Fig 1: FBW7a promotes SOX9 protein turnover in a GSK3-dependent manner through the ubiquitin–proteasome system Endogenous SOX9 protein turnover in HCT116-FBW7 WT and KO cells over the course of 8 h following the addition of 100 ng/ml cycloheximide. ß-Actin indicates total protein loading for each sample.Quantitative PCR analysis of FBW7 and SOX9 mRNA transcripts in Daoy at 8 and 24 h following transfection with 20 nM siFBW7. The FBW7 and SOX9 mRNAs were adjusted to the B2M mRNA prior to being expressed relative to the siScr control. Data are expressed as mean + standard deviation from two independent experiments, with statistical significance determined by multiple-comparison two-way ANOVA with Bonferroni's post-test.Immunoblots of endogenous SOX9 protein in medulloblastoma cell line Daoy and breast cancer cell line HCC1143 following depletion of FBW7 by RNAi (20 nM) for 48 h. Accumulation of cyclin E is used to assess the efficiency of FBW7 knockdown. GAPDH immunoblot is shown as a loading control.Cycloheximide chase of endogenous SOX9 protein over the course of 6 h in glioma cell line U343MG. The cells were transfected with either non-targeting (siScr) or FBW7-specific siRNA for 72 h prior to experiments. Immunoblots of cyclin E, established SCFFBW7 substrate, indicated the efficacy of siFBW, while GAPDH protein was used as total protein loading control for each sample.Western blotting of endogenous SOX9 protein level upon treatment with 10 µM MG132. HEK293 cells were transfected with 1 µg FLAG-FBW7a for 24 h prior to treatment with the proteasome inhibitor. Whole-cell lysates were collected 4 h following MG132 treatment for gel electrophoresis and immunoblotting. GAPDH protein was immunoblotted to indicate total protein loading for each sample.RNAi depletion of FBW7a decreased endogenous SOX9 ubiquitylation in HEK293. The cells were transfected with either scramble (siScr) or siFBW7a for 72 h prior to assessment of endogenous SOX9 ubiquitylation in the absence or presence of MG132. Endogenous SOX9 protein was immunoprecipitated under denaturing condition (1% SDS) from the whole-cell lysate using the pT236-SOX9 antibody and eluted as described in Materials and Methods. Total SOX9, cyclin E, and GAPDH proteins in the whole-cell lysate were immunoblotted.Reconstitution of ubiquitylation reaction in vitro using bead-immobilized IVT HA-SOX9-WT and recombinant, active human SCFFBW7a. Reaction mixture lacking the UbcH3, SCFFBW7a, and IVT HA-SOX9-WT served as control for the experiments. Ubiquitylation was assessed following elution of IVT HA-SOX9-WT from the bead. Immunoblots of SOX9 and FBW7 proteins present in the eluted fraction are shown.
Fig 2: SOX9 interacts with FBW7a through a conserved degron motif phosphorylated by GSK3 General sequence alignment of SOX9 Cdc4 phosphodegron (CPD) motif against other high-affinity motifs present in previously established FBW7a substrates including cyclin E, cMYC, and c-Jun.Immunoblot analysis of the immunoprecipitated HA-SOX9 or FLAG-FBW7a and their pull-down products. Western blots from whole-cell extract (WCE) of the transfected HEK293 show the levels of exogenous HA-SOX9 or FLAG-FBW7a proteins. Cells were treated with 10 µM MG132 for 4 h prior to harvesting and immunoprecipitation. Blots are representative of three independent experiments.FBW7a in vitro binding assay. FBW7a was eluted from the agarose bead-bound SOX9 peptide (encompassing amino acids 231-245), which had been incubated with the recombinant SCFFBW7a for 1 h at 37°C. The agarose bead-bound peptide contains either the non-phosphorylated SOX9 amino acid sequence (SOX9 peptide) or the threonine phosphorylated amino acids (pSOX9 peptide). The input (10%) show the level of the supplemented recombinant SCFFBW7a in the in vitro binding reaction. Blot is representative of two independent experiments.Representative Western blots from three independent repeats of pT236-SOX9 from lysates of HEK293 transfected with either HA-EV or HA-SOX9. Each lysate was divided and left untreated or subjected to lambda phosphatase (?-PPase) treatment for 1 h at 37°C prior to gel electrophoresis. SOX9 blot show the presence of SOX9 protein in both the untreated and the phosphatase-treated SOX9-transfected cell lysates.Bead-immobilized in vitro-translated (IVT) HA-SOX9 wild-type (WT) or T236/240 mutant were either untreated or subjected to in vitro GSK3 kinase reaction for 90 min at 37°C prior to elution and gel electrophoresis. SOX9 immunoblots are representative of three independent experiments.GSK3-mediated phosphorylation of threonine 236 promoted SOX9 interaction with recombinant SCFFBW7a in vitro. IVT HA-SOX9 immobilized on beads were either untreated or phosphorylated with GSK3 as described in (E) prior to further incubation with recombinant SCFFBW7a for in vitro binding assay. Immunoblot show FBW7a protein in the in vitro binding reaction (input) and FBW7a eluted from the untreated and GSK3-phosphorylated SOX9. Blots are representative of three independent experiments.
Fig 3: FBW7a promotes SOX9 protein turnover in a GSK3-dependent manner through ubiquitin–proteasome system Endogenous SOX9 protein turnover in the presence of cycloheximide following RNAi depletion of FBW7a, FBW7ß, or FBW7?. HEK293 cells were transfected with 20 nM of scrambled siRNA or siRNAs specifically targeting FBW7a, FBW7ß, or FBW7? for 72 h prior to chase with the addition of 100 ng/ml cycloheximide. Immunoblot of cyclin E, an established SCFFBW7 substrate, was used to assess the efficacy of RNAi-mediated depletion of FBW7 protein. GAPDH served as a protein loading control.Immunofluorescence staining of D324MED medulloblastoma cells showing nuclear co-localization of HA-SOX9 WT (Alexa Fluor 568; red) and FLAG-FBW7a (Alexa Fluor 488; green). The cell nuclei were counterstained with DAPI (blue). Images are representative of multiple fields taken at 40× objective magnification. Scale bar indicates 20 µm.Endogenous SOX9 protein levels following transfection (24 h) of 100, 250, 500, and 1,000 ng of plasmid expressing FLAG-FBW7a. Changes in SOX9 protein levels were analyzed relative to GAPDH levels using ImageJ. The blots shown are representative of three independent experiments.Expression of FLAG-FBW7a enhance HA-SOX9-WT protein turnover but not HA-SOX9-T236A or HA-SOX9-T236A/T240A protein turnover over a 4-h cycloheximide (100 ng/ml) chase in HEK293 cells. ß-actin protein served as a loading control. The blots shown are representative of four independent experiments.Treatment of HEK293 cells with increasing concentration of the GSK3a/ß inhibitor BIO, increase HA-SOX9-WT protein level in a dose-dependent manner. HA-SOX9-WT and FLAG-FBW7a were co-expressed in HEK293 cells prior to treatment with different concentrations of BIO for 4 h. Whole-cell lysates were collected for Western blotting with anti-HA (SOX9) and anti-FLAG (FBW7) antibodies. GAPDH served as a protein loading control. The blots shown are representative of two independent experiments.RNAi depletion of GSK3ß attenuate FBW7a-induced HA-SOX9-WT turnover in HEK293 cells. HA-SOX9-WT protein turnover was examined following 48 h depletion of siGSK3b (20 nM) in the presence of cycloheximide (100 ng/ml). Immunoblots with GSK3ß antibodies demonstrated depletion of GSK3ß protein with the siRNA. Changes in SOX9 protein levels were analyzed relative to GAPDH levels using ImageJ. The blots shown are representative of three independent experiments.FLAG-FBW7a-WT expression promote poly-ubiquitylation of endogenous SOX9 in HEK293 cells. Expression of FBW7a lacking the F-box domain (?F), or containing R465A mutation did not induce SOX9 poly-ubiquitylation in vivo. Ubiquitylation assay was performed under denaturing condition to disrupt non-covalently linked ubiquitin as described in the Materials and Methods. Expression of different FLAG-FBW7a constructs and endogenous SOX9 protein were examined in the whole-cell lysates. GAPDH protein was used as a loading control. The blots shown are representative of four independent experiments.Reconstitution of SOX9 poly-ubiquitylation by FLAG-FBW7a in vitro. Immobilized IVT HA-SOX9 WT was incubated with FLAG-FBW7a-WT or the ?F mutant for 60 min at 37°C. Reaction mixture lacking the UbE1, an E1 ubiquitin-activating enzyme, or UbcH3, an E2 ubiquitin-conjugating enzyme, served as a negative control for the in vitro reaction. SOX9 poly-ubiquitylation was assessed following elution of the protein from the immobilized beads under denaturing condition as described in Materials and Methods. The blots shown are representative of three independent experiments.
Fig 4: SOX9 interacts with FBW7a through its conserved degron motif phosphorylated by GSK3 Western blotting of total SOX9 protein levels in the cytoplasmic and nuclear fractions of HCT116-FBW7 WT versus KO cells. The ß-actin was used as a loading control.General evolutionary conservation for SOX9 amino acid sequence surrounding the human CPD motif (highlighted in red) of threonine 236–240 across species.Western blotting of FLAG-FBW7a eluted from the immunoprecipitated HA-SOX9 wild-type (WT) or CPD mutants (-T236/240A and -T240A). The HA-SOX9-WT and the CPD mutant constructs were transiently co-expressed for 24 h with FLAG-FBW7a in HEK293 prior to immunoprecipitation with anti-HA antibody. Equal protein expression of FBW7a across the HEK293 cells transfected with different SOX9 constructs was assessed by immunoblotting of the whole-cell extract.Co-expression of FBW7a with HA-SOX9 WT or various other CPD mutant constructs (-T236/240A, -T236A, or -T240A) in HEK293 cells. Whole-cell lysates were collected 24 h following transfection for Western blotting of the total exogenous and the phosphorylated SOX9 proteins using anti-HA and our pT236-SOX9 antibody, respectively. Immunoblot of GAPDH protein was used to indicate protein loading in each lane.Detection of both exogenous and endogenous phosphorylated SOX9 protein from SOX9 immunoprecipitates. HA-SOX9-transfected or non-transfected HEK293 cells were used as sources for exogenous and endogenous SOX9 protein, respectively. Following SOX9 immunoprecipitation with either anti-HA (for exogenous) or anti-SOX9 (for endogenous) antibody, the resulting immunoprecipitates were divided and either treated with ?-phosphatase or left untreated prior to gel electrophoresis and immunoblotting with pT236-SOX9 antibody. The SOX9 protein blot shows the total protein level present in each sample. Treatment of HEK293 with proteasome inhibitor MG132 (10 µM) increased the level of phosphorylated SOX9.Immunoblots of endogenous pT236 and total SOX9 protein 24 h following transfection of D324MED medulloblastoma cell line with either non-targeting scramble RNA (siScr) or increasing concentrations of siRNA against SOX9. GAPDH protein was used to indicated protein loading for each sampleRepresentative immunofluorescence staining depicting high intensity of pT236-SOX9 (Alexa Fluor 488; green) staining in the nucleus (counterstained with DAPI; blue) in Daoy medulloblastoma cells. Transfection of Daoy cells with 20 nM siSOX9 depleted the nuclear staining of pT236-SOX9. Images were taken using a 40× objective. Scale bar: 20 µm.Bead-immobilized IVT HA-SOX9 WT were subjected to in vitro kinase reaction with 1 unit of recombinant active GSK3a, GSK3ß, or their combination (i.e., 0.5 unit for each isoform) for 90 min at 37°C prior to elution and gel electrophoresis. The SOX9 blot shows total SOX9 protein eluted from the beads from each in vitro kinase reaction.
Fig 5: Schematic model of FBW7a-mediated regulation of SOX9 in medulloblastoma cells SCFFBW 7a together with GSK3 regulate SOX9 protein levels through the ubiquitin–proteasome system. FBW7a mutations, or transcriptional downregulation, lead to stabilization of SOX9 protein and SOX9-driven EMT-like reprogramming, migration and drug resistance in medulloblastoma. Small-molecule inhibitors of the PI3K/mTOR pathway can be used to stimulate GSK3/FBW7a-mediated SOX9 turnover and may provide a novel strategy to target SOX9-driven medulloblastoma.
Supplier Page from Abcam for Recombinant human GSK3 beta protein (Active)