Gene Delivery of Manf to Beta-Cells of the Pancreatic Islets Protects NOD Mice from Type 1 Diabetes Development

Abstract

In type 1 diabetes, dysfunctional glucose regulation occurs due to the death of insulin-producing beta-cells in the pancreatic islets. Initiation of this process is caused by the inheritance of an adaptive immune system that is predisposed to responding to beta-cell antigens, most notably to insulin itself, coupled with unknown environmental insults priming the autoimmune reaction. While autoimmunity is a primary driver in beta-cell death, there is growing evidence that cellular stress participates in the loss of beta-cells. In the beta-cell fragility model, partial loss of islet mass requires compensatory upregulation of insulin production in the remaining islets, driving a cellular stress capable of triggering apoptosis in the remaining cells. The Glis3-Manf axis has been identified as being pivotal to the relative fragility or robustness of stressed islets, potentially operating in both type 1 and type 2 diabetes. Here, we have used an AAV-based gene delivery system to enhance the expression of the anti-apoptotic protein Manf in the beta-cells of NOD mice. Gene delivery substantially lowered the rate of diabetes development in treated mice. Manf-treated mice demonstrated minimal insulitis and superior preservation of insulin production. Our results demonstrating the therapeutic potential of Manf delivery to enhance beta-cell robustness and avert clinical diabetes.

Keywords: AAV; Manf; NOD mice; beta-cells; gene delivery; type 1 diabetes.

Figure 1

AAV8.ins-Manf prevented diabetes development in NOD mice. The percentage of non-diabetic NOD mice was controlled after the end of treatment at 30 weeks of age. (A) The AAV8.ins-Manf group (n = 17) and the AAV8.ins-GFP group (n = 14) had 82.4% and 42.8% non-diabetic mice, respectively (p-value of 0.0178). (B) The comparison between two groups with glucagon promoters, AAV8.glu-Manf (n = 14) and AAV8.glu-GFP (n = 14), had a p-value of 0.3045, which show no statistically significant difference. For comparison a Log-rank test was used.

Figure 2

Islets of AAV8.ins-Manf- or AAV8.glu-Manf-treated mice express Manf. (A–D) AAV8.ins-Manf-, AAV8.ins-GFP-, AAV8.glu-Manf-, and AAV8.glu-GFP-treated mice were stained with antibodies: Manf (green), insulin (red), or glucagon (red) and DAPI (blue). (E,F) mean fluorescence intensity (MFI) of Manf in islets of AAV8.ins-Manf-, AAV8.ins-GFP-, AAV8.glu-Manf-, and AAV8.glu-GFP-treated mice. Three to four slides from each mouse (in total 4 mice per group) were stained with antibodies: Manf, insulin, or glucagon and DAPI. Confocal images were captured and analysed using ImageJ software for determining MFI. Results are presented in means ± SEM (n = 4 per group). Unpaired t-tests were performed for comparison, ** denote p < 0.05.

Figure 3

Beta-cell-specific gene delivery of Manf prevented islets from severe insulitis. Light microscope pictures of islets representing no insulitis (A), insulitis grade 1 (C), insulitis grade 2 (E), insulitis grade 3 (G), and insulitis grade 4 (I), are demonstrated as well as graphs of each insulitis grading, comparing the mean percentages of islets in each group (B,D,F,H,J). (K) Total insulitis. Results are presented in means ± SEM (n = 14–17/group). Kruskal-Wallis test, followed by the Dunn’s test was performed for multiple comparisons for finding the significant differences. *, ** and *** denote p < 0.05, p < 0.01 and p < 0.001, respectively.

Figure 4

Beta-cell-specific gene delivery of Manf increased the content of insulin in islets of NOD mice. Light microscope pictures of NOD mice tissue sections immunohistochemically stained for insulin (A); a representative image of islets of AAV8.ins-Manf treated mice and (C); a representative image of islets of AAV8.ins-GFP treated mice) and graphs over the mean percentage of insulin-positive or insulin-negative islets in each group (B,D). (A) Two microscope pictures representing insulin-positive islets, where brown colour indicates insulin. (B) The percentages of insulin-positive islets in treated mice. (C) Two microscope pictures representing insulin-negative islets. (D) The percentages of insulin-negative islets in treated mice. (E) Percentages of insulin-positive and insulin-negative cells in total. Results are presented in means ± SEM (n = 14–17/group). Kruskal–Wallis test followed by Dunn’s test was performed for multiple comparisons. ** denote p < 0.01.

Figure 5

Beta-cell-specific gene delivery of Manf increased the serum levels of insulin in islets of NOD mice. Serum samples were analysed using Ultra-Sensitive Insulin ELISA kit. Results are presented in means ± SEM (n = 11–15/group). Kruskal–Wallis test followed by Dunn’s test was performed for multiple comparisons. * denotes p < 0.05.