Spatial and quantitative datasets of the pancreatic β-cell mass distribution in lean and obese mice

Abstract

A detailed understanding of pancreatic β-cell mass distribution is a key element to fully appreciate the pathophysiology of models of diabetes and metabolic stress. Commonly, such assessments have been performed by stereological approaches that rely on the extrapolation of two-dimensional data and provide very limited topological information. We present ex vivo optical tomographic data sets of the full β-cell mass distribution in cohorts of obese ob/ob mice and their lean controls, together with information about individual islet β-cell volumes, their three-dimensional coordinates and shape throughout the volume of the pancreas between 4 and 52 weeks of age. These data sets offer the currently most comprehensive public record of the β-cell mass distribution in the mouse. As such, they may serve as a quantitative and topological reference for the planning of a variety of in vivo or ex vivo experiments including computational modelling and statistical analyses. By shedding light on intra- and inter-lobular variations in β-cell mass distribution, they further provide a powerful tool for the planning of stereological sampling assessments.

Figure 1. Schematic workflow of the assay design.

Pancreata were isolated and divided into the three primary lobular compartments prior to whole mount immunohistochemistry, agarose embedding and tissue clearing. The processed samples were aligned (COM-AR) in the OPT prior to scanning. Two in-house developed image processing scripts were applied to the collected projection views, DFTA (uniforming alignment values) and CLAHE (equalising the contrast of the insulin labelled islets) before they were reconstructed into tomographic images (Data record A, Data Citation 1). The tomographic images were imported into the Imaris software where they were iso-sufaced (Data record B, Data Citation 1). Volumetric and spatial statistics was extracted in Imaris (Data record C, Data Citation 1) and images for visual reference exported (Data record D, Data Citation 1).

Figure 2. Examples of representative iso-surface rendered OPT images.

(a–j) Islet β-cell distribution in ob/ob (a–e) and lean control (f–j) pancreata (splenic lobe) between 4 and 52 weeks of age. The islet β-cell volumes are reconstructed based on the signal from insulin specific antibody staining (red) and the pancreas outline (gray) is based on the signal from tissue autofluorescense. Scale bar in (j) corresponds to 2 mm in (a–j).

Figure 3. Illustration of lobular separation.

The image shows a photomicrograph of a gut segment from a C57Bl/6 mouse at 8 weeks, encompassing the stomach, duodenum, pancreas and spleen. The broken white lines indicate the pancreas and its three primary, gastric, duodenal and splenic lobular compartments. DL, duodenal lobe; Duo, duodenum; GL, gastric lobe; SL, splenic lobe; Spl, spleen; Sto, stomach. Scale bar is 2 mm.

Figure 4. Schematic illustration depicting the layout of the data records (Data records A-D, Data Citation 1).

Each Data record folder tree is split at level 1 into ‘lean’ (Ctrl) and ‘Obese’ (ob/ob) mice. Level 2 contains the different time points (ages) when the pancreata were isolated. Level 3 represents the individual specimen IDs. Level 4 and 5 only exist for Data record A (Data Citation 1) where level 4 specifies the pancreatic lobe (gastric, duodenal or splenic) and level 5 the visualized channel (anatomy or Insulin respectively).