Transamidation of Wheat Flour Inhibits the Response to Gliadin of Intestinal T Cells in Celiac Disease
2007; Elsevier BV; Volume: 133; Issue: 3 Linguagem: Inglês
10.1053/j.gastro.2007.06.023
ISSN1528-0012
AutoresCarmen Gianfrani, Rosa Anna Siciliano, Angelo Facchiano, Alessandra Camarca, Maria Fiorella Mazzeo, Susan Costantini, V.M. Salvati, Francesco Maurano, Giuseppe Mazzarella, Gaetano Iaquinto, Paolo Bergamo, Mauro Rossi,
Tópico(s)Gastrointestinal disorders and treatments
ResumoBackground & Aims: Celiac disease is characterized by activation of HLA-DQ2/DQ8–restricted intestinal gluten-specific CD4+ T cells. In particular, gluten becomes a better T-cell antigen following deamidation catalyzed by tissue transglutaminase. To date, the only available therapy is represented by adherence to a gluten-free diet. Here, we examined a new enzyme strategy to preventively abolish gluten activity. Methods: Enzyme modifications of the immunodominant α-gliadin peptide p56-68 were analyzed by mass spectrometry, and peptide binding to HLA-DQ2 was simulated by modeling studies. Wheat flour was treated with microbial transglutaminase and lysine methyl ester; gliadin was subsequently extracted, digested, and deamidated. Gliadin-specific intestinal T-cell lines (iTCLs) were generated from biopsy specimens from 12 adult patients with celiac disease and challenged in vitro with different antigen preparations. Results: Tissue transglutaminase–mediated transamidation with lysine or lysine methyl ester of p56-68 or gliadin in alkaline conditions inhibited the interferon γ expression in iTCLs; also, binding to DQ2 was reduced but not abolished, as suggested by in silico analysis. Lysine methyl ester was particularly effective in abrogating the activity of gliadin. Notably, a block in the response was observed when iTCLs were challenged with gliadin extracted from flour pretreated with microbial transglutaminase and lysine methyl ester. Conclusions: Transamidation of wheat flour with a food-grade enzyme and an appropriate amine donor can be used to block the T cell–mediated gliadin activity. Considering the crucial role of adaptive immunity in celiac disease, our findings highlight the potential of the proposed treatment to prevent cereal toxicity. Background & Aims: Celiac disease is characterized by activation of HLA-DQ2/DQ8–restricted intestinal gluten-specific CD4+ T cells. In particular, gluten becomes a better T-cell antigen following deamidation catalyzed by tissue transglutaminase. To date, the only available therapy is represented by adherence to a gluten-free diet. Here, we examined a new enzyme strategy to preventively abolish gluten activity. Methods: Enzyme modifications of the immunodominant α-gliadin peptide p56-68 were analyzed by mass spectrometry, and peptide binding to HLA-DQ2 was simulated by modeling studies. Wheat flour was treated with microbial transglutaminase and lysine methyl ester; gliadin was subsequently extracted, digested, and deamidated. Gliadin-specific intestinal T-cell lines (iTCLs) were generated from biopsy specimens from 12 adult patients with celiac disease and challenged in vitro with different antigen preparations. Results: Tissue transglutaminase–mediated transamidation with lysine or lysine methyl ester of p56-68 or gliadin in alkaline conditions inhibited the interferon γ expression in iTCLs; also, binding to DQ2 was reduced but not abolished, as suggested by in silico analysis. Lysine methyl ester was particularly effective in abrogating the activity of gliadin. Notably, a block in the response was observed when iTCLs were challenged with gliadin extracted from flour pretreated with microbial transglutaminase and lysine methyl ester. Conclusions: Transamidation of wheat flour with a food-grade enzyme and an appropriate amine donor can be used to block the T cell–mediated gliadin activity. Considering the crucial role of adaptive immunity in celiac disease, our findings highlight the potential of the proposed treatment to prevent cereal toxicity. See editorial on page 1025.Celiac disease, the most common food-sensitive enteropathy in humans,1Branski D. Fasano A. Troncone R. Latest developments in the pathogenesis and treatment of celiac disease.J Pediatr. 2006; 149: 295-300Abstract Full Text Full Text PDF PubMed Scopus (49) Google Scholar is caused by the lack of oral tolerance to gliadins and glutenins, protein components of wheat gluten, and to related proteins of rye and barley. The toxicity of prolamins from oats still remains questionable.2Hogberg L. Laurin P. Falth-Magnusson K. et al.Oats to children with newly diagnosed coeliac disease: a randomised double blind study.Gut. 2004; 53: 649-654Crossref PubMed Scopus (130) Google Scholar Celiac disease is strongly associated with HLA class II genes encoding for DQ2 and DQ8 heterodimers.3Sollid L.M. Coeliac disease: dissecting a complex inflammatory disorder.Nat Rev Immunol. 2002; 2: 647-655Crossref PubMed Scopus (794) Google Scholar The involvement of the immune system is well documented, because inappropriate CD4+ T-cell activation in lamina propria, triggered by gluten peptides bound to DQ2 and DQ8 heterodimers, is commonly observed.3Sollid L.M. Coeliac disease: dissecting a complex inflammatory disorder.Nat Rev Immunol. 2002; 2: 647-655Crossref PubMed Scopus (794) Google Scholar The mucosal lesion is mainly induced by the production of interferon (IFN)-γ from these gluten-specific T cells.4Nilsen E. Jahnsen F.L. Lundin K.E. et al.Gluten induces an intestinal cytokine response strongly dominated by interferon gamma in patients with celiac disease.Gastroenterology. 1998; 115: 551-563Abstract Full Text Full Text PDF PubMed Scopus (387) Google Scholar The effects of changes in intestinal permeability5Fasano A. Not T. Wang W. et al.Zonulin, a newly discovered modulator of intestinal permeability, and its expression in coeliac disease.Lancet. 2000; 355: 1518-1519Abstract Full Text Full Text PDF PubMed Scopus (436) Google Scholar as well as activation of innate immune mechanisms1Branski D. Fasano A. Troncone R. Latest developments in the pathogenesis and treatment of celiac disease.J Pediatr. 2006; 149: 295-300Abstract Full Text Full Text PDF PubMed Scopus (49) Google Scholar, 6Gianfrani C. Auricchio S. Troncone R. Adaptive and innate immune responses in celiac disease.Immunol Lett. 2005; 99: 141-145Crossref PubMed Scopus (109) Google Scholar in the pathogenesis of celiac disease are still under debate.DQ2 has a preference for binding peptides containing negatively charged residues at P4, P6, or P7 anchor positions7Johansen B.H. Vartdal F. Eriksen J.A. et al.Identification of a putative motif for binding of peptides to HLA-DQ2.Int Immunol. 1996; 8: 77-182Crossref Scopus (90) Google Scholar, 8van de Wal Y. Kooy Y.M. Drijfhout J.W. et al.Peptide binding characteristics of the coeliac disease-associated DQ(alpha1*0501, beta1*0201) molecule.Immunogenetics. 1996; 44: 246-253Crossref PubMed Scopus (138) Google Scholar, 9Vartdal F. Johansen B.H. Friede T. et al.The peptide binding motif of the disease associated HLA-DQ (alpha 1* 0501, beta 1* 0201) molecule.Eur J Immunol. 1996; 26: 2764-2772Crossref PubMed Scopus (140) Google Scholar; the DQ8 peptide binding motif also has 2 acidic residues that fit the P1 and P9 pockets.10Kwok W.W. Nepom G.T. Raymond F.C. HLA-DQ polymorphisms are highly selective for peptide binding interactions.J Immunol. 1995; 155: 2468-2476PubMed Google Scholar, 11Lee K.H. Wucherpfennig K.W. Wiley D.C. Structure of a human insulin peptide-HLA-DQ8 complex and susceptibility to type 1 diabetes.Nat Immunol. 2001; 2: 501-507Crossref PubMed Scopus (321) Google Scholar However, gluten proteins, characterized by a high content in glutamine and proline residues, do not have many acidic residues. This discrepancy was resolved by the finding that gluten becomes a better T-cell antigen following deamidation.12Sjostrom H. Lundin K.E. 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: 111-115Crossref PubMed Scopus (229) Google Scholar To date, most T-cell gliadin epitopes have been identified following deamidation catalyzed by tissue transglutaminase (tTG), which converts specific glutamine residues into glutamic acid,13Molberg O. McAdam S.N. Korner R. et al.Tissue transglutaminase selectively modifies gliadin peptides that are recognized by gut-derived T cells in celiac disease.Nat Med. 1998; 4: 713-717Crossref PubMed Scopus (969) Google Scholar increasing the affinity of peptides to both DQ212Sjostrom H. Lundin K.E. 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: 111-115Crossref PubMed Scopus (229) Google Scholar, 14Arentz-Hansen H. McAdam S.N. Molberg O. et al.Celiac lesion T cells recognize epitopes that cluster in regions of gliadins rich in proline residues.Gastroenterology. 2002; 123: 803-809Abstract Full Text Full Text PDF PubMed Scopus (301) Google Scholar, 15Arentz-Hansen H. Corner R. Molberg O. et al.The intestinal T cell response to alpha-gliadin in adult celiac disease is focused on a single deamidated glutamine targeted by tissue transglutaminase.J Exp Med. 2000; 191: 603-612Crossref PubMed Scopus (552) Google Scholar, 16Vader W. Kooy Y. Van Veelen P. et al.The gluten response in children with celiac disease is directed toward multiple gliadin and glutenin peptides.Gastroenterology. 2002; 122: 1729-1737Abstract Full Text Full Text PDF PubMed Scopus (373) Google Scholar and DQ817van 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: 1585-1588PubMed Google Scholar molecules. Furthermore, proline residues protect against digestive proteolysis and direct tTG-mediated deamidation of glutamines.18Hausch F. Shan L. Santiago N.A. et al.Intestinal digestive resistance of immunodominant gliadin peptides.Am J Physiol Gastrointest Liver Physiol. 2002; 283: G996-G1003Crossref PubMed Scopus (295) Google Scholar, 19Shan L. Molberg O. Parrot I. et al.Structural basis for gluten intolerance in celiac sprue.Science. 2002; 297: 2275-2279Crossref PubMed Scopus (1237) Google Scholar On the basis of these observations, treatment of gluten with bacterial prolyl endopeptidases was shown to decrease the number of immunostimulatory peptides, highlighting the possibility of developing oral peptidase therapy against celiac disease.20Marti T. Molberg O. Li Q. et al.Prolyl endopeptidase-mediated destruction of T cell epitopes in whole gluten: chemical and immunological characterization.J Pharmacol Exp Ther. 2005; 312: 19-26Crossref PubMed Scopus (115) Google Scholar, 21Cerf-Bensussan N. Matysiak-Budnik T. Cellier C. et al.Oral proteases: a new approach to managing coeliac disease.Gut. 2007; 56: 157-160Crossref PubMed Scopus (37) Google ScholarIn the present work, we examined the possibility of a different enzyme strategy to preventively abolish the stimulatory activity of gliadin while preserving the integrity of the protein structure. We show that transamidation of wheat flour with an appropriate amine group donor can be used to block gliadin immunotoxicity.Materials and MethodsPatientsTwelve HLA-DQ2+ adult patients with celiac disease, 8 treated (range, 18–49 years; mean, 29.4 years) and 4 untreated (range, 18–34 years; mean, 27 years), were enrolled in this study. Celiac disease was diagnosed according to a combination of clinical signs and typical small intestinal histology of crypt hyperplasia and villous atrophy.22Oberhuber G. Granditsch G. Vogelsang H. The histopathology of coeliac disease: time for a standardized report scheme for pathologists.Eur J Gastroenterol Hepatol. 1999; 11: 1185-1194Crossref PubMed Scopus (1346) Google Scholar All patients were informed about the study and gave their consent.ReagentsGliadin, the α-gliadin peptide p56-68, lysine, lysine methyl ester, guinea pig liver tTG (1.5 U/mg), trypsin, pepsin, dithiothreitol, α-cyano-4-hydroxycynnamic acid, angiotensin, and adrenocorticotropic hormone fragment 18-39 were purchased from Sigma (St Louis, MO). The α-gliadin 33-mer peptide19Shan L. Molberg O. Parrot I. et al.Structural basis for gluten intolerance in celiac sprue.Science. 2002; 297: 2275-2279Crossref PubMed Scopus (1237) Google Scholar was synthesized in house (Camarca et al, manuscript in preparation). Microbial transglutaminase (mTG) was purchased from N-Zyme BioTec GmbH (Darmstadt, Germany). RPMI medium, nonessential amino acids, and human serum were from Bio-Whittaker (Bergamo, Italy). All others reagents and solvents were of the highest purity and are available from Carlo Erba (Milan, Italy).Enzyme ReactionsPeptic-tryptic digest of gliadin (PT-gliadin) was prepared by suspending wheat gliadin (100 mg) in 0.1N HCl (10 mL) and incubating with 500 μg pepsin for 2 hours at 37°C with shaking; pH was then adjusted to 7.8 followed by 2-hour incubation with 500 μg trypsin. tTG-mediated deamidation and transamidation reactions were performed in 0.125 mol/L Tris-HCl, pH 8.5, containing 1 mmol/L calcium chloride, 10 mmol/L dithiothreitol, 0.2 μg/μL tTG, and 2 μg/μL substrate for 4 hours with the addition of 20 mmol/L lysine or lysine methyl ester in the transamidation reaction. Peptides were separated from salts and tTG using a Sep-Pak C18 cartridge (Waters, Milford, MA) equilibrated in 0.1% trifluoroacetic acid, eluted with 50% acetonitrile in 0.1% trifluoroacetic acid. Samples were dried and stored at −80°C. mTG treatment of synthetic peptides (2 mg/mL) was performed in water or 0.125 mol/L Tris-HCl, pH 8.5, with 2.5 U/mL enzyme for 2 hours at room temperature and 20 mmol/L lysine or lysine methyl ester. mTG treatment of commercial wheat flour (120 mg/mL) was performed in water containing 0.8 U/mL mTG and 20 mmol/L lysine methyl ester for 2 hours (mild condition) or 2 mol/L lysine methyl ester for 4 hours (strong condition) at 37°C. Control samples were prepared using the same experimental conditions without the amine donors. Gliadin was extracted from flour according to the Osborne fractionation procedure23Kick F. Belitz H.D. Wieser H. et al.Comparative studies of the Osborne protein fraction of wheat varieties with different dough and baking properties.Z Lebens Unters Forsch. 1992; 195: 437-442Crossref PubMed Scopus (7) Google Scholar and stored at −20°C.Mass Spectrometric AnalysesA total of 100 fmol of peptide mixed with a suitable matrix (α-cyano-4-hydroxycynnammic acid 10 mg/mL in 50% acetonitrile), containing 125 fmol/μL adrenocorticotropic hormone and 25 fmol/μL angiotensin as internal standards, was deposited onto a MALDI target plate and dried. Spectra were generated on a matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometer Voyager DE™ PRO (Applied Biosystems, Foster City, CA), operating in the reflectron delay extraction positive-ion mode. Mass spectra were calibrated using the monoisotopic peaks of angiotensin (m/z 931.5154 daltons) and adrenocorticotropic hormone (m/z 2465.1989 daltons) so that the experimental error was <20 ppm. Deamidation and transamidation sites were defined by tandem mass spectrometry (MS/MS) experiments performed on a hybrid quadrupole/orthogonal time-of-flight instrument (Q-Tof Micro; Waters) interfaced with an orthogonal Z-spray source operating in the positive-ion mode. Mass spectra were acquired in the m/z range of 100–2000 daltons.Western Blot AnalysisGliadin (50 μg) was fractionated by 12% sodium dodecyl sulfate/polyacrylamide gel electrophoresis and electroblotted onto polyvinylidene difluoride membrane. After blocking, the membrane was probed with anti-gliadin mouse polyclonal antibody24Rossi M. Maurano F. Caputo N. et al.Intravenous or intranasal administration of gliadin is able to down-regulate the specific immune response in mice.Scand J Immunol. 1999; 50: 177-182Crossref PubMed Scopus (22) Google Scholar followed by incubation with horseradish peroxidase–conjugated secondary antibodies and enhanced chemiluminescence detection. Equivalent protein loading was confirmed by Coomassie blue staining and densitometric analysis by ImageQuant software (Molecular Dynamics, Inc, Sunnyvale, CA).Amino Acid Composition of Gliadin Isolated From Treated FlourGliadin samples were hydrolyzed with 400 μL of 6 mol/L HCl containing 0.02% phenol and 60 nmol of nor-Leu as internal standard at 110°C for 20 hours. HCl was removed under vacuum and samples were taken up in 0.3 mL of 0.2 mol/L lithium citrate buffer, pH 2.2. Aliquots (100 μL) were analyzed using a Biochrom 20 amino acid analyzer (Biochrom, Cambridge, England).Generation of Gliadin-Specific Intestinal T-Cell LinesEndoscopic mucosal explants were digested with collagenase A as previously described.25Troncone R. Gianfrani C. Mazzarella G. et al.Majority of gliadin-specific T-cell clones from celiac small intestinal mucosa produce interferon-gamma and interleukin-4.Dig Dis Sci. 1998; 43: 156-161Crossref PubMed Scopus (54) Google Scholar Intestinal cells (2 × 105 cells/mL) were suspended in RPMI medium supplemented with antibiotics, nonessential amino acids, sodium pyruvate, glutamine, and 10% inactivated human serum (complete medium). Thereafter, cells were stimulated with 1 × 106 irradiated (3500 rad) peripheral blood mononuclear cells and 50 μg/mL tTG-treated PT-gliadin. Forty-eight hours later, cultures were refreshed with complete medium containing 10 ng/mL interleukin (IL)-15 (R&D Systems, Minneapolis, MN). On day 7, intestinal T-cell lines (iTCLs) were restimulated with antigen and autologous irradiated peripheral blood mononuclear cells, followed by addition of fresh medium and IL-15 the day after and at 3- to 4-day intervals. Long-term iTCLs were finally established by restimulation cycles (14 days) with phytohemagglutinin and feeder cells. All iTCLs were found to be 90% CD4+ by fluorescence-activated cell sorter analysis.T-Cell AssaysiTCLs were tested in the resting phase. Antigen-pulsed, HLA-matched, Epstein–Barr virus–transformed B lymphoblastoid cell lines were used as antigen-presenting cells. Irradiated antigen-presenting cells (1 × 105 cells/well) were incubated overnight with different concentrations of peptides or PT-gliadin (50 μg/mL) in 96-well plates. In competition assays, antigen-presenting cells were incubated overnight with native or tTG-deamidated peptide p56-68 (10 and 1 μmol/L, respectively) or PT-gliadin (50 μg/mL) and increasing concentrations of transamidated peptides. After cell washing, 0.3 × 105 T cells were added to each well in a final volume of 200 μL. Culture supernatant aliquots were analyzed for cytokine levels after 24-hour (IL-2) or 48-hour incubation (IFN-γ, IL-4, and IL-10) by enzyme-linked immunosorbent assay.Simulation of DQ2-Peptide ComplexesThe 3-dimensional structure of the DQ2 molecule complexed with peptide p(58-68) E65 was used as a template in the simulations (PDB code: 1S9V).26Kim C.Y. Quarsten H. Bergseng E. et al.Structural basis for HLA-DQ2 mediated presentation of gluten epitopes in celiac disease.Proc Natl Acad Sci U S A. 2004; 101: 4175-4179Crossref PubMed Scopus (360) Google Scholar Amino acids L56 and Q57 were bound to N-terminal peptide using the Biopolymer module of InsightII (Accelrys, San Diego, CA). Additional complexes were created by modifying the sequence of the gliadin peptide. Each DQ2-peptide complex was then optimized with the software InsightII by using 500 steps of energy minimization under conjugate gradient algorithm.27Costantini S. Rossi M. Colonna G. et al.Modelling of HLA-DQ2 and its interaction with gluten peptides to explain molecular recognition in celiac disease.J Mol Graph Model. 2005; 23: 419-431Crossref PubMed Scopus (33) Google Scholar The energy of interaction between the peptide and the DQ2 molecule was then evaluated by using the “Energy/Intermolecular tool” in the Docking module of InsightII. The free binding energies were evaluated by using the program DCOMPLEX.28Liu S. Zhang C. Zhou H. et al.A physical reference state unifies the structure-derived potential of mean force for protein folding and binding.Proteins. 2004; 56: 93-101Crossref PubMed Scopus (161) Google Scholar The HBplus package29McDonald I.K. Thornton J.M. Satisfying hydrogen bonding potential in proteins.J Mol Biol. 1994; 238: 777-793Crossref PubMed Scopus (1858) Google Scholar was used to evaluate the putative formation of H-bonds.Statistical AnalysisThe results are expressed as mean ± SD. Differences among the various treatment groups were determined by one-way analysis of variance (ANOVA). Multiple comparisons of treatment means were made using the Tukey test, and the criterion for significance was P < .05.ResultsModifications Induced by tTG on the α-Gliadin Peptide p56-68 Inhibit IFN Gamma Production in iTCLsStructural modifications induced by tTG at pH 8.5 in the presence of lysine or lysine methyl ester were investigated on the α-gliadin peptide p56-68 (LQLQPFPQPQLPY), which contains an immunodominant epitope.15Arentz-Hansen H. Corner R. Molberg O. et al.The intestinal T cell response to alpha-gliadin in adult celiac disease is focused on a single deamidated glutamine targeted by tissue transglutaminase.J Exp Med. 2000; 191: 603-612Crossref PubMed Scopus (552) Google Scholar In the MALDI-TOF mass spectrum, the peptide cross-linked to lysine originated a strong signal at m/z 1697.94 daltons (Figure 1A), with a shift of 129 daltons from the m/z value of the native peptide (1568.84 daltons), indicating the addition of a single lysine molecule, whereas any trace of the deamidated form (m/z 1569.84 daltons) was undetectable. Similarly, a single form of the adduct with lysine methyl ester or of the deamidated peptide was obtained (data not shown). In the MS/MS spectrum, obtained from the doubly charged ion at m/z 849.56 daltons, generated from p56-68 cross-linked to lysine, the mass difference between the fragment ions y5 and y3 originated from the C-terminal region of the peptide (m/z 746.91 and 392.49 daltons, respectively) can be attributed to the sequence 64PQ65 cross-linked to lysine (Figure 1B). Experiments performed on the different forms of the peptide indicated that both transamidation and deamidation reactions are site specific and modify exclusively Q65 (Table 1). The ability of the various forms of p56-68 to induce IFN-γ expression was tested in iTCLs derived from 12 HLA-DQ2+ patients with celiac disease. iTCLs from all 12 patients produced IFN-γ in their positive controls (tTG-deamidated PT-gliadin), but only 5 of them recognized p56-68 (E65) (Supplementary Table 1; see supplemental material online at www.gastrojournal.org). Interestingly, in the responsive cell lines, incubation with the peptide cross-linked to lysine, p56-68 (Q65-K), significantly decreased, or even abrogated, IFN-γ expression (Supplementary Table 1 [see supplemental material online at www.gastrojournal.org] and Figure 1C). A further decrease was observed following incubation with the peptide cross-linked to lysine methyl ester, p56-68 (Q65-K-CH3).Table 1Identification of the Modified Q Residues in the Deamidated and Transamidated Forms of p56-68PeptidePrecursor ionFragment ion b8Fragment ion b10Fragment ion y3Fragment ion y5p56-68784.98952.661,177.75392.27617.40p56-68 (E65)785.48952.641,178.75392.27618.43p56-68 (Q65-K)849.56952.681,307.59392.49746.91p56-68 (Q65-K-CH3)856.60952.681,321.63392.49760.95NOTE. Diagnostic fragment ions are reported. Precursor ions are doubly charged, whereas fragment ions are all singly charged. Open table in a new tab Ability of Cross-linked Peptides to Interact With HLA-DQ2 HeterodimerA dose-effect response of IFN-γ production was observed for both p56-68 and p56-68 (E65), whereas cross-linked peptides failed to induce a comparable stimulation at any concentration (Supplementary Figure 1A; see supplemental material online at www.gastrojournal.org), thus confirming their reduced stimulatory capacity. Moreover, both p56-68 (Q65-K) and p56-68 (Q65-K-CH3) were unable to inhibit stimulation following coincubation with p56-68 (E65) at any tested concentration (0.01–15 μmol/L) (Supplementary Figure 1B; see supplemental material online at www.gastrojournal.org). Similarly, increasing concentrations of p56-68 (K-CH3) did not inhibit the IFN-γ response induced by p56-68, PT-gliadin, or deamidated PT-gliadin (Supplementary Figure 1B; see supplemental material online at www.gastrojournal.org). To further investigate this issue, the formation of complexes between HLA-DQ2 and the different forms of p56-68 was simulated. Values of free binding energy and energy of interaction suggested that all peptides could fit into the DQ2 pocket but with different affinities in the following order: p56-68 (Q65-K-CH3) ≪ p56-68 (Q65-K) < p56-68 ≪ p56-68 (E65) (Figure 2A). In particular, a large increase of electrostatic energy occurred in p56-68 (E65) as a consequence of adding negatively charged groups. By analyzing the structural environment of the peptide pocket, the side chain of any substituted amino acid in position 65 resulted in proximity of the positive charge of arginine B70 and lysine B71 of the DQ2 molecule, as reported by Kim et al.26Kim C.Y. Quarsten H. Bergseng E. et al.Structural basis for HLA-DQ2 mediated presentation of gluten epitopes in celiac disease.Proc Natl Acad Sci U S A. 2004; 101: 4175-4179Crossref PubMed Scopus (360) Google Scholar Therefore, the negative charge added in p56-68 (E65) could improve the energy of interaction with the HLA-DQ2 molecule, in agreement with previous results.27Costantini S. Rossi M. Colonna G. et al.Modelling of HLA-DQ2 and its interaction with gluten peptides to explain molecular recognition in celiac disease.J Mol Graph Model. 2005; 23: 419-431Crossref PubMed Scopus (33) Google Scholar The p56-68 (Q65-K) peptide, which has 2 charged groups (ie, carboxylic group and amine group of lysine), showed a small decrease of favorable atom contacts (Van der Waals energy) and a little loss of electrostatic energy, in comparison with the native peptide, having the neutral side chain. Finally, p56-68 (Q65-K-CH3), which has only a positively charged group (ie, the amine group of lysine), showed the lowest energy of interaction, mainly due to the loss of electrostatic energy. We also analyzed the putative H-bonds in the peptide-DQ2 complexes. Q65 was involved in 2 H-bonds with the DQ2 molecule (Supplementary Table 2; see supplemental material online at www.gastrojournal.org). Modification of the Q65 side chain affected only the second H-bond, between the side chain nitrogen of Q65 and the side chain oxygen of serine B30. Substitution Q65 → E65 improved the H-bond interaction by means of 2 factors: (1) the oxygen atom of the E65 side chain was a better acceptor than the nitrogen atom of the Q65 side chain and (2) the donor-acceptor distance was lower in the E65 case. Moreover, 2 additional putative H-bonds were possible (Figure 2B, top, and Supplementary Table 2; see supplemental material online at www.gastrojournal.org). Interestingly, one H-bond involved lysine B71 residue, which is present only in the DQ2B1*020x alleles and has an important role in the binding site of DQ2.27Costantini S. Rossi M. Colonna G. et al.Modelling of HLA-DQ2 and its interaction with gluten peptides to explain molecular recognition in celiac disease.J Mol Graph Model. 2005; 23: 419-431Crossref PubMed Scopus (33) Google Scholar Concerning p56-68 (Q65-K), the 2 charged groups of lysine formed H-bonds with the amino acids of DQ2, in particular, the carboxylic group of lysine with the charged side chain of arginine A76 that is conserved in all DQ alleles and its amine group with backbone oxygen of alanine B57 (Figure 2B, bottom, and Supplementary Table 2; see supplemental material online at www.gastrojournal.org). The p56-68 (Q65-K-CH3) only has a positively charged group, which formed an H-bond with asparagine A69, whereas the other H-bond was lost (Supplementary Table 2; see supplemental material online at www.gastrojournal.org).Figure 2Modeling of DQ2-(p56-68) interaction. (A) Bar graphs of the energies of interaction and free binding energies computed for each peptide-DQ2 complex. (Top) Van der Waals (white bars) and electrostatic contributions (grey bars) to the energy of interaction. (Bottom) Free binding energies. (B) Detailed view of the molecular interaction between DQ2 and p56-68 (E65) (top) and p56-68 (Q65-K) peptide (bottom). The C-terminal portion of the peptide is shown as a stick representation, and the amino acids of DQ2 (see labels) involved in H-bonds with the modified amino acid of peptide are shown as a ball and stick representation (atom colors: green, carbon; red, oxygen; blue, nitrogen).View Large Image Figure ViewerDownload Hi-res image Download (PPT)Transamidation of Gliadin Inhibits the Immune Activity In VitroWe next evaluated whether treatment of gliadin with tTG and lysine/lysine methyl ester was able to suppress its immune stimulatory properties. As expected, a more powerful IFN-γ response was generated from most iTCLs with deamidated gliadin (PT-gliadin + tTG) than with native gliadin (PT-gliadin) (Figure 3A). When iTCLs were challenged with gliadin cross-linked to lysine (PT-gliadin + tTG + K), reduced production of IFN-γ was observed in comparison with deamidated gliadin in all examined patients. Interestingly, a stronger reduction was generally detected following incubation of iTCLs with gliadin cross-linked to lysine methyl ester (PT-gliadin + tTG + K-CH3). The statistical evaluation of the results indicated that lysine methyl ester caused a significantly higher inhibition of IFN-γ expression than lysine, with values not different from the negative control (medium; Figure 3B).Figure 3Effect of tTG-mediated transamidation of gliadin on the IFN-γ response of iTCLs. (A) IFN-γ production (pg/mL) of 12 iTCLs isolated from patients with celiac disease. Results are expressed as mean ± SD of triplicate cultures. (B) Per
Referência(s)