Figure 1.
Structural characterization of the T46I-MSP domain.
(a). Far UV CD spectra of the WT- (blue), T46I- (red) and P56S- (black) MSP domains at protein concentrations of 20 µM. (b). Superimposition of the two-dimensional 1H-15N NMR HSQC spectra of the WT- (blue) and T46I- (red) MSP domains at protein concentrations of 100 µM at pH 6.8. Cα (c) and Cβ (d) chemical shifts of the WT- (blue) and T46I- (red) MSP domains. The residues with the disappeared HSQC peaks are indicated by green arrows. (e). The crystal structure of the WT-MSP domain we previously determined (32) to which the T46I residues with their HSQC peaks undetected were mapped. The two mutation residues causing ALS, P56S and T46I, were displayed as spheres.
Figure 2.
Stability of the T46I-MSP domain.
(a). The urea-induced unfolding curves of the WT- (blue) and T46I- (red) MSP domains as reflected by changes of the ellipticity at 222 nm with urea concentrations ranging from 0 to 8 M. One-dimensional NMR spectra of the WT- (b) and T46I- (c) MSP domains at 15°C (green) 25°C (blue) and 45°C (red). (d). Hydrogen-deuterium (H/D) exchange results for the WT-MSP domain. Blue bars: residues with HSQC peaks detectable in the 10 mM phosphate buffer at pH 6.8. Green bars: residues with HSQC peaks detectable 15 min after re-dissolving the lyophilized protein powder in D2O. Red bars: residues with HSQC peaks detectable 2.5 hr after re-dissolving the lyophilized protein powder in D2O. (e). The crystal structure of the MSP domain with the H/D exchange results mapped. The yellow is to indicate the residues undetectable even in the 10 mM phosphate buffer at pH 6.8.
Figure 3.
Interaction between the T46I-MSP domain and the Nir2 peptide.
(a). The ITC titration profile of the binding reaction of the T46I-MSP domain to the Nir2 peptide [upper panel]; and integrated values for reaction heats with subtraction of the corresponding blank results normalized by the amount of ligand injected versus molar ratio of MSP/Nir2 (lower panel). Thermodynamic binding parameters obtained from fitting the data are shown in the box. Superimposition of the two-dimensional 1H-15N NMR HSQC spectra of the WT- (b) and T46I- (d) MSP domains in the absence (black) and presence of the Nir2 peptide at molar ratios 1∶0.5 (green), 1∶1 (blue) and 1∶2 (red). (c). Crystal structure of the VAPA MSP-ORP1 complex to illustrate the MSP-peptide binding interface [12].
Figure 4.
Interaction between the WT-/T46I-MSP domains and EphA4.
(a). The ITC titration profile of the binding reaction of the WT-MSP domain to the ligand-binding domain of EphA4 (upper panel); and integrated values for reaction heats with subtraction of the corresponding blank results normalized by the amount of ligand injected versus molar ratio of MSP/EphA4 (lower panel). Superimposition of the two dimensional 1H-15N NMR HSQC spectra of the WT- (b) and T46I- (d) MSP domains in the absence (blue) and presence of the EphA4 ligand binding domain at molar ratios 1∶2 (green) and 1∶8 (red) (MSP/EphA4). Red letters are used to label residues with disappeared HSQC peaks while blue for residues with shifted HSQC peaks. The MSP structure with the perturbed residues mapped back for the interactions between the WT-MSP and EphA4 (c), and the T46I-MSP and EphA4 (e). Blue is to indicate residues with unperturbed HSQC peaks, while green and red for residues with shifted and disappeared HSQC peaks respectively.
Figure 5.
Interaction between EphA4 and WT-/T46I-MSP domain.
Superimposition of the two-dimensional 1H-15N NMR HSQC spectra of the EphA4 ligand binding domain in the absence (blue) and presence of the WT- (a) and T46I- (c) MSP domains at molar ratios 1∶2 (green) and 1∶8 (red) (EphA4/MSP). Red letters are used to label residues with disappeared HSQC peaks while blue for residues with shifted peaks. Crystal structure of the EphA4 ligand-binding domain we previously determined (37) with the perturbed residues mapped back for the interaction between EphA4 and WT-MSP (b), and T46I-MSP (d). Blue is to indicate residues with unperturbed HSQC peaks, while green and red for residues with shifted and disappeared peaks respectively.
Figure 6.
Docking model for the EphA4-MSP complex.
(a). Superimposition of two lowest energy docking model of the EphA4-MSP complex. (b). The crystal structure we previously determined for the EphA4-ephrinB2 complex [40]. (c). The docking model of the EphA4-MSP complex with the FFAT-containing ORP1 peptide displayed according to the VAPA MSP-ORP1 structure [12].
Figure 7.
Trajectories of MD simulations.
(a–b). Root-mean-square deviations (RMSD) of the backbone atoms for two independent MD simulations of the WT- (blue) and T46I- (red) MSP domains. (c–d). Root-mean-square fluctuations (RMSF) of the Cα atoms computed for two independent simulations (blue for simulation 1 and red for simulation 2) of the WT- (c) and T46I- (d) MSP domains. (e–f). Root-mean-square fluctuations (RMSF) of the Cα atoms computed for two independent simulations for the WT- (blue) and T46I- (red) MSP domains. The average values and standard deviations over 15 ns are computed and displayed. The green arrows are used to indicate the T46I-MSP residues with fluctuations larger than those of the WT-MSP domain.
Figure 8.
(a–b). Structure snapshots (one structure for 1-ns interval) for two independent MD simulations for the WT- (blue) and T46I- (red) MSP domains. (c). The MSP structure in which cyan is utilized to indicate the WT-MSP residues with RMSF values larger than the average. (d). The MSP structure in which red is used to indicate T46I residues whose RMSF values are larger than those of the WT-MSP domain. The two mutation residues causing ALS, P56S and T46I, were displayed as spheres. (e). The MSP structure in which green is used to indicate residues whose hydrogen bonds occupancy difference is larger than 10% between WT and T46I while red is for residues whose hydrogen bonds occupancy difference is larger than 10% between T46I and WT.
Table 1.
Hydrogen Bond Occupancy in MD simulations for the WT- and T46I-MSP Domains.