Background
Celiac disease is a gluten-induced autoimmune disease of the small intestine characterized by small-bowel mucosal villous atrophy with crypt hyperplasia and a profound inflammation in the mucosa. In addition to causing damage to the mucosa in genetically susceptible individuals, gluten also provokes the production of autoantibodies typically found in the sera of untreated celiac disease patients. These autoantibodies recognize exclusively endomysial antigens now identified as transglutaminase 2 (TG2). The autoantibodies are produced locally in the mucosa [
1,
2], and besides being detectable in patient sera, they are also deposited extracellularly
in vivo in the mucosa [
3‐
7]. Furthermore, recent findings suggest that these TG2-targeted mucosal IgA-autoantibody deposits are already present in the early phases of the disease process prior to manifest mucosal lesion [
4,
5] and before autoantibodies appear in the serum [
3‐
6]. After adoption of a gluten-free diet, serum autoantibodies disappear and the small-bowel mucosa heals usually within one year [
8]. At the time the autoantibodies have disappeared from the circulation, there may still be residual autoantibody deposits present in the small-intestinal mucosa which will also in due course disappear on a strict gluten-free diet [
4,
6].
For over twenty years the human small-intestinal organ culture method has been widely used in detecting the toxic effects of wheat gliadin in celiac disease
in vitro [
1,
9‐
15]. In earlier studies the toxicity of gliadin has commonly been demonstrated by an increased density of
lamina proprial lymphocytes [
11‐
13] and reduced epithelial cell height (ECH) [
14,
16] in cultured biopsy samples from untreated and treated celiac disease patients. However, when measuring endomysial autoantibody (EmA) secretion to culture supernatants from biopsies from treated celiac patients, the results have been contradictory. Picarelli and associates [
1] showed that in biopsies derived from treated celiac disease patients gliadin induces secretion of EmA to culture supernatants. In contrast, some studies report that the secretion of EmA can only be achieved in half of [
14,
17] or even no [
18] experiments carried out with treated celiac patient biopsies. And even further, it has been suggested that EmA secretion to the organ culture system is totally independent of gliadin challenge [
18] and histological damage [
14].
Due to these discrepancies among previous studies concerning autoantibody secretion to the organ culture system, our aim was to establish in both short- and long-term treated celiac disease patients whether the antibody secretion to culture supernatants is dependent on the duration of patients' gluten-free diet (GFD). Furthermore, since in celiac disease the small-bowel mucosal extracellular TG2-specific IgA deposits seem to disappear slowly after a gluten-free diet [
6], we hypothesized that the autoantibody secretion to supernatants is related to the existence of mucosal TG2-specific IgA deposits in the cultured small-bowel biopsy samples.
Discussion
The results presented here show that gliadin toxicity, as measured by autoantibody secretion, decrease in ECH and increase in the density of lamina proprial CD25+ cells, can be demonstrated in the treated celiac disease patient organ culture system only if small-intestinal mucosal TG2-specific IgA autoantibody deposits are still present in the cultured biopsy samples. Patients with persisting mucosal autoantibody deposits, normal small-bowel mucosal villous structure and negative serum autoantibody levels are usually those who have been on a gluten-free diet for a short period of time.
Our results explain the discrepancies reported in previous papers on EmA secretion in biopsies derived from treated celiac disease patients. Vogelsang and colleagues [
18] reported that EmA secretion after a 24-hour gliadin challenge was achieved in only 2 out of 18 (11%) biopsy samples derived from celiac disease patients on a strict gluten-free diet with normal mucosal villous morphology. In contrast, biopsies from 60% of patients with dietary lapses and 92% of untreated celiacs with ongoing mucosal inflammation responded to the challenge in respect of EmA secretion. Similar results have been obtained by Biagi and associates [
14] in a small-scale study involving only seven treated celiac disease patients. EmA secretion to the culture supernatant could not be stimulated after a 24-hour gliadin administration in any of the long-term (8–30 years) treated celiac patients with normal mucosal architecture. Samples responding to the gliadin insult were derived from patients still evincing mucosal damage despite a gluten-free diet. In the above-mentioned studies the treated patients unable to react to gliadin as measured by EmA secretion were well and long-term treated and may be hypothesized not to have small-bowel IgA autoantibody deposits. This absence of mucosal deposits could thus explain the discrepancy in results. Furthermore, it has been shown that 24 hours of gliadin treatment might not be a sufficient time to induce EmA secretion, as shown by Picarelli and colleagues [
19] who demonstrated that a 48-hour culture period in the presence of gliadin is needed to achieve EmA secretion in all treated celiac disease patient samples, this also being in agreement with the results presented in the current study (Table
2).
It remains to be elucidated how a gliadin challenge leads to EmA secretion in the organ culture system. Basically there are two different possibilities. Firstly, EmA could be actively secreted by plasma cells in the biopsy specimen [
1]. The inability of long-term treated patient biopsy samples to secrete autoantibodies to the supernatant could be explained by the overall absence of plasma cells and helper T cells as well as the failure of memory B cells to become activated within the timeframe of the organ culture [
17,
18]. Secondly, the secretion of EmA to the culture supernatant might simply be due to detachment of the autoantibodies from the tissue deposits and their release into the supernatant. However, we observed no decrease in the intensity of the mucosal autoantibody deposits after the organ culture. We did nevertheless see changes in ECH and the density of CD25+ cells, which speaks in favor of an active gliadin-induced process, this supporting the first proposition.
Authors' contributions
SS designed the study, carried out the experiments, collected data and drafted the manuscript
KL designed the study, supervised the work, participated in the writing of the manuscript and provided funding
IKS designed the study, analyzed the data and revised the manuscript
OL designed the study and revised the manuscript
PS performed the HLA-typing of patients and revised the manuscript
JP performed the HLA-typing of patients and revised the manuscript
KH performed the HLA-typing of patients and revised the manuscript
HW prepared the PT-gliadin used in the experiments and revised the manuscript
MM originated the idea for the research, provided funding and revised the manuscript
KK supervised the work, studied the patients and participated in the writing of the manuscript
All authors have read and approved the final manuscript.