Fig 1.
Structure of LPA6 and LPA6-LPA docking model.
a) Overall crystal structure of zebrafish LPA6 (Protein Data Bank (PDB) ID: 5XSZ), viewed from the membrane plane. In the left panel, the whole structure is shown as a ribbon model, and the disordered region of ECL2 is illustrated as a dashed line. In the right panel, the whole structure is shown as a surface model from the same viewpoint as the left panel. The gap structure between TM4 and TM5 is indicated with the blue rectangle. b) The LPA6-LPA docking models (Models 1, 2, and 3) viewed from the extracellular side (left), and the membrane plane (right). The protein is shown as a ribbon model and the LPA (2-LPA(18:2)) molecule is depicted by a CPK model. The conserved positively charged residues are shown as stick models.
Fig 2.
Schematic flow-chart of the procedure used in this work for MSM construction.
Fig 3.
Macrostates and MSM groups calculated from the present simulation results.
a) The macrostates observed in the present simulation. The size of the circle corresponds to the population of the microstate. The transitions between the macrostates are indicated by the gray arrows. The thickness of the arrows is proportional to the transition probability. The structures of the centers of the largest clusters are shown with cartoon and CPK models. b) Close-up views of representative structures of the macrostates. The interactions of the ligand-ligand recognition residues in each structure are shown as ball and stick models.
Fig 4.
Free energy landscape projected on the plane spanned by the PC1 and PC2 axes.
a) The explained variance ratios of PC1 to PC5 (gray bar chart), and their cumulative values (gray line). b) Free energy landscape projected on the PC1-PC2 plane. In the left panel, the centers of the clusters and the initial structures are plotted as light and dark blue circles, respectively. The sizes of the light blue circles are proportional to the populations of the clusters. In the middle panel, the centers of the clusters are color-coded according to their macrostates. In the right panel, the center of each microstate is plotted on the free energy map. The groups of the macrostates defined in Fig 3 and the main text (i.e., dissociated, partially-bound, and bound groups) are indicated by magenta lines. c) The relationships between the PC1 values and the distances between the LPA head group and the conserved positive residues (K1885.32, R2676.62, R2817.32, K261.31, and R832.60). The distances are defined as those between the nearest oxygen atom of the head group and the nitrogen atom of the basic residues. d) The ligand conformation and the receptor structure with the highest PC2 value (PC2 = 3.57). The ligand and receptor are shown as CPK and cartoon models, respectively. The definition of the ligand vector and its z value are depicted as the dashed line and blue arrow, respectively. e) The relationship between the PC2 value and the z value of the ligand shown in (d). The difference of the distributions of the ligand z values in the higher and lower PC2 regions (red and blue for PC2 > 0 and < 2, respectively) is shown as a histogram in the right panel.
Fig 5.
Free energy landscape projected on the plane spanned by the PC1 and PC3 axes.
a) Free energy map projected on the PC1-PC3 plane. In the left panel, the centers of the clusters and the initial structures are plotted as light and dark blue circles, respectively. The sizes of the light blue circles are proportional to the populations of the clusters. In the middle panel, the centers of the clusters are color-coded according to their microstate. In the right panel, the positions of the central structures of the macrostates are indicated by circles. A basin in the free energy map with a small PC1 value corresponds to a dissociated state, that with an intermediate PC1 value represents a partially-bound state to the central region with high energy, and that with a large PC1 value corresponds to a bound state. Macrostate 6 corresponds to a basin with a low PC3 value, and 7 corresponds to a basin with a high PC3 value. b) Relationship between the PC3 value and the distance between the LPA head group and R2817.32. The definition of the distance is the same as in Fig 4.
Fig 6.
Reconstituted 100-μs trajectory of the LPA binding process.
a) Plot of the transitions among Macrostates 1–6 of the reconstituted trajectory. The color code corresponds to that used in Fig 3. b) Plot of the transition of the distance between the LPA head group and the center of mass of the receptor protein. Distances less than 25 Å are highlighted in light blue. c) The transition of the TM4-TM5 distance is plotted in light blue, and the moving average between the four frames is plotted in blue. d) The distribution of distances between TM4 and TM5 in the ligand-unbound (gray) and -bound (blue) states. e) The structures with the smallest (orange) and largest (green) TM4-TM5 distances in the ligand-unbound state are shown by cartoon models and cross-sectional views of the electron density surfaces. f) Plot of the distances between the LPA head group and the conserved positive residues (K185, K1885.32, R2676.62, R2817.32, K261.31, and R832.60) around 25 μs and 35 μs. Areas with distances of 2.5 Å or less are highlighted in light blue and light green backgrounds.
Fig 7.
Functional analysis of LPA mutants designed based on the present study.
a) The receptor activities were examined using the alkaline-phosphatase tagged TGF-ɑ shedding assay. b) Activities of LPA and its mutants expressed as RAi (Emax/EC50 relative to WT). Data are mean ± s.e.m. (n = 3 or 4). *P < 0.05; **P < 0.01; ***P < 0.001, one-way ANOVA with Dunnett’s post hoc test. NS, not significant.
Fig 8.
Transition state of the ligand binding process.
a) The structures contained in the three clusters corresponding to the transition states, clusters A, B and C, are plotted on the PC1-PC2 and PC1-PC3 planes, respectively. The central structure of Macrostate 6 is indicated by pink stars. b) The central structures of clusters A, B and C, and the central structure of Macrostate 6 are shown by cartoon and ball and stick models, respectively.
Fig 9.
A schematic model of the LPA6 activation process proposed in this study.