Zhang Y , Li S , Zeng C , Huang G , Zhu X , Wang Q , Wang K , Zhou Q , Yan C , Zhang W , Yang G , Liu M , Lei J , Shi Y .

31930059 National Natural Science Foundation of China (National Science Foundation of China), 81920108015 National Natural Science Foundation of China (National Science Foundation of China), 31621092 National Natural Science Foundation of China (National Science Foundation of China), 31430020 National Natural Science Foundation of China (National Science Foundation of China)

	Fig. 1: Three-dimensional organization of the NPCs in a local region of the X. laevis NE. a) Organization of the NPCs in an original tomogram slice as viewed along the nucleocytoplasmic axis. Some of the representative array-like densities of the LR are indicated by arrowheads. The thickness of the tomographic slice shown here is 8.9 Å. b) Organization of the NPCs in the reconstructed tomogram as viewed along the nucleocytoplasmic axis. As the outer boundary of the NPC, the LR appears to cushion the contacts among neighboring NPC particles. Reconstructions for the individual NPC subunits (CR, IR, NR and LR) and the ribosomes associated with TRAP + OST46 were back-projected onto the original tomograms based on the refined coordinates of the individual particles. Shown here is a section of the NE from a. Scale bar, 50 nm. c) Organization of the NPCs in the reconstructed tomogram as viewed perpendicular to the nucleocytoplasmic axis. In contrast to other ring scaffolds of the NPC, the LR resides in the lumen. 40S: Small ribosome subunit; 60S: Large ribosome subunit; TRAP: translocon-associated protein complex; OST: oligosaccharyl-transferase.
	Fig. 2: Key features of the LR in the cryo-ET reconstruction are confirmed by an independent cryo-EM reconstruction of the X. laevis NPC. a) Reconstruction of a representative NPC particle by sub-tomogram averaging (STA). A top view and a tilt-45° view are shown. The individual subunits of the CR, IR, NR and LR were projected back into the original tomograms to allow reconstruction of a number of NPC particles. Shown here is a representative NPC particle, which contains four ring scaffolds: CR (colored yellow), IR (pink), NR (cyan) and LR (marine). Viewed perpendicular to the NE (left panel), the IR and LR define the inner and outer diameters, respectively, of the cylindrical NPC. b) Reconstruction of the NPC by single particle analysis (SPA). CR, IR and NR were reconstructed using the C8 symmetry. The LR was reconstituted using the refined LR subunit and the C8 symmetry. The LR is highlighted. c) The key structural features of the LR subunit are nearly identical between the STA reconstruction (marine, top panel, accession code EMD-0983) and the SPA reconstruction (gray, bottom panel, accession code EMD-0982). Their overlay is shown in the middle panel. Scale bar, 10 nm.
	Fig. 3: Structure of the LR subunit by the SPA approach. a) Eight LR subunits form a continuous circular scaffold. The overall structure of the LR (colored marine) is viewed along the nucleocytoplasmic axis. b) Structure of the LR subunit. Each butterfly-shaped LR subunit comprises two symmetric wings: Wing-A (colored green) and Wing-B (blue), which interact with each other through an extended interface (orange). The LR subunit has a Finger domain that contacts the fusion and a Grid domain that make up the bulk of the circular LR scaffold. Three mutually perpendicular views are shown. c) Each wing of the LR subunit comprises four elongated, tubular protomers. These protomers (numbered 1–4) associate with each other in a planar fashion. Each protomer contains an arm (colored gold) at one end, a central hub (magenta) and a leg (blue) at the other end. Scale bar, 10 nm. d) The Finger domain directly contacts the fusion of nuclear membranes. The tips of the protomers likely traverse the pore membrane and anchor the LR subunit to the pore. The cross section of the Finger domain has the shape of a diamond as indicated by dotted lines (right panel). e) The Bumper domain is formed between two neighboring LR subunits. Four legs from Wing-A of an LR subunit interact with four legs from Wing-B of the neighboring LR subunit to form the Bumper domain. Two perpendicular views are shown. f) The LR may stabilize both the concave and the convex curvatures of the fusion. The concave curvature relates to the diameter of the fusion within the NE, whereas the convex curvature is defined by the separation of the INM and ONM. Scale bar, 20 nm.
	Fig. 4: Structure of the Bumper domain of the LR by the STA approach. a) The Bumper domain adopts two major conformations. By back-projecting the NPC particles onto the original tomograms, two major conformations of the Bumper domain are seen and named Bumper-6 and -7. The lengths of Bumper-6 and -7 are 29 and 34 nm, respectively. b) Classification of the Bumper domains. The Bumper domains were re-cropped and classified. Bumper-6 and -7 represent 21.5% and 38.4% of the total particles classified. c) Refined structure of Bumper-7 exhibits distinguishing features (upper panel). Two mutually perpendicular views are shown for an isolated Bumper-7 (middle and lower panels). d) Refined structure of Bumper-6. In contrast to Bumper-7, the internal interface of Bumper-6 involves two pairs of promoter tips. e) Mapping the Bumper domains onto the NPC particles. The reconstructions for Bumper-7 (marine), Bumper-6 (red) and the LR subunit (gray) were back-projected onto the original tomograms based on the refined coordinates of the individual particles. Shown here is a representative section of the nuclear membrane. The conformations of nearly all the classified Bumper domains are the same as those of the Bumper domains formed by independently back-projecting the LR subunits onto the original tomograms.
	Fig. 5: The Bumper domain may cushion neighboring NPCs. a) NPCs are generally deformed in the X. laevis nuclear membrane. Statistical analysis of the deformation among neighboring NPCs. For each NPC, the distance to its nearest neighbor is measured and plotted here. The median distance is 136 ± 31 nm. The most frequently observed distance is 127.5 nm, which occurs to 1208 pairs of NPC. The sizes of a symmetric LR is shown in the inset for reference. b) Three representative examples of close contact between neighboring NPCs. The reconstruction for the LR subunit was back-projected onto the original tomograms based on the refined coordinates of the individual LR subunit particles.
	Fig. S1- Representative 0°-tilt micrographs and the associated defocus values for three tilt-schemes used in this study. a) Dose- symmetric tilt-scheme (-60° to 60°). 240 tilt-series were recorded. Shown here (tomo 586) is a representative example. b) Bidirectional tilt-scheme (0° to -60° and 3° to 60°). 874 tilt-series were recorded. Shown here (tomo 453) is a representative example. c) Continuous tilt-scheme (-60° to 60°). 311 tilt-series were recorded. Shown here (tomo 687) is a representative example. In each of the three cases, the 0°-tilt micrograph and the defocus value for the specified tilt-series are shown in the upper and lower panels, respectively.
	Fig. S2- A processing flow chart of the cryo-ET data on the NPC from Xenopus laevis (X. laevis). For detailed description, please refer to the Methods. CR, cytoplasmic ring; IR, inner ring; LR, luminal ring; NR, nuclear ring.
	Fig. S3- Average resolution and anisotropy of the cryo-ET reconstructions for the NPC from X. laevis. a) The STA-based cryo-ET reconstructions of the CR, IR, NR, and LR subunits. The resolution range for each of these reconstructions is color-coded. b) The Fourier shell correlation (FSC) curves for the reconstructions of the CR, IR, NR, and LR subunits. On the basis of the FSC criterion of 0.143, the reconstructions of the CR, IR, NR, and LR subunits have average resolutions of 9.1 Å, 13.1 Å, 13.6 Å, and 15.1 Å, respectively. c) The cryo-ET reconstruction is anisotropic. Shown here are the Fourier shell correlation (FSC) curves of the CR, IR, NR and LR subunits in three different orientations. The FSC curves for tilt 0°, 45°, and 90° (named Tilt-0°, Tilt-45° and Tilt-90°) were calculated with cone masks in Fourier space and colored red, blue, and green, respectively. As a reference, the FSC curve for the global CR subunit is shown and colored black. The resolutions were calculated on the basis of the FSC criterion of 0.143. d) Angular distribution of particles for reconstruction of the CR, IR, NR and LR subunits.
	Fig. S4- Three-dimensional organization of the NPC particles in a local region of the X. laevis NE. Four consecutive tomographic 2D slices are displayed in the left columns of panels a through d, with panel a on the cytoplasmic side and panel d on the nucleoplasmic side. These slices, each measuring 8.89 Ã… in thickness, are evenly spaced with a separation of 17.78 nm between two neighbouring slices. Structural interpretation of the regions is shown in the right columns of panels a through d, where the reconstructions for the NPC subunits and ribosomes were back-projected onto the original tomograms based on the refined coordinates of the individual particles. Ribosomes: 40S: Small ribosome subunit; 60S: Large ribosome subunit; TRAP: translocon-associated protein complex; OST: oligosaccharyl transferase.
	Fig. S5- A flow chart for the processing of the cryo- EM data on the LR subunit of the NPC from X. laevis. Single-particle analysis (SPA) was applied in data processing. For detailed description, please refer to the Methods.
	Fig. S6- Reconstruction of the LR subunit of the X. laevis NPC by the SP A-based cryo-EM approach. a) The SP A cryo-EM reconstruction of the LR subunit. The resolution range is color-coded. Two mutually perpendicular views are shown. The corresponding cylinder representations of angular distribution are shown on the right. b) The FSC curve for the SPA-based reconstruction of the LR subunit. On the basis of the FSC criterion of 0.143, the reconstruction of the LR subunit has an average resolution of 10.7 Ã….
	Fig. S7- Structural comparison of the cryo-ET reconstructions of the NPC. Surface representations of cryo-ET reconstructions are shown to scale for X. laevis (this study), X. laevis (accession code EMD 3005-3008)1, Homo sapiens (EMD 3103)2, Chlamydomonas reinhardtii (EMD 4355)3 and Saccharomyces cerevisiae (EMD 7321)4. Three views of each reconstruction are shown: cross-section side view, 45°-tilt view, and top view from cytoplasmic side. The LR observed in this study is highlighted in marine.
	Fig. S8- Structure of the LR subunit by the STA approach. a) Eight LR subunits form a circular scaffold around the fusion between the inner (INM) and outer nuclear membranes (ONM). The NPC is viewed along the nucleocytoplasmic axis. Two complete LR subunits are highlighted. Within each LR subunit, two wings are colored green and marine. b) Structure of the LR subunit. Each butterfly-shaped LR subunit comprises two symmetric wings: Wing-A (green) and Wing-B (blue), which interact with each other through an extended interface (orange). The LR subunit has a Finger domain and a Grid domain. Three mutually perpendicular views are shown. c) Each wing of the LR subunit comprises four planar, elongated, tubular protomers (numbered 1 through 4). Each protomer contains an arm (colored gold), a hub (magenta), and a leg (blue). The four protomers in each wing form two pairs (1-2 and 3-4), with the arms of each pair connected to each other through their tips. Scale bar, 10 nm. d, The Finger domain directly contacts the fusion of nuclear membranes. e) The Bumper domain is formed between two neighboring LR subunits. Four legs from Wing-A of an LR subunit interact with four legs form Wing-B of the neighboring LR subunit to form the Bumper domain.
	Fig. S9- Cryo-ET analysis of Bumper-6 and Bumper- 7. a) The local resolutions for Bumper-6 and Bumper-7 are color-coded in the cryo- ET reconstructions. Scale bar, 10 nm. b) The FSC curves for Bumper-6 (red) and Bumper-7 (blue). The resolutions for Bumper-6 and -7 are estimated to be 17.0 Ã… and 15.6 Ã…, respectively, on the basis of the FSC criterion of 0.143. c) A local region of the tomogram with the refined NPC particles. The reconstructions for Bumper-7 (marine), Bumper-6 (red), and the LR subunit (grey) were back-projected onto the original tomograms based on the refined coordinates of the individual particles. d) Another local region of the tomogram with the refined NPC particles. Coloring scheme is the same as in panel c.
	Fig. S10- Comparison of the STA and SPA reconstruction of the LR subunit. a) A central slice of the LR reconstruction by the STA approach. The average local resolution in the Finger and Grid domains is 12.6 Ã…. b) A central slice of the LR reconstruction by the SPA approach. For both the STA and SPA reconstructions, similar features are observed for the Finger and the Grid domains. In contrast to STA-based reconstruction, the two Bumper domains are missing in the SPA-based reconstruction, due to application of a considerably smaller alignment mask. c) The FSC curve of the SPA reconstruction with the STA reconstruction shows agreement of up to 13.2 Ã… based on the FSC criterion of 0.143. A common mask was used.
	Fig. S11- Domain structure of four candidate integral membrane proteins in the LR. a) A schematic diagram of the four candidate protein components of the vertebrate LR. The four proteins and the domains within each protein are drawn to scale. GP210 (Pom152 in yeast) is the largest of the three proteins, with 1898 amino acids. GP210 sequentially contains 15 predicted Ig- like domain, a C-domain rich in β-strands, a TM, and about 57 residues at the C- terminus. b) A topology diagram of the four candidate protein components of the LR. For GP210, all sequences N-terminal to the TM are in the lumen of the NE. In contrast, for NDC1, only the sequences connecting the six TMs are in the lumen; for POM121, only the N-terminal 31 residues reside in the lumen. TMEM33 has six predicted TMs and is the vertebrate homologue of yeast Pom33 but has not yet been experimentally confirmed as a Nup. c) The cryo-ET density of Pom152 (the yeast homolog of GP210), colored yellow here, is placed into the reconstruction for one LR protomer (colored grey). Pom152 is thought to contain nine Ig-like domains1.

Akey, Interactions and structure of the nuclear pore complex revealed by cryo-electron microscopy. 1989, Pubmed, Xenbase

Akey, Interactions and structure of the nuclear pore complex revealed by cryo-electron microscopy. 1989, Pubmed , Xenbase
Akey, Architecture of the Xenopus nuclear pore complex revealed by three-dimensional cryo-electron microscopy. 1993, Pubmed , Xenbase
Akey, Structural plasticity of the nuclear pore complex. 1995, Pubmed , Xenbase
Alber, The molecular architecture of the nuclear pore complex. 2007, Pubmed
Beck, The nuclear pore complex: understanding its function through structural insight. 2017, Pubmed
Beck, Snapshots of nuclear pore complexes in action captured by cryo-electron tomography. 2007, Pubmed
Braunger, Structural basis for coupling protein transport and N-glycosylation at the mammalian endoplasmic reticulum. 2018, Pubmed
Castaño-Díez, Dynamo: a flexible, user-friendly development tool for subtomogram averaging of cryo-EM data in high-performance computing environments. 2012, Pubmed
Castaño-Díez, Dynamo Catalogue: Geometrical tools and data management for particle picking in subtomogram averaging of cryo-electron tomograms. 2017, Pubmed
Chadrin, Pom33, a novel transmembrane nucleoporin required for proper nuclear pore complex distribution. 2010, Pubmed
Cohen, Nuclear pore protein gp210 is essential for viability in HeLa cells and Caenorhabditis elegans. 2003, Pubmed
Drummond, Interference with the cytoplasmic tail of gp210 disrupts "close apposition" of nuclear membranes and blocks nuclear pore dilation. 2002, Pubmed , Xenbase
Eibauer, Structure and gating of the nuclear pore complex. 2015, Pubmed , Xenbase
Frenkiel-Krispin, Structural analysis of a metazoan nuclear pore complex reveals a fused concentric ring architecture. 2010, Pubmed , Xenbase
Gerace, Identification of a major polypeptide of the nuclear pore complex. 1982, Pubmed
Görlich, Transport between the cell nucleus and the cytoplasm. 1999, Pubmed
Greber, A major glycoprotein of the nuclear pore complex is a membrane-spanning polypeptide with a large lumenal domain and a small cytoplasmic tail. 1990, Pubmed
Grossman, Functional architecture of the nuclear pore complex. 2012, Pubmed
Hagen, Implementation of a cryo-electron tomography tilt-scheme optimized for high resolution subtomogram averaging. 2017, Pubmed
Hallberg, An integral membrane protein of the pore membrane domain of the nuclear envelope contains a nucleoporin-like region. 1993, Pubmed
Hampoelz, Structure and Assembly of the Nuclear Pore Complex. 2019, Pubmed
Hao, Electron microscopy of Chaetomium pom152 shows the assembly of ten-bead string. 2018, Pubmed
Hoelz, The structure of the nuclear pore complex. 2011, Pubmed
Huang, Structure of the cytoplasmic ring of the Xenopus laevis nuclear pore complex by cryo-electron microscopy single particle analysis. 2020, Pubmed , Xenbase
Jarnik, Toward a more complete 3-D structure of the nuclear pore complex. 1991, Pubmed , Xenbase
Kim, Integrative structure and functional anatomy of a nuclear pore complex. 2018, Pubmed
Knockenhauer, The Nuclear Pore Complex as a Flexible and Dynamic Gate. 2016, Pubmed
Kremer, Computer visualization of three-dimensional image data using IMOD. 1996, Pubmed
Li, Acidic pH-Induced Conformations and LAMP1 Binding of the Lassa Virus Glycoprotein Spike. 2016, Pubmed
Lin, The Structure of the Nuclear Pore Complex (An Update). 2019, Pubmed
Löschberger, Super-resolution imaging visualizes the eightfold symmetry of gp210 proteins around the nuclear pore complex and resolves the central channel with nanometer resolution. 2012, Pubmed , Xenbase
Mahamid, Visualizing the molecular sociology at the HeLa cell nuclear periphery. 2016, Pubmed
Maimon, The human nuclear pore complex as revealed by cryo-electron tomography. 2012, Pubmed
Mansfeld, The conserved transmembrane nucleoporin NDC1 is required for nuclear pore complex assembly in vertebrate cells. 2006, Pubmed , Xenbase
Mastronarde, Automated electron microscope tomography using robust prediction of specimen movements. 2005, Pubmed
Meinema, Long unfolded linkers facilitate membrane protein import through the nuclear pore complex. 2011, Pubmed
Mosalaganti, In situ architecture of the algal nuclear pore complex. 2018, Pubmed
Pfeffer, Structure of the native Sec61 protein-conducting channel. 2015, Pubmed
Ptak, The multifunctional nuclear pore complex: a platform for controlling gene expression. 2014, Pubmed
Rohou, CTFFIND4: Fast and accurate defocus estimation from electron micrographs. 2015, Pubmed
Rothballer, Poring over pores: nuclear pore complex insertion into the nuclear envelope. 2013, Pubmed
Scheres, RELION: implementation of a Bayesian approach to cryo-EM structure determination. 2012, Pubmed
Schwartz, The Structure Inventory of the Nuclear Pore Complex. 2016, Pubmed
Sellés, Nuclear pore complex plasticity during developmental process as revealed by super-resolution microscopy. 2017, Pubmed , Xenbase
Spector, Macromolecular domains within the cell nucleus. 1993, Pubmed
Stavru, NDC1: a crucial membrane-integral nucleoporin of metazoan nuclear pore complexes. 2006, Pubmed
Stoffler, Cryo-electron tomography provides novel insights into nuclear pore architecture: implications for nucleocytoplasmic transport. 2003, Pubmed , Xenbase
Turoňová, Efficient 3D-CTF correction for cryo-electron tomography using NovaCTF improves subtomogram averaging resolution to 3.4Å. 2017, Pubmed
Upla, Molecular Architecture of the Major Membrane Ring Component of the Nuclear Pore Complex. 2017, Pubmed
von Appen, In situ structural analysis of the human nuclear pore complex. 2015, Pubmed
Wan, Structure and assembly of the Ebola virus nucleocapsid. 2017, Pubmed
WATSON, Further observations on the nuclear envelope of the animal cell. 1959, Pubmed
Wozniak, Primary structure analysis of an integral membrane glycoprotein of the nuclear pore. 1989, Pubmed
Wozniak, POM152 is an integral protein of the pore membrane domain of the yeast nuclear envelope. 1994, Pubmed
Zheng, MotionCor2: anisotropic correction of beam-induced motion for improved cryo-electron microscopy. 2017, Pubmed