Secretory IgA mediates retrotranscytosis of intact gliadin peptides via the transferrin receptor in celiac disease - PubMed (original) (raw)

. 2008 Jan 21;205(1):143-54.

doi: 10.1084/jem.20071204. Epub 2007 Dec 31.

Ivan Cruz Moura, Michelle Arcos-Fajardo, Corinne Lebreton, Sandrine Ménard, Céline Candalh, Karima Ben-Khalifa, Christophe Dugave, Houda Tamouza, Guillaume van Niel, Yoram Bouhnik, Dominique Lamarque, Stanislas Chaussade, Georgia Malamut, Christophe Cellier, Nadine Cerf-Bensussan, Renato C Monteiro, Martine Heyman

Affiliations

Secretory IgA mediates retrotranscytosis of intact gliadin peptides via the transferrin receptor in celiac disease

Tamara Matysiak-Budnik et al. J Exp Med. 2008.

Abstract

Celiac disease (CD) is an enteropathy resulting from an abnormal immune response to gluten-derived peptides in genetically susceptible individuals. This immune response is initiated by intestinal transport of intact peptide 31-49 (p31-49) and 33-mer gliadin peptides through an unknown mechanism. We show that the transferrin receptor CD71 is responsible for apical to basal retrotranscytosis of gliadin peptides, a process during which p31-49 and 33-mer peptides are protected from degradation. In patients with active CD, CD71 is overexpressed in the intestinal epithelium and colocalizes with immunoglobulin (Ig) A. Intestinal transport of intact p31-49 and 33-mer peptides was blocked by polymeric and secretory IgA (SIgA) and by soluble CD71 receptors, pointing to a role of SIgA-gliadin complexes in this abnormal intestinal transport. This retrotranscytosis of SIgA-gliadin complexes may promote the entry of harmful gliadin peptides into the intestinal mucosa, thereby triggering an immune response and perpetuating intestinal inflammation. Our findings strongly implicate CD71 in the pathogenesis of CD.

PubMed Disclaimer

Figures

Figure 1.

Figure 1.

IgA and IgA–gliadin complexes are overexpressed in CD. (A) Analysis of IgA immune complexes (IC) isolated from the serum and duodenal secretions of patients with active or treated CD and controls. PEG precipitates containing high molecular weight IC were analyzed by ELISA. Plates were coated with Frazer's fraction (pepsin/trypsin gliadin hydrolysate) or anti-Tgase antibody, followed by anti–human IgA–horseradish peroxidase. Results are presented as OD values obtained with all sera or duodenal secretions tested. A significant increase in IC recognizing gliadin peptides and containing Tgase was observed in the active CD group compared with the treated CD and control groups in both serum and duodenal secretion. Horizontal lines represent medians. *, P < 0.0005 and 0.01 compared with the control and treated CD groups, respectively; #, P < 0.01 and 0.02 compared with the control and treated CD groups, respectively. (B) IgA is concentrated at the apical pole of surface enterocytes in active CD. Cryosections of duodenal mucosa from control subjects, patients with active or treated CD, and patients with refractory celiac sprue were labeled with anti-IgA–FITC antibody and TOPRO-3 (blue nuclei). In controls, treated CD patients, and patients with refractory celiac sprue, IgA staining of epithelial cells was located at the basal pole of villous cells and at the apical and basal poles of crypt cells. In contrast, in patients with active CD, IgA staining was concentrated at the apical pole of the surface epithelium (and of crypt cells; Fig. S1). Results are representative of three controls, seven patients with treated CD, and six patients with active CD. (C) IgA overexpression observed at the apical pole of epithelial cells of patients with active CD was located inside the cell, including the brush border membrane, as shown by its colocalization with cytokeratin and alkaline phosphatase, a marker of apical membranes. No such colocalization was seen in controls. Bars, 50 μm.

Figure 2.

Figure 2.

CD71 expression and colocalization with IgA and p31-49 in duodenal biopsies. (A) Expression of CD71 (immunoperoxidase labeling) on duodenal biopsies from controls, patients with active or treated CD, and patients with refractory celiac sprue. Compared with controls, CD71 was overexpressed in patients with active CD and in patients with refractory celiac sprue. At higher magnification (bottom), strong CD71 expression was observed all over the surface epithelium in patients with active CD, whereas in controls and treated CD patients CD71 expression was only observed at the basal pole of villous epithelial cells and in crypt cells. (B) CD71 overexpression by surface epithelium of patients with active CD was confirmed by immunofluorescent labeling. The fluorophore was a cy5-conjugated secondary antibody (blue staining). Results in A and B are representative of three control subjects, four treated CD patients, eight patients with active CD, and two patients with refractory celiac sprue. (C) Double immunofluorescence labeling of IgA–CD71 in duodenal biopsies. Colocalization (white) of IgA (green) and CD71 (blue) was observed at the apical surface of the epithelium in active CD (n = 5), but not in controls or in treated CD patients (n = 3). (D) Immunogold electron microscopy with double labeling of IgA (10-nm particles; arrowhead) and CD71 (15-nm particles; arrow). In active CD, IgA and CD71 were expressed in the brush border membrane and subepithelial compartments, and IgA–CD71 colocalization was frequent (boxes). No such colocalization was observed in controls (see additional images of one control subject and two patients with active CD in Fig. S2). (E) SIgA can bind CD71 at the cell surface of a B cell line (Daudi cells) known to express CD71 as the only IgA receptor. Cells were incubated for 30 min at 4°C with 500 μg/ml SIgA or pIgA1 in the presence or absence of 500 μg/ml of soluble CD71. IgA was revealed with biotinylated anti-IgA and allophycocyanin-labeled streptavidin (green line). Both pIgA1 and SIgA specifically bound CD71, as the binding was inhibited by soluble CD71 receptors (pink line). The black line indicates the isotope control. (F) Colocalization (yellow) of IgA (green) and p31-49 (red) in duodenal biopsies from two patients with active CD, mounted in Ussing chambers and exposed to p31-49–TAMRA on the apical side for 15 min at 37°C before being fixed, cryosectioned, and stained with anti-IgA–FITC antibodies. No colocalization was found in two treated CD patients or in a control (not depicted). Bars, 50 μm.

Figure 3.

Figure 3.

IgA involvement in intestinal transport and processing of 3H-labeled p31-49. (A) Transport and processing of p31-49 showing typical RP-HPLC elution pattern of 3H-labeled material in the basal compartment of duodenal biopsies incubated for 3 h after apical addition of 3H-labeled p31-49. In controls and treated CD patients, p31-49 was almost completely degraded during transport, as >95% of the total radioactivity was eluted as free 3H-labeled proline in the basal compartment. In contrast, in patients with active CD (n = 7), a large fraction of p31-49 was found on the basal side of the intestinal mucosa, mainly in intact form or as active fragments. Interestingly, in three patients with refractory celiac sprue (flat mucosa and an absence of antigliadin IgA), near-complete degradation of the peptide was observed after intestinal transport, suggesting that a flat mucosa is not responsible for the transport of intact peptide observed in patients with active CD. (B) Mean percentage of tritiated intact p31-49, active fragments, and proline found in the basal compartment after intestinal transport of p31-49 by duodenal biopsies from controls, patients with treated CD, and patients with active CD. The percentage of intact p31-49 plus active fragments crossing the duodenal biopsies (mean ± SD) was significantly higher in active CD (57 ± 18%; n = 17) than in treated CD (23 ± 23%; n = 8) and controls (26 ± 4%; n = 4). *, P < 0.007. (C) Inhibitory effect of pIgA, SIgA, and mIgA on the transport of intact p31-49. To test the involvement of IgA in the transport of intact p31-49, we performed competitive inhibition experiments with different forms of IgA. In a typical RP-HPLC elution profile of 3H-labeled p31-49 obtained in biopsies from the patient with active CD shown in A, 85% of p31-49 was transported intact in basal conditions, whereas this percentage fell sharply in the presence of pIgA and SIgA but not mIgA (mIgA does not bind significantly to CD71; reference 31). (D) Intestinal transport of p31-49 in patients with active CD showing the percentage of intact p31-49 plus active fragments found in the basal compartment of Ussing chambers after blockade with mIgA, dIgA, pIgA, or SIgA. Compared with “peptide alone” (median = 50; n = 7), dIgA (median = 23; n = 3), pIgA (median = 0; n = 3), and SIgA (median = 0; n = 5), but not mIgA (median = 60; n = 3), significantly inhibited the intestinal transport of p31-49. *, P < 0.01 compared with peptide alone. (E) Effect of IgG and Tf on the transport of intact p31-49. 50 μg/ml IgG and 10 μg/ml Tf were preincubated for 30 min on the apical side of duodenal biopsies mounted in Ussing chambers before adding 3H-labeled p31-49. The basal compartment was collected after 3 h and analyzed by radio RP-HPLC to detect p31-49 and its metabolites. No inhibitory effect on p31-49 transport was observed.

Figure 4.

Figure 4.

Inhibitory effect of soluble CD71 on the transport of intact p31-49. (A) Typical HPLC elution profiles of p31-49 after intestinal transport across duodenal biopsies from patients with active CD. 3H-labeled radioactive material is present in the basal compartment of the duodenal biopsies in Ussing chambers 3 h after adding 3H-labeled p31-49 to the apical compartment. In basal conditions, intact p31-49 or active fragments were present in the basal compartment. Soluble CD71 (sCD71) reduced the transport of intact p31-49, whereas soluble CD89 (sCD89) had no effect. (B) Inhibition of intestinal transport of 3H-labeled p31-49 in the presence of sCD71. The histogram shows the mean percentage of intact p31-49 and its active fragments found in the basal compartment after intestinal transport. Compared with peptide alone (median = 50; n = 6), significant inhibition was observed in the presence of 30 μg/ml sCD71 (median= 21; n = 6) but not sCD89 (median = 46; n = 4). The horizontal lines indicate median values, and dotted lines join paired results from the same patient. *, P < 0.01 compared with peptide alone.

Figure 5.

Figure 5.

Duodenal transport of 33-mer in patients with active CD, and competitive inhibition by IgA and soluble CD71. (A) Typical RP-HPLC elution profile of 3H-labeled 33-mer after intestinal transport across duodenal biopsies from a control individual and a patient with active CD, mounted in Ussing chambers. 3H-labeled radioactive material present in the basal compartment 3 h after adding 3H-labeled 33-mer to the apical compartment is shown. The percentages of the different eluted fractions (proline, small and large fragments, and intact 33-mer) were quantified with Radiostar software. The control tissue almost totally degraded the 33-mer peptide, whereas digestion was incomplete in the sample from the patient with active CD. (B, left) Mean percentages of 33-mer and its fragments after intestinal transport. The duodenal mucosa of patients with active CD does not completely degrade 33-mer, as 38 and 22% of intact peptide and large fragments, respectively, were recovered in the basal compartment, compared with 4 and 9% in control subjects. Treated CD patients had an intermediate profile (12 and 8%). (right) dIgA, pIgA, and soluble CD71 (sCD71) significantly inhibited the transport of intact 33-mer but not of large fragments. *, P < 0.04 compared with control; #, P < 0.04 compared with peptide alone.

Figure 6.

Figure 6.

Overview of postulated Tf receptor (CD71)–mediated transport of IgA–gliadin complexes in CD. In healthy individuals, gliadin peptides (resistant to luminal degradation) are taken up nonspecifically by enterocytes and are degraded by lysosomal acid proteases during fluid-phase transcytosis. Very few toxic peptides are delivered into the intestinal lamina propria. In patients with active CD, abnormal expression of CD71 (Tf receptor) at the apical pole of enterocytes allows receptor-mediated uptake of SIgA–gliadin peptide complexes and their protected transport toward the lamina propria and, thus, toward the local immune system. The exact part of the SIgA molecule involved in CD71 binding is not known. Blockade of gliadin peptide entry into the intestinal mucosa might serve as the basis for a novel therapeutic strategy in CD.

References

    1. Fasano, A., I. Berti, T. Gerarduzzi, T. Not, R.B. Colletti, S. Drago, Y. Elitsur, P.H. Green, S. Guandalini, I.D. Hill, et al. 2003. Prevalence of celiac disease in at-risk and not-at-risk groups in the United States: a large multicenter study. Arch. Intern. Med. 163:286–292. - PubMed
    1. Maki, M., K. Mustalahti, J. Kokkonen, P. Kulmala, M. Haapalahti, T. Karttunen, J. Ilonen, K. Laurila, I. Dahlbom, T. Hansson, et al. 2003. Prevalence of Celiac disease among children in Finland. N. Engl. J. Med. 348:2517–2524. - PubMed
    1. Arentz-Hansen, H., R. Korner, O. Molberg, H. Quarsten, W. Vader, Y.M. Kooy, K.E. Lundin, F. Koning, P. Roepstorff, L.M. Sollid, and S.N. Mcadam. 2000. The intestinal T cell response to α-gliadin in adult celiac disease is focused on a single deamidated glutamine targeted by tissue transglutaminase. J. Exp. Med. 191:603–612. - PMC - PubMed
    1. Shan, L., O. Molberg, I. Parrot, F. Hausch, F. Filiz, G.M. Gray, L.M. Sollid, and C. Khosla. 2002. Structural basis for gluten intolerance in celiac sprue. Science. 297:2275–2279. - PubMed
    1. Kim, C.Y., H. Quarsten, E. Bergseng, C. Khosla, and L.M. Sollid. 2004. Structural basis for HLA-DQ2-mediated presentation of gluten epitopes in celiac disease. Proc. Natl. Acad. Sci. USA. 101:4175–4179. - PMC - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources