Fig 2: Prom1 mediates SHH activation through Glis2 The mRNA expression of SHH effector genes (Gli1/2 and Glis1–3) on laser microdissection captured samples was evaluated by real-time RT–PCR analysis. The results are in arbitrary values after normalization for GapDH. Each sample was isolated from distinct mice (n = 3). Statistics was performed using Student's t-test: **P < 0.01. SCs, stem cells; TACs, transit amplifying cells.Representative IF staining of Gli1 (left panel, red) and Glis2 (right panel, red) on the stem cell and transit amplifying cell regions of lower incisor CLE at P7. Samples were counterstained with DAPI (blue). Dotted lines, basement membrane.3D reconstruction of Glis2 staining (see B, red) and its association with DAPI-labeled nuclei (blue). Nuclear transparency was set at 0 or 70% as indicated.Representative IF double staining of Glis2 (red, bottom panel) and AcTub (green, top and bottom panels) along stem cell–transit amplifying cell axis in lower incisor CLE at P7. Samples were counterstained with DAPI (blue). Note the association of Glis2 with primary cilia particularly in the transition zone between stem cell and transit amplifying cell regions.Volume of Sox2-positive region on the lower incisor CLE at P7 was evaluated in WT vs. Glis2 KO mice upon IF staining. Number of animals analyzed is indicated (n). Statistics was performed using Student's t-test: **P < 0.01. Data are presented as mean and standard deviation.Representative IF staining of Prom1 using specific antibodies targeting either its extracellular loop (top panels, green) or cytoplasmic C-terminal end (bottom panels, red) on the stem cell and transit amplifying cell regions of lower incisor CLE at P7 WT and Glis2 KO mice. Samples are counterstained with DAPI (blue). Dotted lines, basement membrane. Arrowheads mark approximate boundary of stem cell regions.Representative IF double staining of Glis2 (red) and AcTub (green) as a marker of primary cilia in stem cell region of WT vs. Prom1 KO mice. Note the reduction of Glis2 expression in Prom1-deficient animals.

Fig 3: Prom1 regulates ciliary dynamics by directly interacting with Hdac6 and Arl13b A, BSize of individual cilia in stem cell and transit amplifying cell regions of WT and Prom1 KO mice (A) and their density in the respective regions (B). Three animals were analyzed for each genotype. The number of quantified cilia is indicated (n). Statistics was performed using Dunnett's test: *P < 0.05; **P < 0.01. Data are presented as mean and standard deviation. SCs, stem cells; TACs, transit amplifying cells.C, DClonogenicity analysis of CLESCs treated with either Hdac6 inhibitor (tubacin) and appropriate controls (niltubacin and DMSO, C) or siArl13b and its siControl. Colonies were stained with crystal violet (D).E, FsiRNA knockdown efficiency of Arl13b (E) and Hdac6 (F) in Ki67-FUCCI CLESCs was evaluated by real-time RT–PCR analysis. The results are in arbitrary values after normalization for GapDH. Statistics was performed in triplicates using one-way ANOVA followed by Bonferroni's test: **P < 0.01. Data are presented as mean and standard deviation.GFlow cytometry analysis and quantification of samples indicated in (E and F).HMutation strategy of Hdac6-binding site on Prom1. The lysine (K) 138 in the first cytoplasmic loop of Prom1 was mutated to glutamine residue (Q). TM, transmembrane domain.IRepresentative IF double staining and 3D reconstruction of the Prom1 (green) and AcTub (red) at the apical domain of polarized MDCK cells expressing Prom1 and its K138Q mutant (top panels). Black-white inserts show Prom1 signals only. Same cells were analyzed by SEM (bottom panels). Note that Prom1 is asymmetrically distributed along the primary cilium (white arrows) as observed in vivo (see Fig 2C) and mutation K138Q significantly impacts on the primary cilium length. MV, microvilli.JIF staining and 3D reconstruction of AcTub-labeled (red) primary cilium at the apical membrane of parental MDCK cells (MDCK) and those cells transfected with Prom1 or its K138Q mutant. The images were used to quantify the primary cilium length (Appendix Fig S6B and C). Arrows indicate primary cilia with more than 4 µm in length.K–MCo-immunoisolation (IS) analyses were performed with paramagnetic beads using antibodies either against Arl13b (K) or Prom1 (AC133) or with anti-FITC antibody as control (L). Recovered materials were analyzed by immunoblotting against the indicated proteins. Brackets and asterisks indicate the plasma membrane and endoplasmic reticulum-associated forms of Prom1, respectively, while arrows show Arl13b. Arrowhead, immunoglobulin. The ratio Arl13b/Prom1 immunoreactivities was quantified (n = 3) (M). Mann-Whitney test: **P < 0.01. Data are presented as mean and standard deviation.

Fig 4: Epithelial Prom1 regulates CLESC maintenance and activation A, BRepresentative images (A) and quantitative analysis (B) of IF staining and 3D reconstruction on Sox2-positive region for WT vs. Prom1 KO mice. Ant: anterior; Post: posterior (A). Statistics was performed using Student's t-test: **P < 0.01. Data are presented as mean and standard deviation.CRepresentative IF double staining of Sox 2 (green) and Ki67 (red) on the stem cells and transit amplifying cell regions of WT vs. Prom1 KO mice on frontal sections. Samples are counterstained with DAPI (blue).DQuantification of Ki67+ cell number in Sox2-positive region based on IF data from WT and Prom1 KO mice (C). Statistics was performed using Student's t-test: **P < 0.01.EIllustration of CLE tissue recombination assay. P7 CLE was microdissected and then dissociated with Dispase II to achieve epithelium and mesenchyme. The Prom1 KO epithelium was then reassociated with WT mesenchyme or vice versa.FStereo images of three explants cultured with indicated recombination conditions before harvesting at day 3. Dotted line indicates the CLE.

Fig 5: Prom1 has a dynamic expression in the incisor CLE primary cilia and nuclei ARepresentative IF staining of Prom1 using specific antibody clone 13A4 targeting extracellular loop (green) on the stem cell and transit amplifying cell regions of lower incisor CLE at P7. Sample is counterstained with DAPI (blue). Dotted lines, basement membrane. SCs, stem cells; TACs, transit amplifying cells; Ams, ameloblasts.B3D reconstruction showing the association of Prom1 (green) with AcTub-labeled (red) primary cilia in stem cell and transit amplifying cell regions. Note that the expression of Prom1 is not limited to primary cilium but also to microvilli.CA representative example of Prom1 association with one primary cilium at the stem cell to transit amplifying cell transition region. Green channel transparency was set up to 70%.DRepresentative IF staining of Prom1 using antibodies directed either its extracellular loop (clone 13A4, green) or cytoplasmic C-terminal end (Biorbyt, Orb129549, red) on transit amplifying cell regions of the WT vs. Prom1 KO mice. Samples are counterstained with DAPI (blue). Note the lack of Prom1 labeling in Prom1 KO mice.E, FThe mRNA (E) and protein (F) profiling on shRNA-mediated Prom1 knockdown (3 different shRNAs were used, marked as NO. 1, 2, and 3) in cultured CLESCs. qRT–PCR results are in arbitrary values after normalization for GapDH. The immunoblotting was performed with anti-Prom1 C-terminal antibody. ß-Actin was used as loading control (F). Molecular mass markers are indicated. Statistics was performed using one-way ANOVA followed by Bonferroni's test: **P < 0.01. Data are presented as mean and standard deviation.GRepresentative IF staining of Prom1 (red) and Sox2 (green) in CLESCs in vitro. Sample is counterstained with DAPI (blue). Arrows indicate colocalization of Prom1 with low–medium level of Sox2 in the cell nuclei.H, ISubcellular fractionation of CLESCs into cytoplasm/membrane (C+M) and nuclear (N) fractions. Materials were analyzed by immunoblotting for Prom1 (H) and Arl13b (I). As controls, membranes were blotted for anti-Lamin A/C (nuclei markers) and GapDH (cytoplasm marker). Molecular mass markers are indicated. Note the presence of full-length Prom1 as well as Arl13b in nuclear fraction.