Antigen-specific immunotherapy combined with a regenerative drug in the treatment of experimental type 1 diabetes

Abstract

Type 1 diabetes is an autoimmune disease caused by the destruction of the insulin-producing β-cells. To revert type 1 diabetes, the suppression of the autoimmune attack should be combined with a β-cell replacement strategy. It has been previously demonstrated that liraglutide, a glucagon-like peptide-1 receptor agonist, restores β-cell mass in type 1 diabetes, via α-cell transdifferentiation and neogenesis. We report here that treatment with liraglutide does not prevent type 1 diabetes in the spontaneous non-obese diabetic (NOD) mouse model, but it tends to reduce leukocytic islet infiltration. However, in combination with an immunotherapy based on tolerogenic liposomes, it is effective in ameliorating hyperglycaemia in diabetic NOD mice. Importantly, liraglutide is not detrimental for the tolerogenic effect that liposomes exert on dendritic cells from patients with type 1 diabetes in terms of membrane expression of molecules involved in antigen presentation, immunoregulation and activation. Moreover, the in vivo effect of the combined therapy was tested in mice humanised with peripheral blood mononuclear cells from patients with type 1 diabetes, showing no adverse effects in leukocyte subsets. In conclusion, the combination therapy with liraglutide and a liposome-based immunotherapy is a promising candidate strategy for type 1 diabetes.

Figure 1

Effect of liraglutide in pre-diabetic NOD mice. (A) Incidence of diabetes (%) in NOD mice (female) treated with liraglutide (Lira, squares, n = 7) or PBS (Sham, triangles, n = 12). No significant differences in T1D incidence were found between groups (Mantel-Cox Log-Rank). Significant differences in the age of the onset were found between Lira and Sham groups (*p < 0.05, Mann–Whitney test). (B) Percentage of islets in each of the infiltration categories, in sham and liraglutide treated groups (n = 3/group): White = 0, no insulitis; Dotted = 1, peri-insular; Striped = 2, mild insulitis (< 25% of the infiltrated islet); Squared = 3, 25–75% of the islet infiltrated; Black = 4, > 75% islet infiltration. Significant differences were found between groups (*p < 0.05, Chi-square test). (C) Representative images of insulitis in islets from H/E-pancreatic cryostat sections (5 μm) from the two groups. Scale bars represent 100 μm.

Figure 2

Effect of liraglutide in diabetic NOD mice. (A) 2 h fasting blood glucose levels (mg/dL) in diabetic NOD mice treated with either PBS (Sham, red, n = 9), PSAB-liposomes (PSAB, grey, n = 6), liraglutide (Lira, green, n = 6) or combined (PSAB + Lira, blue, n = 6). Filled area corresponds to normal blood glucose levels in NOD mice. (B) Area Under the Curve (AUC) of the graph in (A) at day 10, when all mice of Sham, PSAB, and combined groups remain alive. Only responder mice are represented in the combined group. Results are expressed as mean ± SD. Differences were found between Sham and PSAB (*p < 0.05, Mann–Whitney test) and between Sham and combined therapy (**p < 0.01, Mann–Whitney test). (C) Stratification of responders (black squares) and non-responders (white circles) in the group treated with combined therapy regarding the age (weeks) and blood glucose levels (mg/dL) at the onset.

Figure 3

Combined therapy regulates dendritic cell (DCs) phenotype. Relative CD36, TIM4, CD49d, HLA-ABC, HLA-DR, CD54, CD40, CD86, CD25, CCR7, PDL-1, CXCR4 and TLR2 membrane expression in DCs (n = 7) obtained from adult subjects with T1D. Results are mean ± SD. Significant differences were found when comparing culture conditions (*p ≤ 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, Mann–Whitney test).

Figure 4

Gene expression analysis of human DCs co-cultured with PSAB-liposomes and combined therapy. (A) List of differentially expressed genes in both PSAB and combined groups in comparison to control conditions, showing log₂ of Fold Change and adjusted p-value. (B) Relative gene expression of 3 selected genes from A) (FOS, ABCA, CD1D) analysed by qRT-PCR. Gene expression was normalized to GAPDH. Bars show the mean ± S.D. of the Log2 of FC using basal transcription as standard value (Wilcoxon test). Validation of the differential expression of three selected genes from A) (FOS, ABCA, CD1D) by quantitative RT-PCR (n = 3). Results are mean ± SD.

Figure 5

In vivo effect of the combined therapy in PBMC-humanised NSG mice. Total counts (left) and percentage (right) of human cells at the starting point (in patients), in mice from week 1 (w1) to week 4 (w4) post-transplantation and at the endpoint (day 30 of treatment, final). (A) Data of human CD45⁺ cells. (B) Data of human CD3⁺CD4⁺ cell subset. (C) Data of human CD3⁺CD8⁺ cell subset. (D) Data of human memory T regulatory cells (mTreg) subset, identified as CD3⁺CD4⁺CD127^lowCD25⁺CCR4⁺CD45RO⁺. Data refers either to patients (n = 3), combined (n = 7) or control (n = 3) group.