ジャーナル: Proc Natl Acad Sci U S A / 年: 2023 タイトル: Interdigitated immunoglobulin arrays form the hyperstable surface layer of the extremophilic bacterium . 著者: Andriko von Kügelgen / Sofie van Dorst / Keitaro Yamashita / Danielle L Sexton / Elitza I Tocheva / Garib Murshudov / Vikram Alva / Tanmay A M Bharat / 要旨: is an atypical diderm bacterium with a remarkable ability to tolerate various environmental stresses, due in part to its complex cell envelope encapsulated within a hyperstable surface layer (S- ... is an atypical diderm bacterium with a remarkable ability to tolerate various environmental stresses, due in part to its complex cell envelope encapsulated within a hyperstable surface layer (S-layer). Despite decades of research on this cell envelope, atomic structural details of the S-layer have remained obscure. In this study, we report the electron cryomicroscopy structure of the S-layer, showing how it is formed by the Hexagonally Packed Intermediate-layer (HPI) protein arranged in a planar hexagonal lattice. The HPI protein forms an array of immunoglobulin-like folds within the S-layer, with each monomer extending into the adjacent hexamer, resulting in a highly interconnected, stable, sheet-like arrangement. Using electron cryotomography and subtomogram averaging from focused ion beam-milled cells, we have obtained a structure of the cellular S-layer, showing how this HPI S-layer coats native membranes on the surface of cells. Our S-layer structure from the diderm bacterium shows similarities to immunoglobulin-like domain-containing S-layers from monoderm bacteria and archaea, highlighting common features in cell surface organization across different domains of life, with connotations on the evolution of immunoglobulin-based molecular recognition systems in eukaryotes.
Movies collected at the scope were clustered into optics groups based on the XML meta-data of the data-collection software EPU (Thermo Fisher Scientific) using a k-means algorithm implemented in EPU_group_AFIS (https://github.com/DustinMorado/EPU_group_AFIS). Imported movies were motion-corrected, dose-weighted, and Fourier cropped (2x) with MotionCor2 implemented in RELION-3.1. Contrast transfer functions (CTFs) of the resulting motion-corrected micrographs were estimated using CTFFIND4.
粒子像選択
選択した数: 882866 詳細: Initially, side views of S-layer sheets were first manually picked along the edge of the lattice using the helical picking tab in RELION while setting the helical rise to 40 Angstrom. Top and ...詳細: Initially, side views of S-layer sheets were first manually picked along the edge of the lattice using the helical picking tab in RELION while setting the helical rise to 40 Angstrom. Top and tilted views were manually picked at the central hexameric axis. Manually picked particles were extracted in 4x downsampled 100 x 100 boxes and classified using reference-free 2D classification inside RELION3.1. Class averages centered at a hexameric axis were used to automatically pick particles inside RELION3.1. Automatically picked particles were extracted in 4x downsampled 100x100 pixel2 boxes and classified using reference-free 2D classification. Particle coordinates belonging to class averages centered at the hexameric axis were used to train TOPAZ in 5x downsampled micrographs with the neural network architecture conv127. For the final reconstruction, particles were picked using TOPAZ and the previously trained neural network above. Additionally, top, bottom, and side views were picked using the reference-based autopicker inside RELION3.1, which TOPAZ did not readily identify. Particles were extracted in 4x downsampled 100x100 pixel2 boxes and classified using reference-free 2D classification inside RELION3.1. Particles belonging to class averages centered at the hexameric axis were combined, and particles within 30 Angstrom were removed to prevent duplication after alignment. All resulting particles were then re-extracted in 4x downsampled 100x100 pixel2 boxes.
初期モデル
モデルのタイプ: NONE 詳細: All side views and a subset of top and bottom views were used for initial model generation in RELION-3.1. The scaled and lowpass filtered output was then used as a starting model for 3D auto ...詳細: All side views and a subset of top and bottom views were used for initial model generation in RELION-3.1. The scaled and lowpass filtered output was then used as a starting model for 3D auto refinement in a 512x512 pixel2 box.
最終 再構成
使用したクラス数: 1 / 想定した対称性 - 点群: C6 (6回回転対称) / アルゴリズム: FOURIER SPACE / 解像度のタイプ: BY AUTHOR / 解像度: 2.52 Å / 解像度の算出法: FSC 0.143 CUT-OFF / ソフトウェア - 名称: RELION (ver. 3.1) 詳細: Per-particle defocus, anisotropy magnification, and higher-order aberrations were refined inside RELION3.1, followed by another round of focused 3D auto refinement and Bayesian particle ...詳細: Per-particle defocus, anisotropy magnification, and higher-order aberrations were refined inside RELION3.1, followed by another round of focused 3D auto refinement and Bayesian particle polishing. The final map was obtained from 55,345 particles and post-processed using a soft mask focused on the central hexamer, including the dimeric bridge, yielding a global resolution of 2.52 Angstrom according to the gold standard Fourier shell correlation criterion of 0.143. 使用した粒子像数: 55345
初期 角度割当
タイプ: MAXIMUM LIKELIHOOD / ソフトウェア - 名称: RELION (ver. 3.1) / 詳細: Angle assignment was performed within RELION3.1
最終 角度割当
タイプ: MAXIMUM LIKELIHOOD / ソフトウェア - 名称: RELION (ver. 3.1) / 詳細: Angle assignment was performed within RELION3.1
最終 3次元分類
ソフトウェア - 名称: RELION (ver. 3.1)
FSC曲線 (解像度の算出)
-
原子モデル構築 1
精密化
空間: RECIPROCAL / プロトコル: AB INITIO MODEL / 温度因子: 19.48 / 当てはまり具合の基準: Best Fit