Dock-and-lock binding of SxIP ligands is required for stable and selective EB1 interactions
- Department of Biochemistry, Cell Signalling and Systems Biology, Institute of Systems, Molecular and Integrative Biology, University of Liverpool, United Kingdom
- Department of Molecular and Clinical Cancer Medicine, Institute of Systems, Molecular and Integrative Biology, University of Liverpool, United Kingdom
Figures
Figure 1 with 3 supplements
Residues that follow the SxIP motif enhance binding by engaging with the C-terminus of the EB1 EBH domain.
(A) Crystal structure of the EBH domain in the complex with microtubule-actin cross-linking factor (MACF) peptide (PDB ID 3GJO). EB1 is shown as a cartoon with the subunits of the dimer coloured in green and cyan, and a semi-transparent surface. The MACF peptide is coloured orange, with the SxIP motif chains-chains shown as sticks. The zoomed region highlights the binding pocket of EBH formed by the surface of the coiled-coil and the folded C-terminus. (B) Zoom on the partly formed SxIP binding pocket in the structure EBH domain free in solution (PDB ID 3EVI) in the same orientation as (A). The C-terminal region is unfolded. (C) EBH (50 µM) chemical shift changes in the 1H,15N-HSQC spectra induced by 4MACF (blue, 100-fold excess), 6MACF (green, 100-fold excess), and 11MACF (red, 4-fold excess) peptides. (D) Relative intensities of cross-peaks (normalised to the intensity of the C-terminal residue G260) in the 1H,15N-HSQC spectra of the EBH (50 µM) free in solution (black) and in the presence of 4MACF (blue, 100-fold excess), 6MACF (green, 100-fold excess), and 11MACF (red, 4-fold excess) peptides. (E) 'Dock-and-lock’ binding model that explicitly considers the role of EBH C-terminus in the interaction with the SxIP peptide. Initially, the binding pocket is partially formed and only contains SxIP-recognition region. Following the initial binding (‘dock’), the post-SxIP region of the peptide induces the folding of the of the C-terminus and formation of the full binding pocket (‘lock’). The deletion of the EBH C-terminus (orange) removes the ‘Lock’ stage of the binding, thus reducing the affinity of the interaction. The peptide is shown as a coloured bar, with red corresponding to the SxIP and blue to the post-SxIP regions.
-
Figure 1—source data 1
Values of the chemical shift changes in the 1H,15N-HSQC spectra induced by microtubule-actin cross-linking factor (MACF) peptides for Figure 1C.
- https://cdn.elifesciences.org/articles/98063/elife-98063-fig1-data1-v1.xlsx
-
Figure 1—source data 2
Values of the relative intensities of cross-peaks (normalised to the intensity of the C-terminal residue G260) in the 1H,15N-HSQC spectra of the EBH complexes with microtubule-actin cross-linking factor (MACF) peptides for Figure 1D.
- https://cdn.elifesciences.org/articles/98063/elife-98063-fig1-data2-v1.xlsx
Figure 1—figure supplement 1
Post-SxIP residues increase the binding affinity.
(A, B, and C) Superposition of 1H,15N-HSQC spectra of 15N-labelled EBH domain of EB1 (50 µM) recorded at 600 MHz for the free form (black) and in the presence of the peptide (black), (A) 4MACF (5000 µM), (B) 6MACF (5000 µM), and (C) 11MACF (400 µM). The insets show changes in the spectra on the increase of the peptide concentration: (A) 250, 500, 1000, 2500, and 5000 µM of 4MACF, (B) 250, 500, 1000, 2500, and 5000 µM of 6MACF, and (C) 12.5, 25, 75, 82.5, 100, 125, 150, 200, and 400 µM of 11MACF. (D and E) Chemical shift changes (blue) of the residues with the largest induced shifts fitted into the single site binding model (red) to evaluate the dissociation constant KD for (D) 4MACF and (E) 6MACF. (F) Isothermal titration calorimetry (ITC) isotherm for the interaction of EBH with 11MACF (left) and 9MACF (right). The solid line shows the fitting of the data into a single-site binding model.
Figure 1—figure supplement 2
Mapping of the largest chemical shift changes induced by the peptide binding on the surface of EB1 EBH domain (A–C) and EBH EBH-ΔC (D–F) in the complex with the microtubule-actin cross-linking factor (MACF) peptide (green, shown in a stick representation).
The crystal structure was used to represent the complex (PDB ID 3GJO). The EBH C-terminus has been removed for EBHΔC. (A–C) Changes induced in the 1H,15N-HSQC signals of EBH by (A) 4MACF, (B) 6MACF, and (C) 11MACF peptides. (D–F) Changes induced in the 1H,15N-HSQC signals of EBHΔC by (D) 4MACF, (E) 6MACF, and (F) 11MACF peptides.
Figure 1—figure supplement 3
Deletion of the EBH C-terminus reduces the binding affinity.
(A, B, and C) Superposition of 1H,15N-HSQC spectra of 15N-labelled EBH-ΔC domain of EB1 (50 µM) recorded at 600 MHz for the free form (black) and in the presence of the peptide (black): (A) 4MACF (5000 µM), (B) 6MACF (5000 µM), and (C) 11MACF (400 µM). The insets show changes in the spectra on the increase of the peptide concentration: (A) 250, 500, 1000, 2500, and 5000 µM of 4MACF, (B) 250, 500, 1000, 2500, and 5000 µM of 6MACF, and (C) 12.5, 25, 75, 82.5, 100, 125, 150, 200, and 400 µM of 11MACF. (D) Chemical shift changes (blue) of the residues with the largest induced shifts fitted into a single-site binding model (red) to evaluate the dissociation constant KD for 6MACF. (E) Isothermal titration calorimetry (ITC) isotherm for the interaction of EBH with 11MACF. The solid line shows the fitting of the data into a single-site binding model. (F) Thermodynamic parameters (△G, blue; △H, red; –T△S, green) evaluated from the ITC data for the 11MACF binding to the EBH (left) and EBH△C (right).
Figure 2 with 2 supplements
Structure and dynamics of the EB1 EB1 homology (EBH) domain.
(A) Superposition of 20 NMR structures calculated for EBH in the complex with 11MACF peptide (left). Previously reported NMR structures of the free EBH (PDB ID 6EVI, middle) and the EBH complex with a small SxIP-like molecule 1a (PDB ID 6EVI, right) are shown for comparison. The EBH subunits are coloured in green and cyan, the peptide and the small molecule in orange and magenta. (B) Representation of the residues forming the contact interface between the EBH and 11MACF peptide, both shown as cartoon, with side chain displayed for all the residues involved in the contacts between the two molecules. EB1 is coloured in grey and 11MACF in green, with oxygen shown in red and nitrogen shown in blue. (C) Order parameters S2 (left) and exchange contributions into the relaxation rate Rex (right) calculated from the relaxation parameters for the free EBH (top) and EBH in complex with 11MACF peptide (bottom) mapped on the EBH solution structure. The thickness of the tube is proportional to 1 S2 (left) or Rex (right).
Figure 2—figure supplement 1
Relaxation parameters measure at 600 (blue) and 800 (orange) MHz for free EB1 homology (EBH) (left) and EBH in the complex with 11MACF (right).
From top to bottom: relaxation rates R1, R2, ratio R2/R1, 1H-15N NOE, and isotropic correlation time directly calculated for each residue using isotropic model. The secondary structure of the EBH domain is shown schematically at the bottom.
-
Figure 2—figure supplement 1—source data 1
Values of relaxation parameters for the free EB1 homology (EBH) and EBH complex with 11MACF peptide.
- https://cdn.elifesciences.org/articles/98063/elife-98063-fig2-figsupp1-data1-v1.txt
Figure 2—figure supplement 2
Order parameters S2 and exchange rates Rex calculated from the relaxation data for free EB1 homology (EBH) (left) and EBH in complex with the 11MACF peptide (right).
-
Figure 2—figure supplement 2—source data 1
Calculated S2 and Rex parameters for the free EB1 homology (EBH) and EBH complex with 11MACF peptide.
- https://cdn.elifesciences.org/articles/98063/elife-98063-fig2-figsupp2-data1-v1.txt
Figure 3 with 1 supplement
Enhancement of the peptide binding through the substitution of the post-SxIP residues.
(A) Folding of the C-terminus brings the hydrophobic EB1 homology (EBH) residues of that region into contact with the TPQ region of the peptide. Structure of the free EBH (left) and EBH in the complex with 11MACF (right). (B) Proximity of the positively charged RK residues of the peptide to the negatively charged patch on the EBH surface. The peptide is shown as a cartoon, RK side chains are shown in a stick representation. Positive and negative electrostatic surface potential is represented by blue and red colour, respectively. (C) The isothermal titration calorimetry (ITC) titration (top) and the binding isotherm fitted into a single-site binding model (bottom) of the ITC binding experiments for the 11MACF peptide and the mutants 11MACF-LLL, 11MACF-VLL, and 11MACF-VLLRK. The peptide sequences are shown above the graphs, with the changed regions highlighted by the red (mutation) and green (insertion) colours. (D) The thermodynamic parameters ∆G (green), ∆H (blue), and –T∆S (red) calculated from the isothermal titration calorimetry (ITC) data.
-
Figure 3—source data 1
Values of the thermodynamic parameters for Figure 3D.
- https://cdn.elifesciences.org/articles/98063/elife-98063-fig3-data1-v1.xlsx
Figure 3—figure supplement 1
Best scored docking poses obtained for (A) 11MACF-WT, (B) 11MACF-LLL, (C) 11MACF-VLL, and (D) 11MACF-ILL.
The EB1 homology (EBH) structure is shown as a semi-transparent grey surface and cartoon. The peptide is shown in stick representation. Carbon atoms are coloured in magenta, oxygen is shown in red, and nitrogen in blue. Hydrogen bonds are shown as yellow dashed lines and hydrophobic interaction regions are highlighted by green dashed lines. The SxIP region is boxed.
Figure 4 with 4 supplements
Exchange rates and binding energies for the interaction of EB1 homology (EBH) domain with the SxIP ligands.
(A) 11MACF-LLL and 11MACF-VLL peptides induce increased chemical shift changes compared to the 11MACF. Chemical shift differences in the 1H,15N-HSQC spectra between the free EBH and the EBH/11MACF complex (blue), EBH/11MACF and EBH/11MACF-LLL (orange), and EBH/11MACF and EBH/11MACF-VLL (green). (B) Chemical shift changes in the 1H,15N-HSQC spectra on peptide addition observed for the EBH interactions with different peptides illustrated for the Thr249 signal. Superposition of the spectra for the titration of the EBH-ΔC with 11MACF (left), EBH with 11MACF (middle), and EBH with 11MACF-VLL (right). Signals of the free EBH are shown in red, fully bound EBH in blue, and the intermediate titration points are shown in the pale colours. Notice two additional signals that are observed for the EBH/11MACF-VLL titration at the intermediate concentrations corresponding to the non-symmetrical form where EBH dimer binds a single peptide. These signals can only be observed when the exchange between the different forms is very slow. (C) Example of the chemical exchange saturation transfer (CEST) profiles measured for the EBH/11MACF interaction at the irradiation field strength of 12.5, 25, and 50 Hz (left to right) for Asp250. Solid curve represents the fitting of the data into the global exchange model with the dissociation rate 143.6 s–1 calculated with Chemix software. (D) Exchange rates calculated for the EBH interaction with 11MACF peptide from the combination of the NMR data using the two-stage interaction model, where the folding of the EBH C-terminus follows the peptide binding. (E) Free energy contributions into the EBH interaction with the 11MACF peptide (left) and the mutated 11MACF-VLL peptide (right). The SxIP motif itself contributes approximately half of the binding energy (–19.9 kJ/mol), with the second half created by the interaction of the KP and TPQRK regions. The C-terminal EBH region that folds on the peptide contributes –6.6 kJ/mol into the binding. The VLL mutation increases the overall contribution of non-SxIP residues into the binding energy to –28.4 kJ/mol, and the binding energy of the EBH C-terminus to –13.5 kJ/mol.
-
Figure 4—source data 1
Values of the chemical shift changes in the 1H,15N-HSQC spectra induced by microtubule-actin cross-linking factor (MACF) peptides for Figure 4A.
- https://cdn.elifesciences.org/articles/98063/elife-98063-fig4-data1-v1.txt
Figure 4—figure supplement 1
Progressive changes in the HSQC spectra in the complexes with the mutant peptides.
Superposition of (A) the 1H,15N-HSQC spectra of free EBH (blue) and EB1/11MACF complex (red), (B) EB1/11MACF (blue) and EB1/11MACF-LLL (red) complex, (C) EB1/11MACF-LLL (blue) and EB1/11MACF-VLL (red) complex, and (D) EB1/11MACF-VLL (blue) and EB1/11MACF-VLLRK (red) complex. The SxIP region is underlined and the modified post-SxIP region is highlighted in blue. Signals corresponding to the 253FVI255 region are labelled in the spectra, with colours corresponding to the spectrum.
Figure 4—figure supplement 2
Sequence alignment and interactions of SxIP motifs.
(A) Sequence alignment of known SxIP proteins showing large variation of non-SxIP residues. Sequence conservation and consensus sequence are presented at the bottom. Residues are colour-coded according to the properties of the side chains. The figure was created with JalView. (B, C) Isothermal titration calorimetry (ITC) isotherm for the interaction of EBH with 13MACF (B) and CK5P2 (C). The solid line shows the fitting of the data into a single-site binding model.
Figure 4—figure supplement 3
Progressive changes in the 1H,15N-HSQC spectra on peptide addition observed for different EB1 interactions that were used to evaluate the exchange rates with TITAN software for the corresponding EB1 homology (EBH) complexes.
Superposition of the spectra for the titration of (A) EBH-ΔC with 11MACF, (B) EBH-ΔC with 11MACF-VLL, (C) EBH with 11MACF, and (D) EBH with 11MACF-VLL. Signals of the free EBH are shown red, final titration point in blue, and the intermediate titration points are shown in the pale colours. Notice additional signals that are observed for EBH/11MACF-VLL titration at the intermediate concentrations corresponding to the non-symmetrical form where the EBH dimer binds a single peptide. These signals can only be observed when the exchange between different forms is very slow.
-
Figure 4—figure supplement 3—source data 1
Values of the chemical exchange saturation transfer (CEST) measurements for EBH/11MACF complex for Figure 4—figure supplement 3A.
- https://cdn.elifesciences.org/articles/98063/elife-98063-fig4-figsupp3-data1-v1.zip
Figure 4—figure supplement 4
Chemical exchange saturation transfer (CEST) profiles for the residues with the 15N chemical shift differences between the signals of the free and the bound states larger than 1 ppm.
(A) Summary of the CEST profiles observed for EBH/11MACF interaction at the irradiation field strength of 12.5 (blue), 25 (orange), and 50 Hz (green). (B) Illustration of the CEST profile fitting (solid line) into the global exchange model with the dissociation rate 120 s–1 calculated with Chemix software.
Figure 5
High-affinity peptides form comet-like structures that colocalise with EB1 comets.
(A) Live images of HeLa cells transiently transfected with SxIP peptide constructs and EB1 plasmid. Images represent averaging three frames at 1 s intervals. Insets represent zoomed-in regions of the cell to depict SxIP peptides at EB1 comets. Scale bar: 5 µM. (B) Ratio of SxIP comets to EB1 comets present in a cell. (C) EB1 comet number per cell. Lines represent the mean with error bars representing the standard deviation. Changes in comet number with each peptide were not significant when t-test was used with the EB1 construct-only control (EB1). All experiments were performed to an N=3.
-
Figure 5—source data 1
Values of the ratios of SxIP to EB1 in the comets and comet numbers for Figure 5B and C.
- https://cdn.elifesciences.org/articles/98063/elife-98063-fig5-data1-v1.xlsx
Tables
Table 1
Binding parameters of the EB1 homology (EBH) domain interactions with peptides.
| KD(µM)* | ΔG(kJ/mol) | ΔH(kJ/mol) | –TΔS(kJ/mol) | N sites | koff(s–1)† | kon(µM–1 s–1) | |
|---|---|---|---|---|---|---|---|
| EBH/4MACF | (10400±280) | (–11.312±0.066) | |||||
| EBH/6MACF | (1730±240) | (–15.76±0.34) | |||||
| EBH/9MACF | 34.00±0.56 | –25.49±0.04 | –24.95±0.06 | –0.52±0.1 | 1.00±0.04 | ||
| EBH/11MACF | 3.5±1.0 (4.91±0.09) | –31.20±0.81 | –39.6±4.5 | 8.4±3.6 | 1.13±0.17 | 130.2±2.1 (143.6±6.0) | 37±11 |
| EBH/11MACF-LLL | 0.287±0.088 | –37.50±0.87 | –35.8±5.3 | –4.9±1.4 | 1.01±0.11 | ||
| EBH/11MACF-VLL | 0.081±0.009 (0.32±0.012) | –40.50±0.25 | –40.2±1.2 | –1.300±0.091 | 0.906±0.056 | 15.63±0.12 (57.5±8.7) | 192±21 |
| EBH/11MACF-VLLRK | 0.016±0.003 | –44.60±0.47 | –27.80±0.30 | –16.80±0.32 | 1.18±0.14 | ||
| EBH/11MACF-VLLRK150 mM NaCl | 0.198±0.012 | –38.5±1.5 | –34.4±1.8 | –4.18±0.17 | 1.03±0.05 | ||
| EBH/13CK5P2 | 0.021±0.022 | –46.4±5.0 | –36.4±8.8 | –9.6±5.0 | 0.876±0.093 | ||
| EBH-△C/6MACF | (7070±1880) | (–12.27±0.66) | |||||
| EBH-ΔC/11MACF | 41.5±8.8 (26.6±0.51) | –24.62±1.7 | –29.1±1.8 | 1.11±0.92 | 1.08±0.35 | 1900±54 | 45.8±9.8 |
| EBH-ΔC/11MACF-VLL | 18.7±3.1 (25.62±0.82) | –27.00±0.43 | –24.9±4.5 | –2.1±4.8 | 0.99±0.18 | 1541±79 | 82±14 |
| EBH-ΔC/11MACF-VLLRK | 6.29±0.47 | –29.70±0.17 | –24.6±4.6 | –5.1±4.6 | 0.960±0.056 |
-
All experiments were conducted in the buffer containing 50 mM phosphate (pH 6.5), 50 mM NaCl, 0.5 mM TCEP, and 0.02% NaN3. Salt concentration was increased to 150 mM NaCl for EBH/11MACF-VLLRK (marked in the table) to test the ionic strength dependence of the binding.
-
*
KD is determined by ITC and NMR (values in brackets). Average over three technical replicates and standard deviation is reported for ITC. The dissociation constants for 4MACF and 6MACF were determined by fitting chemical shift perturbations; average values over the peaks with the largest perturbations and standard deviation are reported. For other peptides, the NMR binding parameters were determined by the lineshape analysis in TITAN software; the results of the global fit over the well-resolved signals with the largest chemical shift perturbation (Figure 4—figure supplement 3) and standard deviation from the fit are reported.
-
†
koff is determined by the NMR lineshape analysis (standard deviation reported) and CEST (values in brackets with standard deviations from the fit).
Key resources table
| Reagent type (species) or resource | Designation | Source or reference | Identifiers | Additional information |
|---|---|---|---|---|
| Gene (human) | pOPINS, EKAR2G_design1_mTFP_wt_Venus_wt | UniProt | Q15691, MAPRE1 | EBH domain of EB1, expression construct |
| Strain, strain background (Escherichia coli) | BL21(DE3) | NEB | C2530H | Chemically competent cells |
| Cell line (human) | HeLa | AATC | PCS-201–012, RRID:CVCL_0030 | Mycoplasma negative |
| Transfected construct (human) | EKAR2G_design1_mTFP_wt_Venus_wt | Addgene | Plasmid #39813, RRID:Addgene_39813 | FRET vector |
| Transfected construct (human: MAPRE1) | mCherry-EB1-8 | Addgene | Plasmid #55035, RRID:Addgene_55035 | C-terminal mCherry tag |
| Recombinant DNA reagent | pOPINS (plasmid) | Addgene | Plasmid #41115, RRID:Addgene_41115 | SUMO-tag expression vector |
| Peptide, recombinant protein | EBH fragment | This paper | Protein preparation is described in Materials and methods. Plasmid for the expression available from the corresponding author. | |
| Peptide, recombinant protein | SxIP peptides | ChinaPeptides (Shanghai) | Set of synthetic peptides | |
| Commercial assay or kit | In-Fusion HD Cloning | Clontech | Clontech:639647, RRID:SCR_004423 | |
| Chemical compound, drug | CBR-5884 | Sigma-Aldrich | SML1656, RRID:SCR_008988 | |
| Chemical compound, drug | Lipofectamine 3000 | Invitrogen | #L3000015 | Transfection reagent |
| Software, algorithm | SPSS | SPSS | RRID:SCR_002865 | |
| Software, algorithm | Microcal PEAQ-ITC Software, v. 1.41 | Malvern | RRID:SCR_023795 | ITC data analysis |
| Software, algorithm | TopSpin 4.3 | Bruker | RRID:SCR_014227 | NMR data processing |
| Software, algorithm | CCPNmr Analysis v2.4 | CCPN | RRID:SCR_016984 | NMR data analysis software |
| Software, algorithm | Relax 4.1.1 | PMID:18085411 | NMR relaxation analysis | |
| Software, algorithm | ChemEx 2018.10.2 | https://github.com/gbouvignies/chemex; Bouvignies, 2025 | CEST data analysis | |
| Other | DAPI stain | Invitrogen | D1306 | Commercial nuclear stain (1 µg/mL) |
Appendix 1—table 1
NMR restraints and structure statistics for the NMR structure of the EBH/11MACF complex.
| Total restraints used | |
|---|---|
| NOE restraints* | |
| All | 3863 |
| Protein-ligand | 298 |
| Intermonomer | 924 |
| Intrapeptide | 209 |
| Intraresidue | 1133 |
| Sequential (|i – j|=1) | 968 |
| Medium (1 < |i – j|≤4) | 1311 |
| Long range (|i – j|>4) | 177 |
| Dihedral | |
| Φ angles | 65 |
| φ angles | 65 |
| Hydrogen bonds | 90 |
| Structure statistics | |
| Violations | |
| Distance (>0.5 Å) | 41 |
| Dihedral angle (>5°) | 7 |
| Energies (cal/mol) | |
| Overall | –2179 (±208) |
| Bond | 119 (±9) |
| Angle | 479 (±21) |
| Improper | 242 (±29) |
| Dihedral | 868 (±14) |
| Van der Waals | –48 (±28) |
| Electrostatic | –5903 (±98) |
| NOE | 1906 (±119) |
| Geometry – average values | |
| Bond | 7.40x10–3 (±5.7 × 10–4) |
| Angle | 0.91 (±9.96 × 10–2) |
| Improper | 2.50 (±0.30) |
| Dihedral | 41.56 (±0.24) |
| Van der Waals | 428.93 (±83.98) |
| Average pairwise RMSD (Å)† | |
| Heavy atoms | 2.41 (±1.25) |
| Heavy atoms – helical region | 1.51 (±0.99) |
| Backbone | 2.05 (±1.33) |
| Backbone – helical region | 1.08 (±0.93) |
| Ramachandran statistics (%) ‡ | |
| Most favoured regions | 87.0 (99.5) |
| Additional allowed regions | 10.8 (0.3) |
| Generously allowed regions | 1.0 (0.2) |
| Disallowed regions | 1.1 (0) |
-
*
Number in brackets corresponds to the restraints assigned manually.
-
†
Helical region corresponds to residues: Glu192-Glu230 and Pro237-Tyr247.
-
‡
Values within brackets correspond to residues Glu192-Glu230 and Pro237-Tyr247 (helical region).
Additional files
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Dock-and-lock binding of SxIP ligands is required for stable and selective EB1 interactions
eLife 13:RP98063.
https://doi.org/10.7554/eLife.98063.3
