Nuclear Transport of Baculovirus: Revealing the Nuclear Pore Complex Passage ⇑ Shelly Au, Nelly Panté

Journal of Structural Biology xxx (2011) xxx–xxx

Contents lists available at SciVerse ScienceDirect

Journal of Structural Biology

journal homepage: www.elsevier.com/locate/yjsbi

Nuclear transport of baculovirus: Revealing the nuclear pore complex passage ⇑ Shelly Au, Nelly Panté

Department of Zoology, University of British Columbia, 6270 University Boulevard, Vancouver, British Columbia, Canada V6T 1Z4 article info abstract

Article history: Baculoviruses are one of the largest viruses that replicate in the nucleus of their host cells. During an Available online xxxx infection the capsid, containing the DNA viral genome, is released into the cytoplasm and delivers the genome into the nucleus by a mechanism that is largely unknown. Here, we used capsids of the baculo- Keywords: virus Autographa californica multiple nucleopolyhedrovirus in combination with electron microscopy and Electron microscopy discovered this capsid crosses the NPC and enters into the nucleus intact, where it releases its genome. To Electron tomography better illustrate the existence of this capsid through the NPC in its native conformation, we reconstructed Nuclear import the nuclear import event using electron tomography. In addition, using different experimental condi- Nuclear pore complex tions, we were able to visualize the intact capsid interacting with NPC cytoplasmic ﬁlaments, as an initial Baculovirus capsid docking site, and midway through the NPC. Our data suggests the NPC central channel undergoes large- scale rearrangements to allow translocation of the intact 250-nm long baculovirus capsid. We discuss our results in the light of the hypothetical models of NPC function. Ó 2011 Elsevier Inc. All rights reserved.

1. Introduction led to progressive improvements of the three-dimensional (3D) models of the NPC, the latest with a resolution of 6 nm (Beck Cellular compartmentalization is vital, in order for the cell to et al., 2007; Frenkiel-Krispin et al., 2010). According to these studies, function efficiently. The nucleus, being a membrane-enclosed orga- the NPC consists of a membrane-embedded scaffold ring and nelle, contributes significantly to the regulation of numerous cellu- peripheral components that extend into the cytoplasm (eight lar processes. For example, by controlling the access of certain cytoplasmic filaments) and nucleus (the nuclear basket). The macromolecules to the nucleus, nuclear transport can regulate tran- NE-embedded ring has a diameter of 125 nm, is 70 nm in height, scription, DNA replication, and the cell cycle. The flow of proteins, and contains a large central channel of about 50 nm in diameter RNAs, and RNA–protein complexes (such as ribosomal subunits, (Beck et al., 2004, 2007; Frenkiel-Krispin et al., 2010; Stoffler et al., messenger ribonucleoproteins (RNPs), and splicesomal RNPs) into 1999). This central channel acts as a molecular sieve, allowing pas- and out of the nucleus at a rate of up to 1500 molecules per second sive diffusion of ions and molecules smaller than 9 nm in diameter is achieved by nuclear pore complexes (NPCs) embedded within the (or proteins smaller than 40 kDa) and selective facilitated transport nuclear envelope (NE) (Kowalczyk et al., 2011; Ribbeck and Gorlich, of larger cargos up to 39 nm in diameter (Pante and Kann, 2002). The 2001; Strambio-De-Castillia et al., 2010; Wente and Rout, 2010). selectivity is dictated by a signal (stretches of amino acids called Extensive electron microscopy (EM) studies have been carried out nuclear localization signals (NLSs) or nuclear export signals) resid- to elucidate the structure of the NPC, including studies by Ueli Aebi ing on the transported molecule, which is recognized by nuclear and colleagues using several EM techniques and atomic force transport receptors (NTRs) derived from a family of proteins known microscopy (Reviewed in Elad et al., 2009; Lim et al., 2008a,b; Maco as karyopherins or importins (Chook and Suel, 2011; Strambio- et al., 2006; Pante, 2007; Rowat et al., 2008). In particular, using De-Castillia et al., 2010; Terry and Wente, 2009; Wente and Rout, state-of-the-art cryo-electron tomography Aebi and others have 2010). The main components of the NPC are proteins called nucleoporins (Nups). Multiple copies of about 30 different Nups make up Abbreviations: AcMNPV, Autographa californica multiple nucleopolyhedrovirus; EM, electron microscopy; FG, phenylalanine–glycine; GV, granulovirus; LSB, low- the 120-mDa metazoan NPC (Cronshaw et al., 2002). Nups have salt buffer; MBS, modified Barth’s saline; NE, nuclear envelope; NLS, nuclear been classified into three families: integral membrane proteins, localization signal; NPCs, nuclear pore complexes; NPV, nucleopolyhedrovirus; which anchor the NPC in the NE; scaffolding Nups, which form NTRs, nuclear transport receptors; Nups, nucleoporins; RNP, ribonucleoprotein; the NPC scaffold structure; and FG-Nups, which are rich in hydro- TEM, transmission electron microscope; 3D, three-dimensional; WGA, wheat germ agglutinin. phobic phenylalanine–glycine (FG) repeat motifs (Brohawn et al., ⇑ Corresponding author. Fax: +1 604 822 2416. 2009; Tetenbaum-Novatt and Rout, 2010; Walde and Kehlenbach, E-mail address: [email protected] (N. Panté). 2010). The FG-repeat regions of these Nups presumably populate

Please cite this article in press as: Au, S., Panté, N. Nuclear transport of baculovirus: Revealing the nuclear pore complex passage. J. Struct. Biol. (2011), doi:10.1016/j.jsb.2011.11.006 2 S. Au, N. Panté / Journal of Structural Biology xxx (2011) xxx–xxx the central channel of the NPC and form the transport barrier of the (Agri-Food Research Centre, Summerland, B.C., Canada). Baculovi- NPC. Additionally, FG-Nups have a role in the initial docking of the ruses were purified through a continuous 15–60% (w/v) sucrose cargo to the NPC, as interactions of NTRs with peripheral FG-Nups gradient centrifugation, as described by Shoji et al. (1997) and at the NPC cytoplasmic filaments dock the cargo to the NPC (Grun- Obregon-Barboza et al. (2007) with the modifications described in wald and Singer, 2010; Hamada et al., 2011; Lowe et al., 2010; Ya- Au et al. (2010). To remove the viral envelope, purified virus was seen and Blobel, 1999). treated with 1% NP40 for 1 h at 30 °C. After this incubation, the sam- Despite considerable progress in elucidating the structure and ple was washed by centrifugation to remove excess NP40, and dia- composition of the NPC, in identifying nuclear transport signals lyzed for 24 h as described in Au et al. (2010). The successful and their receptors, and in characterizing the interaction of removal of the viral envelope and the integrity of the capsid were FG-Nups and NTRs, the molecular mechanism by which molecules evaluated by examining the capsid preparation under the EM stain- translocate through the NPC central channel remains unclear. Sev- ing with uranyl acetate, as described in Au et al. (2010). eral models have attempted to explain how interaction between Stage VI oocytes were surgically removed from narcotized X. FG-Nups and NTRs allows for efficient and selective translocation laevis, as described by Pante (2006) and Au et al. (2010). Isolated of the cargo through the NPC. Of these models, the Brownian Affin- oocytes were washed with modified Barth’s saline buffer (MBS: ity Gate/Virtual Gating model (Lim et al., 2006; Rout et al., 2000, 88 mM NaCl, 1 mM KCl, 0.82 mM MgSO4, 0.33 mM Ca(NO3)2, 2003) and the Selective Phase model (Frey and Gorlich, 2007; Frey 0.41 mM CaCl2, 10 mM HEPES, pH 7.5) and treated with collage- et al., 2006; Ribbeck and Gorlich, 2002) are the best described; they nase (5 mg/ml) in calcium-free MBS for 1 h to defolliculate the are also the most debated. A common rationale between these two oocytes. models is that FG-containing Nups form a barrier for nuclear import, and transport occurs via association and disassociation of 2.2. Oocyte microinjection NTRs with the FG repeats. Further studies in this area resulted in alternative models such as the Reduction of Dimensionality model Injection needles were made by heating and pulling micropi- (Peters, 2005) and the more recent Forest model (Yamada et al., pettes (Microcaps; Drummond) using a micropipette puller 2010). (Inject-Matic). Microinjection of Xenopus oocytes was performed While biochemical and biophysical measurements and compu- as described in Au et al. (2010) using an oocyte microinjector tational simulations have been developed to support each model, (Inject-Matic). Oocytes were injected with 100 nl of purified capsids these models remain controversial. Direct visualization of large into the cytoplasm at the white rim separating the animal and veg- cargos crossing the NPC central channel may be a good strategy etal hemispheres. As controls, oocytes were mock injected with to test these hypothetical models. In addition, such studies will fur- 100 nl of TE (10 mM Tris, 1 mM EDTA, pH 8.7). ther substantiate the ability of NPCs in setting boundaries for cargo For time-course experiments, oocytes were microinjected with selection, and will provide a clear depiction of the overall structure capsids, incubated at room temperature in MBS for 2, 4, or 8 h, and dimensions of the NPC central channel when occupied by a and prepared for EM as described below. For some experiments, in- large cargo in transit. Because viral capsids are among the largest jected oocytes were incubated at 4 °C instead of room temperature, cargos that enter the nucleus, studies on the nuclear import of viral in order to trap the capsid at intermediate stages of nuclear import. capsids may provide important information that can be used to A biochemical inhibitor of nuclear transport was also used to test the several proposed models for NPC function. At the same depict transport-arrested capsid at the NPC. Oocytes were microin- time, these studies will enhance our knowledge about the replica- jected with the lectin wheat germ agglutinin (WGA) conjugated tion cycle of viruses. with colloidal gold (WGA-gold). Preparation of colloidal gold and In the past, viruses have been invaluable tools for important conjugation of WGA with the gold particles were performed as discoveries in the field of nuclear transport. For example, the NLS described in Pante (2006). As such, 50 nl of WGA-gold was micro- was first discovered in the large T antigen of the simian virus 40 injected into the cytoplasm or 20 nl of WGA-gold was microin- (Kalderon et al., 1984), and the upper size limitation for molecules jected into the nucleus of oocytes, and the oocytes were to transport through NPC was determined using hepatitis B virus incubated at room temperature for 2 h. After the incubation period, capsids with diameters of 36 nm (Pante and Kann, 2002). In this oocytes were microinjected into their cytoplasm with 100 nl of study, we used capsids of the insect virus baculovirus Autographa capsids and further incubated at room temperature for 8 h. californica multiple nucleopolyhedrovirus (AcMNPV) to demonstrate the flexibility of NPCs in nuclear import, and to further dis- 2.3. Preparation of injected oocytes for EM sect the life cycle of this virus. Baculoviruses are large, rod-shaped (30–60 Â 250–300 nm), After microinjection and incubation of oocytes for the indicated enveloped viruses with a DNA genome that requires the nuclear time, oocytes were prepared for embedding and thin-section EM machinery for viral replication (Blissard and Rohrmann, 1990). following detailed protocols in Au et al. (2010). Briefly, oocytes Therefore, the baculovirus genome must enter the host nucleus, were fixed with 2% glutaraldehyde in MBS overnight at 4 °C. Fixed but how this is accomplished remains largely unknown. Using oocytes were then washed in MBS and the animal hemispheres the Xenopus laevis oocyte cell system, which is transcriptionally (which contains the nucleus) were dissected and fixed with 2% glu- inactive, in combination with EM and electron tomography, we taraldehyde in low-salt buffer (LSB: 1 mM KCl, 0.5 mM MgCl2, conclude that, prior to disassembly, the baculovirus AcMNPV 10 mM HEPES, pH 7.5) for 1 h at room temperature. After this fix- capsid vertically interacts with cytoplasmic components of NPCs ation, oocytes were washed with LSB, embedded in 2% low-melting to gradually enter the nucleus fully intact. agarose and post-fixed with 1% osmium tetroxide in LSB for 1 h. Fixed oocytes were sequentially dehydrated in increasing concen- 2. Materials and methods trations of ethanol and embedded in Epon 812 (Fluka) as described by Au et al. (2010). 2.1. Virus and oocytes Thin 50 nm sections through the NE from Epon-embedded oocytes were cut on a Leica Ultracut Ultramicrotome (Leica Micro- Recombinant AcMNPV, propagated in Escherichia coli strain systems) using a diamond knife (Diatome). Sections were collected DH108 and amplified at a multiplicity of infection of 1 in Spodoptera on parlodion/carbon-coated copper EM grids, and stained with 2% frugiperda Sf9 cells were kindly provided by Dr. D. Theilmann uranyl acetate for 15 min and 2% lead citrate for 5 min.

2.4. Electron microscopy and tomography diameter of 30 nm and variable length of 250–300 nm (Fig. 1A). In this preparation, the virus was completely devoid of the envelope Samples were examined using a Hitachi-7600 transmission and the capsids appeared electron dense, which is an indication electron microscope (TEM) operated at an acceleration voltage of that they contain the viral genetic material. These capsids had 80 kV. Micrographs were digitally recorded using a 1-megapixel the expected morphology with two distinct ends: one end blunt AMT Advantage CCD camera (ORCA; Hamamatsu Photonics). and one end conical with a small protuberance (arrows in Fig. 1B). For electron tomography, 200 nm thick sections through the NE We then microinjected these capsids into the cytoplasm of were made, and sections were placed on slot grids coated with 1% Xenopus oocytes and followed their fate by EM after the injected formvar. Single axis tilt series of 200 nm samples were recorded oocytes were embedded in Epon and thin sectioned. To allow automatically over tilt angles ranging from À30° to +30° in two de- observation of the large capsids at the NPCs, oocytes were incu- gree increments, and then from À70° to +70° every degree on a FEI bated at room temperature at different times. Our time-course Tecnai G2 F20 TEM operated at 200 kV. Tomograms were acquired experiment showed that at 2 h post-microinjection, about a quar- using FEI’s TEM tomography software and reconstructed using ter of the capsids were in the cytoplasm, away from the NE (Fig. 2A FEI’s Inspect3D software. and E), while the remaining capsids were already at the NPC (Fig. 2B and E). After 4 h, however, almost all of the capsids were seen docked at the NPC or very close to the NPC (within a distance 3. Results of 100 nm from the NPC; Fig. 2C and F). In the micrographs show- ing capsids at the NPC, the capsids were interacting vertically with 3.1. Purified capsids are imported into Xenopus oocyte nuclei the NPC, and in most of the micrographs we were able to distin- guish the conical end of the capsid at the NPC and the blunt end In order to visualize the nuclear import of baculovirus capsids away from the NPC. Some capsids were also found inside the (referred to as capsids henceforth) by EM, we purified the capsids nucleus after 4 h post-microinjection (Fig. 2D); however, capsids from Sf9 cells infected with baculovirus and microinjected these devoid of DNA (empty capsids, which are not electron dense) or into the cytoplasm of Xenopus oocytes (a system well suited for disassembled capsids were not observed. Eight hours post-micro- the study of nuclear import because it enables us to visualize injection was a sufficient amount of time for most capsids to enter well-preserved NPCs and NPC-arrested cargos). Before microinjec- the nucleus, as both the cytoplasm and nucleus were capsid-free tion, the purified capsids were examined by EM after staining with (data not shown). This also suggests that by 8 h, capsid disassem- uranyl acetate to evaluate the purity and integrity of the capsids. bly has occurred and the DNA genome has been released into the As documented in Fig. 1, the purification of the capsid by our pro- nucleus. Unfortunately, we were unable to detect the order in tocol was effective, yielding typical rod-shaped capsids that had a which these events occurred.

3.2. Capsids remain intact while vertically traversing the NPC

Electron micrographs from oocytes that were incubated for 3.5 h post-microinjection at room temperature showed the capsid vertically traversing the NPC and some of them midway through the NPC (Fig. 3). Remarkably, the NPCs containing capsids in transit appeared less electron-dense than neighbouring NPCs not engaged in capsid nuclear transport. In particular, an area of 8–10 nm in diameter surrounding the capsid appeared completely empty, as if all the material normally ﬁlling the NPC central channel had retracted, presumably to allow movement of the capsid across the NPC. In order to compensate for the fact that thin sections of 50 nm through the NE may not be representative of the NPC as a whole, we obtained thicker sections of 200 nm to generate tomograms of capsids in the midst of being imported into the nucleus through the NPCs. EM tomograms showed that both the capsid and NPC remain intact during this translocation event (Fig. 4 and Supple- mentary video 1). Similarly, an unﬁlled area around the capsid whilst traveling through the central channel can be seen in Fig. 4. Tomographic reconstruction of this event better illustrated an intact capsid traversing the NPC (Supplementary video 2). Via the 3D view, we observed intact capsids in transit through the NPCs, documenting that, in fact, the intact capsid crosses the NPC. Likewise, the NPC was seen to wrap around the capsid, further demonstrating the movement of the capsid within the central channel of the NPC.

3.3. Targeting of capsids to the NPC is delayed at low temperature

Biochemical and physiological inhibitors of nuclear transport Fig.1. Electron micrographs of purified baculovirus AcMNPV capsids stained with have been used extensively to arrest imported molecules at inter- uranyl acetate. The micrograph in (A) documents that the purified capsids were mediate stages of its passage into the nucleus. Accumulation of variable in length. The micrographs in (B) document the morphology of the capsid with its two distinct ends; a blunt end and a conical end with a small protuberance cargos at the NPC cytoplasmic filaments and at the cytoplasmic (arrows). Scale bars, 200 nm in (A) and 50 nm in (B). entrance of the NPC central channel have been observed by EM

Fig.2. Electron micrographs of Xenopus laevis oocytes that have been microinjected with baculovirus AcMNPV capsids and incubated at room temperature for two (A and B) or 4 h (C and D). Bar graphs (E and F) show the percentage of capsids found associated with the NPC, 100 nm away from the NPC, and in the cytoplasm from experiments performed as indicated above. A total of 150 capsids were scored for each condition from three different experiments. Capsids were found in the cytoplasm (A), at the NPC or at 100 nm from the NPC by 2 h post-microinjection (B). Most capsids were docked at the NPC (C) by 4 h post-microinjection, and some were inside the nucleus (D) by this time. Scale bar, 200 nm. n, nucleus; c, cytoplasm. Arrows point to capsids.

Fig.3. Electron micrographs of NPC cross-sections from Xenopus oocytes that have been microinjected with baculovirus AcMNPV capsid and incubated at room temperature for 3.5 h. Capsids of 250–300 nm in length are seen traversing the NPCs. Capsids appear fully intact in its native conformation while crossing the NPC. Note the capsid in the middle panel appear shorter due to the variability in the length of these capsids, as documented in Fig. 1. Scale bar, 100 nm. n, nucleus; c, cytoplasm. when nuclear import is inhibited at 4 °C(Pante, 2007; Pante and the NPC but does not inhibit its initial docking at the NPC. It is sur- Aebi, 1996; Rollenhagen and Pante, 2006; Rollenhagen et al., mised that this condition yielded an increased amount of capsids at 2003). Low temperature inhibits the translocation of cargo through the cytoplasmic face of the NPC; however, we found that in oocytes

Fig.4. Tomographic x-y slices spaced approximately 20 nm through the 3D volume of the tomographic tilt series of a capsid in the midst of being imported into the nucleus through the NPCs. Scale bar, 200 nm. n, nucleus; c, cytoplasm. Arrow points to the capsids. incubated for 2 h at 4 °C post-microinjection, 88% of the capsids glucosamine residues on glycosylated Nups, thereby blocking the remained in the cytoplasm far from the nucleus (Fig. 5A and C), interaction between NTRs and Nups and inhibiting nuclear trans- in contrast to the 23% observed when oocytes were incubated for port (Finlay et al., 1987; Newmeyer and Forbes, 1988). For these 2 h at room temperature (Fig. 2A and E). Similarly, significantly experiments, we conjugated 10 nm gold particles to WGA (to visu- more capsids were found docked at the NPC after 4 h incubation alize the binding of WGA to the NPC) and the WGA-gold complexes at room temperature (Fig. 2D and F) than after 4 h incubation at were pre-microinjected into the cytoplasm of the oocytes. After 2 h 4 °C(Fig. 5B and D). The delayed progress of capsids transiting of incubation at room temperature, the oocytes were again micro- towards the nucleus suggests that metabolic energy is the driving injected into their cytoplasm with capsids and incubated for 8 h at force that allows the capsid to move within the cytoplasm. Our room temperature. After 8 h, WGA-gold particles had accumulated data is in agreement with the previous findings that actin-based at the entrance of the NPC central channel, blocking the transloca- motility drives the capsid towards the nucleus (Charlton and tion of the capsid through the NPC, and the capsids remained at the Volkman, 1993; Ohkawa and Volkman, 1999; Ohkawa et al., 2010). NPC cytoplasmic face at a distance of about 100 nm from the centre of the NPC (Fig. 6A). We also attempted to block the NPC from its nuclear side by 3.4. Initial docking of the capsid occurs at the cytoplasmic filaments of pre-microinjecting WGA-gold into the nucleus of the oocytes. Sim- NPCs ilar to the cytoplasmic injection, nuclear injected oocytes were incubated for 2 h, post-microinjected with capsids, and further The low temperature experiments also demonstrated that the incubated for 8 h at room temperature. Under these conditions, NPC cytoplasmic filaments act as the first binding sites for the cap- the WGA-gold particles were within the NPC central channel, while sid prior to capsid translocation through the NPC. Oocytes that the capsid remained on the cytoplasmic face of the NPC (Fig. 6B). were incubated for 4 h at 4 °C yielded 87% of capsid at the cytoplas- Furthermore, at 8 h post-microinjection we observed capsids in mic face of the NPC at about 100 nm from the center of the NPC, the cytoplasm when the NPCs were blocked by WGA, an occur- and in most of the micrographs the cytoplasmic filaments were rence that was not observed when NPCs were uninhibited. Our clearly depicted (Fig. 5B). In most of the micrographs, we were also data demonstrates that the NPC cytoplasmic filaments are the able to visualize the conical end of the capsid at the NPC, and the initial docking sites for the capsids. blunt end of the capsid away from the NPC (see for example, Fig. 5B left panel). To confirm our results of the initial binding of the capsid to the 4. Discussion NPC cytoplasmic filaments under conditions that do not delay the targeting of the capsid to the NPC, we used WGA, a well character- Despite recent advances in microscopic techniques to help cor- ized inhibitor of nuclear import that binds O-linked N-acetyl relate the structure of the NPC with its function, the molecular

Fig.5. Electron micrograph of Xenopus laevis oocytes that have been microinjected with baculovirus AcMNPV capsids and incubated at 4 °C for 2 h (A) or 4 h (B). Bar graphs (C and D) show the percentage of capsids found associated with the NPC, 100 nm away from the NPC, and in the cytoplasm from experiments performed as indicated above. A total of 150 capsids were scored for each condition from three different experiments. Most capsids were found in the cytoplasm at 2 h post-microinjection (A), while the majority of capsids were docked at the NPC at 4 h post-microinjection (B). No capsids were found inside the nucleus under these conditions. Scale bar, 200 nm. n, nucleus; c, cytoplasm. Arrows point to capsids.

Fig.6. Electron micrograph of Xenopus oocytes that were microinjected with WGA-gold into either the cytoplasm (A) or nucleus (B), incubated at room temperature for 2 h, followed by cytoplasmic injection of baculovirus AcMNPV capsids and further incubated for 8 h. When NPCs were inhibited by WGA-gold particles, capsids remained interacting with the NPC cytoplasmic ﬁlaments 8 h post-microinjection. No capsids were found inside the nucleus when NPCs were inhibited with WGA. Scale bar, 200 nm. n, nucleus; c, cytoplasm. Black arrows point to capsids and white arrowheads point to WGA-gold particles at NPCs.

mechanism of translocation through the NPC remains elusive. using EM yielded contradictory results. The ﬁrst study reported Although several models have been proposed in recent years to capsids from the baculovirus genus granulovirus (GV) docking at explain this mechanism, due to the lack of in vivo experimental set- the NPC of infected cells, but not inside the nucleus (Summers, ups to test these models, they remain controversial and are a major 1971). This suggested a mechanism of DNA nuclear import similar topic of debate. One feasible strategy to test these models focuses to the herpes simplex virus-1, which attaches to the cytoplasmic side on the visualization of large cargos crossing the NPC. We chose to of the NPC and ejects its nucleic acid into the nucleus through the study the nuclear import of baculovirus to provide a more concise NPC, leaving empty capsids at the NPC (Sodeik et al., 1997). Other understanding of both how the NPC functions and the strategy studies using the baculovirus genus nucleopolyhedrovirus (NPV) used by this virus to enter the nucleus. inoculated into larvae observed capsids in both the cytoplasm and With a diameter of 30 nm, the baculovirus capsid is small enough the nucleus (Granados, 1978; Granados and Lawler, 1981; Raghow to cross the NPC without apparent deformation; however, direct and Grace, 1974); however, it was unclear whether the observed demonstration of the actual translocation of the capsid through capsids entered the nucleus during mitosis when the nuclear mem- the NPC has not been reported. Previous attempts to visualize this brane was absent, or via NPCs. Subsequent studies following the

Please cite this article in press as: Au, S., Panté, N. Nuclear transport of baculovirus: Revealing the nuclear pore complex passage. J. Struct. Biol. (2011), doi:10.1016/j.jsb.2011.11.006 S. Au, N. Panté / Journal of Structural Biology xxx (2011) xxx–xxx 7 infection of insect cells in culture reported capsids in both the cyto- finding is in contrast to studies using 14 nm nucleoplasmin-conju- plasm and the nucleus of the infected cells as early as 2 h post-infec- gated gold (Pante and Aebi, 1996), in which gold particles were tion (Carsten et al., 1979), indicating that the capsid crosses the NPC documented to locate at the central channel of the NPC when intact. More recently, experiments with tissue-cultured cells, ar- microinjected into Xenopus oocytes at 4 °C. This difference may rested at G1/S phase revealed capsids on the cytoplasmic side of be due to the size of the cargo, and supports the idea that different the NPC as well as inside the nucleus (van Loo et al., 2001), thus sup- mechanisms of nuclear entry exist and size of the cargo could be porting a mechanism in which the entire baculovirus capsid crosses the main determinant. Consistent with this, splicesomal RNPs con- the NPC and genome release occurs in the nucleus. These contradic- jugated with gold particles were observed at the NPC cytoplasmic tory findings suggest that perhaps the mechanism of nuclear import filaments (like the baculovirus capsid) and not at the NPC central of the baculovirus capsid could be genus specific. In all these studies channel (Rollenhagen and Pante, 2006). This spatial difference reporting capsids inside the nucleus, however, it was unclear could also be the result of a large cargo, such as the viral capsid whether the capsids observed in the nucleus were imported through or the splicesomal RNPs with discreet sites for NTR binding, as the NPC or were made during the course of infection. To decipher this opposed to a single gold particle containing numerous copies of difference and to depict the mode of nuclear entry employed by the the same small protein and multiple sites for NTR binding. baculovirus capsid, we used a non-replicating and non-dividing cell Using microinjection of baculovirus capsids into Xenopus system: X. laevis oocytes. oocytes, in combination with electron tomography, we also docu- Upon microinjection of baculovirus capsids into the Xenopus mented that the central channel of the NPC is able to accommodate oocyte cytoplasm and analysis of the oocytes by EM, we observed a very long (250–300 nm) cargo, which can remain fully intact in its capsids docking at the cytoplasmic side of the NPC in what appears native conformation while traversing the NPC lengthwise. A similar to be NPC cytoplasmic filaments (Figs. 2B and C and 5B). Capsid inter- situation was observed for the NPC translocation of the Balbiani ring action with the NPC appears to be with the conical end, and not with granule, a premessenger RNP complex of very large size synthesized the blunt end. In addition, we also observed capsids in the nucleus of in the larval salivary glands of Chironomus tentans. These granules the injected oocytes (Fig. 2D). As baculoviruses do not replicate in are 50 nm in diameter and consist of an RNP ribbon bent into a Xenopus oocytes, the capsids found inside the nucleus must have ring-like structure. As the granule is exported from the nucleus been imported from the cytoplasm through the NPC. Consistent with through the NPC, the ribbon (25 nm in diameter by 135 nm in this explanation, we depicted the capsid midway through the NPC length) straightens out and has been shown to occupy the central (Fig. 3), and demonstrated by electron tomography (Fig. 4, Supple- channel of the NPC (Mehlin and Daneholt, 1993; Mehlin et al., mentary videos 1 and 2) that the capsid crosses the NPC intact. Thus, 1992, 1995). Both our data with the baculovirus capsid and the pub- our data supports a nuclear entry mode for baculovirus from the lished results for the Balbiani ring granules clearly illustrate the flex- genus NPV that involves translocation of the intact capsid through ibility of the NPC central channel. The data also demonstrates that the NPC, similar to a DNA virus of similar diameter, the hepatitis B the NPC must undergo a large scale of rearrangement to allow such virus capsid (Pante and Kann, 2002; Rabe et al., 2003). large cargos to occupy the NPC central channel. The endocytic route is often necessary for viruses to become When in transit, the capsid appears to occupy the whole NPC competent for nuclear import. For example the acidic environment central channel (Figs. 3 and 4). We also observed some empty of the endosome could trigger exposure of NLSs. Given that we space within the central channel immediately adjacent to the cap- microinjected baculovirus capsids into the cytoplasm of Xenopus sid, unlike the electron dense central channel of neighbouring NPCs oocytes, we bypassed the endocytic route. Since we still observed that are capsid-free. Considering the width of the capsid (30 nm) nuclear import of the injected capsids, our data suggests that either and our measurements of the empty space surrounding the NPC- the NLSs were already exposed on the surface of the capsid, or crossing capsid (about 10 nm from each side of the capsid), the modifications of the capsid in the cytoplasm facilitated this expo- NPC central channel expanded to about 50 nm to allow the capsid sure. However, the putative NLSs on the capsid and the NTRs rec- to pass through it. This is the same value for the dimension of the ognizing these remain largely unknown. Our data further NPC central channel that was deduced from 3D reconstructions of confirms previous suggestions that endocytic conditions are not cryo-EM of NPCs (Beck et al., 2004, 2007; Frenkiel-Krispin et al., necessary in the exposure of NLSs for nuclear import of baculovirus 2010; Stoffler et al., 2003). Our finding indicates that whatever is capsids (Salminen et al., 2005). More recently, baculovirions were normally filling the NPC central channel must completely retract, shown to be able to enter and infect Sf9 insect cells through a leaving the NPC in an open state that allows the capsid to travel non-endocytic pathway, further illustrating the idea that putative across it. This idea of an open state of the NPC was originally pro- NLSs are within the capsid itself (Dong et al., 2010). posed in earlier studies that suggested the concept of a central The delayed progress of capsids transiting through the cytoplasm plug/transporter that resides within the NPC central channel (Akey, towards NPCs when oocytes were incubated at 4 °C suggests that ac- 1990, 1991; Akey and Radermacher, 1993). These studies hypoth- tive transport within the cytoplasm was also hindered. esized that the transporter remains in a closed position when car- P78/83 capsid protein of AcMNPV has been shown to associate with gos dock and further dilates when cargos are in transit through the actin-like structures in the cytoplasm (Charlton and Volkman, NPC. The existence of an NPC central plug/transporter was sup- 1993). More recently, it was demonstrated that when the Arp2/3 ported by early structural analysis of the NPC that documented complex binding region in P78/83 was mutated, viral motility within the presence of a massive particle in the central channel. Because the cytoplasm, as well as viral gene expression was delayed (Ohkawa the size, shape and position of this particle were highly variable, et al., 2010). Therefore, incubating oocytes at low temperature could the central plug/transporter was later proposed to be molecules have impeded active transport via actin within the cytoplasm. in transit (Beck et al., 2004; Elad et al., 2009; Stoffler et al., 2003). The NPC cytoplasmic filaments are the initial docking sites for More recent models of NPC function propose that the central several molecules undergoing nuclear import. Using WGA as an channel is filled by the FG-repeat domains of Nups. The Selective inhibitor of nuclear import, by both cytoplasmic and nuclear injec- Phase model assumes that these domains are cross-linked, forming tion of this inhibitor into Xenopus oocytes, we demonstrated that a hydrogel (Ribbeck and Gorlich, 2002). Although recombinant FG capsids remained at the NPC cytoplasmic filaments when WGA- domains can form a hydrogel in vitro (Frey et al., 2006), it would be gold particles impeded transport through the NPC central channel of importance to further demonstrate how such a hydrogel could (Fig. 6). Capsids were also observed at the NPC cytoplasmic fila- reform after being significantly disrupted, leaving an empty space ments when the oocytes were incubated at 4 °C(Fig. 5B). This of 50 nm in diameter by 70 nm in height (the entire dimensions of

Please cite this article in press as: Au, S., Panté, N. Nuclear transport of baculovirus: Revealing the nuclear pore complex passage. J. Struct. Biol. (2011), doi:10.1016/j.jsb.2011.11.006 8 S. Au, N. Panté / Journal of Structural Biology xxx (2011) xxx–xxx the NPC central channel) that allows the passage of the baculovirus References capsid. Nevertheless, this model suggests an inverse relationship between cargo size and the rate of nuclear import (Lieleg and Rib- Akey, C.W., 1990. Visualization of transport-related configurations of the nuclear pore transporter. Biophys. J. 58, 341–355. beck, 2011); thus, in order to test this model it will be valuable to Akey, C.W., 1991. Probing the structure and function of the nuclear pore complex. measure the rate of nuclear import of the baculovirus capsid using Semin. Cell Biol. 2, 167–177. live-cell imaging. Akey, C.W., Radermacher, M., 1993. Architecture of the Xenopus nuclear pore In another proposed model for NPC function, the FG-repeat complex revealed by three-dimensional cryo-electron microscopy. J. Cell Biol. 122, 1–19. domains of Nups have been suggested to act as a polymer brush Au, S., Cohen, S., Pante, N., 2010. Microinjection of Xenopus laevis oocytes as a that sweeps away macromolecules from a large corona surround- system for studying nuclear transport of viruses. Methods 51, 114–120. ing the entrance of the NPC central channel (Lim et al., 2006, Beck, M., Lucic, V., Forster, F., Baumeister, W., Medalia, O., 2007. Snapshots of nuclear pore complexes in action captured by cryo-electron tomography. 2007a,b). The FG corona acts as a barrier to reject non-binding Nature 449, 611–615. molecules, while molecules capable of binding to the FG corona Beck, M., Forster, F., Ecke, M., Plitzko, J.M., Melchior, F., et al., 2004. Nuclear pore cause local FG repeats to collapse, thereby allowing the molecule complex structure and dynamics revealed by cryoelectron tomography. Science 306, 1387–1390. into the transport channel (Lim et al., 2006). It is possible to imag- Blissard, G.W., Rohrmann, G.F., 1990. Baculovirus diversity and molecular biology. ine that as soon as the conical end of the capsid interacts with the Annu. Rev. Entomol. 35, 127–155. large FG corona surrounding the cytoplasmic entrance of the NPC, Brohawn, S.G., Partridge, J.R., Whittle, J.R., Schwartz, T.U., 2009. The nuclear pore complex has entered the atomic age. Structure 17, 1156–1168. all the FG motifs in the central channel are swept away, leaving Carsterns, E.B., Tjia, S.T., Doerfler, W., 1979. Infection of Spodoptera frugiperda cells the central channel empty and ready to be transited by the capsid. with Autographa californica nuclear polyhedrosis virus. I. Synthesis of To demonstrate that this model is applicable for the NPC translo- intracellular proteins after virus infection. Virololy 99, 386–398. Charlton, C.A., Volkman, L.E., 1993. Penetration of Autographa californica nuclear cation of the baculovirus capsid, it would be worthwhile to polyhedrosis virus nucleocapsids into IPLB Sf 21 cells induces actin cable perform immuno-EM of FG-Nups and see their distribution on formation. Virology 197, 245–254. NPC-containing capsids. Chook, Y.M., Suel, K.E., 2011. Nuclear import by karyopherin-betas: recognition and A more recent proposed model, the Forest model, hypothesises inhibition. Biochim. Biophys. Acta 1813, 1593–1606. Cronshaw, J.M., Krutchinsky, A.N., Zhang, W., Chait, B.T., Matunis, M.J., 2002. that the central channel is separated into two zones – large mole- Proteomic analysis of the mammalian nuclear pore complex. J. Cell Biol. 158, cules move through the inner part of the NPC channel, and smaller 915–927. molecules travel along the outer part of this channel (Yamada Dong, S., Wang, M., Qiu, Z., Deng, F., Vlak, J.M., et al., 2010. Autographa californica multicapsid nucleopolyhedrovirus efficiently infects Sf9 cells and transduces et al., 2010). This model suggests that FG domains converge at mammalian cells via direct fusion with the plasma membrane at low pH. J. the centre of the NPC, forming a central plug/transporter structure Virol. 84, 5351–5359. to allow the movement of larger cargos. The presence of a large Elad, N., Maimon, T., Frenkiel-Krispin, D., Lim, R.Y., Medalia, O., 2009. Structural analysis of the nuclear pore complex by integrated approaches. Curr. Opin. structure residing within the central channel is inconsistent with Struct. Biol. 19, 226–232. the translocation of large cargos, such as baculovirus capsids and Finlay, D.R., Newmeyer, D.D., Price, T.M., Forbes, D.J., 1987. Inhibition of in vitro the Balbiani ring granules. If such a central plug/transporter exists, nuclear transport by a lectin that binds to nuclear pores. J. Cell Biol. 104, 189– 200. its fate when large cargos cross the NPC central channel remains to Frenkiel-Krispin, D., Maco, B., Aebi, U., Medalia, O., 2010. Structural analysis of a be explained. metazoan nuclear pore complex reveals a fused concentric ring architecture. J. In summary, our results indicate that whatever is normally Mol. Biol. 395, 578–586. Frey, S., Gorlich, D., 2007. A saturated FG-repeat hydrogel can reproduce the filling the NPC central channel must be remarkably flexible, and permeability properties of nuclear pore complexes. Cell 130, 512–523. is able to retract itself into the membrane-embedded scaffold ring Frey, S., Richter, R.P., Gorlich, D., 2006. FG-rich repeats of nuclear pore proteins form of the NPC, leaving the entire central channel open for the a three-dimensional meshwork with hydrogel-like properties. Science 314, passage of the large baculovirus capsid. Although our data does 815–817. Granados, R.R., 1978. Early events in the infection of Hiliothis zea midgut cells by a not seem to fit with some of the proposed models for NPC func- baculovirus. Virology 90, 170–174. tion, we do not discard the possibility that the NPC may use dif- Granados, R.R., Lawler, K.A., 1981. In vivo pathway of Autographa californica ferent modes of translocation depending on the size of the cargo baculovirus invasion and infection. Virology 108, 297–308. Grunwald, D., Singer, R.H., 2010. In vivo imaging of labelled endogenous beta-actin being translocated. Thus, some of these models may be valid for mRNA during nucleocytoplasmic transport. Nature 467, 604–607. cargos smaller than the baculovirus capsid. Nevertheless, the Hamada, M., Haeger, A., Jeganathan, K.B., van Ree, J.H., Malureanu, L., et al., 2011. translocation of a long cargo, such as the baculovirus capsid, Ran-dependent docking of importin-b to RanBP2/Nup358 filaments is essential for protein import and cell viability. J. Cell Biol. 194, 597–612. supports a model in which the total length and width of the cen- Kalderon, D., Roberts, B.L., Richardson, W.D., Smith, A.E., 1984. A short amino acid tral channel is completely emptied to accommodate the intact sequence able to specify nuclear location. Cell 39, 499–509. capsid. Kowalczyk, S.W., Kapinos, L., Blosser, T.R., Magalhaes, T., van Nies, P., et al., 2011. Single-molecule transport across an individual biomimetic nuclear pore complex. Nat. Nanotechnol. 6, 433–438. Acknowledgments Lieleg, O., Ribbeck, K., 2011. Biological hydrogels as selective diffusion barriers. Trends Cell Biol. 9, 543–551. Lim, R.Y., Ullman, K.S., Fahrenkrog, B., 2008a. Biology and biophysics of the nuclear We are grateful to Dr. David Theilmann (Agri-Food Research pore complex and its components. Int. Rev. Cell. Mol. Biol. 267, 299–342. Centre, Summerland, B.C., Canada) for providing the Baculovirus Lim, R.Y., Aebi, U., Fahrenkrog, B., 2008b. Towards reconciling structure and infected cells and for helpful discussions. We also thank Bradford function in the nuclear pore complex. Histochem. Cell Biol. 129, 105–116. Lim, R.Y., Koser, J., Huang, N.P., Schwarz-Herion, K., Aebi, U., 2007a. Nanomechanical Ross of the BioImaging Facility at University of British Columbia interactions of phenylalanine–glycine nucleoporins studied by single molecule for help with electron tomography. This work was supported by force–volume spectroscopy. J. Struct. Biol. 159, 277–289. Grants from the Canada Foundation for Innovation, the Canadian Lim, R.Y., Fahrenkrog, B., Koser, J., Schwarz-Herion, K., Deng, J., et al., 2007b. Nanomechanical basis of selective gating by the nuclear pore complex. Science Institutes of Health Research, and the Natural Sciences and 318, 640–643. Engineering Research Council of Canada. N.P. is a Michael Smith Lim, R.Y., Huang, N.P., Koser, J., Deng, J., Lau, K.H., et al., 2006. Flexible Foundation for Health Research Senior Scholar. phenylalanine–glycine nucleoporins as entropic barriers to nucleocytoplasmic transport. Proc. Natl. Acad. Sci. USA 103, 9512–9517. Lowe, A.R., Siegel, J.J., Kalab, P., Siu, M., Weis, K., et al., 2010. Selectivity mechanism Appendix A. Supplementary data of the nuclear pore complex characterized by single cargo tracking. Nature 467, 600–603. Maco, B., Fahrenkrog, B., Huang, N.P., Aebi, U., 2006. Nuclear pore complex structure Supplementary data associated with this article can be found, in and plasticity revealed by electron and atomic force microscopy. Methods Mol. the online version, at doi:10.1016/j.jsb.2011.11.006. Biol. 322, 273–288.

Mehlin, H., Daneholt, B., 1993. The Balbiani ring particle: a model for the assembly Rout, M.P., Aitchison, J.D., Suprapto, A., Hjertaas, K., Zhao, Y., et al., 2000. The yeast and export of RNPs from the nucleus? Trends Cell Biol. 3, 443–447. nuclear pore complex: composition, architecture, and transport mechanism. J. Mehlin, H., Daneholt, B., Skoglund, U., 1992. Translocation of a specific Cell Biol. 148, 635–651. premessenger ribonucleoprotein particle through the nuclear pore studied Rowat, A.C., Lammerding, J., Herrmann, H., Aebi, U., 2008. Towards an integrated with electron microscope tomography. Cell 69, 605–613. understanding of the structure and mechanics of the cell nucleus. BioEssays 30, Mehlin, H., Daneholt, B., Skoglund, U., 1995. Structural interaction between the 226–236. nuclear pore complex and a specific translocating RNP particle. J. Cell Biol. 129, Salminen, M., Airenne, K.J., Rinnankoski, R., Reimari, J., Valilehto, O., et al., 2005. 1205–1216. Improvement in nuclear entry and transgene expression of baculoviruses by Newmeyer, D.D., Forbes, D.J., 1988. Nuclear import can be separated into distinct disintegration of microtubules in human hepatocytes. J. Virol. 79, 2720–2728. steps in vitro: nuclear pore binding and translocation. Cell 52, 641–653. Shoji, I., Aizaki, H., Tani, H., Ishii, K., Chiba, T., et al., 1997. Efficient gene transfer into Obregon-Barboza, V., Del Rincon-Castro, M.C., Cabrera-Ponce, J.L., Ibarra, J.E., 2007. various mammalian cells, including non-hepatic cells, by baculovirus vectors. J. Infection, transfection, and co-transfection of baculoviruses by microprojectile Gen. Virol. 78 (Pt 10), 2657–2664. bombardment of larvae. J. Virol. Methods 140, 124–131. Sodeik, B., Ebersold, M.W., Helenius, A., 1997. Microtubule-mediated transport of Ohkawa, T., Volkman, L.E., 1999. Nuclear F-actin is required for AcMNPV incoming herpes simplex virus 1 capsids to the nucleus. J. Cell Biol. 136, 1007– nucleocapsid morphogenesis. Virology 264, 1–4. 1021. Ohkawa, T., Volkman, L.E., Welch, M.D., 2010. Actin-based motility drives Stoffler, D., Fahrenkrog, B., Aebi, U., 1999. The nuclear pore complex: from baculovirus transit to the nucleus and cell surface. J. Cell Biol. 190, 187–195. molecular architecture to functional dynamics. Curr. Opin. Cell Biol. 11, 391– Pante, N., 2006. Use of intact Xenopus oocytes in nucleocytoplasmic transport 401. studies. Methods Mol. Biol. 322, 301–314. Stoffler, D., Feja, B., Fahrenkrog, B., Walz, J., Typke, D., et al., 2003. Cryo-electron Pante, N., 2007. Contribution of electron microscopy to the study of the nuclear pore tomography provides novel insights into nuclear pore architecture: complex structure, composition and function. In: Méndez-Vilas, A., Diaz, J. implications for nucleocytoplasmic transport. J. Mol. Biol. 328, 119–130. (Eds.), Modern Research and Educational Topics in Microscopy. FORMATEX, Strambio-De-Castillia, C., Niepel, M., Rout, M.P., 2010. The nuclear pore complex: Spain, pp. 144–153. bridging nuclear transport and gene regulation. Nat. Rev. Mol. Cell Biol. 11, 490– Pante, N., Aebi, U., 1996. Sequential binding of import ligands to distinct nucleopore 501. regions during their nuclear import. Science 273, 1729–1732. Summers, M.D., 1971. Electron microscopic observations on granulosis virus entry, Pante, N., Kann, M., 2002. Nuclear pore complex is able to transport uncoating and replication processes during infection of the midgut cells of macromolecules with diameters of about 39 nm. Mol. Biol. Cell 13, 425–434. Trichoplusia ni. J. Ultrastruct. Res. 35, 606–625. Peters, R., 2005. Translocation through the nuclear pore complex: selectivity and Terry, L.J., Wente, S.R., 2009. Flexible gates: dynamic topologies and functions for FG speed by reduction-of-dimensionality. Traffic 6, 421–427. nucleoporins in nucleocytoplasmic transport. Eukaryot. Cell 8, 1814–1827. Rabe, B., Vlachou, A., Pante, N., Helenius, A., Kann, M., 2003. Nuclear import of Tetenbaum-Novatt, J., Rout, M.P., 2010. The mechanism of nucleocytoplasmic hepatitis B virus capsids and release of the viral genome. Proc. Natl Acad. Sci. transport through the nuclear pore complex. Cold Spring Harb. Symp. Quant. USA 100, 9849–9854. Biol. 75, 567–584. Raghow, R., Grace, T.D., 1974. Studies on a nuclear polyhedrosis virus in Bombyx van Loo, N.D., Fortunati, E., Ehlert, E., Rabelink, M., Grosveld, F., et al., 2001. mori cells in vitro. 1. Multiplication kinetics and ultrastructural studies. J. Baculovirus infection of nondividing mammalian cells: mechanisms of entry Ultrastruct. Res. 47, 384–399. and nuclear transport of capsids. J. Virol. 75, 961–970. Ribbeck, K., Gorlich, D., 2001. Kinetic analysis of translocation through nuclear pore Walde, S., Kehlenbach, R.H., 2010. The part and the whole: functions of nucleoporins complexes. EMBO J. 20, 1320–1330. in nucleocytoplasmic transport. Trends Cell Biol. 20, 461–469. Ribbeck, K., Gorlich, D., 2002. The permeability barrier of nuclear pore complexes Wente, S.R., Rout, M.P., 2010. The nuclear pore complex and nuclear transport. Cold appears to operate via hydrophobic exclusion. EMBO J. 21, 2664–2671. Spring Harb. Perspect. Biol. 2, a000562. Rollenhagen, C., Pante, N., 2006. Nuclear import of spliceosomal snRNPs. Can. J. Yamada, J., Phillips, J.L., Patel, S., Goldfien, G., Calestagne-Morelli, A., et al., 2010. A Physiol. Pharmacol. 84, 367–376. bimodal distribution of two distinct categories of intrinsically disordered Rollenhagen, C., Muhlhausser, P., Kutay, U., Pante, N., 2003. Importin beta- structures with separate functions in FG nucleoporins. Mol. Cell. Proteomics 9, depending nuclear import pathways: role of the adapter proteins in the 2205–2224. docking and releasing steps. Mol. Biol. Cell 14, 2104–2115. Yaseen, N.R., Blobel, G., 1999. Two distinct classes of Ran-binding sites on the Rout, M.P., Aitchison, J.D., Magnasco, M.O., Chait, B.T., 2003. Virtual gating and nucleoporin Nup-358. Proc. Natl Acad. Sci. USA 96, 5516–5521. nuclear transport: the hole picture. Trends Cell Biol. 13, 622–628.

Please cite this article in press as: Au, S., Panté, N. Nuclear transport of baculovirus: Revealing the nuclear pore complex passage. J. Struct. Biol. (2011), doi:10.1016/j.jsb.2011.11.006