Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2008 Mar 24;180(6):1133-47.
doi: 10.1083/jcb.200705148. Epub 2008 Mar 17.

Protein Phosphatase 4 Catalytic Subunit Regulates Cdk1 Activity and Microtubule Organization via NDEL1 Dephosphorylation

Affiliations
Free PMC article

Protein Phosphatase 4 Catalytic Subunit Regulates Cdk1 Activity and Microtubule Organization via NDEL1 Dephosphorylation

Kazuhito Toyo-oka et al. J Cell Biol. .
Free PMC article

Abstract

Protein phosphatase 4 catalytic subunit (PP4c) is a PP2A-related protein serine/threonine phosphatase with important functions in a variety of cellular processes, including microtubule (MT) growth/organization, apoptosis, and tumor necrosis factor signaling. In this study, we report that NDEL1 is a substrate of PP4c, and PP4c selectively dephosphorylates NDEL1 at Cdk1 sites. We also demonstrate that PP4c negatively regulates Cdk1 activity at the centrosome. Targeted disruption of PP4c reveals disorganization of MTs and disorganized MT array. Loss of PP4c leads to an unscheduled activation of Cdk1 in interphase, which results in the abnormal phosphorylation of NDEL1. In addition, abnormal NDEL1 phosphorylation facilitates excessive recruitment of katanin p60 to the centrosome, suggesting that MT defects may be attributed to katanin p60 in excess. Inhibition of Cdk1, NDEL1, or katanin p60 rescues the defective MT organization caused by PP4 inhibition. Our work uncovers a unique regulatory mechanism of MT organization by PP4c through its targets Cdk1 and NDEL1 via regulation of katanin p60 distribution.

Figures

Figure 1.
Figure 1.
PP4c dephosphorylates NDEL1 at Cdk1 sites and suppresses Cdk1 activation. (A, top) We examined whether PP4c dephosphorylates phospho-NDEL1 (P-NDEL1) that was phosphorylated by GST-Cdk1 using recombinant proteins. Note the lower mobility of NDEL1 phosphorylated by GST-Cdk1. PP4c efficiently removed phosphate from one of the Cdk1 phosphorylation sites of NDEL1 (T219). (bottom) We tested whether GST-PP4c dephosphorylates P-NDEL1 phosphorylated by Aurora A kinase. GST-PP4c did not show any dephosphorylation activity at the GST–Aurora A phosphorylation site (S251). Western blotting pattern using an anti-NDEL1, an antiphospho-T219 antibody (Cdk1 site), or an antiphospho-S251 antibody (Aurora A site) is shown at the bottom of each. Note that signal by antiphospho-T219 antibody was diminished after dephosphorylation by PP4c. (B) Subcellular distribution of PP4c (left), NDEL1 (middle), and PP4R1 (right) in asynchronously growing HeLa cells at interphase or prophase (representative images of three independent experiments). Arrowheads indicate centrosomes. (C) Examination of PP4c distribution and phosphorylation of cyclin B1 (left) and NDEL1 (right). Synchronously growing HeLa cells were stained with the indicated antibodies for phosphorylated proteins. Arrowheads indicate the centrosomes (representative images of three independent experiments). (D) Persistent expression of PP4c at the centrosome prevented phosphorylation of cyclin B1 (top) and NDEL1 (middle). Synchronously growing HeLa cells transfected with constructs as indicated above the panels were costained with the indicated antibodies. Images were captured at G2 or prophase. The distances of separated centrosomes are summarized at the bottom (representative images of five independent experiments). (bottom) Statistical analysis of centrosomal distances. The p-value was calculated using an unpaired t test (*, P < 0.001; one example of three independent experiments; n = 200). Error bars represent SEM. Bars, 10 μm.
Figure 2.
Figure 2.
Functional relationships between PP4c and Cdk1. (A) PP4c dephosphorylates cyclin B1 (top left) and NDEL1 (top right) in vivo. Synchronously growing HeLa cells cotransfected with constructs as indicated above the panels and were costained with the indicated antibodies. Images were captured at G2. To express active Cdk1 in G2, mitotic Cdk1 was used. Mitotic Cdk1 efficiently phosphorylated cyclin B1 and NDEL1 in vivo in G2. These phosphorylations were clearly abolished by the cotransfection of PP4c. The frequency of phosphorylation is summarized at the bottom (one example of three independent experiments; n = 100), and statistical analysis of phosphorylation-positive cells is shown. (B) Expression of mitotic Cdk1 facilitates recruitment of katanin p60 to the centrosome (top left). Synchronously growing HeLa cells were cotransfected with constructs as indicated above the panels and were costained with the indicated antibodies. Images were captured at G2. Expression of mitotic Cdk1 in G2 HeLa cells revealed augmentation of the concentration of katanin p60 at the centrosome (one example of three independent experiments; n = 100). The MT array appeared sparsely distributed in mitotic Cdk1-expressed HeLa cells (right). Statistical analysis of the fluorescence intensity of p60 is shown at the bottom left. Error bars represent SEM. (A and B) Arrowheads indicate centrosomes. *, P < 0.001; **, P < 0.05. Bars, 10 μm.
Figure 3.
Figure 3.
Disorganization of MTs in PP4c−/− MEF cells. (A) Expression of PP4c in PP4ccko/cko and PP4c −/− MEF cells. (left) Immunofluorescent staining was performed with an anti-PP4c antibody to assess the expression of PP4c 48 h after infection of adeno-Cre (representative of each genotype; n = 50). Uninfected PP4ccko/cko MEF cells were used as controls. Arrowheads indicate centrosomal staining of PP4c. (right) Western blotting analysis of PP4c, PP4R1, and NDEL1 expression in PP4c+/+, PP4ccko/cko, and PP4c −/− MEF cells. Representatives of three independent experiments are shown. (B) Severe MT disorganization in PP4c −/− MEF cells. MEF cells for each genotype were stained with an anti–β-tubulin antibody 48 h after infection of adeno-Cre to PP4ccko/cko MEF cells. PP4c+/+ MEF cells infected with adeno-Cre were used for controls. MT patterns were categorized into four groups as indicated at the bottom of each panel. The relative proportions of each pattern are shown in the bottom panel (one example of three independent experiments; n = 100 for each genotype). (C) MTs in PP4c −/− MEF cells were destabilized. Immunostaining was performed using antiacetylated tubulin 48 h after infection of PP4ccko/cko MEF cells with adeno-Cre. PP4c+/+ MEF cells infected with adeno-Cre were used for controls. MEF cells lacking PP4c revealed a clear reduction of acetylated tubulin (one example of three independent experiments; n = 60 for each genotype). Arrows indicate Cre-positive MEF cells. Bars, 10 μm.
Figure 4.
Figure 4.
Unscheduled activation of Cdk1 in PP4c−/− MEF cells. (A) Aberrant phosphorylation at T219 of NDEL1 in PP4c −/− MEF cells (48 h after infection of PP4ccko/cko MEF cells with adeno-Cre). Uninfected PP4ccko/cko MEF cells were used as controls. (top) T219-phosphorylated NDEL1 was detected by using a phospho-T219–specific monoclonal antibody (N219TP). Arrowheads indicate aberrant N219TP staining and γ-tubulin of PP4c −/− MEF cells in interphase. (middle) Statistical analysis of N219TP-positive cells in interphase (*, P < 0.001; one example of three independent experiments; n = 100). (bottom) Western blotting pattern using a phospho-T219–specific monoclonal antibody. An anti-NDEL1 antibody was used for a control. Note that phosphorylated NDEL1 displayed lower mobility. (B) Unscheduled phosphorylation of cyclin B1 of PP4c −/− MEF cells in interphase 48 h after infection of PP4ccko/cko MEF cells with adeno-Cre. Uninfected PP4ccko/cko MEF cells were used as controls. (top) Antiphospho–cyclin B1 (phospho-S123) staining in PP4c −/− MEF cells. (middle) Statistical analysis of phospho-S123–positive cells (*, P < 0.001; one example of three independent experiments; n = 100). Error bars represent SEM. (bottom) Western blotting pattern using an antiphospho–cyclin B1–specific monoclonal antibody. An anti–cyclin B1 antibody was used for a control. (A and B) Arrowheads indicate the positions of centrosomes. (C) Histone H1 kinase assay. GFP-PACT-Cdk1 and Red-Cre plasmids were transfected to PP4c+/+ or PP4ccko/cko MEF cells. GFP-PACT-Cdk1 was precipitated by an anti-GFP antibody 48 h after transfection followed by a kinase assay using histone H1 as a substrate. Note that GFP-PACT-Cdk1 extracted from PP4c −/− MEF cells displayed higher kinase activity. (D) Flow cytometric analysis of PP4c −/− MEF cells (top) and immunostaining pattern using an antiphosphohistone H3 antibody (bottom) 48 h after infection of PP4ccko/cko MEF cells with adeno-Cre. Uninfected PP4ccko/cko MEF cells were used as controls. Although flow cytometry revealed the enrichment of cell populations in 2n and 4c, phosphohistone H3–positive cells were rarely detected, suggesting that most of the population of PP4c −/− MEF cells was arrested in G2 before entering into prophase. One example of three independent experiments is shown. Bars, 10 μm.
Figure 5.
Figure 5.
Abnormal accumulation of katanin p60 in PP4c−/− MEF cells. (A) Overexpression of katanin p60 in MEFs resulted in defective MT array, which was seen in PP4c −/− MEF cells. Statistical analysis was performed (right; n = 50 in each group). (B) Abnormal katanin p60 accumulation at the centrosome in PP4c −/− MEF cells 48 h after infection of PP4ccko/cko MEF cells with adeno-Cre. Uninfected PP4ccko/cko MEF cells were used as controls. (top) MEF cells of each genotype were stained with an anti–katanin p60 antibody. Fluorescence intensity was calculated using ImageJ software. (bottom) Fluorescence intensity in arbitrary units (*, P < 0.05; one example of three independent experiments; n = 100). Error bars represent SEM. (A and B) Arrowheads indicate the positions of centrosomes. (C) Immunoblotting analysis of sucrose gradient fractions of centrosomal extracts using antibodies against γ-tubulin, phospho-T-219 NDEL1, and katanin p60. Proteins were extracted from uninfected PP4ccko/cko MEF cells or PP4c −/− MEF cells 48 h after infection of PP4ccko/cko MEF cells with adeno-Cre and were subjected to sucrose density gradient fractionations. One example of three independent experiments is shown. Note the presence of phosphorylated NDEL1 with γ-tubulin in PP4c −/− MEF cells and a more restricted distribution of katanin p60 in the central fractions with phosphorylated NDEL1. Bars, 10 μm.
Figure 6.
Figure 6.
Examination of MT dynamics in PP4c−/− MEF cells. (A) Impairment of MT stability at the centrosome after nucleation in PP4c −/− MEF cells 48 h after infection of PP4ccko/cko MEF cells with adeno-Cre. Time (given in seconds) after washout of nocodazole is indicated in the top right corners. Arrows indicate Cre-positive cells. (top) β-Tubulin staining after MT regrowth in PP4ccko/cko MEF cells (DAPI staining only) and PP4c −/− MEF cells (arrows indicate DAPI and RFP-Cre positive). (bottom) Statistical analysis of the percentage of cells with observable MT array from the centrosome (n = 200 for each time point of each genotype). (B) Statistical analysis of the newly emanated MTs from the centrosome (*, P < 0.001; n = 12 for PP4ccko/cko MEF cells and n = 20 for PP4c −/− MEF cells). (C) Statistical analysis of the number of cells with stacked EB1-GFP at the centrosome (*, P < 0.001; one example of three independent experiments; n = 50). (D) Statistical analysis of fluorescence intensity (in arbitrary units) of EB1-GFP at the centrosome is shown (**, P < 0.05; n = 30 for each genotype). (E) Statistical analysis of the number of plus end tips of MTs in the cytoplasm. We calculated the number of EB1-GFP spots per 100 μm2 area in the fixed MEF cells (**, P < 0.05; n = 30 for each genotype). Error bars represent SEM. Bars, 20 μm.
Figure 7.
Figure 7.
Rescue experiments in PP4c-disrupted MEF cells by a Cdk1 inhibitor, additional disruption of Ndel1, and siRNA against katanin p60. (A) Depletion of Cdk1 by siRNA rescued the defect of MTs in PP4c −/− MEF cells (type A). (left) MT array was rescued by the depletion of Cdk1. (middle) T219 phosphorylation of NDEL1 (left) and katanin p60 distribution (right) under control siRNA or Cdk1 siRNA in PP4c −/− MEF cells (RFP-Cre positive). One example of three independent experiments is shown. (right) Statistical analysis of the effect of Cdk1 inhibition on MT defects (n = 200 for each of PP4ccko/cko; control siRNA, PP4ccko/cko; Cdk1 siRNA, PP4c −/−; control siRNA, and PP4c −/−; Cdk1 siRNA). (B) Ndel1 deletion can rescue the defect of MTs in PP4c −/− MEF cells (type A). (left) PP4c −/−/Ndel1−/− MEF cells show a normal MT array compared with PP4c −/−/Ndel1+/+ MEF cells. (middle) S123 phosphorylation of cyclin B1 (left) and katanin p60 distribution (right) in PP4ccko/cko/Ndel1cko/cko MEF cells (RFP-Cre negative) or PP4c −/−/Ndel1−/− MEF cells (RFP-Cre positive). One example of three independent experiments is shown. (right) Statistical analysis of the effect of Ndel1 deletion on MT abnormality (n = 100 for each of PP4ccko/cko/Ndel1cko/cko, PP4c −/−/Ndel1+/+, PP4c+/+/Ndel1−/−, and PP4c −/−/Ndel1−/−). (C) Rescue experiments with siRNA against katanin p60 in PP4c −/− MEF cells (type A). (left) Depletion of katanin p60–rescued MT defects in PP4c −/− MEF cells. (middle) Depletion of katanin p60 did not prevent the aberrant NDEL1 phosphorylation (left) and the unscheduled cyclin B1 phosphorylation (right) in PP4c −/− MEF cells (RFP-Cre positive). One example of three independent experiments is shown. (right) Statistical analysis of the effect of katanin p60 depletion by siRNA against katanin p60 on MT defects (n = 200 each of Cre/ctrlRNAi+, CRE+/ctrlRNAi+, CRE/p60RNAi+, and CRE+/p60RNAi+). (A–C) Arrowheads indicate the positions of centrosomes. Bars, 20 μm.

Similar articles

See all similar articles

Cited by 36 articles

See all "Cited by" articles

References

    1. Abal, M., M. Piel, V. Bouckson-Castaing, M. Mogensen, J.B. Sibarita, and M. Bornens. 2002. Microtubule release from the centrosome in migrating cells. J. Cell Biol. 159:731–737. - PMC - PubMed
    1. Andreassen, P.R., F.B. Lacroix, E. Villa-Moruzzi, and R.L. Margolis. 1998. Differential subcellular localization of protein phosphatase-1 α, γ1, and δ isoforms during both interphase and mitosis in mammalian cells. J. Cell Biol. 141:1207–1215. - PMC - PubMed
    1. Askham, J.M., K.T. Vaughan, H.V. Goodson, and E.E. Morrison. 2002. Evidence that an interaction between EB1 and p150(Glued) is required for the formation and maintenance of a radial microtubule array anchored at the centrosome. Mol. Biol. Cell. 13:3627–3645. - PMC - PubMed
    1. Berrueta, L., J.S. Tirnauer, S.C. Schuyler, D. Pellman, and B.E. Bierer. 1999. The APC-associated protein EB1 associates with components of the dynactin complex and cytoplasmic dynein intermediate chain. Curr. Biol. 9:425–428. - PubMed
    1. Blagden, S.P., and D.M. Glover. 2003. Polar expeditions–provisioning the centrosome for mitosis. Nat. Cell Biol. 5:505–511. - PubMed

Publication types

MeSH terms

Feedback