KinView: a visual comparative sequence analysis tool for integrated kinome research

Abstract

Multiple sequence alignments (MSAs) are a fundamental analysis tool used throughout biology to investigate relationships between protein sequence, structure, function, evolutionary history, and patterns of disease-associated variants. However, their widespread application in systems biology research is currently hindered by the lack of user-friendly tools to simultaneously visualize, manipulate and query the information conceptualized in large sequence alignments, and the challenges in integrating MSAs with multiple orthogonal data such as cancer variants and post-translational modifications, which are often stored in heterogeneous data sources and formats. Here, we present the Multiple Sequence Alignment Ontology (MSAOnt), which represents a profile or consensus alignment in an ontological format. Subsets of the alignment are easily selected through the SPARQL Protocol and RDF Query Language for downstream statistical analysis or visualization. We have also created the Kinome Viewer (KinView), an interactive integrative visualization that places eukaryotic protein kinase cancer variants in the context of natural sequence variation and experimentally determined post-translational modifications, which play central roles in the regulation of cellular signaling pathways. Using KinView, we identified differential phosphorylation patterns between tyrosine and serine/threonine kinases in the activation segment, a major kinase regulatory region that is often mutated in proliferative diseases. We discuss cancer variants that disrupt phosphorylation sites in the activation segment, and show how KinView can be used as a comparative tool to identify differences and similarities in natural variation, cancer variants and post-translational modifications between kinase groups, families and subfamilies. Based on KinView comparisons, we identify and experimentally characterize a regulatory tyrosine (Y177^PLK4) in the PLK4 C-terminal activation segment region termed the P+1 loop. To further demonstrate the application of KinView in hypothesis generation and testing, we formulate and validate a hypothesis explaining a novel predicted loss-of-function variant (D523N^PKCβ) in the regulatory spine of PKCβ, a recently identified tumor suppressor kinase. KinView provides a novel, extensible interface for performing comparative analyses between subsets of kinases and for integrating multiple types of residue specific annotations in user friendly formats.

Figure 1

A) Example benefits of integrative analysis. Placing cancer variants in the context of natural sequence variation and post-translational modifications provides mechanistic insight as to how cancer variants can disrupt functionally significant residues, like phosphorylation sites, to deregulate kinase catalytic activity and rewire signaling networks. Using the classification of the human kinome allows us to consider differences between the kinase groups, families and subfamilies. B) MSAOnt schema. MSAOnt consists of three main classes: AlignedResidue, Insertion and Deletion. Using these three classes, we can fully represent an alignment of sequences to a profile or consensus. It is connected to the ProKinO ProteinKinaseDomain class through the hasMSAElement relation. C) The KinView interface. The interface is divided horizontally into top and bottom regions. Each region has an associated tree structure to select the kinase group, family, subfamily or domains of interest. The pulldown menu above the tree adjusts the residue numbering to match any Human kinase UniProt numbering. After clicking ‘Update’, the natural sequence variation of the selected kinases is displayed using a weblogo. Red circumscribed residue numbers show positions with cancer associated variants, while green circles show positions with experimentally validated post-translational modifications (PTMs). The secondary structure is displayed between the top and bottom regions. Detailed information is displayed by hovering the mouse over a residue number (cancer variants) or green circle (PTMs).

Figure 2

A) KinView comparison of PTK and STK activation segments. The natural sequence variation is displayed in a weblogo format, with cancer variants represented with red circumscribed residue numbers and post-translational modifications represented with green circles. Major phosphorylation sites have the residue type and number of experimentally validated phosphorylation events displayed. Note, the lack of residue numbers below two columns reflect a deletion in PKA relative to the alignment profiles. B) A subset of cancer associated variants in PTKs. Here, we show the common variants effecting positions 201^PKA and 204^PKA in PTKs. The mutant type (MT), count and most commonly associated cancer type are shown. C) A subset of cancer associated variants in STKs. Here, we show the common variants effecting positions 201^PKA and 204^PKA in STKs. The mutant type (MT), count and most commonly associated cancer type are shown.

Figure 3

A) Sequence alignment of model eukaryotic PLK4 activation segments, omitting the ‘DFG’ motif, and ending at the ‘APE’ motif. Conserved Ser/Thr amino acids are outlined in blue, and con-served Y170^PLK4 is outlined in red. B) Coomassie blue staining of 3 μg of PLK4 proteins separated by SDS-PAGE. C) Collision dissociation product ion tandem mass spectra identifying phosphorylation of T177^PLK4 or D) dual phosphorylation of T170^PLK4 and Y177^PLK4 in the same PLK4 phosphoshopeptide. Matched product ions are indicated.

Figure 4

A) Wild type PLK4 (1-269) or D154A^PLK4 mutant PLK4 (1-296) were serially diluted and the indicated amounts (ng) were separated by SDS PAGE. Proteins were transferred to a nitrocellulose membrane and probed with rabbit pT170^PLK4 PLK4 or 6His antibodies. Antibody binding was visualised using goat anti-rabbit secondary antibodies attached to HRP by ECL. B) PLK4 Y177E^PLK4 mutation greatly reduces ³²P incorporation (autophosphorylation) into PLK4 (top panel) and lacks detectable T170^PLK4 phosphorylation before or after ATP addition when probed with pT170^PLK4 PLK4 antibody (middle panel). Equal loading of proteins was confirmed by staining the membrane with Ponceau S (bottom panel).

Figure 5

A) KinView comparison of FGFR family and PDGFR family activation segments. The natural sequence variation is displayed in a weblogo format, with cancer variants represented with red circumscribed residue numbers and post-translational modifications represented with green circles. Major phosphorylation sites have the residue type and number of experimentally validated phosphorylation events displayed. B) A subset of cancer associated variants in FGFR family. Here, we show the common variants effecting positions 646^FGFR1, 652^FGFR1 and 656^FGFR1 in the FGFR family. The mutant type (MT), count and most commonly associated cancer type are shown. Note the high frequency of K656E^FGFR1 variants, which structurally mimic the phosphorylation of the preceding tyrosine. C) A subset of cancer associated variants in PDGFR family. Here, we show the common variants effecting positions 849^PDGFRβ, 855^PDGFRβ and 859^PDGFRβ in the PDGFR family. The mutant type (MT), count and most commonly associated cancer type are shown.

Figure 6

A) KinView selection of kinases that naturally conserve an asparagine at position 220^PKA (top) and those that have mutations to asparagine at position 220^PKA (bottom). Note that the AGC group does not naturally conserve an asparagine at 220^PKA among the 15 organisms included in ProKinO. B) Impaired phosphorylation of D523N^PKCβ PKCβII variant. The mutation to asparagine decreases the priming phosphorylations, which require the activity of PKC, in the activation loop, the C-tail turn motif and the C-tail hydrophobic motif. C) Normalized FRET ratio changes showing PKC activity from COS7 cells co-expressing CKAR and mCherry-tagged PKCβII mutant stimulated with 100μM UTP followed by 200 nM PDBu. Graph on the right shows the signaling output resulting from UTP stimulation (see Methods), quantified and normalized to WT PKC activity. **p<0.01 as compared with WT, using a repeated-measures one-way ANOVA.