
==== Front
Mol Ther
Mol Ther
Molecular Therapy
1525-0016
1525-0024
American Society of Gene & Cell Therapy

S1525-0016(23)00383-0
10.1016/j.ymthe.2023.07.003
Original Article
A dysfunctional miR-1-TRPS1-MYOG axis drives ERMS by suppressing terminal myogenic differentiation
Hüttner Sören S. 1
Henze Henriette 1
Elster Dana 1
Koch Philipp 1
Anderer Ursula 3
von Eyss Björn 1
von Maltzahn Julia Julia.vonmaltzahn@b-tu.de
12∗
1 Leibniz Institute on Aging – Fritz Lipmann Institute, Beutenbergstrasse 11, 07745 Jena, Germany
2 Faculty of Health Sciences Brandenburg, Brandenburg University of Technology Cottbus-Senftenberg, Universitätsplatz 1, 01968 Senftenberg, Germany
3 Department of Cell Biology and Tissue Engineering, Brandenburg University of Technology Cottbus-Senftenberg, Universitätsplatz 1, 01968 Senftenberg, Germany
∗ Corresponding author: Julia von Maltzahn, Leibniz Institute on Aging – Fritz Lipmann Institute, Beutenbergstrasse 11, 07745 Jena, Germany. Julia.vonmaltzahn@b-tu.de
06 9 2023
14 7 2023
31 9 26122632
10 1 2023
10 7 2023
© 2023 The Author(s)
2023
https://creativecommons.org/licenses/by-nc-nd/4.0/ This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Rhabdomyosarcoma is the most common pediatric soft tissue tumor, comprising two major subtypes: the PAX3/7-FOXO1 fusion-negative embryonal and the PAX3/7-FOXO1 fusion-positive alveolar subtype. Here, we demonstrate that the expression levels of the transcriptional repressor TRPS1 are specifically enhanced in the embryonal subtype, resulting in impaired terminal myogenic differentiation and tumor growth. During normal myogenesis, expression levels of TRPS1 have to decrease to allow myogenic progression, as demonstrated by overexpression of TRPS1 in myoblasts impairing myotube formation. Consequentially, myogenic differentiation in embryonal rhabdomyosarcoma in vitro as well as in vivo can be achieved by reducing TRPS1 levels. Furthermore, we show that TRPS1 levels in RD cells, the bona fide model cell line for embryonal rhabdomyosarcoma, are regulated by miR-1 and that TRPS1 and MYOD1 share common genomic binding sites. The myogenin (MYOG) promoter is one of the critical targets of TRPS1 and MYOD1; we demonstrate that TRPS1 restricts MYOG expression and thereby inhibits terminal myogenic differentiation. Therefore, reduction of TRPS1 levels in embryonal rhabdomyosarcoma might be a therapeutic approach to drive embryonal rhabdomyosarcoma cells into myogenic differentiation, thereby generating postmitotic myotubes.

Graphical abstract

Expression levels of the transcriptional repressor TRPS1 are specifically enhanced in embryonal rhabdomyosarcoma, resulting in impaired terminal myogenic differentiation by restricting MYOGENIN expression and tumor growth. However, myogenic differentiation in embryonal rhabdomyosarcoma can be achieved by reducing TRPS1 levels, offering a new therapeutic approach for embryonal rhabdomyosarcoma.

Keywords

rhabdomyosarcoma
TRPS1
myogenic differentiation
RD cells
muscle stem cell
skeletal muscle
ERMS
regeneration
miRNA
==== Body
pmcIntroduction

Rhabdomyosarcoma (RMS) is an aggressive pediatric soft tissue sarcoma of skeletal muscle, accounting for 3% of all childhood cancers and 50% of all pediatric soft tissue sarcomas.1 Interestingly, tumors can arise at every site of the body. RMS mainly originates from deregulated expansion of myogenic progenitor cells, although recently, additional alternative cellular sources from the endothelial or adipogenic lineage have been described.2,3,4 Based on histopathological features, RMS tumors are classified into two major subtypes: embryonal RMS (ERMS) and alveolar RMS (ARMS).5,6 ERMS is the most frequent RMS subtype (incidence rate, 60%–70%) and constitutes a genetically heterogeneous group characterized by mutations in different signaling pathways.7,8,9 In contrast, ARMS is less frequent (approximately 20%–30% of RMS cases) but even more aggressive and homogeneously characterized by the chromosomal translocation t(2;13)/t(1;13), encoding for a novel oncogenic fusion protein, PAX3/7-FOXO1, respectively.7,10,11 Of note, all RMS cells display impaired terminal myogenic differentiation, causing uncontrolled proliferation and tumor growth.12

Terminal myogenic differentiation is required for proper regeneration of skeletal muscle, a process depending on muscle stem cells (MuSCs), a population of tissue-resident stem cells in adult skeletal muscle.13,14,15,16,17 MuSCs remain in a quiescent state, located between the myofiber and the basal lamina, under resting conditions.18,19 They can activate upon respective stimuli (e.g., injury of skeletal muscle) and start to proliferate, differentiate, and fuse to each other or to existing myofibers.20 Myofibers are terminally differentiated, postmitotic, and multinucleated syncytia containing muscle structural proteins, such as myosins and troponins, important for force generation.21 Myogenic differentiation is governed by the well-orchestrated activity and temporal expression of myogenic regulatory factors (MRFs) downstream of Pax7, which comprise Myf5, Myod1, myogenin (Myog), and Mrf4.22 Transient upregulation of Myf5 and Myod1 promotes MuSC activation and myoblast proliferation, followed by upregulation of Myog, promoting myogenic differentiation and fusion of myocytes, whereas Mrf4 has functions in myofiber growth and maintenance.23 The switch from proliferation to differentiation is determined by Myod1-dependent transcription of Myog, which leads to cell cycle exit and expression of genes associated with myocyte fusion and sarcomeric function.24,25,26 This is further illustrated in Myog-deficient mice, which present with severe defects in muscle formation during development and reduced expression of muscle structural genes, such as myosin heavy chain (Mhc), despite normal levels of Myod1 expression.27,28

Importantly, transcriptional deregulation of MRFs has been suggested to be one of the main causes of impaired myogenic differentiation in human RMS.12 However, the underlying mechanisms are largely unknown, although there is emerging evidence that microRNA (miRNA) expression patterns are often deregulated in RMS pathogenesis, causing altered gene expression programs and ultimately leading to inhibition of myogenic differentiation.29,30 For example, while a strong increase in the expression levels of the pro-myogenic miRNAs miR-1/miR-206 and miR-133 is required for myogenic differentiation of myoblasts, the expression of those pro-myogenic miRNAs is strongly reduced in RMS.31,32 Therefore, a detailed molecular understanding of expression and function of the key myogenic factors and their deregulation in RMS (e.g., by transcriptional activators or repressors) needs to be established.

TRPS1 (transcriptional repressor GATA binding 1) is a GATA-like transcription factor with repressor function comprising high conservation among species, including humans and mice.33 TRPS1 contains nine zinc-finger DNA binding domains, of which one shares homology with GATA transcription factors while two zinc fingers are of the IKAROS type and mediate the transcriptional repression function, and six others lack homology to known DNA binding proteins.33 Mutations in the human TRPS1 gene cause tricho-rhino-phalangeal syndrome 1, a disorder mainly affecting development of the skeletal system. Accordingly, TRPS1 has been found to regulate mesenchymal cell fate and chondrogenic differentiation through RUNX2, GLI3, or STAT3.34,35,36,37 However, additional roles of TRPS1 have been identified recently, such as in mammary epithelial cell differentiation, skin wound healing, and, most importantly, its involvement in malignant cell states such as breast cancer, where it regulates estrogen receptor and yes-associated protein (YAP) transcriptional programs.38,39,40,41,42 Moreover, fusions of TRPS1-PLAG1 have been detected in soft tissue myoepithelial tumors, myxoid leiomyosarcomas, and chondroid syringoma, indicating a role of TRPS1 in different cancer types.43,44,45 However, the expression and function of TRPS1 in the context of myogenic differentiation and contribution to RMS development remain unknown so far.

In this study, we show that TRPS1 expression is particularly increased in ERMS. We further demonstrate that reduction of TRPS1 levels is required for myogenic differentiation. However, if TRPS1 levels are maintained at a constantly high level, such as in the human ERMS model cell lines RD, SMS-CTR and JR1, terminal myogenic differentiation is inhibited. We also determined that the persistent TRPS1 expression in RD cells is at least partially mediated through lack of miR-1 expression. Mechanistically, we show that TRPS1 binds to the MYOG promoter in RD cells and that reduction of TRPS1 levels results in enhanced MYOG expression, thereby promoting terminal myogenic differentiation. Ultimately, reduction of TRPS1 levels in RD cells effectively inhibits RD tumor growth in a xenograft mouse model. Thus, we identified a miR-1-TRPS1-MYOG axis, which is dysfunctional in ERMS, thereby contributing to tumor progression.

Results

TRPS1 levels are aberrantly high and sustained in human ERMS

To determine whether TRPS1 is differentially expressed in human RMS tumors compared with healthy skeletal muscle (SkM) control tissue, we performed an immunohistochemistry (IHC) analysis for TRPS1 using an RMS tissue microarray (TMA) (Figure 1A). We found TRPS1 to be frequently expressed in individual cells of the ERMS subtype, while significantly fewer cells of the ARMS subtype and no cells in control tissue displayed TRPS1 expression. Of note, we did not only observe differences in the frequency score between ERMS and ARMS samples but also in the intensity score of tumors displaying positive staining for TRPS1, suggesting that a high proportion of ERMS tumors (22 of 25 analyzed ERMS samples) is characterized by robust TRPS1 expression (Figures 1B and 1C).Figure 1 TRPS1 levels are constantly elevated in human ERMS

(A) Immunohistochemistry (IHC) staining for TRPS1 on human rhabdomyosarcoma (RMS) tissue microarray (TMA). The area within the square marked by the dashed line is magnified in the top right corner inset. Arrowheads indicate TRPS1+ nuclei. Scale bar, 20 μm. SkM, skeletal muscle; ERMS, embryonal RMS; ARMS, alveolar RMS. (B) Quantification of frequency score (1 = 0%–25%, 2 = 25%–50%, 3 = 50%–75%, 4 = 75%–100% positive TRPS1 nuclei) from (A). SkM, n = 8; ERMS, n = 25; ARMS, n = 24; Kruskal-Wallis test; ns, not significant. ∗p < 0.05, ∗∗∗p < 0.001. (C) Quantification of staining intensity score (1 = low, 2 = medium, 3 = high) from (A). ERMS, n = 22; ARMS, n = 16; Mann-Whitney test; ∗p < 0.05. (D) Immunoblot analysis of TRPS1 and GAPDH levels in two independent HSkM myoblast lines (HSkM-1 and HSkM-2), ERMS (RD and SMS-CTR), and ARMS cell lines (RH-30 and RH-41). (E) qRT-PCR analysis of TRPS1 mRNA expression normalized to GAPDH. Data represent the mean + SEM of n = 3, one-way ANOVA and Dunnett’s multiple comparisons tests, ∗p < 0.05. (F) Immunoblot analysis of TRPS1, MHC, and GAPDH levels of HSkM mybolasts cultured in growth medium (GM) or differentiation medium (DM). (G) Densitometric quantification of TRPS1 and (H) MHC normalized to GAPDH from (F). Data represent the mean +SEM of n = 3, one-way ANOVA and Tukey’s multiple comparisons tests, ns = not significant, (∗) p < 0.05, (∗∗) p < 0.01. (I) RT-qPCR analysis of TRPS1 and (J) MYH3 mRNA expression normalized to GAPDH in human skeletal myoblasts (HSkM) cultured in GM or DM. Data represent the mean +SEM of n = 3, one-way ANOVA and Tukey’s multiple comparisons tests, ns = not significant, (∗) p < 0.05, (∗∗∗∗) p < 0.0001. (K) Immunoblot analysis of TRPS1, MHC, and GAPDH levels in human RD cells cultured in GM or DM. (L) Densitometric quantification of TRPS1 and (M) MHC normalized to GAPDH from (K). Data represent the mean +SEM of n = 3, one-way ANOVA and Tukey’s multiple comparisons tests, ns = not significant, (∗) p < 0.05. (N) RT-qPCR analysis of TRPS1 and (O) MYH3 mRNA expression normalized to GAPDH in human RD cells cultured in GM or DM. Data represent the mean + SEM of n = 3, one-way ANOVA and Tukey’s multiple comparisons tests. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗∗p < 0.0001.

To get further insights into the expression and function of TRPS1 in RMS and myogenic differentiation, we examined TRPS1 expression in different human RMS cell lines and proliferating human SkM (HSkM) myoblasts. In contrast to SkM cross-sections, HSkM cells expressed low levels of TRPS1 protein and mRNA, suggesting that TRPS1 is expressed in proliferating myogenic cells. Furthermore, we found very high protein levels and mRNA expression of TRPS1 in ERMS cell lines (RD and SMS-CTR; ∼2-fold enhanced mRNA levels compared with HSkM myoblasts), whereas ARMS cell lines (RH-30 and RH-41) showed very low mRNA expression and no detectable protein levels compared with human myoblasts and ERMS cells (Figures 1D, 1E, and S1A). These findings are supported by reanalysis of publicly available RNA sequencing (RNA-seq), single-cell RNA-seq,46 and proteome RMS tumor expression data,47 indicating that TRPS1 expression is indeed enhanced in ERMS while it is reduced in ARMS (Figures S1C–S1G). In the single-cell RNA-seq data of human RMS tumors, various tumor subpopulations were identified, comprising the full range of embryonic muscle development with varying degrees of stemness and differentiation status.46 By reanalysis of the dataset, we found TRPS1 expression to be enriched in the mesodermal progenitor compartment (Figures S1D–S1F). Interestingly, a higher degree of differentiation correlated with reduced TRPS1 expression (Figure S1E). Next, we asked how TRPS1 expression is modulated during normal myogenic differentiation of human skeletal myoblasts compared with the ERMS model cell lines RD and SMS-CTR. Interestingly, we found that TRPS1 protein and mRNA levels decrease during myogenic differentiation of HSkM myoblasts, coinciding with the expected increase in MHC levels, a marker of terminal myogenic differentiation (Figures 1F–1J). In contrast, RD and SMS-CTR cells did not show a reduction of TRPS1 levels upon serum deprivation used to induce myogenic differentiation; we rather observed a slight increase in TRPS1 mRNA expression and TRPS1 protein levels. Accordingly, MHC levels only increased marginally, corresponding to the inefficient terminal myogenic differentiation in ERMS (Figures 1K–1O and S1G–S1J).

Dynamic Trps1 expression in MuSCs is required for proper myogenic differentiation

To determine whether reduction of TRPS1 levels is a prerequisite for myogenic differentiation, we first investigated the expression of Trps1 during murine muscle development, focusing on the postnatal stages when MuSCs switch from a proliferative to a quiescent state48 (Figure 2A). Interestingly, Trps1 protein was detected in the majority of MuSCs (Pax7+) of the tibialis anterior (TA) muscles at the early postnatal stage (post-natal day 7 (P7)) and the juvenile stage (P21), respectively, while less than half of MuSCs were positive for Trps1 in the adult stage (2 months) (Figures 2B and 2C). Interestingly, expression of Trps1 was mainly found in Ki67+ cells at P7 and P21, while Trps1 protein was identified in Ki67+ and Ki67− MuSCs in the adult stage, a time point when MuSCs become quiescent, suggesting that Trps1 protein is found in proliferating rather than quiescent MuSCs. To investigate this further, we performed cardiotoxin (CTX)-mediated injury of the TA muscle from adult mice, which leads to muscle damage accompanied by activation and proliferation of MuSCs, followed by myogenic differentiation to replace the damaged tissue. Interestingly, we found the highest abundance of Trps1 around 5 days post injury (5 dpi), a time point when MuSCs are highly proliferative49 (Figure S2A), followed by a decline until regeneration was completed at 21 dpi (Figures 2D and S2A). Because Trps1 levels peaked around 5 dpi, we had a closer look at the cell types expressing Trps1 at this specific time point. We found that 95% of Pax7+ MuSCs also contained Trps1 protein, while all Myod1+ cells (100%) representing activated/proliferating MuSCs and myoblasts were positive for Trps1. Remarkably, only a very small proportion of nuclei (10%) of newly formed myofibers, marked by developmental MHC (devMHC), contained Trps1 protein, presumably nuclei derived from myoblasts that just underwent myogenic fusion (Figures 2D–2F). These data demonstrate that Trps1 abundance is reduced during normal myogenic differentiation and that only very few Trps1+ MuSCs were identified in resting SkM, suggesting that Trps1 is found in proliferating but not in quiescent MuSCs. Of note, the data obtained in mice are in line with the results observed in HSkM, suggesting that TRPS1 function is conserved in mice and humans (Figures 1A and S2A).Figure 2 Trps1 levels have to decrease to allow proper myogenic differentiation

(A) Experimental schematic for immunofluorescence (IF) analysis of mouse juvenile tibialis anterior (TA) muscle cross-sections. P, post-natal day. (B) IF images showing Trps1 (red), Pax7 (gray), Ki67 (green), and DAPI (blue) staining of TA cross-sections from (A). White arrows mark Trps1+/Pax7+/Ki67+ nuclei. Scale bar, 10 μm. (C) Quantification of Trps1+/K67+, Trps1+/Ki67−, Trps1−/Ki67+, Trps1−/Ki67− Pax7+ cells from (B). Data represent the mean + SEM of P7 (n = 6), P21 (n = 5), and 2-month-old (n = 4) mice. One-way ANOVA with Tukey’s multiple-comparisons test; ∗∗p < 0.01, ∗∗∗∗p < 0.0001. (D) Experimental schematic for cardiotoxin (CTX)-mediated injury of TA muscle. Hematoxylin and eosin (H&E) and IF staining from the indicated time points in gray are shown in Figure S2A. (E) IF images of TA cross-sections from (D) at 5 days post injury (dpi) showing Trps1 (green), Laminin (gray), and DAPI (blue) together with Pax7 (red), Myod1 (red), or devMHC (red); nuclei are marked by arrowheads. Scale bar, 5 μm. (F) Quantification of Trps1+ among Pax7+, Myod1+, or devMHC+ nuclei at 5 dpi from (E). Data represent the mean + SEM of 4 mice; one-way ANOVA with Tukey’s multiple-comparisons test. ∗∗∗∗p < 0.0001. (G) Experimental schematic for analysis of CTX-induced injury of TA muscles combined with injection of self-delivering siRNAs at 3 and 5 dpi. (H) IF images at 10 dpi showing devMHC (red), Laminin (green), and DAPI (blue) staining of TA cross-sections from (G). Scale bar, 50 μm. (I) Quantification of devMHC+ myofibers within the regenerating area of TA muscle cross-sections from (H). Data represent the mean + SEM of 4 mice, unpaired two-tailed t test; ∗p < 0.05. (J) Experimental schematic for extensor digitorum longus (EDL) single-myofiber isolation and culture combined with siRNA transfection. (K) IF images showing Myog (yellow, white) and DAPI (blue) staining of an EDL myofiber-associated MuSC cluster after siRNA transfection and 72 h of culture; related to (J). Scale bar, 20 μm. (L) Quantification of Myog+ cells per cluster from (K). Data represent the mean + SEM of 4 mice; unpaired two-tailed t test, ∗∗p < 0.01. (M) Immunoblot analysis of TRPS1 and β-actin levels in C2C12 myoblasts stably expressing empty vector (pLeGO) or V5-TRPS1. (N) IF images showing Myog (red, white), Mhc (green, white), and DAPI (blue) staining of empty vector (pLeGO)- or V5-TRPS1-overexpressing C2C12 cells after 3 days in DM. Scale bar, 100 μm. (O) Quantification of the fusion index (ratio of the number of nuclei within Mhc+ cells to all nuclei) from (N). Data represent the mean + SEM of n = 3, unpaired two-tailed t tests, ∗∗∗∗p < 0.0001. (P) Quantification of Myog+ nuclei from (N). Data represent the mean + SEM of n = 3, unpaired two-tailed t tests, ∗∗∗∗p < 0.0001. (Q) qRT-PCR analysis of Myog mRNA expression normalized to B2m in C2C12-pLeGO and C2C12-V5-TRPS1 cells at day 3 of culture in DM. Data represent the mean + SEM of n = 3, unpaired two-tailed t test, ∗p < 0.05.

To examine the relevance of Trps1 for functional regeneration of SkM, we injured the TA muscle of adult mice with CTX and injected a self-delivering small interfering RNA (siRNA) targeting Trps1 into the damaged muscle at the peak of MuSC proliferation, at 3 and 5 dpi, respectively (Figure 2G). Knockdown of Trps1 led to a decrease in the percentage of devMHC+ myofibers at 10 dpi from 53% in muscles injected with si-scr compared with 30% in muscles injected with si-Trps1 (Figures 2H and 2I). In addition, we found a trend toward larger diameters of regenerating myofibers after knockdown of Trps1 and an improved histological outcome, while the number of Myog+ cells was not significantly affected, suggesting enhanced/faster myogenic differentiation and, therefore, regeneration (Figures S2B–S2E). Because injection of a self-delivering siRNA might affect other cell types in addition to MuSCs, we cultured MuSCs on their adjacent myofibers from the extensor digitorum longus (EDL) muscle for 72 h after transfection with a siRNA targeting Trps1 (Figure 2J). Of note, in this established model system, MuSCs can be investigated in their endogenous niche without effects from other mononucleated cells.50,51 After 72 h of culture, clusters of myogenic cells are formed, which arise from a single MuSC. Knockdown of Trps1 resulted in an increased number of cells per cluster (6 si-scr, 8 si-Trps1), suggesting negative regulation of cell proliferation by Trps1 (Figures S2F and S2G). However, reduction of Trps1 levels did not result in enhanced activation of MuSCs per se, as evidenced by a similar number of clusters per myofiber (Figure S2H). Interestingly, we detected changes in the cluster composition after knockdown of Trps1. In particular, we identified an increase in the percentage of Myog+ cells representing a cell population that is further differentiated. This coincided with a reduction in the population of Pax7+/Myod1+ cells, representing committed myoblasts (Figures 2K, 2L, and S2I). This suggests that reduction of Trps1 levels in MuSCs results in an increased ability to proliferate, as evidenced by an enhanced number of cells per cluster arising from a single MuSC after knockdown of Trps1 and an increased ability to differentiate, identified by an increased percentage of Myog+ cells.

Next, we asked whether enhanced TRPS1 levels as observed in ERMS are sufficient to impair myogenic differentiation. Therefore, myogenic differentiation was induced in murine C2C12 myoblasts that stably overexpressed V5-tagged TRPS1. Strikingly, we found severe impairment to fuse and form elongated myotubes in V5-TRPS1-expressing C2C12 cells (Video S1). In particular, we detected a reduction of the fusion index (47% C2C12-pLeGO; 20% C2C12-V5-TRPS1) when cells were allowed to differentiate for 3 days (differentiation medium [DM]), concomitant with a reduction in the percentage of Myog+ cells (50% C2C12-pLeGO, 26% C2C12-V5-TRPS1) (Figures 2M–2P). This was accompanied by a 58% reduction of Myog mRNA levels upon V5-TRPS1 overexpression in C2C12 myoblasts (Figure 2Q). Next, we investigated whether MyoD1, a strong transcriptional inducer of Myog expression, was affected by V5-TRPS1 overexpression in C2C12 cells. Importantly, using immunofluorescence staining to detect MyoD1 and Myog in differentiating C2C12 cells, we found that the number of MyoD1+ cells remained higher in V5-TRPS1 overexpressing cells compared with control C2C12 cells. However, at the same time, the number of cells positive for MyoD1 and Myog was reduced (Figures S2J–S2L), suggesting that V5-TRPS1 overexpression did not affect MyoD1 protein levels but, rather, impaired the transition to MyoD1/Myog double-positive cells.

Video S1. Overexpression of TRPS1 in C2C12 cells

Time lapse microscopy of C2C12 cells expressing pLeGO (control, left video) or pLego-V5-TRPS1 (right video). Induction of differentiation was induced at 0h.

Reduction of TRPS1 levels in ERMS cells allows terminal myogenic differentiation

Because reduction of Trps1 levels in SkM led to increased expression of myogenic differentiation markers and enhanced regeneration, we hypothesized that reduction of TRPS1 levels in RD cells would reinstate the transcriptional regulation of genes important for terminal myogenic differentiation. Therefore, we first performed RNA-seq of RD cells after acute doxycycline-induced, shRNA-mediated reduction of TRPS1 levels using two independent shRNAs targeting TRPS1 (Figures 3A–3G, S3A, and S3B). We identified an intersection of 45 protein-coding genes, which were differentially expressed (log2 fold change [log2FC] > 0.5 or log2FC < −0.5, padj < 0.05) in both RD cell lines expressing a shRNA targeting TRPS1 compared with a non-targeting control shRNA. Of the 45 identified genes, 36 were commonly upregulated, while 7 genes were commonly downregulated (Figure 3C). By gene set enrichment analysis (GSEA), we identified enrichment of the hallmark gene set “myogenesis” as well as the Gene Ontology (GO) gene set “contractile fiber” among the top 5 ranks for both datasets with reduced TRPS1 expression (Figures 3D–3G and S3C). Accordingly, we observed an increase in the mRNAs of genes associated with muscle structure and contraction, such as MHC3 and troponin C1 (TNNC1), supporting our notion that reduction of TRPS1 levels indeed permits terminal myogenic differentiation in ERMS. This was accompanied by a reduction in the levels of mRNAs associated with cell cycle genes, such as cyclin D1 (CCND1) (Figure 3H). Immunofluorescence analysis further demonstrated that RD cells with reduced TRPS1 levels displayed a strong increase in formation of well-aligned, myotube-like, MHC+ structures, coinciding with a reduction in the percentage of proliferating cells (Figures 3I–3K). We next examined whether reduction of TRPS1 levels in other ERMS model cell lines and patient-derived ERMS tumor cells would also result in enhanced myogenic differentiation. Therefore, we first determined TRPS1 mRNA expression and TRPS1 protein levels in two additional ERMS lines, SMS-CTR and JR1, confirming aberrantly high TRPS1 protein and mRNA levels compared with human skeletal myoblasts (Figures S4A and S4B). We then tested whether a single transfection with siRNA targeting TRPS1 would be sufficient to induce myogenic differentiation in SMS-CTR and JR1 cells. Indeed, reduction of TRPS1 expression enhanced myogenic differentiation, as measured by an increase in the percentage of MYOG+ nuclei in both cell lines (Figures S4C–S4F). We further tested whether targeting TRPS1 in patient-derived ERMS tumor cells as well as in another established ERMS line (CT10) would result in induction of myogenic differentiation (Figure S4G–S4N). Indeed, we found that transfection with siRNA targeting TRPS1 yielded in the induction of myogenic differentiation in CT10 cells as well as in patient-derived ERMS cells (Figures S4H–S4J, S4M, and S4N). Of note, we also observed a reduction in the percentage of Ki67+ cells after transfection with a siRNA targeting TRPS1 in patient-derived tumor cells, suggesting reduced proliferation (Figures S4L and S4M). These data suggest that TRPS1 is one of the main causes of impaired terminal myogenic differentiation in ERMS cells and that reduction of TRPS1 levels in ERMS cells permits myogenic differentiation.Figure 3 Transcriptomics analysis reveals TRPS1-suppressed genes to be associated with terminal myogenic differentiation

(A) Experimental schematic for transcriptomics analysis of RD-i-shRNA cells. Cells were incubated with doxycycline (DOX) in GM for 2 days for immunoblot (IB) analysis and subsequently in DM for 10 days, followed by RNA isolation for RNA-seq, qRT-PCR, and IF analysis. (B) IB analysis of TRPS1 and GAPDH levels in RD-i-shRNA cells after 2 days of culture in GM supplemented with 0.1% (v/v) ethanol (EtOH) or 1 μM DOX. (C) Heatmap depicting 45 differentially expressed genes (DEGs) common for both i-shTRPS1 conditions in DM with padj < 0.05, log2FC > 0.5, and log2FC < −0.5. DESeq2-normalized counts were centered and standardized for each gene (Z score). Samples and genes were hierarchically clustered based on pairwise correlations. (D−G) Gene set enrichment analysis (GSEA) comparing the DEGs of i-shTRPS1#1 (D) and i-shTRPS1#2 (E) with i-shNT for the hallmark gene set “HALLMARK_MYOGENESIS” and (F) i-shTRPS1#1 and (G) i-shTRPS1#2 with i-shNT for the GO term “GO_CONTRACTILE_FIBER,” respectively. The top 5 categories are shown in Figure S3C. (H) qRT-PCR analysis of CCND1, MYH3, and TNNC1 mRNA expression normalized to B2M in RD-i-shRNA cells at day 10 of culture in DM from (A). Data represent the mean + SEM of n = 3, two-way ANOVA and Dunnett’s multiple comparisons tests; ∗p < 0.05, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. (I) IF images showing MHC (green, white) and Ki67 (red, white) staining of RD-i-shRNA cells at 10 days of culture in DM with DOX from (A). Scale bar, 50 μm. (J and K) Quantification of nuclei within MHC+ cells (J) and Ki67+ nuclei (K) from (I). Data represent the mean + SEM of n = 3, one-way ANOVA and Tukey’s multiple-comparisons tests; ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Next, we asked whether a temporal reduction of TRPS1 expression is sufficient to allow terminal myogenic differentiation. Therefore, we took advantage of the reversibility of the doxycycline-inducible shRNA system. After an initial differentiation period under doxycycline-induced expression of the shRNAs, we cultured the RD cells for an additional 4 days without doxycycline and assayed terminal myogenic differentiation. Indeed, we found that parameters such as an increase in the percentage of MHC+ cells and a reduction in the percentage of Ki67+ nuclei upon TRPS1 knockdown were stable after doxycycline removal compared with continuous doxycycline treatment (Figures S3E–S3J). This suggests that a reduction of TRPS1 expression during the early phases of terminal myogenic differentiation is sufficient for formation of myotube-like structures and reduced proliferation, making TRPS1 a promising candidate for treatment of ERMS. Interestingly, we did not observe a regain of TRPS1 expression after doxycycline removal, suggesting that most cells underwent terminal myogenic differentiation and therefore do not express TRPS1 anymore.

miR-1 regulates TRPS1 levels in RD cells

The analysis of the transcriptomic datasets indicated an activated miR-1 and MYOD1 signature upon reduced TRPS1 expression, suggesting miR-1 and MYOD1 to be potential upstream regulators of the observed gene expression changes (Figure S3D). In line with this prediction, miR-1 deficiency is known to contribute to development of RMS.32,52 Intriguingly, we identified a potential miR-1 interaction with the 3′ untranslated region (UTR) of the TRPS1 mRNA using in silico prediction tools (Figure S5A). Because miR-1 expression is strongly induced during myogenic differentiation, exerting pro-myogenic activity, and deregulation of miR-1 expression is a common hallmark of RMS cells,32 we investigated whether miR-1 controls TRPS1 levels in RD cells. Therefore, we transfected non-targeting (NT) or miR-1 miRNA mimics into RD cells and assessed TRPS1 protein levels 2 days later (Figure 4A). We detected a strong reduction of TRPS1 protein levels (78%) in cells transfected with the miR-1 mimic compared with cells transfected with the NT miRNA mimic, suggesting that miR-1 indeed controls TRPS1 levels in RD cells (Figures 4B and 4C). Next, we asked whether miR-1 also regulates TRPS1 levels during myogenic differentiation of myoblasts. Therefore, we transfected an antagomiR targeting miR-1 into human myoblasts and monitored TRPS1 levels after induction of differentiation (Figure 4D). Strikingly, we found TRPS1 protein levels to be increased 5-fold when cells were transfected with antagomiR-1 compared with an NT control, followed by differentiation for 5 days, suggesting that miR-1 also controls TRPS1 levels during normal myogenesis (Figures 4E and 4F). In line with this, inhibition of miR-1 during differentiation of human myoblasts resulted in a 38% reduction of MHC levels compared with cells transfected with an NT control. The reduction of MHC levels coincided with the respective changes in morphological appearance for less differentiated cells (Figures 4E, 4G, and S5B). These data indicate that miR-1 regulates TRPS1 levels in ERMS and during normal myogenesis and thereby affects the state of myogenic differentiation. To characterize the miR-1-TRPS1 signaling axis in more detail, we combined replenishment of miR-1 levels with loss of function or gain of function of TRPS1 in RD cells and investigated their ability to undergo myogenic differentiation. However, we did not observe additionally enhanced myogenic differentiation of RD cells based on MYOG and MHC marker proteins when we reduced TRPS1 expression in addition to replenishing miR-1 levels (Figures S5C–S5F). We conclude that miR-1 and TRPS1 act in a directly connected signaling axis, thereby affecting expression of shared target genes; e.g., MYOG and MHC. Because resubstituting miR-1 levels regulates many targets in RD cells,32 we asked whether a reduction of TRPS1 protein levels is required for induction of myogenic differentiation triggered by miR-1. To address this, we transfected an miR-1 mimic or the respective control into RD cells stably overexpressing V5-TRPS1 or an empty vector control (Figures 4H and S5G). Interestingly, overexpression of V5-TRPS1 in RD cells abrogated the miR-1-mediated induction of MYOG (28% RD-pLeGO, 17% RD-V5-TRPS1) (Figures 4I and 4J). Subsequently, we observed a reduction of MHC abundance in V5-TRPS1-overexpressing cells after transfection with the miR-1 mimic compared with control cells (30% RD-pLeGO, 18% RD-V5-TRPS1) (Figures 4K and 4L). Taken together, these results demonstrate a role of TRPS1 as a critical downstream target of miR-1 and that reduction of TRPS1 levels is required for efficient terminal myogenic differentiation.Figure 4 TRPS1 is a critical downstream target of miR-1

(A) Experimental schematic of miRNA mimic transfection into RD cells in GM for IB analysis. (B) IB analysis of TRPS1 and GAPDH levels after miRNA mimic transfection from (A). (C) Densitometric quantification of TRPS1 normalized to GAPDH from (B). Data represent the mean + SEM of n = 3, unpaired two-tailed t test, ∗∗∗∗p < 0.0001. (D) Experimental schematic of antagomiR transfection into human myoblasts followed by IB analysis. (E) IB analysis of TRPS1, MHC, and GAPDH levels in HSkM myoblasts at day 5 of culture in DM after antagomiR transfection from (D). (F and G) Densitometric quantification of TRPS1 (F) and MHC (G) normalized to GAPDH. Data represent the mean + SEM of n = 3, one-way ANOVA and Tukey’s multiple comparisons tests; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. (H) Experimental schematic for miRNA mimic transfection and subsequent analysis by IF analysis in RD-pLeGO and RD-V5-TRPS1 cells; IB analysis is shown in Figure S4G. (I) IF images showing MYOG (red, white) and DAPI (blue) staining of RD-pLeGO and RD-V5-TRPS1 cells after transfection with miRNA mimic for 2 days in GM from (H). Scale bar, 50 μm. (J) Quantification of MYOG+ nuclei from (I). Data represent the mean + SEM of n = 3, two-way ANOVA and Tukey’s multiple comparisons tests; ∗p < 0.05, ∗∗∗p < 0.001. (K) IF images showing MHC (green, white) and DAPI (blue) staining of RD-pLeGO and RD-V5-TRPS1 cells after transfection with miRNA mimic for 2 days in GM and an additional 3 days in DM from (H). Scale bar, 50 μm. (L) Quantification of nuclei within MHC+ cells from (K). Data represent the mean + SEM of n = 3, two-way ANOVA and Tukey’s multiple comparisons tests; ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. (M) Schematic of the 3′ UTR of TRPS1 mRNA showing regions used for the miRNA reporter assay (top panel) and quantification of luciferase activity normalized to β-galactosidase after co-transfection of reporter constructs with miRNA mimics in RD cells for 48 h (bottom panel). Data represent the mean + SEM of n = 3, two-way ANOVA and Sidak’s multiple comparisons tests; ∗p < 0.05, ∗∗p < 0.01. (N) Workflow of metabolic labeling of newly synthesized proteins with L-azidohomoalanine (AHA) and TRPS1 enrichment by immunoprecipitation (IP) (top panel) and detection of metabolically labeled TRPS1 in NT or miR-1 mimic-transfected RD cells after 48 h by IB analysis using streptavidin-HRP, including the whole-cell lysate (WCL) (bottom panel).

Then, we investigated whether miR-1 directly affects TRPS1 abundance at the protein level. Therefore, we used a miRNA reporter assay containing various regions of the TRPS1 mRNA 3′ UTR, which can be tested for sensitivity to regulation by miR-1 (Figure 4M). Transfection of RD cells with the miR-1 mimic led to reduced luciferase activity when the co-transfected reporter plasmids contained either a large fragment of the TRPS1 UTR (TRPS1 3′ UTR2) containing the miR-1 seed region or the isolated miR-1 seed region alone. No changes in luciferase activity were observed after miR-1 mimic co-transfection with reporter constructs containing the TRPS1 3′ UTR1 fragment missing the miR-1 seed region or the mutated seed region alone. Thus, miR-1 directly interacts with the seed region within the human 3′ UTR of TRPS1 mRNA. Because we did not observe changes in TRPS1 mRNA levels after miR-1 mimic transfection (Figure S5H), the ability of miR-1 to regulate TRPS1 protein synthesis was analyzed using a metabolic labeling approach. We detected a reduced amount of nascent TRPS1 protein in RD cells after transfection with the miR-1 mimic, suggesting that miR-1 regulates TRPS1 abundance by translational control (Figure 4N).

In conclusion, we identified an interrelated miR-1-TRPS1 axis in RD cells, with loss of miR-1 expression contributing to increased TRPS1 protein levels in ERMS and, therefore, impaired myogenic differentiation.

TRPS1 binds to the MYOG promoter, impairing induction of MYOG expression in RD cells

We then sought to reveal how TRPS1 mediates suppression of terminal myogenic differentiation. MYOG is a known inducer of myogenic differentiation described to be sufficient to overcome the impaired myogenic differentiation of RMS cells.53,54,55 We identified MYOD1, a well-known transcriptional inducer of MYOG expression during myogenic differentiation, as a potential activated upstream regulator in RD cells when TRPS1 levels were experimentally reduced (Figure S3D). Indeed, we identified a ∼2.5-fold increase in MYOG protein levels upon reduction of TRPS1 expression, but, unexpectedly, we did not observe changes in MYOD1 protein levels (Figures 5A–5E). In line with these data, immunofluorescence experiments demonstrated a similar percentage of MYOD1+ nuclei under all conditions (∼62%) and an ∼2-fold increase of MYOD1+/MYOG+ nuclei in RD cells (21% i-shNT, 49% i-shTRPS1#1, 38% i-shTRPS#2) upon reduction of TRPS1 expression, while almost all cells positive for MYOG were also positive for MYOD1 (∼99%) (Figures 5F–5I). Consistent with this, we found a ∼2.5-fold increase in MYOG mRNA after knockdown of TRPS1 on day 2 and day 3 of differentiation, reflecting the time point of early myogenic differentiation (Figure 5J). However, we did not observe corresponding changes in MYOD1 mRNA levels (Figure 5K). Together, these data suggest that TRPS1 is a negative regulator of MYOG mRNA transcription potentially binding to a regulatory region of the MYOG gene, either together with or independent of MYOD1.Figure 5 TRPS1 represses expression of MYOG through binding to its promoter, thereby impairing binding of MYOD1

(A) Experimental schematic for analysis of early myogenic events in RD-i-shRNA cells. (B) IB analysis of TRPS1, MYOG, MYOD1, and GAPDH levels in RD-i-shRNA cultured for 3 days in DM with DOX from (A). (C–E) Densitometric quantification of TRPS1 (C), MYOG (D), and MYOD1 (E) normalized to GAPDH from (B). Data represent the mean + SEM of n = 3, one-way ANOVA and Tukey’s multiple-comparisons tests; ∗∗p < 0.01, ∗∗∗p < 0.001. (F) IF images showing MYOG (red, white), MYOD1 (green, white), and DAPI (blue) staining of RD-i-shRNA cells after 3 days in DM from (A). Scale bar, 50 μm. (G–I) Quantification of MYOD1+ nuclei (G), MYOD1+/MYOG+ nuclei (H), and the percentage of MYOD1+/MYOG+ nuclei from all MYOG+ nuclei (I) from (F). Data represent the mean + SEM of n = 3, one-way ANOVA and Tukey’s multiple-comparisons tests; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗∗p < 0.0001. (J and K) qRT-PCR analysis of MYOG (J) and MYOD1 (K) mRNA expression normalized to B2M in RD-i-shRNA cells from (A). Data represent the mean + SEM of n = 3, two-way ANOVA with Dunnett’s multiple-comparisons test; comparisons with RD-i-shNT are indicated. ∗∗∗∗p < 0.0001. (L) HOMER known motif analysis of TRPS1 binding sites identified by ChIP-sequencing of RD cells (the top 15 ranks are listed in Figure S5K). (M) Overlap of the genomic binding sites of the transcription factors (TFs) TRPS1 and MYOD1 in RD cells. ChIP-seq data of MYOD1 genomic binding sites were obtained from MacQuarrie et al.56 (N) IF images showing the proximity ligation assay (PLA) signal (red, white) and DAPI (blue) using anti-V5 and anti-MYOD1 antibodies in RD-TRPS1-KO cells stably expressing empty vector (pLeGO) or V5-TRPS1. (O) Quantification of PLA puncta per nucleus from (N). Data represent the mean + SEM of n = 3, two-way ANOVA and Sidak’s multiple-comparisons tests; ∗∗∗∗p < 0.0001. (P) Integrated Genomics Viewer (IGV) showing TRPS1 ChIP-seq reads at the MYOG locus. H3K27ac marks were derived from published ChIP-seq data in RD cells from Gryder et al.57 (Q) ChIP-qPCR using anti-TRPS1 antibody or normal rabbit IgG and primers to amplify either the MYOG locus or U2 control region in RD cells. Data represent the mean + SEM of n = 3, two-way ANOVA and Sidak’s multiple-comparisons tests; ∗∗∗∗p < 0.0001. (R) IGV showing the MYOG locus and a common binding site for TRPS1 and MYOD1 marked in red. (S) ChIP-qPCR using an anti-MYOD1 antibody or normal mouse IgG and primers to amplify the MYOG locus or U2 control region in RD-i-shRNA cells treated with DOX for 2 days in GM and 3 days in DM. Data represent the mean + SEM of n = 3, two-way ANOVA and Sidak’s multiple-comparisons tests, ∗p < 0.05.

To determine the genomic binding sites of TRPS1 in RD cells, we performed chromatin immunoprecipitation sequencing (ChIP-seq) experiments using parental RD cells and RD-TRPS1 knockout (KO) cells as controls, which we generated by CRISPR-Cas9 as described previously.40 Of note, RD-TRPS1-KO cells recapitulated the phenotype we observed after reducing TRPS1 expression using shRNA mediated approaches, displaying increased presence of markers of myogenic differentiation, such as MYOG and MHC (Figures S6A–S6H). Indeed, we identified GATA-related binding motifs within TRPS1 binding sites in RD cells by HOMER known motif analysis, including a previously described TRPS1 binding motif40 (Figure S6I). Among them, a substantial subset of TRPS1 binding sites contained motifs of the MRFs MYOG (26%) and MYOD1 (21%) (Figure 5L). This prompted us to compare the newly identified TRPS1 binding sites with already published MYOD1 binding sites.56 We identified 3,859 genomic binding sites, which are shared between TRPS1 and MYOD1 in RD cells. This accounts for ∼1/4 of all TRPS1 binding sites and ∼1/14 of all MYOD1 binding sites (Figure 5M). Next, we validated the nuclear co-localization of TRPS1 and MYOD1 by proximity ligation assays (PLA) (Figures 5N and 5O) and direct binding of TRPS1 to the MYOG locus by confirmatory ChIP-qPCR experiments (Figures 5P and 5Q). We identified enrichment of endogenous TRPS1 at the MYOG locus compared with the negative control region U2 when using an anti-TRPS1 antibody. Interestingly, TRPS1 and MYOD1 share an overlapping binding site at the MYOG promoter, suggesting that TRPS1 binding might affect the ability of MYOD1 to bind to the MYOG promoter and thereby impair expression of MYOG (Figure 5R). Consequently, we performed ChIP-qPCR of the MYOG promoter in cells with normal or reduced TRPS1 levels using an anti-MYOD1 antibody. Indeed, we determined increased binding of MYOD1 to the MYOG promoter when TRPS1 expression was reduced (Figure 5S), probably allowing increased transcriptional activation of MYOG.

Reduction of TRPS1 levels improves myogenic differentiation of RD cells in vivo

Next, we asked whether reduction of TRPS1 expression in ERMS tumor cells leads to enhanced myogenic differentiation in their endogenous niche in SkM and, thereby, reduced tumor growth in vivo. Therefore, we first injected RD cell lines with normal or reduced TRPS1 expression into TA muscle explants from healthy mice and cultured the injected explant ex vivo for 3 days under differentiation-inducing conditions (Figure 6A). Indeed, when we engrafted RD cells with reduced TRPS1 expression, we observed an ∼3-fold increase in cells positive for devMHC, representing newly formed myofibers (Figures 6B and 6C). Of note, all newly formed myofibers displayed at least the contribution of one RD cell, as indicated by the appearance of human LAMIN A/C, suggesting that reduction of TRPS1 expression in RD cells also increases their myogenic differentiation in the endogenous niche.Figure 6 TRPS1 suppresses myogenic differentiation in ERMS, thereby promoting tumor growth in vivo

(A) Experimental schematic for RD cell differentiation in TA muscle explants. RD cells stably expressing inducible shRNA (i-shRNA) were treated with DOX for 2 days in GM, followed by injection into TA muscle explants, which were cultured for an additional 3 days in DM in the presence of DOX. TA explants were then prepared for IF analysis. (B) IF images showing devMHC (red/white), LAMIN A/C (green), Laminin (gray), and DAPI (blue) staining of TA muscle explant cross-sections from (A). Scale bar, 50 μm. (C) Quantification of devMHC+ expressing cells within LAMIN A/C+ nuclei from (B). Data represent the mean + SEM of n = 3, one-way ANOVA and Tukey’s multiple-comparisons tests; ∗p < 0.05, ∗∗p < 0.01. (D) Measurement of xenograft tumor volume. 107 RD cells stably expressing shRNA were subcutaneously injected into the flanks of NOD.SCID mice. Data represent the mean ± SEM of shRenilla (n = 3), shTRPS1#1 (n = 4), and shTRPS1#2 (n = 4) mice. Two-way ANOVA and Tukey’s multiple-comparisons test, ∗∗∗∗p < 0.0001. (E) Images of RD xenograft tumors after dissection at day 52. (F) Quantification of tumor weight from (E). Data represent the mean + SEM of shRenilla (n = 3), shTRPS1#1 (n = 4), and shTRPS1#2 (n = 4). One-way ANOVA and Tukey’s multiple-comparisons tests; ∗p < 0.05, ∗∗p < 0.01. (G) Images of cross-sections from RD xenograft tumors stained with antibodies directed against Ki67 (green); nuclei are counterstained in blue. Scale bar, 100 μm. (H) Quantification of Ki67+ nuclei from (G). Data represent the mean + SEM of shRenilla (n = 3), shTRPS1#1 (n = 4), and RD-shTRPS1#2 (n = 4). One-way ANOVA and Tukey’s multiple comparisons tests, ∗∗p < 0.01. (I) qRT-PCR analysis of MYOG mRNA expression normalized to B2M in tumor samples from (E). Data represent the mean + SEM of shRenilla (n = 3), shTRPS1#1 (n = 4), and RD-shTRPS1#2 (n = 4); one-way ANOVA and Tukey’s multiple-comparisons tests. (J) qRT-PCR analysis of MYH3 mRNA expression normalized to B2M in tumor samples from (E). Data represent the mean + SEM of shRenilla (n = 3), shTRPS1#1 (n = 4), and RD-shTRPS1#2 (n = 4); one-way ANOVA and Tukey’s multiple-comparisons tests. (K) Images of cross-sections from RD xenograft tumors stained with H&E and IHC for MYOG. Scale bar, 20 μm. Data represent the mean + SEM of shRenilla (n = 3), shTRPS1#1 (n = 4), and RD-shTRPS1#2 (n = 4); one-way ANOVA and Tukey’s multiple-comparisons tests, ∗∗p < 0.01. (L) Quantification of MYOG+ nuclei from (K).

To assess whether reduction of TRPS1 expression in ERMS can decrease the oncogenic potential, we performed rhabdosphere assays and in vivo xenograft transplants of RD cells with reduced TRPS1 expression (Figures S7A–S7D). 3D non-adhesive rhabdosphere growth is an established assay to analyze the self-renewal capacity of ERMS tumor-propagating cells under restrictive conditions.58,59 Interestingly, reduction of TRPS1 expression resulted in a 46% reduction in rhabdosphere formation in vitro, suggesting that the aberrant TRPS1 expression in ERMS might also lead to enhanced tumor growth in vivo (Figures S7E and S7F). Therefore, we injected the respective RD cell lines subcutaneously into non-obese diabetic (NOD)/severe combined immunodeficiency (SCID) mice and investigated tumor growth and composition of the tumor. As expected, RD cells harboring endogenous TRPS1 expression (shRenilla) developed large tumors (879 mm3), while RD cells with reduced TRPS1 expression formed significantly smaller tumors (shTRPS1#1, 276 mm3; shTRPS1#2, 265 mm3). This difference in tumor volume was accompanied by a reduction in tumor weight (shTRPS1#1, 33%; shTRPS1#2, 54%) and percentage of Ki67+ cells (Figures 6D–6H). Of note, the histological analysis of the tumors by H&E staining revealed a higher cytoplasmic proportion (eosinophil) of tumors derived from RD cells with reduced TRPS1 expression (shTRPS1#1 and shTRPS#2) compared with tumors formed from RD cells with endogenous TRPS1 expression, hinting towards enhanced myogenic differentiation upon reduction of TRPS1 expression also in vivo. In line with this conclusion, we observed a 2.8-fold increase in the number of MYOG+ nuclei in tumors derived from RD cells with reduced TRPS1 expression (shTRPS1) compared with RD cells with endogenous TRPS1 expression (shRenilla), supporting our hypothesis that reduction of TRPS1 expression also improves myogenic differentiation in ERMS in vivo (Figures 6K and 6L). This is supported by enhanced expression of MYOG and MYH3 in tumors derived from RD cells with reduced TRPS1 expression (Figures 6I and 6J). Additionally, no differences in the amount of cleaved PARP were detected in tumor lysates by immunoblot analysis, indicating that apoptosis is not the main cause of reduced tumor growth following reduction of TRPS1 expression (Figure S7G).

In conclusion, our data suggest that the increased TRPS1 levels in ERMS cells cause enhanced tumor growth by repressing terminal myogenic differentiation. Furthermore, we show that TRPS1 mRNA is a direct miR-1 target in RD cells and that miR-1 deficiency contributes to the enhanced TRPS1 levels observed in ERMS. Moreover, TRPS1 was found to directly bind to the MYOG promoter, thereby affecting binding of MYOD1 and repressing MYOG transcription.

Discussion

Deregulation of myogenic differentiation is one of the contributing factors to the uncontrolled proliferation of RMS tumor cells. However, the underlying molecular mechanisms are unknown or highly diverse, not at least because of various oncogenic mechanisms within the specific RMS tumor subtypes. Here, we show that levels of TRPS1, a transcriptional repressor, are specifically enhanced in the human ERMS model cell lines RD, SMS-CTR, CT10, and JR1. We demonstrate that TRPS1 prevents terminal myogenic differentiation through repression of MYOG, thereby promoting tumor growth in vivo. Furthermore, we identified TRPS1 mRNA as a direct downstream target of miR-1, whose expression is lost in RD cells, resulting in aberrantly high TRPS1 protein abundance. Our data also suggest a role of Trps1 in normal myogenesis as a negative regulator of myogenic differentiation of MuSCs in physiological contexts, such as juvenile muscle growth and muscle regeneration after acute injury in mice.

We identified increased TRPS1 expression and TRPS1 protein levels specifically in human ERMS tumors and RMS cell lines (Figures 1A–1E). Interestingly, high TRPS1 expression was found in a subset of cells in human ERMS tumors (Figures 1A–1C and S1A), while TRPS1 levels were homogenously high in cells isolated from ERMS tumors; e.g., RD cells or patient-derived ERMS cells (Figures S1B and S4G). Because TRPS1-expressing ERMS cells are unable to undergo myogenic differentiation, therapeutic approaches should focus especially on this subpopulation as a supportive therapy in addition to routine chemotherapy for ERMS treatment, similar to approaches targeting EGFR in ERMS aiming to prevent disease recurrence.46 The RMS subtype-specific expression of TRPS1 might be due to the presence of the oncogenic PAX3/7-FOXO1 fusion gene in ARMS cells, a strong inhibitor of myogenic differentiation itself, rendering the cells independent of TRPS1 expression.4 Of note, PAX3-FOXO1 does not directly control TRPS1 expression because overexpression of the fusion gene in mouse NIH 3T3 cells or human RD cells did not alter TRPS1 protein levels.60,61,62 However, frequent chromosomal aberrations, including amplification of chromosome 8 containing the TRPS1 gene, have been described for PAX3/7-FOXO1-negative RMSs.63 Another plausible cause of the enhanced expression of TRPS1 specifically in ERMS is the heterogeneous source of cells of origin for the ERMS and ARMS subtypes. While ERMS tumors probably arise at an earlier stage of the myogenic lineage, ARMS tumors arise at later stages,47,64 which could reflect differences in endogenous TRPS1 expression before oncogenic transformation. This hypothesis is in line with the highly dynamic regulation of Trps1 protein levels we identified in the differentiation process of myogenic progenitor cells in mice. While we detected Trps1 protein in Pax7+ myogenic progenitor cells in TA muscles of juvenile mice, representing proliferating cells at this developmental stage,48 no Trps1 protein was detected in adult murine SkM or healthy HSkM tissue, a state when MuSCs are quiescent19 (Figures 1A, 2A–2C, and S2A).

We identified a decrease in TRPS1 expression during myogenic differentiation in mice or humans, indicating that the required reduction of TRPS1 expression is a conserved mechanism to allow myogenic differentiation (Figures 1F–1J and 2D–2F). To functionally address the role of Trps1 in early myogenic differentiation, we either reduced Trps1 expression in single myofiber-associated MuSCs or forced expression in myoblasts. Of note, reduction of Trps1 expression in myofiber-associated MuSCs increased the number of cells per cluster and changed the cluster composition towards further differentiated cells accompanied by increased levels of MYOG, a transcription factor well known to promote myogenic differentiation65 (Figures 2J–2K and S2C–S2F). Forced expression of TRPS1 in mouse myoblasts severely impaired terminal myogenic differentiation resembling the impaired differentiation in RD cells, probably by competing with Myod1 for binding at the myogenin locus (Figures 2M–2Q and 2G–2I). These data suggest that regulation of Trps1 levels is required for proper control of myogenic lineage progression. Indeed, reduction of Trps1 expression in injured TA muscles causes accelerated regeneration of SkM, as observed here (Figures 2G–2H and S2B). However, it needs to be determined whether the enhanced early regeneration upon reduction of Trps1 expression comes at the expense of stem cell exhaustion and consecutive impairments of regeneration in later stages.66,67

Based on the finding that Trps1 modulates the differentiation dynamics during normal myogenic processes, we focused on the aberrant TRPS1 expression in human ERMS. We discovered that increased TRPS1 levels in RD cells are causally linked to the lack of miR-1 expression, suggesting that miR-1 regulates TRPS1 expression. Deregulation of miRNA expression, including miR-1/miR-206 and miR-133, which have essential functions during myogenesis,31 has been reported as a general feature of RMSs.29,30,68,69 Essentially, restoring miR-1/miR-206 function strongly enhanced myogenic differentiation and reduced RMS tumor growth in vivo.70,71 Here, we provide experimental evidence that TRPS1 mRNA is a direct target of miR-1, which controls TRPS1 protein levels through post-transcriptional regulation. Reporter assays demonstrated a functional interaction of miR-1 with the 3′ UTR mRNA of TRPS1, and metabolic labeling of newly synthesized proteins revealed reduced levels of nascent TRPS1 protein after miR-1 mimic transfection in RD cells. Interestingly, a previous study reported induction of expression of MYOG and the terminal differentiation marker MyL1 upon expression of miR-1 in RD cells.32 Importantly, overexpression of V5-TRPS1 reduced the ability of miR-1 to induce expression of MYOG and subsequent upregulation of MHC in RD cells, which identifies an interrelated miR-1-TRPS1-MYOG axis (Figures 4 and S5).

Importantly, in ERMS (e.g., in RD cells), TRPS1 levels are constantly elevated, thereby perturbing the myogenic differentiation program. Consistent with this, we could demonstrate that reduction of TRPS1 expression promotes upregulation of MYOG levels and consequentially increased terminal myogenic differentiation in RD cells, ultimately reducing xenograft tumor growth in vivo. This is in line with reports describing upregulation of MYOG to coincide with upregulation of the cell cycle inhibitors p21 and p57 in myogenic progenitor cells.72,73,74 Interestingly, MYOG protein levels are considered to be rather low in ERMS and high in ARMS,75,76 suggesting a potential link to the increased levels of TRPS1 in ERMS tumors. Importantly, increased expression of MYOG has been demonstrated to induce terminal myogenic differentiation and reduce ERMS tumor growth.53,54,55

This let us speculate that ERMS tumor cells do not entirely lose their intrinsic potential for myogenic differentiation but rather display impaired transcriptional regulation, as also postulated by others.12 For instance, MacQuarrie et al.56 compared the genome-wide binding sites of MYOD1 in RD and muscle cells. They identified reduced MYOD1 binding in RD cells at specific sites associated with motifs of transcriptional regulators, which were expressed at lower levels in ERMS; e.g., RUNX1, MEF2C, JDP2, and NFIC.56 However, overexpression of these transcriptional regulators enhanced differentiation of RD cells.56 Similarly, suppressed expression of CASZ1 in ERMS cells has been shown recently to impair activation of transcriptional differentiation networks driven by MYOD1, including expression of MYOG. Also, here, re-expression of CASZ1 was sufficient to enhance myogenic differentiation of ERMS cells.54 In the present study, we used genome-wide approaches to identify TRPS1 targets and DNA binding sites in RD cells, revealing that subsets of regions bound by TRPS1 also contain binding motifs of MRFs, such as MYOD1 and MYOG (Figure 5L). Specifically, TRPS1 and MYOD1 share common binding sites, and both proteins are located in close proximity in the nucleus (Figures 5M–5O). Furthermore, TRPS1 and MYOD1 bind to the MYOG locus in RD cells, suggesting that binding of TRPS1 impairs transcription of MYOG by competing with MYOD1 for binding at the MYOG locus. Upon reduction of TRPS1 levels, we observed enhanced binding of MYOD1 to the MYOG locus, which at least partially contributes to the increased MYOG expression (Figures 5P–5S). Given the abundance of MYOD1 and MYOG binding motifs within the TRPS1 binding sites, it is likely that TRPS1 co-regulates multiple target genes beyond MYOG that are required for terminal myogenic differentiation. However, the precise mode of transcriptional repression of MYOG by TRPS1 needs to be established. A recent study uncovered a role of TRPS1 in controlling long-range enhancer-promoter chromatin interactions to regulate YAP transcriptional activity in breast cancer cells, which was associated with TRPS1 binding at H3K27ac-rich regions.40 In line with this, TRPS1 binding at the MYOG locus in RD cells was found at a local depression of a larger H3K27ac stretch, which was identified recently as a super-enhancer region critical for myogenic function.54 However, whether the negative regulation of myogenic differentiation by TRPS1 depends on negative regulation of YAP transcriptional activity in the context of RMS needs to be investigated.40

In conclusion, we identified TRPS1 as a player in myogenic differentiation, which was not known so far. We demonstrate that TRPS1 prevents terminal myogenic differentiation of RD cells. Lack of miR-1 expression contributes to elevated levels of TRPS1, which binds to the MYOG locus and impairs induction of MYOG expression by reducing binding of MYOD1 to the MYOG promotor. These findings may help us to explore new therapeutic options targeting the miR-1-TRPS1-MYOG axis to induce differentiation of ERMS and improve treatment outcomes and quality of life for patients suffering from ERMS.

Materials and methods

Experimental model and subject details

All experiments involving live animals in this study were approved by the “Thüringer Landesamt für Verbraucherschutz,” Bad Langensalza (FLI-17-005, FLI-17-015, FLI-18-001, and 03-048-16).

Mice

Adult C57BL/6J mice were either bred in house or purchased from Janvier. NOD.CB17-Prkdc scid/NCrCrl, hereafter referred to as NOD.SCID mice, were purchased from Janvier. Upon change of housing location, mice were allowed to acclimatize for 1 week before any procedures were performed. All mice were housed at the animal facility of the Leibniz Institute on Aging – Fritz Lipmann Institute under specific pathogen-free (SPF) conditions. The mice were maintained in groups of 2–5 mice in individually ventilated cages (IVCs) with nesting material, enrichment, and 12-h day-night cycles at room temperature and received standard chow and water ad libitum according to directive 2010/63 EU and GV SOLAS. Mice were checked daily.

Cells

Human ERMS RD cells and human ARMS RH-30 and RH-41 cells were purchased from the ATCC (CCL-136) and DSMZ (ACC 489 and 592), respectively. SMS-CTR, CT10, and JR1 cells were generously provided by Peter Houghton and Janet Shipley. RMS cells were cultured in growth medium (DMEM [Thermo Fisher Scientific, 61965026] or RPMI 1640 [Thermo Fisher Scientific, 11875093] supplemented with 10% fetal bovine serum [FBS; Thermo Fisher Scientific, 10270-106] and 1% penicillin-streptomycin [Thermo Fisher Scientific, 15140122]) or DM (DMEM [Thermo Fisher Scientific, 61965026] supplemented with 2% HS [Thermo Fisher Scientific, 26050-088] and 1% penicillin-streptomycin [Thermo Fisher Scientific, 15140122]). Patient-derived ERMS tumor cells (A52/94) were obtained from a 14-year-old female diagnosed with ERMS and cultured in RPMI 1640 (Thermo Fisher Scientific, 11875093) supplemented with 10% FBS (Thermo Fisher Scientific, 10270-106) as part of the project “Einrichtung einer Tumorzellbank für maligne Kindertumore” in 1994 at the Charite (Berlin, Germany).

Human Lenti-X(293T) cells were purchased from Takara (632180) and cultured in DMEM (Thermo Fisher Scientific, 61965026) supplemented with 10% FBS (Thermo Fisher Scientific, 10270-106) and 1% penicillin-streptomycin (Thermo Fisher Scientific, 15140122).

Mouse C2C12 myoblasts were purchased from the ATCC (CRL-1772) and cultured in growth medium (DMEM [Sigma-Aldrich, D6046] supplemented with 10% FBS [Thermo Fisher Scientific, 10270-106] and 1% penicillin-streptomycin [Thermo Fisher Scientific, 15140122]) or DM (DMEM [Sigma-Aldrich, D5796] supplemented with 2% HS [Thermo Fisher Scientific, 26050-088] and 1% penicillin-streptomycin [Thermo Fisher Scientific, 15140122]).

HSkM myoblasts were purchased from Thermo Fisher Scientific (A12555) and cultured on collagen (Corning, 354236)-coated cell culture dishes in growth medium (Ham’s F-10 [Thermo Fisher Scientific, 31550] supplemented with 20% FBS [Thermo Fisher Scientific, 10270-106], 2.5 ng bFGF [Thermo Fisher Scientific, 13256029], and 1% penicillin-streptomycin (Thermo Fisher Scientific, 15140122]) or DM (DMEM [Thermo Fisher Scientific, 61965026] supplemented with 5% HS [Thermo Fisher Scientific, 26050-088] and 1% penicillin-streptomycin [Thermo Fisher Scientific, 15140122]).

All cells were cultured in a humidified incubator at 37°C and 5% CO2.

Generation of TRPS1 KO RD cells

CRISPR-mediated TRPS1 KO in RD cells was performed as described recently.40 The TRPS1 KO was validated by the absence of the TRPS1 protein using two different antibodies that recognize TRPS1 (Abcam, ab209664) and specifically its N terminus (Bethyl Laboratories, A303-563A). Genomic DNA was isolated, and the region flanking the sgRNA targeting site was amplified by PCR (Fw, 5′-AAGGCTGGCTTCAATTATGAAA-3′; Rev, 5′-TGTTTCCTCTGTAGCCTTTGGT-3′). To further verify the KO, the amplicon was cloned into the pJet sequencing vector (Thermo Fisher Scientific, K1232) and subjected to Sanger sequencing, revealing frameshifts leading to premature stop codons.

Lentivirus infection

To generate cell lines stably expressing a transgene, cells were infected with a lentivirus produced in LentiX cells and selected by antibiotics resistance as described previously.40 Plasmids used for virus production are listed in Table S6.

Transient transfection and treatment of cells

siRNA, miRNA mimics, and miRNA inhibitors (Tables S4 and S5) were transfected using the RNAiMax reagent (Thermo Fisher Scientific, 13778150) following the manufacturer’s instructions.

Acute TRPS1 knockdown in RD-i-shRNA cells was induced by supplementing the growth medium with 1 μg/mL doxycycline (Sigma-Aldrich, D3447).

Analysis of TMAs and xenograft tumor sections by IHC

RMS TMAs were purchased from US Biomax (SO2082a). IHC was performed after deparaffinization of specimens, antigen retrieval using Tris-EDTA + 0.05% Tween 20 (pH 9), incubation with blocking reagent, primary antibody (Table S1) incubation overnight at 4°C, and secondary antibody and DAB substrate incubation according to the manufacturer’s instructions (Vector Laboratories, Sk-4103). Hematoxylin was used as a nuclear counterstain. The staining was analyzed by light microscopy using an Axio.D1 Observer microscope (Carl Zeiss).

RNA isolation and qRT-PCR

Total RNA was isolated from cells by chloroform-phenol extraction using the TRI reagent (Sigma-Aldrich, 93289) according to the manufacturer’s instructions. After reverse transcription (Bio-Rad, 170-8891), specific cDNAs were quantified by qRT-PCR using the SYBR Green Kit (Bio-Rad, 170-8862) and respective primers (Table S2) on a Stratagene 3000 cycler (Agilent Technologies) with the MX Pro software. For gene expression analysis, at least 3 independent biological replicates (cDNAs) were measured in technical duplicates or triplicates. The obtained Ct values of target genes and housekeeping genes were used to calculate relative expression according to the 2−ΔΔCt formula.77

RNA-seq and expression analysis

RNA was isolated from RD-i-shRNA cells, which were treated with doxycycline (1 μg/mL) in growth medium for 2 days and DM for 10 days. Sequencing of RNA samples was performed using Illumina’s next-generation sequencing methodology.78 In detail, total RNA was quantified and quality checked using the 2100 Bioanalyzer instrument in combination with the RNA 6000 Nano Kit (both Agilent Technologies). Libraries were prepared from 180 ng of input material (total RNA) using the NEBNext Ultra II Directional RNA Library Preparation Kit in combination with the NEBNext Poly(A) mRNA Magnetic Isolation Module and NEBNext Multiplex Oligos for Illumina (Index Primer Set 1/2/3/4) following the manufacturer’s instructions (New England Biolabs). Quantification and quality checking of libraries was done using the 2100 Bioanalyzer Instrument and DNA 7500 Kit (Agilent Technologies). Libraries were pooled and sequenced in eight lanes of the HiSeq 2500 system run in 51 cycle/single-end/high-output mode. Sequence information was converted to FASTQ format using bcl2fastq v.2.20.0.422. The reads were aligned with STAR 2.7.2b (parameters: “--alignIntronMax 100000 --outSJfilterReads Unique --outSAMmultNmax 1 --outFilterMismatchNoverLmax 0.04”)79 to the Homo sapiens reference genome (GRCh38 with Ensembl gene annotation release 99). For each annotated gene, reads that mapped uniquely to one genomic position were counted with FeatureCounts 1.6.5 (multi-mapping or multi-overlapping reads were discarded, and stranded mode was set to “-s 2”).80 Quality assessment of the reads and the described analysis steps was done with MultiQC.81 Differentially expressed genes (DEGs) were determined with R 3.6.1 using the package DESeq2 1.26.0.82 Only genes that had at least 1 read count in any of the analyzed samples of a particular comparison were subjected to DESeq2. The sample group “i-shTRPS1#1” or “i-shTRPS1#2” was compared with the group “i-shNT” in a pairwise fashion. For each gene in each comparison, the p value was calculated using the Wald significance test. Resulting p values were adjusted for multiple testing using the Benjamini and Hochberg correction. The log2FC values were shrunk with the DESeq2 function lfcShrink (type = ”normal”) to control for variance of log2FC estimates for genes with low read counts. Genes with an adjusted p < 0.05 were considered differentially expressed. The final DEG set was formed as the intersection of DEGs from either pairwise analysis, keeping only genes with the biotype “protein coding,” and was analyzed using the iDEP web tool (http://ge-lab.org/idep/), Ingenuity Pathway Analysis (IPA; QIAGEN), and the GSEA83,84,85 software using default parameters.

ChIP-seq and ChIP-qPCR

ChIP-seq was performed as described previously.40 Briefly, RD and RD-TRPS1-KO cells were fixed in 1% formaldehyde for 10 min. Nuclei were released using hypotonic buffer with protease inhibitors and lysed in ChIP lysis buffer. The chromatin was sonified in 1.5 mL thin-walled reaction tubes (Diagenode) using a Bioruptor (Diagenode) with 60 cycles (30 s ON, 30 s OFF), yielding chromatin fragment sizes of approximately 200 bp. For immunoprecipitation, 500 μg chromatin was incubated with 10 μg of a non-commercial anti-TRPS1 antibody, raised in rabbit and validated previously,40 or control immunoglobulin G (IgG) coupled to magnetic protein A dynabeads, for 6 h at 4°C. The respective antibodies are listed in Table S1. DNA libraries for sequencing were prepared according to the manufacturer’s instructions using the NEBNext Ultra II DNA Library Prep Kit for Illumina (NEB, E7645L) and Dual Index Primers (NEBNext Multiplex Oligos for Illumina, NEB, E7600S). Libraries were purified using 0.9-fold magnetic beads and finally resuspended in EB buffer. Quantification and quality checks of libraries were done using the 2100 Bioanalyzer instrument and a DNA 7500 Kit (Agilent Technologies). Libraries were pooled and sequenced in one lane of the HiSeq 2500 system run in 51 cycle/single-end/high-output mode. Sequence information was converted to FASTQ format using bcl2fastq v.2.20.0.422. Processing of raw data and bioinformatic analysis were conducted as described previously.40 H3K27ac and MYOD1 ChIP-seq data in RD cells were retrieved from Gryder et al.57 (GSE84628) and MacQuarrie et al.56 (GSE50413), respectively.

ChIP-qPCR was performed using the SimpleChIP Plus Sonication Chromatin IP Kit (Cell Signaling Technology, 56383) according to the manufacturer’s instructions. Briefly, cells were cross-linked as described above. Nuclei were extracted and lysed using the kit’s components. For chromatin immunoprecipitation, 2 μg of antibody or the respective IgG control were incubated together with 10 μg cross-linked and sheared chromatin overnight at 4°C under constant rotation. Subsequently, magnetic beads were added to the samples, followed by incubation at 4°C under constant rotation for 2 h. Immunoprecipitates were then washed and eluted. DNA was purified using spin columns and quantified by qPCR assay. ChIP-qPCR primers are listed in Table S2.

PLA

The PLA (Sigma-Aldrich, DUO92008) was carried out following the manufacturer’s protocol. RD-TRPS1-KO cells expressing V5-TRPS1 were cultured in chamber slides and fixed in 4% paraformaldehyde (PFA) for 10 min at room temperature. Specific anti-V5 antibodies and anti-MYOD1 antibodies were incubated overnight at 4°C. Immunofluorescent puncta per nuclei were analyzed with an Axio.D1 Observer microscope (Carl Zeiss). Antibodies are listed in Table S1.

Protein isolation and immunoblotting

Proteins were obtained by cell lysis using RIPA buffer containing PhosSTOP (Roche, 4906845001) and Protease STOP (Roche, 04693132001) inhibitors. Cleared protein lysates were separated on Bis-Tris gels by SDS-PAGE and transferred to polyvinylidene fluoride (PVDF) membranes. Membranes were incubated in 5% skim milk in TBS-T as blocking reagent and incubated with primary and secondary antibodies diluted in blocking reagent. Protein bands were visualized on a LAS camera system (Thermo Fisher Scientific) using the enhanced chemiluminescence (ECL) detection kit (Thermo Fisher Scientific, 32209). For densitometric quantification of protein bands, FIJI software86 was used. Target protein expression was normalized to GAPDH from at least 3 independent biological replicates (3 protein lysates). Antibodies are listed in Table S1.

Immunofluorescence

Immunofluorescence (IF) staining of muscle cross-sections has been described previously in detail.87 Of note, to detect devMHC by IF, the samples were not PFA fixed. Images were captured using an Axio Observer microscope (Carl Zeiss). Cells were fixed in 4% PFA for 10 min at room temperature directly in the well and processed for IF analysis following the principal steps of this protocol. Antibodies are listed in Table S1.

Mouse CTX injury and siRNA injection

To induce muscle damage and regeneration, TA muscles of 2- to 4-month-old C57BL/6J mice were injected with 50 μL 20 μM CTX (Latoxan, L8102), followed by preparation of muscle cross-sections and IF analysis as described elsewhere.87 Self-delivering siRNAs (Table S4), either scrambled control or targeting Trps1, were injected into the CTX-damaged muscles on days 3 and 5 after injury to induce knockdown in vivo. TA muscles were analyzed at 10 dpi by IF microscopy.

EDL single-myofiber culture and siRNA transfection

Single EDL myofibers were isolated from 2- to 4-month-old wild-type C57BL/6J mice. Myofiber-associated MuSCs were siRNA transfected and cultured for 72 h under floating conditions in a humidified incubator at 37°C and 5% CO2. Myogenic cells were analyzed by IF microscopy using a Carl Zeiss Observer microscope. Experimental details have been reported previously.50

TA muscle explants

The lower hindlimbs of 7-month-old NOD.SCID mice were isolated from the knee to the ankle, and all tissue but the TA muscle attached to the bone were removed. Beforehand, RD-i-shRNA cells were treated with 1 μg/mL doxycycline in growth medium for 48 h. 750,000 RD-i-shRNA cells in 50 μL DM containing doxycycline were then injected into the isolated TA explants and cultured in a 12-well plate with doxycycline-containing DM for 72 h. Afterward, TA explants were processed for cross-sectioning and IF staining.

Xenografts

107 RD cells stably expressing control shRNA (shRenilla) or one of two different shRNAs targeting TRPS1 (shTRPS1#1 or shTRPS1#2) were injected in 250 μL Matrigel (Sigma-Aldrich, E6909)/medium (1:1) subcutaneously into the flanks of 9-week-old NOD.SCID mice. Tumor growth was continuously measured using a caliper. After dissection, tumors were weighed, parts of the tumors were PFA fixed and paraffin embedded, and 5-μm sections were prepared using a microtome (Leica). Deparaffinization and standard H&E histological analysis were carried out using an automated slide stainer (Leica). Tumor sections were subjected to IHC staining as described above using universal heat-induced antigen retrieval (HIER) reagent (Abcam, ab208572) and anti-MYOG antibody (Table S1), nuclei were counterstained using DAPI. Parts of the unfixed tumors were used for protein lysates using RIPA buffer for immunoblotting.

Rhabdosphere growth

RD-shRenilla and RD-shTRPS1#1 cells were grown in serum-free medium (Neurobasal [Gibco, 21103049] + 20 ng/mL bFGF [R&D Systems, 233-FB-025], 10 ng/mL epidermal growth factor [EGF; R&D Systems, 236-EG-200], and 2× B27 [Invitrogen, 17504044]) on ultra-low attachment plates (Corning) following the protocol by Walter et al.59 After 10 passages, spheres were dissociated using Accutase (Sigma-Aldrich, A6964-100ML), and 20,000 cells were seeded into 24 well plates and cultured for 3 days, and the spheres per well were counted.

miRNA reporter assay

Luciferase vectors for miRNA reporter assays were constructed with the pMIR-REPORT miRNA Expression Reporter Vector System (Thermo Fisher Scientific, AM5795) following the manufacturer’s instructions. The TRPS1 3′ UTR fragments UTR1 and UTR2 were PCR amplified. miR-1 seed regions within the TRPS1 3′ UTR, either wild type or mutated, were generated by DNA oligonucleotide annealing. Inserts were ligated to the SpeI- and MluI-digested pMIR-REPORT luciferase vector. Oligonucleotide sequences are listed in Table S6.

150,000 RD cells were seeded in 24-well plates filled with 0.5 mL growth medium, and incubated overnight at 37°C. The cells were then transfected with 0.5 μg of the reporter vectors (pMIR-REPORT and pbeta-GAL) together with the miRNA mimics (1 nM final) using Lipofectamine 2000 (Thermo Fisher Scientific, 11668019) according to the manufacturer’s instructions.

12 h post transfection, cells were washed once in PBS and then lysed in reporter lysis buffer according to the manufacturer’s instructions (Promega, E4030). 20 μL lysate was added to a white 96-well plate with a flat closed bottom, and firefly luciferase activity was measured at room temperature using firefly luciferase substrate solution in a plate-reading luminometer (Mithras) with injector function. The measurement was set up as follows: injection of 100 μL luciferase substrate solution, 2-s shaking/delay time, 10-s measurement. For measurement of β-galactosidase activity, 30 μL of lysate, 20 μL 1× reporter lysis buffer, and 50 μL 2× assay buffer (Promega, E2000) were mixed in a 96-well plate and incubated at 37°C for 30 min. The absorbance was measured at 405 nm and used for normalization of luciferase activity.

Metabolic labeling

RD cells were transfected with miR-NT or miR-1 miRNA mimic. 43 h post transfection, the cells were starved for 1 h with methionine-free medium (Gibco, 21013-024). The medium was then replaced with methionine-free medium supplemented with 200 μM AHA (L-azidohomoalanine) (Thermo Fisher Scientific, C10102), an amino acid analog, and cells were incubated for 4 h at 37°C. Whole-cell protein lysates were prepared, and immunoprecipitation (IP) was performed using 2 μg anti-TRPS1 antibody (Abcam, ab209664) for 200 μg total protein. A Click-IT reaction was performed to biotinylate the incorporated AHA, following the manufacturer's instructions (Invitrogen, C33372), and the beads were then washed and eluted. The IP eluates as well as input protein samples were subjected to immunoblotting. For detection of biotinylated TRPS1 protein in the IP samples, a streptavidin-horseradish peroxidase (HRP) conjugate was used.

Statistical analysis

Mouse and cell line experiments were performed at least in biological triplicates, and numbers are indicated in the figure legends. The results are shown as the mean ± SEM unless indicated otherwise in the figure legends. Statistical significance was calculated using the GraphPad Prism software using the statistical test indicated in the figure legends, with ∗p ≤ 0.05, ∗∗p ≤ 0.01, ∗∗∗p ≤ 0.001, ∗∗∗∗p ≤ 0.0001, and ns (not significant). RNA-seq approaches were conducted in biological triplicates per experimental group, and statistical analysis is described in the respective section. ChIP-seq was performed from parental RD cells (wild type [WT]) and RD-TRPS1-KO cells using the same anti-TRPS1 antibody; RD cells were used as input.

Supplemental information

Document S1. Figures S1–S7 and Tables S1–S6

Document S2. Article plus supplemental information

Data availability

RNA-seq and ChIP-seq data generated and discussed in this study have been deposited in NCBI’s Gene Expression Omnibus88 and are accessible through GEO Series accession number GSE197552 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE197552).

Acknowledgments

The authors gratefully acknowledge support from the FLI Core Facilities Imaging, FLI Mouse, FLI Core Services Histology, as well as SPARK FLI. They especially thank Marco Groth, Ivonne Görlich, and Karol Szafranski from the core facilities CF DNA Sequencing and CF Life Science Computing for excellent support. The authors would like to thank Christina Picker and Christine Poser for excellent technical assistance. The authors thank Janet Shipley and Peter Houghton for providing JR1 and SMS-CTR cells. The authors thank Said Hashemolhosseini for providing protein lysates and cDNA of RH-41 cells. This work was supported by grants from the 10.13039/501100001659 Deutsche Forschungsgemeinschaft (MA-3975/2-1), the 10.13039/501100005972 Deutsche Krebshilfe (DKH-JvM-861005), the Wilhelm-Sander Foundation, and SPARK FLI (to J.v.M.). The FLI is a member of the Leibniz Association and is financially supported by the Federal Government of Germany and the State of Thuringia.

Author contributions

S.S.H. and J.v.M. designed and performed most experiments, analyzed data, and interpreted results. H.H. and D.E. performed experiments and analyzed data. U.A. isolated and provided patient-derived ERMS cells. B.v.E. and P.K. analyzed bioinformatic data. S.S.H. and J.v.M. wrote the manuscript.

Declaration of interests

The authors declare no competing interests.

Supplemental information can be found online at https://doi.org/10.1016/j.ymthe.2023.07.003.
==== Refs
References

1 Amer K.M. Thomson J.E. Congiusta D. Dobitsch A. Chaudhry A. Li M. Chaudhry A. Bozzo A. Siracuse B. Aytekin M.N. Epidemiology, Incidence, and Survival of Rhabdomyosarcoma Subtypes: SEER and ICES Database Analysis J. Orthop. Res. 37 2019 2226 2230 10.1002/jor.24387 31161653
2 Drummond C.J. Hanna J.A. Garcia M.R. Devine D.J. Heyrana A.J. Finkelstein D. Rehg J.E. Hatley M.E. Hedgehog Pathway Drives Fusion-Negative Rhabdomyosarcoma Initiated From Non-myogenic Endothelial Progenitors Cancer Cell 33 2018 108 124.e5 10.1016/j.ccell.2017.12.001 29316425
3 Hatley M.E. Tang W. Garcia M.R. Finkelstein D. Millay D.P. Liu N. Graff J. Galindo R.L. Olson E.N. A mouse model of rhabdomyosarcoma originating from the adipocyte lineage Cancer Cell 22 2012 536 546 10.1016/j.ccr.2012.09.004 23079662
4 Skapek S.X. Ferrari A. Gupta A.A. Lupo P.J. Butler E. Shipley J. Barr F.G. Hawkins D.S. Rhabdomyosarcoma Nat. Rev. Dis. Primers 5 2019 1 10.1038/s41572-018-0051-2 30617281
5 Patton R.B. Horn R.C. Jr. Rhabdomyosarcoma: clinical and pathological features and comparison with human fetal and embryonal skeletal muscle Surgery 52 1962 572 584 14484432
6 Enterline H.T. Horn R.C. Jr. Alveolar rhabdomyosarcoma; a distinctive tumor type Am. J. Clin. Pathol. 29 1958 356 366 10.1093/ajcp/29.4.356 13533329
7 Ognjanovic S. Linabery A.M. Charbonneau B. Ross J.A. Trends in childhood rhabdomyosarcoma incidence and survival in the United States, 1975-2005 Cancer 115 2009 4218 4226 10.1002/cncr.24465 19536876
8 Shern J.F. Chen L. Chmielecki J. Wei J.S. Patidar R. Rosenberg M. Ambrogio L. Auclair D. Wang J. Song Y.K. Comprehensive genomic analysis of rhabdomyosarcoma reveals a landscape of alterations affecting a common genetic axis in fusion-positive and fusion-negative tumors Cancer Discov. 4 2014 216 231 10.1158/2159-8290.CD-13-0639 24436047
9 Seki M. Nishimura R. Yoshida K. Shimamura T. Shiraishi Y. Sato Y. Kato M. Chiba K. Tanaka H. Hoshino N. Integrated genetic and epigenetic analysis defines novel molecular subgroups in rhabdomyosarcoma Nat. Commun. 6 2015 7557 10.1038/ncomms8557 26138366
10 Barr F.G. Galili N. Holick J. Biegel J.A. Rovera G. Emanuel B.S. Rearrangement of the PAX3 paired box gene in the paediatric solid tumour alveolar rhabdomyosarcoma Nat. Genet. 3 1993 113 117 10.1038/ng0293-113 8098985
11 Davis R.J. D'Cruz C.M. Lovell M.A. Biegel J.A. Barr F.G. Fusion of PAX7 to FKHR by the variant t(1;13)(p36;q14) translocation in alveolar rhabdomyosarcoma Cancer Res. 54 1994 2869 2872 8187070
12 Keller C. Guttridge D.C. Mechanisms of impaired differentiation in rhabdomyosarcoma FEBS J. 280 2013 4323 4334 10.1111/febs.12421 23822136
13 de Morree A. Klein J.D.D. Gan Q. Farup J. Urtasun A. Kanugovi A. Bilen B. van Velthoven C.T.J. Quarta M. Rando T.A. Alternative polyadenylation of Pax3 controls muscle stem cell fate and muscle function Science 366 2019 734 738 10.1126/science.aax1694 31699935
14 Pawlikowski B. Pulliam C. Betta N.D. Kardon G. Olwin B.B. Pervasive satellite cell contribution to uninjured adult muscle fibers Skelet. Muscle 5 2015 42 10.1186/s13395-015-0067-1 26668715
15 Sambasivan R. Yao R. Kissenpfennig A. Van Wittenberghe L. Paldi A. Gayraud-Morel B. Guenou H. Malissen B. Tajbakhsh S. Galy A. Pax7-expressing satellite cells are indispensable for adult skeletal muscle regeneration Development 138 2011 3647 3656 10.1242/dev.067587 21828093
16 Lepper C. Partridge T.A. Fan C.M. An absolute requirement for Pax7-positive satellite cells in acute injury-induced skeletal muscle regeneration Development 138 2011 3639 3646 10.1242/dev.067595 21828092
17 Murphy M.M. Lawson J.A. Mathew S.J. Hutcheson D.A. Kardon G. Satellite cells, connective tissue fibroblasts and their interactions are crucial for muscle regeneration Development 138 2011 3625 3637 10.1242/dev.064162 21828091
18 Mauro A. Satellite cell of skeletal muscle fibers J. Biophys. Biochem. Cytol. 9 1961 493 495 13768451
19 Ancel S. Stuelsatz P. Feige J.N. Muscle Stem Cell Quiescence: Controlling Stemness by Staying Asleep Trends Cell Biol. 31 2021 556 568 10.1016/j.tcb.2021.02.006 33674167
20 Schmidt M. Schüler S.C. Hüttner S.S. von Eyss B. von Maltzahn J. Adult stem cells at work: regenerating skeletal muscle Cell. Mol. Life Sci. 76 2019 2559 2570 10.1007/s00018-019-03093-6 30976839
21 Sweeney H.L. Hammers D.W. Muscle Contraction Cold Spring Harb. Perspect. Biol. 10 2018 a023200 10.1101/cshperspect.a023200
22 Hernández-Hernández J.M. García-González E.G. Brun C.E. Rudnicki M.A. The myogenic regulatory factors, determinants of muscle development, cell identity and regeneration Semin. Cell Dev. Biol. 72 2017 10 18 10.1016/j.semcdb.2017.11.010 29127045
23 Bentzinger C.F. Wang Y.X. Rudnicki M.A. Building muscle: molecular regulation of myogenesis Cold Spring Harb. Perspect. Biol. 4 2012 a008342 10.1101/cshperspect.a008342
24 Deato M.D.E. Marr M.T. Sottero T. Inouye C. Hu P. Tjian R. MyoD targets TAF3/TRF3 to activate myogenin transcription Mol. Cell 32 2008 96 105 10.1016/j.molcel.2008.09.009 18851836
25 Liu Q.C. Zha X.H. Faralli H. Yin H. Louis-Jeune C. Perdiguero E. Pranckeviciene E. Muñoz-Cànoves P. Rudnicki M.A. Brand M. Comparative expression profiling identifies differential roles for Myogenin and p38alpha MAPK signaling in myogenesis J. Mol. Cell Biol. 4 2012 386 397 10.1093/jmcb/mjs045 22847234
26 Ganassi M. Badodi S. Ortuste Quiroga H.P. Zammit P.S. Hinits Y. Hughes S.M. Myogenin promotes myocyte fusion to balance fibre number and size Nat. Commun. 9 2018 4232 10.1038/s41467-018-06583-6 30315160
27 Nabeshima Y. Hanaoka K. Hayasaka M. Esumi E. Li S. Nonaka I. Nabeshima Y. Myogenin gene disruption results in perinatal lethality because of severe muscle defect Nature 364 1993 532 535 10.1038/364532a0 8393146
28 Hasty P. Bradley A. Morris J.H. Edmondson D.G. Venuti J.M. Olson E.N. Klein W.H. Muscle deficiency and neonatal death in mice with a targeted mutation in the myogenin gene Nature 364 1993 501 506 10.1038/364501a0 8393145
29 Bersani F. Lingua M.F. Morena D. Foglizzo V. Miretti S. Lanzetti L. Carrà G. Morotti A. Ala U. Provero P. Deep Sequencing Reveals a Novel miR-22 Regulatory Network with Therapeutic Potential in Rhabdomyosarcoma Cancer Res. 76 2016 6095 6106 10.1158/0008-5472.CAN-16-0709 27569217
30 Megiorni F. Cialfi S. McDowell H.P. Felsani A. Camero S. Guffanti A. Pizer B. Clerico A. De Grazia A. Pizzuti A. Deep Sequencing the microRNA profile in rhabdomyosarcoma reveals down-regulation of miR-378 family members BMC Cancer 14 2014 880 10.1186/1471-2407-14-880 25427715
31 Chen J.F. Mandel E.M. Thomson J.M. Wu Q. Callis T.E. Hammond S.M. Conlon F.L. Wang D.Z. The role of microRNA-1 and microRNA-133 in skeletal muscle proliferation and differentiation Nat. Genet. 38 2006 228 233 10.1038/ng1725 16380711
32 Rao P.K. Missiaglia E. Shields L. Hyde G. Yuan B. Shepherd C.J. Shipley J. Lodish H.F. Distinct roles for miR-1 and miR-133a in the proliferation and differentiation of rhabdomyosarcoma cells FASEB J. 24 2010 3427 3437 10.1096/fj.09-150698 20466878
33 Malik T.H. Shoichet S.A. Latham P. Kroll T.G. Peters L.L. Shivdasani R.A. Transcriptional repression and developmental functions of the atypical vertebrate GATA protein TRPS1 EMBO J. 20 2001 1715 1725 10.1093/emboj/20.7.1715 11285235
34 Zhang Y. Xie R.L. Gordon J. LeBlanc K. Stein J.L. Lian J.B. van Wijnen A.J. Stein G.S. Control of mesenchymal lineage progression by microRNAs targeting skeletal gene regulators Trps1 and Runx2 J. Biol. Chem. 287 2012 21926 21935 10.1074/jbc.M112.340398 22544738
35 Wuelling M. Kaiser F.J. Buelens L.A. Braunholz D. Shivdasani R.A. Depping R. Vortkamp A. Trps1, a regulator of chondrocyte proliferation and differentiation, interacts with the activator form of Gli3 Dev. Biol. 328 2009 40 53 10.1016/j.ydbio.2009.01.012 19389374
36 Napierala D. Sam K. Morello R. Zheng Q. Munivez E. Shivdasani R.A. Lee B. Uncoupling of chondrocyte differentiation and perichondrial mineralization underlies the skeletal dysplasia in tricho-rhino-phalangeal syndrome Hum. Mol. Genet. 17 2008 2244 2254 10.1093/hmg/ddn125 18424451
37 Suemoto H. Muragaki Y. Nishioka K. Sato M. Ooshima A. Itoh S. Hatamura I. Ozaki M. Braun A. Gustafsson E. Fässler R. Trps1 regulates proliferation and apoptosis of chondrocytes through Stat3 signaling Dev. Biol. 312 2007 572 581 10.1016/j.ydbio.2007.10.001 17997399
38 Cornelissen L.M. Drenth A.P. van der Burg E. de Bruijn R. Pritchard C.E.J. Huijbers I.J. Zwart W. Jonkers J. TRPS1 acts as a context-dependent regulator of mammary epithelial cell growth/differentiation and breast cancer development Genes Dev. 34 2020 179 193 10.1101/gad.331371.119 31879358
39 Serandour A.A. Mohammed H. Miremadi A. Mulder K.W. Carroll J.S. TRPS1 regulates oestrogen receptor binding and histone acetylation at enhancers Oncogene 37 2018 5281 5291 10.1038/s41388-018-0312-2 29895970
40 Elster D. Tollot M. Schlegelmilch K. Ori A. Rosenwald A. Sahai E. von Eyss B. TRPS1 shapes YAP/TEAD-dependent transcription in breast cancer cells Nat. Commun. 9 2018 3115 10.1038/s41467-018-05370-7 30082728
41 Yang J. Liu X. Huang Y. He L. Zhang W. Ren J. Wang Y. Wu J. Wu X. Shan L. TRPS1 drives heterochromatic origin refiring and cancer genome evolution Cell Rep. 34 2021 108814 10.1016/j.celrep.2021.108814
42 Mascharak S. Talbott H.E. Januszyk M. Griffin M. Chen K. Davitt M.F. Demeter J. Henn D. Bonham C.A. Foster D.S. Multi-omic analysis reveals divergent molecular events in scarring and regenerative wound healing Cell Stem Cell 29 2022 315 327.e6 10.1016/j.stem.2021.12.011 35077667
43 Panagopoulos I. Gorunova L. Andersen K. Lund-Iversen M. Lobmaier I. Micci F. Heim S. NDRG1-PLAG1 and TRPS1-PLAG1 Fusion Genes in Chondroid Syringoma Cancer Genomics Proteomics 17 2020 237 248 10.21873/cgp.20184 32345665
44 Zhu G. Benayed R. Ho C. Mullaney K. Sukhadia P. Rios K. Berry R. Rubin B.P. Nafa K. Wang L. Diagnosis of known sarcoma fusions and novel fusion partners by targeted RNA sequencing with identification of a recurrent ACTB-FOSB fusion in pseudomyogenic hemangioendothelioma Mod. Pathol. 32 2019 609 620 10.1038/s41379-018-0175-7 30459475
45 Arias-Stella J.A. 3rd Benayed R. Oliva E. Young R.H. Hoang L.N. Lee C.H. Jungbluth A.A. Frosina D. Soslow R.A. Antonescu C.R. Novel PLAG1 Gene Rearrangement Distinguishes a Subset of Uterine Myxoid Leiomyosarcoma From Other Uterine Myxoid Mesenchymal Tumors Am. J. Surg. Pathol. 43 2019 382 388 10.1097/PAS.0000000000001196 30489320
46 Patel A.G. Chen X. Huang X. Clay M.R. Komorova N. Krasin M.J. Pappo A. Tillman H. Orr B.A. McEvoy J. The myogenesis program drives clonal selection and drug resistance in rhabdomyosarcoma Dev. Cell 57 2022 1226 1240.e8 10.1016/j.devcel.2022.04.003 35483358
47 Stewart E. McEvoy J. Wang H. Chen X. Honnell V. Ocarz M. Gordon B. Dapper J. Blankenship K. Yang Y. St. Jude Children's Research Hospital – Washington University Pediatric Cancer Genome Project Identification of Therapeutic Targets in Rhabdomyosarcoma through Integrated Genomic, Epigenomic, and Proteomic Analyses Cancer Cell 34 2018 411 426.e19 10.1016/j.ccell.2018.07.012 30146332
48 Bachman J.F. Chakkalakal J.V. Insights into muscle stem cell dynamics during postnatal development FEBS J. 289 2022 2710 2722 10.1111/febs.15856 33811430
49 Bentzinger C.F. Wang Y.X. Dumont N.A. Rudnicki M.A. Cellular dynamics in the muscle satellite cell niche EMBO Rep. 14 2013 1062 1072 10.1038/embor.2013.182 24232182
50 Huttner S.S. Hayn C. Ahrens H.E. Schmidt M. Henze H. von Maltzahn J. Single Myofiber Culture Assay for the Assessment of Adult Muscle Stem Cell Functionality Ex Vivo J. Vis. Exp. 2021 10.3791/62257
51 Pasut A. Jones A.E. Rudnicki M.A. Isolation and culture of individual myofibers and their satellite cells from adult skeletal muscle J. Vis. Exp. 2013 e50074 10.3791/50074
52 Li L. Sarver A.L. Alamgir S. Subramanian S. Downregulation of microRNAs miR-1, -206 and -29 stabilizes PAX3 and CCND2 expression in rhabdomyosarcoma Lab. Invest. 92 2012 571 583 10.1038/labinvest.2012.10 22330340
53 Nanni P. Nicoletti G. Palladini A. Astolfi A. Rinella P. Croci S. Landuzzi L. Monduzzi G. Stivani V. Antognoli A. Opposing control of rhabdomyosarcoma growth and differentiation by myogenin and interleukin 4 Mol. Cancer Ther. 8 2009 754 761 10.1158/1535-7163.MCT-08-0678 19372547
54 Liu Z. Zhang X. Lei H. Lam N. Carter S. Yockey O. Xu M. Mendoza A. Hernandez E.R. Wei J.S. CASZ1 induces skeletal muscle and rhabdomyosarcoma differentiation through a feed-forward loop with MYOD and MYOG Nat. Commun. 11 2020 911 10.1038/s41467-020-14684-4 32060262
55 Pomella S. Sreenivas P. Gryder B.E. Wang L. Milewski D. Cassandri M. Baxi K. Hensch N.R. Carcarino E. Song Y. Interaction between SNAI2 and MYOD enhances oncogenesis and suppresses differentiation in Fusion Negative Rhabdomyosarcoma Nat. Commun. 12 2021 192 10.1038/s41467-020-20386-8 33420019
56 MacQuarrie K.L. Yao Z. Fong A.P. Diede S.J. Rudzinski E.R. Hawkins D.S. Tapscott S.J. Comparison of genome-wide binding of MyoD in normal human myogenic cells and rhabdomyosarcomas identifies regional and local suppression of promyogenic transcription factors Mol. Cell. Biol. 33 2013 773 784 10.1128/MCB.00916-12 23230269
57 Gryder B.E. Yohe M.E. Chou H.C. Zhang X. Marques J. Wachtel M. Schaefer B. Sen N. Song Y. Gualtieri A. PAX3-FOXO1 Establishes Myogenic Super Enhancers and Confers BET Bromodomain Vulnerability Cancer Discov. 7 2017 884 899 10.1158/2159-8290.CD-16-1297 28446439
58 Hayes M.N. McCarthy K. Jin A. Oliveira M.L. Iyer S. Garcia S.P. Sindiri S. Gryder B. Motala Z. Nielsen G.P. Vangl2/RhoA Signaling Pathway Regulates Stem Cell Self-Renewal Programs and Growth in Rhabdomyosarcoma Cell Stem Cell 22 2018 414 427.e6 10.1016/j.stem.2018.02.002 29499154
59 Walter D. Satheesha S. Albrecht P. Bornhauser B.C. D'Alessandro V. Oesch S.M. Rehrauer H. Leuschner I. Koscielniak E. Gengler C. CWS Study Group CD133 positive embryonal rhabdomyosarcoma stem-like cell population is enriched in rhabdospheres PLoS One 6 2011 e19506 10.1371/journal.pone.0019506
60 Khan J. Bittner M.L. Saal L.H. Teichmann U. Azorsa D.O. Gooden G.C. Pavan W.J. Trent J.M. Meltzer P.S. cDNA microarrays detect activation of a myogenic transcription program by the PAX3-FKHR fusion oncogene Proc. Natl. Acad. Sci. USA 96 1999 13264 13269 10.1073/pnas.96.23.13264 10557309
61 Begum S. Emami N. Cheung A. Wilkins O. Der S. Hamel P.A. Cell-type-specific regulation of distinct sets of gene targets by Pax3 and Pax3/FKHR Oncogene 24 2005 1860 1872 10.1038/sj.onc.1208315 15688035
62 Barber T.D. Barber M.C. Tomescu O. Barr F.G. Ruben S. Friedman T.B. Identification of target genes regulated by PAX3 and PAX3-FKHR in embryogenesis and alveolar rhabdomyosarcoma Genomics 79 2002 278 284 10.1006/geno.2002.6703 11863357
63 Dehner C.A. Armstrong A.E. Yohe M. Shern J.F. Hirbe A.C. Genetic Characterization, Current Model Systems and Prognostic Stratification in PAX Fusion-Negative vs. PAX Fusion-Positive Rhabdomyosarcoma Genes (Basel) 12 2021 1500 10.3390/genes12101500
64 Abraham J. Nuñez-Álvarez Y. Hettmer S. Carrió E. Chen H.I.H. Nishijo K. Huang E.T. Prajapati S.I. Walker R.L. Davis S. Lineage of origin in rhabdomyosarcoma informs pharmacological response Genes Dev. 28 2014 1578 1591 10.1101/gad.238733.114 25030697
65 Cao Y. Kumar R.M. Penn B.H. Berkes C.A. Kooperberg C. Boyer L.A. Young R.A. Tapscott S.J. Global and gene-specific analyses show distinct roles for Myod and Myog at a common set of promoters EMBO J. 25 2006 502 511 10.1038/sj.emboj.7600958 16437161
66 Paris N.D. Soroka A. Klose A. Liu W. Chakkalakal J.V. Smad4 restricts differentiation to promote expansion of satellite cell derived progenitors during skeletal muscle regeneration Elife 5 2016 e19484 10.7554/eLife.19484
67 Su Y. Yu Y. Liu C. Zhang Y. Liu C. Ge M. Li L. Lan M. Wang T. Li M. Fate decision of satellite cell differentiation and self-renewal by miR-31-IL34 axis Cell Death Differ. 27 2020 949 965 10.1038/s41418-019-0390-x 31332295
68 Gasparini P. Ferrari A. Casanova M. Limido F. Massimino M. Sozzi G. Fortunato O. MiRNAs as Players in Rhabdomyosarcoma Development Int. J. Mol. Sci. 20 2019 5818 10.3390/ijms20225818
69 Missiaglia E. Shepherd C.J. Aladowicz E. Olmos D. Selfe J. Pierron G. Delattre O. Walters Z. Shipley J. MicroRNA and gene co-expression networks characterize biological and clinical behavior of rhabdomyosarcomas Cancer Lett. 385 2017 251 260 10.1016/j.canlet.2016.10.011 27984116
70 Taulli R. Bersani F. Foglizzo V. Linari A. Vigna E. Ladanyi M. Tuschl T. Ponzetto C. The muscle-specific microRNA miR-206 blocks human rhabdomyosarcoma growth in xenotransplanted mice by promoting myogenic differentiation J. Clin. Invest. 119 2009 2366 2378 10.1172/JCI38075 19620785
71 Yan D. Dong X.D.E. Chen X. Wang L. Lu C. Wang J. Qu J. Tu L. MicroRNA-1/206 targets c-Met and inhibits rhabdomyosarcoma development J. Biol. Chem. 284 2009 29596 29604 10.1074/jbc.M109.020511 19710019
72 Zhang P. Wong C. Liu D. Finegold M. Harper J.W. Elledge S.J. p21(CIP1) and p57(KIP2) control muscle differentiation at the myogenin step Genes Dev. 13 1999 213 224 10.1101/gad.13.2.213 9925645
73 Andrés V. Walsh K. Myogenin expression, cell cycle withdrawal, and phenotypic differentiation are temporally separable events that precede cell fusion upon myogenesis J. Cell Biol. 132 1996 657 666 10.1083/jcb.132.4.657 8647896
74 Mademtzoglou D. Asakura Y. Borok M.J. Alonso-Martin S. Mourikis P. Kodaka Y. Mohan A. Asakura A. Relaix F. Cellular localization of the cell cycle inhibitor Cdkn1c controls growth arrest of adult skeletal muscle stem cells Elife 7 2018 e33337 10.7554/eLife.33337
75 Kumar S. Perlman E. Harris C.A. Raffeld M. Tsokos M. Myogenin is a specific marker for rhabdomyosarcoma: an immunohistochemical study in paraffin-embedded tissues Mod. Pathol. 13 2000 988 993 10.1038/modpathol.3880179 11007039
76 Sebire N.J. Malone M. Myogenin and MyoD1 expression in paediatric rhabdomyosarcomas J. Clin. Pathol. 56 2003 412 416 10.1136/jcp.56.6.412 12783965
77 Pfaffl M.W. A new mathematical model for relative quantification in real-time RT-PCR Nucleic Acids Res. 29 2001 e45 10.1093/nar/29.9.e45 11328886
78 Bentley D.R. Balasubramanian S. Swerdlow H.P. Smith G.P. Milton J. Brown C.G. Hall K.P. Evers D.J. Barnes C.L. Bignell H.R. Accurate whole human genome sequencing using reversible terminator chemistry Nature 456 2008 53 59 10.1038/nature07517 18987734
79 Dobin A. Davis C.A. Schlesinger F. Drenkow J. Zaleski C. Jha S. Batut P. Chaisson M. Gingeras T.R. STAR: ultrafast universal RNA-seq aligner Bioinformatics 29 2013 15 21 10.1093/bioinformatics/bts635 23104886
80 Liao Y. Smyth G.K. Shi W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features Bioinformatics 30 2014 923 930 10.1093/bioinformatics/btt656 24227677
81 Ewels P. Magnusson M. Lundin S. Käller M. MultiQC: summarize analysis results for multiple tools and samples in a single report Bioinformatics 32 2016 3047 3048 10.1093/bioinformatics/btw354 27312411
82 Love M.I. Huber W. Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2 Genome Biol. 15 2014 550 10.1186/s13059-014-0550-8 25516281
83 Subramanian A. Tamayo P. Mootha V.K. Mukherjee S. Ebert B.L. Gillette M.A. Paulovich A. Pomeroy S.L. Golub T.R. Lander E.S. Mesirov J.P. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles Proc. Natl. Acad. Sci. USA 102 2005 15545 15550 10.1073/pnas.0506580102 16199517
84 Mootha V.K. Bunkenborg J. Olsen J.V. Hjerrild M. Wisniewski J.R. Stahl E. Bolouri M.S. Ray H.N. Sihag S. Kamal M. Integrated analysis of protein composition, tissue diversity, and gene regulation in mouse mitochondria Cell 115 2003 629 640 10.1016/s0092-8674(03)00926-7 14651853
85 Mootha V.K. Lindgren C.M. Eriksson K.F. Subramanian A. Sihag S. Lehar J. Puigserver P. Carlsson E. Ridderstråle M. Laurila E. PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes Nat. Genet. 34 2003 267 273 10.1038/ng1180 12808457
86 Schindelin J. Arganda-Carreras I. Frise E. Kaynig V. Longair M. Pietzsch T. Preibisch S. Rueden C. Saalfeld S. Schmid B. Fiji: an open-source platform for biological-image analysis Nat. Methods 9 2012 676 682 10.1038/nmeth.2019 22743772
87 Ahrens H.E. Henze H. Schuler S.C. Schmidt M. Huttner S.S. von Maltzahn J. Analyzing Satellite Cell Function During Skeletal Muscle Regeneration by Cardiotoxin Injury and Injection of Self-delivering siRNA In Vivo J. Vis. Exp. 2019 10.3791/60194
88 Edgar R. Domrachev M. Lash A.E. Gene Expression Omnibus: NCBI gene expression and hybridization array data repository Nucleic Acids Res. 30 2002 207 210 10.1093/nar/30.1.207 11752295
