._ 430: 59–63 * Fesus L _et al_. 1991 _Eur. J. Cell Biol._ 56: 170–177 * Piacentini M . 1995 _Curr. Top. Microbiol._ 200: 163–176 * Fesus L _et al_. 1989 _FEBS Lett._ 245: 150–154

* Colizzi V _et al_. 1999 _Immunol. Today_ 20: 134–138 * Marttinen A _et al_. 1997 _Eur. J. Clin. Invest._ 27: 135–140 * Piacentini M _et al_. 1999 _J. Pathol._ 189 1: 92–98 Article  CAS 

Google Scholar  * Johnson TS _et al_. 1997 _J. Clin. Invest._ 90: 2950–2960 Article  CAS  Google Scholar  * Thompson CB . 1995 _Science_ 267: 1456–1462 Article  CAS  Google Scholar  *

Piredda L _et al_. 1997 _Cell Death Differ._ 4: 463–472 Article  CAS  Google Scholar  * Mirza A _et al_. 1997 _Am. J. Physiol._ 272: G281–G288 * Maki M . 1996 _Lancet_ 348: 1046–1047 * Uhlig

H _et al_. 1998 _Autoimmunity_ 28: 185–195 Article  CAS  Google Scholar  * Picarelli A _et al_. 1996 _Lancet_ 348: 1065–1067 * Picarelli A . 2001 _J. Intern Med._ 249: 181–188 Download

references ACKNOWLEDGEMENTS This work was supported by grants from ‘AIDS’, Ricerca Corrente and Finalizzata Projects from the Ministero Sanita and UE. AUTHOR INFORMATION AUTHORS AND

AFFILIATIONS * Department of Biology, University of Rome, “Tor Vergata”, Italy M G Farrace & M Piacentini * Department of Clinical Science, University of Rome ‘La Sapienza’, Italy A

Picarelli, M Di Tola & L Sabbatella * Instituto ‘Proda’, Roma, Italy O P Marchione * IRCCS ‘L. Spallanzani’, Rome, Italy G Ippolito & M Piacentini Authors * M G Farrace View author

publications You can also search for this author inPubMed Google Scholar * A Picarelli View author publications You can also search for this author inPubMed Google Scholar * M Di Tola View

author publications You can also search for this author inPubMed Google Scholar * L Sabbatella View author publications You can also search for this author inPubMed Google Scholar * O P

Marchione View author publications You can also search for this author inPubMed Google Scholar * G Ippolito View author publications You can also search for this author inPubMed Google

Scholar * M Piacentini View author publications You can also search for this author inPubMed Google Scholar CORRESPONDING AUTHOR Correspondence to M Piacentini. RIGHTS AND PERMISSIONS

Reprints and permissions ABOUT THIS ARTICLE CITE THIS ARTICLE Farrace, M., Picarelli, A., Di Tola, M. _et al._ Presence of anti-‘tissue’ transglutaminase antibodies in inflammatory

intestinal diseases: an apoptosis-associated event?. _Cell Death Differ_ 8, 767–770 (2001). https://doi.org/10.1038/sj.cdd.4400880 Download citation * Issue Date: 01 July 2001 * DOI:

https://doi.org/10.1038/sj.cdd.4400880 SHARE THIS ARTICLE Anyone you share the following link with will be able to read this content: Get shareable link Sorry, a shareable link is not

currently available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative