Determination of anti-omega-gliadin antibodies in serologic tests for coeliac disease

Background: Serologic detection of coeliac disease in the general populatio n or in subjects belonging to risk groups implies the use of a test with hi gh efficiency, large-scale use, and low cost. The enzyme-linked immunosorbe nt assay (ELISA) technique is the most appropriate assay for performing thi s kind of studies. Even though anti-gliadin determination has been the test most frequently used as the first step in screening procedures, many false -positive results produced a low-specificity test. In a previous work a sel ective recognition of omega-gliadins, mainly by IgA antibodies, was observe d. Results also indicated that omega-gliadins can be useful as antigens in serologic detection of coeliac disease. We therefore wanted to analyse the anti-gliadin antibody reactivity by using purified gliadins and to evaluate the actual performance of the anti-omega-gliadin antibody test. Methods: A population consisting of 105 coeliac patients, 81 healthy controls, and 73 subjects in a disease control group was analysed. Anti-endomysium (EMA), b oth IgG and IgA anti-omega-gliadins, and anti-tissue transglutaminase (tTG) antibodies were determined. Results: Concordant results, positive and nega tive, in the EMA and IgG and IgA anti-gliadin determinations were observed in 220 of 259 samples from the total population analysed. The three assays showed high efficiency, being 96.9%, 90.7%, and 91.1% for EMA and anti-omeg a-gliadins IgG and IgA, respectively. Anti-tTG determination was performed on 103 samples (69 controls and 34 coeliac patients), finding 4 false resul ts (2 false positive and 2 false negative), whereas anti-omega-gliadins sho wed 10 false results (5 false negative and 5 false positive), 3 of which we re coincident with anti-tTG determination. To compare the reactivity of ant i-gliadin antibodies, alpha-, beta-, gamma- and omega-gliadins were isolate d under non-denaturing conditions by acid preparative electrophoresis and c ation-exchange fast protein liquid chromatography (FPLC) and used in an ind irect ELISA test. The composition of these fractions was analysed by means of capillary electrophoresis, showing no cross-contamination among them. Co nclusions: The comparison of results using purified gliadins shows that ome ga-gliadins present a differential reactivity that has not previously been documented. Results using omega-gliadins isolated by either preparative ele ctrophoresis or FPLC were similar. Tests using the purified omega-gliadin f raction present the best performance when anti-gliadin antibodies are evalu ated.