Schistosoma mansoni infection in preschool-aged children: development of immunoglobulin E and immunoglobulin G4 responses to parasite allergen-like proteins

Abstract

Specific immunoglobulin E (IgE) responses are upregulated during chronic schistosome infection and during allergy. These responses are tightly regulated during schistosomiasis. We have previously shown that IgE regulation depends on the extent and length of exposure to individual parasite allergen-like proteins. Here we compare the development of IgE and immunoglobulin G4 (IgG(4)) responses to the differentially expressed allergen-like proteins SmTAL1 and SmTAL2 among preschool-aged children from 2 villages with different levels of Schistosoma mansoni transmission. We found a lack of SmTAL1 responsiveness among all children, but evidence for IgG(4)-dependent IgE-SmTAL2 desensitization in both villages, occurring earlier among children from the village where the level of transmission was greater. Findings provide insights into the development and regulation of allergic-type immune responses.

Figure 1.

Predicted probability for TAL2 immunoglobulin E (A) and TAL2 immunoglobulin G₄ (B) responsiveness among infants in Bugoigo (dashed line) and Piida (solid line). A total of 180 children in Bugoigo (68.7%) donated sera, 98 (56.0%) of whom had detectable infection. A total of 121 children in Piida (73.8%) donated sera, 37 (30.6%) of whom had detectable infection. A, A statistically significant age-village interaction was observed (χ² [3 df] = 16.01; P = .001); age was modeled as a categorical variable because of departure from linearity for Piida estimates (P = .01). Model-predicted odds ratios (ORs) were as follows: male sex, 0.68 (95% confidence interval [CI], .36–1.29); village (Piida), 0.77 (95% CI, .26–2.32); age 2.1–3 years, 0.82 (95% CI, .27–2.55); age 3.1–4 years, 0.52 (95% CI, .14–1.97); age 4.1–6 years, 0.41 (95% CI, .11–1.54); age 2.1–3 years*Piida interaction term, 7.29 (95% CI, 1.33–40.05); 3.1–4 years*Piida interaction term, 25.36 (95% CI, 4.35–147.68); and age 4.1–6 years*Piida interaction term, 14.18 (95% CI, 2.43–82.76). B, No significant age-village interaction was observed (χ² [3 df] = 0.374; P = .541); age was modeled as a continuous variable because there was no departure from linearity (P = .274). Model-predicted ORs were as follows: male sex, 0.44 (95% CI, .20–.94); village (Piida), 0.01 (95% CI, .003–.02); and age, 2.03 (95% CI, 1.54–2.67).