Fig 1.
Identification of recombinant CTB-Ins-GAD and CTB-GAD-Ins expressed in silkworm.
(A) Schematic structure of the CTB-Ins-GAD fusion gene. AcMNPVPh-Pro: Autographa californica multiple nuclear polyhedrosis virus polyhedrin promoter, CTB: cholera toxin B subunit, L: linker peptide (GPGP), Ins: human insulin, GAD: triple copies of the glutamic acid decarboxylase 65 epitope (GAD65531-545), and LB and RB: left and right border, respectively, of the donor WT pFastBac1. The hemolymph from silkworm pupae was analyzed for the expression of the CTB-Ins-GAD fusion protein using anti-CTB (B) or anti-insulin (C) primary antibodies. The boiled and unboiled samples were diluted 50- and 10-fold, respectively. Quantitative analysis of protein production in BmN cells (D) and silkworm pupae (E). Data are presented as the mean concentrations ± standard deviations (SD) on each day. (F) Reactivity of fusion protein with the GM1 ganglioside and a native bacterial CTB control. (G) Pentamer and monomers analyzed. Approximately equal amounts of each sample were used to measure A492 signal levels. Data represent the mean A492 value ± SD of each sample.
Fig 2.
Suppression of insulitis and diabetes.
(A) Representative histopathological pancreatic islets from normal mice or experimental NOD mice, more than 25 pancreatic islets per mouse were scored, 1–4: normal mice (n = 6), CTB-GAD (n = 6), CTB-Ins-GAD (n = 6) and WT treated mice (n = 6), respectively. (B and C) Semiquantitative analysis of pancreatic islet insulitis scores. Six mice per group were individually tested in two separate experiments. * p<0.05; ** P<0.01. Data represent the mean values ± SD of each sample. (D) Suppression of diabetes. Five-week-old female NOD mice were fed CTB-Ins-GAD (n = 18), CTB-GAD-Ins (n = 17), CTB-GAD (n = 18), or WT (n = 18) alone three to four times per week until 35 weeks of age. Diabetes was confirmed by hyperglycemia (>200 mg/dL glucose) for 2 consecutive weeks. Data are combined from two independent experiments.
Fig 3.
Blood glucose levels in the NOD mice treated with (A) CTB-Ins-GAD (n = 10), (B) CTB-GAD-Ins (n = 10), (C) CTB-GAD (n = 10), or (D) WT (n = 10) alone were monitored twice a week until 50 days starting on the onset of diabetes with glucose levels between 200 mg/dl and 249 mg/dl. The mice that developed endpoint are shown as open circles (○) and others are shown as filled squares (■).
Fig 4.
Serum antibodies in NOD mice with different treatments.
Five-week-old female NOD mice were fed CTB-Ins-GAD (n = 6), CTB-GAD-Ins (n = 6), CTB-GAD (n = 6), or WT (n = 6) alone three to four times per week until 10 weeks of age. The serum antibodies were detected. (A) Anti-CTB, anti-insulin, anti-GAD65 IgG serum antibody titers; (B) anti-CTB IgG1, anti-insulin IgG1, and anti-insulin IgG2a; (C) anti-GAD65 IgG1, anti-GAD65 lgG2a; (D) anti-CTB IgA and IgE serum titers in mice fed CTB-Ins-GAD, CTB-GAD-Ins, CTB-GAD, or WT. Results are presented as the mean titer values ± SD. Six mice per group were individually tested in two separate experiments. *** P<0.001; NS: no significant difference.
Fig 5.
ELISPOT analysis of IL-4/IFN-γ-producing T cells and cytokine assay.
Splenocytes from the same 10-week-old NOD mice used to examine the development of insulitis were pooled into a given treatment group. In the ELISPOT assay, splenocytes (106/well) from CTB-Ins-GAD-, CTB-GAD-Ins-, CTB-GAD-, or WT-immunized mice were stimulated with insulin (A and B) or GAD65 peptide (C and D). T cells producing IL-4 (B and D) or IFN-γ (A and C) were detected. The number of spots was quantified in triplicate wells for each group. IFN-γ or IL-4 concentrations in serum (E and F) and pancreas (G and H) was measured via ELISA. Six mice per group were individually tested in two separate experiments. All data are expressed as the means ± SD. * p<0.05; ** P<0.01; *** P<0.001.
Fig 6.
CTB-Ins-GAD and CTB-GAD-Ins induce a greater increase in CD4+CD25+Foxp3+ T cells.
(A and B) Lymphocyte (1×106) isolated from CTB-GAD-, CTB-Ins-GAD-, CTB-GAD-Ins-, or WT-fed NOD mice at 10 weeks of age were treated with the mouse regulatory T cell staining kit. These treated lymphocyte were examined using a FACScan flow cytometer to determine CD4+CD25+Foxp3+Treg cell proportions. The gate was set based on CD3+CD4+. Results are presented as the mean values ± SD. Six mice per group were evaluated in two separate experiments. * p<0.05; ** P<0.01.
Fig 7.
CTB-Ins-GAD and CTB-GAD-Ins treatment induced regulatory T cell differentiation.
Proliferation analysis: splenocytes (1×107cells) isolated from CTB-GAD, CTB-Ins-GAD-, CTB-GAD-Ins, WT-fed, and normal mice at 10 weeks of age were cultured in 24-well plates with RPMI 1640 medium and stimulated with GAD65 (A), insulin (B), or PHA (C). Cells were stained immunohistochemically, and the stained cell proportions (representative of proliferative ability) were determined according to the protocol. (D) Migration analysis: splenocytes were cultured in the upper chamber of Transwell plates with 0.6 mL normal culture medium, 150 IU/mL IL-2, 20 μg/ml insulin and 20 μg/ml GAD65 placed in the bottom chamber at 37°C with 5% CO2 for 14 h. The migrated cells were fixed, stained, and numbered under the microscope. (E) CTB-Ins-GAD and CTB-GAD-Ins effects in adoptive transfer. Splenocytes (106/well) from CTB-Ins-GAD-, CTB-GAD-Ins-, CTB-GAD-, or WT-immunized mice were mixed with diabetogenic splenocytes (1×107) from diabetic NOD mice, and then co-transferred to 8-week-old NOD/SCID mice (n = 6 for all four groups) using a diabetogenic splenocyte-only injection as a control. The development of diabetes in the recipients was monitored for 18 weeks. Six mice per group were individually tested in two separate experiments. * p<0.05; ** P<0.01; NS: no significant difference.