Antibodies against Entamoeba histolytica in human milk and serum in Kenya.

A modified enzyme-linked immunosorbent assay (ELISA) technique has been developed for the detection of secretory immunoglobulin A (SIgA) antibodies against Entamoeba histolytica. By substituting phosphate-buffered saline-Triton for phosphate-buffered saline-Tween as the antibody incubating buffer, unspecific absorption of SIgA was avoided. As revealed by chessboard titrations, the optimal amount of antigen for coating was higher in SIgA than in IgG ELISA. A purified antigen from the membrane pellet fraction of E. histolytica gave equally good reactivity with SIgA and IgG antibodies and was used throughout. A total of 283 milk and 232 serum samples from three areas in Kenya were tested. The samples were collected in maternity hospitals on one of days 1 to 3 after parturition. All milk samples were tested for total SIgA. In one of the study areas (Machakos), the mean level of SIgA was significantly lower than in the other two areas (Mombasa and Nairobi). Eighty-seven of the milk samples (31%) were reactive in the test. The rate of positives was higher in Mombasa, where the SIgA levels were highest. Since both the frequency and the level of serum antibodies were similar in the three study areas, it is likely that the higher milk reactivity in Mombasa was mainly due to the higher SIgA concentrations observed. Antibodies were detected in 32 (14%) of the sera, mostly in low or moderate titers. Surprisingly few mothers had detectable antibodies in both milk and serum. In fact, the majority of positives were reactive in either milk or serum, with a predominance of milk positives. The background for this is probably complex, containing components such as differences in immunoglobulin concentrations in the samples, diversities in local and systemic antigenic stimulation responses, and level of immunological memory.