Differential response of TC-32 cells to dilutions of Wnt-1 and Wnt-3a CM. After 3 h incubation, whole-cell lysates were immunoblotted with antibodies to Dvl-2, Dvl-3, and HSP70, the last serving as a loading control. Control cells (Ctl.) were incubated with serum-free culture medium for 3 h before processing. Arrows highlight doublet bands indicative of phosphorylation.

CK1δ and CK1ε both contribute to Dvl phosphorylation but have contrasting roles in neurite outgrowth. (A) TC-32 cells were treated with siRNA reagents targeting expression of luciferase (negative control), CK1δ, and/or CK1ε, and subsequently incubated for 3 h with serum-free culture fluid or 1:10 dilution of Wnt-3a CM. Cell lysates were immunoblotted for Dvl-2, CK1δ, CK1ε, and HSP70. (B) Neurite outgrowth analysis in TC-32 cells treated with siRNA reagents directed against luciferase, CK1δ, or CK1ε, followed by 3 h incubation in the presence or absence of Wnt-3a. The percentage of cells with long neurites was determined and normalized to the percentage observed in cells treated with luciferase siRNA in the absence of Wnt-3a. Results are the means ± SD of three independent experiments. **, P < 0.01; *, P < 0.05. (C) Representative image of phalloidin 488–stained TC-32 cell treated with CK1ε siRNA and no Wnt-3a. Bar, 20 µm. (D) Representative images of phalloidin 488–stained TC-32 cells treated with CK1ε siRNA vs. CK1ε + CK1δ siRNA. Bar, 20 µm. (E) Neurite outgrowth analysis in TC-32 cells treated with siRNA reagents directed against luciferase, CK1δ, and/or CK1ε. Results are presented as described in B. ***, P < 0.001. (F) Immunoblot analysis of CK1δ, CK1ε, and HSP70 in TC-32 cell lysates after siRNA treatment described in E.

Contrasting centrosomal localization of endogenous CK1δ and CK1ε. (A) TC-32 cells were cultured in RPMI medium (Control) in the absence or presence of Wnt-3a, fixed in methanol, and stained for CK1δ, the centrosomal marker γ-tubulin, and DNA (DAPI). Arrows point to colocalized signals. Bars, 10 µm. (B) TC-32 cells were cultured as in A, fixed in formaldehyde, and stained for CK1ε, the centrosomal marker pericentrin, and DNA (DAPI). Bars: (top panels) 20 µm; (bottom panels) 5 µm.

C-terminal domain of CK1δ is necessary and sufficient for centrosomal localization. (A) Schematic diagram of CK1δ and CK1ε protein sequences. Numbers of amino acid (aa) residues in domains from mouse (m) and human (h) proteins are indicated along with the percent sequence identity of CK1δ and CK1ε domains in each species. Domain sequences were obtained at http://www.uniprot.org and amino acid sequence analysis was performed with resources at http://www.ebi.ac.uk/Tools/clustalw2/index.html. (B) Immunofluorescent staining of HeLa cells stably transfected with lentiviral vector encoding Myc-tagged full-length mouse CK1δ or CK1ε, or chimeras in which their C-terminal domains were interchanged (mCK1δ/ε contains the C-terminal domain of CK1ε; mCK1ε/δ contains the C-terminal domain of CK1δ). Cells were fixed in methanol, stained for Myc, pericentrin, and DNA (DAPI). Bars: (top two rows) 10 µm; (bottom two rows) 5 µm. Arrows point to colocalized signals. Magnified area corresponds to the box in adjacent panel to the left. (C) Immunofluorescent signal in HeLa cells transiently expressing EGFP, δCT-EGFP, or εCT-EGFP. Cells were fixed in formaldehyde and co-stained with pericentrin antibody and DAPI. Bars, 5 µm. Arrows point to colocalized signals. Magnified area corresponds to the box in adjacent panel to the left.

Centrosomal localization signal of CK1δ was delineated by deletion mutant analysis. (A) Schematic diagram of EGFP fusion proteins containing varying segments from the C-terminal domain of mouse CK1δ (boundaries of segments are indicated by amino acid residue numbers). (B) Centrosomal localization of δCT-EGFP fusion proteins. For each of the indicated δCT-EGFP fusion proteins, colocalization with pericentrin was ascertained in ∼30 cells after transient transfection of HeLa cells. Semi-quantitative analysis was based on the intensity of EGFP signal that colocalized with pericentrin relative to EGFP signal elsewhere in the cell. Intense signal that colocalized with pericentrin was scored as ++ (black), colocalizing signal intensity comparable to that seen elsewhere in the cell was + (dark gray), weak signal was +/− (light gray), and no signal was − (white). The bar graph displays the percentage of cells in each category for all the fusion proteins. See also Fig. S2.

δCT-EGFP, but not εCT-EGFP, displaced CK1δ from the centrosome and inhibited Wnt-3a–dependent neurite outgrowth. (A) Representative confocal micrographs of TC-32 cells that were cotransfected with Myc-hCK1δ and the indicated EGFP constructs, and subsequently probed for Myc and EGFP distribution along with pericentrin and DNA (DAPI). Arrowheads highlight pericentrin signals, arrows indicate centrosomal localization of Myc-hCK1δ. Bars, 10 µm. (B) Semi-quantitative analysis of Myc-hCK1δ colocalization with pericentrin when coexpressed with the indicated EGFP constructs. Percentage of cells with clear colocalization is shown in black, questionable colocalization in gray, and no colocalization in white. Approximately 30 cells were analyzed in each treatment group. (C and D) Immunoblot analysis of the various EGFP derivatives transiently coexpressed in the centrosomal displacement experiments. (E) Wnt-3a–dependent neurite outgrowth in TC-32 cells transiently expressing EGFP, δCT-EGFP, or εCT-EGFP. The presence of neurites was quantified in ∼30 cells expressing the indicated EGFP proteins and incubated in the absence or presence of 100 ng/ml Wnt-3a for 3 h. Results are expressed as the mean ± SD of three independent experiments. ***, P < 0.001. (F) Immunoblot analysis of EGFP derivatives expressed in the neurite outgrowth experiments.

mCK1ε/δCT(278–364) rescued Wnt-3a–dependent neurite outgrowth otherwise inhibited by CK1δ siRNA. (A) Wnt-3a–dependent neurite outgrowth in TC-32 cells treated with hCK1δ siRNA targeting 3′-UTR sequence and transiently transfected with empty vector (pcDNA3.3 Ctl.) or construct expressing mouse full-length CK1δ, full-length CK1ε, or chimera consisting of CK1ε N-terminal and kinase domains plus residues 278–364 from mouse CK1δ C-terminal domain. The presence of neurites was quantified in ∼30 cells expressing Myc-tagged CK1 protein for each treatment group. Data are from one of two experiments with similar results. (B) Immunoblot analysis of ectopically expressed Myc-tagged CK1 proteins, endogenous CK1δ, and HSP70 in the neurite outgrowth experiment. Results illustrate the efficacy and specificity of knockdown with siRNA directed against human CK1δ 3′-UTR sequence.