Aims To develop methods for isolation of enterovirus strains from subjects with preclinical Type 1 diabetes and evaluate if their presence in stools is associated with beta-cell damage.
Methods The study subjects were participants of the Finnish Type 1 Diabetes Prediction and Prevention Study (DIPP). The prospectively followed birth cohort comprised 12 children who turned positive for diabetes-associated autoantibodies during the follow-up (case children) and 53 controls matched for date of birth, sex and HLA-DQB1 alleles. Altogether, 878 stool samples were analysed for the presence of enterovirus RNA by RT-PCR followed by virus isolation and partial sequencing of viral genome. Enterovirus antibodies and RNA were simultaneously analysed from serum.
Results Eleven enterovirus infections were diagnosed in case children and 42 infections in control children by the presence of viral RNA in stools. The proportion of children who were repeatedly enterovirus RNA-positive stools was higher among case than control children (42% vs. 11% of children; P = 0.02). Combined serum (antibody and RT-PCR) and stool analyses indicated at least one enterovirus infection in 83% of the case children before the appearance of autoantibodies, while only 42% of the control children had infection by the same age (P = 0.006). Twelve enterovirus strains were isolated from case children and 38 strains from control children.
Conclusions This protocol makes it possible to isolate a large number of enterovirus strains from prediabetic subjects. The findings suggest that enterovirus infections may be associated with the beta-cell damaging process.
Diabet. Med. 21, 156–164 (2004)