Nuclear export receptor CRM1 recognizes diverse conformations in nuclear export signals.

Nuclear export receptor CRM1 binds highly variable nuclear export signals (NESs) in hundreds of different cargoes. Previously we have shown that CRM1 binds NESs in both polypeptide orientations (Fung et al., 2015). Here, we show crystal structures of CRM1 bound to eight additional NESs which reveal diverse conformations that range from loop-like to all-helix, which occupy different extents of the invariant NES-binding groove. Analysis of all NES structures show 5-6 distinct backbone conformations where the only conserved secondary structural element is one turn of helix that binds the central portion of the CRM1 groove. All NESs also participate in main chain hydrogen bonding with human CRM1 Lys568 side chain, which acts as a specificity filter that prevents binding of non-NES peptides. The large conformational range of NES backbones explains the lack of a fixed pattern for its 3-5 hydrophobic anchor residues, which in turn explains the large array of peptide sequences that can function as NESs.

Introduction

The chromosome region maintenance 1 protein (CRM1) or Exportin-1 (XPO1) binds 8–15 residues-long nuclear export signals (NESs) in hundreds of different protein cargoes (Fornerod et al., 1997; Fukuda et al., 1997; Stade et al., 1997; Ossareh-Nazari et al., 1997; Xu et al., 2012a; Kırlı et al., 2015; Thakar et al., 2013). NES sequences are very diverse but each usually has 4–5 hydrophobic residues (often Leu/Val/Ile/Phe/Met; labeled Ф0–5) that bind hydrophobic pockets (labeled P0–P4) in a hydrophobic groove formed by HEAT repeats 11 and 12 of CRM1 (9–16). The hydrophobic anchor residues are arranged in many ways, currently described by ten consensus patterns for corresponding NES classes 1a, 1b, 1c, 1d, 2, 3, 1a-R, 1b-R, 1c-R and 1d-R (Figure 1A) (Fung et al., 2015).

Figure 1.

Structures of CRM1-bound NESs that match the potentially all-helical class 3 pattern.

(A) Current NES sequence patterns (Φ is Leu, Val, Ile, Phe or Met and X is any amino acid). Potential amphipathic α-helices, predicted with hydrophobic patterns of i, i + 4, i + 7 or i, i + 3, i + 7 or i, i + 3, i + 7, i + 10, are shaded grey. (B) Structure of PKINES(Ф0L) (dark blue, PDB ID: 3NBY), RevNES (pink, 3NBZ) and CPEB4NES (yellow, 5DIF) bound to the NES-binding groove of CRM1 (grey surface). NESs are shown in cartoon representations with their Φ side chains shown as sticks. (C) The overall structure of the engineered CRM1 (grey)-Ran•GTP (orange)-RanBP1 (purple)-mDia2NES (pale green) complex. The structure of (D) mDia2NES (pale green), (E) CDC7NES (green-cyan) and (F) X11L2NES (forest) bound to the NES-binding groove of CRM1 in the engineered CRM1-Ran-RanBP1 complex. *The X11L2NES sequence matches the class 3 pattern, but binds CRM1 according to the new hydrophobic pattern Φ0XXΦ1XXXΦ2XXΦ3XXXΦ4 that we termed the class 4 pattern. mDia2NES is not shown in the leftmost panel of (D) to view the five hydrophobic pockets (P0–P4) of the CRM1 groove. Rightmost panels of (D–F): overlays of 3.0σ positive densities of kick OMIT mFo-DFc maps (calculated with peptides omitted) and final coordinates of the NES peptides. Middle panels of (D–F): black dashes show CRM1-NES hydrogen bonds and polar contacts.

DOI: http://dx.doi.org/10.7554/eLife.23961.002

DOI: http://dx.doi.org/10.7554/eLife.23961.003

DOI: http://dx.doi.org/10.7554/eLife.23961.004

DOI: http://dx.doi.org/10.7554/eLife.23961.005

Data from triplicate experiments of direct (titration of CRM1 to FITC-PKINES in the presence of excess RanGTP) or competition titrations (titration of MBP-NES to FITC-PKINES and CRM1 with excess RanGTP) are plotted as mean ± SD and fitted binding curves are overlaid on top. MBP-NES competes with FITC-PKINES for CRM1. Residuals are plotted at the bottom. Dissociation constants are reported below the graphs with ranges in brackets representing the 68.3% confidence interval. *Experiments done on different days were fitted using a corresponding triplicate set of direct titrations (traces in black).

DOI: http://dx.doi.org/10.7554/eLife.23961.006

(A) The difference densities (mFo-DFc contoured at 2.5σ) for mDia2NES when the dataset was refined without the NES peptide. White circle on the left panel shows good side chain density for an arginine side chain. The image in the right panel shows the structure rotated 90 degrees to show strong side chain density for a hydrophobic residue pointing into the P2 pocket of CRM1. (B) Left panels: difference densities (mFo-DFc contoured at 3.0σ; shown in dashed white circles) for the incorrect model of FMRP-1bNES. Right panels: difference densities (mFo-DFc contoured at 3.0σ) for the correct model of FMRP-1bNES are shown at the same positions as the left panels. The absence of difference densities at side chain positions provides support for assignment of 11LRALERLQID20 to the NES peptide structure.

DOI: http://dx.doi.org/10.7554/eLife.23961.007

Differential bleaching experiments of FMRP-1bNES mutant YLKEVDQLRALER show that it does not bind CRM1, thus providing support that assignment of 11LRALERLQID20 to the FMRP-1bNES is most likely correct. Triplicate direct tirations of CRM1 into FITC-PKINES in the presence of excess RanGTP is shown in black, and triplicate competition titrations of MBP-FMRP-1bNES mutant to CRM1 and FITC-PKINES is shown in teal. The KD values obtained are shown below the graph and numbers in brackets represent 68.3% confidence level (U=underteremined, NB=not binding).

DOI: http://dx.doi.org/10.7554/eLife.23961.008

2mFo-DFc map meshes contoured at 1.0σ and kicked OMIT map meshes contoured at 3.0σ are overlaid on the final coordinates of the indicated structures with NES peptides shown as sticks.

DOI: http://dx.doi.org/10.7554/eLife.23961.009

(A) Pull-down binding assay (~10 µg immobilized GST-CDC7NES and the longer GST-CDC7NESext, 2.5 µM CRM1 and 7.5 µM RanGTP in a reaction volume of 200 µL) show that both peptides binding to CRM1. (B) Structure of the longer CDC7NESext peptide (deep teal) bound to the CRM1 groove (grey surface). (C) 2mFo-DFc map (mesh) contoured at 0.8σ and kick OMIT map (mesh) at 2.5σ overlaid onto the final coordinates of CDC7NESext peptide (sticks).

DOI: http://dx.doi.org/10.7554/eLife.23961.010

Sequences of wild type mDia2NES, the mDia2NES(Φ0A) and mDia2NES(AA) mutants are shown along with those for wild type PKINES and the PKINES(Φ4A) mutant. mDia2NES(Φ0A) and PKINES(Φ4A) were mutated to each have only 3 Ф side chains, leucines in positions Ф1, Ф2 and Ф3. Arg1189 and Arg1191 side chains of mDia2NES make electrostatic interactions with the CRM1 groove, and were both mutated to alanines to generate the mDia2NES(AA) mutant. Affinities of MBP-mDia2NES and PKINES mutants binding to CRM1 were measured using differential bleaching competition assays. Fitted binding curves are overlaid onto data points plotted as mean ± SD and residuals are plotted in the bottom. KD values are reported with ranges in brackets representing the 68.3% confidence interval. *Experiments done on different days were fitted using a corresponding triplicate set of direct titrations (in black). mDia2NES(Φ0A) and PKINES(Φ4A) bind CRM1 very differently even though they both have the same 3 Ф (Leu) side chains. mDia2NES(Φ0A) retains CRM1 binding (KD = 9.1 [7.5, 11.1] μM, vs wild type mDia2NESKD = 1.7 [1.3, 2.1] μM) but the PKINES(Φ4A) mutant no longer binds CRM1 (KD = 150 [100, 240] μM vs wild type PKI NESKD = 35 [26, 42] nM). mDia2NES(AA) binds CRM1 2-fold tighter (KD = 800 [600, 1100] nM), suggesting that electrostatic interactions involving Arg1189 and Arg1191 of the NES are not critical for CRM1 binding. Loss of the electrostatic interactions may be compensated by the gain of helical propensity, which may lead to stabilization of the mDia2NES helix.

DOI: http://dx.doi.org/10.7554/eLife.23961.011

Data collection and refinement statistics.

Data collection and refinement statistics (cont.).

Crystallization conditions of CRM1-NES complexes.

Previous structures of CRM1 bound to five different NESs showed virtually identical NES-binding grooves (Fung et al., 2015; Monecke et al., 2009; Dong et al., 2009; Güttler et al., 2010). NESs from Snurportin-1 (SNUPNNES; class 1c) and Protein Kinase A Inhibitor (PKINES; class 1a) bind CRM1 as α-helix followed by a short β-strand, while the proline-rich NES from HIV-1 REV (RevNES; class 2) adopts mostly extended conformation (Figure 1B) (Güttler et al., 2010). The majority of CRM1-NES interactions involve NES hydrophobic anchor side chains, with very few polar and main chain interactions. Previously, we studied NESs with the Φ1XXΦ2XXXΦ3XXΦ4 (class 3) pattern where the i, i + 3, i + 7, i + 10 Φ positions suggested a single long amphipathic helix. However, it is perplexing that a long all-helical peptide could fit in the narrow tapering CRM1 groove. Structures of such NESs from kinase RIO2 and cytoplasmic polyadenylation element-binding protein 4 (hRio2NES, CPEB4NES) showed that they do not adopt all-helical conformations but unexpectedly adopt helix-strand conformations that bind CRM1 in the opposite or minus (−) polypeptide direction to that of SNUPNNES, PKINES and RevNES ((+) NESs) (Fung et al., 2015). hRio2NES and CPEB4NES were hence reclassified as class 1a-R NESs (Figure 1A).

We do not understand the extent of NES structural diversity nor how NESs with different hydrophobic patterns that presumably reflect different secondary structures all bind to the seemingly invariant and three-dimensionally constrained CRM1 groove. Here, eight new structures of CRM1 bound to diverse NESs show several different and unexpected NES backbone conformations that share only a common one-turn helix element. All NESs also participate in hydrogen bonding with Lys568 of CRM1, and mutagenic/structural analysis identifies Lys568 as a selectivity filter that blocks binding of non-NES peptides.

Results

CRM1-bound NESs adopt diverse conformations

We study three NESs that uniquely match the all-helical class 3 pattern (Ф1XXФ2XXXФ3XXФ4). Because most previously studied NESs have substantial helical content, we also study five NESs that match class 2 (Ф1XФ2XXФ3XФ4) and class 1b (Ф1XXФ2XXФ3XФ4) patterns, where the hydrophobic residue positions do not suggest an amphipathic helix (Figure 1A). All eight NESs bind CRM1 in the presence of RanGTP with dissociation constants (KDs) of 670 nM-20 μM, and were crystallized bound to the previously described engineered CRM1-RanGppNHp-Yrb1p complex (Figure 1—figure supplement 1, Figure 1—source data 1, 2 and 3) (Fung et al., 2015). Details of how the NES peptides were modelled can be found in methods section and in Figure 1—figure supplements 2 and 3. NES-bound CRM1 grooves in the structures (2.1–2.4 Å resolution) resemble the PKINES-bound CRM1 groove (Cα/all-atom rmsds 0.5 Å/1.1 Å for 85 groove residues), and all NESs use 4–5 hydrophobic anchor residues to bind P0-P4 hydrophobic pockets of CRM1 (Figures 1D–F and 2) (Güttler et al., 2010).

Figure 1—figure supplement 1.

Binding affinities of NES to CRM1 measured by differential bleaching experiments.

Data from triplicate experiments of direct (titration of CRM1 to FITC-PKINES in the presence of excess RanGTP) or competition titrations (titration of MBP-NES to FITC-PKINES and CRM1 with excess RanGTP) are plotted as mean ± SD and fitted binding curves are overlaid on top. MBP-NES competes with FITC-PKINES for CRM1. Residuals are plotted at the bottom. Dissociation constants are reported below the graphs with ranges in brackets representing the 68.3% confidence interval. *Experiments done on different days were fitted using a corresponding triplicate set of direct titrations (traces in black).

DOI: http://dx.doi.org/10.7554/eLife.23961.006

Figure 1—figure supplement 2.

Modeling strategy for NESs with weak Φ side chain densities.

(A) The difference densities (mFo-DFc contoured at 2.5σ) for mDia2NES when the dataset was refined without the NES peptide. White circle on the left panel shows good side chain density for an arginine side chain. The image in the right panel shows the structure rotated 90 degrees to show strong side chain density for a hydrophobic residue pointing into the P2 pocket of CRM1. (B) Left panels: difference densities (mFo-DFc contoured at 3.0σ; shown in dashed white circles) for the incorrect model of FMRP-1bNES. Right panels: difference densities (mFo-DFc contoured at 3.0σ) for the correct model of FMRP-1bNES are shown at the same positions as the left panels. The absence of difference densities at side chain positions provides support for assignment of 11LRALERLQID20 to the NES peptide structure.

DOI: http://dx.doi.org/10.7554/eLife.23961.007

Figure 1—figure supplement 3.

Mutagenesis of FMRP-1bNES to validate the sequence assignment of the NES.

Differential bleaching experiments of FMRP-1bNES mutant YLKEVDQLRALER show that it does not bind CRM1, thus providing support that assignment of 11LRALERLQID20 to the FMRP-1bNES is most likely correct. Triplicate direct tirations of CRM1 into FITC-PKINES in the presence of excess RanGTP is shown in black, and triplicate competition titrations of MBP-FMRP-1bNES mutant to CRM1 and FITC-PKINES is shown in teal. The KD values obtained are shown below the graph and numbers in brackets represent 68.3% confidence level (U=underteremined, NB=not binding).

DOI: http://dx.doi.org/10.7554/eLife.23961.008

Figure 2.

Structures of NESs with non-helical sequence patterns.

(A–E) FMRPNES (light orange), SMAD4NES (salmon), HDACNES (slate), PaxNES (pale cyan) and FMRP-1bNES (deep teal) bound to the CRM1 groove. Black dashes show CRM1-NES hydrogen bonds and polar contacts, and unoccupied CRM1 hydrophobic pockets are labeled. *HDAC5NES and PaxNES sequences match the class 1b pattern, but both peptides bind CRM1 using Ф residues that match class 1a pattern. Average displacements of each Φ Cα in the eight NESs (including mDia2NES, CDC7NES, X11L2NES in Figure 1) from the equivalent Φ Cα of PKINES(Ф0L) are 1.3 ± 0.6 (Φ4), 0.8 ± 0.5 (Φ3), 0.7 ± 0.4 (Φ2), 0.9 ± 0.3 (Φ1) and 1.8 ± 0.9 (Φ0) Å.

DOI: http://dx.doi.org/10.7554/eLife.23961.012

2mFo-DFc map meshes contoured at 1.0σ and kicked OMIT map meshes contoured at 3.0σ are overlaid on the final coordinates of the indicated structures with NES peptides shown as sticks.

DOI: http://dx.doi.org/10.7554/eLife.23961.013

(A) Sequences of class 1b NESs generated from PKINES and FMRPNES. The binding of mutants to CRM1 were tested in pull-down assay with ~7.5 µg immobilized GST-NESs, 2.5 µM CRM1 and 7.5 µMRanGTP in 200 µL reaction volumes. The reactions were extensively washed and visualized with SDS/PAGE and Coomassie staining. (B) Leptomycin B (LMB) sensitive nuclear export activity of FMRP-1bNES was verified by expressing EYFP2-SV40NLS-FMRP-1bNES in HeLa cells and imaged with spinning disk confocal microscope (60X).

DOI: http://dx.doi.org/10.7554/eLife.23961.014

Class 3 NESs from mouse diaphanous homolog 3 (mDia2NES: 1179SVPEVEALLARLRAL1193) and the cell division cycle 7-related protein kinase (CDC7NES: 456QDLRKLCERLRGMDSSTP473) are indeed all-helix peptides, both forming 3-turn α-helices that occupy only the wide part of the CRM1 groove (Figure 1D,E, Figure 1—figure supplement 4). The last residue of the mDia2 protein (Leu1193) binds the CRM1 P3 pocket leaving the P4 pocket empty. CDC7NES is far from the protein C-terminus but structures of longer peptides suggest that CDC7NESexits the groove after Met468 or Φ3 (Figure 1—figure supplement 5). The mDia2NES and CDC7NESsequence patterns should thus be Φ0XXΦ1XXXΦ2XXΦ3. The Φ4 anchor position is clearly not used in mDia2NES and CDC7NES even though Φ4 is key for activities of several other NESs (Wen et al., 1995; Meyer et al., 1996; Richards et al., 1996; Scott et al., 2002; Tsukahara and Maru, 2004). The number of Φ anchor residues necessary for CRM1 binding can vary between 3–5 (Figure 1—figure supplement 6). A third class 3-matching NES from beta-amyloid binding protein X11L2 (X11L2NES: 55SSLQELVQQFEALPGDLV72) binds differently (Figure 1F, Figure 1—figure supplement 4). 57LQELVQQFEAL67 forms a 3-turn α-helix, 68PGDL71 forms a type I β-turn, and X11L2NES therefore exhibits a new Φ0XXΦ1XXXΦ2XXΦ3XXXΦ4 (class 4) pattern.

Figure 1—figure supplement 4.

Electron densities of the NES peptides.

2mFo-DFc map meshes contoured at 1.0σ and kicked OMIT map meshes contoured at 3.0σ are overlaid on the final coordinates of the indicated structures with NES peptides shown as sticks.

DOI: http://dx.doi.org/10.7554/eLife.23961.009

Figure 1—figure supplement 5.

A longer CDC7NES peptide binds the CRM1 groove without interacting with the P4 pocket.

(A) Pull-down binding assay (~10 µg immobilized GST-CDC7NES and the longer GST-CDC7NESext, 2.5 µM CRM1 and 7.5 µM RanGTP in a reaction volume of 200 µL) show that both peptides binding to CRM1. (B) Structure of the longer CDC7NESext peptide (deep teal) bound to the CRM1 groove (grey surface). (C) 2mFo-DFc map (mesh) contoured at 0.8σ and kick OMIT map (mesh) at 2.5σ overlaid onto the final coordinates of CDC7NESext peptide (sticks).

DOI: http://dx.doi.org/10.7554/eLife.23961.010

Figure 1—figure supplement 6.

Variable number of Ф side chains are needed to have an active NES.

Sequences of wild type mDia2NES, the mDia2NES(Φ0A) and mDia2NES(AA) mutants are shown along with those for wild type PKINES and the PKINES(Φ4A) mutant. mDia2NES(Φ0A) and PKINES(Φ4A) were mutated to each have only 3 Ф side chains, leucines in positions Ф1, Ф2 and Ф3. Arg1189 and Arg1191 side chains of mDia2NES make electrostatic interactions with the CRM1 groove, and were both mutated to alanines to generate the mDia2NES(AA) mutant. Affinities of MBP-mDia2NES and PKINES mutants binding to CRM1 were measured using differential bleaching competition assays. Fitted binding curves are overlaid onto data points plotted as mean ± SD and residuals are plotted in the bottom. KD values are reported with ranges in brackets representing the 68.3% confidence interval. *Experiments done on different days were fitted using a corresponding triplicate set of direct titrations (in black). mDia2NES(Φ0A) and PKINES(Φ4A) bind CRM1 very differently even though they both have the same 3 Ф (Leu) side chains. mDia2NES(Φ0A) retains CRM1 binding (KD = 9.1 [7.5, 11.1] μM, vs wild type mDia2NESKD = 1.7 [1.3, 2.1] μM) but the PKINES(Φ4A) mutant no longer binds CRM1 (KD = 150 [100, 240] μM vs wild type PKI NESKD = 35 [26, 42] nM). mDia2NES(AA) binds CRM1 2-fold tighter (KD = 800 [600, 1100] nM), suggesting that electrostatic interactions involving Arg1189 and Arg1191 of the NES are not critical for CRM1 binding. Loss of the electrostatic interactions may be compensated by the gain of helical propensity, which may lead to stabilization of the mDia2NES helix.

DOI: http://dx.doi.org/10.7554/eLife.23961.011

Structures of NESs with non-helical patterns are also informative. The previous structure of RevNES (class 2) suggested that its three prolines may constrain against a helical conformation (Figure 1B) (Güttler et al., 2010). Class 2 NESs in the Mothers against decapentaplegic homolog 4 protein (SMAD4NES: 133YERVVSPGIDLSGLTLQ149) and the fragile X mental retardation protein (FMRPNES: 423YLKEVDQLRLERLQI437) have few to no prolines but still bind CRM1 with mostly loop-like structures (Figure 2A,B, Figure 2—figure supplement 1). Histone deacetylase 5 (HDAC5NES: 1082EAETVSAMALLSVG1095) and Paxillin (PaxNES: 264RELDELMASLSDFKFMAQ281) have NESs with non-α-helical class 1b pattern (Ф1XXФ2XXФ3XФ4), but the peptides bind according to the class 1a pattern instead (Figure 2C,D, Figure 2—figure supplement 1). This left no other experimentally verified NES in the databases that unambiguously match class 1b pattern (Kosugi et al., 2008; Xu et al., 2012c). We engineered a class 1b NES by adding an alanine to FMRPNES (FMRP-1bNES; YLKEVDQLRALERLQID), which forms a short 1.5-turn 310 helix followed by a 3-residue β-strand (Figure 2E, Figure 2—figure supplements 1 and 2). The 310 helix favorably presents Φ1 and Φ2 into the CRM1 groove and natural class 1b NESs are likely to bind similarly.

Figure 2—figure supplement 1.

Electron densities of the NES peptides.

2mFo-DFc map meshes contoured at 1.0σ and kicked OMIT map meshes contoured at 3.0σ are overlaid on the final coordinates of the indicated structures with NES peptides shown as sticks.

DOI: http://dx.doi.org/10.7554/eLife.23961.013

Figure 2—figure supplement 2.

Engineering class 1b NESs.

(A) Sequences of class 1b NESs generated from PKINES and FMRPNES. The binding of mutants to CRM1 were tested in pull-down assay with ~7.5 µg immobilized GST-NESs, 2.5 µM CRM1 and 7.5 µMRanGTP in 200 µL reaction volumes. The reactions were extensively washed and visualized with SDS/PAGE and Coomassie staining. (B) Leptomycin B (LMB) sensitive nuclear export activity of FMRP-1bNES was verified by expressing EYFP2-SV40NLS-FMRP-1bNES in HeLa cells and imaged with spinning disk confocal microscope (60X).

DOI: http://dx.doi.org/10.7554/eLife.23961.014

Structural requirements for an NES

Structures of >13 different CRM1-bound NESs are now available, and may be sorted into five or six groups according to peptide backbone conformations (Figure 3A). Class 1 NESs are helix-strand peptides with either α-helices (class 1a, 1c) or 310 helices (class 1b). Class 1-R NESs are strand-helix peptides, class 2 NESs are mostly loop-like and class 3 NESs are all-helix peptides. The helix-β-turn X11L2NES structure revealed a new Φ0XXΦ1XXXΦ2XXΦ3XXXΦ4 (class 4) pattern.

Figure 3.

NESs adopt diverse conformations to bind CRM1, sharing only a small one turn of the helix structural element.

(A) Overlay of 13 different CRM1-bound NESs: mDia2NES, CDC7NES, X11L2NES, FMRPNES, SMAD4NES, HDACNES, PaxNES and FMRP-1bNES, SNUPNNES (3GB8), PKINES(Φ0L) (3NBY), RevNES (3NBZ), hRio2NES (5DHF), CPEB4NES (5DIF) (shown as ribbons with Φ residues in sticks; their CRM1 H12A helices were superimposed). A grey rectangle highlights the only secondary structural element shared by all 13 NESs: a single turn of helix. (B) Ramachandran plot of phi/psi angles of the four residues in each of the conserved one-turn of helix. Arrows indicate changes in psi angles along the polypeptide direction. For example, (+) HDAC5NES: Φ2 Ψ = −43.5°, Φ2+1Ψ = −21.7°, Φ2+2Ψ = −1.3°, Φ3 Ψ = 133.8° and (−) CPEB4NES: Φ2+2Ψ = −40.2°, Φ2+1Ψ = −31.0°, Φ2 Ψ = 16.5°, Φ2-1Ψ = 111.5°. *Residues in SNUPNNES plotted are Φ2+1, Φ2+2, Φ2+3 and Φ3. (C) Detailed view of niche motifs in (+) NESs, FMRPNES and mDia2NES.

DOI: http://dx.doi.org/10.7554/eLife.23961.015

Structures of CPEB4NES (5DIF), hRio2NES (5DHF) and PKINES-Flip3 mutant (5DH9) bound to CRM1, showing hydrogen bonds made by the Lys579 side chain of CRM1. Niche motifs observed in (+) NESs are absent here, but in CPEB4NES and PKINES-Flip3 mutant, Lys579 is still coordinated by two main chain carbonyls from Ф1-1 and Ф2-1.

DOI: http://dx.doi.org/10.7554/eLife.23961.016

(A) Pull-down binding assays with ~5 µg immobilized GST-NESs, 500 nM wild type CRM1, the CRM1(K568A) or CRM1(K568M) mutant and 1.5 µM RanGTP in 1 mL reaction volumes. Most of the NESs showed decreased binding to the K568A and K568M mutants compared to wild type CRM1. Because of the limited dynamic range of detection in pull-down assays, it is unclear if binding of hRio2NES and X11L2NES to mutant versus wild type CRM1 proteins are different. (B) Binding affinity of FITC-PKINES to CRM1, CRM1(K568A) and CRM1(K568M) measured using differential bleaching. Different concentration regimes were used for different mutants to obtain meaningful data points and at least three titrations were performed for each CRM1. Fitted binding curves are overlaid onto data points plotted as mean ± SD and residuals are plotted in the bottom. KD values are reported with ranges in brackets representing the 68.3% confidence interval.

DOI: http://dx.doi.org/10.7554/eLife.23961.017

The only common secondary structural element in the NES structures is one turn of NES helix at Φ2X2-3Φ3 (grey box, Figure 3A). This conserved turn of helix is flanked on one side by additional turns of helix (classes 1, 1-R) or by loops (class 2), and on the other side by β-strands (classes 1, 1-R, 2) or β-turn (class 4), or the helix ends as the chain terminates or exits the groove (class 3) (Figure 3A). Dihedral (psi) angles in the 1-turn of helix gradually increase in progression from helical to β-strand conformations (Figure 3B).

In all (+) NESs, main chain carbonyls of Φ2+1 and Φ3 residues in the 1-turn helix element hydrogen bonds with the CRM1 Lys579 (or CRM1/CRM1 Lys568) side chain, much like niche3/4 motifs where carbonyls of residues i and i + 2 or i + 3 coordinate a cationic group (Torrance et al., 2009). The Φ3-Lys579 hydrogen bond is possible only because the β-strand psi angle turns Φ3 carbonyl towards Lys579 (Figure 3C). NES helix-Lys579 hydrogen bonds are absent in (−) NESs as backbone carbonyls point in the opposite direction. Here, carbonyls of the N-terminal β-strand hydrogen bond with Lys579 (Figure 3—figure supplement 1). Therefore, another common structural feature of NESs is hydrogen bonding between NES backbone and CRM1 Lys579 (CRM1 Lys568). Mutations of CRM1 Lys568 impair NES binding. Mutants CRM1(K568A) and CRM1(K568M) bind FITC-PKINES two to three orders of magnitude weaker than wild type CRM1, supporting the importance of Lys568-NES interactions (Figure 3—figure supplement 2).

Figure 3—figure supplement 1.

NES main chain hydrogen bonds with CRM1 Lys579 in (−) NESs.

Structures of CPEB4NES (5DIF), hRio2NES (5DHF) and PKINES-Flip3 mutant (5DH9) bound to CRM1, showing hydrogen bonds made by the Lys579 side chain of CRM1. Niche motifs observed in (+) NESs are absent here, but in CPEB4NES and PKINES-Flip3 mutant, Lys579 is still coordinated by two main chain carbonyls from Ф1-1 and Ф2-1.

DOI: http://dx.doi.org/10.7554/eLife.23961.016

Figure 3—figure supplement 2.

The Lys568 of CRM1 is important for NES binding.

(A) Pull-down binding assays with ~5 µg immobilized GST-NESs, 500 nM wild type CRM1, the CRM1(K568A) or CRM1(K568M) mutant and 1.5 µM RanGTP in 1 mL reaction volumes. Most of the NESs showed decreased binding to the K568A and K568M mutants compared to wild type CRM1. Because of the limited dynamic range of detection in pull-down assays, it is unclear if binding of hRio2NES and X11L2NES to mutant versus wild type CRM1 proteins are different. (B) Binding affinity of FITC-PKINES to CRM1, CRM1(K568A) and CRM1(K568M) measured using differential bleaching. Different concentration regimes were used for different mutants to obtain meaningful data points and at least three titrations were performed for each CRM1. Fitted binding curves are overlaid onto data points plotted as mean ± SD and residuals are plotted in the bottom. KD values are reported with ranges in brackets representing the 68.3% confidence interval.

DOI: http://dx.doi.org/10.7554/eLife.23961.017

In summary, an active NES (1) can use many different backbone conformations to present 3–5 hydrophobic anchor residues into 3–5 CRM1 hydrophobic pockets (P0 and/or P4 are sometimes not used), (2) has one turn of helix with helix-strand transition that binds the central portion of the CRM1 groove and (3) has backbone conformation that can hydrogen bond with Lys568 of CRM1.

CRM1 Lys568 is a selectivity filter for NES recognition

What then are CRM1 groove features that selectively recognize the key NES features? Arrangement of hydrophobic pockets in the groove likely selects NESs with suitably placed Φ residues. Groove shape, tapering and most constricted at CRM1 Lys579 (CRM1 Lys568), likely selects for NES helices that transition to strands or NES helices that end (Figure 3C). Is groove-constricting CRM1 Lys568 perhaps key for differentiating active from false positive NES sequences? We tested mutants CRM1(K568A) and CRM1(K568M) for interactions with three previously identified false positive NESs that match NES consensus but do not bind CRM1: peptides from Hexokinase-2 (Hxk2pep: 18DVPKELMQQIENFEKIFTV36, class 3 match), Deformed Epidermal Autoregulatory Factor 1 homolog (DEAF1pep:452SWLYLEEMVNSLLNTAQQ469; class 1a-R match) and COMM domain-containing protein 1 (COMMD1pep; 173KTLSEVEESISTLISQPN190; class 3 match) (Figure 4A) (Xu et al., 2012c, 2015). Wild type CRM1 does not bind the peptides but CRM1(K568A) binds Hxk2pep and DEAF1pep, and CRM1(K568M) binds DEAF1pep but not Hxk2pep, suggesting that Lys568 is important in filtering out false positive NESs (Figure 4A).

Figure 4.

CRM1 Lys568 is a selectivity filter.

(A) False positive NES sequences with Φ residues of consensus matches underlined. Pull-down binding assay of ~5 µg immobilized GST-NESs and 7.5 µM RanGTP with 2.5 µM of either wild type CRM1 or mutant CRM1(K568A) or CRM1(K568M) in 200 µL reactions. (C–D) Structures of Hxk2pep (pink) (C) and DEAF1pep (purple) (D) bound to CRM1(K579A). Left panels, peptides were removed to show the surface of the mutant CRM1(K579A) groove. Middle panels, peptides bound to the mutant CRM1(K579A). Right panels, CRM1(K579A)-bound Hxk2pep and DEAF1pep superimposed onto the PKINES(Ф0L)-bound CRM1 groove (3NBY; CRM1 H12A helices were aligned; PKINES not shown) to show steric clash of the Hxk2pep and DEAF1pep peptides with the CRM1 Lys568 side chain.

DOI: http://dx.doi.org/10.7554/eLife.23961.018

DOI: http://dx.doi.org/10.7554/eLife.23961.019

2mFo-DFc map meshes contoured at 0.8σ and kicked OMIT map meshes contoured at 3.0σ are overlaid on the final coordinates of the indicated structures with NES peptides shown as sticks.

DOI: http://dx.doi.org/10.7554/eLife.23961.020

(A) Hxk2pep-CRM1(K579A) (Hxk2pep in pink cartoon and CRM1(K579A) shown as grey surface) superimposed onto the PKINES(Ф0L)-bound CRM1 groove (wild type CRM1, 3NBY; blue surface), the PaxNES- bound CRM1 groove (pale cyan), the FMRPNES-bound CRM1 (wheat) and the mDia2NES-bound CRM1 (pale green) by aligning their CRM1 H12A helices. The superpositions show steric clash of the Hxk2pep peptide with the CRM1 Lys568 or CRM1 Lys579 side chains. (B) DEAF1pep-CRM1(K579A) (DEAFpep in magenta cartoon and CRM1(K579A) shown as grey surface) superimposed onto the PKINES(Ф0L)-bound CRM1 groove (wild type CRM1, 3NBY; blue surface), the hRio2NES-bound CRM1 (orange, 5DHF) and the CPEB4NES-bound CRM1 (yellow, 5DIF). The N-terminus of DEAF1pep clashes with of CRM1 Lys568 orCRM1 Lys579 side chains. Zoomed in views are rotated to show proximity of DEAF1pep Ф1XФ2 to the CRM1 Lys579 side chain.

DOI: http://dx.doi.org/10.7554/eLife.23961.021

Data collection and refinement statistics and crystallization conditions.

Both CRM1(K579A)-bound Hxk2pep and DEAF1pep are all-helix peptides (Figure 4B,C, Figure 4—figure supplement 1, Figure 4—source data 1). The fourth turn of the Hxk2pep helix packs into hydrophobic space widened by removal and rearrangement of the Lys579 and Glu582 side chains, respectively (Figure 4B). The 2.5-turn α-helix of DEAF1pepbinds in the (−) direction and is slightly longer than helices in true (−) NESs (Figure 4C). Superpositions of Hxk2pep and DEAF1pep onto wild type CRM1 grooves show the fourth turn of the Hxk2pep helix and the N-terminus of the DEAF1pep helix clashing with CRM1 Lys579/CRM1 Lys568 side chains (Figure 4B,C, Figure 4—figure supplement 2). The rest of the mutant CRM1(K579A) groove is highly similar to the wild type groove. Therefore, the key feature of the wild type groove that prevents Hxk2pep and DEAF1pep binding is Lys568, which is not only a critical hydrogen bond donor for binding NESs, but its long side chain also blocks binding of sequences that do not meet NES structural requirements.

Figure 4—figure supplement 1.

Electron densities of the HXK2pep and DEAFpep false positive NES peptides bound to the CRM1(K579A) mutant.

2mFo-DFc map meshes contoured at 0.8σ and kicked OMIT map meshes contoured at 3.0σ are overlaid on the final coordinates of the indicated structures with NES peptides shown as sticks.

DOI: http://dx.doi.org/10.7554/eLife.23961.020

Figure 4—figure supplement 2.

Additional overlays of Hxk2pep and DEAF1pep onto NES-bound wild type CRM1 grooves.

(A) Hxk2pep-CRM1(K579A) (Hxk2pep in pink cartoon and CRM1(K579A) shown as grey surface) superimposed onto the PKINES(Ф0L)-bound CRM1 groove (wild type CRM1, 3NBY; blue surface), the PaxNES- bound CRM1 groove (pale cyan), the FMRPNES-bound CRM1 (wheat) and the mDia2NES-bound CRM1 (pale green) by aligning their CRM1 H12A helices. The superpositions show steric clash of the Hxk2pep peptide with the CRM1 Lys568 or CRM1 Lys579 side chains. (B) DEAF1pep-CRM1(K579A) (DEAFpep in magenta cartoon and CRM1(K579A) shown as grey surface) superimposed onto the PKINES(Ф0L)-bound CRM1 groove (wild type CRM1, 3NBY; blue surface), the hRio2NES-bound CRM1 (orange, 5DHF) and the CPEB4NES-bound CRM1 (yellow, 5DIF). The N-terminus of DEAF1pep clashes with of CRM1 Lys568 orCRM1 Lys579 side chains. Zoomed in views are rotated to show proximity of DEAF1pep Ф1XФ2 to the CRM1 Lys579 side chain.

DOI: http://dx.doi.org/10.7554/eLife.23961.021

Discussions

Class 1a, 1b, 1c, 2, 3, 4 and 1a-R NES structures show 5–6 distinct backbone conformations that match their respective hydrophobic sequence patterns. We can infer structures of remaining NES classes: class 1d NESs (Ф1XXФ2XXXФ3XФ4) are likely α-helix-strand and other (−) NES classes are likely the reverse of their (+) counterparts. Symmetrical class 2, 3 and 4 patterns are identical in both (+)/(−) directions but (−) class 3 and 4 NESs, lacking β-strands to hydrogen bond with CRM1 Lys568, may not be ideal NESs.

Structures of many diverse NES sequences suggest how one unchanging peptide-bound CRM1 groove can recognize up to a thousand different peptides. Dependence of 3–5 hydrophobic residues in 8–15 residues-long NES arises from the substantial binding energy of anchor hydrophobic side chains interacting with 3–5 CRM1 hydrophobic pockets. However, lack of contact with NES backbone allows anchor side chains to be presented in many conformations including both N- to C-terminal orientations, explaining broad specificity defined by highly variable spacings between anchors. Interestingly, NES conformation is not entirely unrestrained, as CRM1 groove constriction imposes either exit/termination of the NES chain or its continuation in extended configuration. Solutions for the broadly specific NES recognition contrast with those of analogous systems. MHC I and II proteins, each recognizing at least hundreds of different peptide antigens, use many peptide main chain contacts for affinity with only a few supplementary peptide side chain interactions (Zhang et al., 1998; Madden, 1995). The result here is a conformational selection of particular lengths of extended peptides binding in conserved N- to C-terminal orientation, with little sequence restriction. The Calmodulin-helical peptide system is yet another contrast, as the binding domain uses its flexible fold to adapt to various helical ligands (Tidow and Nissen, 2013; Hoeflich and Ikura, 2002).

CRM1-NES structures expanded the six NES patterns derived from peptide library studies to the eleven patterns shown in Figures 1A and 3A. The ever-expanding set of NES patterns suggests that no fixed hydrophobic pattern likely describes the NES. Furthermore, only ~50% of consensus-matching previously reported NESs that were tested actually bound CRM1, contributing to the inefficiency of available NES predictors (with precision of 50% at 20% recall rate) (Xu et al., 2012b, 2015; Kosugi et al., 2014; Fu et al., 2011). The many available NES structures, diversity of NES conformations and the structurally conserved one-turn helix NES element revealed here will enable development of structure- rather than sequence-based NES predictors (Raveh et al., 2011; Schindler et al., 2015; Trellet et al., 2013; Yan et al., 2016). There is a need to identify many more CRM1 cargoes as apoptosis of different cancer cells upon CRM1 inhibition by the drug Selinexor (in clinical trials for a variety of cancers) and other inhibitors (Parikh et al., 2014; Mendonca et al., 2014; Das et al., 2015; Alexander et al., 2016; Gounder et al., 2016; Abdul Razak et al., 2016; Lapalombella et al., 2012; Inoue et al., 2013; Etchin et al., 2013; Tai et al., 2014; Cheng et al., 2014; Kim et al., 2016; Hing et al., 2016; Vercruysse et al., 2016) appears to be driven by nuclear accumulation of different sets of NES-containing cargoes, but identities of most of these apoptosis-causing cargoes are still unknown.

Finally, we find that the CRM1 Lys568 side chain acts as a filter that physically selects for NESs with helices that transition to strands or end at the narrow part of the CRM1 groove. Interestingly, Lys568 interacts electrostatically with CRM1 Glu571, which is mutated to glycine or lysine in chronic lymphocytic leukemia and lymphomas with poor prognosis (Puente et al., 2011; Jardin et al., 2016; Camus et al., 2016). Disease mutations will abolish Glu571-Lys568 contacts and possibly affect NES binding and selectivity.

Materials and methods

Crystallization of CRM1-Ran-RanBP1-NES complexes

CRM1 (CRM1 residues 1–1058, △377–413, 537DLTVK541 to GLCEQ, V441D), Yrb1p (yeast RanBP1 residues 62–201), human Ran full length and various NESs were expressed and purified as described in Fung et al. (Fung et al., 2015). CRM1 (K579A) mutant was expressed in pGex-TEV and purified like CRM1. Crystallization, data collection and processing, and solving of the structures were also performed in the same manner as previously described. X-ray/stereochemistry weight and X-ray/ADP weight were both optimized by phenix.refine in PHENIX (RRID:SCR_014224).

NES peptides are modeled according to the positive difference density (2mFo-DFc map) at the binding groove after refinement of the CRM1-Ran-RanBP1 model. In all structures, there are good electron densities for the NES main chain and directions of side chain density in the helical portion of the peptides allow unambiguous determination that they are all oriented in the positive (+) NES orientation. Side chain assignments of the NES peptides are guided by (1) densities of Ф side chains that point into the binding groove, (2) densities for long non-Ф side chains such as arginine, phenylalanine and methionine and (3) physical considerations such as steric clashes.

For example, to model the bound mDia2NES peptide (sequence: GGSY-1179SVPEVEALLARLRAL1193), we made use of the obvious electron densities (mFo-DFc map) for long side chains to guide sequence assignment. There is a strong side chain density suitable for an arginine side chain on the peptide (white dashed circle in Figure 1—figure supplement 2A). There are only two arginine residues in the mDia2NES peptide, Arg1189 and Arg1191. If the long side chain density is assigned to Arg1189, then Arg1191 would end up pointing into the binding groove – a very energetically unfavorable and unlikely situation. Furthermore, Ala1188 would end up in the P2 pocket of CRM1 where there is an obvious density for a longer hydrophobic side chain (left panel, Figure 1—figure supplement 2A). On the other hand, when Arg1191 is assigned to the long and continuous side chain density (adjacent to helix H12A of CRM1), the remaining side chains in the NES end up in positions that are consistent the electron densities.

For the FMRP-1bNES (sequence: 1GGS-YLKEVDQLRALERLQID20), there are no obvious long side chain densities that could help with modeling. There are however obvious densities for several side chains in the first two turn of the NES helix. These side chain densities are consistent with two possible sequence assignments: 5LKEVDQLRAL14 or the more C-terminal 11LRALERLQID20. We tested modeling of FMRP-1bNES by refining both peptide models and by testing a mutant peptide that should distinguish between the two models. Ten cycles of PHENIX refinement of the 5LKEVDQLRAL14 model resulted in positive and negative difference densities (mFo-DFc map) at several NES side chains, which suggested an incorrect assignment (left panels, Figure 1—figure supplement 2B). In contrast, different densities are absent when the 11LRALERLQID20 model is refined (right panels, Figure 1—figure supplement 2B). The final FMRP-1bNES structure was therefore modelled as 11LRALERLQID20. The sequence assignment of FMRP-1bNES was also tested using a mutant FMRP-1bNES that has the sequence YLKEVDQLRALER. If the NES is 5LKEVDQLRAL14, FMRP-1bNES mutant YLKEVDQLRALER should bind well to CRM1. However, if 11LRALERLQID20 is the FMRP-1bNES, mutant YLKEVDQLRALER should not bind CRM1 as the C-terminal half of the NES or 17LQID20 which includes Ф3 and Ф4 is missing. Results in Figure 1—figure supplement 3 show that FMRP-1bNES mutant YLKEVDQLRALER does not bind CRM1, providing further support that the NES is indeed 11LRALERLQID20 as currently assigned.

NES activity assays

Pull-down binding assays, in vivo NES activity assay and differential bleaching experiments for determining binding affinities were all performed the same way as described in Fung et al. (2015). The data were analyzed in PALMIST (Scheuermann et al., 2016) and plotted with GUSSI (Brautigam, 2015).

Accession codes

Structures and crystallographic data have been deposited at the PDB: 5UWI (CRM1-HDAC5NES), 5UWH (CRM1-PaxNES), 5UWO (CRM1-FMRP-1bNES), 5UWJ (CRM1-FMRPNES), 5UWU (CRM1-SMAD4NES) 5UWP (CRM1-mDia2NES), 5UWQ (CRM1-CDC7NES), 5UWR (CRM1-CDC7NES ext), 5UWS (CRM1-X11L2NES), 5UWT (CRM1(K579A)-Hxk2pep), 5UWW (CRM1(K579A)-DEAF1pep).

In the interests of transparency, eLife includes the editorial decision letter and accompanying author responses. A lightly edited version of the letter sent to the authors after peer review is shown, indicating the most substantive concerns; minor comments are not usually included.

Thank you for submitting your article "CRM1 recognizes diverse Nuclear Export Signals conformations" for consideration by eLife. Your article has been reviewed by two peer reviewers, and the evaluation has been overseen by a Reviewing Editor and John Kuriyan as the Senior Editor. The following individuals involved in review of your submission have agreed to reveal their identity: Gino Cingolani (Reviewer #1) and Douglas Barrick (Reviewer #2).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Summary:

This "Research Advances" manuscript describes continued structural analysis of NES sequences bound to CRM1, a nuclear export receptor. In the previous 2015 eLife paper, the authors published two crystal structures of bound sequences and showed that they can interact in different ways, reversing chain orientation. In the present study, they solve eight additional structures from diverse sequences, and characterize additional variation in the mode of binding. The results demonstrate that a 'folded' secondary structure is not a prerequisite to bear a functional NES. Instead, sidechains from the NES sequences arranged in conformations that match the tight and restrained groove generated at the HEAT:HEAT interface are the key determinants. The results help to define different sequence motifs for binding, and how they translate into structure. The present study is certainly an advance from the previous work. As such, the manuscript is favorable for the format of "Research Advances" in eLife.

Essential revisions:

1) One major comment to be addressed in revision is that the authors should do a better job in Figure 1 showing the sequences in each NES subfamily, their expected structures, and their actual structures (perhaps a string of h's, s's, and t's). And importantly, indicate which sequences are from the current study, and which are from previous studies. Without a better guide, most readers won't remember which NES sequence has which structure, sequence motif, and orientation.

2) Although all of the crystal structures are solved at a resolution ranging from 2.0~2.5 Å, the electron densities for some of the NES peptides may not have sufficient quality for reliably model building (some of the sidechain densities for the critical hydrophobic residues are missing, e.g. FMRP-1b, mDia-2, etc.). The overall statistics (R factor and R-free factor) cannot reflect this because the peptide only constitutes a small portion of the entire structure. Moreover, the B factors for some of the peptides are extremely high compared to the CRM1*-Ran-RanBP1 protein, suggesting overfitting of the peptides to some extent. The authors need to address this point by stating the rationales for modelling the peptides as they reported.

Essential revisions:

1) One major comment to be addressed in revision is that the authors should do a better job in

To better show the sequence patterns of each NES subfamily, we changed the font in Figure 1A of the manuscript to the monospace Courier New font. To incorporate information about potential α-helices within consensus sequences of the different NES classes or subfamily, we shaded these regions grey. These regions of the NESs are predicted to be amphipathic α-helices because their hydrophobic residues form patterns of i, i+4, i+7 or i, i+3, i+7 or i, i+3, i+7, i+10. This explanation of how we predict α-helical parts of NESs is now included in 1) the legend for Figure 1, 2) the revised Introduction where we added new text to better explain the unexpected previously solved structures of CRM1-bound hRio2NES and CPEB4NES which has the potentially all-helical class 3 pattern, and 3) the Results section where we revised the text to include rationale for the potentially non-helical class 2 and class 1b NESs.

We also added information into Figure 1B, D-F and Figure 2 of the manuscript so that they can serve as better guides to clearly distinguish previous structures from new structures, and connect sequence motifs with peptide structure/orientation:

1) We added prominently displayed PDB IDs to the structures in Figure 1B of the manuscript to clearly indicate that all three peptides are previously published structures.

2) In Figure 1B, we also added the sequence of each NES and their respective NES class, marked Φ0-Φ4 residues in the sequences and labeled one Φ residue of each peptide structure to connect the structures to the sequences.

3) In Figure 1D-F and Figure 2, we clearly indicated the NES class or subfamily for each NES peptide, marked the Φ0-Φ4 residues in the sequences, and one Φ residue in each peptide structure.

Changes in the text of the manuscript:

We revised the legend for Figure 1 according to the changes above: “(A) NES sequence patterns (Φ is Leu, Val, Ile, Phe or Met and X is any amino acid). Potential amphipathic α-helices, predicted with hydrophobic patterns of i, i+4, i+7 or i, i+3, i+7 or i, i+3, i+7, i+10, are shaded grey.”

We also revised text in the Introduction and Results of the manuscript. The revised text in Introduction reads: “Previously, we studied NESs with the Φ1XXΦ2XXXΦ3XXΦ4 (class 3) pattern where the i, i+3, i+7, i+10 Φ positions suggested a single long amphipathic helix. […] Structures of such NESs from kinase RIO2 and cytoplasmic polyadenylation element-binding protein 4 (hRio2NES, CPEB4NES) showed that they do not adopt all-helical conformations but unexpectedly adopt helix-strand conformations that bind CRM1 in the opposite or minus (-) polypeptide direction to that of SNUPNNES, PKINES and RevNES ((+) NESs) (16).”

The revised text in Results reads: “We study three NESs that uniquely match the all-helical class 3 pattern (Ф1XXФ2XXXФ3XXФ4). Because most previously studied NESs have substantial helical content, we also study five NESs that match class 2 (Ф1XФ2XXФ3XФ4) and class 1b (Ф1XXФ2XXФ3XФ4) patterns, where the hydrophobic residue positions do not suggest an amphipathic helix (Figure 1A).”

2) Although all of the crystal structures are solved at a resolution ranging from 2.0~2.5 Å, the electron densities for some of the NES peptides may not have sufficient quality for reliably model building (some of the sidechain densities for the critical hydrophobic residues are missing, e.g. FMRP-1b, mDia-2, etc.). The overall statistics (R factor and R-free factor) cannot reflect this because the peptide only constitutes a small portion of the entire structure. Moreover, the B factors for some of the peptides are extremely high compared to the CRM1*-Ran-RanBP1 protein, suggesting overfitting of the peptides to some extent. The authors need to address this point by stating the rationales for modelling the peptides as they reported.

We thank the reviewers for pointing out the need to provide clear rationale for the way we modeled the NES peptides. As they noted, densities are missing for the Φ1 and Φ3 side chains of mDia2NES and for the Φ1 side chain of FMRP-1bNES.

To model the bound mDia2NES peptide (sequence: GGSY-1179SVPEVEALLARLRAL1193), we made use of the obvious electron densities (mFo-DFc map) for long side chains to guide sequence assignment. There is strong side chain density suitable for an arginine side chain on the peptide (white dashed circle in Figure 1—figure supplement 2A). There are only two arginine residues in the mDia2NES peptide, Arg1189 and Arg1191. If the long side chain density is assigned to Arg1189, then Arg1191 would end up pointing into the binding groove – a very energetically unfavorable and unlikely situation. Furthermore, Ala1188 would end up in the P2 pocket of CRM1 where there is obvious density for a longer hydrophobic side chain (left panel of Figure 1—figure supplement 2A). On the other hand, when Arg1191 is assigned to the long and continuous side chain density (adjacent to helix H12A of CRM1), the remaining side chains in the NES end up in positions that are consistent the electron densities.

For the FMRP-1bNES (sequence: 1GGS-YLKEVDQLRALERLQID20), there are no obvious long side chain densities that could help with modeling. There are however obvious densities for several side chains in the first two turn of the NES helix. These side chain densities are consistent with two possible sequence assignments: 5LKEVDQLRAL14 or the more C-terminal 11LRALERLQID20. We tested modeling of FMRP-1bNES by refining both peptide models and by testing a mutant peptide that should distinguish between the two models. Ten cycles of PHENIX refinement of the 5LKEVDQLRAL14 model resulted in positive and negative difference densities (mFo-DFc map) at several NES side chains, which suggested an incorrect assignment (left panels, Figure 1—figure supplement 2B). In contrast, difference densities are absent when the 11LRALERLQID20 model is refined (right panels, Figure 1—figure supplement 2B). The final FMRP-1bNES structure was therefore modelled as 11LRALERLQID20.

The sequence assignment of FMRP-1bNES was also tested in an additional experiement using a mutant FMRP-1bNES that has the sequence YLKEVDQLRALER. If the NES is 5LKEVDQLRAL14, FMRP-1bNES mutant YLKEVDQLRALER should bind well to CRM1. However, if 11LRALERLQID20 is the FMRP-1bNES, mutant YLKEVDQLRALER should not bind CRM1 as the C-terminal half of the NES or 17LQID20 which includes Ф3 and Ф4 is missing. New experimental results in Figure 1—figure supplement 3 show that FMRP-1bNES mutant YLKEVDQLRALER does not bind CRM1, providing further support that the NES is indeed 11LRALERLQID20 as currently assigned.

We also agree with the reviewers that B-factors of the NES peptides, which are mostly between 60-90 Å2, with the expection of CDC7NES-ext and Hxk2pep at 90-100 Å2, are high compared to the overall average B-factors for the proteins in the crystalsin all structures). We examined the average B-factors for 29 CRM1 residues that line the NES groove, and noticed that they are in the 50-70 Å2 range, higher than average B-factors for the rest of CRM1 but lower than NES B-factors. The high NES B-factors are likely due to partial occupancy of the NES peptides in the CRM1 grooves. Our peptide models remain reliable despite high B-factors as kick OMIT densities calculated with NES peptides omitted clearly show presence of the NES peptides and the omit densities also match the final peptide models well (Figure 1—figure supplement 4 and Figure 2—figure supplement 1).

Changes in the text of the manuscript:

We added a sentence in main text of the manuscript (Results section) to refer to the Methods section for details on our modeling strategy: “All eight NESs bind CRM1 in the presence of RanGTP with dissociation constants (KDs) of 670 nM-20 μM, and were crystallized bound to the previously described engineered CRM1-RanGppNHp-Yrb1p complex (Figure 1—figure supplement 1, Figure 1—source data 1, 2, 3) (Xu et al., 2012). Details of how the NES peptides were modelled can be found in the Methods section and in Figure 1—figure supplement 2, 3.”

We also added the following description of how peptides are modeled to the Methods section:

“NES peptides are modeled according to the positive difference density (2mFo-DFc map) at the binding groove after refinement of the CRM1-Ran-RanBP1 model. […] Results in Figure 1—figure supplement 3 show that FMRP-1bNES mutant YLKEVDQLRALER does not bind CRM1, providing further support that the NES is indeed 11LRALERLQID20 as currently assigned.”