Abstract

We have applied serial block-face scanning electron microscopy (SBF-SEM) using a Zeiss SIGMA-VP SEM and a Gatan 3View system to measure parameters that describe the architecture of pancreatic islets of Langerhans, microscopic endocrine organs about 200 to 300 micrometers in size, which secrete insulin and glucagon for control of blood glucose. By analyzing entire mouse islets, we show that it is possible to determine (1) the distributions of alpha and beta cells, (2) the organization of blood vessels and pericapillary spaces, and (3) the ultrastructure of the individual secretory cells. Our results show that the average volume of a beta cell is nearly twice that of an alpha cell, and the total mitochondrial volume is about four times larger. In contrast, nuclear volumes in the two cell types are found to be approximately equal. Although the cores of alpha and beta secretory granules have similar diameters, the beta granules have prominent halos resulting in overall diameters that are twice those of alpha granules. Visualization of the blood vessels revealed that every secretory cell in the islet is in contact with the pericapillary space, with an average contact area of 9.5% of the cell surface area. Our data show that consistent results can be obtained by analyzing small numbers of islets. Due to the complicated architecture of pancreatic islets, such precision cannot easily be achieved by using TEM of thin sections. A combination of 2D and 3D analyses of tissue volume ultrastructure acquired by serial block face scanning electron microscopy (SBF-SEM) can greatly shorten the time required to obtain quantitative information from big data sets that contain many billions of voxels. Thus, to analyze the number of organelles of a specific type, or the total volume enclosed by a population of organelles within a cell, we have shown that it is possible to estimate the number density or volume fraction of that organelle using a stereological approach to analyze randomly selected 2D slices through the cells, and to combine such estimates with precise measurement of 3D cell volumes by delineating the plasma membrane in successive slices. The validity of such an approach can be easily tested since the entire 3D tissue volume is available in the SBF-SEM data set. We have applied this hybrid 3D/2D technique to determine the number of secretory granules in alpha and beta cells of mouse pancreatic islets of Langerhans, and have been able to estimate the total insulin content of beta cells. These results are in agreement with measured values. The spatial resolution of SBF-SEM normal to the block face is currently limited to approximately 25 nanometers by the minimum slice thickness that can be removed using the ultramicrotome that is built into the SEM's specimen stage. We have carried out Monte Carlo simulations of electron trajectories within the block face to determine whether it is possible to obtain sub-25 nanometer z-resolution by recording backscattered images at different beam energies to probe different sub-surface depths within the block. Results show the feasibility of achieving a z-resolution of around 10 nanometers by combining two or more backscattered images for electrons with primary energy between 1 keV and 3.5 keV. We have tested this capability on well-defined test specimens, and are now applying the technique to determine cellular ultrastructure with improved z-resolution. I another study, we have compared the performance of SBF-SEM with STEM tomography, by considering the 3-D ultrastructure of human blood platelets. We find that many features of the complex membranes composing the platelet organelles can be determined from both approaches. STEM tomography provides a higher spatial resolution (3 nm isotropic resolution) relative to SBF-SEM ( 5 nm in the plane of the block face and 25 nm perpendicular to the block face). On the other hand, SBF-SEM enables visualization of large numbers of entire platelets, each of which extends 2 m in the minimum dimension, whereas STEM tomography can only visualize a fraction of the platelet volume due to a rapid non-linear loss of signal in specimens of thickness greater than approximately 1.5 m. In addition, acquisition and image processing times are considerably shorter for SBF-SEM than for STEM tomography. SBF-SEM is capable of producing large 3D images of cellular ultrastructure, but the labor required to manually segment EM images into their semantic components hinders further data analysis. Currently, software pipelines incorporating deep neural networks offer state-of-the-art performance for automated segmentation. However, even state-of-the-art automated segmentation tools require extensive manual correction for many data sets of interest to the structural biology and systems biology communities, and are therefore impractical for image analysis. Our lab is designing novel neural networks and incorporating them into a segmentation software pipeline to improve automated segmentation performance for EM data sets taken from multiple biological systems. We are beginning to develop a design framework and software for constructing segmentation neural networks, and to test these methods on large 3D data sets generated in our laboratory. Whereas previous work in the field of deep learning in SBF-SEM has tended to focus on the identification of cell membranes for mapping neuronal circuits in brain, our approach aims to segment intracellular volumes into multiple classes of organelles for a diverse range of cell types. Preliminary results for blood platelets show considerable promise.

  Funding Agency

Agency
National Institute of Health (NIH)
Institute
National Institute of Biomedical Imaging and Bioengineering (NIBIB)
Type
Investigator-Initiated Intramural Research Projects (ZIA)
Project #
1ZIAEB000086-05
Application #
9555657
Study Section
Project Start
Project End
Budget Start
Budget End
Support Year
5
Fiscal Year
2017
Total Cost
Indirect Cost

  Institution

Name
Biomedical Imaging & Bioengineering
Department
Type
DUNS #
City
State
Country
Zip Code