Article Text

Cross-reacting antibodies in coeliac disease?
  1. Department of Gastroenterology,
  2. Ullevaal Hospital,
  3. 0407 Oslo,
  4. Norway

Statistics from

See article on page 168

Most patients with coeliac disease have antibodies to wheat gliadin, reticulin, and endomysium. In 1997 a seminal paper showed that at least a substantial fraction of anti-endomysial antibodies (EMA) recognises the endogenous enzyme tissue transglutaminase (tTG).1 However, antibodies recognising other antigens can also be found and in this issue (see page 168) Krupičková and colleagues attempt to characterise these antibodies.

Anti-gliadin antibodies (AGA) were isolated from coeliac serum samples using (semi)purified α-gliadin as the substrate. These antibodies were tested for specificity using a synthetic α-gliadin peptide competition assay. This important study is the first of its kind. An interesting finding is that the antibody responses are directed towards a limited set of epitopes. These epitopes do not overlap with peptides recognised by small intestinal HLA-DQ restricted T cells,2 3 but our current knowledge is too limited to judge whether this is important. One of the epitopes was the VLPVQQQQF peptide, which corresponds to α-gliadin residues 22–30. The glutamins were important, as substitution with glutamic acid removed the inhibitory function of VLPVQQQQF. Conversion of Qs to Es (deamidation) by tTG has been implicated recently in the pathogenesis of coeliac disease.4 5 However, tTG acts on T cell epitopes of gliadin by specific deamidation of only some of the glutamins and it probably would have been more realistic if only one or some of the Qs in the 22–30 peptide had been replaced.

Although recent studies have focused on coeliac antibodies recognising tTG, other antibody specificities have been reported. Mäkiet al described several antigens recognised by EMA but which were not recognised by AGA.6 Börneret al isolated various components from different animal tissues using serum from patients with coeliac disease.7 It is unlikely that the antigens reported in these two studies are tTG. As mentioned by Krupičková and colleagues, antibodies cross-reacting with gliadin, enterocytes and calreticulin can also be found. Interestingly, these authors have shown that some of the same peptides that interfere with AGA binding to α-gliadin also effect binding to enterocytes and calreticulin. By doing a sequence similarity search they identified corresponding sequences in α-gliadin and calreticulin. However, it would have been reassuring to see whether synthetic calreticulin peptides could also inhibit binding of AGA to α-gliadin, and further details of the calreticulin preparation used are essential. As with any new and unexpected finding it would also be good to see the cross-reactivity between gliadin and calreticulin reproduced by other investigators.

Can one now conclude that coeliac disease is an autoimmune condition directed against enterocytes and calreticulin? There are some difficulties in accepting such a hypothesis. Calreticulin is an abundant Ca2+ binding protein which is expressed in every cell in higher organisms.8 It has a retrieval signal for endoplasmic reticulum (ER) and the ER is considered to be the major cellular site of localisation. Small amounts of calreticulin may also be present at extra-ER sites including the cell surface. At any rate, an immunological cross-reaction would presumably manifest itself in different organs. Coeliac disease is seen more frequently in IgA deficient individuals, so at least IgA antibodies are not necessary for the disease. If the cross-reactivities were attributable to IgG antibodies, they could give rise to the complement activation known to be present in coeliac lesions.9 We know that IgG-AGA are not specific as they can be found both in healthy subjects and those with coeliac disease. However, this may not necessarily reflect what is going on in the small intestine. Conversely, cross-reactive AGA could have an effect during the first phases of disease pathogenesis directly following α gliadin challenge,10 where the observed phenomenon might fit with a rapid antibody recognition event. Whether cross-reactive antibodies recognising enterocytes are capable of inducing apoptosis is still an open question.

Finally, a comment can be made on the usage of the “molecular mimicry” model as an explanation for the putative autoimmune component in coeliac disease. As Michael Bevan defined it, molecular mimicry is important for the induction phase of autoimmunity, where an infectious agent triggers an autoimmune loop, which persists even after the infection has been cleared. The complete remission seen in almost all coeliac patients after withdrawal of cereal proteins from the diet is difficult to reconcile with this concept.

See article on page 168


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles