Improved Visualization of Vertebrate Nuclear Pore Complexes by Field Emission Scanning Electron Microscopy

Structure

Ways & Means Improved Visualization of Vertebrate Nuclear Pore Complexes by Field Emission Scanning Electron Microscopy Lihi Shaulov1 and Amnon Harel1,* 1Department of Biology, Technion - Israel Institute of Technology, Haifa 32000, Israel *Correspondence: [email protected] DOI 10.1016/j.str.2012.01.022

SUMMARY

Field emission scanning electron microscopy (FESEM) can provide high-resolution three-dimensional surface imaging of many biological structures, including nuclear envelopes and nuclear pore complexes (NPCs). For this purpose, it is important to preserve NPCs as close as possible to their native morphology, embedded in undamaged nuclear membranes. We present optimized methodologies for FESEM imaging in a cell-free reconstitution system and for the direct visualization of mammalian cell nuclei. The use of anchored chromatin templates in the cell-free system is particularly advantageous for imaging fragile intermediates inhibited at early stages of assembly. Our new method for exposing the surface of mammalian nuclei results in an unprecedented quality of NPC images, avoiding detergentinduced and physical damage. These new methodologies pave the way for the combined use of FESEM imaging with biochemical and genetic manipulation, in cell-free systems and in mammalian cells.

INTRODUCTION A hallmark of eukaryotic cells is the compartmentalization of nucleic acid synthesis and processing within the nucleus, and their separation from protein synthesis in the cytoplasm. This separation is achieved by the double membranes of the nuclear envelope (NE) and the selective transport channels of nuclear pore complexes (NPCs) embedded within this barrier (Gerace and Burke, 1988; Hetzer and Wente, 2009; Onischenko and Weis, 2011). NPCs, composed of individual proteins called nucleoporins, are huge gateways that provide the only means for the exchange of molecules between the nucleus and the cytoplasm. They are characterized by a massive central framework with 8-fold rotational symmetry and additional filamentous structures extending from the nuclear and cytoplasmic facades (Fahrenkrog et al., 2004; Fernandez-Martinez and Rout, 2009; Harel and Gruenbaum, 2011; Vasu and Forbes, 2001). The central framework is anchored via integral membrane nucleoporins to a highly curved pore-membrane domain connecting the inner and outer nuclear membranes.

Being one of the largest and most complex macromolecular assemblies known in cells, the NPC presents a formidable challenge for structural determination. NPC structure has been studied for over five decades by various techniques of transmission electron microscopy (for review, see Lim et al., 2008). Significant progress has been made in recent years by cryoelectron tomography of whole NPCs and X-ray crystallography of individual nucleoporins and their complexes (for review, see Brohawn et al., 2009; Hoelz et al., 2011; Maimon and Medalia, 2010). Despite these important strides, there is a large resolution gap between the structures obtained by these two methods and a detailed atomic level structural model of the whole NPC is currently beyond our reach. Another important strategy for structural studies of the NE and NPCs is three-dimensional surface imaging by high-resolution field emission scanning electron microscopy (FESEM). Pioneering work by Terry Allen and colleagues has provided striking images of both the cytoplasmic and nucleoplasmic facades of NPCs in manually removed NEs from amphibian oocytes (Goldberg and Allen, 1992, 1995). Additional views of internal substructures of the NPC were revealed by various techniques of fracturing and proteolytic treatment and by studies of early Drosophila embryos at the syncytial stage of development (Goldberg and Allen, 1993, 1996; Kiseleva et al., 2001). The visualization of NPCs in tissue culture cells by FESEM requires the exposure of the nuclear surface and has mainly relied on protocols employing detergents, which result in various degrees of damage to the nuclear membranes (Allen et al., 2007a; Drummond et al., 2006; Walther et al., 2003). Importantly, FESEM imaging does not depend on averaging techniques and can reveal subtle differences between individual NPCs, while preserving them close to their native morphology, embedded within the NE. In multicellular organisms, NPC assembly occurs during two separate stages of the cell cycle. Open mitosis dictates the coordinated formation of NPCs and the double membranes of the NE at telophase (Burke and Ellenberg, 2002; Gu¨ttinger et al., 2009; Hetzer et al., 2005). New NPCs are also inserted into the intact, expanding NE during interphase, when NPC numbers are doubled (D’Angelo et al., 2006; Maul et al., 1972). Much of the mechanistic insight on NPC assembly has been gleaned from in vitro reconstitution systems based on amphibian egg extracts (Forbes et al., 1983; Harel et al., 2003b; Lohka and Masui, 1983; Macaulay and Forbes, 1996; Rotem et al., 2009; Walther et al., 2003). These cell-free extracts can be used to reconstitute nuclear assembly in the test tube and allow direct

Structure 20, 407–413, March 7, 2012 ª2012 Elsevier Ltd All rights reserved 407

Structure Scanning Electron Microscopy of Nuclear Pores

Figure 1. Nuclear Reconstitution on Anchored Chromatin Templates (A) Schematic representation of nuclear reconstitution in the cell-free egg extract system using anchored chromatin templates. Chromatin is preattached to silicon chips and incubated in a reconstitution mixture containing: cytosol, membrane vesicles and an ATP regenerating system. Membrane vesicles are targeted to the surface of chromatin and start fusing to form flattened sheets, toward the formation of a continuous double-membrane envelope. The fully assembled NE contains mature NPCs embedded within the membranes. (B) FESEM imaging of nuclear reconstitution on anchored chromatin templates. Normal assembly (left) results in an intact NE with abundant NPCs after 1 hr. The addition of 5 mM BAPTA to the reconstitution mixture at t = 0, results in smoothly sealed membranes with no NPCs (middle). When 15 mM of full-length importin b is added at t = 0, membrane vesicles are targeted to chromatin but fail to fuse and form a continuous envelope (right). See also Figure S1.

access to the surface of growing or mature NEs by FESEM. In the Xenopus reconstitution system, FESEM has been extensively used in conjunction with immunodepletion, specific inhibitors of assembly and the addition of recombinant proteins (Goldberg et al., 1997; Harel et al., 2003a, 2003b; Rotem et al., 2009; Walther et al., 2003; Wiese et al., 1997). Here, we present optimized methodologies for the visualization of NEs and NPCs by FESEM. First, we describe the use of anchored chromatin templates in the Xenopus cell-free reconstitution system. This method has distinct advantages for the imaging of nuclear assembly intermediates and for subsequent immunogold labeling and localization of specific proteins of interest. In parallel, we developed a new method for exposing the outer surface of nuclei from mammalian tissue culture cells, without the use of detergent. This treatment facilitates highly improved FESEM imaging, displaying large expanses of undamaged NPCs embedded in intact membranes. RESULTS Anchored Chromatin Templates for Nuclear Assembly FESEM analysis of nuclear assembly in the Xenopus cell-free reconstitution system has previously been performed by lowspeed centrifugation of the samples onto silicon chips, followed by fixation (Allen et al., 2007b; Goldberg et al., 1997; Harel et al., 2003b). A disadvantage of this method is that excess membranous material can be pelleted down onto the NE surface.

When aborted assembly intermediates are concerned, an incomplete or pore-free NE often results in fragile structures that are easily damaged by centrifugation. To circumvent these problems and limit the amount of membranes included in the reaction mixture, we used the anchored chromatin assembly method. This method was originally developed for the light microscope and involves the prior attachment of chromatin templates to a solid support before the addition of egg extract components (Macaulay and Forbes, 1996). A schematic representation of the stepwise assembly process is shown in Figure 1A. We have already used a similar setup, while omitting the membrane vesicle fraction from the assembly mixture, to study the first ‘‘chromatin seeding’’ step in postmitotic NPC assembly (Rotem et al., 2009; see also Figure 2A below). When the full extract system, including membranes, is used, normal NEs are formed with NPCs embedded in the nuclear membranes (Figure 1B, left). The overall shape of these anchored nuclei remains flat and elongated, because the tethered chromatin is restricted from expansion into a spherical nucleus (Macaulay and Forbes, 1996; see also Shaulov et al., 2011). FESEM Imaging of Aborted Assembly Intermediates The calcium chelator BAPTA (1,2-bis-[o-Aminophenoxy]ethane-N,N,N’,N’-tetra acetic acid) is a known inhibitor of nuclear assembly, which has been shown by transmission EM to cause the formation of a double membrane, pore-free NE (Macaulay and Forbes, 1996). When BAPTA-inhibited assembly

408 Structure 20, 407–413, March 7, 2012 ª2012 Elsevier Ltd All rights reserved

Structure Scanning Electron Microscopy of Nuclear Pores

observed within these NE ‘‘patches’’ (Figure S1B). This observation has important implications for the continued mechanistic debate on NPC assembly (see Discussion).

Figure 2. Immunogold Labeling of Exposed Chromatin and Mature NPCs (A) A single anchored chromatin template incubated in membrane-free cytosol is shown at low magnification in the top-left panel. Details of the surface topology at higher magnification can be seen in an in-lens image in the topright panel. Following immunolabeling with anti-Nup107 and gold-conjugated secondary antibodies, the positions of gold particles are identified in a backscatter electron detector image of the same area (bottom left). A combined view with pseudocolored gold particle positions superimposed on the in-lens image is shown in the bottom-right panel. (B) A single anchored chromatin template incubated in the full reconstitution mixture, including membranes, is shown at low magnification after NE assembly was completed (left). Note that the nuclear surface area is much expanded compared to the chromatin template that was incubated in membrane-free cytosol, shown in (A). Higher magnification reveals mature NPCs embedded in the NE and immunolabeled with anti-Nup62 antibodies (right). Pseudocolored gold particle positions, determined from the backscatter electron detector, are superimposed on the in-lens image.

intermediates are viewed by FESEM, a smooth membrane surface is revealed, with no sign of architectural elements resembling any part of the mature NPC (Figure 1B, middle). When an excess of the nuclear transport receptor importin b is added into the reconstitution mixture, nuclear assembly is inhibited at an earlier stage. Membrane vesicles of different sizes are observed on the surface of chromatin, but vesicle-vesicle fusion appears to be inhibited and a continuous NE is not formed (Figure 1B, right). This is in agreement with our previous observation made with a fluorescent membrane dye (Harel et al., 2003a). Figure S1 (available online) demonstrates that the amount of membranes in the anchored assembly assay can be titrated down until they become limiting. Importantly, when limited areas of flattened membrane sheets are formed, mature NPCs can be

Immunogold Labeling of Anchored Nuclei Immunogold labeling is a powerful technique that can complement the high-resolution morphological analysis provided by FESEM imaging (Allen et al., 2007a, 2008; Rotem et al., 2009). Assembly on anchored chromatin templates is particularly suited for this purpose as it allows repeated incubations and washes. Anchored chromatin templates incubated in membrane-free cytosol facilitate chromatin ‘‘seeding’’ by the ELYS/Nup107160 complex, but no further steps in the assembly process (Rotem et al., 2009). Figure 2A shows the three-dimensional surface imaging of such templates by FESEM, together with immunolocalization of the seeding module with affinity purified anti-Nup107 antibodies. The precise localization to specific substructures observed in the in-lens image is achieved by the identification of gold-conjugated secondary antibodies via a backscatter electron detector (Allen et al., 2008; Rotem et al., 2009). The same technique can be applied to fully assembled NEs containing mature NPCs, as shown in Figure 2B. Here, affinity purified anti-Nup62 antibodies exclusively label the NPCs (Figure 2B, right). The staining pattern is consistent with previous studies that have indicated that Nup62 is anchored to the central pore region and has a flexible and mobile FG repeat domain (Schwarz-Herion et al., 2007). Standard controls, like omitting the primary antibodies, result in very low background labeling levels (see Experimental Procedures). In addition, it is sometimes possible to perform specific immunodepletion controls for the target protein(s) in question (Shaulov et al., 2011). As in other immunohistochemistry techniques, the quality of the data depends on the quality and specificity of the antibodies used. A New Method for Imaging Mammalian Cell Nuclei To visualize the NPCs of mammalian cells we first tested the available published methods, which are based on the use of permeabilizing detergents and/or dry fracturing (Allen et al., 2007a, 2008; Drummond et al., 2006). Figure S2 demonstrates the results we obtained with a protocol using 0.5% Triton X-100 on adherent cells. These results are comparable to previously published images from mammalian cells and exhibit NPCs that appear to be constricted or damaged and nuclear membranes that are not preserved in an intact form (Figure S2). As an alternative approach, we developed a protocol based on repeated hypotonic treatment and pipetting of detached cells, without the use of detergent. This protocol was tested on a primary culture of mouse lung fibroblasts and two widely used cell lines: HeLa and COS-1 (see Experimental Procedures). Although the finer details of the protocol had to be optimized for each cell type, the results were very similar and represent a dramatic improvement in the imaging of mammalian NEs and NPCs (Figures 3 and 4). These images are comparable in quality only to those obtained in the past from amphibian oocytes (see Discussion). A distinct advantage of FESEM is the ease and accessibility of searching large areas over a wide range of magnifications. A field of treated cells is shown at a low magnification in Figure 3A, with several examples of partially exposed

Structure 20, 407–413, March 7, 2012 ª2012 Elsevier Ltd All rights reserved 409

Structure Scanning Electron Microscopy of Nuclear Pores

Figure 3. Direct Surface Imaging of Nuclei from Mouse Lung Fibroblasts (A) Cells from a primary culture of mouse lung fibroblasts were subjected to hypotonic treatment and spun down onto silicon chips as described in Experimental Procedures. A field of cells is shown at low magnification with white circles marking four spherical bodies that were identified at higher magnification as nuclei. These nuclei were largely detached from the main cell body, or had significant portions of exposed NEs. (B) A detached nucleus shown at intermediate magnification. (C) Selected areas of intact NEs shown at a higher magnification. Mature NPCs are embedded in the undamaged membrane surface of the NE. See also Figure S2 to compare these results with those obtained by a previously published procedure.

or completely detached nuclei, which can be identified by zooming in to higher magnifications (Figures 3B and 3C). Although many nuclei can be identified, only a fraction of these is suitable for high-resolution imaging because of specimen charging. Special care should be taken with the critical point drying and chromium coating steps of the preparation to overcome this difficulty (see Experimental Procedures). Higher magnification of mouse lung fibroblast nuclei (Figure 3C), as well as HeLa and COS-1 cell nuclei (Figure 4), shows large expanses of well-preserved NEs with abundant NPCs embedded in intact membranes. The cytoplasmic facades of these NPCs exhibit a clear indication of the cytoplasmic filaments and in some cases a large particle can be observed in the central transport channel. Other NPCs provide a glimpse of the internal nuclear basket structure. Distinct differences between individual NPCs in the preparation can be observed in the gallery of magnified panels (bottom of Figure 4) and in the stereo pair representation shown in Figure S3. DISCUSSION One of the major challenges in current research in the nuclear transport field is placing individual nucleoporins within structural models of the fully assembled NPC. Recognition of the modular nature of the architectural scaffold of the NPC has led to increasingly large atomic resolution structures of the building blocks being solved and to various models attempting to connect these individual pieces into a full scaffold (Brohawn et al., 2009; Hoelz et al., 2011; Schwartz, 2005; Vetter, 2009). From the mechanistic point of view, it is still unclear how and in what order the different nucleoporin subcomplexes interact with each other and assemble the mature NPC. The only structural clues provided by previous FESEM imaging are proposed architectural intermediates (e.g., ‘‘stabilizing pores’’ and ‘‘star-rings’’) identified in the Xenopus reconstitution system (Goldberg et al., 1997). However, these putative assembly intermediates were only observed at low abundance, even in the presence of inhib-

itors, and were never correlated with the presence of specific protein components. Improved FESEM methodologies presented in this study will provide powerful tools for further investigation of NPC assembly and structure in cell-free systems as well as in mammalian cell research. The current consensus regarding the mechanism of postmitotic NPC assembly relates only to the first step in the process: chromatin ‘‘seeding’’ by the ELYS/Nup107-160 complex (Harel et al., 2003b; Rasala et al., 2008; Rotem et al., 2009; Walther et al., 2003). Opinions are divided as to the next steps and the involvement of membrane components in the assembly mechanism (for review, see Gu¨ttinger et al., 2009). We have recently demonstrated a dominant negative effect of an internal fragment of the membrane nucleoporin POM121 on NPC assembly in the Xenopus cell-free system (Shaulov et al., 2011). In the postmitotic mode of assembly, this fragment prevents the formation of a continuous NE leaving large regions of exposed chromatin, docked membrane vesicles and limited areas of fused membrane sheets, with no sign of NPCs (Shaulov et al., 2011). Remarkably, when we titrated down the amount of membranes in the anchored assembly assay (Figure S1B of this study), some NPCs were formed within limited areas of flattened membrane sheets. These observations, together with the recent analysis of ‘‘cold intermediates’’ assembled at 14 C (Fichtman et al., 2010), suggest that membrane components play a critical role at an early stage of assembly and that an inner/outer membrane fusion event may be needed for NPC formation. FESEM imaging of anchored chromatin assembly reactions will be important for further analysis of NPC biogenesis in the Xenopus cell-free reconstitution system. Many of the early assembly intermediates mentioned above are extremely fragile and could not be analyzed by previously published FESEM methods. An important future goal will be to design potential dominant negative constructs of additional nucleoporins, based on crystallographic data, and attempt to disrupt later stages in the assembly pathway. Mammalian nucleoporins are preferred targets for knockdown or knockout by genetic manipulation. However, the FESEM methodologies published to date could not be used for a reliable morphological characterization of affected cells

410 Structure 20, 407–413, March 7, 2012 ª2012 Elsevier Ltd All rights reserved

Structure Scanning Electron Microscopy of Nuclear Pores

Figure 4. High-Resolution Imaging of Nuclei from HeLa and COS-1 Cells (A) HeLa cells were processed for high-resolution surface imaging by FESEM as in Figure 3. See Experimental Procedures for details of optimizing the protocol for different cell types. An area of the intact NE, replete with NPCs, is shown at a high magnification with a selected gallery of individual NPCs further magnified at the bottom. (B) Similar analysis of COS-1 cells, with an intact NE area shown at parallel magnifications. Note that in both cases some NPCs contain large particles in the central channel area, whereas others display an apparently open channel or parts of the internal basket structure. When present, the central particles are not always oriented symmetrically in the middle of the channel and appear to have varying dimensions. See also Figure S3.

by direct visualization of NEs and NPCs. The new method presented here can pave the way for more experiments that will rely on high-resolution surface imaging of mammalian nuclei. Figures 3 and 4 of this study represent the highest quality of surface imaging showing mammalian NPCs close to their native morphology, i.e., embedded in undamaged nuclear membranes. This is comparable only to the quality of images obtained from manually removed NEs of amphibian oocytes (Goldberg and Allen, 1992, 1995). Although oocytes have many advantages, they exemplify a static preembryonic system that is very different from somatic cells. The mouse lung fibroblasts used in our study represent cells that can be derived from transgenic and knockout mice. HeLa and COS-1 cells represent cell lines that can be transfected with siRNAs or mutant forms of mammalian nucleoporins and related proteins. Thus, our improved FESEM methodologies lay the foundation for new attempts to decipher the complex structure-function relationships of the NPC by genetic manipulation of mammalian cells. EXPERIMENTAL PROCEDURES Egg Extracts and Nuclear Reconstitution Preparation of demembranated sperm chromatin, Xenopus egg extracts, and the fractionation of extracts and preparation of membrane-free cytosol were performed as previously described (Harel et al., 2003a; Rotem et al., 2009). For each new preparation of egg extracts, the optimal ratio between the membrane vesicle and cytosolic fractions was determined both in standard reconstitution assays (in solution) and in anchored nuclear assembly. For the standard solution assays, this ratio is generally 1:20–1:25 (membranes: cytosol), whereas the optimal ratio for anchored reactions is about 1:100. All assembly mixtures contained an ATP-regeneration system and 5 mg/ml nocodazole and were incubated for 1 hr at room temperature in a humidified chamber (Rotem et al., 2009). To prepare anchored assembly reactions for FESEM analysis, silicon chips (Ted Pella, USA) were pretreated with 0.2 mg/ml poly-lysine for 15 min and washed with H2O. Chromatin was decondensed in crude nucleoplasmin (Macaulay and Forbes, 1996) and allowed to settle onto the silicon chips at a final concentration of 1,500 sperm units/ml in 1XELB (10 mM HEPES [pH 7.6], 50 mM KCl, 2.5 mM Mg Cl2). Tethered chromatin templates were washed once in 1XELB and blocked in 5% BSA-ELB for 20 min. For the inhibition of nuclear assembly 5 mM BAPTA (Calbiochem) or 15 mM full-length untagged Xenopus importin b (Rotem et al., 2009) were added to the reconstitution mixture at t = 0. To analyze chromatin ‘‘seeding,’’ the membrane vesicle fraction was omitted from the assembly mixture and membrane-free cytosol was used. All templates were fixed in 4% formaldehyde (Electron Microscopy Sciences) in 1XELB for 10 min, and further processed for FESEM analysis.

Cell Culture and Exposure of Mammalian Cell Nuclei Mouse lung fibroblasts, HeLa cells, and COS-1 cells were grown under standard conditions to 80% confluency, detached by trypsinization, washed in PBS, and subjected to hypotonic treatment. Alternatively, detached cells were frozen in medium supplemented with fetal calf serum and DMSO, stored under liquid nitrogen, and subsequently thawed for treatment. No significant differences in the efficiency of the procedure were detected between thawed and directly treated cells for all three cell types. One million cells were used for four 5 3 5 mm2 silicon chips. Silicon chips were pretreated with poly-lysine and washed in H2O as in the anchored chromatin assembly assay. Cells were gently resuspended in 1 ml hypotonic buffer (15 mM Tris [pH 7.4], 10 mM NaCl, 3 mM MgCl2), incubated on ice for 5–20 min and then passed twice through a syringe connected to a 21 gauge needle. This procedure was repeated a second time in the same buffer supplemented with 10% glycerol and the cells were spun down and resuspended in PBS+10% glycerol. For each cell type, the length of incubation and the exact handling procedure needed to be optimized empirically. Thus, one short incubation (5 min) and gentle handling were sufficient to release COS-1 cell nuclei, whereas HeLa cell nuclei required two separate incubations of 20 min and more aggressive handling. From the analysis of a large number of fields such as the one shown in Figure 3A, the percentage of cells that produce exposed nuclei was estimated to range between 3% (in HeLa cells) and 20% (in COS-1 cells). In each case, there is a fine balance between the number of nuclei that are released or exposed and the degree of damage to the nuclear envelopes. Additional considerations relate to specimen charging and chromium coating (see below) and therefore the finer details of the procedure need to be adjusted for each cell type. Successful exposure and preservation of the NEs could only be determined by viewing the samples at high magnification by FESEM. We were unable to modify this procedure and omit the trypsinization step, in order to apply it to adherent cells. All the centrifugations were for 3 min at 800 3 g, except the last stage of centrifugation onto silicon chips, which was for 10 min at 800 3 g. After cells and nuclei were spun down onto silicon chips they were fixed in 3% glutaraldehyde (Electron Microscopy Sciences) in PBS and further processed for FESEM analysis. Immunogold Labeling Primary antibodies were affinity purified polyclonal anti-xNup62 (Meier et al., 1995) and anti-xNup107 (Rotem et al., 2009). Before attempting to perform immunogold labeling for FESEM on anchored chromatin or nuclei, equivalent anchored templates were analyzed by indirect immunofluorescence on coverslips as in (Rotem et al., 2009). An approximate dilution range for each primary antibody was determined in these initial trials and then tested with secondary 18 nm gold-conjugated goat anti-rabbit (Jackson ImmunoResearch) by FESEM analysis. All incubations were performed at room temperature in a humidified chamber. Antibodies were diluted in 0.5% BSA in PBS. Washes were performed in 24-well plates. Background staining levels of the secondary gold-conjugated antibodies were determined in identical reactions in which the primary antibody was omitted. These background levels ranged between 1%–3% of the number of gold particles on equivalent chromatin or NE surface areas. Silicon chips were postfixed in 3.7% formaldehyde, 0.2%

Structure 20, 407–413, March 7, 2012 ª2012 Elsevier Ltd All rights reserved 411

Structure Scanning Electron Microscopy of Nuclear Pores

glutaraldehyde in buffer E (80 mM PIPES [pH 6.8], 150 mM sucrose, 1 mM MgCl2) for 30 min at room temperature.

Fichtman, B., Ramos, C., Rasala, B., Harel, A., and Forbes, D.J. (2010). Inner/Outer nuclear membrane fusion in nuclear pore assembly: biochemical demonstration and molecular analysis. Mol. Biol. Cell 21, 4197–4211.

Field Emission Scanning Electron Microscopy Processing of fixed samples on silicon chips was performed essentially as described (Allen et al., 2007b; Rotem et al., 2009) including postfixation in 1% osmium tetroxide and dehydration through a graded ethanol series. Critical-point drying was performed from high-purity liquid CO2 on a CPD030 apparatus (Bal-Tec). Samples were sputter-coated with 1–3 nm of chromium on an EMITECH K575X apparatus. COS-1 cell nuclei were particularly susceptible to specimen charging and required a thicker chromium coat to overcome this problem, whereas a thin coat is advantageous for the detection of gold particles. Samples were then viewed on a Zeiss ULTRA plus field emission scanning electron microscope, using an in-lens detector for secondary electrons to provide the surface structure and an EsB detector for backscattered electrons to localize gold particles.

Forbes, D.J., Kirschner, M.W., and Newport, J.W. (1983). Spontaneous formation of nucleus-like structures around bacteriophage DNA microinjected into Xenopus eggs. Cell 34, 13–23.

SUPPLEMENTAL INFORMATION

Goldberg, M.W., and Allen, T.D. (1995). Structural and functional organization of the nuclear envelope. Curr. Opin. Cell Biol. 7, 301–309.

Supplemental Information includes three figures and can be found with this article online at doi:10.1016/j.str.2012.01.022.

Goldberg, M.W., and Allen, T.D. (1996). The nuclear pore complex and lamina: three-dimensional structures and interactions determined by field emission in-lens scanning electron microscopy. J. Mol. Biol. 257, 848–865.

ACKNOWLEDGMENTS

Goldberg, M.W., Wiese, C., Allen, T.D., and Wilson, K.L. (1997). Dimples, pores, star-rings, and thin rings on growing nuclear envelopes: evidence for structural intermediates in nuclear pore complex assembly. J. Cell Sci. 110, 409–420.

The authors thank Eugenia Klein for advice on the generation of stereo pair images and Rita Gruber for help with the assembly diagram. We also thank Ulrike Kutay for the gift of anti-Nup107 antibodies, Lauren Weil and Beatriz Fontoura for the gift of mouse lung fibroblasts, Dana Savulescu for COS-1 cells, and Lena Kliouchnikov for HeLa cells. This work was supported by grants from the Israel Science Foundation (1072/10) and the Russell Berrie Nanotechnology Institute - Technion to A.H. Received: October 2, 2011 Revised: January 29, 2012 Accepted: January 31, 2012 Published: March 6, 2012 REFERENCES Allen, T.D., Rutherford, S.A., Murray, S., Gardiner, F., Kiseleva, E., Goldberg, M.W., and Drummond, S.P. (2007a). Visualization of the nucleus and nuclear envelope in situ by SEM in tissue culture cells. Nat. Protoc. 2, 1180–1184. Allen, T.D., Rutherford, S.A., Murray, S., Sanderson, H.S., Gardiner, F., Kiseleva, E., Goldberg, M.W., and Drummond, S.P. (2007b). Generation of cell-free extracts of Xenopus eggs and demembranated sperm chromatin for the assembly and isolation of in vitro-formed nuclei for Western blotting and scanning electron microscopy (SEM). Nat. Protoc. 2, 1173–1179. Allen, T.D., Rutherford, S.A., Murray, S., Drummond, S.P., Goldberg, M.W., and Kiseleva, E. (2008). Scanning electron microscopy of nuclear structure. Methods Cell Biol. 88, 389–409. Brohawn, S.G., Partridge, J.R., Whittle, J.R., and Schwartz, T.U. (2009). The nuclear pore complex has entered the atomic age. Structure 17, 1156–1168. Burke, B., and Ellenberg, J. (2002). Remodelling the walls of the nucleus. Nat. Rev. Mol. Cell Biol. 3, 487–497. D’Angelo, M.A., Anderson, D.J., Richard, E., and Hetzer, M.W. (2006). Nuclear pores form de novo from both sides of the nuclear envelope. Science 312, 440–443. Drummond, S.P., Rutherford, S.A., Sanderson, H.S., and Allen, T.D. (2006). High resolution analysis of mammalian nuclear structure throughout the cell cycle: implications for nuclear pore complex assembly during interphase and mitosis. Can. J. Physiol. Pharmacol. 84, 423–430. Fahrenkrog, B., Ko¨ser, J., and Aebi, U. (2004). The nuclear pore complex: a jack of all trades? Trends Biochem. Sci. 29, 175–182. Fernandez-Martinez, J., and Rout, M.P. (2009). Nuclear pore complex biogenesis. Curr. Opin. Cell Biol. 21, 603–612.

Gerace, L., and Burke, B. (1988). Functional organization of the nuclear envelope. Annu. Rev. Cell Biol. 4, 335–374. Goldberg, M.W., and Allen, T.D. (1992). High resolution scanning electron microscopy of the nuclear envelope: demonstration of a new, regular, fibrous lattice attached to the baskets of the nucleoplasmic face of the nuclear pores. J. Cell Biol. 119, 1429–1440. Goldberg, M.W., and Allen, T.D. (1993). The nuclear pore complex: threedimensional surface structure revealed by field emission, in-lens scanning electron microscopy, with underlying structure uncovered by proteolysis. J. Cell Sci. 106, 261–274.

Gu¨ttinger, S., Laurell, E., and Kutay, U. (2009). Orchestrating nuclear envelope disassembly and reassembly during mitosis. Nat. Rev. Mol. Cell Biol. 10, 178–191. Harel, A., and Gruenbaum, Y. (2011). Nuclear pore structure: warming up the core. Cell 146, 191–193. Harel, A., Chan, R.C., Lachish-Zalait, A., Zimmerman, E., Elbaum, M., and Forbes, D.J. (2003a). Importin beta negatively regulates nuclear membrane fusion and nuclear pore complex assembly. Mol. Biol. Cell 14, 4387–4396. Harel, A., Orjalo, A.V., Vincent, T., Lachish-Zalait, A., Vasu, S., Shah, S., Zimmerman, E., Elbaum, M., and Forbes, D.J. (2003b). Removal of a single pore subcomplex results in vertebrate nuclei devoid of nuclear pores. Mol. Cell 11, 853–864. Hetzer, M.W., and Wente, S.R. (2009). Border control at the nucleus: biogenesis and organization of the nuclear membrane and pore complexes. Dev. Cell 17, 606–616. Hetzer, M.W., Walther, T.C., and Mattaj, I.W. (2005). Pushing the envelope: structure, function, and dynamics of the nuclear periphery. Annu. Rev. Cell Dev. Biol. 21, 347–380. Hoelz, A., Debler, E.W., and Blobel, G. (2011). The structure of the nuclear pore complex. Annu. Rev. Biochem. 80, 613–643. Kiseleva, E., Rutherford, S., Cotter, L.M., Allen, T.D., and Goldberg, M.W. (2001). Steps of nuclear pore complex disassembly and reassembly during mitosis in early Drosophila embryos. J. Cell Sci. 114, 3607–3618. Lim, R.Y., Aebi, U., and Fahrenkrog, B. (2008). Towards reconciling structure and function in the nuclear pore complex. Histochem. Cell Biol. 129, 105–116. Lohka, M.J., and Masui, Y. (1983). Formation in vitro of sperm pronuclei and mitotic chromosomes induced by amphibian ooplasmic components. Science 220, 719–721. Macaulay, C., and Forbes, D.J. (1996). Assembly of the nuclear pore: biochemically distinct steps revealed with NEM, GTP gamma S, and BAPTA. J. Cell Biol. 132, 5–20. Maimon, T., and Medalia, O. (2010). Perspective on the metazoan nuclear pore complex. Nucleus 1, 383–386. Maul, G.G., Maul, H.M., Scogna, J.E., Lieberman, M.W., Stein, G.S., Hsu, B.Y., and Borun, T.W. (1972). Time sequence of nuclear pore formation in phytohemagglutinin-stimulated lymphocytes and in HeLa cells during the cell cycle. J. Cell Biol. 55, 433–447.

412 Structure 20, 407–413, March 7, 2012 ª2012 Elsevier Ltd All rights reserved

Structure Scanning Electron Microscopy of Nuclear Pores

Meier, E., Miller, B.R., and Forbes, D.J. (1995). Nuclear pore complex assembly studied with a biochemical assay for annulate lamellae formation. J. Cell Biol. 129, 1459–1472.

Shaulov, L., Gruber, R., Cohen, I., and Harel, A. (2011). A dominant-negative form of POM121 binds chromatin and disrupts the two separate modes of nuclear pore assembly. J. Cell Sci. 124, 3822–3834.

Onischenko, E., and Weis, K. (2011). Nuclear pore complex-a coat specifically tailored for the nuclear envelope. Curr. Opin. Cell Biol. 23, 293–301.

Vasu, S.K., and Forbes, D.J. (2001). Nuclear pores and nuclear assembly. Curr. Opin. Cell Biol. 13, 363–375.

Rasala, B.A., Ramos, C., Harel, A., and Forbes, D.J. (2008). Capture of AT-rich chromatin by ELYS recruits POM121 and NDC1 to initiate nuclear pore assembly. Mol. Biol. Cell 19, 3982–3996.

Vetter, I.R. (2009). Nup120: one more piece in the NPC puzzle. Structure 17, 1036–1038.

Rotem, A., Gruber, R., Shorer, H., Shaulov, L., Klein, E., and Harel, A. (2009). Importin beta regulates the seeding of chromatin with initiation sites for nuclear pore assembly. Mol. Biol. Cell 20, 4031–4042. Schwartz, T.U. (2005). Modularity within the architecture of the nuclear pore complex. Curr. Opin. Struct. Biol. 15, 221–226. Schwarz-Herion, K., Maco, B., Sauder, U., and Fahrenkrog, B. (2007). Domain topology of the p62 complex within the 3-D architecture of the nuclear pore complex. J. Mol. Biol. 370, 796–806.

Walther, T.C., Alves, A., Pickersgill, H., Loı¨odice, I., Hetzer, M., Galy, V., Hu¨lsmann, B.B., Ko¨cher, T., Wilm, M., Allen, T., et al. (2003). The conserved Nup107-160 complex is critical for nuclear pore complex assembly. Cell 113, 195–206. Wiese, C., Goldberg, M.W., Allen, T.D., and Wilson, K.L. (1997). Nuclear envelope assembly in Xenopus extracts visualized by scanning EM reveals a transport-dependent ‘envelope smoothing’ event. J. Cell Sci. 110, 1489– 1502.

Structure 20, 407–413, March 7, 2012 ª2012 Elsevier Ltd All rights reserved 413