Fig. 11

NEK2 phosphorylation in PDZ domain promotes open conformation of DVL3. a: PDZ sequence alignment of the three human DVL isoforms. The conserved serine residues phosphorylated specifically by NEK2 are highlighted in magenta. b: Crystal structure of DVL2 PDZ domain bound to internal ligand (PDB: 3CC0). The two serines prone to NEK2 phosphorylation are shown in magenta, the bound peptide in cyan, and PDZ residues of the binding loop or close to the binding loop in orange. c: NMR titrations of DVL3 C-terminal peptide to DVL2 wildtype PDZ or S281E/S298E phosphomimicking mutant. Magnified insets show the qualitative differences in peptide binding for PDZ residues annotated in (b). d: Schematic depiction of the FRET efficiency experiment of the ECFP DVL3 FlAsH III construct. In the default conformation, the DVL molecule is less phosphorylated and closed/compact, which is reflected in the close proximity of the FRET pair (ECFP and FlAsH (F) tags) – leading to the high intramolecular FRET efficiency (depicted as orange dash lines). In the presence of the active kinase (e.g. NEK2), DVL is phosphorylated in the PDZ domain, which promotes the open/loose conformation by the disruption of the C-terminus/PDZ domain interaction. This is reflected in the high proximity of ECFP and F tags thus leading to the low intramolecular FRET efficiency (no orange dash lines). e: HEK293 cells were transfected in the indicated combinations and the intramolecular FRET efficiency of the ECFP-DVL3 FlAsH III construct (schematically depicted in D) was measured. The data represent median ± interquartile range from three independent experiments (numbers of analyzed cells are indicated below). Statistical significance was confirmed by One-way ANOVA and Tukey’s post test (** p < 0.01, n.s. - not significant). FRET eff. Stands for Förster-Resonance-Energy-Transfer efficiency, ECFP for Enhanced Cyan Fluorescent Protein, F for FlAsH tag, and kd for kinase dead