Aspects of autoantibody epitopes in type 1 diabetes

Type 1 diabetes (T1D) is strongly associated with autoantibodies against insulin (IAA), glutamic acid decarboxylase 65 (GADA), insulinoma-associated protein 2 (IA-2A) and the most recently identified zinc T8 transporter (ZnT8A). Together or alone, they are important both to predict T1D and to classify diabetes at the time of clinical onset. The single nucleotide polymorphism (SNP) rs13266634 in the SLC30A8 gene encodes either an arginine (R) or a tryptophan (W) (R325W) at the amino acid (aa) position 325 in the ZnT8 protein. Autoantibodies that recognize ZnT8-arginine (ZnT8RA), ZnT8-tryptophan (ZnT8WA), or both, but restricted by the polymorphic site of the ZnT8 protein are common in newly diagnosed T1D patients. However, the epitope specificity and affinity of ZnT8A are poorly understood. The autoantigenicity of the ZnT8 polymorphism is unique and comparable protein sequence variations are not found for insulin, GAD65 or IA-2.
Neuropeptide Y (NPY) was reported as a minor autoantigen in T1D. This neurotransmitter has a SNP, rs16139, at aa position 7 in the NPY gene. The major allele is coding for leucine (L), and the minor for proline (P). The latter has been associated with impaired glucose tolerance and increased risk for type 2 diabetes (T2D). However, the possible autoantigenicity of the NPY polymorphism in T1D has not been investigated.
The overall aim of this thesis was to investigate the autoantibody epitopes of ZnT8 and NPY in newly diagnosed T1D patients as well as in patients with long-term T1D and T2D.
Our data in paper I and IV suggests that the polymorphic 325 variant is a major determinant of a conformation-dependent ZnT8A epitope. However, human sera with ZnT8-specific autoantibodies against either the R or the W variant did not recognize ZnT8 (318-331) peptides. It was therefore suggested that the conformational epitope recognized by the autoantibodies requires yet other amino acids beyond the 318-331 peptide of ZnT8. In addition, T1D patients with specific ZnT8WA displayed higher affinity compared to patients with specific ZnT8RA as demonstrated in reciprocal competitive displacement experiments. We suggest that future studies should include the ZnT8A variant specificity in both humoral and cellular tests to understand the possible role of autoantibody affinity to predict T1D.
In order to investigate the presence of NPY autoantibody (NPYA) variants (L or P at aa position 7), we developed radiobinding assays for both variants. We identified NPY-LA (23%) and NPY-PA (19%) in long-term T1D and 12% and 23% in T2D patients, respectively. The frequency of NPYA in newly diagnosed T1D patients at 1-18 years of age was 17% for NPY-LA and 24% for NPY-PA. In statistical regression analyses adjusted for gender, HLA and autoantibody status, NPYA were more common in children with older age at onset compared to children at younger age at onset. We suggest that NPYA may prove useful in the screening of young adults and in patients with long-term diabetes. The autoantibody response against the L7P polymorphic site was rarely specific to any of the two amino acid variants.
This thesis has revealed novel insights of the ZnT8A affinity to the 325-epitope and the possible importance of autoantibody affinity in T1D. Novel insights also include NPY as a minor autoantigen of significance to diabetes etiology and pathogenesis. Therefore, both epitope-specific ZnT8 and NPY autoantibodies are suggested to be included in future attempts to identify islet autoimmunity and to predict the clinical onset of autoimmune T1D.