Department of Microbiology, The University of Chicago, The Philippine Ministry of Health, Schistosomiasis Research and Control Project, Department of Pathology, The Pritzker School of Medicine, The University of Chicago, Chicago, Illinois 60637, Philippines
At present, there is no consensus that purified schistosome egg antigens offer any advantage in the diagnosis of schistosomiasis by enzyme linked immunosorbent assay (ELISA). Previously, we demonstrated by multiple techniques that the major serologic antigens in Schistosoma japonicum soluble egg antigen (SEA) are glycoproteins, and that the glycoproteins with highest specificity and sensitivity are hydrophobes. We therefore tested these materials for their specificity, sensitivity and cost effectiveness in the ELISA. In this study we used five SEA fractions that varied in their purity and antigenicity. The order of immunologic specific activity in the ELISA, measured by titration of a standard sera pool, was: hydrophobic glycoproteins (highest), crude SEA glycoproteins, hydrophilic glycoproteins, crude SEA, and SEA proteins (lowest). Complexity (purity) of these materials were (in rank order), hydrophilic glycoproteins (purest), hydrophobic glycoproteins, crude glycoproteins, SEA proteins, and crude SEA (most complex). Epidemiologic sensitivity in the ELISA was tested on limited but well characterized populations. At high antigen coating concentration (0.5 µg/well), the only antigen fraction with poor sensitivity was SEA proteins. There was little difference in epidemiologic sensitivity between the purer fractions with highest immunologic sensitivity (hydrophobic glycoproteins and crude SEA glycoproteins) and the crude SEA which possesses intermediate immunologic sensitivity. Differences in epidemiologic sensitivity were most pronounced when wells were coated at an antigen concentration (0.1 µg/well) where crude SEA began to fail. Specificity for all preparations, assessed by reactivity with sera from patients with other trematode infections and with cestode and nematode infections, was excellent. The clinical sensitivity of the ELISA employing crude S. japonicum SEA is so high, and the specificity so good, that the increased immunologic sensitivity of partially purified antigens had little effect on epidemiologic sensitivity. This is not true for the S. mansoni ELISA where crude antigens had inferior sensitivity and specificity.