prot26271-sup-0001-TableS1.pdfPDF document, 459.4 KB

Table S1 Dataset of protein kinases (PK) in active conformation and pseudokinases (PsK), related to plant receptor-like kinase (RLK), human tyrosine kinase-like (TKL), and tyrosine kinase (TK) families, used in this study.

prot26271-sup-0002-TableS2.xlsxExcel 2007 spreadsheet , 29.5 KB

Table S2 All versus all structural comparisons of protein kinases (PDB entries shaded in red) and pseudokinases (PDB entries shaded in green), closely related to RLK, TKL, and TK families are provided in separate sheets. The analysis was done using DALI web server. The RMSD values of kinase–kinase, pseudokinase–pseudokinase, and kinase–pseudokinase structural comparisons are demarcated in red, green, and white cells, respectively.

prot26271-sup-0003-TableS3.docxWord 2007 document , 23.9 KB

Table S3 Dataset of 19 protein kinases (PK) in inactive conformation, related to plant RLK, human TKL, and TK families, used in this study.

prot26271-sup-0004-TableS4.pdfPDF document, 341.6 KB

Table S4 Maximum collectivity values (κmax) and Cα-backbone flexibility measures (calculated from area under the curve) for the selected nonzero normal mode, for each kinase and pseudokinase in the dataset, derived from ANM-NMA.

prot26271-sup-0005-TableS5.pdfPDF document, 335.6 KB

Table S5 Community membership of kinase and pseudokinase community maps, derived from residue-by-residue cross-correlated motions, computed using all-atom normal mode analysis.

prot26271-sup-0006-FigureS1.tifTIFF image, 8.9 MB

Figure S1 Amino acid sequence identity heatmaps of kinases (shaded in red) and pseudokinases (shaded in green), closely related to RLK, TKL, and TK families. The color scales, alongside heatmaps, indicate range of percentage sequence identities (low to high violet to green) between kinases and pseudokinases of respective families. Protein names are provided with prefixes to indicate source organism (At: Arabidopsis thaliana, Zm: Zea mays, Hs: Homosapien sapien). Kinase–kinase and pseudokinase–pseudokinase sequence identities are demarcated by red and green squares, respectively, on the heatmap.

prot26271-sup-0007-FigureS2.tifTIFF image, 5.5 MB

Figure S2 Cα-backbone fluctuation patterns of 51 “active” protein kinases and 26 pseudokinases, derived from ANM-NMA using ProDy. (A) Secondary structure organization (helices, strands, loops) of a kinase-like domain has been provided, as in Figure 1A. Helices (α) and strands (β), indicated as rectangles and arrows, respectively, are numbered. N-terminal lobe (gray), hinge region (blue), and C-terminal lobe (yellow) with catalytic loop and activation segment are indicated. (B–G) Individual flexibility profile for each kinase (PK) or pseudokinase (PsK) related to receptor-like kinase (RLK), tyr-kinase like (TKL), or Tyr Kinase (TK) family are color-coded, according to legend provided with each plot. The respective PDB entries used for each category (RLK-PK/PsK, TKL-PK/PsK, and TK-PK/PsK) are mentioned in Table S1. Normalized square fluctuations for each entry, calculated from the first 10 consecutive nonzero modes, are plotted as a function of aligned amino acid residue positions. For clarity, a replica of secondary structure organization in S1A is provided along the X-axis of each plot.

prot26271-sup-0008-FigureS3.tifTIFF image, 2.8 MB

Figure S3 Residue-wise mean of normalized square fluctuations, calculated from the last 10 consecutive nonzero normal modes or fast modes (91–100), for protein kinases (dark red) and pseudokinases (green), related to each kinase family, are plotted as a function of aligned amino acid residue positions. A replica of secondary structure organization from Figure 1A is plotted along X-axes of all profiles. Sub-structural elements of interest, labeled on top, are shaded in respective plots, to clearly indicate residue positions, peak profiles, and similarities or differences in fluctuation patterns.

prot26271-sup-0009-FigureS4.tifTIFF image, 990.9 KB

Figure S4 Residue-wise normalized square fluctuations, calculated from the first 10 (slow modes) and last 10 (fast modes) consecutive nonzero normal modes, are depicted for the kinase-like domain (residue positions: 68–345) of SelO pseudokinase (PDB code: 6eac). A secondary structure organization of SelO kinase-like domain is plotted along X-axes of the profile. Helices (α) and strands (β) are indicated as rectangles and arrows, respectively, according to the PDB entry. N-terminal lobe (gray), hinge region (blue), and C-terminal lobe (yellow) are indicated.

prot26271-sup-0010-FigureS5.tifTIFF image, 1 MB

Figure S5 Individual flexibility profiles for the inactive conformation of 19 protein kinases related to RLK, TKL, and TK families are color-coded, according to legend provided with each plot. The respective PDB entries used for each category are listed in Table S3. Normalized square fluctuations for each entry calculated from the first 10 consecutive nonzero modes are plotted as a function of aligned amino acid residue positions. For clarity, a replica of secondary structure organization is provided along the X-axis of each plot.

prot26271-sup-0011-FigureS6.tifTIFF image, 3.1 MB

Figure S6 Individual flexibility profiles for groups of randomly selected 10 protein kinases and 10 pseudokinases, related to RLK, TKL, and TK families. Residue-wise normalized square fluctuations, calculated from the first 10 consecutive nonzero modes, are color-coded (see legend alongside) and plotted as a function of aligned amino acid residue positions. For clarity, a replica of secondary structure organization is provided along the X-axis of each plot. The respective PDB entries used in this analysis are listed in Table S1.

prot26271-sup-0012-FigureS7.tifTIFF image, 1.1 MB

Figure S7 Mode overlap analysis for RET kinase and JAK2 pseudokinase. (A) Heatmap showing the overlap between the first 10 nonzero normal modes for JAK2 and RET. The color bar denotes the overlap from 0 to 1 (purple to yellow). (B) Residue-wise coarse-grained normalized square fluctuations for mode 6 of RET and mode 1 of JAK2. Secondary structure organization of the kinase-like domain has been depicted along the X-axis of both plots. Important sub-structures are also indicated.

prot26271-sup-0013-MovieS1.mp4MPEG-4 video, 26.4 MB

Movie S1 Distinct large-scale Cα-backbone fluctuations in RET kinase and JAK2 pseudokinase. Tertiary structure representations of RET kinase (PDB code 6nja, left) and JAK2 pseudokinase (PDB code 4fvq, right) are depicted with eigen vectors (black arrows) using NMWiz. N-lobe (gray), hinge (blue), and C-lobe (yellow) for both structures, activation segment in RET (red) and pseudo-activation segment in JAK2 (green) are color coded. For clarity, eigen vectors are only shown for nodes whose fluctuation RMSD is >1 Å. Fluctuations are shown for the mode displaying κmax in RET (mode 2) and JAK2 (mode 23), as listed in Table S4. The imaginary axis between topologically equivalent E and F helices in JAK2 pseudokinase (right) is represented by a dotted line.

prot26271-sup-0014-MovieS2.mp4MPEG-4 video, 9.7 MB

Movie S2 Large-scale Cα-backbone fluctuations in the N-lobe of JAK2 pseudokinase appear important for dimerization with JAK2 kinase domain. Static tertiary structures of JAK2 pseudokinase, C-helix facing (top left), and αC-β4 loop facing (bottom left), are provided along with Cartesian coordinate axes. Pseudokinase–kinase dimer interface residues are indicated in green circles and mutations in pseudokinase domain that lead to constitutive activation of JAK2 are marked in orange circles (bottom left). Residues that serve as both interface residues and mutational hotspots are indicated in green circles and outlined in orange. The abovementioned residues are marked with their eigen vectors (black arrows). Corresponding dynamic structures with imaginary axes are provided alongside (top right and bottom right), which indicate the collective motions of interface residues in JAK2 pseudokinase. For clarity, eigen vectors are only shown for nodes whose fluctuation RMSD is >1 Å.

prot26271-sup-0015-MovieS3.mp4MPEG-4 video, 16 MB

Movie S3 Protein backbone fluctuations for the most correlated slow mode in RET kinase (mode 6, left) and JAK2 pseudokinase (mode 1, right). Cartoon representations of RET kinase (PDB code: 6nja) and JAK2 pseudokinase (PDB code: 4fvq) structures are depicted with eigen vectors (using NMWiz). Besides, N-lobe (gray), hinge (blue), and C-lobe (yellow), activation segments in RET and JAK2 are colored in red and green, respectively. For clarity, eigen vectors are only shown for nodes whose RMSD is >1 Å.

prot26271-sup-0016-MovieS4.mp4MPEG-4 video, 67.6 MB

Movie S4 Differences in large-scale concerted motions within pairs of kinase and pseudokinase, sharing highest amino acid sequence identity, within each family (as depicted in Figure 3). Cα-backbone fluctuations or collective motions for the mode displaying maximum collectivity (κmax) was visually analyzed using NMWiz, as explained in Figure 2 (Table S4). Kinase and pseudokinase tertiary structures are color coded, as described previously: N-lobe (gray), hinge (blue), C-lobe (yellow), activation segments in kinases (red-BRI1, BRAF, and EGFR) and pseudo-activation segments in pseudokinases (green-BIR2, KSR2, and HER3). For clarity, eigen vectors (black arrows) are only shown for nodes whose fluctuation RMSD is >1 Å. Imaginary axes in pseudokinases are represented by dotted lines.