Figure 1.
PCA analysis on 11 hPrP structures reveal structural perturbations correlated with prion disease.
(A) Projection of hPrP NMR ensembles onto PC1 and PC2. (B) Projection of hPrP NMR ensembles onto PC1 and PC3. For (A) and (B), each point on the conformer plot represents an NMR model, and the models are colored to reflect NMR ensembles. For each NMR ensemble, the NMR representative model that has been selected by OLDERADO [39] is indicated by a black triangle. Ovals indicate dominant clusters that represent the hPrP diseases of CJD (green oval), FFI (blue oval), GSS (red oval), as well as the set of WT and variant proteins (WT+V, black oval). The ovals representing hPrP disease are also labeled, with the PDB code of their corresponding NMR ensemble in brackets. 2K1D models which cluster separately from the rest of the 2K1D ensemble are circled (dashed brown oval). (C) Eigenvalue contribution of PCs to variance of the dataset. Further analysis of this dataset is demonstrated in Figure 4A.
Figure 2.
PCA analysis results of 11 hPrP structures.
(A) Contribution of each residue of hPrP to the first three principal components. Subdomains of concerted displacement in each PC are indicated by colored boxes and labeled. (B) Subdomains of concerted displacement in each of the PCs are highlighted against the reference structure 1QLZ (WT hPrP), and color-coded by their first appearance in a PC. From our dataset, pathogenic mutations causing familial disease (D178N, E200K, Q212P, causing FFI, CJD, and GSS, respectively) are indicated (black boxes), as well as nonpathogenic variants (M129V, M166V or M166C, S170N, R220K, E221C)(blue boxes). (C–E) Structural interpolation of atomic displacements from the mean structure for PC1,PC2, and PC3, respectively (reference structure 1QLZ). Subdomains exhibiting displacement in each PC are indicated by arrows, and the arrows are color-coded to match the boxed subdomains in (A). (See also Figure S5).
Figure 3.
PCA analysis of the WT hPrP subset.
(A) Contribution of each residue to the first three principal components (reference structure 1QLZ). Each subdomain of concerted displacement is indicated by a box that is color-coded across all 3 PCs. (B) Eigenvalue contribution of PCs to variance of the dataset.
Figure 4.
Comparative analysis of conformer plots, residue contribution, and structural interpolation of hPrP mutant NMR ensembles structures versus WT and variant hPrP.
Each row of the figure represents one PCA analysis and contains, from left to right, a conformer plot, residue contribution plot, and structural interpolation diagram. An explanation of the conformer plots is provided in Figure 1. Residue contribution to each PC is color-coded by PC (red = PC1, green = PC2, purple = PC3) in all residue contribution plots. For structural interpolation diagrams, PC1 is represented as equidistant atomic displacements from the mean structure (reference 1QLZ), and corresponding subdomains are indicated (red arrows). (A) Combined set of WT, variant and mutant hPrP NMR ensembles. The conformer plot is identical to Figure 1A. In the conformer plot, NMR ensembles of mutant structures are encircled in green, red, and blue ovals and labeled by their corresponding human disease, as well as the PDB code corresponding to the NMR ensemble (in brackets). The set of WT and variant hPrP structures (encircled by the black oval) have been labeled as WT+V. For rows (B–D), each analysis consists of the set of WT+V and an NMR ensemble from each of the CJD, FFI, and GSS mutant structures, respectively. The NMR ensemble of the mutant structure is encircled by an oval (color-coded to (A)), and labeled by the human disease it represents, and the PDB code corresponding to the NMR ensemble (in brackets). (B) CJD mutant (PDB 1FO7) and WT+V, (C) FFI mutant (PDB 2K1D) and WT+V, (D) GSS mutant (PDB 2KUN) and WT+V. (See also Figure S1).
Figure 5.
PCA analysis of the 21 NMR ensembles of WT PrP structures.
(A–C) Projection of the structures, including chPrP, onto PCs 1–3 (D–F) Projection of structures, excluding chPrP, onto PCs 1-3. For (A–F), each point on the conformer plot represents an NMR model, and the models are colored to reflect NMR ensembles. For each NMR ensemble, the NMR representative model that has been selected by OLDERADO [39] is indicated by a black triangle. Identifiable clusters of NMR ensembles have been labeled by the species they represent, with the corresponding PDB code of the ensemble in brackets. (G) Regions of concerted displacement in PC1 and PC2 of the residue contribution plot. (H) Regions of concerted displacement are labeled (black boxes) onto the primary structure (reference structure 1QLZ (hPrP)). Residues that do not contribute to the core alignment are shaded in black.
Figure 6.
Projection of mammalian PrP NMR ensembles onto PCs 1–3.
(A–C) Mammalian PrP structures (n = 13 species) (D–F) TSE-non-susceptible mammals (n = 5 species, including Sheep R168 variant). For all conformer plots, each point on the plot represents an NMR model, and the models are colored to reflect NMR ensembles. For each NMR ensemble, the NMR representative model that has been selected by OLDERADO [39] is indicated by a black triangle. Identifiable clusters of NMR ensembles have been labeled by the species they represent, with the corresponding PDB code of the ensemble in brackets. (See also Figure S4).
Figure 7.
Residue contribution to PCs of TSE-non-susceptible, TSE-susceptible, and combined dataset of mammalian PrP.
(A) The combined mammalian dataset (n = 13 species). (B) TSE-non-susceptible mammals (n = 5 species, including Sheep R168 variant). (C) TSE-susceptible mammals (n = 9 species, including Sheep H168 variant). Notably, sheep has been included in both species counts, as the sheep polymorphism R168 is non-susceptible, while H168 is susceptible. For all conformer plots, structures are colored by PDB name to reflect NMR ensembles, and identifiable clusters of NMR ensembles have been labeled by the species they represent. (See also Figure S3).
Figure 8.
Conformationally variable subdomains in hPrP.
Subdomains are colored in cyan, and labeled by region. Important polymorphisms and disease-linked (DLMs) mutations in each section are also depicted.
Table 1.
Summary of PCA analyses on PrP datasets.