Structural Biology of Nucleocytoplasmic Transport

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26

ARI

1 May 2007

17:35

Structural Biology of Nucleocytoplasmic Transport Atlanta Cook,1 Fulvia Bono,1 Martin Jinek,1 and Elena Conti1,2 1

European Molecular Biology Laboratory, D-69117 Heidelberg, Germany; email: [email protected], [email protected], [email protected]

2

Max Planck Institute of Biochemistry, D-82152 Martinsried, Germany; email: [email protected]

Annu. Rev. Biochem. 2007. 76:647–71

Key Words

The Annual Review of Biochemistry is online at biochem.annualreviews.org

exportin, importin, karyopherin, nucleocytoplasmic transport, nuclear pore complex, RanGTP, TAP:p15

This article’s doi: 10.1146/annurev.biochem.76.052705.161529 c 2007 by Annual Reviews. Copyright  All rights reserved 0066-4154/07/0707-0647$20.00

Abstract In eukaryotic cells, segregation of DNA replication and RNA biogenesis in the nucleus and protein synthesis in the cytoplasm poses the requirement of transporting thousands of macromolecules between the two cellular compartments. Transport between nucleus and cytoplasm is mediated by soluble receptors that recognize specific cargoes and carry them through the nuclear pore complex (NPC), the sole gateway between the two compartments at interphase. Nucleocytoplasmic transport is specific not only in terms of cargo recognition, but also in terms of directionality, with nuclear proteins imported into the nucleus and RNAs exported from it. How is directionality achieved? How can the receptors be both specific and versatile in recognizing a multitude of cargoes? And how can their interaction with NPCs allow fast translocation? We describe the molecular mechanisms underlying nucleocytoplasmic transport as they have been revealed by structural studies of the receptors and regulators in different steps of transport cycles.

647

ANRV313-BI76-26

ARI

1 May 2007

17:35

Contents

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

INTRODUCTION . . . . . . . . . . . . . . . . . THE KARYOPHERIN PROTEIN FAMILY . . . . . . . . . . . . . . . . . . . . . . . . . NUCLEAR IMPORT . . . . . . . . . . . . . . . Importin β . . . . . . . . . . . . . . . . . . . . . . . Cargo Recognition by Importin β . Importin β Adaptors: Classical NLS Recognition by Importin α . . . . . . . . . . . . . . . . . . . . Importin β Adaptors: m3 G Cap Recognition by Snurportin1 . . . Transportin: Binding of M9-Containing Cargo . . . . . . . . . RanGTP Binding to Importin β and Cargo Dissociation . . . . . . . . RanGTP-Mediated Cargo Dissociation from Transportin . Disruption of NLS:Importin α Complexes . . . . . . . . . . . . . . . . . . . . NUCLEAR EXPORT . . . . . . . . . . . . . . Nuclear Export by Cse1 . . . . . . . . . . The Exportin Crm1 . . . . . . . . . . . . . . THE RAN CYCLE . . . . . . . . . . . . . . . . . RanGDP: The Off State . . . . . . . . . . RanGAP-Catalyzed GTP Hydrolysis . . . . . . . . . . . . . . . . . . . . RCC1-Mediated Nucleotide Exchange . . . . . . . . . . . . . . . . . . . . . Recycling RanGDP to the Nucleus: NTF2 . . . . . . . . . . . . . . . NUCLEOPORINS . . . . . . . . . . . . . . . . . mRNA EXPORT . . . . . . . . . . . . . . . . . . . The TAP:p15 mRNA Export Factor: Nucleoporin Binding. . . The TAP:p15 mRNA Export Factor: Cargo Binding . . . . . . . . . Adaptors and Regulators in the mRNA Export Pathway . . . . . . . .

648 650 651 651 651

653 655 655 655 656 657 657 657 658 659 659 659 661 662 662 663 663 663 665

INTRODUCTION NPC: nuclear pore complex

648

In eukaryotic cells, DNA replication and RNA transcription are spatially separated from protein synthesis by a double membrane, the nuCook et al.

clear envelope. Consequently, thousands of macromolecules are transferred between the nuclear and cytosolic compartments at interphase (reviewed in References 1–7). RNA transcripts are exported from the nucleus to reach the ribosomes in the cytoplasm. Conversely, proteins, such as histones, polymerases, and transcription factors, are imported into the nucleus to reach the genetic material. Nucleocytoplasmic transport occurs through the nuclear pore complex (NPC), a large assembly that spans the nuclear envelope (reviewed in References 8–10). The NPCs are impermeable to most macromolecules, with the notable exception of nucleocytoplasmic transport factors. These proteins can shuttle through the NPCs, carrying other macromolecules across. The majority of nucleocytoplasmic transport factors belong to the family of karyopherin β proteins, also known as importin β-like proteins after the first receptor identified (importin β) (reviewed in References 11 and 12). β-karyopherins that mediate nuclear import are generally known as importins, whereas those mediating nuclear export are known as exportins. Importins associate with their macromolecular cargo in the cytoplasm, either directly or indirectly via adaptor proteins. They dock to components of the NPC (known collectively as nucleoporins), translocate to the opposite side of the nuclear envelope, and release their cargo there (Figure 1). Cargo release is achieved by association of the importins with the GTPase Ran in the GTP-bound form (RanGTP). Nuclear export is essentially the reciprocal process, with cargo recognition occurring in the nucleus in the presence of RanGTP and cargo complexes dissociating in the cytoplasm upon GTP hydrolysis (Figure 1). In contrast to RanGTP, the GDP-bound form of Ran (RanGDP) is unable to bind to βkaryopherins. Ran is present at high concentrations in the nucleus in the GTP-bound form, and mainly in the GDP-bound form in the cytoplasm (reviewed in References 13 and

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26

ARI

1 May 2007

17:35

Figure 1 Schematic drawing of nuclear import and export processes mediated by β-karyopherins (also known as β-like importins and exportins). The two nuclear pore complexes (NPCs) show the characteristic nuclear basket and cytoplasmic fibrils. (bottom left) An importin binds a cargo in the cytoplasm and releases it upon binding RanGTP in the nucleus (upper left). (upper right) An exportin binds both cargo and RanGTP in the nucleus and releases them upon conversion of RanGTP into RanGDP (bottom right).

14). This compartmentalization depends on the localization of the proteins that regulate the nucleotide state of Ran (Figure 2). GTP hydrolysis requires a Ran GTPase-activating protein (RanGAP) that is present in the cytoplasm. Conversely, the exchange of GTP to GDP is catalyzed by a guanine nucleotide exchange factor (GEF) (known as regulator of chromosome condensation 1 or RCC1) that is bound to chromatin in the nucleus. Additional factors contribute to the Ran cycle. RanGAP needs cofactors (the Ran-binding proteins, or RanBPs) to act on the karyopherin-bound complexes that reach the cytoplasmic side of the nuclear envelope. Following hydrolysis, RanGDP is recycled back to the nucleus by a dedicated transport factor (nuclear transport factor 2

or NTF2) that bears no resemblance to the karyopherins (reviewed in Reference 15). With the determination of the structures of different transport factors and their regulators (recapitulated in Reference 16), we now understand at the atomic level the molecular mechanisms that underlie crucial steps in nucleocytoplasmic transport. In this review, we first describe how a set of β-karyopherins recognizes their cargoes and dissociates from them, how they dock to nucleoporins, and how the regulatory state of Ran is maintained. We then describe the current structural understanding of the messenger RNA (mRNA) export pathway, which depends on an altogether different transport factor (TAP:p15) and which is, for the most part, still uncharted territory. www.annualreviews.org • Nucleocytoplasmic Transport

Karyopherin: a nuclear transport receptor belonging to the importin β family of HEAT repeat proteins Importin: a β-karyopherin that carries cargo into the nucleus Exportin: a β-karyopherin that carries cargo out of the nucleus TAP:p15: the mRNA export factor

649

ARI

1 May 2007

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26

17:35

Figure 2 Schematic drawing of the Ran cycle. The high concentration of RanGDP (GDP-bound form of Ran) in the cytosol is maintained by RanGAP (GTPase-activating protein), which is bound to the cytoplasmic fibrils of the nuclear pore complex. With the help of other factors, it acts on the RanGTP (GTP-bound form of Ran) that enters the cytoplasm (via binding to exportins and importins). The high concentration of RanGTP in the nucleus is maintained by regulator of chromosome condensation 1 (RCC1), a chromatin-bound guanine exchange factor (RanGEF), which acts on the RanGDP that enters the nucleus [with its dedicated transport factor nuclear transport factor 2 (NTF2)].

THE KARYOPHERIN PROTEIN FAMILY

GEF: guanidine exchange factor RanGTP: the small GTPase Ran in its GTP-bound form RanGDP: the small GTPase Ran in its GDP-bound form

650

The family of β-karyopherins includes 14 members in budding yeast and at least 20 in humans (reviewed in References 11 and 12). β–karyopherins are relatively large proteins (around 100 kDa in molecular weight). They share weak sequence homology overall, with identity typically between 15% and Cook et al.

20%. The sequence similarity is strongest in the N-terminal half, where RanGTP binds, thus reflecting the functional similarity of the members of this family. β-karyopherins are HEAT-repeat proteins, as they are made up of tandem repeats of an ∼40 amino acid motif (17) that was first identified in Huntingtin, elongation factor 3, PR65/A subunit of protein phosphatase 2A and the TOR

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26

ARI

1 May 2007

17:35

lipid kinase. The consensus for HEAT motifs is rather degenerate, making it difficult to predict the exact number of repeats and their boundaries in a given sequence. The members of the β-karyopherin family with known crystal structures (the import factors importin β and transportin, and the export factor Cse1) have 19 to 20 HEAT repeats that are arranged in tandem, both in sequence and in structure. Each HEAT motif folds into a pair of αhelices (known as the A and B helices). Consecutive HEAT motifs stack together in a parallel fashion, usually with a slight clockwise twist. In β-karyopherins, this produces molecules with an overall superhelical architecture. The inner concave surface of the βkaryopherin superhelix is formed by the B helices, whereas the A helices form the outer convex surface. The β-karyopherin superhelices have an intrinsic flexibility that appears to arise from the modular architecture of their structure and to be important for the versatility of these molecules in recognizing different cargoes (18). For convenience, we refer to the superhelical structures of these proteins in terms of an N-terminal arch (HEAT repeats 1 to 8) and a C-terminal arch (HEAT repeats 9 to the C terminus).

NUCLEAR IMPORT Importin β Importin β (also known as karyopherin β1 or Kap95 in yeast) is perhaps the most extensively studied member of the β-karyopherin family (reviewed in Reference 19). It was first identified as the transport factor for proteins carrying classical nuclear localization signals (NLSs), such as those of nucleoplasmin or the SV40 T antigen (20–23). Importin β does not bind these classical NLSs directly, but binds importin α, which in turn binds the NLS (Figure 3a). Importin α (also known as karyopherin α or Kap60 in yeast) is thus an adaptor, with an importin β-binding (IBB) domain at the N terminus of the protein

connected to an NLS-binding domain. Although this is considered as the classical import pathway, the use of adaptors to bind cargoes seems to be the exception rather than the rule (Figure 3b). Most β-karyopherins bind their cargoes directly, and even importin β is able to recognize cargo substrates without the need for an adaptor, as in the case of the sterol regulatory element-binding protein 2 (SREBP-2) or the parathyroid hormonerelated protein (PTHrP).

Cargo Recognition by Importin β The three structures of cargo-bound importin β known to date (with the IBB domain of importin α, with SREBP-2, and with PTHrP) (24–26) show that the cargoes bind at the inner concave surface of the superhelix. The IBB domain of importin α binds mainly as a long helix to the C-terminal arch of importin β, at the B helices of HEAT repeats 7 to 19 (24). Importin β curls around the IBB helix in a tight spiral to form a structure reminiscent of a snail shell (Figures 3b and 4). A small portion of the IBB domain is in an extended conformation and binds to a loop from HEAT repeat 8 containing several conserved acidic residues, known as the acidic loop. SREBP-2 also binds between HEAT 7 and 19, although the detailed interactions are different to those that hold the IBB domain (26). SREBP-2 binds as a dimer of the helix-loophelix zipper (HLHZ) domain. The HLHZ is enclosed by two extended helices from HEAT repeats 7 and 17, holding the protein like a pair of “molecular chopsticks.” Whereas the positively charged IBB domain makes largely electrostatic interactions with acidic residues of importin β, SREBP-2 makes mostly hydrophobic contacts with aromatic residues of importin β. The superhelical conformation of importin β differs in its complexes with the IBB domain or SREBP2, reflecting the intrinsic plasticity of the transport factor that allows it to mold around different molecular shapes (schematized in Figure 3b). www.annualreviews.org • Nucleocytoplasmic Transport

RanGAP: GTPase-activating protein of Ran RanBP: Ran-binding protein HEAT repeat: a helical repeat motif identified in Huntingtin, elongation factor 3, PR65/A subunit of protein phosphatase 2A and the TOR lipid kinase NLS: nuclear localization signal

651

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26 ARI 1 May 2007

652

17:35

Cook et al.

ANRV313-BI76-26

ARI

1 May 2007

17:35

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

Finally, the structure of an importin β fragment (encompassing HEAT repeats 1–11) with PTHrP shows that this cargo binds the inner surface of the karyopherin at an altogether different site, within the N-terminal arch. PTHrP does not bind as a folded molecule but contains a linear import signal recognized in an extended conformation, its main chain hydrogen bonded to an array of asparagine and glutamine residues of importin β (25). These interactions are reminiscent of NLS recognition by importin α (see below).

Importin β Adaptors: Classical NLS Recognition by Importin α Importin α contains a 50-kDa NLS-binding domain downstream of the IBB domain (Figure 3a). The NLS-binding domain is made up of 10 armadillo (ARM) repeats (Figure 4). The ARM repeat is an ∼40-amino acid motif consisting of three α-helices (H1, H2, and H3) (17). The repeats stack together in tandem, with an approximate 30◦ rotation between each motif to produce an elongated molecule with a superhelical twist. The H3 helices form the inner concave surface of the molecule in much the same way as the B helices of the HEAT motifs in importin β. The inner surface of importin α is the site of NLS recognition (27) (Figure 4). Classical NLSs can be monopartite, such as the SV40 T antigen NLS, or bipartite, such

as the nucleoplasmin NLS. A monopartite NLS consists of a cluster of three to five positively charged residues. A bipartite NLS has an additional smaller cluster of lysine/arginine residues, separated from the monopartite-like cluster by a linker of 10 to 12 residues. Structures of both the mouse and yeast homologues of importin α have been solved with a variety NLSs and show a high degree of conservation of the recognition mechanism (28, 29). NLSs bind in an extended conformation, with the polypeptide backbone making hydrogen bonds with a series of conserved asparagines on the importin α H3 helices. The side chains of the NLS peptides insert into a series of pockets formed by an array of conserved tryptophan residues that also protrude from the H3 helices. The tryptophans engage in hydrophobic interactions with the aliphatic part of the lysine side chains of the NLS. The positively charged tips of the NLS lysines interact electrostatically with surrounding acidic residues of importin α. There are two clusters of lysine-binding pockets distributed on the surface of importin α (making a larger binding site and a smaller binding site), mirroring the distribution of positively charged residues in the NLS. The distance between these two binding sites explains the requirement for a linker region in bipartite signals of at least 10 residues, the minimum number required to span the distance between the two sites in an extended conformation.

ARM repeat: a helical repeat motif also found in armadillo proteins such as β-catenin

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−− Figure 3 Schematic drawing of import pathways. (a) The classical nuclear localization signal (NLS) transport cycle. (bottom left) The NLS present in the cargo to be imported is recognized by the adaptor molecule importin α (impα) via its armadillo-repeat domain (ARM). The importin β-binding (IBB) domain of importin α binds in a helical conformation to importin β (impβ). The cargo:impα:impβ complex is transported to the nucleus via the nuclear pore complex (NPC), where it is dissociated upon the formation of importin β:RanGTP complex. In the absence of impβ, the IBB domain of impα is autoinhibitory because it binds at the NLS-binding site (upper left). (upper right) The adaptor impα is recycled back by a dedicated receptor, the exportin Cse1/CAS. Upon hydrolysis of RanGTP to RanGDP, impα is released in the cytoplasm, ready for another round of NLS import. (b) Schematic representation of import receptors highlight the different shapes adopted on binding to different cargoes. The drawings are based on crystal structures and show impβ upon impα-mediated NLS binding, snurportin-mediated m3 G-cap binding, and direct binding of the transcription factor SREBP-2. Another karyopherin, transportin 1 (tpn1), is bound to an M9-containing cargo. The adaptors are shown in pink. www.annualreviews.org • Nucleocytoplasmic Transport

653

ARI

1 May 2007

17:35

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26

Figure 4 Structures in nuclear import pathways. (a) Importin β [impβ, Protein Data Bank (PDB) code 2bku] is shown bound to RanGTP and to the importin β-binding domain of the adaptor impα (PDB code 1qgk). Transportin/karyopherin β2 (tpn1, PDB code 1qbk) is shown bound to RanGTP and to the M9 signal of the cargo hnRNPA1 (PDB code 2h4m). (b) The structure of the armadillo repeat domain (ARM) of the adaptor impα (PDB code 1bk6) is shown bound to the nuclear localization signal (NLS) of the SV40 T antigen, and the m3 G cap-binding domain of the adaptor snurportin (PDB code 1xk5) is shown bound to its signal.

654

Cook et al.

ANRV313-BI76-26

ARI

1 May 2007

17:35

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

Importin β Adaptors: m3 G Cap Recognition by Snurportin1 UsnRNPs (splicosomal ribonucleoproteins containing the UsnRNAs) undergo a maturation cycle in the cytoplasm that results in the incorporation of an m3 G-cap on the RNA (reviewed in Reference 30). This cap acts as an NLS that is specifically recognized by snurportin1, allowing it to distinguish between mature UsnRNPs and immature UsnRNPs with an m7 G-cap (31). Snurportin1 mediates nuclear import of spliceosomal UsnRNPs in an importin β-dependent manner (32). It interacts directly with importin β via an N-terminal IBB-like domain (33). The structure of an m3 G cap analogue bound to the C-terminal domain of snurportin shows that the cap moiety is sandwiched by two hydrophobic residues in the cap-binding pocket (34) (Figure 4). This provides a hydrophobic environment that perfectly fits a trimethylated guanosine. Binding of the smaller monomethylated cap would require energetically unfavorable dehydration, providing an understanding of how untimely import of incorrectly assembled UsnRNPs might be prevented.

Transportin: Binding of M9-Containing Cargo Transportin 1 (also known as karyopherin-β2 in yeast) is an importin that recognizes signal sequences present in many mRNA-binding proteins. These proteins must be imported into the nucleus to mediate the nuclear export of mRNA. The best-known NLS recognized by transportin is the M9 sequence of the mRNA-binding protein hnRNPA1 (35, 36). The M9 NLS is 38 amino acid residues long and has an overall positive charge. The structure of the transportin:M9 complex shows that the NLS binds as a linear extended peptide to the C-terminal arch of transportin (37) (Figure 3b and 4). As with cargo binding to importin β, recognition of the M9 NLS occurs at the in-

ner concave surface of transportin, but the overall shape of the superhelix is quite different. The two arches of transportin have a distinct kink, so that instead of the tight snaillike conformation observed in the importin β:IBB complex, transportin has a rather open Z-like structure (Figure 4). Binding of the M9 NLS involves general electrostatic complementarity with negatively charged residues of transportin, similar to IBB recognition by importin β. Two motifs in the M9 peptide form additional crucial interactions. The first is a proline-tyrosine motif that is found toward the C terminus of the NLS and is usually preceded by a basic residue two to five amino acids upstream. The second is a hydrophobic motif that contains a glycine (or another small residue) and is found toward the N terminus of the NLS. These structural constraints have allowed Chook and colleagues (37) to identify a large number of potential transportindependent import cargoes, whose prediction had been hampered by the low overall sequence similarity to the M9 NLS.

RanGTP Binding to Importin β and Cargo Dissociation In the nucleus, the binding of RanGTP to the β-karyopherins leads to cargo dissociation (Figure 1). Ran is a 24-kDa protein belonging to the Ras superfamily of small GTPases and contains the characteristic guanine nucleotide-binding domain (G domain). The G domain features two loop regions known as switch I and switch II, so named because they undergo marked conformational rearrangements upon switching between the GTP- and GDP-bound states (38). In addition, it has a C-terminal extension of ∼40 amino acids that is conserved in all Ran orthologues. The structure of RanGTP bound to importin β (39, 40) shows that the G domain of Ran binds at the N-terminal arch, fitting neatly within the inner concave surface of the superhelix (Figure 4). Ran binds at three separate sites. The B helices of the first three HEAT repeats contact the Ran switch regions. www.annualreviews.org • Nucleocytoplasmic Transport

655

ARI

1 May 2007

17:35

A second contact is made in the middle portion of the karyopherin molecule, where the acidic loop at HEAT repeat 8 interacts with a patch of basic residues on the surface of the G domain. The B helices of HEAT repeats 12 to 15 in the C-terminal arch provide a third contact area. The C-terminal extension of Ran is not involved in binding importin β and is detached from the G domain in all the RanGTP structures known to date. How does RanGTP binding cause the dissociation of cargoes such as the IBB domain of importin α, SREBP-2, or PTHrP? Although the different functional states have been studied by crystallizing proteins from different species (yeast and mammalian proteins), the structural results can be compared because of the high level of sequence similarity between these orthologues. In the case of PTHrP, the binding site for this cargo at the N-terminal arch overlaps almost entirely with that of Ran. Binding of the two proteins onto the karyopherin is therefore mutually exclusive. The IBB and SREBP-2 binding sites also have considerable overlap with the RanGTP interaction site at the conserved acidic loop within HEAT 8. Thus, the binding of RanGTP interferes with the cargo by steric interference, particularly in the middle of the molecule (Figure 4). In addition, importin β has a longer helical pitch (i.e., a more open conformation) in the RanGTP complex as compared with the tightly curled conformation that is observed in the IBB complex. Therefore, binding of RanGTP induces a conformational change in the karyopherin superhelix that may be incompatible with optimal cargo binding and might therefore further promote its release.

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26

RanGTP-Mediated Cargo Dissociation from Transportin The overall mode of RanGTP recognition in transportin is very similar to that observed in importin β, perhaps not surprisingly given that the greatest sequence similarity between these two proteins is in the Ran-binding re656

Cook et al.

gion. Although the details of the interactions vary, RanGTP also binds transportin at the concave surface of the N-terminal arch (41) (Figure 4). The switch I and switch II regions of Ran dock to the first three HEAT repeats of transportin, and the basic patch of the Ran G domain docks to a long loop with acidic residues at HEAT 8. The acidic loop of transportin is also conserved across species but is much longer than the corresponding loop in importin β (65 residues instead of 12). In contrast to importin β, the acidic loop of transportin does not contribute to binding the NLS. A mutant of transportin in which the loop has been removed is still able to bind the M9 NLS, and indeed removal of this flexible feature on the surface of the molecule was necessary to crystallize the M9:transportin complex (37). However, the loop contributes to cargo dissociation because its removal allows transportin to bind the M9 NLS and RanGTP simultaneously (42). What is the mechanism of cargo dissociation from transportin? In the RanGTP-bound structure, the loop extends across to the Cterminal arch, occupying much of the surface that binds the NLS in the M9-bound structure (Figure 4). The interaction of the loop with the NLS-binding site likely depends on RanGTP. In the absence of RanGTP, the interaction of the acidic loop with the negatively charged surface of the NLS-binding site would not be electrostatically favorable, allowing the basic M9-containing cargo to interact. In the presence of RanGTP, the interaction of the basic surface patch of Ran with the acidic loop may alter the electrostatic potential of the loop region, allowing it to occlude the NLS-binding site and expel the cargo. Thus, cargo dissociation is probably governed by structural rearrangement of the acidic loop. Overall conformational rearrangements do not appear to play a significant role in transportin because the conformational changes of the superhelix in the RanGTP or M9-bound states are not as dramatic as those observed in importin β (Figure 4).

ANRV313-BI76-26

ARI

1 May 2007

17:35

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

Disruption of NLS:Importin α Complexes The mode of RanGTP binding explains the release of cargoes that are bound directly to the β-karyopherin. However, it does not necessarily explain NLS release in adaptor-mediated nuclear import (reviewed in Reference 43). For example, RanGTP binding to importin β releases the IBB domain of importin α (Figure 3a), but how does the NLS dissociate from the ARM domain of importin α? One possible mechanism is that in the absence of importin β, the IBB domain of importin α has a conformation that is unfavorable for NLS binding. The structure of full-length importin α (44) shows that the IBB domain is largely disordered in the absence of importin β. However, a nine-residue segment of the IBB domain binds in the pockets at the larger NLS-binding site. This leads to an autoinhibited conformation of importin α, explaining the observation that NLS peptides have a higher affinity for importin α in the absence of the IBB sequence (45). Other factors are proposed to affect NLS release such as the nucleoporin, Nup2 in yeast, and the vertebrate orthologue, Nup50 (46– 50). These proteins localize to the nuclear side of the NPC and share high sequence identity at the N terminus, where importin α binds. The structure of the N-terminal region of Nup50 in complex with importin α shows that the nucleoporin binds over an extended surface on the ARM-repeat domain, including the minor NLS-binding site (51). Thus, Nup2 and Nup50 may accelerate the release of NLS peptides (at least the bipartite ones) through steric interference with the NLS-binding site. A third contribution to NLS release from importin α is the mechanism of its recycling to the cytoplasm by Cse1 (see below).

NUCLEAR EXPORT Nuclear Export by Cse1 Yeast Cse1 is the only exportin for which complete structural information is available at

present. Cse1 (known as CAS in humans) has only one known export cargo, importin α, but is nevertheless essential for cell survival (52, 53). Cse1 binds importin α with high affinity in the presence of RanGTP in the nucleus and releases it on entry to the cytoplasm (52, 53) (Figure 3a). The structures of Cse1 in both the cargo-bound (nuclear) form and in the unbound (cytosolic) form have been determined (Figure 5), providing insights into the mechanisms of cargo association and dissociation (54, 55). Cse1 is composed of 19 consecutive HEAT repeats. In the ternary export complex, RanGTP binds at the N-terminal arch of the superhelix, somewhat similarly to the importin complexes. Importin α contacts both the N-terminal and C-terminal arches of Cse1 (Figure 5). Importin α also forms extensive interactions with RanGTP, displacing it from the plane of the N-terminal arch in Cse1 as compared to the importin:Ran complexes. These interactions partly explain the mutual need of Ran and cargo in the formation of the export complex. Importin α does not present a linear “export signal” to Cse1 but is recognized at multiple sites on the ARM repeat domain. The C-terminal part of the ARM repeat contacts an insertion at HEAT 8 of Cse1. In contrast to the importins, the insertion at HEAT 8 of Cse1 does not have significant acidic character, folds into two α-helices instead of a loop, and is not involved in binding the positively charged surface of RanGTP. The Cse1 insertion is nevertheless conserved across species and essential for complex formation (54). The IBB domain of importin α also makes extensive contacts. The IBB domain is folded back across the NLS-binding site, resembling the autoinhibited conformation observed in the structure of full-length importin α (44). Recycling of importin α to the cytoplasm thus can only occur when the NLS has been released, preventing futile transport cycles. The unbound form of Cse1 has a more closed conformation as compared with the bound state (54) (Figure 5). The N-terminal and C-terminal arches of Cse1 interact www.annualreviews.org • Nucleocytoplasmic Transport

657

ARI

1 May 2007

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26

17:35

Figure 5 Structures in a nuclear export pathway. (a) The exportin Cse1 in the nuclear state in its complex with RanGTP and importin α (impα, PDB code 1wa5) and (b) Cse1 in the cytosolic state (i.e., unbound, PDB code 1z3h).

intramolecularly, with HEAT repeats 1–3 contacting repeats 14–16. This forms a ringlike structure that is not compatible with binding of either RanGTP or importin α. The displacement of the N-terminal arch away from the C-terminal arch is required for binding of the export cargo, and this conformation is likely induced and/or stabilized by RanGTP. Different conformations and accessibility of binding sites in the presence and absence of RanGTP thus play an important role in governing the binding affinities for importin α to importin β and Cse1 in the cytoplasmic and nuclear compartments.

The Exportin Crm1

NES: nuclear export signal

658

One of the most studied nucleocytoplasmic transport factors is the exportin Crm1 (also known as Xpo1 in yeast or exportin-1). In humans, Crm1-dependent export is sensitive to leptomycin B, a small molecule that covalently attaches to a cysteine residue and interferes with cargo binding (56, 57). Crm1 cargoes generally contain a leucine-rich nuclear exCook et al.

port signal (NES) (58–60). However NESs are more difficult to identify correctly compared with classical NLSs because they share sequence similarity to regions that form the hydrophobic cores of many proteins. The present structural data on Crm1 come from a combination of low- and highresolution techniques. The crystal structure of a C-terminal fragment of Crm1 has provided the atomic model for a region of the protein that is predicted to encompass HEAT repeats 14 to 19 (61). By fitting this structure into an electron microscopy map of the full-length ¨ protein, Muller and colleagues (61) have modeled the remaining part of the molecule. The results suggest that Crm1 has a ring-like structure similar to that observed for Cse1 in the unbound state. The modeling also identifies a region that is predicted to be a loop in the eighth HEAT repeat and that is proposed to regulate cargo binding by an allosteric mechanism (61). Understanding the determinants of cargo binding by Crm1 and whether it undergoes conformational regulation will ultimately require the structure of an export

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26

ARI

1 May 2007

17:35

complex. These studies have so far been hampered because most NESs bind with low affinities and need additional adaptor proteins (62, 63). The structural data on human Crm1 are consistent with results from small-angle X-ray scattering (SAXS), showing that the yeast orthologue of Crm1 has a similar compact shape in the unbound conformation (64). Given this similarity, it is tempting to speculate that the unbound, closed conformation and its opening upon cargo binding might be a general feature of exportins. However, SAXS studies of the export factor for tRNAs (Xpo-t) indicate that at least at low resolution Xpo-t is more similar to importin β than it is to the exportin Cse1 (64). Thus, there is no simple rule of conformational regulation that might distinguish an exportin from an importin, and indeed, certain karyopherins act as importins for one type of cargo and exportins for another (65, 66).

THE RAN CYCLE RanGDP: The Off State The dissociation of export complexes in the cytoplasm occurs upon hydrolysis of RanGTP to RanGDP (Figure 2). Similarly, the importin:RanGTP complexes that are recycled back to the cytoplasm are dissociated by GTP hydrolysis, allowing the importins to start a new round of cargo recognition and import. In contrast to RanGTP, RanGDP does not bind to karyopherins. The dissociation constant for the complex between RanGTP and importin β, for example, is subnanomolar (67), whereas the Kd for RanGDP is about 10 μM (39). Upon the GTP to GDP transition, Ran undergoes significant conformational changes. The structure of Ran in the GDPbound state (68) shows that the switch regions are displaced from the conformation that holds them near the γ-phosphate of the nucleotide in the GTP-bound form (Figure 6a). This conformational change is similar to that

observed with other members of the Ras family (38). The second and perhaps most dramatic change involves remodeling of the C-terminal extension of Ran. This region is disordered in the RanGTP complex but folds back against the G domain as an αhelix in the GDP form, with the conserved DEDDDL motif (which is still disordered in the RanGDP structure) likely packing against a conserved basic patch that is found on the surface of Ran (69). This conformation of the C-terminal helix provides a steric barrier that prevents RanGDP from binding to the β-karyopherins. The C-terminal extension of Ran seems to be a defining feature of the nucleotide state of Ran because a C-terminal truncation mutant of Ran is able to bind to βkaryopherins even in the GDP-bound form (70).

RanGAP-Catalyzed GTP Hydrolysis Hydrolysis of GTP to GDP in Ran is promoted by proteins that are localized to the cytoplasmic compartment (Figure 2). RanBP1 and RanBP2 promote dissociation of RanGTP from the karyopherins as they exit the NPC and stimulate RanGAP-mediated hydrolysis (71, 72). RanBP1 is a 23-kDa cytoplasmic protein that contains a single Ran-binding domain (RanBD) (73), whereas RanBP2 (also known as Nup358) is a component of the cytoplasmic fibrils of the NPC and contains four RanBDs (74). The structure of a complex of RanGTP with the first RanBD of RanBP2 reveals that the RanBD has a pleckstrin homology (PH) domain fold (69). The PH domain of RanBD binds to the switch I region of the G domain, and an extended region at the N terminus of the RanBD extends across the surface of the G domain. The C-terminal region of RanGTP also wraps around the RanBD, making extensive contacts in a molecular “embrace.” It has been observed that importin β, RanGTP, and RanBP1 form a stable ternary complex (67, 75, 76), which may represent an intermediate in the disassembly of karyopherin:RanGTP www.annualreviews.org • Nucleocytoplasmic Transport

659

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26 ARI 1 May 2007

660

17:35

Cook et al.

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26

ARI

1 May 2007

17:35

complexes on the cytoplasmic side of the NPC. Indeed, the binding site for the RanBD on RanGTP does not overlap with that of importin β. Furthermore, the unique C terminus of Ran extends into solution in karyopherin:RanGTP complexes (39, 41), suggesting how it can be accessed by the RanBD. As with other small GTPases like Ras, the intrinsic rate of GTP hydrolysis by Ran is very low (kcat = 1.8 × 10−5 s−1 ) (77). RanGAP stimulates the GTPase activity of Ran by several orders of magnitude to approximately 2–10 s−1 (77). RanGAP proteins from different species share an N-terminal catalytic (GAP) domain (78). The structure of the GAP domain forms a crescent-shaped arch consisting of 11 leucine-rich repeats (LRR) (79). RanGAP proteins of higher eukaryotes contain an additional conserved C-terminal domain of approximately 230 amino acids. The C-terminal domain is covalently linked to the ubiquitin-like protein SUMO-1. This modification localizes RanGAP to the cytoplasmic face of the NPC through an interaction with RanBP2 (80, 81). In the structure of the RanGAP:RanBP1: RanGTP ternary complex, Ran binds to the edge of the RanGAP arch (82) (Figure 6b). RanGTP interacts with 7 of the 11 LRRs, making numerous ionic interactions with residues in the intrarepeat loops. A protruding loop in the third LRR acts as a “footrest,” stabilizing the switch II region of Ran. In

other small GTPases, such as Ras, an activating arginine residue is supplied in trans by the GAP protein to stabilize the transition state of the hydrolysis reaction (reviewed in Reference 83). The structure of the RanGAP:RanGTP:RanBP1 complex reveals that there is no such “arginine finger” at the active site (82). Instead, RanGAP appears to enhance the GTPase activity of Ran by stabilizing the switch II region and orienting the catalytic glutamine of Ran. Superposition of the structure of the ternary RanGAP:RanBP1:Ran complex with karyopherin:RanGTP complexes shows that binding of karyopherins and RanGAP to RanGTP is mutually exclusive and explains previous observations that karyopherins protect RanGTP from both RanGAP-mediated GTP hydrolysis and RCC1-mediated nucleotide exchange (67, 84).

LRR: leucine-rich repeat

RCC1-Mediated Nucleotide Exchange G proteins bind guanine nucleotides with high affinity and release them very slowly. GEFs increase the rate of nucleotide dissociation by several orders of magnitude, allowing rapid GDP release after GTP hydrolysis and reloading of the G protein with GTP (38) (Figure 2). The RanGEF activity is provided by the nuclear protein RCC1 in metazoans (85, 86) and its homologue Prp20 in yeast

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−− Figure 6 Structures in the Ran cycle. In all panels, GDP and GTP molecules are shown in stick format with carbon atoms colored yellow, and magnesium ions are shown as purple spheres. (a) Comparison of RanGDP (left) (coordinates kindly provided by K. Scheffzek) and RanGTP as observed in the RanGTP:RanBD1 complex (right) (PDB code 1rrp). The GDP- and GTP-bound conformations of switch I and II regions and the C-terminal extension are shown in different colors. (b) Structure of the RanGAP:RanGTP:RanBP1 complex (PDB code 1k5d). The binding of RanGTP to RanGAP orients the catalytic glutamine (Gln69, shown in stick format) toward the γ-phosphate of the GTP. (c) Structure of nucleotide-free Ran bound to RCC1 (PDB code 1i2m). A sulfate ion (shown in stick format) is bound to the P loop. RCC1 inserts a β-hairpin (the β-wedge) into the active site of Ran, distorting the switch II region and the P loop. (d) RanGDP binding to the recycling factor NTF2 (PDB code 1a2k). Phe72 from the switch II region of Ran (shown in stick format) is inserted into the hydrophobic cavity of the NTF2 barrel. Abbreviations: NTF2, nuclear transport factor 2; PDB, Protein Data Bank; RanBP, Ran-binding protein; RanGAP, Ran GTPase-activating protein; RanGDP, GDP-bound form of Ran; RanGTP, GTP-bound form of Ran; RCC1, regulator of chromosome condensation 1. www.annualreviews.org • Nucleocytoplasmic Transport

661

ARI

1 May 2007

17:35

(87, 88). RCC1 localizes to the nucleus by virtue of its association with chromatin via the core histones H2A and H2B (89). RCC1 is an ∼45-kDa protein with a sevenbladed β-propeller fold (90). The structure of RCC1 in complex with nucleotide-free Ran (91) reveals that Ran binds to one face of the β-propeller, in accordance with mutagenesis studies (92) (Figure 6c). In the complex, the C-terminal extension of Ran is already in the triphosphate conformation, detached from the G domain and disordered. This suggests that RCC1 might play an active role in inducing the C-terminal switch in Ran. A key element of the RCC1-Ran interface is a protruding β-hairpin insertion in the third propeller blade of RCC1. This hairpin acts as a “β-wedge,” inserting between the switch II region and the P loop of Ran. This forces the switch II region to adopt a conformation different from those in the GTP- and GDPbound states and displaces residues that contribute to Mg2+ binding. The P loop is a region that is critical for nucleotide binding in G proteins (38). In the Ran:RCC1 complex, the P loop moves toward the guanine basebinding site. The phosphate-binding site in this structure is occupied by an oxyanion, most likely a sulfate. This has also been observed in other GTPase:GEF complexes such as the Rac1:Tiam1 (93) and the RhoA:LARG complexes (94). It has been postulated that these complexes mimic an intermediate of the exchange reaction, i.e., a low-affinity ternary GTPase:nucleotide:GEF complex (91).

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26

Recycling RanGDP to the Nucleus: NTF2 The continuous efflux of karyopherin: RanGTP complexes from the nucleus depletes the nuclear levels of RanGTP. RanGDP therefore has to be reimported into the nucleus where nucleotide exchange is catalyzed (Figure 2). The nuclear import of RanGDP is mediated by NTF2 (95, 96). NTF2 specifically recognizes Ran in its GDP-bound state (67, 97–102). 662

Cook et al.

The crystal structure of NTF2 reveals that the molecule is a homodimer with an overall barrel-like structure. The monomer folds into a curved antiparallel β-sheet that on one side is flanked by α-helices and on the other packs against the β-sheet of the other monomer (103). The NTF2 barrel opens at one end, forming a distinctive cavity. The structure of the NTF2:RanGTP complex reveals that this hydrophobic cavity of NTF2 interacts extensively with the switch II region of Ran (104) (Figure 6d ). How does NTF2 discriminate against RanGTP? Superposition of Ran in the GTPand GDP-bound states indicates that the switch regions of RanGTP would sterically clash with NTF2 (69). Comparison of the RanGDP:NTF2 and Ran:RCC1 complexes shows that Ran cannot bind to NTF2 and RCC1 simultaneously (91). This is consistent with the observation that NTF2 inhibits RCC1-catalyzed dissociation of GDP from Ran (105) and suggests that RanGDP needs to dissociate from the NTF2 complex prior to nucleotide exchange. As with the karyopherins, NTF2 is able to translocate through the NPC by interacting with nucleoporins (98, 106–108).

NUCLEOPORINS One of the functional criteria that defines a nuclear transport factor is the ability to interact with the NPC, in particular with phenylalanine-glycine rich sequences, known as FG repeats, that are characteristic of many nucleoporins. The structures of FG repeats in complex with importin β or with NTF2 show that NPC recognition occurs at sites on the transport factors distinct from those involved in cargo and/or Ran binding. In the case of importin β, for example, the FG repeats bind at two sites on the outer convex surface of the superhelix (109). This observation explains how the transport factors can interact with nucleoporins both in the presence and in the absence of bound proteins and are therefore able to shuttle through the NPC

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26

ARI

1 May 2007

17:35

with cargo and to recycle back to the original compartment without it. The FG-containing moiety binds with a similar conformation in complexes with both importin β and NTF2, despite the fact that the two transport factors have very different structures (109, 110). In both cases, the FGcontaining peptide folds in a tight turn, with the phenylalanine side chain protruding into a hydrophobic pocket on the surface of the transport factor. The paucity of the contacts is consistent with the observation that the interaction between transport factors and FG repeats are generally of low affinity, reflecting the need for movement through the NPC rather than localization (111). FG-repeat nucleoporins can be classified as FxFG or GLFG types (where x is any residue) based on the local sequence surrounding the FG motif and have a different distribution within the NPC. Crystallographic analysis indicates that these two types have comparable interactions with importin β and bind to overlapping sites (109, 112). Thus, the distribution of different families of FG-containing nucleoporins may not be important for directionality in transport by the karyopherin and NTF2 families of transport factors. A similar mode of FG-repeat recognition is also observed for the mRNA export factor (see below).

mRNA EXPORT The TAP:p15 mRNA Export Factor: Nucleoporin Binding The bulk of mRNA export is independent of Ran and of members of the karyopherin-β family (Figure 7). The mRNA export factor is a heterodimer containing a conserved protein, known as TAP (or NXF1) in metazoans and Mex67 in yeast (reviewed in Reference 113). TAP associates with p15 (also known as NXT1), a protein that shares significant sequence homology with NTF2 (114). In yeast, the corresponding Mex67-associated protein is Mtr2 (115). Yeast Mtr2 is not only functionally homologous to human p15, but also

structurally homologous despite lacking any discernable sequence similarity (116). TAP is a multidomain protein, composed of an N-terminal cargo-binding region and a C-terminal nucleoporin-binding region (117, 118) (Figure 7b). The C-terminal region of TAP consists of two domains, one of which has low sequence similarity to NTF2 and is required for the interaction with p15, and the other shows limited homology to a ubiquitinassociated (UBA) fold. The crystal structure of the NTF2-like domain of TAP in complex with p15 shows a heterodimer that is reminiscent of the NTF2 homodimer (119). However, because of relatively minor structural differences, TAP:p15 has no binding pocket at the corresponding structural position of the Ran-binding pocket in NTF2. Consistently, the TAP:p15 NTF2-like heterodimer is unable to bind RanGDP. The TAP:p15 NTF2-like structure provides a binding site for FG-repeat nucleoporins. The FG-binding pocket is found at a single site on the TAP domain. No FG-binding sites are present in p15, which contributes indirectly by forming a single structural unit with TAP. A second FG-binding site on TAP is provided by the UBA-like domain (120). Both FG-binding domains contribute to NPC association and are required synergistically to export cellular mRNAs. Mutation of the FG-binding sites in either of these domains impairs nuclear rim association and mRNA export activity (114).

RBD: RNA-binding domain

The TAP:p15 mRNA Export Factor: Cargo Binding The N-terminal region of TAP consists of an RNA-binding domain (RBD, also known as RNA-recognition motif or RRM domain) and a leucine-rich repeat (LLR) domain (121) (Figure 7b). The RBD has the structural and functional characteristic of a typical RBD, although it lacks the two characteristic sequence motifs RNP1 and RNP2 (reviewed in Reference 122) and as such was not predicted on the basis of sequence analysis. The RBD of TAP is not strictly required for general RNA export www.annualreviews.org • Nucleocytoplasmic Transport

663

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26 ARI 1 May 2007

664

17:35

Cook et al.

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26

ARI

1 May 2007

17:35

activity (123) and is unlikely to be present in the yeast orthologue Mex67. However, it is required with the downstream LRR domain for the binding of the constitutive transport element (CTE) RNA to human TAP. The CTE RNA is present in simian type D retroviruses and mediates the export of the viral unspliced genomic RNA to the host cytoplasm via the direct interaction with TAP (117, 124, 125). The requirement of the RBD and LRR domains of TAP to bind the CTE RNA is reminiscent of the structural and biochemical properties of the U2B” and U2A’ components of the spliceosomal complex that bind to the U2 small nuclear RNA (121, 126). Given the similarity of the LRR domain of TAP to U2A’ and its conservation across species, it might serve as a protein-protein interaction module (possibly for RBD-containing adaptors) similar to that observed in the spliceosomal complex. In contrast to CTE RNA recognition, the recognition of cellular mRNAs by TAP is likely not direct but mediated by adaptor molecules that are associated with mRNA.

Adaptors and Regulators in the mRNA Export Pathway Cellular mRNAs are associated with proteins in the form of ribonucleoprotein particles (mRNPs) (Figure 7). The protein composition of a given mRNP changes as it proceeds from transcription to processing and export (reviewed in References 127–129). The TREX complex, for example, is recruited at

early stages of mRNP processing and is removed before nuclear export; the exon junction complex associates upon splicing and stays stably bound after export to the cytoplasm; and the helicase Dbp5 travels together with mRNA and is also found associated with components on the NPC. Although we are still far from understanding the exact molecular mechanisms of how mature mRNPs are recognized and how they are delivered to cytoplasmic ribosomes, some of the crucial players other than TAP:p15 have been identified. The protein Aly/REF (also known as Yra1 in yeast) is an adaptor protein between the export factor and mRNA (130–132). Aly/REF has a conserved RBD (133, 134) (Figure 7c), flanked by Gly-Arg rich N-terminal and Cterminal sequences. These flanking sequences interact with the N-terminal region of TAP and RNA (135, 136). The recruitment of Aly/REF to the mRNA prior to nuclear export is probably mediated by UAP56 (known as Sub2 in yeast) (137, 138). UAP56 belongs to the DExD-box family of RNA-dependent ATPases (139) (Figure 7c). UAP56 is essential for the export of poly(A)+ RNA from the nucleus to the cytoplasm in yeast, Drosophila and Caenorhabditis elegans (137, 140). Interestingly, knockdown of Aly/REF in cultured Drosophila cells shows little effect on mRNA export, raising the possibility that UAP56 may interact with other proteins in addition to Aly/REF (141). Other candidates as adaptors for mRNA export are members of the SR family of splicing factors (142).

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−− Figure 7 Structures in the mRNA export pathway. (a) Schematic of some of the proteins and interactions involved in the export of mRNA through the nuclear pore complex (NPC). (b) The mRNA export factor TAP:p15. The N-terminal cargo-binding region of TAP contains an RNA-binding domain (RBD) and leucine-rich repeat (LRR) domains (PDB code 1fo1). The C-terminal half of TAP contains a nuclear transport factor 2 (NTF2)-like domain that dimerizes with p15 and binds an FG-containing nucleoporin peptide (PDB code 1jn5). The C-terminal UBA-like domain also binds an FG peptide (PDB code 1oai). (c) The RBD of Aly/REF (PDB code 1no8). The helicase domain of the DEAD-box protein UAP56 is in complex with ADP and magnesium (PDB code 1xtj). The N-terminal domain of the nucleoporin Nup159 is involved in binding the DEAD-box protein, Dbp5 (PDB code 1xip). Abbreviations: EJC, exon junction complex; NPC, nuclear pore complex. www.annualreviews.org • Nucleocytoplasmic Transport

665

ARI

1 May 2007

17:35

Proteins of the DExD-box family are involved in all steps of mRNA metabolism, from splicing to decay (143). Another member of this family that plays a role in mRNA nuclear export is the protein Dbp5 (144, 145). In yeast, Dpb5 is associated with the cytoplasmic filaments of the NPC through its interactions with the nucleoporin Nup159 (CAN in humans) (144, 146, 147) and with Gle1, which enhances the ATPase and helicase activity of Dbp5 (148, 149). The structure of the N-terminal domain of Nup159 reveals a β-propeller domain that has been shown by mutational analysis to bind Dbp5 via a set of conserved residues (150) (Figure 7c). It is possible that these interactions at the

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26

cytoplasmic side of the nuclear envelope play a role in imparting directionality to the mRNA export process, but the exact mechanism of mRNA dissociation from the export factor is currently unclear. Although several structures of protein domains involved in the mRNA export pathway have been determined (Figure 7), there is no information available at present on how these domains mediate specific macromolecular interactions. More generally, understanding how the mRNA export machinery is linked to the proteins that mediate upstream and downstream steps in the gene expression pathway is perhaps the most challenging problem ahead.

SUMMARY POINTS 1. The majority of nuclear import and export traffic is carried by proteins of the karyopherin β family. Structurally, karyopherins share a superhelical architecture of tandem HEAT repeats. Functionally, they share the ability to interact with the regulator RanGTP and with FG components of NPCs. 2. Structures of the same karyopherin bound to different protein cargoes have shown how receptors can be versatile in cargo recognition. The superhelical architecture of the receptor has an inherent plasticity that allows it to adopt shapes complementary to different cargoes with an induced-fit type of mechanism. 3. RanGTP binding triggers the dissociation of importin:cargo complexes and the association of exportin:cargo complexes. A conserved insertion at HEAT repeat 8 of the karyopherins plays a pivotal role in the RanGTP-mediated response. 4. The directionality of transport by the karyopherin family relies on the presence of a steep RanGTP gradient between the cytoplasm and the nucleus. The intricate series of protein-protein interactions that are responsible for generating this gradient have been elucidated at the atomic level. 5. The export factor that mediates mRNA export is the TAP:p15 heterodimer. TAP is a multidomain protein that is not related to the karyopherin family and that does not respond to the presence of Ran. However, β-karyopherins and TAP:p15 both recognize the FG components of NPCs with a similar mode of binding, indicating a unifying mechanism for different nucleocytoplasmic transport factors.

FUTURE ISSUES 1. In order to determine the general principles of Ran-dependent transport, the structure of each karyopherin bound to its cargoes is required. This will reveal the mechanisms

666

Cook et al.

ANRV313-BI76-26

ARI

1 May 2007

17:35

by which nuclear import/export signals, particularly of cargoes such as NESs, tRNAs, or pre-microRNAs, are recognized. 2. In addition to these general principles, exceptions have been described that have not been structurally tackled as yet. For example, certain cargoes can be dissociated without Ran, and in some cases, the same β-karyopherin can mediate both import and export.

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

3. Perhaps the most important issue to be resolved entails the mRNA export pathway: how different mRNAs are recognized for export, how they are released in the cytoplasm, and how the nuclear transport machinery performs cross talk with complexes involved in mRNA metabolism upstream and downstream of NPC translocation.

ACKNOWLEDGMENTS We thank Nicola Graf for preparation of schematic figures and Yuh-Min Chook for access to PDB coordinates prior to release. We are also grateful to Anja Strasser and Esben Lorentzen for critical reading of the manuscript. We apologize, in advance, to those investigators whose work was inadvertently overlooked.

LITERATURE CITED 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24.

Fried H, Kutay U. 2003. Cell Mol. Life Sci. 60:1659–88 Damelin M, Silver PA, Corbett AH. 2002. Methods Enzymol. 351:587–607 Kuersten S, Ohno M, Mattaj IW. 2001. Trends Cell Biol. 11:497–503 Macara IG. 2001. Microbiol. Mol. Biol. Rev. 65:570–94 Conti E, Izaurralde E. 2001. Curr. Opin. Cell Biol. 13:310–19 Chook YM, Cingolani G, Conti E, Stewart M, Vetter I, Wittinghofer A. 1999. Trends Cell Biol. 9:310–11 ¨ Gorlich D, Kutay U. 1999. Annu. Rev. Cell Dev. Biol. 15:607–60 Tran EJ, Wente SR. 2006. Cell 125:1041–53 Maco B, Fahrenkrog B, Huang NP, Aebi U. 2006. Methods Mol. Biol. 322:273–88 Schwartz TU. 2005. Curr. Opin. Struct. Biol. 15:221–26 Mosammaparast N, Pemberton LF. 2004. Trends Cell Biol. 14:547–56 Strom AC, Weis K. 2001. Genome Biol. 2:REVIEWS3008 Weis K. 2003. Cell 112:441–51 Dasso M. 2002. Curr. Biol. 12:R502–8 Stewart M. 2000. Cell Struct. Funct. 25:217–25 Suel KE, Cansizoglu AE, Chook YM. 2006. Methods 39:342–55 Andrade MA, Petosa C, O’Donoghue SI, Muller CW, Bork P. 2001. J. Mol. Biol. 309:1–18 Conti E, Muller CW, Stewart M. 2006. Curr. Opin. Struct. Biol. 16:237–44 Harel A, Forbes DJ. 2004. Mol. Cell 16:319–30 ¨ Gorlich D, Kostka S, Kraft R, Dingwall C, Laskey RA, et al. 1995. Curr. Biol. 5:383–92 Imamoto N, Shimamoto T, Kose S, Takao T, Tachibana T, et al. 1995. FEBS Lett. 368:415–19 Chi NC, Adam EJ, Adam SA. 1995. J. Cell Biol. 130:265–74 Radu A, Blobel G, Moore MS. 1995. Proc. Natl. Acad. Sci. USA 92:1769–73 Cingolani G, Petosa C, Weis K, Muller CW. 1999. Nature 399:221–29 www.annualreviews.org • Nucleocytoplasmic Transport

667

ARI

1 May 2007

17:35

25. Cingolani G, Bednenko J, Gillespie MT, Gerace L. 2002. Mol. Cell 10:1345–53 26. Lee SJ, Sekimoto T, Yamashita E, Nagoshi E, Nakagawa A, et al. 2003. Science 302:1571– 75 27. Conti E, Uy M, Leighton L, Blobel G, Kuriyan J. 1998. Cell 94:193–204 28. Conti E, Kuriyan J. 2000. Structure 8:329–38 29. Fontes MR, Teh T, Kobe B. 2000. J. Mol. Biol. 297:1183–94 30. Will CL, Luhrmann R. 2001. Curr. Opin. Cell Biol. 13:290–301 31. Fischer U, Luhrmann R. 1990. Science 249:786–90 32. Huber J, Cronshagen U, Kadokura M, Marshallsay C, Wada T, et al. 1998. EMBO J. 17:4114–26 33. Palacios I, Hetzer M, Adam SA, Mattaj IW. 1997. EMBO J. 16:6783–92 34. Strasser A, Dickmanns A, Luhrmann R, Ficner R. 2005. EMBO J. 24:2235–43 35. Pollard VW, Michael WM, Nakielny S, Siomi MC, Wang F, Dreyfuss G. 1996. Cell 86:985–94 36. Bonifaci N, Moroianu J, Radu A, Blobel G. 1997. Proc. Natl. Acad. Sci. USA 94:5055–60 37. Lee BJ, Cansizoglu AE, Suel KE, Louis TH, Zhang Z, Chook YM. 2006. Cell 126:543–58 38. Vetter IR, Wittinghofer A. 2001. Science 294:1299–304 ¨ 39. Vetter IR, Arndt A, Kutay U, Gorlich D, Wittinghofer A. 1999. Cell 97:635–46 40. Lee SJ, Matsuura Y, Liu SM, Stewart M. 2005. Nature 435:693–96 41. Chook YM, Blobel G. 1999. Nature 399:230–37 42. Chook YM, Jung A, Rosen MK, Blobel G. 2002. Biochemistry 41:6955–66 43. Goldfarb DS, Corbett AH, Mason DA, Harreman MT, Adam SA. 2004. Trends Cell Biol. 14:505–14 44. Kobe B. 1999. Nat. Struct. Biol. 6:388–97 45. Harreman MT, Cohen PE, Hodel MR, Truscott GJ, Corbett AH, Hodel AE. 2003. J. Biol. Chem. 278:21361–69 46. Booth JW, Belanger KD, Sannella MI, Davis LI. 1999. J. Biol. Chem. 274:32360–67 47. Hood JK, Casolari JM, Silver PA. 2000. J. Cell Sci. 113(Pt. 8):1471–80 48. Lindsay ME, Plafker K, Smith AE, Clurman BE, Macara IG. 2002. Cell 110:349–60 49. Solsbacher J, Maurer P, Vogel F, Schlenstedt G. 2000. Mol. Cell. Biol. 20:8468–79 50. Gilchrist D, Mykytka B, Rexach M. 2002. J. Biol. Chem. 277:18161–72 51. Matsuura Y, Stewart M. 2005. EMBO J. 24:3681–89 ¨ 52. Kutay U, Bischoff FR, Kostka S, Kraft R, Gorlich D. 1997. Cell 90:1061–71 53. Solsbacher J, Maurer P, Bischoff FR, Schlenstedt G. 1998. Mol. Cell. Biol. 18:6805–15 54. Cook A, Fernandez E, Lindner D, Ebert J, Schlenstedt G, Conti E. 2005. Mol. Cell 18:355–67 55. Matsuura Y, Stewart M. 2004. Nature 432:872–77 56. Kudo N, Matsumori N, Taoka H, Fujiwara D, Schreiner EP, et al. 1999. Proc. Natl. Acad. Sci. USA 96:9112–17 57. Wolff B, Sanglier JJ, Wang Y. 1997. Chem. Biol. 4:139–47 58. Fischer U, Huber J, Boelens WC, Mattaj IW, Luhrmann R. 1995. Cell 82:475–83 59. Fornerod M, Ohno M, Yoshida M, Mattaj IW. 1997. Cell 90:1051–60 60. Wen W, Meinkoth JL, Tsien RY, Taylor SS. 1995. Cell 82:463–73 61. Petosa C, Schoehn G, Askjaer P, Bauer U, Moulin M, et al. 2004. Mol. Cell 16:761–75 62. Kutay U, Guttinger S. 2005. Trends Cell Biol. 15:121–24 63. Englmeier L, Fornerod M, Bischoff FR, Petosa C, Mattaj IW, Kutay U. 2001. EMBO Rep. 2:926–32 64. Fukuhara N, Fernandez E, Ebert J, Conti E, Svergun D. 2004. J. Biol. Chem. 279:2176–81

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26

668

Cook et al.

ANRV313-BI76-26

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107.

ARI

1 May 2007

17:35

¨ Mingot JM, Kostka S, Kraft R, Hartmann E, Gorlich D. 2001. EMBO J. 20:3685–94 Yoshida K, Blobel G. 2001. J. Cell Biol. 152:729–40 ¨ Gorlich D, Pante N, Kutay U, Aebi U, Bischoff FR. 1996. EMBO J. 15:5584–94 Scheffzek K, Klebe C, Fritz-Wolf K, Kabsch W, Wittinghofer A. 1995. Nature 374:378– 81 Vetter IR, Nowak C, Nishimoto T, Kuhlmann J, Wittinghofer A. 1999. Nature 398:39–46 Nilsson J, Weis K, Kjems J. 2002. J. Mol. Biol. 318:583–93 ¨ Bischoff FR, Gorlich D. 1997. FEBS Lett. 419:249–54 Bischoff FR, Krebber H, Smirnova E, Dong W, Ponstingl H. 1995. EMBO J. 14:705–15 Coutavas E, Ren M, Oppenheim JD, D’Eustachio P, Rush MG. 1993. Nature 366:585–87 Delphin C, Guan T, Melchior F, Gerace L. 1997. Mol. Biol. Cell 8:2379–90 Chi NC, Adam EJ, Visser GD, Adam SA. 1996. J. Cell Biol. 135:559–69 Lounsbury KM, Macara IG. 1997. J. Biol. Chem. 272:551–55 Klebe C, Bischoff FR, Ponstingl H, Wittinghofer A. 1995. Biochemistry 34:639–47 Haberland J, Gerke V. 1995. FEBS Lett. 357:173–77 Hillig RC, Renault L, Vetter IR, Drell T, Wittinghofer A, Becker J. 1999. Mol. Cell 3:781–91 Mahajan R, Delphin C, Guan T, Gerace L, Melchior F. 1997. Cell 88:97–107 Matunis MJ, Coutavas E, Blobel G. 1996. J. Cell Biol. 135:1457–70 Seewald MJ, Korner C, Wittinghofer A, Vetter IR. 2002. Nature 415:662–66 Scheffzek K, Ahmadian MR. 2005. Cell Mol. Life Sci. 62:3014–38 Floer M, Blobel G. 1996. J. Biol. Chem. 271:5313–16 Bischoff FR, Ponstingl H. 1991. Nature 354:80–82 Bischoff FR, Ponstingl H. 1991. Proc. Natl. Acad. Sci. USA 88:10830–34 Aebi M, Clark MW, Vijayraghavan U, Abelson J. 1990. Mol. Gen. Genet. 224:72–80 Kadowaki T, Goldfarb D, Spitz LM, Tartakoff AM, Ohno M. 1993. EMBO J. 12:2929–37 Nemergut ME, Mizzen CA, Stukenberg T, Allis CD, Macara IG. 2001. Science 292:1540– 43 Renault L, Nassar N, Vetter I, Becker J, Klebe C, et al. 1998. Nature 392:97–101 Renault L, Kuhlmann J, Henkel A, Wittinghofer A. 2001. Cell 105:245–55 Azuma Y, Renault L, Garcia-Ranea JA, Valencia A, Nishimoto T, Wittinghofer A. 1999. J. Mol. Biol. 289:1119–30 Worthylake DK, Rossman KL, Sondek J. 2000. Nature 408:682–88 Kristelly R, Gao G, Tesmer JJ. 2004. J. Biol. Chem. 279:47352–62 ¨ Ribbeck K, Lipowsky G, Kent HM, Stewart M, Gorlich D. 1998. EMBO J. 17:6587–98 Smith A, Brownawell A, Macara IG. 1998. Curr. Biol. 8:1403–6 Moore MS, Blobel G. 1994. Proc. Natl. Acad. Sci. USA 91:10212–16 Nehrbass U, Blobel G. 1996. Science 272:120–22 Weis K, Dingwall C, Lamond AI. 1996. EMBO J. 15:7120–28 Schlenstedt G, Saavedra C, Loeb JD, Cole CN, Silver PA. 1995. Proc. Natl. Acad. Sci. USA 92:225–29 Richards SA, Carey KL, Macara IG. 1997. Science 276:1842–44 Paschal BM, Delphin C, Gerace L. 1996. Proc. Natl. Acad. Sci. USA 93:7679–83 Bullock TL, Clarkson WD, Kent HM, Stewart M. 1996. J. Mol. Biol. 260:422–31 Stewart M, Kent HM, McCoy AJ. 1998. J. Mol. Biol. 277:635–46 Yamada M, Tachibana T, Imamoto N, Yoneda Y. 1998. Curr. Biol. 8:1339–42 Paschal BM, Gerace L. 1995. J. Cell Biol. 129:925–37 Clarkson WD, Corbett AH, Paschal BM, Kent HM, McCoy AJ, et al. 1997. J. Mol. Biol. 272:716–30 www.annualreviews.org • Nucleocytoplasmic Transport

669

ARI

1 May 2007

17:35

108. Clarkson WD, Kent HM, Stewart M. 1996. J. Mol. Biol. 263:517–24 109. Bayliss R, Littlewood T, Stewart M. 2000. Cell 102:99–108 110. Bayliss R, Leung SW, Baker RP, Quimby BB, Corbett AH, Stewart M. 2002. EMBO J. 21:2843–53 ¨ 111. Ribbeck K, Gorlich D. 2001. EMBO J. 20:1320–30 112. Bayliss R, Littlewood T, Strawn LA, Wente SR, Stewart M. 2002. J. Biol. Chem. 277:50597–606 113. Izaurralde E. 2002. Eur. J. Cell Biol. 81:577–84 114. Braun IC, Herold A, Rode M, Conti E, Izaurralde E. 2001. J. Biol. Chem. 276:20536–43 115. Santos-Rosa H, Moreno H, Simos G, Segref A, Fahrenkrog B, et al. 1998. Mol. Cell. Biol. 18:6826–38 116. Fribourg S, Conti E. 2003. EMBO Rep. 4:699–703 117. Braun IC, Rohrbach E, Schmitt C, Izaurralde E. 1999. EMBO J. 18:1953–65 118. Bachi A, Braun IC, Rodrigues JP, Pante N, Ribbeck K, et al. 2000. RNA 6:136–58 119. Fribourg S, Braun IC, Izaurralde E, Conti E. 2001. Mol. Cell 8:645–56 120. Grant RP, Neuhaus D, Stewart M. 2003. J. Mol. Biol. 326:849–58 121. Liker E, Fernandez E, Izaurralde E, Conti E. 2000. EMBO J. 19:5587–98 122. Varani G, Nagai K. 1998. Annu. Rev. Biophys. Biomol. Struct. 27:407–45 123. Herold A, Suyama M, Rodrigues JP, Braun IC, Kutay U, et al. 2000. Mol. Cell. Biol. 20:8996–9008 124. Gruter P, Tabernero C, von Kobbe C, Schmitt C, Saavedra C, et al. 1998. Mol. Cell 1:649–59 125. Kang Y, Cullen BR. 1999. Genes Dev. 13:1126–39 126. Price SR, Evans PR, Nagai K. 1998. Nature 394:645–50 127. Moore MJ. 2005. Science 309:1514–18 128. Vinciguerra P, Stutz F. 2004. Curr. Opin. Cell Biol. 16:285–92 129. Reed R, Hurt E. 2002. Cell 108:523–31 130. Bruhn L, Munnerlyn A, Grosschedl R. 1997. Genes Dev. 11:640–53 131. Strasser K, Hurt E. 2000. EMBO J. 19:410–20 132. Stutz F, Bachi A, Doerks T, Braun IC, Seraphin B, et al. 2000. RNA 6:638–50 133. Perez-Alvarado GC, Martinez-Yamout M, Allen MM, Grosschedl R, Dyson HJ, Wright PE. 2003. Biochemistry 42:7348–57 134. Wittekind M, Gorlach M, Friedrichs M, Dreyfuss G, Mueller L. 1992. Biochemistry 31:6254–65 135. Zenklusen D, Vinciguerra P, Strahm Y, Stutz F. 2001. Mol. Cell. Biol. 21:4219–32 136. Rodrigues JP, Rode M, Gatfield D, Blencowe BJ, Carmo-Fonseca M, Izaurralde E. 2001. Proc. Natl. Acad. Sci. USA 98:1030–35 137. Strasser K, Hurt E. 2001. Nature 413:648–52 138. Luo ML, Zhou Z, Magni K, Christoforides C, Rappsilber J, et al. 2001. Nature 413:644– 47 139. Shi H, Cordin O, Minder CM, Linder P, Xu RM. 2004. Proc. Natl. Acad. Sci. USA 101:17628–33 140. Gatfield D, Le Hir H, Schmitt C, Braun IC, Kocher T, et al. 2001. Curr. Biol. 11:1716–21 141. Gatfield D, Izaurralde E. 2002. J. Cell Biol. 159:579–88 142. Huang Y, Gattoni R, Stevenin J, Steitz JA. 2003. Mol. Cell 11:837–43 143. Rocak S, Linder P. 2004. Nat. Rev. Mol. Cell Biol. 5:232–41 144. Schmitt C, von Kobbe C, Bachi A, Pante N, Rodrigues JP, et al. 1999. EMBO J. 18:4332– 47

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

ANRV313-BI76-26

670

Cook et al.

ANRV313-BI76-26

ARI

1 May 2007

17:35

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

145. Tseng SS, Weaver PL, Liu Y, Hitomi M, Tartakoff AM, Chang TH. 1998. EMBO J. 17:2651–62 146. Hodge CA, Colot HV, Stafford P, Cole CN. 1999. EMBO J. 18:5778–88 147. Strahm Y, Fahrenkrog B, Zenklusen D, Rychner E, Kantor J, et al. 1999. EMBO J. 18:5761–77 148. Alcazar-Roman AR, Tran EJ, Guo S, Wente SR. 2006. Nat. Cell Biol. 8:711–16 149. Weirich CS, Erzberger JP, Flick JS, Berger JM, Thorner J, Weis K. 2006. Nat. Cell Biol. 8:668–76 150. Weirich CS, Erzberger JP, Berger JM, Weis K. 2004. Mol. Cell 16:749–60

www.annualreviews.org • Nucleocytoplasmic Transport

671

AR313-FM

ARI

8 May 2007

21:56

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

Contents

Annual Review of Biochemistry Volume 76, 2007

Mitochondrial Theme The Magic Garden Gottfried Schatz p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p673 DNA Replication and Transcription in Mammalian Mitochondria Maria Falkenberg, Nils-Göran Larsson, and Claes M. Gustafsson p p p p p p p p p p p p p p p p p p p679 Mitochondrial-Nuclear Communications Michael T. Ryan and Nicholas J. Hoogenraad p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p701 Translocation of Proteins into Mitochondria Walter Neupert and Johannes M. Herrmann p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p723 The Machines that Divide and Fuse Mitochondria Suzanne Hoppins, Laura Lackner, and Jodi Nunnari p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p751 Why Do We Still Have a Maternally Inherited Mitochondrial DNA? Insights from Evolutionary Medicine Douglas C. Wallace p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p781

Molecular Mechanisms of Antibody Somatic Hypermutation Javier M. Di Noia and Michael S. Neuberger p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p1 Structure and Mechanism of Helicases and Nucleic Acid Translocases Martin R. Singleton, Mark S. Dillingham, and Dale B. Wigley p p p p p p p p p p p p p p p p p p p p p p 23 The Nonsense-Mediated Decay RNA Surveillance Pathway Yao-Fu Chang, J. Saadi Imam, Miles F. Wilkinson p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 51 Functions of Site-Specific Histone Acetylation and Deacetylation Mona D. Shahbazian and Michael Grunstein p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 75 The tmRNA System for Translational Surveillance and Ribosome Rescue Sean D. Moore and Robert T. Sauer p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p101 Membrane Protein Structure: Prediction versus Reality Arne Elofsson and Gunnar von Heijne p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p125 v

AR313-FM

ARI

8 May 2007

21:56

Structure and Function of Toll Receptors and Their Ligands Nicholas J. Gay and Monique Gangloff p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p141 The Role of Mass Spectrometry in Structure Elucidation of Dynamic Protein Complexes Michal Sharon and Carol V. Robinson p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p167 Structure and Mechanism of the 6-Deoxyerythronolide B Synthase Chaitan Khosla, Yinyan Tang, Alice Y. Chen, Nathan A. Schnarr, and David E. Cane p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p195 Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

The Biochemistry of Methane Oxidation Amanda S. Hakemian and Amy C. Rosenzweig p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p223 Anthrax Toxin: Receptor Binding, Internalization, Pore Formation, and Translocation John A.T. Young and R. John Collier p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p243 Synapses: Sites of Cell Recognition, Adhesion, and Functional Specification Soichiro Yamada and W. James Nelson p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p267 Lipid A Modification Systems in Gram-negative Bacteria Christian R.H. Raetz, C. Michael Reynolds, M. Stephen Trent, and Russell E. Bishop p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p295 Chemical Evolution as a Tool for Molecular Discovery S. Jarrett Wrenn and Pehr B. Harbury p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p331 Molecular Mechanisms of Magnetosome Formation Arash Komeili p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p351 Modulation of the Ryanodine Receptor and Intracellular Calcium Ran Zalk, Stephan E. Lehnart, and Andrew R. Marks p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p367 TRP Channels Kartik Venkatachalam and Craig Montell p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p387 Studying Individual Events in Biology Stefan Wennmalm and Sanford M. Simon p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p419 Signaling Pathways Downstream of Pattern-Recognition Receptors and Their Cross Talk Myeong Sup Lee and Young-Joon Kim p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p447 Biochemistry and Physiology of Cyclic Nucleotide Phosphodiesterases: Essential Components in Cyclic Nucleotide Signaling Marco Conti and Joseph Beavo p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p481 The Eyes Absent Family of Phosphotyrosine Phosphatases: Properties and Roles in Developmental Regulation of Transcription Jennifer Jemc and Ilaria Rebay p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p513 vi

Contents

AR313-FM

ARI

8 May 2007

21:56

Assembly Dynamics of the Bacterial MinCDE System and Spatial Regulation of the Z Ring Joe Lutkenhaus p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p539 Structures and Functions of Yeast Kinetochore Complexes Stefan Westermann, David G. Drubin, and Georjana Barnes p p p p p p p p p p p p p p p p p p p p p p p p563 Mechanism and Function of Formins in the Control of Actin Assembly Bruce L. Goode and Michael J. Eck p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p593

Annu. Rev. Biochem. 2007.76:647-671. Downloaded from arjournals.annualreviews.org by University of Massachusetts - Amherst on 05/04/10. For personal use only.

Unsolved Mysteries in Membrane Traffic Suzanne R. Pfeffer p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p629 Structural Biology of Nucleocytoplasmic Transport Atlanta Cook, Fulvia Bono, Martin Jinek, and Elena Conti p p p p p p p p p p p p p p p p p p p p p p p p p p647 The Postsynaptic Architecture of Excitatory Synapses: A More Quantitative View Morgan Sheng and Casper C. Hoogenraad p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p823 Indexes Cumulative Index of Contributing Authors, Volumes 72–76 p p p p p p p p p p p p p p p p p p p p p p p p849 Cumulative Index of Chapter Titles, Volumes 72–76 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p853 Errata An online log of corrections to Annual Review of Biochemistry chapters (if any, 1997 to the present) may be found at http://biochem.annualreviews.org/errata.shtml

Contents

vii