Fig 1.
Family and subfamily divergent residues.
(A) Kinase domain phylogeny for 2,094 different kinase domains, spanning 9 species. Each sequence has been coloured according to its group membership. (B) Explanation of the score used to identify divergent residues. RC: conservation of the residue within the family of interest. AC: extent to which the ancestral residues are conserved between sister clades in the phylogeny. p(AC): a measure of confidence in the ancestral sequence prediction. A full explanation is given in the Materials and methods section. (C–F) Examples of residues predicted to be functionally divergent in the SRPK (PDB: 1WBP), CDC7 (PDB: 4F9A), CAMK2 (PDB: 5H9B), and PLK (PDB: 4B6L with peptide binding modelled) families. In these 4 examples, kinase residues have been numbered according to their position in the protein kinase domain (Pfam: PF00069). The peptides and/or proteins that physically interact with the kinase have been coloured in yellow. AC, ancestral conservation; AGC, (PKA, PKG, PKC); AUR, Aurora Kinase; CAMK, Calcium- and Calmodulin-regulated kinases; CAMK2, Calcium/calmodulin-dependent protein kinase type 2; CDC7, Cell Division Cycle 7-related protein kinase; CK1, Casein Kinase 1; CK 2, Casein Kinase 2; CLK, CDK-Like Kinase; CMGC, (cyclin-dependent kinases (CDKs), mitogen-activated protein kinases (MAP kinases), glycogen synthase kinases (GSK) and CDK-like kinases (CLKs)); PDB, Protein Data Bank; PHK, Phosphorylase Kinase; Pfam, Protein families; PLK,; Polo-like kinase RC, recent conservation; SRPK, SR-rich protein kinase; STE, sterile mutant; TK, tyrosine kinase; TKL, Tyrosine Kinase-Like.
Fig 2.
Aggregated analysis of sequence divergence across kinase families (left) and subfamilies (right).
For each kinase domain position, the total number of ‘switches’ (score > 95th percentile of scores) was counted across all families or subfamilies considered (see Materials and methods). (Top) Results mapped to kinase structures (mouse protein kinase A: PDB 1ATP). The kinases are represented in complex with an ATP molecule (green, orange, blue, red) and a substrate-mimicking inhibitor (PKIA, yellow). Darker shades of red or blue represent a higher number of switches (see colour bar, right). (Middle) Total number of switches mapped to the kinase primary sequence, with secondary structure elements represented above the bar plot. A domain position is considered to be ‘frequently switching’ if the number of switches lies above a 90th percentile threshold for the kinase domain. The threshold is 8 for families and 7 for subfamilies. (Bottom) The values for each domain position have been grouped according to the functional category (‘catalytic’, ‘regulatory’, ‘proximal’, etc.) and the distribution plotted separately at the family and subfamily level. The numbers in brackets represent the number of residues in each category. PDB, Protein Data Bank; PKIA, cAMP-dependent protein kinase inhibitor alpha; SDR, specificity-determining residue.
Fig 3.
Evolution of kinase specificity at the group, family, and subfamily level.
(A) Differences in S/T kinase specificity models at the group, family, and subfamily levels. The Frobenius distance was calculated for all possible pairwise comparisons within and between groups, families, and subfamilies. (B) Representative kinase pairs belonging to different groups (left), families (centre), and subfamilies (right). Frobenius distances for each of the 3 pairs are given beneath the logos. (C) A simplified tree of life with 3 important divergence times (plant-opisthokont, fungi-metazoa, chordate-nonchordate) marked. (D) Phylogenetic estimation of kinase ages at the group, family, and subfamily level for S/T kinases. AGC, (PKA, PKG, PKC); CMGC, cyclin-dependent kinases (CDKs), mitogen-activated protein kinases (MAP kinases), glycogen synthase kinases (GSK) and CDK-like kinases (CLKs); mya, million years ago; PRKCG, Protein Kinase C gamma; RSK, Ribosomal S6 Kinases; RSKp70, Ribosomal S6 Kinases p70; RSKp90, Ribosomal S6 Kinases p90; S/T, serine/threonine.
Fig 4.
Enrichment of S/T phosphorylation motifs across several species.
Binomial p-values were calculated for each motif and each species considered. The heat map cells are coloured according to the extent of enrichment for that particular motif and species (see legend, bottom right). The numbers in the column labels correspond to the sample size of unique S/T phosphorylation sites (15-mer). Prokaryotic phosphosite samples are coloured in purple. (Top) Enrichment of 3 common phosphorylation signatures (S/T-P, R-x-x-S/T, and S/T-x-x-D/E). (Bottom) Enrichment of 29 motifs discovered using the motif-x tool. Motifs in which the effector kinase has already been described in the literature are coloured in red. S/T, serine/threonine.