Microbial Seromarkers in Coeliac Disease

Coeliac disease is an enteropathy caused by an immune response to ingested gluten in genetically predisposed individuals. It is characterized by the presence of histological damage in the small-bowel mucosa and serum antibodies targeted against transglutaminase 2 (TG2-abs) and endomysium (EmA). The disease develops gradually from an early stage involving only mucosal inflammation and/or coeliac disease-associated autoantibodies to crypt hyperplasia and villous atrophy. Currently the only scientifically accepted treatment is a lifelong gluten-free diet. The normal mucosal architecture is usually restored upon treatment, but a subgroup of patients continues to have persistent clinical symptoms and histological damage; i.e., nonresponsive coeliac disease (NRCD).

Coeliac disease is considered to have a multifactorial background. Nearly all patients carry the human leukocyte antigen (HLA) haplotypes DQ2 and/or DQ8, and over 40 loci outside the HLA region have also been associated with the disease. However, only a fraction of predisposed individuals develops the disease, indicating the presence of environmental modifiers. Intestinal microbiota has an essential role in the development and function of the immune system, and alterations in the microbial composition compared with non-coeliac controls have been associated with coeliac disease. Furthermore, increased serological responses towards commensal microbiota, including yeast Saccharomyces cerevisiae and Pseudomonas fluorescens and Bacteroides caccae bacteria have been reported in patients with active coeliac disease, as well as a decrease in the antibody levels during gluten-free diet. However, the time of the appearance of these microbial antibodies and whether they are associated with the clinical picture of coeliac disease remains unclear. There are also very limited data on the prevalence of microbial antibodies in the at-risk relatives of coeliac disease patients.

The aim of this dissertation was to assess seroreactivity to IgA and IgG class anti- Saccharomyces cerevisiae antibodies (ASCA) and IgA class antibodies against the I2 protein of Pseudomonas fluorescens (anti-I2 antibodies) and the TonB-linked outer membrane protein of Bacteroides caccae (anti-OmpW) in patients with different types and stages of coeliac disease, the first-degree relatives of coeliac disease patients and non-coeliac controls. Further, association of microbial antibody positivity with HLA and distribution of seropositivity among families were evaluated. The dissertation consists of three individual studies.

In Study I, the microbial antibodies were measured in 44 coeliac disease patients with early-stage disease and normal mucosal morphology, at the time when they developed diagnostic atrophy (n=16), and after one year on a gluten-free diet (n=33). It was found that five out of six patients seronegative to coeliac autoantibodies at an early stage already showed seropositivity to at least one of the microbial antibodies, and that the frequency of ASCA positivity decreased significantly on a gluten-free diet. Moreover, ASCA and anti-OmpW antibody levels were higher at the early stage than on a gluten-free diet, whereas the anti-I2 antibody levels were elevated at the time of coeliac disease diagnosis.

In Study II, the microbial antibodies of 20 NRCD patients on a gluten-free diet for a median of 3.5 years were compared with those of 58 newly diagnosed coeliac patients before the initiation of dietary treatment, 55 of whom also gave serum samples after one year showing a beneficial response to gluten-free diet, and with those of 80 non-coeliac blood donors. The results showed that seropositivity to ASCA was most frequent in the NRCD group, and that the ASCA IgA and IgG levels were significantly higher in NRCD patients than in the treated diet-responsive patients and controls. Neither seropositivity rate to anti-I2 or anti-OmpW antibodies nor antibody levels distinguished NRCD from diet-responsive coeliac disease, but the anti-I2 levels were higher in NRCD patients than in the controls.

In Study III, ASCA, anti-I2 and anti-OmpW antibodies were measured in 463 first-degree relatives of coeliac disease patients, 49 of whom were seropositive to TG2-ab and/or EmA. The results were compared with the 58 diet-responsive coeliac disease patients and the 80 controls also included in Study II. The microbial antibody levels were observed to be higher in the TG2-ab/EmA positive than the autoantibody negative relatives. Additionally, seropositivity was more common among the autoantibody negative relatives than among the controls but less common than among the coeliac disease patients. ASCA and anti-I2 antibody levels were also higher in all relatives than in the controls, whereas with anti-OmpW this was seen only in autoantibody positive relatives. Seropositivity to microbial antibodies was not significantly associated with HLA DQ haplotype.

This dissertation demonstrates that adaptive immune response to microbial antibodies may already be present at the early stage of coeliac disease before the development of small-bowel mucosal atrophy, and, in some patients, even before the emergence of disease-associated autoantibodies. The autoantibody negative first- degree relatives also had increased seroreactivity to microbial antibodies. Therefore, measuring these markers might help to identify those at the highest risk among genetically predisposed individuals, in some cases prior to the emergence of coeliac disease-associated autoantibodies. Combining different microbial antibodies increases the detection rate of at-risk individuals. Elevated ASCA levels are associated with NRCD, while the antibody levels decrease along with the mucosal healing. ASCA could thus serve as an additional noninvasive marker in the follow- up of coeliac disease. Whether the antibody response and the targeted microbes have a causal role in the coeliac disease pathogenesis remains a subject for further research.