Kinetics of functional beta cell mass decay in a diphtheria toxin receptor mouse model of diabetes

Abstract

Functional beta cell mass is an essential biomarker for the diagnosis and staging of diabetes. It has however proven technically challenging to study this parameter during diabetes progression. Here we have detailed the kinetics of the rapid decline in functional beta cell mass in the RIP-DTR mouse, a model of hyperglycemia resulting from diphtheria toxin induced beta cell ablation. A novel combination of imaging modalities was employed to study the pattern of beta cell destruction. Optical projection tomography of the pancreas and longitudinal in vivo confocal microscopy of islets transplanted into the anterior chamber of the eye allowed to investigate kinetics and tomographic location of beta cell mass decay in individual islets as well as at the entire islet population level. The correlation between beta cell mass and function was determined by complementary in vivo and ex vivo characterizations, demonstrating that beta cell function and glucose tolerance were impaired within the first two days following treatment when more than 50% of beta cell mass was remaining. Our results illustrate the importance of acquiring quantitative functional and morphological parameters to assess the functional status of the endocrine pancreas.

Figure 1

Optical projection tomography shows pancreatic beta cell loss in the RIP-DTR mouse. (a,b) Isosurface rendered OPT images of representative splenic lobes of the RIP-DTR mouse at 15 days after sham (a) or diphtheria toxin (b) treatment. Three-dimensional surface renderings are based on anti-insulin staining (red) and intrinsic autofluorescence portraying the anatomy (grey). (c) Graph of the average total pancreatic beta cell volume illustrating severe beta cell loss after diphtheria toxin treatment. (d) Graph showing that beta cell loss in the RIP-DTR pancreas is seen in all three lobes. (e,f) OPT images displaying pseudo-colours yellow (>5 × 10⁶ μm³), red (1–5 × 10⁶ μm³), and white (<1 × 10⁶ μm³) to indicate islets of different sizes in the splenic lobe of control (e) and RIP-DTR (f) pancreas. (g) Graph depicting the overall number of islets in the pancreas of sham or diphtheria toxin treated mice. (h) Graph showing the average cumulative beta cell volume of all islets within their respective size categories. Data presented as mean ± SE with n = 4 animals per group. Statistical significance indicated as ^*P < 0.05, ^**P < 0.01, ^***P < 0.001. Scale bars = 2 mm.

Figure 2

Repeated in vivo confocal imaging details the destruction of RIP-DTR islets over time. (a,b) Consecutively collected photographs (a) and confocal images (b) of engrafted RIP-DTR positive and negative islets in the eye of a diphtheria toxin-resistant control mouse. Yellow bar indicates diphtheria toxin treatment, dashed yellow circles in (a) show examples of individual islets followed over time in (b) using backscattered light to monitor morphological changes. (c) Graph showing the relative islet volume of RIP-DTR positive and negative islets in the anterior chamber of the eye over time following DT treatment. Islet volumes are shown averaged per mouse (grey lines, n = 4–11 islets per eye) and averaged per animal (black lines, n = 4 mice). (d) Relative decrease in islet volume (day 8) of individual RIP-DTR positive islets plotted against their initial starting volume (day 0). Linear regression analysis shows that no size-dependent sensitivity to DT-treatment exists (R² < 0.001, n = 20 islets). Islet size categories as defined for OPT data analysis (“S” = small, “M” = medium, “L” = large, see Fig. 1) are displayed for comparison purposes. Data presented as mean ± SD. Statistical significance indicated as ^***P < 0.001. Confocal images are shown as maximum intensity projections. Scale bars = 50 μm.

Figure 3

Metabolic features of the RIP-DTR mouse. (a,b) Unfasted blood glucose levels (a) and 4 hours fasted plasma insulin levels (b) measured daily following diphtheria toxin (solid line) or sham treatment (dashed line) of RIP-DTR mice. Control, n = 10 and DT, n = 5. (c,d) Graphs of glucose tolerance tests performed one week before and 40 hours after DT treatment of RIP-DTR positive and negative mice of mixed genders. Area under the curve (A.U.C.) was calculated, showing a significant difference 40 hours after treatment. Control, n = 9 and DT, n = 13. Data presented as mean ± SE. Statistical significance indicated as ^*P < 0.05, ^**P < 0.01, ^***P < 0.001. n.s.: nonsignificant.

Figure 4

Changes in beta cell mass and insulin expression patterns during RIP-DTR islet destruction. (a) Photographs and confocal images of RIP-DTR islets engrafted in the anterior chamber of the eye of RIP-DTR mice over the course of four days following diphtheria toxin administration (yellow bar). Reflected light images of the encircled islet are shown over time. (b) Quantification of confocal images per animal (grey lines, n = 3–10 islets) and overall mean ± SD (black line, n = 4). Graph shows a decrease of relative islet volumes, representing a gradual beta cell destruction. (c–e) Histological examination of paraffin sections from the RIP-DTR pancreas collected two days after diphtheria toxin treatment depicts islets have a normal morphology based on haematoxylin and eosin staining (c) whereas anti-insulin staining (green) shows a large spread in the fluorescence intensity (d,e). Graph shows individual insulin positive cells analysed (circles) and mean ± interquartile range (n = 3 mice, intensities are normalised per mouse to the average intensity). Confocal images are shown as maximum intensity projections. Scale bars = 50 μm.

Figure 5

Insulin content and intracellular Ca²⁺ signalling of RIP-DTR islets one day after diphtheria toxin treatment. (a) Representative confocal images of isolated islets one day after DT or sham treatment of RIP-DTR mice. Reflected light signal shows that DT-treated islets have preserved insulin granule scattering properties and similar morphology compared to controls. (b) Quantification of whole islet insulin content analysed by ELISA and corrected for total DNA content (mean ± SE, n = 8–9 samples per condition). (c) Representative traces from [Ca²⁺]_i measurements of islets loaded with Fura-2 AM and perfused for 2,000 s with 3 or 11 mM glucose and 25 mM KCl. (d–f) Analysis of the [Ca²⁺]_i traces (n = 10–12 islets per condition) for response time to the stimulus (time to reach 50% of the peak) and relative increase of fluorescence (peak height). Whiskers represent tukey (d) or min/max values (e,f). Islets originated from 3 mice per condition. Statistical significance indicated as ^*P < 0.05, ^**P < 0.01, ^***P < 0.001, n.s.: nonsignificant. Confocal images are shown as maximum intensity projections. Scale bar = 50 µm.