MYOD-SKP2 axis boosts tumorigenesis in fusion negative rhabdomyosarcoma by preventing differentiation through p57Kip2 targeting

Abstract

Rhabdomyosarcomas (RMS) are pediatric mesenchymal-derived malignancies encompassing PAX3/7-FOXO1 Fusion Positive (FP)-RMS, and Fusion Negative (FN)-RMS with frequent RAS pathway mutations. RMS express the master myogenic transcription factor MYOD that, whilst essential for survival, cannot support differentiation. Here we discover SKP2, an oncogenic E3-ubiquitin ligase, as a critical pro-tumorigenic driver in FN-RMS. We show that SKP2 is overexpressed in RMS through the binding of MYOD to an intronic enhancer. SKP2 in FN-RMS promotes cell cycle progression and prevents differentiation by directly targeting p27^Kip1 and p57^Kip2, respectively. SKP2 depletion unlocks a partly MYOD-dependent myogenic transcriptional program and strongly affects stemness and tumorigenic features and prevents in vivo tumor growth. These effects are mirrored by the investigational NEDDylation inhibitor MLN4924. Results demonstrate a crucial crosstalk between transcriptional and post-translational mechanisms through the MYOD-SKP2 axis that contributes to tumorigenesis in FN-RMS. Finally, NEDDylation inhibition is identified as a potential therapeutic vulnerability in FN-RMS.

Fig. 1. SKP2 is highly expressed in RMS.

a Box plot depicting SKP2 expression across skeletal muscle, rhabdomyosarcoma, mixed normal tissues and other human tumors (one-way ANOVA corrected with two-stage step-up method of Benjamini, Krieger and Yekutieli for multiple comparisons). AML Acute Myeloid Leukemia, CLL Chronic Lymphocytic Leukemia. Number of patients from each dataset is reported in Source Data File. Box plots show 25th to 75th quartiles, black bar shows the median, and whiskers go down to the smallest value and up to the largest. b Box plot of SKP2 gene expression in a cohort of Fusion-Positive Rhabdomyosarcoma (FP-RMS) (n = 33) and Fusion-Negative Rhabdomyosarcoma (FN-RMS) (n = 25) patients (GSE66533) compared to skeletal muscle tissues (n = 40, GSE9103) (one-way ANOVA). Box plots show 25th to 75th quartiles, black bar shows the median, and whiskers go down to the smallest value and up to the largest. c Representative SKP2 immunohistochemical staining in RMS primary tumors (FP-RMS n = 6, FN-RMS n = 7) and normal skeletal muscle (n = 2). Scale Bar = 100 μm. d mRNA levels (RT-qPCR) of SKP2 in FP-RMS (red) and FN-RMS (blue) cell lines and in Human Skeletal Muscle Myoblasts (HSMM, gray; normal control cells) were normalized to GAPDH levels and reported as fold increase over HSMM values. n = 3 independent experiments, data presented as mean values ± SD. e Representative western blot (n = 3 independent experiments) of the indicated proteins in HSMM (normal control cells), FP-RMS and FN-RMS cell lines. α-TUBULIN is the loading control. f Representative profile of ChIP-seq read densities of H3K27ac at SKP2 locus on a panel of FP-RMS (red), FN-RMS (blue) primary samples and cell lines and on skeletal muscle tissues (yellow). TSS Transcription start site. Source data are provided as a Source Data file.

Fig. 2. SKP2 is regulated by a MYOD bound enhancer.

a Representative profile of ChIP-seq read densities of MYOD (green), H3K27ac (yellow) and PAX3-FOXO1 (red) at SKP2 locus on RD (FN-RMS) and RH4 (FP-RMS) cells. TSS, Transcription start site. b Pearson’s correlation analysis of MYOD1 versus SKP2 expression in FN-RMS (n = 58) and FP-RMS (n = 54) primary tumors. Line represents linear regression of data. Pearson’s correlation and p values are reported in the figure. P-value was calculated by a two-tailed Pearson correlation test. Error bands represent the 95% confidence interval for the Loess curve fit of the data. c, e Representative western blot (n = 3 independent experiments) of the indicated proteins on RD and JR1 FN-RMS cells transfected with either Scrambled (siSCR) or MYOD siRNA (siMYOD.1) at 24 h (h) and 48 h post-transfection. Vinculin is the loading control. d, f mRNA levels (RT-qPCR) of MYOD1 and SKP2 on cells treated as in (c, e) were normalized to GAPDH levels and expressed as fold increase over siSCR. n = 3 (RD, JR1, RH4) and n = 4 (RH30) independent experiments, data presented as mean values ± SD, Student’s two-tailed t-test. g Workflow schematic model of Chromosome Conformation Capture (3 C) assay. Created with BioRender.com. h (left) Representative electrophoretic run and (right) band intensity quantification of amplification products from 3 C assay on RD (FN-RMS) and RH4 (FP-RMS) cells. P1-P2, primers were designed on SKP2 promoter and intronic enhancer, respectively (see g). As loading control a region lacking restriction sites at SKP2 enhancer has been used. As 3 C positive control we used MYOD looping between SNAI2 distal enhancer and promoter as described in our previous work. A negative control was selected as reported in Supplementary Methods. n = 3 independent experiments, data presented as mean values ± SD, one-way ANOVA. Source data are provided as a Source Data file.

Fig. 3. SKP2 depletion induces p27^Kip1-dependent cell cycle arrest reducing growth.

a Representative western blot on cells transfected with Scrambled (siSCR) or SKP2 siRNA (siSKP2). α-TUBULIN is the loading control. b mRNA levels of SKP2, CDKN1A and CDKN1B on cells treated as in (a) were reported as fold increase over siSCR (1 arbitrary unit, not reported). Data presented as mean values ± SD, Student’s two-tailed t-test. c Representative western blot of co-Immunoprecipitation of endogenous SKP2. (*) indicates IgG. d Representative confocal pictures of Proximal Ligation Assay (PLA) for SKP2 and p27^Kip1. Proximity/interactions are visualized as red dots (upper). DAPI (blue) stained nuclei and WGA (green) membrane glycoproteins. p27^Kip1 (middle) or SKP2 (lower) antibodies alone were PLA negative controls. Scale Bar = 10 μm. e Representative flow cytometry cell cycle diagrams of cells treated as in (a). f Histogram depicts fold changes of the percentage of siSKP2-cells in cell cycle phases over siSCR-cells (1 arbitrary unit, not reported). Data presented as mean values ± SD, Student’s two-tailed t-test. g Representative western blot on cells transfected with SCR, SKP2, p27^Kip1 or SKP2 + p27^Kip1 siRNA 72 h post-transfection. Vinculin is the loading control. h Growth curve analysis of cells treated as in (g). Data presented as mean values ± SD, two-way ANOVA (upper siSKP2+sip27^Kip1 vs siSCR, middle siSKP2+sip27^Kip1 vs siSKP2, lower siSKP2 vs siSCR). i Dose-response curves of cells treated with increasing concentration of SMIP004. Data presented as mean values ± SEM. j Representative western blot on cells treated with SMIP004 20 µM for 48 h. Vinculin is the loading control. k mRNA levels of SKP2, CDKN1A and CDKN1B on cells treated as in (j) were reported as fold increase over DMSO (1 arbitrary unit, not reported). Data presented as mean values ± SD, Student’s two-tailed t-test. l Growth curve analysis of cells treated as in (j). Data presented as mean values ± SD, Student’s two-tailed t-test. m Histogram depicts fold changes of the percentage of SMIP004-treated cells in cell cycle phases over DMSO cells (1 arbitrary unit, not reported). Data presented as mean values ± SD, Student’s two-tailed t-test. All the presented data derive from n = 3 independent experiments. Source data are provided as a Source Data file.

Fig. 4. SKP2 suppression leads to myogenic differentiation.

a Representative western blot on cells transfected with either Scrambled (siSCR) or SKP2 siRNA (siSKP2). Vinculin, GAPDH and β-Actin are the loading controls. b mRNA levels of MYOG, MYOD1 and CDKN1C on cells treated as in (a) were reported as fold increase over siSCR (1 arbitrary unit, not reported). Data presented as mean values ± SD, Student’s two-tailed t-test. c mRNA levels of MyH2 on cells treated as in (a) in growth medium (GM) for 4 days post-transfection were expressed as fold increase over siSCR (1 arbitrary unit). Data presented as mean values ± SD, Student’s two-tailed t-test. d Representative immunofluorescence on cells treated as in (a) for 6 days in GM showing expression of Myosin Heavy Chain (MyHC) (green). DAPI (blue) stained nuclei. Scale Bar = 100 μm. e Histogram depicts the quantification of MyHC-positive cells treated as in (d). Data presented as mean values ± SD, Student’s two-tailed t-test. f Representative western blot on cells infected with either shSCR, shSKP2.1 or shSKP2.2. Vinculin is the loading control. g Representative immunofluorescence on cells treated as in (f) for 6 days, showing expression of MyHC (green). DAPI (blue) stained nuclei. Scale Bar = 100 μm. h Histogram depicts the quantification of MyHC-positive cells treated as in (g). Data presented as mean values ± SD, two-way ANOVA. i Representative western blot on cells treated with SMIP004 20 μM for 168 h. Vinculin is the loading control. j mRNA levels of MYOG, MYOD1 and CDKN1C on cells treated as in (i) were reported as fold increase over DMSO (1 arbitrary unit, not reported). Data presented as mean values ± SD, Student’s two-tailed t-test. k mRNA levels of MyH2 on cells treated as in (i), cultured in GM, were expressed as fold increase over DMSO (1 arbitrary unit). Data presented as mean values ± SD, Student’s two-tailed t-test. l Representative immunofluorescence of cells treated as in (i) showing expression of MyHC (green). DAPI (blue) stained nuclei. Scale Bar = 100 μm. m Histogram depicts the quantification of MyHC-positive cells treated as in (i). Data presented as mean values ± SD, Student’s two-tailed t-test. All presented data derive from n = 3 independent experiments. Source data are provided as a Source Data file.

Fig. 5. SKP2 depletion-induced p57^Kip2 increase is needed for myogenic differentiation.

a Representative western blot (n = 3 independent experiments) of the indicated proteins on RD and JR1 cells treated for 8 h with either vehicle (DMSO), 10.0 or 7.5 μM of MG132. Vinculin is the loading control. b Representative western blot (n = 3 independent experiments) of co-Immunoprecipitation of endogenous SKP2 in RD and JR1 cells showing SKP2 and p57^Kip2. c Representative confocal pictures (n = 3 independent experiments) of Proximal Ligation Assay (PLA) for SKP2 and p57^Kip2 on RD and JR1 cells. SKP2/p57^Kip2 direct close proximity/interactions are visualized as red dots (upper). Nuclei were stained with DAPI (blue) and membrane glycoproteins with WGA (green) (see Methods). PLA negative controls were performed by using either p57^Kip2 (middle) or SKP2 (lower) antibodies alone. Scale Bar = 10 μm. d Representative western blot (n = 3 independent experiments) of the indicated proteins on RD and JR1 cells transfected with either SCR, SKP2, p57^Kip2 or SKP2 + p57^Kip2 siRNA at 72 h post-transfection. Vinculin is the loading control. e Representative immunofluorescence of RD and JR1 cells treated as in (d) and cultured for 6 days in growth medium (GM, supplemented with 10% serum), showing expression of Myosin Heavy Chain (MyHC) (green). Nuclei were stained with DAPI (blue). Scale Bar = 100 μm. f Histogram depicts the quantification of MyHC-positive cells treated as in (e). n = 3 independent experiments, data presented as mean values ± SD, two-way ANOVA. Source data are provided as a Source Data file.

Fig. 6. SKP2 silencing reduces stemness and tumorigenicity in vitro and suppresses growth in vivo.

a Representative western blot (n = 3 independent experiments) of the indicated proteins on RD and JR1 cells infected with either Scrambled (shSCR), SKP2.1 (shSKP2.1) or SKP2.2 (shSKP2.2) lentiviral shRNA at 72 h post-selection. Vinculin is the loading control. b Representative light microscopy pictures of soft agar colony formation assay on RD and JR1 cells treated as in (a) and grown for 2 weeks. Scale bar = 50 μm. c Histogram depicts the quantification of soft agar colony numbers per field. n = 3 (RD) and n = 4 (JR1) independent experiments, data presented as mean values ± SD, one-way ANOVA. d Representative light microscopy pictures of single cell colony formation assay on RD and JR1 cells treated as in (a) and grown for 2 weeks. Scale bar = 50 μm. e Histogram depicts the quantification of colony numbers per field. n = 3 independent experiments, data presented as mean values ± SD, one-way ANOVA. f Representative light microscopy pictures of β-Galactosidase staining of RD and JR1 cells transfected with either SCR or SKP2 siRNA. Scale bar = 100 μm. g Histogram depicts the quantification of the percentage of senescent cells per field. n = 3 independent experiments, data presented as mean values ± SD, Student’s two-tailed t-test. h Images of JR1 shSCR, shSKP2.1, and shSKP2.2 tumors explanted from mice post euthanasia at 48 days post-inoculation. i Tumor volume of shSCR (n = 11), shSKP2.1 (n = 5) and shSKP2.2 (n = 6) JR1 xenografts assessed by caliper measurement represented in mm³ followed for 48 days post-inoculation. Data presented as mean values ± SD, two-way ANOVA. j Tumor weight of shSCR (n = 11), shSKP2.1 (n = 5) and shSKP2.2 (n = 6) JR1 xenografts. Box plots show 25th to 75th quartiles, black bar shows the median, and whiskers go down to the smallest value and up to the largest. One-way ANOVA k Representative images (n = 3 independent experiments) of H&E, p27^Kip1, p57^Kip2, MYOG, MyHC, and Ki67 immunohistochemistry of tumor sections from JR1 xenografts expressing either shSCR or shSKP2.2. Scale Bars = 100 μm. Source data are provided as a Source Data file.

Fig. 7. NEDDylation inhibition prevents SKP2 functions and suppresses tumor growth in vivo.

a Representative western blot (n = 3 independent experiments) of the indicated proteins on RD and JR1 cells treated for 48 h with either vehicle (DMSO) or MLN4924, at the reported Growth Inhibition (GI) concentrations (refer to Fig. S7A for details). Vinculin is the loading control. b Growth curve analysis of RD and JR1 cells treated as in (a). n = 3 independent experiments, data presented as mean values ± SD, Student’s two-tailed t-test. c Representative western blot (n = 3 independent experiments) of the indicated proteins on RD and JR1 cells treated as in (a). Vinculin is the loading control. d Representative light microscopy pictures of β-Galactosidase staining (blue) of RD and JR1 cells treated with vehicle (DMSO) or GI₅₀ MLN4924. Scale bar = 100 μm. e Histogram depicts the quantification of the percentage of senescent cells per field of RD and JR1 cells treated as in (d). n = 3 independent experiments, data presented as mean values ± SD, Student’s two-tailed t-test. f Histogram depicts the quantification of Caspase 3/7 activity at the reported MLN4924 concentration and time points calculated as fold increase over vehicle (DMSO) (n = 3 independent experiments, data presented as mean values ± SD, two-way ANOVA). g Images of RD tumors explanted from mice treated with either vehicle or MLN4924 for 3 weeks. h Tumor volume of RD xenografts treated as in (g) (vehicle n = 5, MLN4924 n = 5) assessed by caliper measurement represented in mm³ followed for 3 weeks treatment. Data presented as mean values ± SD, two-way ANOVA. i Tumor weight of transplanted RD xenografts treated as in (g). Box plots show 25th to 75th quartiles, black bar shows the median, and whiskers go down to the smallest value and up to the largest. One-way ANOVA. j Representative images (n = 3 independent experiments) of H&E, p27^Kip1, p21^Cip1, Cleaved Caspase and Ki67 immunohistochemistry of tumor sections from RD xenografts treated with either vehicle or MLN4924. Scale Bars = 100 μm. Source data are provided as a Source Data file.

Fig. 8. SKP2 effects are mediated by p27^Kip1 and p57^Kip2 degradation.

a A model showing that in myoblasts and RMS cells that are both proliferating, H3K27ac (Ac) enrichment at SKP2 DNA regulatory regions allows MYOD, which is bound on these regions, to induce the expression of SKP2 thus supporting proliferation (left). Upon differentiation cues, reduction of H3K27ac enrichment results in a closed chromatin conformation, impairing MYOD ability to induce SKP2 transcription. These events promote differentiation, which is inhibited in RMS (right). b A model showing that in RMS cells, MYOD induces the expression of SKP2, which in turn associates with the SCF/CRL1 complex to promote ubiquitylation and proteasomal degradation of p27^Kip1 and p57^Kip2 by direct interaction and p21^Cip1 in an indirect way, promoting oncogenic features (left). These SKP2 functions can be suppressed by SKP2 silencing (upper right) or by its functional inhibition by the SKP2 inhibitor SMIP004 or the NEDDylation inhibitor MLN4924 (lower right) to block RMS oncogenic properties. NAE NEDD8-activating enzyme, Ub Ubiquitin, N8 NEDD8. Created with BioRender.com.