The cation efflux transporter ZnT8 (Slc30A8) is a major autoantigen in human type 1 diabetes

Abstract

Type 1 diabetes (T1D) results from progressive loss of pancreatic islet mass through autoimmunity targeted at a diverse, yet limited, series of molecules that are expressed in the pancreatic beta cell. Identification of these molecular targets provides insight into the pathogenic process, diagnostic assays, and potential therapeutic agents. Autoantigen candidates were identified from microarray expression profiling of human and rodent pancreas and islet cells and screened with radioimmunoprecipitation assays using new-onset T1D and prediabetic sera. A high-ranking candidate, the zinc transporter ZnT8 (Slc30A8), was targeted by autoantibodies in 60-80% of new-onset T1D compared with <2% of controls and <3% type 2 diabetic and in up to 30% of patients with other autoimmune disorders with a T1D association. ZnT8 antibodies (ZnTA) were found in 26% of T1D subjects classified as autoantibody-negative on the basis of existing markers [glutamate decarboxylase (GADA), protein tyrosine phosphatase IA2 (IA2A), antibodies to insulin (IAA), and islet cytoplasmic autoantibodies (ICA)]. Individuals followed from birth to T1D showed ZnT8A as early as 2 years of age and increasing levels and prevalence persisting to disease onset. ZnT8A generally emerged later than GADA and IAA in prediabetes, although not in a strict order. The combined measurement of ZnT8A, GADA, IA2A, and IAA raised autoimmunity detection rates to 98% at disease onset, a level that approaches that needed to detect prediabetes in a general pediatric population. The combination of bioinformatics and molecular engineering used here will potentially generate other diabetes autoimmunity markers and is also broadly applicable to other autoimmune disorders.

Fig. 1.

Identification of candidate autoantigens. Pancreas specificity is plotted against the pancreatic abundance (pancreatic EST clones per all tissues ESTs) and against the level of mRNA expression in human islets from microarray data (SI Table 1). Approximate positions are shown for diabetes autoantigens (filled symbols).

Fig. 2.

ZnT8 autoantibody assays in new-onset T1D subjects. (A) The levels of autoreactivity to the ORF (aa1–369), N-terminal (amino acids 1–74), C-terminal (amino acids 264–369) constructs, and a single-chain fusion of the N and C termini (B). (C) The overlap in autoantibody prevalence to each construct.

Fig. 3.

Uniquiness of ZnTA. ZnT8 C-terminal autoreactivity was measured in GADA-, IA2A-, IAA-, and ICA-negative new-onset T1D subjects, nondiabetic subjects who were 21-hydroxylase antibody-positive with or without Addison's disease and transglutaminase antibody-positive relatives of T1D patients with celiac disease.

Fig. 4.

Overlapping prevalence of ZnT8A, GADA, IA2A, and IAA at onset. (A) Seropositive individuals evaluated with three-autoantibody standard or with ZnT8A substituted for GADA, IA2A, or IAA. The ZnT8A assay incorporates both C-terminal and N/C assays in the one measurement. (B) Seropositive individuals evaluated with four-autoantibody standard.

Fig. 5.

Relationship of ZnT8A to age at diabetes onset. New-onset T1D subjects ranging from <1 yr to 18 yr old were grouped in 2-yr intervals to illustrate the prevalence of antibodies to ZnT C-terminal and N/C constructs in relation one another and to GADA, IA2A, and IAA.

Fig. 6.

ZnT8A in prediabetes. (A–C) The cumulative incidence of autoantibodies in 27 subjects followed to T1D is shown for ZnT C-terminal autoreactivity relative to GADA, IA2A, and IAA. (D) ZnT8 C-terminal autoantibodies were determined on the first serum samples to test positive for GADA, IA2, or IAA. Survival analysis compares the disease outcome (mean ± SEM) for ZnT8A-seropositive versus -negative individuals. Subjects remaining in both studies at each time point are indicated.