Serum IgG levels to bovine insulin in type I diabetes mellitus.

ABSTRACT

OBJECTIVES: This study was undertaken to determine humoral immune response to bovine insulin in Iranian children with type 1 diabetes mellitus. METHODS: Serum samples were taken from 93 children aged 4-17 years with type I diabetes mellitus from two centers in Iran (the Iranian Association of Diabetes in Tehran and Center for Diabetes Research in Hamedan), 17 apparently healthy siblings of the diabetic patients (related controls), 28 apparently healthy age- and sex- matched controls (unrelated controls), 14 patients aged 11-15 years with auto-immune thyroiditis, and 45 patients with type II diabetes (aged 44-68 years). Samples were then examined for specific IgG to bovine insulin by enzyme-linked immunosorbent assay (ELISA). A questionnaire on medical history, duration of exclusive and non-exclusive breast feeding and daily intake of dairy products was completed before bleeding. RESULTS: Duration of exclusive and non-exclusive breast-feeding showed no significant difference between patients with type I diabetes, related and unrelated controls and thyroid patients. Diabetic children, however, had significantly higher serum levels of anti-bovine insulin IgG than did unrelated and related healthy controls and patients with type II diabetes (P< 0.01). There was no significant difference between healthy siblings of diabetic children and unrelated controls. In type I diabetic patients and their healthy siblings, serum levels of IgG to bovine insulin were inversely correlated with the duration of non-exclusive breast feeding (rs= -0.37, P= 0.016 and rs= -0.53, P= 0.049, respectively). There was no correlation between serum levels of IgG to bovine insulin with daily intake of dairy products. Bovine insulin cross-reacted with human insulin as judged by ELISA inhibition assay. CONCLUSION: The emergence of anti-insulin antibodies in Iranian patients with type I DM, which is associated with the duration of breast-feeding is less likely to be due to early exposure of infants with the proteins found in cow's milk. One speculation could be that the production of antibodies to insulin in type I diabetes may just be a physiologic response (probably to increase the half-life of the circulating insulin). The importance of anti-insulin antibodies in type I diabetes mellitus needs further studies.