Zinc transporter 8 (ZnT8) autoantibodies

The discovery of the latest major humoral autoantigen in T1D, ZnT8, was the result of a bioinformatics strategy to define new candidate autoantigens. ZnT8 is a member of a large conserved family of cation efflux proteins that plays a vital role in exporting Zn++ into the lumen of the beta cell secretory granule where it enables insulin crystallization. ZnT8 autoantibodies are detected in ~70% of newly diagnosed T1D patients. In prediabetic individuals, ZnT8 autoantibodies appear in the prodromal phase years prior to clinical disease and their measurement improves the accuracy of disease risk prediction. Autoreactivity to ZnT8 is unique with regard to a key amino acid residue at position 325 encoded by polymorphisms in the SLC30A8 gene, either arginine (R), tryptophan (W) or glutamine (Q). Prediabetic children harboring ZnT8 autoantibodies that are homozygous for either the R or W variant allele carry the greatest disease risk. Sensitive and specific ZnT8 autoantibody assays have been developed to capture antibodies targeting the ZnT8 major epitopes.

The discovery of ZnT8

 Schematic diagram of ZnT8 in the secretory granule membrane. ZnT8 functions as a dimer to form a pore between transmembrane domains IV and V. Allelic variants encoding amino acid 325 are indicated in the C-terminal domain (Arg, Trp, Glu).
Schematic diagram of ZnT8 in the secretory granule membrane. ZnT8 functions as a dimer to form a pore between transmembrane domains IV and V. Allelic variants encoding amino acid 325 are indicated in the C-terminal domain (Arg, Trp, Glu).
ZnT8 was initially identified in a bioinformatics screen for novel autoantigens in T1D using comparison of microarray expression profiles spanning multiple tissues, species and functional groups to assemble an array of candidate autoantigens. A gene index was calculated based on tissue specificity with consideration for abundance and distribution of transcripts between pancreas and islet tissues, beta, alpha and ductal cell lines, and endocrine vs exocrine expression. Further criteria imposed were that new antigens adhere to common features of the existing primary T1D antigens: high levels of predominantly beta cell expression, localization to the secretory granule, affiliation with the regulated secretory pathway, alternative splicing and protein topology often featuring at least one transmembrane domain. The predictive capacity of the ranking algorithm was supported by all the known major (and minor) antigens scoring high in the screen. ZnT8 placed highly among them, had not previously been proposed as an autoantigen and was subsequently tested for immunoreactivity [1].

ZnT8 structure and function

ZnT8 is a member of a conserved subfamily of nine genes that function in cation diffusion efflux which are included in the solute carrier 30 (SLC30A) super-family of genes [2]. The ZnT8 gene, SLC30A8, resides on chromosome 8 and encodes 369 amino acids. The topography of Zn++ transporters typically includes 6 transmembrane domains (which form the membrane pore) flanked by N- and C-terminal domains facing the cytosol and a “histidine loop” which binds Zn++ [3].

The Zn++ transporters often display tissue-specific expression with distinct intracellular localization [4]. ZnT8 is predominantly confined to the pancreatic islet beta cell with more modest expression in the alpha and ductal cells [5][6] and is thus more tissue-specific than IA-2, phogrin and GAD65. This feature led to the proposition that it may be a potential biomarker for residual beta cell mass, as insulin therapy promotes insulin autoantibody prevalence and titer, however there is no apparent correlation between residual C-peptide levels and ZnT8 antibodies [7]. ZnT8 spans the secretory granule membrane where it catalyzes the import of Zn++ ions from the cytosol into the lumen in exchange for protons translocated by the vesicular proton pump. Lumenal zinc concentrations approach 20 mM (the highest in the body) and facilitate insulin processing/maturation, crystallization and hexamer formation and eventual secretion [8][9][10].

ZnT8 autoantibody detection

Autoreactivity to ZnT8 was demonstrated by cloning the cDNA in a coupled in vitro transcription/translation vector to generate 35S-labeled ZnT8 antigen for immunoprecipitation with T1D sera. Early estimates showed that between 60-80% of sera derived from newly diagnosed T1D patients (depending on age) contained ZnT8 autoantibodies and thus their prevalence overlapped that of insulin, GAD65 and IA-2 autoantibodies. Notably ~26% of sera from newly diagnosed T1D patients negative for the other islet autoantibodies were positive for ZnT8 autoantibodies, so employment of the ZnT8 assay significantly reduced the number of autoantibody negative T1D individuals found at diagnosis (from 5.8% to 1.8%) [1][11]. Derivative ZnT8 N- and C-terminal domain, and concatenated “loop” antigen probes indicated the vast majority of immunoreactivity resided in the C-terminal cytoplasmic tail (aa275-369) with some immunoreactivity in the N-terminal domain [1][12]. The RIAs were optimized accordingly, avoiding the hydrophobic transmembrane domains in the optimized ZnT8 antigen probe. The ZnT8 autoantibody assay was submitted for external validation in the Diabetes Antibody Standardization Program (DASP) 2007 to confirm its reliability and utility.

ZnT8 autoantibody profiles in T1D

Measurement of ZnT8 autoantibodies across prospective longitudinal sera serum samples demonstrated ZnT8 antibodies add significant value for prediction of progression to overt disease in first-degree relatives of T1D individuals and among individuals with high-risk HLA genotypes (DR3 and/or DR4) [1][12][13].

In such studies, ZnT8 antibodies emerge many years prior to disease onset, but typically appear later than insulin or GAD65 autoantibodies [1][12][14][15][16]. The prevalence of ZnT8 antibodies peaks (80%) in late adolescence, at 12-16 years [1][17]. Among latent autoimmune diabetes in adults (LADA), ~42% of patients test positive for ZnT8 antibodies [18]. Prevalence of antibodies to ZnT8 (all variants, described below) rapidly decline post-disease onset to 6.7% within 20 years duration of disease with a profile weakly correlating to that of IA-2 [7][11][18]. Among the Gold Medalists (T1D duration more than 50 years) ZnT8 autoantibodies are virtually absent.

ZnT8 polymorphisms and epitopes

Immunoreactivity to ZnT8 is complex as there are three variants defined by the amino acid residue at position 325. A common nonsynonymous single nucleotide polymorphism (SNP), rs13266634, encodes an amino acid replacement (CGG)R>(TGG)W. Another SNP, rs16889462, encodes (CAG)Q at the same position although this allele is rare. Genome wide association studies demonstrated strong association of the former SNP with type 2 diabetes [19].

In T1D the predominant variant alleles encode two single determinants of immunoreactivity implying autoimmunity to ZnT8 targets self molecules. Individuals harboring ZnT8 autoantibodies display binding to either 325R or 325W or a complex ZnT8 profile of both and residue 325 alternate epitopes (below). Antibody responses restricted to ZnT8 325R and 325W variants strictly segregate with the alleles encoding them [17]. Homozygocity for either the 325R or 325W allele confers an increased rate of progression to T1D in ZnT8 autoantibody positive children compared to heterozygous children [12]. Homozygocity at the 325R allele has also been reported to be associated with new onset patients younger than 5 years [20].

ZnT8 (C-terminal domain) epitope mapping studies with near saturation mutagenesis characterized a third major conformational epitope including residues R332, E333, K336 and K340 [21]. The question of changes in ZnT8 epitope specificity post disease onset is unresolved and although large cohorts have been monitored for ZnT8 variant specific autoimmunity, the profiles are complex [22].

ZnT8 assay development

The existence of three ZnT8 variants led to the development of composite RIAs to capture antibodies to all ZnT8 “isotypes”. Concatenated probes including both major variants (325R and 325W) were externally evaluated in DASP 2009 by 26 labs internationally and achieved remarkable concordance. ZnT8 probes linking 2 copies of each monomer (325R, 325W, 325Q) or the dimer (325R + 325W) have been shown to have higher sensitivity without sacrificing specificity [7][23]. Alternate radio binding assays have likewise been developed using monomer probes; a “triple mix” of the three ZnT8 variants (325R, 325W, and 325Q) vs a plasmid encoding the major two (or three) variants sequentially [24]. Both RBA formats simplify ZnT8 antibody detection in lieu of individual assays when measurement of immunoreactivity targeting specific ZnT8 epitopes is not pertinent.

Non-radio active platforms for ZnT8 autoantibody detection have been fashioned, based on the RIA technology to promote accessibility of the assay to a wider range of laboratories. A commercially available ZnT8 antibody enzyme-linked immunosorbent assay (ELISA) has been developed (RSR) which is equally reliable to the RIAs (76% sensitivity, 97% specificity) and performed well in the Islet Autoantibody Standardization Program in 2012. The ZnT8 antibody ELISA has since been incorporated into a format for detection of ZnT8, IA-2 and GAD65 autoantibodies simultaneously. Luciferase (Gaussia) immunoprecipitation system (LIPS) assays have likewise been fabricated for detection of ZnT8 antibodies albeit with less specificity and sensitivity [25].

Concluding remarks

The discovery of ZnT8 as the fourth major autoantigen in T1D has enhanced the collective diagnostic sensitivity of islet autoantibodies such that up to ~95% of patients can be identified at disease onset. Autoantibodies to ZnT8 detected in the circulation of at-risk individuals are strongly predictive of diabetes development, especially in those harboring homozygous alleles for ZnT8 isotypes. As such, ZnT8 antibody measurements have been incorporated into selection criteria for clinical trials and offer a potential therapeutic target for diabetes prevention and restoration of immune tolerance in prediabetic subjects.


  1. ^ Wenzlau JM et al. The cation efflux transporter ZnT8 (Slc30A8) is a major autoantigen in human type 1 diabetes. 2007. Proc Natl Acad Sci U S A 23;104(43):17040-45.

  2. ^ Kambe T et al. Overview of mammalian zinc transporters. 2004. Cell Mol Life Sci;61(1):49-68.

  3. ^ Chimienti F et al. Identification and cloning of a beta-cell-specific zinc transporter, ZnT-8, localized into insulin secretory granules. 2004. Diabetes;53(9):2330-7.

  4. ^ Kambe T et al. The Physiological, Biochemical, and Molecular Roles of Zinc Transporters in Zinc Homeostasis and Metabolism. 2015. Physiol Rev;95(3):749-84.

  5. ^ Juhl K et al. Mouse Pancreatic Endocrine Cell Transcriptome Defined in the Embryonic Ngn3-Null Mouse. 2008. Diabetes;57(10), 2755-61

  6. ^ Murgia C et al. Diabetes-linked zinc transporter ZnT8 is a homodimeric protein expressed by distinct rodent endocrine cell types in the pancreas and other glands. 2008. Nutr Metab Cardiovasc Dis;19(6):431-9.

  7. ^ Wenzlau JM et al. Kinetics of the post-onset decline in zinc transporter 8 autoantibodies in type 1 diabetic human subjects. 2010. J Clin Endocrinol Metab;95(10):4712-19.

  8. ^ Formby B et al. Relationship between insulin release and 65zinc efflux from rat pancreatic islets maintained in tissue culture. 1984. Diabetes 33(3), 229–34.

  9. ^ Foster MC et al. Elemental composition of secretory granules in pancreatic islets of Langerhans.1993.

  10. ^ Li YV. Zinc and insulin in pancreatic beta-cells.2014. Endocrine;45(2):178-89.

  11. ^ Andersson C et al. Triple specificity of ZnT8 autoantibodies in relation to HLA and other islet autoantibodies in childhood and adolescent type 1 diabetes. 2013. Pediatr Diabetes;14(2):97-105.

  12. ^ Achenbach P et al. Autoantibodies to zinc transporter 8 and SLC30A8 genotype stratify type 1 diabetes risk. 2009. Diabetologia;52(9):1881-8.

  13. ^ De Grijse J et al. Predictive power of screening for antibodies against insulinoma-associated protein 2 beta (IA-2beta) and zinc transporter-8 to select first-degree relatives of type 1 diabetic patients with risk of rapid progression to clinical onset of the disease: implications for prevention trials. 2009. Diabetologia;53(3):517-24.

  14. ^ Yu L et al. Antiislet autoantibodies usually develop sequentially rather than simultaneously. 1996. J Clin Endocrinol Metab;81(12):4264-7.

  15. ^ Leslie RD, Delli Castelli M. Age-dependent influences on the origins of autoimmune diabetes: evidence and implications. 2004. Diabetes;53(12):3033-40.

  16. ^ Achenbach P et al. Natural history of type 1 diabetes. 2005. Diabetes;54 Suppl 2:S25-31.

  17. ^ Wenzlau JM et al. A common nonsynonymous single nucleotide polymorphism in the SLC30A8 gene determines ZnT8 autoantibody specificity in type 1 diabetes. 2008. Diabetes;57(10):2693-97.

  18. ^ Vaziri-Sani F et al. ZnT8 autoantibody titers in type 1 diabetes patients decline rapidly after clinical onset. 2010. Autoimmunity;43(8):598-606.

  19. ^ Sladek R et al. A genome-wide association study identifies novel risk loci for type 2 diabetes. 2007. Nature;445(7130):881-5.

  20. ^ Gohlke H et al. SLC30A8 (ZnT8) Polymorphism is Associated with Young Age at Type 1 Diabetes Onset. 2008. Rev Diabet Stud;5(1):25-7.

  21. ^ Wenzlau JM et al. Mapping of conformational autoantibody epitopes in ZnT8. 2011. Diabetes Metab Res Rev;27(8):883-6.

  22. ^ Wenzlau JM et al. Mapping of conformational autoantibody epitopes in ZNT8. 2011. Diabetes Metab Res Rev;27(8):883-6. Wenzlau JM, Frisch LM, Hutton JC, Fain PR, Davidson HW.

  23. ^ Lampasona V et al. Diabetes antibody standardization program: first proficiency evaluation of assays for autoantibodies to zinc transporter 8. 2011. Clin Chem;57(12):1693-702.

  24. ^ Vaziri-Sani F et al. A novel triple mix radiobinding assay for the three ZnT8 (ZnT8-RWQ) autoantibody variants in children with newly diagnosed diabetes. 2011 J Immunol Methods;371(1-2):25-37.

  25. ^ Ustinova J et al. Development of a luciferase-based system for the detection of ZnT8 autoantibodies. 2014. Journal of Immunological Methods;405, 67–73.


Nobody has commented on this article

Commenting is only available for registered Diapedia users. Please log in or register first.