Figure 1.
Identification of Miz1 as a positive regulator of Hh signaling.
(A) Illustration of Miz1 mutants. Wildtype (wt) Miz1 (total 803 amino acid residues) contains a POZ domain (amino acid residue 1–104) and a Myc-interacting domain (amino acid residue 641–715) POZ domain is deleted in Miz1ΔPOZ mutant, and the Myc-interacting domain is deleted in Miz1Δ641–715 mutant. (B) Miz1 interacts with Smo, and such interaction is independent of the POZ domain of Miz1. HEK293 cells were transiently transfected with Myc-tagged Miz1 or a POZ domain deletion mutant Miz1ΔPOZ along with Flag-Smo as indicated. Co-immunoprecipitation experiments were performed using the Flag antibody. The presence of Miz1 full length and Miz1ΔPOZ was detected using the Myc antibody. (C) Miz1 also interacts with Gli2, and the POZ domain of Miz1 is required for the interaction between Miz1 and Gli2. HEK293 cells were transiently transfected with HA-Gli2 together with vector, Flag-Miz1, or Flag- Miz1ΔPOZ as indicated. Co-immunoprecipitation experiments were performed using the Flag antibody. The presence of Gli2 was detected using the HA antibody. (D) Miz1 promotes the activation of Gli activity. C3H10T1/2 cells were transfected with the Hh-responsive reporter (9× Gli-luciferase) together with one of the corresponding constructs as indicated. The Gli reporter activity was determined using the ratio of firefly/Renilla luciferase light-units. The value in cells transfected with pcDNA3 was normalized to 1 and used as control, and the relative Gli activity was obtained by comparing to the control. Data shown in the graph represent the mean ± SEM (n = 3, **: p<0.001, ns: not significant, compared to the pcDNA3 sample, t test). (F) Miz1 activates Gli activity downstream of Smo receptor. Smo−/− MEF cells were transfected with the Hh-responsive reporter (9×Gli-luciferase) and pRL-TK together with one of the corresponding constructs as indicated. Gli reporter activities were determined as described in (D). Data shown in the graph represent the mean ± SEM (n = 3, **: p<0.001, ns: not significant, compared to the pcDNA3 sample, t test).
Figure 2.
Knockdown of endogenous Miz1 attenuates Hh signaling.
(A) Knock-down of Miz1 diminishes SAG-dependent activation of Gli activity. Stable NIH 3T3 cells infected with either a scrambled shRNA (Scr) or Miz1 shRNA lentiviruses (#3 and #5) were transfected with the Hh-responsive reporter and treated with DMSO or SAG. The Gli reporter activity was determined using the ratio of firefly/Renilla luciferase light-units. The value in Scr control cells treated with DMSO was set to 1, and relative Gli activities for all other conditions was normalized accordingly. Data shown in the graph represent the mean ± SEM (n = 3, **: p<0.001, compared to the SAG-stimulated Scr sample, t test). The knockdown efficiency of Miz1 was verified and shown in upper panels. Specifically, endogenous Miz1 was immunoprecipitated from cell lysates and subsequently detected by immunoblotting using the Miz1 antibody. β-actin was used as a loading control. (B) and (C) Loss of Miz1 decreases the mRNA expression of Hh target genes. The scrambled control (Scr) and Miz1 knockdown (#3 and #5) cells were treated with DMSO or SAG for 24 hours. The mRNA levels of Gli1 (B) and Ptc1 (C) were analyzed using qRT-PCR. Data shown in both graphs represent the mean ± SEM (n = 3, **, p<0.001, compared to the Scr sample, t test).
Figure 3.
Miz1 regulates Hh signaling in primary cilia.
(A) Confocal images showing the localization of Miz1 in primary cilia. NIH 3T3 cells treated with DMSO or SAG were co-stained with antibodies for Miz1 (green color) and acetylated-tubulin (red color, to indicate primary cilia). Cell nuclei were stained with DAPI (blue color). Co-localization is shown in the overlay panel. Middle and bottom panels correspond to the regions in the dashed white boxes in the upper panels. (B) Quantification of the average intensity of Miz1 in primary cilia. Total 101 and 103 cilia were quantified for DMSO and SAG treated group, respectively (**, p<0.001, compared to the DMSO sample, t test). (C) Knockdown of Miz1 expression inhibits activation-dependent recruitment of Smo into primary cilia. NIH 3T3 cells expressing GFP-Smo were stably infected with scrambled (Scr) control or Miz1 shRNAs (#3 and #5). Cells treated with DMSO or SAG were stained with anti-acetylated tubulin (red color, to indicate primary cilia), and cell nuclei were indicated by DAPI staining (blue color). GFP-Smo was shown in green color. Overlay images are shown in upper panels while the single-channel GFP-Smo images are shown directly below. Arrowheads in each image indicate primary cilia. (D) Quantification of the average intensity of GFP-Smo in primary cilia. Data shown in the graph represent the mean ± SEM (n = 37, 204, 267, 268, 215, and 212 cilia from left to right; **: p<0.001, t test). (E) Knockdown of Miz1 inhibits the accumulation of Gli2 at the tip of primary cilia upon Hh activation. NIH 3T3 cells were stably infected with lentiviral particles encoding either scrambled control shRNA (Scr) or shRNAs against Miz1 (#3 and #5). Cells were treated with DMSO or SAG for 24 hours and subsequently co-stained with antibodies for Gli2 (green color) and detyrosinated-tubulin (GluTu, red color, to indicate primary cilia). Cell nuclei were stained with DAPI (blue color). Overlay images are shown in upper panels for each condition, and the enlarged images of areas enclosed by the dashed white boxes are shown in the panels directly below. (F) Quantification of the percentage Gli2-containing primary cilia. Data shown in the graph represent the mean ± SEM (n = 276, 406, 214, 239, 246, and 307 cilia from left to right. **: p<0.001, t test).
Figure 4.
Miz1 is transported into nucleus upon Hh activation.
(A) NIH 3T3 cells treated with DMSO, 0.25 μM SAG, or 0.25 μM SAG plus 1 μM GDC-0449 (SAG+GDC) were fractionated into cytosol (Cyto) and nuclear (Nu) fractions. Endogenous Miz1 was immunoprecipitated from each fraction first and analyzed using immunoblotting. The presence of Miz1 was detected using the Miz1 antibody. Lactate dehydrogenase (LDH) and Histone H2B (H2B) were used as protein markers of cytosol and nuclear fractions, respectively. β-actin was used as the loading control. (B) Graphic representation of data shown in panel (A). The amount of Miz1 detected in the nuclear fraction was normalized to that of cytosol, and this number (Nu/Cyto) for the DMSO treated group was set to 1. Data shown in the graph represent the mean ± SEM (n = 3; **: p<0.001, t test). (C) Smo is required for Miz1 translocation into the nucleus. Smo−/− MEF cells treated with DMSO or SAG were fractionated into cytosol (Cyto) and nuclear (Nu) fractions. Endogenous Miz1 was immunoprecipitated from each fraction first and analyzed using immunoblotting. Note that the amount of Miz1 in the cytosol or nucleus is not changed upon SAG treatment in Smo−/− MEF.
Figure 5.
Miz1 is required for nuclear translocation of Gli2.
(A) NIH 3T3 cells were stably infected with lentivirus encoding either the scrambled control shRNA (Scr) or shRNAs against Miz1 (#3 and #5). Cells treated with DMSO or 0.25 μM SAG for 24 hours were stained with the Gli2 antibody. Cell nuclei were stained with DAPI. Red dashed circles adapted from corresponding DAPI images indicate the regions of nuclei. (B) Graphic representation of data shown in panel (A). The amount of Gli2 in the nucleus was measured as the intensity of Gli2 fluorescence staining within the DAPI-positive nuclear regions [circled areas in images shown in (A)]. Data represent the mean ± SEM (n = 203, 127, 211, 210, 200 and 241 cells from left to right, respectively. **: p<0.001, t test).
Figure 6.
Knockdown of Miz1 inhibits cell proliferation and tumorigenesis in PZp53MED1 medulloblastoma cells.
(A) and (B) The rate of cell proliferation was measured in stable scrambled control (Scr) and Miz1 knockdown (#3 and #5) PZp53MED1 cells using WST-1 (A) and 3H-thymidine incorporation assays (B). Data shown in the graphs represent the mean ± SEM (n = 3; **: p<0.001, compared to the Scr sample, t test). (C) and (D) The levels of mRNA expression for Gli1 (C) and Ptch1 (D) were measured using qRT-PCR in control and Miz1 knockdown PZp53MED1 cells. Data shown in the graphs represent the mean ± SEM (n = 4 for Gli1 and n = 3 for Ptch1; *: p<0.05; and **: p<0.001, compared to the Scr sample, t test). (E) SCID mice were inoculated subcutaneously with stable control and Miz1 knockdown PZp53MED1 cells. Image shows the tumors dissected from mice at the end of the experiment. ND represents non-detectable tumor growth around the injection sites. (F) Quantitative representation of tumor weights measured. Data in the graph represent the mean ± SEM (n = 5 for each group; *: p<0.05, compared to the Scr sample, t test).