Aims/hypothesis: The aim of this study was to assess the prevalence of glucokinase gene mutations in Italian children with MODY and to investigate genotype/phenotype correlations of the mutants.
Methods: Screening for sequence variants in the glucokinase gene was performed by denaturing gradient gel electrophoresis and direct sequencing in 132 children with maturity onset diabetes of the young (MODY) and in 9 children with chronic fasting hyperglycaemia but without laboratory evidence for Type I (insulin-dependent) diabetes mellitus and with normoglycaemic parents ("non-classical" MODY).
Results: Altogether 54 mutations were identified in the MODY group (54/132 or 41%) and 3 among the "non-classical" MODY individuals (3/9 or 33%). Paternity testing indicated that the latter mutations have arisen de novo. Mean fasting plasma glucose concentrations of the children with the mutant glucokinase was in the expected impaired fasting glucose range. In contrast, results of the oral glucose tolerance test showed a wide range from normal glucose tolerance (Group 1: 2-h OGTT = 6.7 +/- 1.1 mmol/l; 11 patients) to diabetes (Group 2: 2-h OGTT = 11.5 +/- 0.5 mmol/l; 9 patients), with the remaining in the impaired glucose tolerance range. Disruptive mutations (i.e. nonsense, frameshifts, splice-site) were equally represented in Groups 1 and 2 and were not clearly associated with an impaired first-phase insulin response. Surprisingly, 5 out of 11 children (or 45%) in Group 1 were found to be overweight but no children in Group 2 were overweight. Sensitivity index (SI), calculated by a recently described method, was found to be significantly lower in Group 2 than in Group 1 (SI Group 2 = 0.0013 +/- 0.0009 ml Kg(-1) min(-1)/muU/ml; SI Group 1 = 0.0068 +/- 0.0048, p < 0.0035).
Conclusion/interpretation: Mutations in glucokinase are the first cause of MODY among Italian children selected through a low threshold limit of fasting plasma glucose (i. e. > 5.5 mmol). The lack of correlation between the molecular severity of glucokinase mutations, insulin secretion at intravenous glucose tolerance test and differences in glucose tolerance suggests that factors outside the beta cell are also involved in determining post-load glucose concentrations in these subjects. Our results seem to indicate that the differences observed in the 2-h responses at the OGTT among children with MODY 2 could be related to individual differences in insulin sensitivity.