Major DQ8-restricted T-cell epitopes for human GAD65 mapped using human CD4, DQA1* 0301, DQB1* 0302 transgenic IA (null) NOD mice.

J Liu, LE Purdy, S Rabinovitch, AM Jevnikar… - Diabetes, 1999 - Am Diabetes Assoc
J Liu, LE Purdy, S Rabinovitch, AM Jevnikar, JF Elliott
Diabetes, 1999Am Diabetes Assoc
The 65KD isoform of GAD is considered to be a major target autoantigen in many humans
with autoimmune prediabetes or diabetes. The major histocompatibility complex class II
allele DQA1* 0301, DQB1* 0302, which encodes HLA-DQ8, confers susceptibility to type 1
diabetes and occurs in up to 80% of affected individuals. To map T-cell epitopes for GAD65
restricted to the diabetes-associated DQ8 heterodimer, we generated transgenic NOD mice
expressing HLA-DQ8 and human CD4 while having the mouse class II gene (IA (beta)) …
The 65KD isoform of GAD is considered to be a major target autoantigen in many humans with autoimmune prediabetes or diabetes. The major histocompatibility complex class II allele DQA1*0301, DQB1*0302, which encodes HLA-DQ8, confers susceptibility to type 1 diabetes and occurs in up to 80% of affected individuals. To map T-cell epitopes for GAD65 restricted to the diabetes-associated DQ8 heterodimer, we generated transgenic NOD mice expressing HLA-DQ8 and human CD4 while having the mouse class II gene (IA(beta)) deleted. These mice were immunized with full-length purified recombinant GAD65, and the fine specificity of T-cell responses was mapped by examining recall responses of bulk splenocytes to an overlapping set of 20-mer peptides encompassing the entire GAD65 protein. Four different peptides (P121-140, P201-220, P231-250, and P471-490) gave significant T-cell recall responses. P201-220 and P231-250 have been shown previously to bind DQ8, whereas the other two peptides had been classified as nonbinders. Interestingly, the peptide giving the greatest response (P201-220) encompasses residues 206-220 of GAD65, a region that has been shown to be a dominant T-cell epitope in wild-type IA(g7) NOD mice. Overlap in this T-cell epitope likely reflects structural similarities between DQ8 and IA(g7). The fine specificity of antibody responses in the GAD65-immunized mice was also examined by testing the antisera by enzyme-linked immunosorbent assay (ELISA) against the same overlapping set of peptides. The two dominant B-cell epitopes were P361-380 and P381-400; P121-140 and P471-490 appeared to correspond to both B- and T-cell epitopes. Although the NOD human CD4, DQ8, IA(null) transgenic mice generated in these studies do not develop autoimmune diabetes either spontaneously or after cyclophosphamide treatment, they can be used to map DQ8-restricted T-cell epitopes for a variety of human islet autoantigens. They can also be used to test T-cell-specific reagents, such as fluorescently labeled DQ8 tetramers containing GAD65 peptides or other beta-cell peptides, which we believe will be useful in analyzing human immune responses in diabetic and prediabetic patients.
Am Diabetes Assoc