M-Phase Phosphoprotein 9 regulates ciliogenesis by modulating CP110-CEP97 complex localization at the mother centriole

Abstract

The primary cilium is elongated from the mother centriole and has diverse signaling roles during development and disease. The CP110-CEP97 complex functions as a negative regulator of ciliogenesis, although the mechanisms regulating its mother centriole localization are poorly understood. Here we show that M-Phase Phosphoprotein 9 (MPP9) is recruited by Kinesin Family Member 24 (KIF24) to the distal end of mother centriole where it forms a ring-like structure and recruits CP110-CEP97 by directly binding CEP97. Loss of MPP9 causes abnormal primary cilia formation in growing cells and mouse kidneys. After phosphorylation by Tau Tubulin Kinase 2 (TTBK2) at the beginning of ciliogenesis, MPP9 is targeted for degradation via the ubiquitin-proteasome system, which facilitates the removal of CP110 and CEP97 from the distal end of the mother centriole. Thus, MPP9 acts as a regulator of ciliogenesis by regulating the localization of CP110-CEP97 at the mother centriole.

Fig. 1

MPP9 represents the ring-like structures at the distal ends of the centrioles. a Immunofluorescence of MPP9 (green) and Centrin-3 (red, upper) or C-Nap1 (red, lower) in U2OS cells. b Immunofluorescence of MPP9 (green) and acetylated-tubulin (Acet-Tub, red) in mitotic HeLa cells. DNA was stained with DAPI (white). c Immunoblots of MPP9 during the cell cycle in hTERT RPE-1 cells. Tubulin was used as a loading control. Relative amounts of MPP9 were quantified and normalized to tubulin. d 3D-SIM images of U2OS cells triple-immunostained with antibodies against MPP9 (green), CP110 (purple), and Centrin-3 (red). Arrows show distal ends of the centrioles. e 3D-SIM images of MPP9-GFP (green) overexpressing U2OS cells co-immunostained with antibodies against CP110 (red) and C-Nap1 (purple). Arrows show proximal ends of the centrioles. f–i 3D-SIM images of U2OS cells co-immunostained with antibodies against MPP9 (f, green) and Acet-Tub (f, red) or MPP9 (h, red) and CEP164 (h, green). g, i are the intensity plots of the rings from f, h, respectively. j Immuno-electron microscopy images. U2OS cells were labeled with anti-MPP9 antibodies followed by anti-rabbit IgG-gold (10 nm) secondary antibodies. Green, gold particles. Schematics of immuno-electron microscopy images are shown. k Schematic of MPP9 localization at centrosomes. Scale bars: 5 μm (b); 1 μm (a); 500 nm (d, e)

Fig. 2

Depletion of MPP9 leads to aberrant ciliogenesis. a Immunostaining of acetylated-tubulin (Acet-Tub, red) and MPP9 (green) in normal and serum-free medium-treated hTERT RPE-1 cells. b Quantification of the percentage of cells with MPP9 at distal ends of the mother centrioles after serum starvation. c Immunoblots showing depletion of MPP9 by siRNAs (#1 and #2) transfection and rescue of MPP9 expression by overexpressing Flag-tagged siRNA-resistant MPP9 (Flag-ResMPP9). Tubulin was used as a loading control. Relative amounts of MPP9 were quantified and normalized to tubulin. d Quantification of ciliogenesis in control-siRNA, MPP9-siRNA, or CP110-siRNA, or MPP9-siRNA together with Flag-ResMPP9-rescued hTERT RPE-1 cells. e Immunostaining of Centrin-3 (red) and Arl13B (green) in MPP9-siRNA-treated and Flag-ResMPP9-rescued hTERT RPE-1 cells. DNA was stained with DAPI (blue). f Whole mpp9^+/+ and mpp9^–/– embryos at embryonic day (E) 10.5. g The male mpp9^+/+ (left) and mpp9^–/– (right) mice at 1 month after birth. h Line graph showing the body weight of mpp9^+/+ and mpp9^–/– mice at the indicated times (n = 6 mice for each group). i Immunoblots of MPP9 and CEP97 in the kidneys of mpp9^+/+ or mpp9^–/– mice at 1 month after birth. GAPDH was used as a loading control. j Kidneys from mpp9 ^+/+ and mpp9 ^–/– mice (n = 3 mice for each group) at 1 month were stained with Acet-Tub (red) and DAPI (DNA; blue). White dashed lines represent the border of each renal tubule. k The percentage of ciliated cells from j. For b, d, h, and k, bars represent the means ± S.E.M for three independent experiments. n.s., not significant, *p < 0.05, **p < 0.01, ***p < 0.001, as determined by unpaired two-tailed Student’s t-test (b, h, and k), one-way ANOVA analysis (d). Scale bars: 1 mm (f); 5 μm (e main, j); 1 μm (a)

Fig. 3

KIF24 recruits MPP9 at the distal end of the mother centriole in hTERT RPE-1 cells. a Lysates from hTERT RPE-1 cells were subjected to immunoprecipitation (IP) with anti-MPP9 antibody, and immunoblotting with the indicated antibodies. b Immunoblots showing depletion of MPP9 or KIF24 by siRNA in hTERT RPE-1 cells. Tubulin was used as a loading control. c Immunostaining of acetylated-tubulin (Acet-Tub, red) and MPP9 (green) in the control- or KIF24-siRNA-treated hTERT RPE-1 cells. d Quantification of the percentage of cells with the indicated number of MPP9 dots from c. e Immunostaining of Acet-Tub (red) and KIF24 (green) in the control- or MPP9-siRNA-treated hTERT RPE-1 cells. f Quantification of the fluorescence intensity of KIF24 at centrosomes from e (n = 100 cells for each group). g Immunostaining of MPP9 (green) and Flag (red) in KIF24-3’UTR siRNA-treated or/and Flag-KIF24-rescued hTERT RPE-1 cells. h Immunoblots showing depletion of KIF24 by 3’UTR siRNA transfection or/and rescue of KIF24 expression by overexpression of Flag-KIF24. Tubulin was used as a loading control. i Quantification of the percentage of cells with 4-dot MPP9 in control-, KIF24-3’UTR siRNA, or/and Flag-KIF24-rescued hTERT RPE-1 cells. j Immunoblots showing depletion of MPP9 or/and KIF24 by siRNA in hTERT RPE-1 cells. Tubulin was used as a loading control. k Quantification of ciliogenesis in control-, MPP9-, or/and KIF24-siRNA-treated hTERT RPE-1 cells. l Immunoblots showing depletion of MPP9 or KIF24 by siRNA in U2OS cells. Tubulin was used as a loading control. m Quantification of the percentage of cells with the indicated number of MPP9 dots in control- and KIF24-siRNA-treated U2OS cells. For d, f, i, k, and m, bars represent the means ± S.E.M for three independent experiments. n.s., not significant, *p < 0.05, **p < 0.01, as determined by unpaired two-tailed Student’s t-test (d, f, k, m) and one-way ANOVA analysis (i). Scale bars: 1 μm (c, e, g)

Fig. 4

MPP9 recruits CEP97 and CP110 at the distal end of the mother centriole in hTERT RPE-1 cells. a Immunoblots showing depletion of MPP9, CP110, or CEP97 by siRNA in hTERT RPE-1 cells. GAPDH was used as a loading control. b Immunostaining of acetylated-tubulin (Acet-Tub, red) and CP110 (green) in the control- or MPP9-siRNA-treated hTERT RPE-1 cells. c Quantification of the percentage of cells with the indicated number of CP110 dots from b. d Immunostaining of Acet-Tub (red) and CEP97 (green) in the control- or MPP9-siRNA-treated hTERT RPE-1 cells. e Quantification of the percentage of cells with the indicated number of CEP97 dots from d. f Immunoblots showing expression of the indicated proteins in MPP9-depleted hTERT RPE-1 cells after treatment with DMSO or MG132 for 4 h. Tubulin was used as a loading control. Relative amounts of MPP9, CEP97, and CP110 were quantified and normalized to tubulin. g Immunostaining of Centrin-3 (red) and CEP97 (green, upper) or CP110 (green, lower) after treating with DMSO or MG132 for 4 h in MPP9-depleted hTERT RPE-1 cells. h Immunostaining of CEP97 (red) and Flag-siRNA-resistant MPP9 (Flag-ResMPP9, green) or Flag-ResMPP9-△451-500 (green) in MPP9-depleted hTERT RPE-1 cells. i Immunoblots showing expression of the indicated proteins in hTERT RPE-1 cells transfected with MPP9-siRNA and Flag-siRNA-resistant MPP9 wild-type (Flag-ResMPP9-WT) or lacking 451–500 aa mutant (Flag-ResMPP9-△451–500). Tubulin was used as a loading control. Relative amounts of CEP97 and CP110 were quantified and normalized to tubulin. j Immunostaining of Centrin-3 (red) and Arl13B (green) after transfection of Flag-ResMPP9-WT or Flag-ResMPP9-△451–500 in MPP9-depleted hTERT RPE-1 cells. DNA was stained with DAPI (blue). k Quantification of ciliogenesis in MPP9-depleted hTERT RPE-1 cells overexpressing Flag-ResMPP9-WT or Flag-ResMPP9-△451–500. For c, e, k, bars represent the means ± S.E.M for three independent experiments. n.s., not significant, *p < 0.05, **p < 0.01, as determined by unpaired two-tailed Student’s t-test (c, e), and one-way ANOVA analysis (k). Scale bars: 5 μm (j main); 1 μm (b, d, g, h, j insets)

Fig. 5

TTBK2 phosphorylates MPP9 mainly at S629 site. a Immunoblots of lysates from HEK293T cells co-overexpressing Flag-MPP9 and the indicated cilia-related kinase. Tubulin was used as a loading control. b Immunoblots of lysates from HEK293T cells co-overexpressing Flag-tagged MPP9 full-length (FL) or the indicated MPP9 truncation mutants with TTBK2-GFP-WT or TTBK2-GFP-KD. Tubulin was used as a loading control. WT, wild-type; KD, kinase-dead. c Immunoblots of lysates from HEK293T cells co-overexpressing Flag-tagged MPP9 (401–800 aa) and TTBK2-GFP-WT or TTBK2-GFP-KD after treatment with λ-PPase. Tubulin was used as a loading control. d Mass spectrometric analysis showing phosphorylation sites of MPP9 at S629 and S636. e Immunoblots of lysates from HEK293T cells co-overexpressing TTBK2-GFP and Flag-MPP9 (401–800 aa)-WT, or the indicated unphosphorylatable mutants. Tubulin was used as a loading control. f GST-MPP9 (401–800 aa)-WT, or the indicated mutants were subjected to a TTBK2 kinase assay in vitro followed by autoradiography. Coomassie blue (CBB) staining showed GST-tagged proteins. Relative amounts of phosphorylated MPP9 signals were quantified. g Lysates of HEK293T cells co-overexpressing Flag-tagged MPP9-WT or the S629A mutant and TTBK2-GFP were subjected to immunoprecipitation (IP) and immunoblotting with anti-Flag and anti-Phospho-S629 antibodies. Relative amounts of S629-phosphorylated MPP9 were quantified. h Immunostaining of S629-phosphorylated MPP9 (red) and Flag in Flag-MPP9 (green, upper) or Flag-CEP170 (green, lower) overexpressing hTERT RPE-1 cells. i Immunostaining of S629-phosphorylated MPP9 (red) and Flag in Flag-MPP9 (green) or Flag-MPP9–629A (green) overexpressing hTERT RPE-1 cells after depleting MPP9. j Quantifications of S629-phosphorylated MPP9 signals in i (n = 50 cells for each group). k Immunostaining of S629-phosphorylated MPP9 (green) and Centrin-3 (red) in control- or TTBK2-siRNA transfected hTERT RPE-1 cells. l Quantifications of S629-phosphorylated MPP9 signals in k (n = 50 cells for each group). m Lysates of control-siRNA or TTBK2-siRNA transfected hTERT RPE-1 cells after serum starvation were subjected to immunoprecipitation and immunoblotting with the indicated antibodies. Relative amounts of S629-phosphorylated MPP9 were quantified and normalized to MPP9. For j, l, bars represent the means ± S.E.M for three independent experiments. ***p < 0.001, as determined by unpaired two-tailed Student’s t-test. Scale bars: 1 μm (h, i, k)

Fig. 6

Phosphorylation of MPP9 by TTBK2 promotes UPS-mediated degradation of MPP9. a Immunoblots of lysates from hTERT RPE-1 cells after serum starvation. Tubulin was used as a loading control. b Immunoblots following immunoprecipitation (IP) with an anti-Flag antibody using lysates from hTERT RPE-1 cells overexpressing the indicated proteins. c Immunoblots of MPP9 after serum starvation in hTERT RPE-1 cells. Tubulin was used as a loading control. Relative amounts of MPP9 were quantified and normalized to tubulin. d Immunostaining of MPP9 (green) and ODF2-HA (red) in hTERT RPE-1 cells. Arrows, the distal ends of the mother centrioles. e Quantification of the fluorescence intensity of MPP9 at the distal ends of the mother centrioles from d (n = 40 cells for each group). f Immunoblots of MPP9 and S629-phosphorylated MPP9 in cycling hTERT RPE-1 cells. g Immunoblots following immunoprecipitation with an anti-MPP9 antibody using lysates from control- or TTBK2-siRNA-treated hTERT RPE-1 cells. h Immunoblots of TTBK2 and MPP9 after serum starvation in control- or TTBK2-siRNA-treated hTERT RPE-1 cells. Tubulin was used as a loading control. i Immunoblots of lysates from HEK293T cells co-overexpressing the indicated proteins. Tubulin was used as a loading control. Relative amounts of Flag-tagged MPP9 were quantified and normalized to tubulin. WT, wild-type; KD, kinase-dead. j Immunoblots of lysates from HEK293T cells co-overexpressing the indicated proteins. Tubulin was used as a loading control. Relative amounts of Flag-MPP9 were quantified and normalized to tubulin. k Immunoblots following immunoprecipitation with an anti-Flag antibody using lysates from HEK293T cells overexpressing the indicated proteins. l Mass spectrometry analysis showing the ubiquitination of MPP9 at K632 (blue). Yellow, phosphorylation sites. m Immunoblots of lysates from HEK293T cells co-overexpressing Flag-tagged MPP9-WT or the K632R mutant with TTBK2-GFP-WT or TTBK2-GFP-KD. Tubulin was used as a loading control. Relative amounts of Flag-MPP9 were quantified and normalized to tubulin. n Immunoblots following immunoprecipitation with an anti-Flag antibody using lysates from HEK293T cells overexpressing the indicated proteins. For e, bars represent the means ± S.E.M for three independent experiments. n.s., not significant, **p < 0.01, as determined by one-way ANOVA analysis. Scale bar: 1 μm (d)

Fig. 7

MPP9 facilitates the removal of the CEP97-CP110 complex from centrosomes after phosphorylation. a Immunostaining of MPP9 (green) and acetylated-tubulin (Acet-tub, red) or CP110 (red) in hTERT RPE-1 cells after serum starvation and serum re-addition. b Line graph showing the percentage of ciliated hTERT RPE-1 cells, the cells presenting with 4 dots of MPP9, and cells with CP110 or CEP97 on both centrioles. c Immunoblots of lysates from hTERT RPE-1 cells stably overexpressing Flag-tagged MPP9 wild-type (WT) or the indicated unphosphorylatable mutants after serum starvation. Tubulin was used as a loading control. Relative amounts of MPP9 were quantified and normalized to tubulin. d Immunostaining of Arl13B (red, upper) and γ-tubulin (red, upper) or CEP97 (red, lower) in hTERT RPE-1 cell lines stably overexpressing Flag-tagged MPP9 or the indicated unphosphorylatable mutants (green). 2, 629, and 636 A. e Quantification of the percentage of cells with cilia (left) and cells with CEP97 dots on both centrioles (right) from d. f Immunoblots following immunoprecipitation (IP) with an anti-Flag antibody using lysates from HEK293T cells overexpressing the indicated proteins. g Schematic model depicting the function of MPP9 in cilia formation control. After serum starvation, MPP9 is phosphorylated by TTBK2 at S629 and S636, which promotes its ubiquitination at K632 and degradation via UPS. Subsequently, the degradation of MPP9 causes the removal of CEP97 and CP110 from the distal end of the mother centriole and initiates cilia formation. For b, e, bars represent the means ± S.E.M for three independent experiments. n.s., not significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, as determined by one-way ANOVA analysis (e). Scale bars: 1 μm (a, d)