Normal incidence of diabetes in NOD mice tolerant to glutamic acid decarboxylase

Abstract

Experiments in nonobese diabetic (NOD) mice that lacked expression of glutamic acid decarboxylase (GAD) in beta cells have suggested that GAD represents an autoantigen essential for initiating and maintaining the diabetogenic immune response. Several attempts of inducing GAD-specific recessive tolerance to support this hypothesis have failed. Here we report on successful tolerance induction by expressing a modified form of GAD under control of the invariant chain promoter resulting in efficient epitope display. In spite of specific tolerance insulitis and diabetes occurred with normal kinetics indicating that GAD is not an essential autoantigen in the pathogenesis of diabetes.

Figure 1.

Characterization of GAD65 transgenic mice. (A) Diagram of the transgenic GAD65 construct. (B) Details of the transgenic construct being used. (a) K^b-leader (underlined) fusioned with amino acid 46 (***) of the GAD65 protein. (b) Amino acid 585 (***) of GAD65 fusioned with the transmembrane, cytoplasmic tail of the LAMP1 protein (underlined). The lysosomal targeting signal is in boldface. XXX indicate stop codons. (C) Expression of GAD65 mRNA in lymphoid compartments of 6-wk-old female NOD or GAD65-tg. mice by reverse transcriptase PCR using fivefold serial dilutions. β-actin cDNA was amplified as an internal control. (D) Epitope display of thymic dendritic cells. 11 different GAD65 reactive T cell hybridomas recognizing I-A^g7 restricted GAD65 epitopes were tested against thymic DCs from GAD65tg. (black bars) and NOD control mice (open bars). Stimulation indices were calculated as cpm sample/cpm control. Background values were between 300 cpm to 3,300 cpm/well depending on the hybridoma used. Stimulation indices for peptide pulsed DCs were between 10–70 for the hybridomas shown.

Figure 1.

Characterization of GAD65 transgenic mice. (A) Diagram of the transgenic GAD65 construct. (B) Details of the transgenic construct being used. (a) K^b-leader (underlined) fusioned with amino acid 46 (***) of the GAD65 protein. (b) Amino acid 585 (***) of GAD65 fusioned with the transmembrane, cytoplasmic tail of the LAMP1 protein (underlined). The lysosomal targeting signal is in boldface. XXX indicate stop codons. (C) Expression of GAD65 mRNA in lymphoid compartments of 6-wk-old female NOD or GAD65-tg. mice by reverse transcriptase PCR using fivefold serial dilutions. β-actin cDNA was amplified as an internal control. (D) Epitope display of thymic dendritic cells. 11 different GAD65 reactive T cell hybridomas recognizing I-A^g7 restricted GAD65 epitopes were tested against thymic DCs from GAD65tg. (black bars) and NOD control mice (open bars). Stimulation indices were calculated as cpm sample/cpm control. Background values were between 300 cpm to 3,300 cpm/well depending on the hybridoma used. Stimulation indices for peptide pulsed DCs were between 10–70 for the hybridomas shown.

Figure 1.

Characterization of GAD65 transgenic mice. (A) Diagram of the transgenic GAD65 construct. (B) Details of the transgenic construct being used. (a) K^b-leader (underlined) fusioned with amino acid 46 (***) of the GAD65 protein. (b) Amino acid 585 (***) of GAD65 fusioned with the transmembrane, cytoplasmic tail of the LAMP1 protein (underlined). The lysosomal targeting signal is in boldface. XXX indicate stop codons. (C) Expression of GAD65 mRNA in lymphoid compartments of 6-wk-old female NOD or GAD65-tg. mice by reverse transcriptase PCR using fivefold serial dilutions. β-actin cDNA was amplified as an internal control. (D) Epitope display of thymic dendritic cells. 11 different GAD65 reactive T cell hybridomas recognizing I-A^g7 restricted GAD65 epitopes were tested against thymic DCs from GAD65tg. (black bars) and NOD control mice (open bars). Stimulation indices were calculated as cpm sample/cpm control. Background values were between 300 cpm to 3,300 cpm/well depending on the hybridoma used. Stimulation indices for peptide pulsed DCs were between 10–70 for the hybridomas shown.

Figure 1.

Characterization of GAD65 transgenic mice. (A) Diagram of the transgenic GAD65 construct. (B) Details of the transgenic construct being used. (a) K^b-leader (underlined) fusioned with amino acid 46 (***) of the GAD65 protein. (b) Amino acid 585 (***) of GAD65 fusioned with the transmembrane, cytoplasmic tail of the LAMP1 protein (underlined). The lysosomal targeting signal is in boldface. XXX indicate stop codons. (C) Expression of GAD65 mRNA in lymphoid compartments of 6-wk-old female NOD or GAD65-tg. mice by reverse transcriptase PCR using fivefold serial dilutions. β-actin cDNA was amplified as an internal control. (D) Epitope display of thymic dendritic cells. 11 different GAD65 reactive T cell hybridomas recognizing I-A^g7 restricted GAD65 epitopes were tested against thymic DCs from GAD65tg. (black bars) and NOD control mice (open bars). Stimulation indices were calculated as cpm sample/cpm control. Background values were between 300 cpm to 3,300 cpm/well depending on the hybridoma used. Stimulation indices for peptide pulsed DCs were between 10–70 for the hybridomas shown.

Figure 2.

T cell responses of GAD65 transgenic animals against GAD65 and its peptide epitopes. (A) Proliferation assay of draining lymph node cells of 10-wk-old NOD mice 8 d after immunization with GAD65 in complete (black bars) and incomplete Freund's adjuvants (open bars) into foot pads. Representative result from five independent experiments. (B and C) Proliferative response of draining lymph node cells of NOD control mice (black bars) and two GAD65 transgenic lines (gray bars: line 14; open bars: line 29) to GAD65 and its peptide epitopes (all 10 μg/ml) 8 d after immunization with GAD65 in CFA. CF, culture filtrate protein of M. tuberculosis as positive control. (C) Titration of the positive proliferative recall responses as seen in B. (▪ NOD controls; ⋄ GAD65tg. line 14; Δ GAD65tg. line 29).

Figure 2.

T cell responses of GAD65 transgenic animals against GAD65 and its peptide epitopes. (A) Proliferation assay of draining lymph node cells of 10-wk-old NOD mice 8 d after immunization with GAD65 in complete (black bars) and incomplete Freund's adjuvants (open bars) into foot pads. Representative result from five independent experiments. (B and C) Proliferative response of draining lymph node cells of NOD control mice (black bars) and two GAD65 transgenic lines (gray bars: line 14; open bars: line 29) to GAD65 and its peptide epitopes (all 10 μg/ml) 8 d after immunization with GAD65 in CFA. CF, culture filtrate protein of M. tuberculosis as positive control. (C) Titration of the positive proliferative recall responses as seen in B. (▪ NOD controls; ⋄ GAD65tg. line 14; Δ GAD65tg. line 29).

Figure 2.

T cell responses of GAD65 transgenic animals against GAD65 and its peptide epitopes. (A) Proliferation assay of draining lymph node cells of 10-wk-old NOD mice 8 d after immunization with GAD65 in complete (black bars) and incomplete Freund's adjuvants (open bars) into foot pads. Representative result from five independent experiments. (B and C) Proliferative response of draining lymph node cells of NOD control mice (black bars) and two GAD65 transgenic lines (gray bars: line 14; open bars: line 29) to GAD65 and its peptide epitopes (all 10 μg/ml) 8 d after immunization with GAD65 in CFA. CF, culture filtrate protein of M. tuberculosis as positive control. (C) Titration of the positive proliferative recall responses as seen in B. (▪ NOD controls; ⋄ GAD65tg. line 14; Δ GAD65tg. line 29).

Figure 3.

T cell responses of GAD65 transgenic animals against GAD65 and its peptide epitopes. (A) Percent of CD4⁺ cells from lymph nodes (draining lymph nodes: open bars; mesenteric lymph nodes: gray bars) of NOD control and GAD65tg. mice 8 d after immunization with GAD65 in CFA as measured by FACS^® analysis. Percent values of background tetramer staining (animals just immunized with CFA [0.034–0.062%]) were subtracted from each sample (n = 4). (B) Representative FACS-plot of I-A^g7-p286 tetramer staining of lymph node cells from NOD control mice (B) and GAD65tg. mice (C) immunized with GAD65 in CFA (top two panels) or CFA alone (bottom two panels). (D) Number of IFN-γ–secreting cells from draining lymph nodes of NOD-control (open bars) and GAD65tg. mice (gray bars) after immunization with GAD65 as determined by ELISPOT analysis.

Figure 3.

T cell responses of GAD65 transgenic animals against GAD65 and its peptide epitopes. (A) Percent of CD4⁺ cells from lymph nodes (draining lymph nodes: open bars; mesenteric lymph nodes: gray bars) of NOD control and GAD65tg. mice 8 d after immunization with GAD65 in CFA as measured by FACS^® analysis. Percent values of background tetramer staining (animals just immunized with CFA [0.034–0.062%]) were subtracted from each sample (n = 4). (B) Representative FACS-plot of I-A^g7-p286 tetramer staining of lymph node cells from NOD control mice (B) and GAD65tg. mice (C) immunized with GAD65 in CFA (top two panels) or CFA alone (bottom two panels). (D) Number of IFN-γ–secreting cells from draining lymph nodes of NOD-control (open bars) and GAD65tg. mice (gray bars) after immunization with GAD65 as determined by ELISPOT analysis.

Figure 3.

T cell responses of GAD65 transgenic animals against GAD65 and its peptide epitopes. (A) Percent of CD4⁺ cells from lymph nodes (draining lymph nodes: open bars; mesenteric lymph nodes: gray bars) of NOD control and GAD65tg. mice 8 d after immunization with GAD65 in CFA as measured by FACS^® analysis. Percent values of background tetramer staining (animals just immunized with CFA [0.034–0.062%]) were subtracted from each sample (n = 4). (B) Representative FACS-plot of I-A^g7-p286 tetramer staining of lymph node cells from NOD control mice (B) and GAD65tg. mice (C) immunized with GAD65 in CFA (top two panels) or CFA alone (bottom two panels). (D) Number of IFN-γ–secreting cells from draining lymph nodes of NOD-control (open bars) and GAD65tg. mice (gray bars) after immunization with GAD65 as determined by ELISPOT analysis.

Figure 3.

T cell responses of GAD65 transgenic animals against GAD65 and its peptide epitopes. (A) Percent of CD4⁺ cells from lymph nodes (draining lymph nodes: open bars; mesenteric lymph nodes: gray bars) of NOD control and GAD65tg. mice 8 d after immunization with GAD65 in CFA as measured by FACS^® analysis. Percent values of background tetramer staining (animals just immunized with CFA [0.034–0.062%]) were subtracted from each sample (n = 4). (B) Representative FACS-plot of I-A^g7-p286 tetramer staining of lymph node cells from NOD control mice (B) and GAD65tg. mice (C) immunized with GAD65 in CFA (top two panels) or CFA alone (bottom two panels). (D) Number of IFN-γ–secreting cells from draining lymph nodes of NOD-control (open bars) and GAD65tg. mice (gray bars) after immunization with GAD65 as determined by ELISPOT analysis.

Figure 4.

Cells mediating GAD65-specific tolerance. (A) NOD mice were lethally irradiated and reconstituted with bone marrow from NOD (open bars) or GAD65tg. mice (gray bars). Proliferative recall responses after immunization with GAD65. CF, culture filtrate protein of M. tuberculosis as positive control. Representative result out of three independent experiments. (B) Thymectomized NOD mice were engrafted with E14 fetal thymi from NOD (open bars) or GAD65tg. mice (gray bars) and, after ablative irradiation, reconstituted with NOD bone marrow. Proliferative recall responses after immunization with GAD65.

Figure 4.

Cells mediating GAD65-specific tolerance. (A) NOD mice were lethally irradiated and reconstituted with bone marrow from NOD (open bars) or GAD65tg. mice (gray bars). Proliferative recall responses after immunization with GAD65. CF, culture filtrate protein of M. tuberculosis as positive control. Representative result out of three independent experiments. (B) Thymectomized NOD mice were engrafted with E14 fetal thymi from NOD (open bars) or GAD65tg. mice (gray bars) and, after ablative irradiation, reconstituted with NOD bone marrow. Proliferative recall responses after immunization with GAD65.

Figure 5.

Spontaneous anti-insulin immune response, insulitis, and IDDM in GAD65 transgenic mice. (A) Insulin-autoantibodies in 14-wk-old, female GAD transgenic, NOD mice as well as in controls (NOR, non obese diabetes resistant mice; C57/B6, Balb/c mice) measured by radiobinding assay. There was no statistical difference between 14w old NOD mice and GAD65-tg. animals. (B) Histological grading of insulitis in pancreas sections of 4 and 10 week old mice (n = 4 per group). (C) Cumulative incidence of development of IDDM in NOD mice (n = 30; black curve) and GAD65 transgenic mice (n = 34; gray curve). Data of two independent GAD65 founder lines were combined. There was no statistical difference in diabetes onset between transgenic and nontransgenic mice according to Kaplan-Meier Analysis.

Figure 5.

Spontaneous anti-insulin immune response, insulitis, and IDDM in GAD65 transgenic mice. (A) Insulin-autoantibodies in 14-wk-old, female GAD transgenic, NOD mice as well as in controls (NOR, non obese diabetes resistant mice; C57/B6, Balb/c mice) measured by radiobinding assay. There was no statistical difference between 14w old NOD mice and GAD65-tg. animals. (B) Histological grading of insulitis in pancreas sections of 4 and 10 week old mice (n = 4 per group). (C) Cumulative incidence of development of IDDM in NOD mice (n = 30; black curve) and GAD65 transgenic mice (n = 34; gray curve). Data of two independent GAD65 founder lines were combined. There was no statistical difference in diabetes onset between transgenic and nontransgenic mice according to Kaplan-Meier Analysis.

Figure 5.

Spontaneous anti-insulin immune response, insulitis, and IDDM in GAD65 transgenic mice. (A) Insulin-autoantibodies in 14-wk-old, female GAD transgenic, NOD mice as well as in controls (NOR, non obese diabetes resistant mice; C57/B6, Balb/c mice) measured by radiobinding assay. There was no statistical difference between 14w old NOD mice and GAD65-tg. animals. (B) Histological grading of insulitis in pancreas sections of 4 and 10 week old mice (n = 4 per group). (C) Cumulative incidence of development of IDDM in NOD mice (n = 30; black curve) and GAD65 transgenic mice (n = 34; gray curve). Data of two independent GAD65 founder lines were combined. There was no statistical difference in diabetes onset between transgenic and nontransgenic mice according to Kaplan-Meier Analysis.