| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
* Immunology and
Internal Medicine, Division of Gastroenterology and Hepatology, Mayo Clinic and Foundation, Rochester, MN 55905
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, and IL-10. Finally, intestinal lymphocytes
isolated from gluten-fed HCD4/DQ8 mice displayed an activated
phenotype. These data suggest that this HLA class II transgenic murine
model of gluten sensitivity may provide insight into the initiation of
the MHC class II-restricted gluten sensitivity in celiac
disease. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Understanding the immunopathologic aspects of CS has been hindered due to the lack of an animal model for the disease. Transgenic mice expressing human class II molecules (in the absence of endogenous mouse class II) have been used to study other autoimmune disorders, including rheumatoid arthritis and multiple sclerosis (15, 16, 17, 18). We have studied the effect of gluten in double transgenic mice expressing HLA-DQ and human CD4 genes in the absence of endogenous class II and CD4 genes. We determined the effect of specific MHC molecules on the development of gluten sensitivity and the role of specific gliadin epitopes as a way to provide insight into the initiation of celiac disease.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cosmids containing either the DQ8 or DQ6 gene were microinjected into (B10 x SWR)F1-derived embryos. Transgene-positive mice were then crossed with the class II knockout mouse from a B6 x 129 background (courtesy of Drs. C. Benoist and D. Mathis, Institute de Genetique et de Biologie Moleculaire et Cellulaire, University of Louis Pasteur, Strasbourg, France). The founder mice were backcrossed to B10 mice. The resulting mice expressing human class II and lacking endogenous mouse class II were designated AB0DQ8 or AB0DQ6. These mice were then mated to mice expressing human CD4 (hCD4+, courtesy of R. Flavell, Yale University School of Medicine, New Haven, CT) lacking mouse CD4 (CD40, courtesy of T. Mak, Ontario Cancer Institute, University of Toronto, Toronto, Ontario, Canada). The double transgenic, double knockout mice were intercrossed to maintain the line. Thus, these lines have similar background but are not congenic. Transgene-negative full siblings were used as controls. These mice have an intact immune system and are capable of generating both a Th1 and a Th2 response (16, 19, 20, 21). Surface expression of the human class II and HCD4 transgenes on peripheral blood leukocytes was monitored by flow cytometry. Mice were maintained in the pathogen-free immunogenetics mouse colony at the Mayo Clinic (Rochester, MN).
Antigens
Crude gluten was purchased from Sigma-Aldrich (St. Louis, MO). Overlapping peptides of gliadin were synthesized at the Peptide Core Facility of the Mayo Clinic. These peptides were 20 aa in length, with a 10-aa residue overlap with the previous peptide. Rice cereal was used as a negative control.
Monoclonal Abs
Cultured supernatants from cell lines producing mAbs for the
HLA-DQ
chain (IVD12), HLA DR (L227),
DQ6 chain (61.11.1), and CD4 (GK1.5) were prepared in our
laboratory.
Proliferation assays
Mice were injected with 20 µg crude gluten or 200 µg
purified peptide emulsified in CFA in the tail and hind footpad. The
draining lymph nodes were harvested 7 days postinjection and the cells
were isolated as described previously (22). Lymphocyte
proliferation was determined by incorporation of
[3H]thymidine. Results are expressed as
difference (
cpm) in incorporation and calculated as cpm =
mean triplicate counts in Ag-containing cultures minus mean triplicate
counts of cultures with medium alone.
Cytokine determination
Mice were immunized with 500 µg crude gluten or rice i.p. with CFA on day 0. On day 7, mice were fed 2 mg gluten (rice) by gavage. On day 14, intestinal T cells were isolated as previously described (23). Briefly, intestines were washed six times in CMF containing FCS for 15 min. Intestines were then incubated in a CMF/FBS/EDTA solution at 37°C for 15 min. The intestines were then treated with RPMI 10 and collagenase for 2 h. The resulting supernatants from the CMF/FBS/EDTA incubation and collagenase treatment were then combined and centrifuged. The resulting cell pellet was treated with pan T (anti-Thy1) Abs (Dynabeads; Dynal Biotech, Oslo, Norway). Intestinal T cells were incubated at a concentration of 1 x 106 cells/ml in a 0.5-ml volume in 24-well plates (Costar, Cambridge, MA). T cells were then incubated with gluten in the presence of irradiated spleen cells for 72 h. The cultures were then centrifuged, the cells were subjected to FACS analysis, and the supernatants were tested for cytokines. Cells were harvested after a 72-h incubation with gluten at 37°C and centrifuged, and the supernatants were collected for cytokine analysis.
Cytokine secretion was analyzed by ELISA using mini-kits for IL-2,
IL-4, IL-6, TNF-, IFN-
, IL-10, and TGF- from Duoset (Genzyme
Diagnostics, Cambridge, MA). All ELISAs were performed on 96-well
flat-bottom microtiter plates (Immulon 2; Dynatech Laboratories,
Chantilly, VA). Plates were read in a microtiter autoreader (Bio-Rad,
Pleasanton, CA) at 450 nm. Quantification of cytokine levels was
determined by comparison with standards.
Characterization of T cell population responding to gluten
Intestinal T cells were isolated as previously described (22). Briefly, intestines were washed six times in calcium, magnesium-free HBSS (CMF) containing FCS for 15 min. Intestinal pieces were then incubated in RPMI and collagenase twice for 1 h. T cells were purified using anti-Thy1.2 Dynabeads. They were then incubated for 72 h in the presence of gluten. mAbs for the following cell surface markers were used for flow cytometry: CD4, CD25, CD45RB, CD62 ligand (BD PharMingen, San Diego, CA).
Detection of anti-gliadin Abs
Serum was isolated from mice after immunization and subsequent feeding with gluten or rice. Anti-gliadin IgG and IgA Abs were detected by ELISA using a kit from Inova Diagnostics (San Diego, CA) using IgG and IgA standards for comparison.
Statistical analysis
A two-tailed Student’s t test was used to compare the means of responses in all assays, which were performed in triplicate. Three to five mice were used in each experimental group. A p value of <0.05 was considered significant.
Histological examinations
Cross-sections from the proximal and distal intestine were preserved in formalin and stained with H&E. Specimens were examined by light microscopy for changes in villous architecture, epithelial change, and inflammation.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The ability of the transgenic mice to respond to gluten was
examined using standard T cell proliferation assays.
HCD4/DQ8 and HCD4/DQ6 mice were injected with
gluten in the base of the tail and hind footpad (n =
5). One week later, draining lymph nodes were harvested and examined
for T cell responsiveness. As shown in Fig. 1
, both HCD4/DQ8 and
HCD4/DQ6 mice responded to gluten, but the T cells from
HCD4/DQ8 mice mounted a much greater proliferative response
than cells from the HCD4/DQ6 mice (cpm vs cpm). No
proliferation was seen in HLA class II-deficient mice (data not shown),
and no activity was detected in either strain of mice when rice was
used as the grain protein (Fig. 1).
|
To determine the subpopulation of T cells responsible for
proliferation to gluten, mAbs to CD4 and DQ were added to the wells.
Proliferation was inhibited (Fig. 2) when
anti-CD4 was added to the well. In addition, proliferation was
inhibited when anti-DQ mAb was added to the well, whereas
anti-DR had no effect. Taken together, these data indicate that the
in vitro response in HCD4/DQ8 transgenic mice is mediated by
CD4-expressing, DQ-restricted T lymphocytes.
|
Cultured supernatants from intestinal HCD4/DQ8 and
HCD4/DQ6 mice were assayed for the production of various
cytokines. No IFN- was found in the supernatants from either
transgenic mouse after immunization and gluten feeding. Similarly,
minimal amounts of IL-2 were seen in these cultures. IL-6 was expressed
at high levels in lymphocyte cultures from HCD4/DQ8 mice and
to a somewhat lesser extent in cultures from HCD4/DQ6 mice
after exposure to gluten (Fig. 3). IL-10
was produced at the same level in lymphocytes isolated from both
strains of mice after gluten immunization (Fig. 3B). The
level of IL-4 produced by cell culture in these mice was minimally
affected by gluten exposure. Interestingly, high levels of TGF- were
induced by gluten exposure in cultures from both strains of mice;
however, the HCD4/DQ8 cultures expressed more TGF- than
the HCD4/DQ6 mice (Fig. 3C). Similar results were
seen in cultures of splenic lymphocytes. Thus, contact with gluten
induces a T regulatory cytokine response in both strains of transgenic
mice, with greater levels of the cytokines produced by the sensitized
HCD4/DQ8 mice as compared with the less sensitized
HCD4/DQ6 mice.
|
The T cell response to specific antigenic determinants of gliadin
was studied using overlapping peptides. Lymph node cells from the
HCD4/DQ8 mice were able to respond to a greater variety of
synthetic peptides when compared with HCD4/DQ6 mice (Fig. 4). In particular, peptides corresponding
to amino acid residues 1–20 and 10–30 induced significant T cell
proliferation in the HCD4/DQ8 mice. Additionally, a peptide
spanning amino acids 206–217 also resulted in a strong proliferative
response.
|
We tested whether simulated deamidation of specific
residues in gliadin peptide 206–217 (SGQGSFQPSQQN) would increase the
antigenicity of the peptide in these transgenic mice. The amino acid of
the native gliadin peptides (capital letters) are replaced by glutamic
acid E. Peptides that were synthesized with deamidated
residues (italics) at positions 208(SGEGSFQPSQQN),
215(SGQGSFQPSEQN), and
216(SGQGSFQPSQEN) resulted in T cell responsiveness (Fig. 5), indicating that these residues may be
involved in regulating the immune response to gliadin and that
deamidation may be an important factor in inducing or amplifying the
response. Cells from DQ6 mice did not recognize these
altered epitopes.
|
Patients with celiac disease have anti-gliadin IgG and IgA
present in their sera. To determine whether the transgenic mice also
produced Abs to gliadin, HCD4/DQ8 and HCD4/DQ6
mice were bled and their sera were analyzed for the presence of IgG and
IgA Abs to gliadin. High levels of anti-gliadin IgG were found in
the HCD4/DQ8 mice. The HCD4/DQ6 mice produced
anti-gliadin IgG Abs, but the levels were barely detectable and
similar to levels in rice-fed HCD4/DQ8 mice (Fig. 6). No anti-gliadin IgA was
detectable in either strain of transgenic mouse. Endomysial IgA Abs are
recognized in human celiac disease. No endomysial IgA or IgG Abs were
detected in the mouse sera.
|
Several of the gluten-fed DQ8 mice showed poor growth
and developed a wasted appearance. However, no gross changes in
intestinal morphology or gross inflammation was identified in these
mice by the time of sacrifice. There was no change in villous
architecture, inflammation, or enterocyte height, characteristics most
often associated with celiac disease. Despite this, lymphocyte
activation by gluten was identified in the mucosa. Intestinal
lymphocytes were subjected to FACS analysis and stained for a variety
of activation markers (Fig. 7). Gluten
stimulated a population of cells in HCD4/DQ8 mice that have
the following phenotype:
CD4+CD25+CD62Llow.
This phenotype is consistent with an activated population of cells.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In humans, the HLA class II Ags have been associated with a number of autoimmune diseases, including CS, insulin-dependent diabetes mellitus, rheumatoid arthritis, and multiple sclerosis. Of these, celiac disease has the strongest HLA association (5). In fact, very few patients without the specific haplotypes have been described (8, 25). Although a particular HLA haplotype is virtually required to produce celiac disease, it is clear that those HLA alleles alone will not produce disease (8, 26).
We have demonstrated that both HCD4/DQ8 and HCD4/DQ6 transgenic mice elicit in vitro responses to gluten that are HLA-DQ restricted and dependent on CD4+ cells. The HCD4/DQ8 in vitro responses to gluten were stronger than those of the HCD4/DQ6 mice, although both strains were able to produce a response to the crude gluten preparation. In addition, the HCD4/DQ8 mice responded to a number of gliadin-derived peptides, while the HCD4/DQ6 responses were minimal in number and degree. Several of the gliadin peptides to which the HCD4/DQ8 mice respond have been implicated as being pathogenic in celiac disease in humans (11, 27).
While several studies have suggested that native unaltered peptides of gliadin were antigenic in celiac disease, gliadin lacks the negatively charged amino acids needed to bind to the recognition sites on the DQ2/8 molecules (28, 29, 30). It has since been recognized that the gliadin molecules (or peptide fragments derived therefrom) undergo changes that alter the binding characteristics of gliadin and, ultimately, its ability to induce a cellular immune response. Much recent work has described the probable role of tTG in the development or perpetuation of gluten sensitivity in celiac disease. tTG is involved in cell-to-cell signaling and extracellular matrix formation, and is released in the gut mucosa in the setting of inflammation (31, 32, 33, 34). It is also the target for the potent humoral autoantibody response seen in most patients with untreated celiac disease. In celiac disease, it is believed that tTG plays an important role in creating new epitopes of the gliadin molecule, with which the immune system must now contend (35). Gliadin is a preferred substrate for tTG, and it is certain that the enzyme deamidates the glutamine residues of gliadin to glutamic acid (11). Gliadin contains many glutamine residues, and peptides of gliadin can be made more antigenic when specific residues are deamidated (11, 12, 13, 14). This study demonstrates that the effect of this deamidated peptide is entirely restricted to the human disease-associated DQ genotype in the transgenic mouse.
tTG is released from fibroblasts in the setting of inflammation; this also occurs with infection or other perturbations of the mucosal integrity that might serve as triggering or enabling events for the development of enteropathy from gliadin sensitivity (31, 32, 33, 34, 35, 36). Deamidation of the specific peptides is believed to result in improved binding to specific pockets in the DQ molecule.
Several studies have reported that naturally occurring gliadin epitopes cannot be recognized by T cells cloned from the intestinal mucosa patients with established disease, although they are recognized by T cells derived from peripheral blood (37, 38, 39). The assertion that deamidation of the gliadin peptides is essential to celiac disease is based almost exclusively on results of experiments in which T cells cloned from the active established lesion of celiac disease are used as a measure of responsiveness (10, 11, 12, 24, 39, 40, 41, 42, 43). The exception to this is the 11-mer native peptide of gliadin (206–217) that can induce proliferation in DQ8 T cell lines in its native form (41). Interestingly, a similar peptide overlapping 206–217 induced a jejunal inflammation when instilled into the jejunum of a patient with celiac disease (24). Our observations in the DQ8 mouse suggest that both native gliadin peptides and deamidated peptides can evoke T cell responses in a DQ-restricted fashion. T cell cultures derived from the biopsies of children with recent onset of disease respond to a narrower range of gluten peptides than adults (42). Thus, it is possible that naturally occurring epitopes of gliadin may be involved in the initiation of gluten sensitivity. The deamidated epitopes become important in the development or maintenance of the enteropathy, possibly as the result of epitope spreading or a second underlying predisposition or trigger to immune dysfunction.
This mouse model provides a method to study the earliest process in the initiation of gluten hypersensitivity before the chronic lesion has become established. The ability to generate an HLA-DQ-restricted response to naturally occurring epitopes supports the concept that naturally occurring peptides are responsible for the initiation of disease (or at least gluten sensitivity) (24). This does not exclude the possibility that altered peptides (deamidation by tTG or otherwise) are also antigenic either during initiation of disease or later in the disease. We demonstrated that DQ8 can present these naturally occurring peptides to T cells and that CD4+CD25+ activated cells are seen in response to gluten in the small intestine of the mouse. It is, of course, possible that the mouse tTG may deamidate the peptides in vivo. If that is so, it is occurring in the absence of obvious inflammation. The absence of circulating tTG Abs (data not shown) and the use of synthetically generated peptides preclude the issues previously used to explain the conflicting findings using purified gluten proteins.
The failure of these mice to develop enteropathy may be due to a number of factors. These transgenic mice may lack some other genes that contribute to the development of celiac disease. It is known that in humans the MHC genes contribute just 40% of the risk of transmission of celiac disease in families (26). These mice may lack another genetic factor or environmental stimulus such as bacterial or viral pathogen that may be involved in triggering enteropathy. It is also possible that the exposure to gluten was not of sufficient duration to induce appreciable damage.
Celiac disease is associated with a predominantly Th1 or Th0 type of
cytokine response in established disease (44). The
cytokine profile in these mice resembles that of a regulatory phenotype
and thus differs from that seen in established celiac disease. The
regulatory cytokine response seen may reflect an appropriate and
effective regulatory response in the setting of increased sensitivity,
as occurs in the gluten-immunized DQ8 mouse. Regulatory T cells
secreting TGF- and IL-10 would certainly down-regulate inflammation,
and overcoming this may be one of the first steps to producing the
celiac lesion. TGF- has complex effects in the intestinal mucosa,
including regulation of T and B cell responses. It is also important in
maintaining or healing the epithelial structure (45, 46, 47).
Recent work with another mouse model wherein TGF- receptor signaling
has been rendered defective in the gut mucosa results in an
inflammatory enteropathy that has histologic features which mimic the
enteropathy of celiac disease (48). These mice also have
circulating Abs to tTG (49). The specific regulatory
responses generated may be part of the intact control mechanisms that
occur during the mucosal response to gluten even in the sensitized
mouse. This regulation may prevent (or protect the mouse from
developing) the enteropathy associated with CS. This implies that this
mouse models gluten sensitivity that is HLA determined but lacks the
probable second defect/trait or changes in the mucosal environment
required for enteropathy. It is possible that the mouse background
genes are not prone to inflammation or that a second necessary luminal
trigger for the conversion from sensitivity to enteropathy (other than
gluten) may be absent in the specific pathogen-free environment in
which these animals are kept. A particular microbial environment may be
required to stimulate a pathologic immune response.
The study of celiac disease has been hampered by the lack of readily usable animal models in which to study both the immunogenetics and pathogenesis of the gluten sensitivity and the subsequent enteropathy. This class II transgenic mouse model provides a potentially powerful tool with which to study the mechanism underlying the initiation of disease, i.e., the gliadin sensitivity. Our model presents a unique opportunity to dissect the genetic basis for predicting the human–food interaction that leads to celiac disease. This gluten-sensitive mouse model provides a means to examine potential interventions, such as induction of anergy, vaccination, or other therapies that may be aimed at preventing the loss of or induction of tolerance. The ability to demonstrate an HLA-restricted response demonstrates that this component of celiac disease can be reproduced without the other human genetic factors involved in celiac disease.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Joseph A. Murray, Division of Gastroenterology and Hepatology, Mayo Clinic and Foundation, 200 1st Street, SW, Rochester, MN 55905.
3 Abbreviations used in this paper: CS, celiac sprue; tTG, tissue transglutaminase; CMF, calcium, magnesium-free HBSS.
Received for publication May 10, 2002. Accepted for publication September 9, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
heterodimer explains the divergent HLA-DR associations observed in various Caucasian populations. Tissue Antigens 36:136.[Medline]
/ heterodimer. J. Exp. Med. 169:345.1*0501,1*0201) restricted T-cells isolated from the small intestinal mucosa of celiac disease patients. J. Exp. Med. 178:187.-gliadin in adult celiac disease is focused on a single deamidated glutamine targeted by tissue transglutaminase. J. Exp. Med. 191:603..1*0501,1*0201) molecule. Eur. J. Immunol. 26:2764.[Medline]
.1*0501,1*0201) molecule. Immunogenetics 44:246.[Medline]
-gliadin presented by the celiac disease-associated HLA-DQ (1*0501, 1*0201) heterodimer. Hum. Immunol. 39:243.[Medline]
in promoting T helper cell type 1 responses in the small intestine in coeliac disease. Gut 48:425. expression in normal intestinal mucosa and in active celiac disease. J. Pediatr. Gastroenterol. Nutr. 29:308.[Medline]
/Smad signaling defects in inflammatory bowel disease: mechanisms and possible novel therapies for chronic inflammation. J. Clin. Invest. 108:523.[Medline]
1 preserves epithelial barrier function: identification of receptors, biochemical intermediates, and cytokine antagonists. J. Cell. Physiol. 181:55.[Medline]
signaling in the intestine contributes to tissue injury in inflammatory bowel disease. Gut 49:190. signaling in small intestine. Gastroenterology 122:A23.This article has been cited by other articles:
|
|
 |
|
  I. L. Huibregtse, E. V. Marietta, S. Rashtak, F. Koning, P. Rottiers, C. S. David, S. J. H. van Deventer, and J. A. Murray Induction of Antigen-Specific Tolerance by Oral Administration of Lactococcus lactis Delivered Immunodominant DQ8-Restricted Gliadin Peptide in Sensitized Nonobese Diabetic Ab{degrees} Dq8 Transgenic Mice J. Immunol., August 15, 2009; 183(4): 2390 - 2396. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. de Kauwe, Z. Chen, R. P. Anderson, C. L. Keech, J. D. Price, O. Wijburg, D. C. Jackson, J. Ladhams, J. Allison, and J. McCluskey Resistance to Celiac Disease in Humanized HLA-DR3-DQ2-Transgenic Mice Expressing Specific Anti-Gliadin CD4+ T Cells J. Immunol., June 15, 2009; 182(12): 7440 - 7450. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. F. Verdu, X. Huang, J. Natividad, J. Lu, P. A. Blennerhassett, C. S. David, D. M. McKay, and J. A. Murray Gliadin-dependent neuromuscular and epithelial secretory responses in gluten-sensitive HLA-DQ8 transgenic mice Am J Physiol Gastrointest Liver Physiol, January 1, 2008; 294(1): G217 - G225. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Senger, F. Maurano, M. F. Mazzeo, M. Gaita, O. Fierro, C. S. David, R. Troncone, S. Auricchio, R. A. Siciliano, and M. Rossi Identification of Immunodominant Epitopes of {alpha}-Gliadin in HLA-DQ8 Transgenic Mice following Oral Immunization J. Immunol., December 15, 2005; 175(12): 8087 - 8095. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Stickler, N. Rochanayon, O. J. Razo, J. Mucha, W. Gebel, N. Faravashi, R. Chin, S. Holmes, and F. A. Harding An In Vitro Human Cell-Based Assay to Rank the Relative Immunogenicity of Proteins Toxicol. Sci., February 1, 2004; 77(2): 280 - 289. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. H. Cross and P. T. Golumbek Neurologic manifestations of celiac disease: Proven, or just a gut feeling? Neurology, May 27, 2003; 60(10): 1566 - 1568. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
