The purpose of this study was to determine whether avenin could cause immune activation of cultured duodenal biopsies taken from patients with coeliac disease. The parameters investigated included quantification of IFN-γ and IL-2 mRNA and also measurement of secreted cytokines in culture supernatants. Avenin failed to induce a statistically significant cytokine response in the eight coeliac patients studied: no significant increase in the level of mRNA or protein for IFN-γ or IL-2 was observed. In contrast, gliadin addition to duodenal biopsy cultures from a further nine treated coeliac patients caused immune activation in these subjects.
In gliadin stimulated coeliac biopsy cultures, production of IFN-γ was more prominent than that of IL-2. An increase in IFN-γ mRNA was observed in all nine patients with very high levels present in four subjects. Moreover, an increase in IFN-γ protein was found in these four patients. In contrast, although IL-2 mRNA was increased in six subjects after gliadin stimulation, the mean IL-2 message level was 14-fold lower than quantities of mRNA for IFN-γ. Furthermore, only low levels of IL-2 protein were present and an increase in this cytokine was found in two subjects following gliadin activation. These findings were specific for biopsies from coeliac subjects and no cytokine products were observed in the gliadin stimulated biopsy cultures from control subjects.
The four hour culture period was chosen based on published cytokine mRNA kinetic experiments.16
Nilsen et al
found that optimal mRNA production for IFN-γ and other cytokines (including IL-2, IL-4, IL-6, and tumour necrosis factor α) following gluten stimulation of duodenal organ cultures was between two and six hours. However, such short term cultures are not optimal for cytokine protein production and in their study Nilsen and colleagues16
were unable to detect cytokines in the culture supernatant after two or six hours of gluten stimulation. Failure to detect soluble cytokines might be explained by delayed synthesis of these proteins or because secreted cytokine bound rapidly to specific receptors. None the less, IFN-γ was detected by immunohistochemical staining of fresh biopsy tissue in patients with untreated coeliac disease.16
In the current study, after four hours of organ culture with gliadin, IFN-γ protein production was detected in four subjects and interestingly these patients had the highest levels of mRNA for IFN-γ. Moreover, two of these patients also secreted IL-2 protein. The finding of secreted cytokines is in keeping with reports of increased tissue expression of several key immune response molecules (HLA-DR and intercellular adhesion molecule 1 (ICAM-1)) within two hours of gluten stimulated biopsy culture.22
IFN-γ is known to cause upregulation of both molecules on epithelial cells.23–26
Some coeliac patients showed evidence of mucosal damage and two patients were EMA positive. However, spontaneous or gliadin induced IFN-γ production did not correlate with the severity of the patient's mucosal lesion or EMA status. For example, patient Nos 1, 2, 7, and 8 produced the highest levels of IFN-γ mRNA following gliadin stimulation and yet all had normal or mildly damaged mucosae and were EMA negative. In contrast, patient Nos 4 and 6 whose biopsies showed severe villous blunting, produced little IFN-γ mRNA, even after gliadin stimulation. These results are of interest as in earlier studies levels of cytokine mRNA and protein were higher in patients with untreated coeliac disease.16,27–31
The different findings may relate to the time point chosen for cytokine detection and the relatively small number of study subjects. Although there was a trend for more severe histological damage in biopsies cultured with gliadin, as each prolamin result was compared with spontaneous cytokine synthesis, this did not affect the validity of the results.
The finding of a marked IFN-γ mRNA and protein response in coeliac tissue is in keeping with several previous reports. These studies were based on investigation of fresh biopsy tissue,29
intestinal tissue after organ culture,16
and T cells cloned from duodenal biopsies.32–34
In addition, IFN-γ was produced by peripheral blood mononuclear cells from coeliac subjects in response to unfractionated gliadin35
or gliadin peptide stimulation.36
Thus like many other chronic inflammatory disorders such as Crohn's disease, rheumatoid arthritis, and insulin dependent diabetes mellitus,37
the so-called Th1 response seems to be dominant in coeliac disease. This is keeping with the concept that a delayed-type hypersensitivity response to gluten is central to the pathogenesis of coeliac disease. However, precisely how this leads to the coeliac mucosal lesion is not understood.
Organ culture of treated coeliac intestinal biopsies, in the presence of gluten or gluten fractions, is a traditional in vitro experimental approach used in the study of coeliac disease. Until recently, these studies relied on detailed morphometric evaluation of the tissue, examining features such as reduced enterocyte height38,39
and increased infiltration of the epithelial layer by lymphocytes.15
However, in these experiments sizable properly orientated biopsies are crucial to the success of the study. With the increasing use of endoscopic forceps rather than the Crosby capsule to obtain biopsies, few studies now rely on morphometry to reflect tissue changes. In more recent studies, immunohistochemical staining of biopsy tissue has been employed and changes in the level of expression of molecules such as HLA-DR, ICAM-I, CD25,22
and various cytokines27,28
have been investigated. However, a level of subjectivity is inevitable in this experimental procedure. It was for this reason that in our study objective quantitative methods were used, employing realtime PCR for determination of mRNA and ELISA for measurement of secreted cytokines. A disadvantage of this approach is the restriction on the number of experiments which can be performed: a total of four biopsies were required to obtain sufficient mRNA for analysis of one culture condition—for example, studying the effect of avenin. Thus it was not possible to compare the response to two different cereal fractions in the same individual.
In earlier organ culture studies of prolamin toxicity, a concentration of 1 mg/ml of protein was commonly employed and found to be sufficient to induce a mucosal response.22,28,38,40,41
In the present study, even when 5 mg/ml of avenin were added to the biopsy cultures, there was no evidence of cytokine production. An equivalent amount of gliadin did activate the coeliac mucosa.
Because avenin accounts for only 5–15% of the total protein in oats (whereas wheat, barley, and rye prolamins constitute 40–50%), it has been suggested that larger quantities of oats may still be toxic to coeliac patients. However, the findings of this study argue against this and demonstrate that purified avenin is not immunogenic to the coeliac mucosa. These results concur with several studies which reported that oats caused no adverse clinical effects and did not result in immune activation.7,8,10
Moreover, they are in agreement with a short clinical challenge study of two patients with dermatitis herpetiformis given 2.5 g of purified avenin for five days without adverse effect.42
In conclusion, the findings of this study provide evidence that the immunogenic sequences in gliadin are not present in avenin and support the belief that oats are safe for consumption by coeliac patients.