A Fence-like Coat for the Nuclear Pore Membrane

Molecular Cell

Article A Fence-like Coat for the Nuclear Pore Membrane Erik W. Debler,1,2 Yingli Ma,1,2 Hyuk-Soo Seo,1 Kuo-Chiang Hsia,1 Thomas R. Noriega,1 Gu¨nter Blobel,1,* and Andre´ Hoelz1,* 1Laboratory

of Cell Biology, Howard Hughes Medical Institute, The Rockefeller University, 1230 York Avenue, New York, NY 10065, USA authors contributed equally to this work *Correspondence: [email protected] (G.B.), [email protected] (A.H.) DOI 10.1016/j.molcel.2008.12.001 2These

SUMMARY

We recently proposed a cylindrical coat for the nuclear pore membrane in the nuclear pore complex (NPC). This scaffold is generated by multiple copies of seven nucleoporins. Here, we report three crystal structures of the nucleoporin pair Seh1 Nup85, which is part of the coat cylinder. The Seh1 Nup85 assembly bears resemblance in its shape and dimensions to that of another nucleoporin pair, Sec13 Nup145C. Furthermore, the Seh1 Nup85 structures reveal a hinge motion that may facilitate conformational changes in the NPC during import of integral membrane proteins and/or during nucleocytoplasmic transport. We propose that Seh1 Nup85 and Sec13 Nup145C form 16 alternating, vertical rods that are horizontally linked by the three remaining nucleoporins of the coat cylinder. Shared architectural and mechanistic principles with the COPII coat indicate a common evolutionary origin and support the notion that the NPC coat represents another class of membrane coats. d

d

d

d

d

d

INTRODUCTION The nuclear pore complex (NPC) mediates the selective exchange of macromolecules between the nucleus and cytoplasm and represents one of the largest proteinaceous assemblies in the eukaryotic cell (Hoelz and Blobel, 2004; Pemberton and Paschal, 2005; Reichelt et al., 1990). Electron microscopic reconstructions have revealed that the NPC is a cylindrical structure that consists of a central core with 8-fold rotational symmetry across a nucleocytoplasmic axis and 2-fold symmetry in the plane of the nuclear envelope (Beck et al., 2004; Hinshaw et al., 1992; Yang et al., 1998). This symmetric core is linked to the asymmetric ‘‘cytoplasmic filaments’’ and a ‘‘nuclear basket’’ structure (Fahrenkrog et al., 2004). The NPC is composed of 30 different nucleoporins (nups), which are organized into several subcomplexes (Cronshaw et al., 2002; Rout et al., 2000; Suntharalingam and Wente, 2003). Over the last decade, a detailed mechanistic understanding at the atomic level has been accomplished for some of the mobile transport factors and their involvement in nucleocytoplasmic transport (Chook and Blobel, 2001; Cook et al., 2007). However, a similar level of understanding regarding the structure and assembly of the NPC

remains largely elusive, despite its central importance for eukaryotic life. In cells with open mitosis, the NPC is disassembled either into individual nups or various subcomplexes (Belgareh et al., 2001; Loı¨odice et al., 2004; Vasu et al., 2001). Similar subcomplexes were also obtained by dissecting intact NPCs using nonionic detergents and a range of salt concentrations (Suntharalingam and Wente, 2003). In yeast, a well-characterized subcomplex consists of Nup84, Nup85, Nup120, Nup133, Nup145C, Sec13, and Seh1 (Allen et al., 2001; Lutzmann et al., 2002; Siniossoglou et al., 2000). Negative-stain electron microscopy on this heptamer assembled from recombinant proteins revealed a 400 A˚ long Y-shaped complex and established the relative position of its members (Lutzmann et al., 2002). The seven nups are arranged in a linear fashion with Nup133 and Nup84 at the base and the Sec13 Nup145C pair in the center, followed by Nup120 and the Seh1 Nup85 pair at the upper arms of the Y (Figure 1A). All members of the yeast heptamer are well conserved; however, the vertebrate complex contains two additional members, Nup37 and Nup43, forming a nonamer (Fontoura et al., 1999; Loı¨odice et al., 2004; Vasu et al., 2001; Walther et al., 2003). The deletion or immunodepletion of any nup from these complexes has dramatic consequences on the architecture and function of the NPC, as well as the organization of the nuclear envelope (Baı¨ et al., 2004; Boehmer et al., 2003; Dockendorff et al., 1997; Siniossoglou et al., 1996, 2000; Vasu et al., 2001). A notable exception is Seh1, which is nonessential for cell growth, indicating that its close homolog Sec13 may perform an overlapping or redundant function (Siniossoglou et al., 1996). Although the spatial arrangement of the heptamers in the symmetric NPC core has not been determined, the heptamers have been suggested to serve as ‘‘membrane-curving modules,’’ similar to the members of the COPI, COPII, and clathrin coats (Devos et al., 2004). The heptamers were proposed to form two separated rings on the cytoplasmic and nucleoplasmic periphery, each containing eight copies of the heptamer (Alber et al., 2007). Interestingly, Sec13 is shared between the NPC and the COPII cage, where it forms the outer coat layer in complex with Sec31 (Fath et al., 2007; Lederkremer et al., 2001; Salama et al., 1997; Stagg et al., 2006, 2007, 2008). Recently, crystal structures of Sec13 Sec31 and Sec13 Nup145C have been determined (Fath et al., 2007; Hsia et al., 2007). Remarkably, the molecular model of the COPII cage includes an elongated, curved Sec13 Sec31 hetero-octamer that possesses similar architectural features to the Sec13 Nup145C hetero-octamer, which indicates a common evolutionary origin and underlines the physiological relevance of the hetero-octameric assemblies d

d

d

d

d

d

Molecular Cell 32, 815–826, December 26, 2008 ª2008 Elsevier Inc. 815

Molecular Cell A Fence-like Coat for the Nuclear Pore Membrane

Figure 1. Overview of the Seh1 Nup85 Structure d

(A) Schematic representation of the heptameric complex and the approximate localization of its seven nups (Lutzmann et al., 2002). (B) Domain structures of Seh1 and Nup85. For Seh1, the WD40 repeats (orange) and the numbering relative to yeast Seh1 are indicated. For Nup85, the unstructured N-terminal region (gray), the domain invasion motif (DIM) (purple), the a-helical solenoid domain (blue), and the C-terminal a-helical region (light pink) are indicated. The numbering is relative to yeast Nup85. The bars above the domain structures mark the crystallized fragments. (C) Ribbon representation of the Seh1dNup85 hetero-octamer, showing Seh1 in yellow and orange and Nup85 in green and blue. A 90 rotated view is shown on the right. The three pseudo-2-fold axes (black ovals) and the overall dimensions are indicated. The Seh1dNup85 hetero-octamer forms a slightly bent rod. (D) Schematic representation of the Seh1dNup85 hetero-octamer. Magenta lines indicate interaction surfaces.

(Devos et al., 2004; Fath et al., 2007; Hsia et al., 2007; Stagg et al., 2008). We hypothesized that the elongated, curved Sec13 Nup145C hetero-octamer may form a vertical rod in a coat for the nuclear pore membrane (Hsia et al., 2007). In order to gain further evidence for and insights into the architecture of a coat for the nuclear pore membrane in atomic detail, d

we determined crystal structures of the Seh1 Nup85 complex and complemented these results with solution studies. Our analyses show that Seh1 Nup85 occurs as oligomers of heterotetramers in three different crystal forms under vastly different crystallization conditions. Importantly, we also detected oligomerization in solution. Based on structural, biochemical, and

816 Molecular Cell 32, 815–826, December 26, 2008 ª2008 Elsevier Inc.

d

d

Molecular Cell A Fence-like Coat for the Nuclear Pore Membrane

Table 1. Crystallographic Analysis of Crystal Form 1 Crystal 1 Native

Crystal 2 SeMet

Crystal 3 [Ta6Br12]2+

Crystal 4 PCMBa

Crystal 5 K2OsCl6

Synchrotron

ALSb

ALSb

ALSb

ALSb

ALSb

Beamline

BL8.2.2

BL8.2.1

BL8.2.2

BL8.2.2

BL8.2.2

Space group

P21

P21

P21

P21

P21

Cell dimensions a, b, c (A˚)

a = 79.8

a = 79.6

a = 79.1

a = 79.7

a = 79.5

b = 166.2

b = 166.8

b = 163.5

b = 165.3

B = 165.7

c = 188.9

c = 189.8

c = 189.9

c = 190.1

c = 188.9

a = g = 90, b = 93.0

a = g = 90, b = 93.2

a = g = 90, b = 92.8

a = g = 90, b = 93.5

a = g = 90, b = 92.8

Se Peak

Ta Peak

Hg Peak

Os Peak

Data Collection

a, b, g ( ) Wavelength (A˚)

1.00000

0.97950

1.25500

1.00800

1.14030

Resolution (A˚)

50.02.9

50.03.4

50.05.0

50.03.1

50.04.0

Rsym (%)c

7.9 (79.0)

10.8 (60.2)

12.7 (58.6)

7.6 (59.3)

10.3 (75.5)

c

19.9 (1.8)

17.4 (2.2)

8.6 (1.6)

15.0 (1.8)

12.8 (1.3)

97.7 (78.0)

100.0 (100.0)

88.8 (61.6)

98.5 (87.5)

98.8 (93.1)

6.7 (5.0)

4.6 (4.5)

3.3 (2.4)

3.8 (3.3)

3.7 (3.3)

0.440/0.310

0.694/0.529

Completeness (%)

c

Redundancyc Phasing power (iso/ano)

0.322/1.353

Refinement Resolution (A˚)

20.02.9

No. reflections

97,295

Test set

9722 (9.0%)

Rwork / Rfree (%)

24.6 / 26.5

No. atoms

25,239

Rms deviations Bond lengths (A˚)

0.009

Bond angles ( )

1.6

Ramachandran plotd Most favored (%)

82.3

Additionally allowed (%)

13.9

Generously allowed (%)

0.7

Disallowed (%)

0.1

a

PCMB, para-Chloromercuribenzoic acid. b ALS, Advanced Light Source, Lawrence Berkeley National Laboratory. c Highest-resolution shell is shown in parentheses. d As determined by Procheck (Laskowski et al., 1993).

computational analyses, we suggest that Seh1 Nup85 exists as a hetero-octamer in the NPC. Architectural and mechanistic similarities of Seh1 Nup85 and Sec13 Nup145C with COPII coat proteins provide further evidence for a common evolutionary ancestry and suggest that the coat for the nuclear pore membrane represents another class of membrane coats. d

d

d

RESULTS

isomorphous replacement with anomalous scattering (MIRAS) from the monoclinic crystal form (Figure S1 available online) and was refined to 2.9 A˚ resolution with an Rfree of 26.5% (Table 1). The structures of the two orthorhombic space groups were solved by molecular replacement and refined to 3.2 A˚ (Rfree of 28.1%) and 3.75 A˚ (Rfree of 27.2%) resolution, respectively (Tables S1 and S2). In all three crystal forms, we observed higher-order structures of the Seh1 Nup85 pair; while the asymmetric units of crystal forms 1 and 2 each contained an elongated, curved Seh1 Nup85 hetero-octamer with a molecular weight of 400 kDa (Figures 1C and S2A), crystal form 3 harbored an elongated heterododecamer (600 kDa) with a slight helical twist (Figure S2A). These oligomers are assembled from compact Seh1 Nup85 heterotetramers via the same Seh1 Seh1 interface (Figure 1C). The d

Architectural Overview of the Seh1 Nup85 Structure The yeast Nup85 construct comprising residues 1–570 was coexpressed with full-length yeast Seh1 (Figure 1B). We will refer to this Nup85 fragment as Nup85 in the remainder of the text. The Seh1 Nup85 complex crystallized in three unrelated crystal forms. The structure of Seh1 Nup85 was determined by multiple d

d

d

d

d

d

Molecular Cell 32, 815–826, December 26, 2008 ª2008 Elsevier Inc. 817

Molecular Cell A Fence-like Coat for the Nuclear Pore Membrane

Figure 2. Detailed Structural Analysis of Nup85 and Seh1 (A) A ribbon representation of the Nup85 structure is shown in rainbow colors along the polypeptide chain from the N to the C terminus. The N-terminal domain invasion motif (DIM), the a-helical solenoid domain, and their secondary structure elements are indicated. (B) The structure of the Seh1dNup85 heterodimer. The Nup85DIM (magenta), the Nup85 a-helical solenoid domain (blue), the Nup85 aQ-aR connector (red), the Seh1 b propeller (yellow), the disordered Seh1 5CD loop (gray dots), and the Seh1 2CD loop (orange) are indicated; a 90 rotated view is shown on the right. Dotted lines represent disordered regions. (C) Schematic representation of the Seh1dNup85 interaction. The Seh1 2CD loop, the Nup85 aQ-aR connector, and the DIM region are highlighted in orange, red, and pink, respectively.

818 Molecular Cell 32, 815–826, December 26, 2008 ª2008 Elsevier Inc.

Molecular Cell A Fence-like Coat for the Nuclear Pore Membrane

Figure 3. Surface Properties Seh1 Nup85 Heterodimer

of

the

d

(A) Surface rendition of the Seh1dNup85 complex. The surface is colored according to the proteins (Seh1, yellow; Nup85, blue) and their participation in various interactions: with Seh1 of the adjacent heterodimer, orange; with Nup85 of the adjacent heterodimer, green; and with Seh1 of the adjacent heterotetramer, purple. (B) Nup85 is colored according to sequence conservation, from 40% similarity (yellow) to 100% identity (red). (C) Surface rendition of Nup85, colored according to the electrostatic potential, from red (15 kBT/e) to blue (+15 kBT/e).

and 2E). The complementation of a sixbladed b propeller by a DIM was previously observed in the Sec13 Sec31 and Sec13 Nup145C structures (Fath et al., 2007; Hsia et al., 2007). The b strands 6E of Nup85 and 7D of Seh1 each provide a Velcro closure (Paoli, 2001) for blades 6 and 7, respectively (Figure 2E). Seh1 features two long loops, namely 2CD (the loop between strands C and D of blade 2) and 5CD. The 2CD loop mediates hetero-octamer and heterododecamer formation, respectively. By contrast, the large 5CD loop is not involved in any contacts with neighboring Seh1 or Nup85 molecules, and the major part of this surface-exposed loop is invisible in the electron density, presumably due to disorder. Altogether, 78 and 86 residues of Nup85 and Seh1, respectively, contribute to a large buried surface area of 5180 A˚2 at the heterodimer interface (Figures S3–S6). The significance of this interaction is further substantiated by high sequence conservation and a primarily hydrophobic character (Figures S3–S6). d

d

heterotetramer, in turn, is formed from two Seh1 Nup85 pairs that are related by a pseudo-2-fold symmetry via a large interface (Figures 1C and 1D). Furthermore, the interfaces within the oligomers are evolutionarily conserved, suggesting their physiological relevance. d

The Seh1 Nup85 Pair Nup85 adopts an a-helical fold in the form of a U-shaped solenoid structure that is preceded by a small domain invasion motif (DIM) (Figures 2A–2C and Movies S1 and S2). The 470-residue solenoid domain consists of 22 a helices. Helices aA–aD form the descending part of the U, helices aE–aK pack into a six-helix bundle at the base of the U, helices aL–aQ form the lower ascending arm, while the helices aR–aV at the top of the ascending arm extensively interact with the bottom face of the Seh1 b propeller. A salient feature is the extended 30-residue aQ–aR connector that contributes numerous contacts between two neighboring Nup85 molecules in the tetramer (see below). The overall structure of the seven-bladed b propeller, which is formed by Seh1 and the DIM of Nup85, conforms to the canonical b propeller fold (Napetschnig et al., 2007; Paoli, 2001) (Figures 2D d

Seh1 Nup85 Heterotetramer In all three crystal forms, two Seh1 Nup85 pairs are tightly associated with each other longitudinally in an antiparallel and complementary fashion (Figures 1C, 1D, and S2A). Superimposition of the seven crystallographically independent heterotetramers reveals that they are closely related with a root-meansquare deviation of only 1.1 A˚ over 1386 Ca atoms (Figure S2B). Therefore, the heterotetramer structure appears to be neither influenced by the various crystal packing environments nor by the differing crystallization conditions. The dimer-dimer interface buries a huge surface area of 5100 A˚2 that includes 101 residues on each heterodimer (Figures 3, 4, S3, and S4). Notably, the long aQ-aR connector extends to the neighboring Nup85 molecule and reinforces the interaction between the two nucleoporin d

d

(D) The b propeller domain of Seh1 in complex with the Nup85DIM. Seh1 is shown in yellow, and the six blades are indicated. The Nup85DIM contributes one strand to blade 6 and three strands to blade 7, completing the b propeller. (E) Schematic representation of the Seh1 b propeller and its interaction with the Nup85DIM.

Molecular Cell 32, 815–826, December 26, 2008 ª2008 Elsevier Inc. 819

Molecular Cell A Fence-like Coat for the Nuclear Pore Membrane

Figure 4. Interfaces in the Seh1 Nup85 Hetero-Octamer d

(A) A large surface area of 5100 A˚2 is buried upon heterotetramer formation and extends over almost the entire length of a Seh1dNup85 heterodimer. a helices that mediate heterotetramer formation are indicated. The Nup85 aQ-aR connector segments (red) are located at the center and contribute many contacts to the interface. (B) The Seh1dSeh1 dimerization interface located at the center of the Seh1dNup85 hetero-octamer is significantly smaller than the interface in (A). The two interacting Seh1 b propeller domains and the two adjacent Nup85 molecules are colored yellow and blue, respectively. The locations of the pseudo-2-fold axes of symmetry that run through both interfaces are indicated (black ovals).

pairs (Figure 4A). A large number of apolar residues impart a hydrophobic character to the upper and lower ends of the interface (Figure 3C). Computational analysis of the heterotetrameric complex by the PISA server confirms that this assembly possesses high thermodynamic stability in solution (Krissinel and Henrick, 2007). Altogether, these findings suggest that Seh1 Nup85 heterotetramerization also occurs in vivo. d

Seh1 Nup85 Hetero-Octamer The hetero-octameric assembly of Seh1 Nup85 in crystal form 1 has overall dimensions of 280 A˚ 3 110 A˚ 3 50 A˚ (Figure 1C and Movies S3 and S4). The association of two heterotetramers is exclusively mediated by the homodimerization of Seh1 (Figure 4B). In detail, the 2CD loop contributes the major part to the interface, while the interblade loop 1D2A participates only to a minor extent. Although only 400 A˚2 of surface area are buried at this interface, the shape complementarity parameter Sc of 0.72 is high and falls into the range typically observed for protein oligomeric interfaces (0.70–0.74) (Lawrence and Colman, 1993). Importantly, 7 out of 16 Seh1 residues are charged and form several salt bridges at the interface that are likely to reinforce the heterooctameric assembly via electrostatic interactions (Figure S5C). d

d

Conformational Flexibility of Seh1 Nup85 Oligomers All higher-order oligomers are assembled via the same interface between heterotetramers. Consistent with the relatively small size of this tetramer-tetramer interface, we observe structural plasticity in this region in various crystal forms (Figure 5). The d

two hetero-octameric assemblies of crystal forms 1 and 2 both form elongated, curved rods that are related by an 35 hinge motion around the center of the hetero-octamer. In crystal form 2, the curvature of the hetero-octamer is more pronounced, with a slightly decreased buried surface area of 340 A˚2 (Figure 5A and Movie S5). The involved loop regions of the interface maintain their main-chain conformations, while adjustments of several side-chain residues take place to accommodate the related hetero-octamer conformations. Notably, residues with long but well-defined side chains, such as arginine, lysine, and glutamate residues, are prevalent at the interface. In crystal form 3, the heterododecamer possesses two identical tetramer-tetramer interfaces. Comparison of crystal forms 1 and 3 reveals that two adjacent heterotetramers in the heterododecamer display the same overall curvature but that one of the heterotetramers is rotated around its long axis by 80 (Figure 5B and Movie S6), now burying a surface area of 950 A˚2. The rigid body movement of the heterotetramer causes a register shift in the interface, whereby hydrogen bond donors and acceptors switch their partners, maintaining a similar hydrogen bond network in both structures. Interestingly, such intermolecular rearrangements were also observed in the Nup58/45 tetramer a´k et al., 2007). (Melc Oligomerization of Seh1 Nup85 in Solution In order to probe the oligomerization state of Seh1 Nup85 in solution, we employed multiangle light scattering, analytical ultracentrifugation, and analytical size exclusion chromatography. Seh1 Nup85 elutes from a gel filtration column as two peaks with apparent molecular weights corresponding to heterotetramers and hetero-octamers. However, gel filtration coupled with multiangle light scattering revealed that Seh1 Nup85 predominantly occurs as a heterodimer (92 kDa) that exists in equilibrium with a minor population of a heterotetrameric species (184 kDa) (Figure S7A). Consistent with this result, analytical

820 Molecular Cell 32, 815–826, December 26, 2008 ª2008 Elsevier Inc.

d

d

d

d

Molecular Cell A Fence-like Coat for the Nuclear Pore Membrane

Figure 5. Flexibility of the Seh1 Nup85 Hetero-Octamer d

(A) The hetero-octamers of crystal forms 1 and 2 are related by an 35 hinge motion around the center of the hetero-octamer. On the right, a schematic of the two conformations is shown, where Seh1 and Nup85 are displayed as balls and cylinders, respectively. (B) Crystal form 3 harbors a heterododecamer in the asymmetric unit (small ribbon representation). The two interfaces between the heterotetramers are identical. In the upper two neighboring heterotetramers of crystal form 3 (boxed in the heterododecamer), one heterotetramer is rotated by 80 around its long axis with respect to crystal form 1. For clarity, a stripe of black lines marks the relative orientations of the heterotetramers. The alignment of the different structures was based on the lower heterotetramer of crystal form 1.

Comparison of Seh1 Nup85 with Sec13 Nup145C Superimposition of Seh1 on Sec13 reveals high structural similarity (Figure 6A). Additional elements in Seh1 with respect to Sec13 refer to the Velcro closure b strands of blades 6 and 7, as well as the insertions in the 2CD (8 residues) and 5CD loops (41 residues). Notably, the 2CD loop of Seh1 plays a prominent role in mediating hetero-octamer formation in Seh1 Nup85. Although the a-helical domains of Nup85 and Nup145C share a similar fold, they cannot be superimposed onto each other and show marked differences. The solenoid domain in Nup85 is considerably twisted in the region most distant from the b propeller (Figure 6A). Furthermore, four additional a helices, as well as differing orientations of topologically equivalent a helices, are present in Nup85. On the level of higher-order oligomers, the organization of the two complexes is substantially different (Figure 6B). Seh1 Nup85 forms a compact heterotetramer with a large interface involving two b propellers and two a-helical solenoid domains, whereas the elongated Sec13 Nup145C heterotetramer is assembled via an interface between two solenoid domains. In Seh1 Nup85, hetero-octamer formation is realized by a small b propeller homodimerization interface that is not present in Sec13, while two elongated Sec13 Nup145C heterotetramers overlap with each other along a large interface that engages two b propellers and two a-helical solenoid domains. Although the Sec13 residues that mediate Sec13 Nup145C hetero-octamer formation are conserved in Seh1, an analogous mode of homodimerization is sterically prevented in Seh1, as this region is already utilized for heterotetramer formation in Seh1 Nup85. Strikingly, the two disparate hetero-octamer architectures of the two complexes arrive at an elongated rod of similar overall dimensions and a comparable curvature (Figure 6B), suggesting similar but nonidentical roles of the two nucleoporin pairs in the NPC architecture. d

d

ultracentrifugation corroborated the Seh1 Nup85 heterodimer as the primary species in solution with a molecular weight of 104 kDa (Figure S7B). However, due to the small heterotetramer fraction in solution combined with the low protein concentration dictated by analytical ultracentrifugation, the heterotetramer was not detected by this technique. d

The Hybrid Sec13 Nup85 Complex Yields Insights into the Architectural Redundancy within the NPC Deletion studies in yeast established that Seh1 is nonessential for cell growth and indicate that Sec13 could perform an overlapping or redundant function (Siniossoglou et al., 1996). Structure-based sequence alignment, in fact, demonstrates that most of the Seh1 residues that interact with Nup85 are either identical or conserved in Sec13 (Hsia et al., 2007) (Figure S3), which would enable Sec13 to substitute for Seh1. In order to test this hypothesis, we coexpressed yeast Sec13 and Nup85. We were able to purify the hybrid complex and characterize it through analytical size exclusion chromatography (Figure S8). In comparison with Seh1 Nup85, the significantly reduced amount of soluble protein in our bacterial expression system points toward a less stable Sec13 Nup85 complex. Consistent with this observation, coexpression of Nup85 with Seh1 and Sec13 together exclusively resulted in Seh1 Nup85 complexes (data not shown). Finally, the coexpression of Seh1 and Nup145C did not yield a complex (data not shown). Altogether, replacement of Seh1 with Sec13 can be tolerated on the molecular level, in agreement with the nonlethality of the genetic Seh1 deletion (Siniossoglou et al., 1996). d

d

d

d

d

d

d

d

d

d

d

Molecular Cell 32, 815–826, December 26, 2008 ª2008 Elsevier Inc. 821

Molecular Cell A Fence-like Coat for the Nuclear Pore Membrane

Figure 6. Comparison of Seh1 Nup85 and Sec13 Nup145C d

d

(A) Comparison of the Seh1dNup85 (left) and the Sec13dNup145C (middle) heterodimers. Superimposition is based on the b propellers (right). (B) Comparison of the Seh1dNup85 (first and second panel) and the Sec13dNup145C heterooctamers (third and fourth panel).

(Figure 4) that mediate tetramer and octamer formation, respectively. As a consequence, the interface between the two heterodimers amounts to only 1800 A˚2 in the 3.5 A˚ structure. By contrast, detailed analysis of the dimer-dimer interface in the Seh1 Nup85 heterotetramers determined here uncovered an interface of more than 5000 A˚2 with characteristics typical of macromolecular complexes, suggesting a physiological relevance of the observed oligomerization. Indeed, dimerization of the Seh1 Nup85 pair also occurs in solution, albeit to a small extent. Although neither the crystalline state nor a diluted solution of Seh1 Nup85 can faithfully mimic the environment of Seh1 Nup85 in the NPC, the tetramer is likely to occur in the NPC, given the high local concentrations of Seh1 Nup85, as well as the presence of neighboring nucleoporins that interact with Seh1 Nup85 in the NPC. Due to the 8-fold rotational symmetry of the NPC (Hinshaw et al., 1992), experimental estimates for the copy number of Nup85 (30.4 ± 4) and Seh1 (27.2 ± 11.2) per yeast NPC (Rout et al., 2000) suggest a stoichiometry of 32 Seh1 Nup85 pairs in the NPC. A plausible higher-order structure of Seh1 Nup85 in the NPC is immediately gleaned from the Seh1 Nup85 arrangement in the three crystal structures, where adjacent Seh1 Nup85 heterotetramers are aligned linearly. Strikingly, the length of the Seh1 Nup85 hetero-octamer (280 A˚) is in good agreement with the height of the yeast NPC (300 A˚), as determined by electron microscopy (Yang et al., 1998). Furthermore, the bent hetero-octamer would fit the nuclear pore membrane curvature, while its internal symmetry conforms to the 2-fold symmetry of the core of the NPC in the plane of the nuclear envelope (Hinshaw et al., 1992). The observation of an elongated, curved Seh1 Nup85 heterooctamer fully supports our previously proposed architecture of a coat for the nuclear pore membrane (Hsia et al., 2007). According to this model, eight vertical poles of Seh1 Nup85 would alternate with a second set of eight vertical poles of Sec13 Nup145C. These hetero-octamers would form the vertical connections of four antiparallel stacked rings, each composed of eight heptamers. Horizontal connections between the poles, d

d

d

d

d

d

d

DISCUSSION

d

d

The structure of Seh1 Nup85 provides fundamental insights into the architecture and the dynamic nature of the coat for the nuclear pore membrane, as well as insights into a common evolutionary origin of the NPC coat and the COPII cage of coated vesicles. Crystallographic analyses revealed that Seh1 Nup85 forms oligomers of a compact heterotetramer under three vastly different crystallization conditions and in three different crystal packing environments. A fourth crystal form was reported during the review of this manuscript that also contains the heterotetramer in the asymmetric unit of the crystal (Brohawn et al., 2008). However, this structure at a resolution of 3.5 A˚ with an Rfree of 36.9% reveals sequence register shifts in the Seh1 strand 5D and in the Nup85 helices aE and aI; possesses 86 truncated side chains per heterodimer; and misses numerous features such as the two Velcro closure b strands (Figure 2D) and a total of 17 loops, including the critical aQ-aR connector and the 2CD loop d

d

822 Molecular Cell 32, 815–826, December 26, 2008 ª2008 Elsevier Inc.

d

d

d

d

d

Molecular Cell A Fence-like Coat for the Nuclear Pore Membrane

mediated by the remaining three nucleoporins of the heptamer, would then give rise to a picket fence-like coat for the nuclear pore membrane. However, this model is based on the assumption that the oligomerization of the two complexes also occurs in the fully assembled NPC and is not disrupted by neighboring nups. This assumption is plausible, given the extensive interfaces that are evolutionarily conserved, hydrophobic to a large extent, complementary in shape, and given the fact that both Seh1 Nup85 and Sec13 Nup145C also oligomerize in solution. X-ray crystallography, as well as cryoelectron microscopy, have been the prime methods of not only visualizing the structure of complexes with known stoichiometry, but also identifying novel and unanticipated interactions between macromolecules that could not readily be detected in solution and that often lead to a paradigm shift in the respective field. Prominent examples refer to the EPO receptor dimer in the unliganded state (Livnah et al., 1999), to the asymmetric CDK/cyclin-like dimer of the EGFR kinase domain (Zhang et al., 2006), and—most relevant to the work presented here—to the hetero-octamer of Sec13 Sec31 in the COPII coat (Stagg et al., 2006). The underlying theme of all of these examples is the biological context, such as the cell-surface membrane or membrane-bound vesicles, that favors the observed interactions. Accordingly, we envision that the Sec13 Nup145C and Seh1 Nup85 hetero-octamers are plausible assemblies in the nuclear pore complex. Structures of additional nucleoporin complexes and of higher-order assemblies are needed to confirm these interactions. The observed flexibility of Seh1 Nup85 oligomers (Figure 5) may be important for the function of the NPC. In fact, evidence for conformational plasticity and deviations of the NPC from an 8-fold rotational symmetry has been reported (Akey, 1995; Beck et al., 2007; Hinshaw and Milligan, 2003). Such plasticity would be required for the proposed import of integral membrane proteins to the inner nuclear membrane (King et al., 2006). We envisage that hinge motions, such as those observed for the Seh1 Nup85 pole, could provide a mechanism for such rearrangements. Intriguingly, a hinge motion between two adjacent b propellers in the related Sec13 Sec31 coat layer has recently been proposed to accommodate a wide range of different size and different shape cargo in COPII-coated vesicles (Stagg et al., 2008). The different Seh1 Nup85 conformations provide evidence for such motions and suggest that similar mechanistic principles operate in both membrane coat classes. The Sec13 Nup145C and Seh1 Nup85 hetero-octamers exhibit interfaces at their distal ends that would allow for the attachment of additional heterotetramers. Indeed, a chimeric complex of human Sec13 and yeast Nup145C forms a small amount of heterododecamers in solution (Hsia et al., 2007). As for Seh1 Nup85, we were able to trap a heterododecamer in one of our crystal forms (Figures 5B and S2A). However, the agreement of the height of the hetero-octamers with that of the yeast NPC core suggests that only hetero-octamers occur in the fully assembled NPC. The uncapped ends of the hetero-octamers may provide sites for the attachment of asymmetric nups to the symmetric NPC core. Precedence for such a capping mechanism is provided by the COPII complex, where the distal ends of the Sec13 Sec31 hetero-octamers interact with the center portion of adjacent d

d

d

d

d

d

d

d

d

d

d

d

d

Sec13 Sec31 hetero-octamers (Fath et al., 2007). Although Sec13 Sec31 hetero-octamer formation could not be detected in solution or in the crystal, cryo-EM studies demonstrated that the hetero-octamer unit does, in fact, occur in the COPII coat (Fath et al., 2007; Stagg et al., 2006). Thus, in analogy to the COPII cage, numerous weak interactions with adjacent nucleoporins in the NPC core can lead to a strong and stable assembly. A proteomics-based computational approach provided an alternative model for the architecture of the NPC (Alber et al., 2007). Although this study also describes a core scaffold that coats the nuclear envelope membrane, the architecture of the NPC coat in our model is different. The scaffold in the computational model contains two separated rings at the nucleoplasmic and cytoplasmic periphery, which together harbor 16 heptamers, inconsistent with their previously determined stoichiometry (Rout et al., 2000). These peripheral ‘‘outer’’ rings sandwich two centrally located ‘‘inner’’ rings containing Nup157, Nup170, Nup188, and Nup192. By contrast, four continuous, stacked rings would be formed by 32 heptamers in our model. Neither the heterotetramers nor the hetero-octamers of Seh1 Nup85 and Sec13 Nup145C would be compatible with the proteomics-based computational model. Discrepancies between the two models may, in part, derive from the fact that each nup of the heptamer was assumed to be present in 16 copies per NPC in the calculation of the proteomics-based model (Alber et al., 2007). Hence, a cylindrical coat composed of 32 heptamers, as we propose, could not have been attained. d

d

d

d

Conclusions The detailed molecular architecture of the symmetric core of the NPC is unknown at present. However, crystal structures of the large hetero-octameric assemblies of Sec13 Nup145C and Seh1 Nup85 yielded insights into the architecture of the NPC core at the atomic level. Our findings suggest that 32 copies of a well-characterized heptameric subcomplex form a continuous yet porous fence-like coat for the nuclear pore membrane. According to this model, Sec13 Nup145C and Seh1 Nup85 hetero-octamers would form vertical rods in the NPC. Striking similarities of these complexes with elements of the COPII coat of vesicles in the secretory pathway underline their proposed common evolutionary origin (Devos et al., 2004) and suggest that the NPC coat represents another class of membrane coats, in addition to the established COPI, COPII, and clathrin coats for membrane-bound vesicles (Stagg et al., 2007). Finally, the structures of Seh1 Nup85 revealed a hinge motion that may facilitate conformational changes of the NPC during import of integral membrane proteins and/or during nucleocytoplasmic transport. The structure determination of additional nucleoporin complexes is needed to further unravel the molecular architecture of the symmetric core of the NPC piece by piece and holds great promise to ultimately lead to a complete atomic model of the entire NPC, a central machinery for all eukaryotic life. d

d

d

d

d

EXPERIMENTAL PROCEDURES Protein Expression, Purification, and Crystallization DNA fragments encoding full-length yeast Seh1 and full-length yeast Sec13 were cloned into a modified pET28a expression vector (Novagen) that

Molecular Cell 32, 815–826, December 26, 2008 ª2008 Elsevier Inc. 823

Molecular Cell A Fence-like Coat for the Nuclear Pore Membrane

contained a PreScission protease site directly after the N-terminal hexahistidine tag (Hoelz et al., 2003). DNA fragments encoding yeast Nup85 (residues 1–570) and yeast Nup145C (125–555) were cloned into the pETDuet-1 expression vector (Novagen). For the expression of the various complexes, E. coli BL21-CodonPlus (DE3)-RIL cells (Stratagene) were cotransformed with the appropriate expression vectors. Protein expression was carried out in LB medium and induced by the addition of 0.5 mM IPTG at 18 C for 16 hr. Cells were harvested by centrifugation and resuspended in a buffer containing 20 mM TRIS (pH 8.0), 500 mM NaCl, 5 mM b-mercaptoethanol, and protease inhibitor cocktail (Roche). The cells were lysed with a cell disrupter (Avestin), and the lysate was centrifuged for 60 min at 40,000 3 g. The lysate was then applied to a Ni-NTA column (QIAGEN) and eluted via an imidazole gradient. Fractions containing the complex were pooled and concentrated. The protein was purified over a 16/60 Superdex 200 column (GE Healthcare). Fractions containing the complex were pooled and cleaved with PreScission protease (GE Healthcare) for 36 hr. The complex was further purified over a 5/50 MonoQ column (GE Healthcare) and eluted via an NaCl gradient. Fractions containing the complex were pooled, concentrated, and finally purified over a 16/60 Superdex 200 column (GE Healthcare). Crystals of Seh1 Nup85 (15 mg/ml) were grown at 21 C in hanging drops containing 1 ml of protein and 1 ml of reservoir solution in the space groups P21 (crystal form 1), P212121 (crystal form 2), and P21212 (crystal form 3), respectively. The crystallization conditions consisted of 9% (w/v) PEG 10,000, 0.1 M MES (pH 6.1) (crystal form 1); of 0.8 M sodium citrate (pH 7.1), 0.2 M NaCl, 0.1 M TRIS (pH 7.2) (crystal form 2); and of 12% (w/v) PEG 3350 and 4% tacsimate (pH 6.0) (crystal form 3). In the case of the monoclinic crystal form, microseeding substantially improved the growth of large single crystals. Typically, they grew to their maximum size of 1000 3 300 3 200 mm3 within 2–3 months. For cryoprotection, crystals were stabilized in 10% (w/v) PEG 10,000, 0.1 M MES (pH 6.1), 23% (v/v) glycerol (crystal form 1); 0.9 M sodium citrate (pH 7.1), 0.2 M NaCl, 0.1 M TRIS (pH 7.2), 23% (v/v) glycerol (crystal form 2); and 14% (w/v) PEG 3350 and 4% tacsimate (pH 6.0) and 23% (v/v) glycerol (crystal form 3). Cryoprotected crystals were then flash-cooled in liquid propane. X-ray diffraction data were collected at beamlines 8.2.1 and 8.2.2 at the Advanced Light Source (ALS) and at the GM/CA-CAT and NE-CAT beamlines at the Advanced Photon Source (APS). X-ray intensities were processed using HKL2000 (Otwinowski and Minor, 1997) (Tables 1, S1, and S2). The CCP4 program package (CCP4, 1994) was used for subsequent calculations. d

Structure Determination Initial phases were determined by multiple isomorphous replacement with anomalous scattering (MIRAS) using X-ray diffraction data obtained from seleno-L-methionine (SeMet)-labeled and several heavy-metal derivatized crystals of the monoclinic crystal form (Figure S1 and Table 1). Phasing was carried out in SHARP (de La Fortelle and Bricogne, 1997), followed by density modification in DM (CCP4, 1994), with solvent flattening and histogram matching. This procedure yielded an interpretable electron density map of high quality. The initial model was built into the electron density map of the monoclinic crystal form using O (Jones et al., 1991) and refined using CNS (Bru¨nger et al., 1998). The final model was refined to 2.90 A˚ resolution with an Rwork of 24.6% and an Rfree of 26.5% (Table 1). No electron density was observed for the N-terminal 43 and C-terminal 19 residues, the loop residues 127–131 and 442–449 in Nup85, and residues 249–290 of Seh1. These residues are presumed to be disordered and, therefore, were omitted from the final model. Phasing of the X-ray diffraction data derived from the other two crystal forms was carried out with molecular replacement using the program Phaser (McCoy et al., 2007) and the coordinates of the refined Seh1 Nup85 complex from the monoclinic crystal form. The molecular replacement solutions were independently confirmed by experimental phasing of SeMet-labeled, potassium osmate(VI)-derivatized (K2OsO4) and [Ta6Br12]2+ cluster-derivatized protein crystals (Tables S1 and S2). The models were refined using CNS to 3.75 and 3.2 A˚ resolution with Rwork and Rfree factors of 24.4 and 27.2% (crystal form 2) and 26.3 and 28.1% (crystal form 3), respectively. The stereochemical quality of all models was assessed with PROCHECK (Laskowski et al., 1993) and Molprobity (Davis et al., 2007). Leu188 in Seh1 is the only outlier in the d

structure. However, this residue is well defined in the electron density map and is located in a canonical b turn type II0 (Richardson, 1981). Data collection and refinement statistics are shown in Tables 1, S1, and S2. Ultracentrifugation and Multiangle Light Scattering Sedimentation velocity experiments were performed at 4 C and 20 C in a Beckman Optima XL-I analytical ultracentrifuge at a rotor speed of 50,000 rpm. Double-sector cells were loaded with 400 ml of the protein sample (2 mg/ml Seh1 Nup85 in a solution containing 20 mM Tris (pH 8.0), 100 mM NaCl, and 5 mM DTT) and 410 ml of the reference solutions, respectively. The reference solution was 20 mM Tris buffer (pH 8.0), 100 mM NaCl, and 5 mM DTT. Data were recorded with absorbance detection at wavelength 300 nm. The partial specific volume and the solvent density were calculated using the SEDNTERP program. The SEDFIT analysis program was used to analyze the absorbance profiles and to calculate the sedimentation coefficient distribution c(s), which was then transformed into a molar mass distribution c(M) (Schuck, 2000). Purified protein was characterized by multiangle light scattering following size exclusion chromatography. Protein (52 mg/ml) was injected onto a Superdex 200 HR10/30 size exclusion chromatography column (GE Healthcare) equilibrated in gel filtration buffer (20 mM Tris [pH 8.0], 100 mM NaCl, and 5 mM DTT). The chromatography system was coupled to an 18 angle light scattering detector (DAWN HELEOS) and refractive index detector (Optilab rEX) (Wyatt Technology). Data were collected every 1 second at a flow rate of 0.5 ml/min. Data analysis was carried out using the program ASTRA V, yielding the molar mass and mass distribution (polydispersity) of the sample. d

Illustrations and Figures Figures were generated using PyMOL (www.pymol.org). The electrostatic potential was calculated using APBS (Baker et al., 2001). Sequence alignments were generated using ClustalX (Jeanmougin et al., 1998) and colored with ALSCRIPT (Barton, 1993). ACCESSION NUMBERS The atomic coordinates and structure factors of the three crystal structures have been deposited to the Protein Data Bank with the accession codes 3F3F, 3F3G, and 3F3P. SUPPLEMENTAL DATA The Supplemental Data include eight figures, two tables, and six movies and can be found with this article online at http://www.cell.com/molecular-cell/ supplemental/S1097-2765(08)00840-X. ACKNOWLEDGMENTS a´k, V. Nagy, J. Napetschnig, A. Patke, and P. StavWe thank O. Dreesen, I. Melc ropoulos for discussions and comments on the manuscript; S. Etherton for help with editing the manuscript; D. King for mass spectrometry analysis; P. Stavropoulos and E. Hurt for providing material; and J. Champagne for assistance with the light scattering analysis. In addition, we thank C. Ralston (ALS), M. Becker and R. Fischetti (GM/CA-CAT), K. Rajashankar and N. Sukumar (NE-CAT), and W. Shi (NSLS) for support during data collection. N-terminal protein sequencing was performed by the Protein Center of the Rockefeller University; analytical ultracentrifugation was carried out by the Wadsworth Center Biochemistry Core Facility. E.W.D. is the Dale F. and Betty Ann Frey Fellow of the Damon Runyon Cancer Research Foundation, DRG-1977-08. A.H. was supported by a grant from the Leukemia and Lymphoma Society. Received: September 8, 2008 Revised: November 21, 2008 Accepted: December 2, 2008 Published: December 24, 2008

824 Molecular Cell 32, 815–826, December 26, 2008 ª2008 Elsevier Inc.

Molecular Cell A Fence-like Coat for the Nuclear Pore Membrane

REFERENCES

Fahrenkrog, B., Koser, J., and Aebi, U. (2004). The nuclear pore complex: A jack of all trades? Trends Biochem. Sci. 29, 175–182.

Akey, C.W. (1995). Structural plasticity of the nuclear pore complex. J. Mol. Biol. 248, 273–293.

Fath, S., Mancias, J.D., Bi, X., and Goldberg, J. (2007). Structure and organization of coat proteins in the COPII cage. Cell 129, 1325–1336.

Alber, F., Dokudovskaya, S., Veenhoff, L.M., Zhang, W., Kipper, J., Devos, D., Suprapto, A., Karni-Schmidt, O., Williams, R., Chait, B.T., et al. (2007). The molecular architecture of the nuclear pore complex. Nature 450, 695–701.

Fontoura, B.M., Blobel, G., and Matunis, M.J. (1999). A conserved biogenesis pathway for nucleoporins: Proteolytic processing of a 186-kilodalton precursor generates Nup98 and the novel nucleoporin, Nup96. J. Cell Biol. 144, 1097–1112.

Allen, N.P., Huang, L., Burlingame, A., and Rexach, M. (2001). Proteomic analysis of nucleoporin interacting proteins. J. Biol. Chem. 276, 29268–29274. Baı¨, S.W., Rouquette, J., Umeda, M., Faigle, W., Loew, D., Sazer, S., and Doye, V. (2004). The fission yeast Nup107–120 complex functionally interacts with the small GTPase Ran/Spi1 and is required for mRNA export, nuclear pore distribution, and proper cell division. Mol. Cell. Biol. 24, 6379–6392. Baker, N.A., Sept, D., Joseph, S., Holst, M.J., and McCammon, J.A. (2001). Electrostatics of nanosystems: Application to microtubules and the ribosome. Proc. Natl. Acad. Sci. USA 98, 10037–10041. Barton, G.J. (1993). ALSCRIPT: A tool to format multiple sequence alignments. Protein Eng. 6, 37–40. Beck, M., Forster, F., Ecke, M., Plitzko, J.M., Melchior, F., Gerisch, G., Baumeister, W., and Medalia, O. (2004). Nuclear pore complex structure and dynamics revealed by cryoelectron tomography. Science 306, 1387–1390. Beck, M., Lucic, V., Forster, F., Baumeister, W., and Medalia, O. (2007). Snapshots of nuclear pore complexes in action captured by cryo-electron tomography. Nature 449, 611–615. Belgareh, N., Rabut, G., Bai, S.W., van Overbeek, M., Beaudouin, J., Daigle, N., Zatsepina, O.V., Pasteau, F., Labas, V., Fromont-Racine, M., et al. (2001). An evolutionarily conserved NPC subcomplex, which redistributes in part to kinetochores in mammalian cells. J. Cell Biol. 154, 1147–1160. Boehmer, T., Enninga, J., Dales, S., Blobel, G., and Zhong, H. (2003). Depletion of a single nucleoporin, Nup107, prevents the assembly of a subset of nucleoporins into the nuclear pore complex. Proc. Natl. Acad. Sci. USA 100, 981–985. Brohawn, S.G., Leksa, N.C., Spear, E.D., Rajashankar, K.R., and Schwartz, T.U. (2008). Structural evidence for common ancestry of the nuclear pore complex and vesicle coats. Science 322, 1369–1373.

Hinshaw, J.E., and Milligan, R.A. (2003). Nuclear pore complexes exceeding eightfold rotational symmetry. J. Struct. Biol. 141, 259–268. Hinshaw, J.E., Carragher, B.O., and Milligan, R.A. (1992). Architecture and design of the nuclear pore complex. Cell 69, 1133–1141. Hoelz, A., and Blobel, G. (2004). Cell biology: Popping out of the nucleus. Nature 432, 815–816. Hoelz, A., Nairn, A.C., and Kuriyan, J. (2003). Crystal structure of a tetradecameric assembly of the association domain of Ca2+/calmodulin-dependent kinase II. Mol. Cell 11, 1241–1251. Hsia, K.C., Stavropoulos, P., Blobel, G., and Hoelz, A. (2007). Architecture of a coat for the nuclear pore membrane. Cell 131, 1313–1326. Jeanmougin, F., Thompson, J.D., Gouy, M., Higgins, D.G., and Gibson, T.J. (1998). Multiple sequence alignment with Clustal X. Trends Biochem. Sci. 23, 403–405. Jones, T.A., Zou, J.Y., Cowan, S.W., and Kjeldgaard, M. (1991). Improved methods for building protein models in electron density maps and the location of errors in these models. Acta Crystallogr. A 47, 110–119. King, M.C., Lusk, C.P., and Blobel, G. (2006). Karyopherin-mediated import of integral inner nuclear membrane proteins. Nature 442, 1003–1007. Krissinel, E., and Henrick, K. (2007). Inference of macromolecular assemblies from crystalline state. J. Mol. Biol. 372, 774–797. Laskowski, R.A., MacArthur, M.W., Moss, D.S., and Thornton, J.M. (1993). PROCHECK: A program to check the stereochemical quality of protein structures. J. Appl. Cryst. 26, 283–291. Lawrence, M.C., and Colman, P.M. (1993). Shape complementarity at protein/ protein interfaces. J. Mol. Biol. 234, 946–950.

Bru¨nger, A.T., Adams, P.D., Clore, G.M., DeLano, W.L., Gros, P., GrosseKunstleve, R.W., Jiang, J.S., Kuszewski, J., Nilges, M., Pannu, N.S., et al. (1998). Crystallography & NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr. D Biol. Crystallogr. 54, 905–921.

Lederkremer, G.Z., Cheng, Y., Petre, B.M., Vogan, E., Springer, S., Schekman, R., Walz, T., and Kirchhausen, T. (2001). Structure of the Sec23p/24p and Sec13p/31p complexes of COPII. Proc. Natl. Acad. Sci. USA 98, 10704– 10709.

CCP4. (1994). The CCP4 suite: Programs for protein crystallography. Acta Crystallogr. D Biol. Crystallogr. 50, 760–763.

Livnah, O., Stura, E.A., Middleton, S.A., Johnson, D.L., Jolliffe, L.K., and Wilson, I.A. (1999). Crystallographic evidence for preformed dimers of erythropoietin receptor before ligand activation. Science 283, 987–990.

Chook, Y.M., and Blobel, G. (2001). Karyopherins and nuclear import. Curr. Opin. Struct. Biol. 11, 703–715. Cook, A., Bono, F., Jinek, M., and Conti, E. (2007). Structural biology of nucleocytoplasmic transport. Annu. Rev. Biochem. 76, 647–671. Cronshaw, J.M., Krutchinsky, A.N., Zhang, W., Chait, B.T., and Matunis, M.J. (2002). Proteomic analysis of the mammalian nuclear pore complex. J. Cell Biol. 158, 915–927. Davis, I.W., Leaver-Fay, A., Chen, V.B., Block, J.N., Kapral, G.J., Wang, X., Murray, L.W., Arendall, W.B., III, Snoeyink, J., Richardson, J.S., and Richardson, D.C. (2007). MolProbity: All-atom contacts and structure validation for proteins and nucleic acids. Nucleic Acids Res. 35, W375–W383.

Loı¨odice, I., Alves, A., Rabut, G., Van Overbeek, M., Ellenberg, J., Sibarita, J.B., and Doye, V. (2004). The entire Nup107–160 complex, including three new members, is targeted as one entity to kinetochores in mitosis. Mol. Biol. Cell 15, 3333–3344. Lutzmann, M., Kunze, R., Buerer, A., Aebi, U., and Hurt, E. (2002). Modular selfassembly of a Y-shaped multiprotein complex from seven nucleoporins. EMBO J. 21, 387–397. McCoy, A.J., Grosse-Kunstleve, R.W., Adams, P.D., Winn, M.D., Storoni, L.C., and Read, R.J. (2007). Phaser crystallographic software. J. Appl. Cryst. 40, 658–674.

de La Fortelle, E., and Bricogne, G. (1997). Maximum-likelihood heavy-atom parameter refinement in the multiple isomorphous replacement and multiwavelength anomalous diffraction methods. Methods Enzymol. 276, 472–494.

a´k, I., Hoelz, A., and Blobel, G. (2007). Structure of Nup58/45 suggests Melc flexible nuclear pore diameter by intermolecular sliding. Science 315, 1729– 1732.

Devos, D., Dokudovskaya, S., Alber, F., Williams, R., Chait, B.T., Sali, A., and Rout, M.P. (2004). Components of coated vesicles and nuclear pore complexes share a common molecular architecture. PLoS Biol. 2, e380.

Napetschnig, J., Blobel, G., and Hoelz, A. (2007). Crystal structure of the N-terminal domain of the human protooncogene Nup214/CAN. Proc. Natl. Acad. Sci. USA 104, 1783–1788.

Dockendorff, T.C., Heath, C.V., Goldstein, A.L., Snay, C.A., and Cole, C.N. (1997). C-terminal truncations of the yeast nucleoporin Nup145p produce a rapid temperature-conditional mRNA export defect and alterations to nuclear structure. Mol. Cell. Biol. 17, 906–920.

Otwinowski, Z., and Minor, W. (1997). Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276, 307–326. Paoli, M. (2001). Protein folds propelled by diversity. Prog. Biophys. Mol. Biol. 76, 103–130.

Molecular Cell 32, 815–826, December 26, 2008 ª2008 Elsevier Inc. 825

Molecular Cell A Fence-like Coat for the Nuclear Pore Membrane

Pemberton, L.F., and Paschal, B.M. (2005). Mechanisms of receptor-mediated nuclear import and nuclear export. Traffic 6, 187–198. Reichelt, R., Holzenburg, A., Buhle, E.L., Jr., Jarnik, M., Engel, A., and Aebi, U. (1990). Correlation between structure and mass distribution of the nuclear pore complex and of distinct pore complex components. J. Cell Biol. 110, 883–894. Richardson, J.S. (1981). The anatomy and taxonomy of protein structure. Adv. Protein Chem. 34, 167–339. Rout, M.P., Aitchison, J.D., Suprapto, A., Hjertaas, K., Zhao, Y., and Chait, B.T. (2000). The yeast nuclear pore complex: Composition, architecture, and transport mechanism. J. Cell Biol. 148, 635–651. Salama, N.R., Chuang, J.S., and Schekman, R.W. (1997). Sec31 encodes an essential component of the COPII coat required for transport vesicle budding from the endoplasmic reticulum. Mol. Biol. Cell 8, 205–217. Schuck, P. (2000). Size-distribution analysis of macromolecules by sedimentation velocity ultracentrifugation and lamm equation modeling. Biophys. J. 78, 1606–1619. Siniossoglou, S., Wimmer, C., Rieger, M., Doye, V., Tekotte, H., Weise, C., Emig, S., Segref, A., and Hurt, E.C. (1996). A novel complex of nucleoporins, which includes Sec13p and a Sec13p homolog, is essential for normal nuclear pores. Cell 84, 265–275. Siniossoglou, S., Lutzmann, M., Santos-Rosa, H., Leonard, K., Mueller, S., Aebi, U., and Hurt, E. (2000). Structure and assembly of the Nup84p complex. J. Cell Biol. 149, 41–54.

Stagg, S.M., Gurkan, C., Fowler, D.M., LaPointe, P., Foss, T.R., Potter, C.S., Carragher, B., and Balch, W.E. (2006). Structure of the Sec13/31 COPII coat cage. Nature 439, 234–238. Stagg, S.M., LaPointe, P., and Balch, W.E. (2007). Structural design of cage and coat scaffolds that direct membrane traffic. Curr. Opin. Struct. Biol. 17, 221–228. Stagg, S.M., LaPointe, P., Razvi, A., Gurkan, C., Potter, C.S., Carragher, B., and Balch, W.E. (2008). Structural basis for cargo regulation of COPII coat assembly. Cell 134, 474–484. Suntharalingam, M., and Wente, S.R. (2003). Peering through the pore: Nuclear pore complex structure, assembly, and function. Dev. Cell 4, 775–789. Vasu, S., Shah, S., Orjalo, A., Park, M., Fischer, W.H., and Forbes, D.J. (2001). Novel vertebrate nucleoporins Nup133 and Nup160 play a role in mRNA export. J. Cell Biol. 155, 339–354. Walther, T.C., Alves, A., Pickersgill, H., Loiodice, I., Hetzer, M., Galy, V., Hulsmann, B.B., Kocher, T., Wilm, M., Allen, T., et al. (2003). The conserved Nup107–160 complex is critical for nuclear pore complex assembly. Cell 113, 195–206. Yang, Q., Rout, M.P., and Akey, C.W. (1998). Three-dimensional architecture of the isolated yeast nuclear pore complex: Functional and evolutionary implications. Mol. Cell 1, 223–234. Zhang, X., Gureasko, J., Shen, K., Cole, P.A., and Kuriyan, J. (2006). An allosteric mechanism for activation of the kinase domain of epidermal growth factor receptor. Cell 125, 1137–1149.

826 Molecular Cell 32, 815–826, December 26, 2008 ª2008 Elsevier Inc.