Clinical Immunology (2006) 121, 40 — 46
available at www.sciencedirect.com
www.elsevier.com/locate/yclim
Evidence of a selective epitope loss of anti-transglutaminase immunoreactivity in gluten-free diet celiac sera: A new tool to distinguish disease-specific immunoreactivities C. Tiberti a,*, M. Bonamico b, F. Dotta c, A. Verrienti a, M. Di Tola a, E. Liu d, M. Ferri b, R. Nenna b, A. Picarelli a, G.S. Eisenbarth d a
Dept. of Clinical Sciences, University of Rome bLa SapienzaQ, Policlinico Umberto I, Viale del Policlinico 155, 00161 Rome, Italy b Pediatrics, University bLa SapienzaQ, Policlinico Umberto I, Viale del Policlinico 155, 00161 Rome, Italy c Depts. of Internal Medicine and of Endocrine and Metabolic Sciences, University of Siena, Italy d Barbara Davis Center for Childhood Diabetes, University of Colorado Health Sciences, Center, CO 80309-0482, USA Received 3 March 2006; accepted with revision 16 May 2006 Available online 23 June 2006
KEYWORDS Celiac disease; Human tissue transglutaminase; Epitopes; Autoimmunity
Abstract The aim of the present study was to evaluate the epitope specific humoral human tissue transglutaminase (tTG) immunoreactivity against 3 different human recombinant tTG constructs [(full-length tTG (a.a. 1—687), tTG (a.a. 227—687); tTG (a.a. 473—687)] before and after the introduction of a gluten-free diet (GFD). To this end, sera from 64 celiac disease (CD) subjects on a gluten-containing diet (44 f, 20 m) and after 0.6 F 0.3 years and 2.1 F 1.3 years of GFD were studied using a quantitative radioimmunoprecipitation assay. All 64 CD patients at diagnosis were full-length anti-tTG (a.a. 1—687)Ab positive. These Abs significantly decreased in frequency and titer after 6 months and 2 years of GFD. However, at low titers, 64.1% (41/64) of CD patients were still fl-tTG (a.a. 1—687)Ab positive after 2 years of GFD. At disease diagnosis, 70.3% (45/64) of the CD patients had Abs directed against fragments (227—687) and/or (473—687) of the tTG protein. This percentage, after 2 years of GFD, significantly decreased to 18.7%, whereas almost 50% of GFD patients had no tTG (227—687) and tTG (473—687) fragment reactivity, but only persistent, low-titer full-length tTG (1—687)Abs. We suggest that the selective loss of immunoreactivity against tTG (227—687) and tTG (473—687) fragments in CD patients with a
* Corresponding author. Fax: +39 06 49972619. E-mail address:
[email protected] (C. Tiberti). 1521-6616/$ — see front matter D 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.clim.2006.05.009
Epitope loss of anti-transglutaminase immunoreactivity
41
GFD, could be due to quantitative decrease of autoreactivity driven by tTG-gliadin interaction underlying celiac disease pathogenesis. D 2006 Elsevier Inc. All rights reserved.
Introduction Celiac disease (CD) is an immune-mediated enteropathy triggered, in genetically susceptible individuals, by the ingestion of gluten [1]. CD is characterized by villous atrophy and inflammatory cell infiltration of the lamina propria and may appear with different clinical presentation, including gastrointestinal manifestations (classic CD), or minor, transitory, extraintestinal symptoms (atypical CD), or a total absence of symptoms (silent CD) [2,3]. In 1997, human tissue transglutaminase (tTG) was identified as a dominant autoantigen in CD [4]. This protein belongs to a family of calcium-dependent enzymes able to catalyze the post-translational modification of proteins, with formation of highly resistant isopeptide bonds implicated in several biological processes [5]. Tissue transglutaminase autoantibodies (tTG-Abs) are detectable in most celiac patients at diagnosis and are an important complement to the intestinal biopsy in monitoring the specific dietary disease treatment [4,6]. It was demonstrated that at CD diagnosis the tTG autoreactive domains are age- and sex-dependent [7], and that the tTGAbs are directed against multiple conformational epitopes of the protein [8,9]. To date, limited information is available on the humoral anti-tTG immune response occurring after a GFD and in particular on the modulation of the tTG epitope immunoreactivity. The aim of the present study was to evaluate in classic CD patients the epitope specific humoral immunoreactivity against 3 different human recombinant tTG constructs before and after gluten-free diet.
Materials and methods Subjects
and crosslinking of extracellular matrix proteins, are involved calcium binding region (a.a. 446—453) and active site located at Cys 227 of the protein. In the other enzymatic activity, a GTP/ATP hydrolysis activity associated with signal transduction in cell cycle regulation, the N-terminal 185 amino acids of the tTG protein are involved. The study of threedimensional structure of tTG demonstrated the presence of at least 4 critical amino acid residues involved in the preservation of dominant tTG epitopes (a.a. 1—13, a.a. 228—347, a.a. 473—496, a.a. 649—687) [8]. On the basis of these findings, we constructed 3 fragments of human tTG molecule, full-length tTG (a.a. 1—687) or fl-tTG, the truncated tTG (a.a. 227—687) and tTG (a.a. 473—687), containing the functional domains of the protein as well as the critical amino acid residues involved in the preservation of dominant tTG epitopes. In addition, in the construction of the tTG fragments and in order to set up a highly sensitive antibody radioimmunoassay, we aimed at obtaining antigenic constructs that, after transcription and translation in presence of radiolabeled methionine, could contain a sufficient number of methionine residues. It is not known whether the truncated tTG molecules utilized in our study occur in vivo, however, they account for protein domains shown to be, in various studies, major targets of anti-tTG antibodies in celiac disease [7—10]. The three tTG constructs investigated in the study, amplified by polymerase chain reaction (PCR) from full-length human tTG cDNA clone AM55153 (kindly provided by Dr. Peter J.A. Davis, University of Texas Medical School, Houston, TX, USA) were prepared as previously reported [7] and sequenced with vector promoter primers Sp6 and T7 as well as internal primers for the tTG molecule using ABI 377 DNA sequencer (PE Applied Byosistems, Foster City, CA, USA). No deletions or truncations were found in the three tTG constructs.
Autoantibody radioimmunoassays Sera from 64 classic CD subjects on a gluten-containing diet (44 females, 20 males, mean age at diagnosis 13.1 F 13.1, age range 1.0—54.0 years; mean age females 14.0 F 13.2, age range 1.3—54.0 years, mean age males 11.0 F 13.0, age range 1.0—52.0 years), and after 0.6 F 0.3 years (range 0.25—1.0 years) and 2.1 F 1.3 years (range 1.1—8,9 years) of GFD, were analyzed. Diagnosis of CD was made according to criteria established by the European Society of Pediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) [2]. The 64 patients were identified as classic CD for presence at diagnosis of a gluten-sensitive enteropathy accompanied by gastrointestinal symptoms (diarrhea, weight loss, abdominal pain, or vomiting). None of the CD patients in the study had selective IgA deficiency.
tTG constructs utilized in the study Human tissue transglutaminase has two distinct enzyme activities, each related to different protein domains. In one enzymatic activity, where tTG plays a role in cell adhesion
Each tTG cDNA fragment was transcribed and translated in vitro in the presence of [35S]-methionine (NENk Life Science Products Inc.) using the TNT-coupled rabbit reticulocyte system (Promega) with Sp6 RNA polymerase. Autoantibodies against the different tTG constructs were detected according to a previously described quantitative radioimmunoprecipitation assay procedure [11]. Briefly, each [35S]-methionine human tTG construct (50.000 cpm/ tube) was incubated with serum at 1:25 dilution overnight at 48C. Twenty-five microliters of goat anti-human IgA-Agarose (Sigma) was added to this solution to separate antibodybound labeled products from free ones and leave to incubate at 48C for 1 h. After elimination of the specific radioactivity by washing with buffer solution and aspiration of the supernatant, 100 Al of scintillation liquid (Packard) was added to each tube to resuspend the pellet. Subsequently the suspension was carefully transferred to a scintillation vial and counted in a h-counter. Autoantibody levels were expressed as an index defined as follows:
42
C. Tiberti et al.
(sample cpm — negative standard control cpm)/(positive standard control cpm — negative standard control cpm). Values above the 99th percentile of 112 healthy control sera (67 females, 45 males, mean age 16.0 F 14.2 years, range 1.1—50.0 years) were considered positive for autoantibody presence and calculated separately for each tTG fragment. Autoantibody positivity was considered for Ab indexes above 0.050 for fl-tTG (a.a. 1—687), 0.166 for tTG (a.a. 227—687) and 0.126 for tTG (a.a. 473—687), respectively. The intraassay coefficients of variation were 5.4% for fl-tTG (a.a. 1— 687), 8.2% for tTG (a.a. 227—687) and 8.9% for tTG (a.a. 473—687).
EMA detection IgA endomysium antibodies (EMA) were identified by indirect immunofluorescence analysis on cryostat sections of the third distal portion of the monkeys’ esophagus (Eurospital) [12]. According to the manufacturer’s instructions, serum samples were diluted 1:5 and incubated for 30 min. Fluorescein isothiocyanate (FITC)-conjugated anti-human IgA, diluted 1:100 was added and sections were incubated for 30 min. The slides containing the stained sections were mounted with an opportune mounting medium and then observed to a fluorescence microscope (Olympus BX60) at 400. The presence of EMA, highlighted by a reticulin-like fluorescence staining along the smooth-muscle bundles of the muscolaris mucosae compartment, was evaluated by two observers (agreement rate 98.8%) who were unaware of patient’s conditions.
Statistical analysis The Mann-Whitney test was used to determine differences between groups. Categorical variables were analyzed using Fisher’s Exact Test. A two-tailed P b 0.05 was considered significant. Statistical was performed utilizing True Epistat (Round Rock, Richardson, TX, USA).
Results The results of the study are shown in Table 1 and Fig. 1. In particular, Table 1 reports for each of the tTG fragments Table 1
investigated, the number and percentage of CD patients detected as Ab-positive at disease diagnosis, and after a mean of 6 months and 24 months of gluten-free diet. The mean Ab indexes (FSD) reported in Table 1 are related to all the 64 patients analyzed in the study. The column indicated in Table 1 as tTG (227—687) and/or (473—687) is related to those CD patients having antibodies directed against one or both of these two tTG fragments.
Full-length tissue transglutaminase antibodies [fl-tTG (a.a. 1--687)Abs] At diagnosis All 64 CD patients at diagnosis (64/64, 100%) had autoantibodies directed against fl-tTG (a.a. 1—687) with a mean Ab index of 0.804 F 0.422. fl-tTG (1—687)Ab frequency and titer did not differ between female and male celiac patients. After 6 months of GFD Autoantibodies reacting with fl-tTG (a.a. 1—687) were detected in CD patients at a significantly lower frequency [81.2% (52/64) vs. 100%(64/64), P = 0.003] and titer (0.389 F 0.382 vs. 0.804 F 0.422, P b 0.0001) compared to disease diagnosis. Similar results were found for celiac female (0.382 F 0.395 vs. 0.779 F 0.416, P b 0.0001) and male (0.406 F 0.361 vs. 0.859 F 0.439, P b 0.0001) for autoantibody titers, however, Ab frequency decreased significantly in CD females [75% (33/44) vs. 100% (44/44), P = 0.0005], but not in CD males [95% (19/20) vs. 100% (20/20)]. After 2 years of GFD Autoantibodies reacting with fl-tTG (a.a. 1—687) were detected in CD patients at significantly lower mean titer (0.207 F 0.273 vs. 0.389 F 0.382, P b 0.0001) and frequency [64.1% (41/64) vs. 81.2% (52/64), P = 0.046] at 2 years compared to 6 months of GFD. Significantly lower mean Ab titers, but not frequencies, were found in CD females [0.221 F 0.286 vs. 0.382 F 0.395, P = 0.03; 59.1% (26/44) vs. 75% (33/44)] and males [0.175 F 0.246 vs. 0.406 F 0.361, P = 0.02; 75%(15/20) vs. 95%(19/20)] compared to 6 months of GFD, however, CD male Ab frequency significantly decreased with respect to disease diagnosis [75% (15/20) vs. 100% (20/20), P = 0.047].
tTG construct immunoreactivities in 64 classic celiac patients
At diagnosis
6 months of gluten-free diet 24 months of gluten-free diet
no. Ab+ %Ab+ Ab index (mean F SD) no. Ab+ %Ab+ Ab index (mean F SD) no. Ab+ %Ab+ Ab index (mean F SD)
tTG fl (1—687)
tTG (227—687)
tTG (473—687)
tTG (227—687) and/or (473—687)
64 100% a 0.804 F 0.422 b 52 81.2% c 0.389 F 0.382 d 41 64.1% e 0.207 F 0.273 f
31 48.4% g 0.300 F 0.395 h 7 10.9% i 0.076 F 0.187 j 3 4.7% 0.036 F 0.114
43 45 67.2% k 70.3% s 0.367 F 0.332 m 24 25 37.5% n 39.1% t 0.116 F 0.231 p 12 12 18.7% q 18.7% u 0.044 F 0.161 r
a vs. c P = 0.003; d vs. f P = 0.046; b vs. d P b 0.0001; c vs. e P b 0.0001; g vs. i P b 0.0001; h vs. j P b 0.0001; k vs. n P = 0.001; m vs. p P b 0.0001; n vs. q P = 0.03; p vs. r P = 0.004; s vs. t P = 0.001; t vs. u P = 0.019.
Epitope loss of anti-transglutaminase immunoreactivity
43
Figure 1 Reactivity against full-length tTG (1—687), tTG (227—687) and tTG (473—687) constructs in 64 classic celiac disease patients (CD). The y axis indicates autoantibody indexes relative to each CD patient. The x axis indicate the different times of gluten-free diet analyzed in the study (disease diagnosis, a mean time of 6 months and 2 years of gluten-free diet, respectively). The y axis shows the autoantibody titers expressed as an index. The horizontal dashed lines represent the limit of positivity of the RIA method, defined by 99th percentile of 112 normal sera. Bars indicate median value.
tTG (a.a. 227--687)Abs At diagnosis Autoantibodies reacting with tTG (227—687) were detected in 48.4% (31/64) of the CD patients, with a mean Ab titer of 0.300 F 0.395. tTG (227—687)Ab frequency and mean Ab titers were not significantly different between female and male CD patients. After 6 months of GFD tTG (a.a. 227—687)Abs were detected in patients with CD at a significantly lower frequency [10.9% (7/64), P b 0.0001] and titer (0.076 F 0.187 vs. 0.300 F 0.395, P b 0.0001) with respect to disease diagnosis. Similar results were found in CD females [9.1% (4/44) vs. 47.7% (21/44) P = 0.0001; 0.084 F 0.216 vs. 0.268 F 0.318, P = 0.002] and males [15% (3/20) vs. 50% (10/20), P = 0.04; 0.058 F 0.104 vs. 0.377 F 0.544, P = 0.01].
After 6 months of GFD tTG (a.a. 473—687)Abs were detected in CD patients at a significantly lower frequency [37.5% (24/64), P = 0.001] and titer (0.116 F 0.231, P b 0.0001) compared to disease diagnosis. Similar results were found in CD females [34.1% (15/44) vs. 61.4% (27/44), P = 0.0184; 0.107 F 0.246 vs. 0.359 F 0.359, P = 0.0002] and males [45% (9/20) vs. 80% (16/20), P = 0.048; 0.135 F 0.2 vs. 0.385 F 0.27, P = 0.002]. After 2 years of GFD tTG (a.a. 473—687)Abs significantly decreased in classic CD patients both in frequency [18.7% (12/64) vs. 37.5% (24/64), P = 0,0298] and mean Ab titer (0.044 F 0.161 vs. 0.116 F 0.231, P = 0.004) compared to 6 months of GFD. However, the frequency and mean Ab titers of these Abs did not significantly decrease in female as well as in male CD patients with respect to 6 months of GFD.
EMA After 2 years of GFD tTG (a.a. 227—687)Abs did not significantly decrease in CD patients, both in frequency [4.7% (3/64)] as well as in mean titer (0.036 F 0.114 vs. 0.076 F 0.187) compared to 6 months of GFD. Similar results were found in CD females and males.
tTG (a.a. 473--687)Abs At diagnosis Autoantibodies reacting with tTG (a.a. 473—687) were detected in 67.2% (43/64) of patients with classic CD. The mean Ab titer of the patients was 0.367 F 0.332. tTG (473— 687)Ab frequency and mean Ab titers were not significantly different between female and male CD patients.
At disease diagnosis EMA were detected in 98.4% (63/64) of CD patient sera. All the patients found EMA positive were positive also for tTG (a.a. 1—687)Abs, whereas 68.3% (43/63) and 49.2%(31/63) of them were tTG (a.a. 473— 687) and tTG (227—687)Ab positive, respectively. The sole patient found EMA negative was tTG (a.a. 1—687)Ab positive but tTG (a.a. 473—687) and tTG (227—687)Ab negative. After 6 months of GFD, EMA frequency significantly decreased compared to disease diagnosis [43.7% (28/64) P b 0.0001]. All the patients found EMA positive were positive also for tTG (a.a. 1—687)Abs, whereas 64.3% (18/28) and 17.9% (5/28) of them were tTG (a.a. 473—687) and tTG (227—687)Ab positive, respectively. After 2 years of GFD EMA frequency significantly decreased with respect to 6 months of GFD [18.7% (12/64) P =
44
C. Tiberti et al.
Figure 2 Examples of tTG (1—687) (squares), tTG (227—687) (circles) and tTG (473—687 (triangles) autoantibody individual patterns of reactivity in 4 celiac patient sera at disease diagnosis and after gluten-free diet (GFD). Full and empty symbols indicate autoantibody positivity and negativity, respectively. The x axis indicates the different times of GFD. The y axis shows the autoantibody titers expressed as an index.
0.004]. All the patients found EMA positive were positive also for tTG (a.a. 1—687)Abs, whereas 58.3% (7/12) and 25%(3/12) of them were tTG (a.a. 473—687)Ab and tTG (227—687)Ab positive, respectively.
Discussion In the present study, we demonstrated that the elimination of gluten from the diet of 64 newly diagnosed CD patients resulted in a significant decrease of the humoral immune response in all the CD patients investigated with a specific epitope loss of anti-transglutaminase immunoreactivity. The fl-tTG (a.a. 1—687) mean Ab titer of these CD patients decreased by approximately a factor of 4 times relative to disease diagnosis, providing evidence of a strict adherence of the patients to the GFD. However, even if the anti-tTG immune response decreased significantly, 64.1% of CD patients were still fl-tTG (a.a. 1—687)Ab positive after 2 years of GFD. This finding is not surprising if we consider the higher sensitivity and specificity of the RIA method in detecting anti-tTG immunoreactivity with respect to other methods [13]. In a previous study, analyzing CD patients on a gluten-free diet, we found that the radioimmunoprecipitation assay was able to detect four times more Ab-positive patients compared to an ELISA assay [14]. At disease diagnosis, 70.3% (45/64) of the CD patients investigated in this study had Abs directed against fragments (227—687) and/ or (473—687) of the tTG protein, both containing the Cterminal domain of the protein. After 2 years of GFD, the percentage of the patients having antibodies directed against one or both of the 2 tTG fragments decreased to 18.7%, whereas 46.9% of GFD patients were tTG (227—687) and tTG (473—687)Ab negative, but full-length tTG (1—687)Ab positive supporting the hypothesis that this latter type of tTG
immunoreactivity is likely to be directed against the Nterminal and/or conformational domains of the protein. There is now clear evidence that CD4 T cell responses to HLA-DQ2- and DQ-8-bound gluten-derived peptides are the primary mechanism underlying celiac disease. To fit the requirements for high affinity binding to HLA-DQ2 or—DQ8, gluten glutamines are converted into glutamic acids by the enzymatic action of tissue transglutaminase. The formation of tTG-gliadin complexes seems to be important for disease development, with a B cell response initially directed against gliadin that spreads to the autoantigen tTG [6,15,16]. Supporting this hypothesis was the presence of IgA antibodies recognizing not only single gliadin or tTG proteins, but also neoepitopes of tTG-gliadin complexes which, like gliadin, are limited to the intestine [5,17]. The loss of tTG (227—687) fragment immunoreactivity after 6 months of GFD is quite dramatic, with only 10.9% of celiac patients having antibodies reacting with this protein domain. On the other hand, there are persistent low titer fl tTG (1—687)Abs detected in CD patients after 2 years of GFD, probably directed only against N-terminal and/or conformational tTG epitopes of the protein. Fig. 2 shows a few individual patterns relative to the specific epitope loss of anti-transglutaminase immunoreactivity in celiac patients at diagnosis and after GFD. The results of the present study confirm, in agreement with other studies, that several portions encompassing virtually the entire tTG protein are target of humoral autoimmunity and have an antigenic relevance [8,9]. The appearance of multiple epitopes of humoral immune response was demonstrated in several autoimmune diseases [8]. In particular, in type 1 diabetes, there was a sequential temporal spreading of the humoral autoimmunity from immunodominant epitopes in the preclinical phase of the disease, to less immunogenic
Epitope loss of anti-transglutaminase immunoreactivity domains at disease diagnosis [18,19]. It is possible that a similar mechanism occurs also in celiac patients, leading to the formation of an early, broad humoral immune response directed against gliadin, tTG and tTG-gliadin complexes, and, after the elimination of gluten from the diet, to a restricted immunoreactivity, probably directed only against gliadin-independent tTG domains. The clinical relevance of this hypothesis is that in most cases, the persistence of fltTG (a.a. 1—687)Abs in patient sera after 2 years from CD diagnosis could not necessarily be due to an inappropriate gluten-free diet but probably only to a residual immune response against specific tTG antigens, thus suggesting the use of other serological markers to verify the real clinical status of the patients and/or their adherence to the GFD. After 24 months of gluten-free diet, 8/41 of CD patients with persisting anti-full-length tTG (1—687)Abs had Ab titers clearly comparable to median at disease diagnosis. Analyzing the anti-tTG immune response relative to these 8 patients (6 f, 2 m; age range 3.9—35years), we found that all of them had really high antibody titers at disease diagnosis. After 6 months of gluten-free diet, 6/8 of these patients had tTGAb titers still comparable to those at disease diagnosis, whereas after 2 years of gluten-free diet, the tTG (1—687)Ab titers started to decrease more consistently. It is possible to explain these results with an initial, poor dietary compliance of these patients and with the need of a prolonged period of time to decrease their tTGAb titers at the levels of the other subjects investigated. Human tTG protein was identified as the main, if not the sole, target of EMA, that are connective tissue antibodies directed against the endomysium or the loose connective tissue around smooth muscle. EMA, detected by immunofluorescence using as substrate monkey esophagus, were considered for years the gold-standard for a specific and sensitive detection of humoral autoimmunity in serum of untreated as well GFD CD patients [1]. Since EMA disappear in most cases within a year after gluten withdrawal, their detection is thought to help in assessing compliance to a GFD. However, it was demonstrated that they are poor reliable markers of dietary compliance and similar controversial results were reported also for tTG-Abs [20]. Interestingly, in our subset of sera, after 2 years of GFD, almost 60% of fl tTG (1—687)Ab-positive CD patients reacting also with tTG (227—687) and/or tTG (473—687) constructs were EMA positive, in comparison with 16.7% CD patients that were EMA positive and reacted with fl tTG (1—687), but not with tTG (227—687) and/or tTG (473—687) constructs ( P = 0.038). It could be of value, in future studies, to evaluate if presence of tTG (227—687) and/or tTG (473—687) antibodies in GFD CD patients, may represent a more specific marker of non-strict adherence to a gluten-free diet in comparison to fl tTG (1—687) or EMA detection. In conclusion, the results of this study provide clear evidence that the decrease of the anti-tTG humoral immune response in GFD CD patients occurs through a specific differential loss of tTG epitope immunoreactivity.
Acknowledgments This study is dedicated to Umberto Di Mario. The work was supported by grants from the Italian Ministry of Scientific
45 Research, the Italian University Public Education and Research Ministry, the Promoter Foundation ONLUS, the Immune Tolerance Network and Autoimmunity Prevention Center, the NIH grant (DK32083) and the CEDAR grant Narcon. The excellent technical assistance of Federica Lucantoni and Mariangela Suma were greatly appreciated.
References [1] L.M. Sollid, Coeliac disease, Nat. Rev., Immunol. 2 (2002) 647 – 655. [2] I.D. Hill, M.H. Dirks, G.S. Liptak, R.B. Colletti, A. Fasano, S. Guandalini, E.J. Hoffenberg, K. Horvath, J.A. Murray, M. Pivor, E.D. Seidman, Guideline for the diagnosis and treatment of celiac disease in children: recommendations of the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition, J. Pediatr. Gastroenterol. Nutr. 40 (2005) 1 – 19. [3] G. Bottaro, F. Cataldo, N. Rotolo, M. Spina, G.R. Corazza, The clinical pattern of subclinical/silent celiac disease: an analysis on 1026 consecutive cases, Am. J. Gastroenterol. 93 (1999) 691 – 696. [4] W. Dieterich, T. Enhis, M. Bauerm, P. Donner, U. Volta, E.O. Riecken, D. Schuppan, Identification of tissue transglutaminase as the autoantigen of celiac disease, Nat. Med. 3 (1997) 725 – 726. [5] D. Schuppan, W. Dieterich, E.O. Riecken, Exposing gliadin as a tasty food for lymphocytes, Nat. Med. 6 (1998) 666 – 667. [6] W. Dieterich, E. Laag, H. Schfpper, U. Volta, A. Ferguson, H. Gillett, E.O. Riecken, D. Schuppan, Autoantibodies to tissue transglutaminase as predictors of celiac disease, Gastroenterology 115 (1998) 1317 – 1321. [7] C. Tiberti, F. Bao, M. Bonamico, A. Verrienti, A. Picarelli, M. Di Tola, M. Ferri, E. Vecci, F. Dotta, G.S. Eisenbarth, U. Di Mario, Celiac-disease associated transglutaminase autoantibody target domains at diagnosis are age and sex dependent, Clin. Immunol. 109 (2003) 318 – 324. [8] J. Seissler, U. Wohlrab, C. Wuensche, W.A. Sherbaum, B.O. Boehm, Autoantibodies from patients with coeliac disease recognize distinct functional domains of the autoantigen tissue transglutaminase, Clin. Exp. Immunol. 125 (2001) 216 – 221. [9] D. Sblattero, F. Florian, E. Azzoni, T. Zyla, M. Park, V. Baldas, T. Not, A. Ventura, A. Bradbury, R. Marzari, The analysis of the fine specificity of coeliac disease antibodies using tissue transglutaminase fragments, Eur. J. Biochem. 269 (2002) 5175 – 5181. [10] K. Nakachi, M. Powell, G. Swift, M.A. Amoroso, R. AnanievaJordanova, C. Arnold, J. Sanders, J. Furmaniak, B. Ress Smith, Epitopes recognised by tissue transglutaminase antibodies in celiac disease, J. Autoimmun. 22 (2004) 53 – 63. [11] M. Bonamico, C. Tiberti, A. Picarelli, P. Mariani, D. Rossi, E. Cipolletta, M. Greco, M. Di Tola, L. Sabbatella, B. Carabba, M. Magliocca, P. Strisciuglio, U. Di Mario, Radioimmunoassay to detect antitransglutaminase autoantibodies is the most sensitive and specific screening method for celiac disease, Am. J. Gastroenterol. 96 (2001) 1536 – 1540. [12] A. Picarelli, L. Sabbatella, M. Di Tola, S. Vetrano, C. Casale, M.C. Anania, B. Porowska, M. Vergari, R. Schiaffini, P. Gargiulo, Anti-endomysial antibody of IgG1 isotype detection strongly increases the prevalence of celiac disease in patients affected by type I diabetes mellitus, Clin. Exp. Immunol. 142 (2005) 111 – 115. [13] E. Liu, M. Li, F. Bao, D. Miao, M.J. Rewers, G.S. Eisenbarth, E.J. Hoffenberg, Need for quantitative assessment of transglutaminase autoantibodies for celiac disease in screening-identified children, J. Pediatr. (2005 (April)) 494 – 499. [14] M. Bonamico, C. Tiberti, A. Picarelli, P. Mariani, D. Rossi, E. Cipolletta, M. Greco, M. Di Tola, L. Sabbatella, B. Carabba, M.
46
[15] [16]
[17]
[18]
C. Tiberti et al. Magliocca, P. Strisciuglio, U. Di Mario, Radioimmunoassay to detect antitransglutaminase autoantibodies is the most sensitive and specific screening method for celiac disease, Am. J. Gastroenterol. 96 (2001) 1536 – 1540. L.M. Sollid, Coeliac disease: dissecting a complex inflammatory disorder, Nat. Rev. 2 (2002) 647 – 655. F. Koning, L. Gilissen, C. Wijmenga, Gluten: a two edged sword. Immunopathogenesis of celiac disease, Springer Semin. Immunopathol. 27 (2005) 217 – 232. T. Freitag, H. Schulze-Koops, G. Niedobitek, G. Melino, D. Shuppan, The role of the immune response against tissue transglutaminase in the pathogenesis of celiac disease, Autoimmun. Rev. 3 (2004) 13 – 20. H.E. Naserke, A.G. Ziegler, V. Lampasona, E. Bonifacio, Early
development and spreading of autoantibodies to epitopes of IA-2 and their association with progression to type 1 diabetes, J. Immunol. 161 (1998) 6963 – 6969. [19] P. Sfhnlein, M. Mqller, K. Syren, U. Hartmann, B.O. Bum, H.M. Meinck, M. Knip, H.K. Akerblom, W. Ritcher, Epitope spreading and a varying but not disease-specific GAD 65 antibody response in type 1 diabetes. The Childhood Diabetes in Finland Study Group, Diabetologia 43 (2000) 210 – 217. [20] K. Vahedi, F. Mascart, J.Y. Mary, J.E. Laberenne, Y. Bouhnik, M.C. Morin, A. Ocmant, C. Velly, J.F. Colombel, C. Matuchanski, Reliability of antitransglutaminase antibodies as predictors of gluten-free diet compliance in adult celiac disease, Am. J. Gastroenterol. 98 (2003) 1079 – 1087.
