The Genetic Architecture of Type 1 Diabetes

Abstract

Type 1 diabetes (T1D) is classically characterised by the clinical need for insulin, the presence of disease-associated serum autoantibodies, and an onset in childhood. The disease, as with other autoimmune diseases, is due to the interaction of genetic and non-genetic effects, which induce a destructive process damaging insulin-secreting cells. In this review, we focus on the nature of this interaction, and how our understanding of that gene-environment interaction has changed our understanding of the nature of the disease. We discuss the early onset of the disease, the development of distinct immunogenotypes, and the declining heritability with increasing age at diagnosis. Whilst Human Leukocyte Antigens (HLA) have a major role in causing T1D, we note that some of these HLA genes have a protective role, especially in children, whilst other non-HLA genes are also important. In adult-onset T1D, the disease is often not insulin-dependent at diagnosis, and has a dissimilar immunogenotype with reduced genetic predisposition. Finally, we discuss the putative nature of the non-genetic factors and how they might interact with genetic susceptibility, including preliminary studies of the epigenome associated with T1D.

Keywords: Type 1; diabetes; genes; genetic.

Figure 1

Evolution of pathogenicity over time. In patients with genetic disposition to disease, the clinical presentation of Type 1 diabetes (T1D) is predated by environmental exposure leading to changes in the immune system and development of autoantibodies. Note that the destruction of tissue—insulin producing cells in the case of T1D—predates the presentation of the illness, and continues after presentation also. Genetic impact can operate at different stages throughout the disease process.

Figure 2

Potential therapeutic options in the prevention and treatment of T1D. Currently, the only options available are the treatment of symptoms and beta cell transplantation; however, in the future, genetic stratification may allow secondary preventative measures to be employed.

Figure 3

Genetic risk score (GRS) between T1D, Type 2 diabetes (T2D), latent autoimmune diabetes in adulthood (LADA), and LADA-restricted cases and controls. The GRS distributions were compared between subjects with T1D (n = 1990), T2D (n = 1960), LADA (n = 978), LADA restricted (n = 309), LADA GADA-only (n = 669), and Bone Mineral Density in Childhood Study (BMDCS) controls (n = 1057). The Violin plots show the distribution of GRS for the five groups: (a) using the T1D single nucleotide polymorphisms (SNPs) for the five groups and (b) using the T2D SNPs. A multiple comparison test (Wilcoxon rank sum test) was performed to calculate pair-wise differences, and key differences are highlighted. (*** p < 0.00001, ** p < 0.0001, * p < 0.05) [54].

Figure 4

We used (MZ twins discordant for T1D in three different cell types, to define either differentially methylated CpG positions (DMPs), or differentially variable CpG positions (DVPs), as illustrated [80].