A two locus model for juvenile diabetes

Summary As stated before, the aim of this analysis was t o determine if a two locus plus environment disease model for IDDX mould give predicted values of various population parameters in agreelllent with observed values, or whether the model could be ruled out by obvious discrepancy between sonie predicted and observed values. It is assumed that the HLA linked ‘;disease’ allele D, follows a recessive mode of inheritance, with an allele frequency of approximately 357;. The results show that the addition of a second ‘disease’ locus gives predicted values that, are certainly not in disagreement with observed values. This in no way proves that this is the mechanism operating with IDDN but certainly makes it a feasible possibility. The discrepancies with observed population parameters of the one locus plus environment dominant and recessive models, plus the discrepancy between concordance rate in monozygous twins and incidence in sibs of probands, plus the observations of association of the Lewis blood group with IDDX, provide evidence in favour of the present two locus model. At the moment no real distinction can be made as t o whether it is more likely that the hypothesized second ‘disease’ locus would follow a dominant or recessive mode of inheritance. A further point of interest from the study by Vague et al. (1978) is the demonstration that not only IDDSI but also maturity onset diabetes shows an association with the Lewis negative blood group. This opens up exciting possibilities concerning the relationship of the two diseases. It may be that susceptibility to maturity onset diabetes is determined either solely by one locus (D2) or by D, plus other loci, plus environmental effects, while susceptibility to IDDM is determined by two (or possibly more) loci (D1 and D,) plus environmental effects. The observation by MacDonaltl(1974) of an equal incidence of maturity‐onset diabetes among the ancestors of juvenile‐diabetics and non‐diabetics would argue against such an hypothesis. However, further investigation of this question is called for in view of Vague et al (1978) observation. Obserx‐ations from the mouse are also supportive of the involvement of more than one locus in determining the degree of severity of diabetes. Coleman (1978) states that in the mouse, ‘The severity and clinical course of the diabetes produced depends on the interaction of the mutant, gene with the inbred background rather than the action of the gene itself. Thus on one genetic background a well‐compensated, maturity onset type diabetes, compatible with nearnormal life is observed whereas on another inbred background the syndrome presents as a juvenile‐type diabetes with insulinopenia, islet cell degeneration, marked hyperglycaemia, sonie ketosis and a much shoxtened lifespan.’ I n these studies the diabetic ‘disease’ locus being studied follows a recessive mode of inheritance and ‘the modifiers leading to islet hypertrophy and well‐compensated diabetes compatible with a near normal lifespan are dominant to those factors causing severe diabetes’ (Coleman, 1978).