Circulating anti-hypothalamus antibodies in celiac patients: tissue transglutaminase friend or foe?

Abstract

Celiac disease (CD) is an autoimmune disease with inflammatory characteristics, having a condition of chronic malabsorption, affecting approximately 1% of the population at any age. In recent years, a concrete correlation between eating disorders and CD has emerged. Hypothalamus plays a central role in determining eating behaviour, regulating appetite and, consequently, food intake. One hundred and ten sera from celiac patients (40 active and 70 following a gluten-free diet) were tested for the presence of autoantibodies against primate hypothalamic periventricular neurons by immunofluorescence and by a home-made ELISA assay. In addition, ghrelin was measured by ELISA. As control, 45 blood serums from healthy age matched were analysed. Among active CD, all patients resulted positive for anti-hypothalamus autoantibodies and sera showed significantly higher levels of ghrelin. All of the free-gluten CD were negative for anti-hypothalamus autoantibodies and had low levels of ghrelin, as well as healthy controls. Of interest, anti-hypothalamic autoantibodies directly correlate with anti-tTG amounts and with mucosal damage. In addition, competition assays with recombinant tTG showed a drastically reduction of anti-hypothalamic serum reactivity. Finally, ghrelin levels are increased in CD patients and correlated with anti-tTG autoantibodies and anti-hypothalamus autoantibodies. This study demonstrates for the first time the presence of anti-hypothalamus antibodies and their correlation with the severity of the CD. It also allows us to hypothesize the role of tTG as a putative autoantigen expressed by hypothalamic neurons.

Keywords: Active CD; Free-gluten CD; Anti-hypothalamus autoantibodies; Anti-tTG autoantibodies; Mucosal damage; Ghrelin.

Fig. 1

Evaluation of the presence of anti-hypothalamus IgG autoantibodies in CD patients and in healthy donors. A Anti-hypothalamus autoantibodies are present in sera from active CD patients, but they are not in gluten-free diet CD patients (p < 0.0001), neither in healthy donors (p < 0.0001). Anti-hypothalamic autoantibodies were detected by IIF method. A positive and a negative control serum were analysed in each analytical session. The intensity of the fluorescence was determined by two different readers blind to subject characteristic and a scale from 0 to 3 was assigned (0 no reaction, 1 weak or uncertain fluorescence, 2 moderate fluorescence, 3 high fluorescence). Samples scored positive if a 2 or 3 fluorescence reaction was observed. B A correlation of tTG-IgA autoantibodies vs. anti-hypothalamus IgG can be depicted. Untreated and treated CD patients are characterised by positive (> 9 ng/ml) and negative (< 9 ng/ml) anti-tTG autoantibody amounts respectively. A vertical line represents the cut-off value of 7 ng/ml for anti-tTG IgA autoantibodies (as suggested by the manufacturer’s protocols). Dashed lines indicate the 95% confidence interval of the best-fit line (continuous lines)

Fig. 2

tTG molecule is able to interfere with autoantibody binding. As tTG molecules are expressed by different brain tissues, and tTG is thought to play a role in oxidative damage, experiments were set up to investigate whether they may interfere with the binding of reactive autoantibodies to hypothalamic tissues. Fourty sera from active (untreated CD) and 40 from gluten-free diet (treated CD) sera were titred before and after tTG depletion. Numbers of sera patients for each dilution are shown

Fig. 3

Evaluation of the presence of ghrelin in sera of CD patients and healthy donors. A Serum amount of ghrelin was measured by a specific ELISA test. As can be depicted, a gluten-free diet (treated CD) significantly decreases the total serum ghrelin concentration measured during the active phase of CD (untreated CD). Ghrelin levels in treated CD patients and healthy donor do not significantly differ. B A correlation of tTG autoantibodies vs. ghrelin serum amounts can be depicted. CD population can be clustered in three groups: CD patients following a gluten-free diet are anti-tTG negative and express a lower amount of ghrelin. Among active CD, two groups can be highlighted: those that were characterised by an intermediate level of both anti-tTG autoantibodies and ghrelin and that showed high levels of both anti-tTG autoantibodies and ghrelin. A vertical line represents the cut-off value of 7 ng/ml for anti-tTG IgA autoantibodies (as suggested by the manufacturer’s protocols). Dashed lines indicate the 95% confidence interval of the best-fit line (continuous lines)

Fig. 4

A significant positive correlation among anti-hypothalamus autoantibodies vs. ghrelin exists. The quantitative evaluation of ghrelin and the qualitative presence of anti-hypothalamus antibodies detected in the total CD population, independently by diet show a positive correlation

Fig. 5

Anti-hypothalamus autoantibodies and ghrelin levels correlate with mucosal damage. A As expected, anti-tTG autoantibody values correlated mucosal damage (classified according to the modified Marsh). B A significant positive correlation was observed when anti-hypothalamus autoantibodies were compared to mucosal damage (modified Marsh classification). C Finally, it seems ghrelin production reflects the local mucosal damage