Year: 2024
Type: article
Abstract: Volume electron microscopy has become an important set of techniques for determining the nanoscale 3D ultrastructure of cells and biological tissues prepared by heavy atom staining and embedding in epon or other plastic materials. In the focused-ion-beam scanning electron microscope (FIB-SEM), thin layers of material are successively removed from the surface of the specimen block by Ga ion abrasion and the surface layer is imaged with the backscattered electron signal [1-3]. Experience has shown that 5-10 nm spatial resolution is achievable using FIB-SEM, and the resolution is approximately isotropic parallel and perpendicular to the block-face. Images are typically acquired with primary electron beam energies of ∼1.5 keV, and a fluence of ∼600 electrons/nm2, producing images of higher quality than can be achieved with the alternative 3D volume EM method of serial block face (SBF)-SEM. Figure 1 illustrates the 3D ultrastructure seen with FIB-SEM from a specimen of isolated blood platelets prepared using the protocol developed by Deerinck et al. [4]. Slices through the 3D volume, parallel to the block-face (x-y) and perpendicular to the block-face (x-z), reveal similar resolution in the two orthogonal planes (estimated as 5-10 nm), from which the subcellular ultrastructure of blood platelets can be visualized and segmented, including alpha granules, mitochondria, and canalicular system [5]. Interestingly, the z-resolution in the Monte Carlo (CASINO) simulations (Figure 2) is predicted to yield poorer z-resolution than that observed experimentally in the FIB-SEM [6, 7]. The stain in the simulation is modeled as atomic 3% of Pb in a matrix of light atoms corresponding to epon embedding resin, as determined from a previous study [8]. When a comparison is made with SBF-SEM, the electron fluence tolerated by the FIB-SEM is much greater — at least 600 electrons/nm2 in the FIB-SEM compared with only ∼ 20 electrons/nm2 in the SBF-SEM. At doses exceeding 20 electrons/nm2 the block-face is found to collapse in SBF-SEM causing irregular cutting by the in-situ diamond knife, which can be attributed to ionization damage by the scanned electron probe resulting in loss of light atoms including H and O. We are currently investigating the cause of the apparent inconsistencies between the results from FIB-SEM and SBF-SEM using images acquired with both techniques, as well as from scanning transmission electron microscopy (STEM) and further Monte Carlo simulations [9]. FIB-SEM backscattered electron images from an embedded block of stained blood platelets showing typical slices through a platelet in the x-y plane and in the x-z plane. Similar quality is evident in the two orthogonal planes demonstrating nearly isotropic spatial resolution. Monte Carlo (CASINO) simulation [6, 7] of an epon block with square regions of Pb stain at an atomic concentration of 3%, contained within 5-nm thick non-overlapping layers situated at different depths below the block-face at distances from 0-5 nm, 5-10 nm, 10-15 nm, 15 -20 nm, 20-25 nm, 25-30 nm, 30-35 nm, and 35-40 nm. Five squares are visible suggesting that the z-resolution is around 25 nm, which is commensurate with the minimum slice thickness that can be removed from the block-face in SBF-SEM. However, the predicted z-resolution is poorer than can be actually achieved with FIB-SEM.
Source: Microscopy and Microanalysis
Language: English
Subfield: Structural Biology
Domain: Life Sciences
Open Access status: bronze
OpenAlex ID: W4401005593