(A) Left, comparison of SDS-PAGE gel mobility and steady-state protein level of myc epitope-tagged Plk2 kinase dead (K108M), wild-type (WT), and constitutively active (T236E) forms expressed in COS7 cells. Right, Plk2 (K108M) or (T236E) were immunoprecipitated using myc agarose and visualized on SDS-PAGE with Coomassie staining for mass spectrometric analysis. Schematic illustrates proposed open and closed conformations of differently migrating species. (B) Trypic peptides identified by LC-MS/MS with potential phosphorylation sites denoted by red and asterisk. Amino acid (aa) residue of sites is also indicated. The @ symbol next to methionines indicates that the methionine is oxidized (16 mass units higher than expected), which readily occurs during the process of running and staining/destaining the gel. (C) Positions of potential phosphorylation sites within the primary structure of Plk2 (arrows). Positions of kinase-dead (K108M, red) and constitutively active (T236E, green) mutants are indicated with lines. Sites identified in Plk3 are shown at bottom (blue) aligned with homologous residues in Plk2 (but designated with amino acid numbering for Plk3).

(A) Expression and gel mobility of myc epitope-tagged phosphosite mutants and KD-, WT-, and CA-Plk2 controls. Lysates from COS7 cells transfected as indicated at top of blots were analyzed by SDS-PAGE and immunoblotted with myc antibodies. Cotransfection of GFP demonstrates equivalent transfection efficiency and loading. (B) Quantification of data in (A); n = 4–14 replicates per construct. (C) Example immunoblots of WT and selected Plk2 mutants expressed in COS7 cells for 24 hrs and treated with cycloheximide for the indicated times. The calculated half-life from (D) is indicated to the right. (D) Half-life determination of Plk2 constructs from (C), plotted as percent remaining protein on a log scale against time (n = 3–6 replicates per time point per construct). *p<0.05. **p<0.01, ***p<0.005.

(A–G) Cultured hippocampal neurons (DIV 16) were transfected with constructs as indicated and immunostained 24 hr post-transfection with an anti-myc antibody to visualize exogenous Plk2. Scale bar in (A) represents 20 µm. (H) Quantification of data from A–G.

(A–G) DIV 16 hippocampal neurons were cotransfected with GFP and phosphosite mutants or controls as indicated. After 24 hrs, neurons were fixed and immunostained with anti-GFP antibodies. Images were captured on a laser scanning confocal microscope for analysis of spine density. Representative dendrites are enlarged below each neuron image. (H) Quantification of data in (A–G); n = 5–8 neurons per group; *p< 0.05 relative to WT-Plk2. Scale bar in (A) represents 20 µm.

Numbering in amino acid residues. Red boxes indicate known phosphorylation sites as annotated at right of sequences.

(A) Immunoblot of the Plk2 substrate SPAR (tagged with myc-epitope) cotransfected with each phosphosite mutant and KD, WT, and CA-Plk2 controls. Mutant pairs S299A/S299E and S588A/S588E exhibited differential levels of SPAR between the non-phosphorylatable and phosphomimetic mutants. (B) Representative blot of Plk2 controls and 4 phosphosite mutants identified in (A) cotransfected with myc-SPAR for quantification. (C) Quantification of SPAR cotransfected with Plk2 controls and phosphosite mutants; n = 3–4 replicates per condition; *p<0.05 and **p<0.005 relative to SPAR cotransfected with WT-Plk2.

(A) DIV 16 hippocampal neurons were cotransfected with GFP and phosphosite mutants or controls as indicated. After 24 hrs, neurons were fixed and immunostained with anti-GFP to identify transfected cells and with anti-SPAR antibodies to visualize endogenous SPAR. Representative dendrites are enlarged below each neuron image. White arrows indicate SPAR puncta that colocalize with GFP-filled spines. (B) Quantitation of data from (A); n = 7–11 neurons per transfection; *p< 0.05 and **p<0.005 relative to SPAR in neurons transfected with WT-Plk2. Scale bar represents 20 µm.

(A) When inactive, the polo box domain (PBD) consisting of polo boxes (PB) 1 and 2 forms an intramolecular inhibitory loop that creates a conformationally closed state and allosterically blocks the enzymatic active site. (B) Activation of Plk2 by mutants (T236E or S588E) disrupts PBD-kinase domain interactions to form an open, active state. (C) S588A mutation weakens but does not fully disrupt PBD-kinase interaction, resulting in a conformationally closed but “loose” state that may be more easily activated in the presence of substrate. (D) S299E mutation may result in a similar “loose” and closed conformation similar to S588A (left), or may directly cause allosteric opening of the enzymatic active site (right) to increase kinase activity. In contrast, S299A mutation has no apparent effect and is functionally wild-type (not shown).