4Pi microscopy reveals an impaired three-dimensional mitochondrial network of pancreatic islet β-cells, an experimental model of type-2 diabetes

Insulin production in pancreatic beta-cells is critically linked to mitochondrial oxidative phosphorylation. Increased ATP production triggered by blood glucose represents the beta-cells' glucose sensor. Type-2 diabetes mellitus results from insulin resistance in peripheral tissues and impaired insulin secretion. Pathology of diabetic beta-cells might be reflected by the altered morphology of mitochondrial network. Its characterization is however hampered by the complexity and density of the three-dimensional (3D) mitochondrial tubular networks in these cell types. Conventional confocal microscopy does not provide sufficient axial resolution to reveal the required details; electron tomography reconstruction of these dense networks is still difficult and time consuming. However, mitochondrial network morphology in fixed cells can also be studied by 4Pi microscopy, a laser scanning microscopy technique which provides an similar to 7-fold improved axial resolution (similar to 100 nm) over conventional confocal microscopy. Here we present a quantitative study of these networks in insulinoma INS-1E cells and primary beta-cells in Langerhans islets. The former were a stably-transfected cell line while the latter were transfected with lentivirus, both expressing mitochondrial matrix targeted redox-sensitive GFP. The mitochondrial networks and their partial disintegration and fragmentation are revealed by carefully created iso-surface plots and their quantitative analysis. We demonstrate that beta-cells within the Langerhans islets from diabetic Goto Kakizaki rats exhibited a more disintegrated mitochondrial network compared to those from control Wistar rats and model insulinoma INS-1E cells. Standardization of these patterns may lead to development of morphological diagnostics for Langerhans islets, for the assessment of beta-cell condition, before their transplantations. (C) 2010 Elsevier B.V. All rights reserved.