3D FIB-SEM reconstruction of microtubule–organelle interaction in whole primary mouse β cells

Abstract

Microtubules play a major role in intracellular trafficking of vesicles in endocrine cells. Detailed knowledge of microtubule organization and their relation to other cell constituents is crucial for understanding cell function. However, their role in insulin transport and secretion is under debate. Here, we use FIB-SEM to image islet β cells in their entirety with unprecedented resolution. We reconstruct mitochondria, Golgi apparati, centrioles, insulin secretory granules, and microtubules of seven β cells, and generate a comprehensive spatial map of microtubule–organelle interactions. We find that microtubules form nonradial networks that are predominantly not connected to either centrioles or endomembranes. Microtubule number and length, but not microtubule polymer density, vary with glucose stimulation. Furthermore, insulin secretory granules are enriched near the plasma membrane, where they associate with microtubules. In summary, we provide the first 3D reconstructions of complete microtubule networks in primary mammalian cells together with evidence regarding their importance for insulin secretory granule positioning and thus their supportive role in insulin secretion.

High-precision targeting workflow for volume electron microscopy

“…For both techniques, the achievable X,Y resolution is comparable (3-4 nm). However, the use of an ion beam enables finer sectioning resolution, down to 3-4 nm (Wei et al, 2012;Xu et al, 2017;Hoffman et al, 2020;Müller et al, 2021). For this reason, while SBEM is mostly suitable for the imaging of very large volumes (tissues or small organisms) in a nonisotropic fashion, FIB-SEM is currently considered best suited to automatically acquire relatively small volumes (ranging from subcellular to a few cells) at high isotropic resolution.…”

Robust Segmentation of Cellular Ultrastructure on Sparsely Labeled 3D Electron Microscopy Images using Deep Learning

“…The ultrastructure of tumor cells and the tumor microenvironment is very different from the widely studied neural and other normal cells (Baba and Câtoi 2007, Zink et al, 2004, Coman and Anderson., 1955. Therefore, segmentation methodologies designed for these cells cannot be directly applied for cancer cell ultrastructure segmentation (Shen et al, 2017, Linsley et al, 2018, Vergara et al, 2020, Müller et al, 2021. To overcome these difficulties, we propose a simple yet effective Residual U-Net (ResUNet) deep learning model for segmentation of cancer ultrastructure (Figure 1D).…”

Granular detail of β cell structures for insulin secretion

“…Yet, the FIB-SEM data suggest that the effect of glucose is not simply to disinhibit this effect. Glucose stimulated an approximately threefold increase in the number of microtubules and an approximately threefold decrease in the average length of each microtubule, so that total tubulin polymer density remained unchanged (4). The images also show that microtubule ends and insulin secretory granules are enriched near the plasma membrane and indicate an important role for microtubules in positioning the granules for exocytosis.…”