Deletion of the mitochondrial flavoprotein apoptosis inducing factor (AIF) induces beta-cell apoptosis and impairs beta-cell mass

Abstract

Background: Apoptosis is a hallmark of beta-cell death in both type 1 and type 2 diabetes mellitus. Understanding how apoptosis contributes to beta-cell turnover may lead to strategies to prevent progression of diabetes. A key mediator of apoptosis, mitochondrial function, and cell survival is apoptosis inducing factor (AIF). In the present study, we investigated the role of AIF on beta-cell mass and survival using the Harlequin (Hq) mutant mice, which are hypomorphic for AIF.

Methodology/principal findings: Immunohistochemical evaluation of pancreata from Hq mutant mice displayed much smaller islets compared to wild-type mice (WT). Analysis of beta-cell mass in these mice revealed a greater than 4-fold reduction in beta-cell mass together with an 8-fold increase in beta-cell apoptosis. Analysis of cell cycle dynamics, using BrdU pulse as a marker for cells in S-phase, did not detect significant differences in the frequency of beta-cells in S-phase. In contrast, double staining for phosphorylated Histone H3 and insulin showed a 3-fold increase in beta-cells in the G2 phase in Hq mutant mice, but no differences in M-phase compared to WT mice. This suggests that the beta-cells from Hq mutant mice are arrested in the G2 phase and are unlikely to complete the cell cycle. beta-cells from Hq mutant mice display increased sensitivity to hydrogen peroxide-induced apoptosis, which was confirmed in human islets in which AIF was depleted by siRNA. AIF deficiency had no effect on glucose stimulated insulin secretion, but the impaired effect of hydrogen peroxide on beta-cell function was potentiated.

Conclusions/significance: Our results indicate that AIF is essential for maintaining beta-cell mass and for oxidative stress response. A decrease in the oxidative phosphorylation capacity may counteract the development of diabetes, despite its deleterious effects on beta-cell survival.

Figure 1. Decreased β-cell mass in Hq mutant mice.

(A). Representative Western blots (panel 1,3) and PCR analyzes (panel 2,4) of AIF expression in isolated mouse (panel 1,2) and human (panel 3,4) islets. Actin was used as loading control/ house keeping gene. Western blots/ PCRs are representatives of three independent experiments from three mice or from 3 organ donors, respectively. (B) Histological analysis by insulin staining in green and glucagon staining in red show a normal islet cellular composition and smaller islets in 2-week-old Hq mutant mice. (C,D) Analysis of β-cell mass (C) or β-cell mass divided by body weight (D) of WT and Hq mutant mice at 2 and 9 weeks of age. Values are representative of 5 slides spanning the whole pancreas of each mouse and 4 mice for each group at each age (magnification x125). (E) Cell cycle characteristics of β-cells from WT mice and Hq mutant mice as measured by BrdU and pHH3 staining. BrdU⁺insulin⁺ cells are counted as β-cells at S phase (see example in F, upper panel). pHH3⁺ (with punctuated pattern) insulin+ cells are counted as β-cells at G2 phase (see example in F, Hq mice lower right panel). pHH3⁺ (with strong nuclear expression) insulin+ cells were counted as β-cells at M phase (see example in F, WT mice lower left panel). Data are shown as mean±SE. *P<0.05 in Hq mutant mice vs. WT mice. For section 1 of Figure 1A, the following copyright limitations apply: Copyright © 2006, American Diabetes Association. Published by American Diabetes Association. All rights reserved.

Figure 2. AIF depletion in mice leads to increased β-cell apoptosis.

(A, B) Triple staining for TUNEL in red, insulin in green and 4′,6-diamidino-2-phenylindole (DAPI) in blue was performed on fixed, paraffin-embedded sections from isolated islets from 9-week old mice treated for 2 h with or without 50 µM H₂O₂. (B) Results are expressed as percentage of TUNEL-positive β-cells±SE. The mean number of β-cells counted was 700 for each treatment condition. (C) Glucose tolerance test with 2 mg/g BW glucose: Fasting and glucose stimulated plasma glucose levels are significantly lower in 12–16 week old Hq mutant mice (n = 9) compared to age-matched WT mice. Data are shown as mean + SE. *p<0.05 in Hq mutant mice vs. wt mice. **p<0.05 in H₂O₂ treated vs. untreated control.

Figure 3. AIF depletion leads to increased β-cell apoptosis in human islets without affecting insulin secretion.

(A) Isolated human pancreatic islets were exposed to siRNA to AIF (siAIF) or scrambled control siRNA (siScr) for 3 days. The knockdown efficiency was determined by Western blot analysis. Actin was used as a loading control on the same membrane after stripping. This Western blot is representative of three independent experiments from three different organ donors. The density of expression levels were quantified after scanning and normalised to actin levels. (B–D) 3 days after transfection, islets were exposed to 50 µM H₂O₂ for 2 h. (B) Islet sections were prepared for analysis of β-cell apoptosis by the TUNEL assay. Islets were double-stained for insulin in green and counterstained for DAPI in blue. Results are means±SE of the percentage of TUNEL-positive β-cells. The mean number of β-cells counted was 3400 for each treatment condition. (C,D) GSIS: after the H₂O₂ treatment, islets were washed and basal and stimulated insulin secretion analyzed during successive 1-h incubations at 2.8 mM (basal) and 16.7 mM (stimulated) glucose. Data are normalized to insulin content. (D) Stimulatory index denotes the ratio between stimulated and basal values of insulin secretion. (B–D) All assays were performed in triplicate or quadruplicate in three independent experiments from 3 different organ donors, respectively. *p<0.05 to siScr control, **p<0.05 in H₂O₂ treated vs. untreated control.