Differential regulation of autophagy by STAU1 in alveolar rhabdomyosarcoma and non‐transformed skeletal muscle cells

Abstract

Purpose

Recent work has highlighted the therapeutic potential of targeting autophagy to modulate cell survival in a variety of diseases including cancer. Recently, we found that the RNA-binding protein Staufen1 (STAU1) is highly expressed in alveolar rhabdomyosarcoma (ARMS) and that this abnormal expression promotes tumorigenesis. Here, we asked whether STAU1 is involved in the regulation of autophagy in ARMS cells.

Methods

We assessed the impact of STAU1 expression modulation in ARMS cell lines (RH30 and RH41), non-transformed skeletal muscle cells (C2C12) and STAU1-transgenic mice using complementary techniques.

Results

We found that STAU1 silencing reduces autophagy in the ARMS cell lines RH30 and RH41, while increasing their apoptosis. Mechanistically, this inhibitory effect was found to be caused by a direct negative impact of STAU1 depletion on the stability of Beclin-1 (BECN1) and ATG16L1 mRNAs, as well as by an indirect inhibition of JNK signaling via increased expression of Dual specificity phosphatase 8 (DUSP8). Pharmacological activation of JNK or expression silencing of DUSP8 was sufficient to restore autophagy in STAU1-depleted cells. By contrast, we found that STAU1 downregulation in non-transformed skeletal muscle cells activates autophagy in a mTOR-dependent manner, without promoting apoptosis. A similar effect was observed in skeletal muscles obtained from STAU1-overexpressing transgenic mice.

Conclusions

Together, our data indicate an effect of STAU1 on autophagy regulation in ARMS cells and its differential role in non-transformed skeletal muscle cells. Our findings suggest a cancer-specific potential of targeting STAU1 for the treatment of ARMS.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13402-021-00607-y.

Introduction

RMS is the most frequent childhood sarcoma [1] and accounts for 4-8 % of all pediatric malignancies [2]. RMS tumours arise in muscle beds and mainly metastasize to the head, neck, genitourinary tract and extremities [3–5]. Alveolar rhabdomyosarcoma (ARMS) is one of the major subtypes of RMS [6–8], which mostly occurs in young adults and represents approximately 30 % of RMS cases [9]. ARMS is characterized by the presence of oncogenic fusion genes which are generated by chromosomal translocations, t(2;13)(q35;q14) and t(1;13)(p36;q14), between, respectively, the PAX3 and PAX7 genes (which encode transcription factors of the paired box family) and the FOXO1 locus (coding for a member of the forkhead family of transcription factors) [10, 11]. These oncogenic fusion genes have been observed in ~ 80 % of ARMS cases. The first translocation involving PAX3 and FOXO1 located on chromosomes 2 and 13, respectively, has been observed in 55 % of the ARMS cases. The second translocation involves PAX7 and FOXO1, located on chromosomes 1 and 13, respectively. This translocation has been reported in ~ 20 % of ARMS cases. Evidence shows that ARMS patients carrying the former translocation (PAX3/FOXO1) have a poorer prognosis compared to those carrying the latter (PAX7/FOX1) translocation. The 4-year survival rate of the former group is 8 % while the survival rate of the later group is 75 %. Also, in both groups, the fusion protein expression levels negatively affect patient survival [12]. The remaining 20 % of ARMS cases lack PAX/FOXO1 fusion genes, and it has been found that their gene expression patterns and clinical outcomes are not distinguishable from those of embryonal RMS (ERMS). Therefore, their classification as ARMS tumours is disputed [13]. The fusion-negative subtype of ARMS exhibits a favourable prognosis. According to the available literature, the 5-year survival rate for non-metastatic fusion-negative ARMS is 70–80 % [14]. However, despite advances that have been made in therapy, the 3-year event-free survival rate for all ARMS cases (fusion-positive and fusion negative) remains ~ 30 % [15]. Such poor statistics highlight the need for either improving current or developing novel therapeutic approaches.

RNA-binding proteins (RBPs) exhibit a wide range of functions including in RNA transcription, processing and localization [16]. Due to their essential roles in the regulation of multiple cell functions, their impact on cancer development and progression has recently gained increasing attention [17]. Therefore, over the past several years, many RBPs have emerged as molecular biomarkers and as therapeutic targets in respectively cancer diagnosis and treatment [18, 19]. In this context, we previously reported a role of the RBP Staufen1 (STAU1) in RMS oncogenesis [20]. STAU1 is a multifunctional RBP involved in several aspects of RNA metabolism including mRNA transport, localization, stability, translation and splicing [21–24]. Our previous findings revealed that STAU1 is highly expressed in ARMS and that its transient silencing in vitro reduces the viability and metastasis of ARMS cells and in vivo partially inhibits tumour formation [20]. However, the mechanism(s) by which STAU1 exert(s) its effects on ARMS remained unclear.

Autophagy is a catabolic event that promotes cell survival by eliminating protein aggregates and damaged organelles, thereby supplying energy for cells [25–27]. Therefore, autophagy dysregulation has been reported to be associated with several diseases including cancer [28]. A growing body of evidence indicates a crucial role of autophagy in the development of anoikis-resistance in solid tumours [29] and enhancement of their metastasis [30]. Given this notion, elevated autophagy in tumour cells is considered to promote cancer progression and metastasis, which negatively affects patient survival. Therefore, targeting autophagy during cancer treatment has been suggested as a novel approach to enhance the clinical efficacy of cancer therapy [31]. In this context, recent studies showed that upregulated autophagy in ARMS cells promotes their survival, whereas genetic or pharmacological inhibition of autophagy reduced their viability, suggesting a positive impact of autophagy on disease progression [32, 33]. Based on our previous work focusing on STAU1, we set out to investigate the impact of STAU1 in the regulation of autophagy in ARMS. Our findings uncover a novel role of STAU1 in the regulation of autophagy in ARMS and highlight its differential effects in malignant and non-transformed muscle cells.

Materials and methods

Constructs and reagents

Lentiviral constructs and siRNAs were purchased from GE Healthcare dharmacon Inc., Addgene, and System Biosciences (Supplementary Materials and Methods). All chemical reagents used in this study are listed in the supplementary materials and methods section.

Cell culture, transfection, lentivirus production and infection

RH30 cells (CRL-2061) and C2C12 myoblasts (CRL-1772) were purchased from the ATCC, and RH41(ABC0-TC0969) cells were purchased from ACCEGEN Biotech. Primary human skeletal muscle cells, SKMC (CC-2561) and HSMM (CC-2580), were obtained from Lonza. Immortalized skeletal muscle cell lines, HSMM-C2 and HSMM-C3, were generated in Dr. Pantic’s lab at the University of Padua, Italy, as reported before [34]. All cell lines were cultured as recommended by the suppliers and in the presence of 0.2 % MycoZap™. Also, cells were regularly tested for mycoplasma contamination.

To generate knockdown cell lines, pTRIPZ-LV and pLKO-LV plasmids were used according to the protocol of the 3rd generation lentiviral packaging system [35]. Lentiviral particles were collected, filtered and stored at -80 ˚C. For transduction, 0.2 × 10⁶ cancer cells were cultured in 6-well plates and after 24 h, medium was replaced with 2 ml of the cell-specific medium, containing 500 µl lentivirus aliquot and 8 µg/ml Sequebrene. Seventy-two hours later, puromycin was added (concentration varied based on the cell type) to select transduced cells. Note: for doxycycline-inducible knockdown cells, the cells were treated with 20 µg/ml doxycycline 96 h before the start of each experiment.

To transiently overexpress STAU1 in normal myoblasts, transfection and transduction were performed as mentioned above. Seventy-two hours post-transduction, cells were collected for further analysis. Moreover, for performing siRNA-mediated knockdown of DUSP8 in RH30 cells, scrambled-siRNA and siDUSP8 were directly transfected into the cells in the presence of transfection reagent Lipofectamine RNAiMAX as per the manufacturer’s instructions (Thermo Fisher Scientific). Three days later, cells were collected for further experiments.

Western blotting

Cells and muscle tissues were lysed in 1x RIPA buffer (20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM Na2-EDTA, 1 mM EGTA, 1 % NP-40, 1 % sodium deoxycholate, 2.5 mM sodium pyrophosphate, 1 mM β-glycerophosphate, 1 mM Na3VO4, and 1 µg/ml leupeptin) and 1x muscle lysis buffer (20 mM HEPES, 10 mM NaCl, 1.5 mM MgCl₂, 1 mM EDTA, 20 % Glycerol, and 0.1 % Triton X-100) supplemented with Protease Inhibitor Cocktail (PIC) and PhosSTOP, respectively. Protein concentrations were calculated according to a BCA assay (Thermo Fisher Scientific). For Western blot analysis, 20 µg protein samples were separated by SDS-gel electrophoresis and then transferred onto 0.2 μm nitrocellulose membranes (BioRad). The membranes were blocked with 5 % milk powder dissolved in 1x TBST (137 mM NaCl, 2.7 mM KCl, 19 mM Tris-base, and 0.1 % TWEEN 20) and incubated with specific primary antibodies overnight at 4˚C. The next day, the membranes incubated with appropriate HRP-conjugated secondary antibodies diluted in 1 % milk powder in 1x TBST for 1 h at RT. The HRP substrate chemiluminescent reaction was performed at RT in light. X-ray films were quantified using ImageJ (NIH version 1.0) and/or Image Lab. To ensure that the band intensities were proportional of the amounts of target protein, different amounts of protein (5 µg, 10 µg and 20 µg) were loaded on the gels. The level of β-actin served to control for variations in protein loading. Band detection was within the linear range (a list of primary and secondary antibodies is provided in the supplementary materials and methods section).

RNA extraction, reverse transcription and quantitative real‐time PCR

mRNA was isolated from cells using TRIzol and a RNA Purification kit as per the manufacturer’s instructions (Thermo Fisher Scientific). Two µg RNA was used for the synthesis of complementary DNA (cDNA) according to the Super Script® II First-Strand Synthesis System (Thermo Fisher Scientific). Gene expression was quantified by RT-qPCR using gene-specific primers. All primers were designed in our laboratory and synthesized by Life Technologies (all primers used in the RT-qPCR assays are listed in Supplementary Table S1). Data were analyzed based on the Livak and Schmittgen’s 2^− ΔΔCT method and normalized to 3-phosphate dehydrogenase (GAPDH) or β-actin as reference genes [36].

RNA immunoprecipitation assay

Cells were grown for 2 days and fixed in 1 % formaldehyde for 10 min at RT. The reaction was quenched with 0.25 M glycine in 1x PBS for 10 min. Next, the cells were scraped, washed, and resuspended in 1x RIPA buffer. The samples were sonicated by four 15-s pulses and centrifuged (16,000 g for 10 min at 4 °C). A/G plus beads (40 µl/sample; Santa Cruz Biotechnology, Inc.) were washed 3–4 times with low stringency RIPA buffer (50 mM Tris-HCL pH: 7.4, 150 mM NaCl, 1 % NP40, 0.25 % Sodium Deoxycholate, 1 mM EDTA and 1x Protease inhibitor cocktail). Beads were resuspended in low stringency RIPA buffer containing 200 µg/ml competitor tRNA and 40 µg/ml salmon sperm DNA buffer and then incubated at 4 °C for 1 h. Complexes were immunoprecipitated using 3 µg of specific antibodies and protein A/G plus beads overnight at 4 °C. The next day, samples were washed 3–4 times with low stringency RIPA buffer followed by two washes with TE buffer (10 mM Tris-HCl, pH 8.0 and 1 mM EDTA). Subsequently, beads containing the immunoprecipitated samples were resuspended in 100 µl elution buffer (50 mM Tris-HCl, pH 7.0, 5 mM EDTA, 10 mM DTT and 1 % SDS), after which formaldehyde-induced cross-linking was reversed by incubating the samples at 70 °C for 5 h. The efficiency of immunoprecipitation was assessed by Western blotting of STAU1 protein. RT-qPCR analyses of the mRNAs of interest were performed after RNA extraction with TRIzol.

mRNA stability assay

Cells were grown for 96 h and then treated with 10 µg/ml Actinomycin D for 2, 4 and 6 h. RNA samples were isolated and used for the synthesis of cDNA as mentioned above. RT-qPCR assays were performed to determine the percentage of remaining mRNA of the genes of interest. In this experiment, c-Myc and β-actin mRNAs were used as positive and negative controls, respectively.

Annexin V-based apoptosis assay

After 96 h incubation, cells were collected and washed with 1x Annexin buffer (0.1 M HEPES/NaOH (pH 7.4) 1.4 M NaCl and 25 mM CaCl₂). Next, 1 × 10⁶ cells were stained with 12.5 µg/ml AnnexinV-fluorescein isothiocyanate (AnnexinV) and 5 µg/ml LIVE/DEAD™ fixable violet dead cell stain solution for 15 min at RT in the dark. The stained cells were re-suspended in 500 µl of 1x Annexin buffer. Fluorescent intensities for AnnexinV and Violet stain were quantified using a BD FACSCelesta™ flow cytometer (Spectron Corporation) at wavelengths of 488 nm and 405 nm, respectively.

Autophagy detection assay

To visualize changes in autophagy, an autophagy detection kit was used as per the manufacturer’s instructions (Abcam) and Chloroquine (100 µM) treatment was used as a positive control. Briefly, four days after culturing cells, they were re-suspended in 100 µl FACS buffer (1x PBS, 1 % FBS and 1 % 0.5 M EDTA) containing Autophagy Green stain (1:1000 dilution) and incubated for 30 min at 37 °C in the dark. Next, the cells were fixed with 4 % paraformaldehyde (PFA), resuspended in 500 µl FACS buffer and analyzed using the above-mentioned flow cytometer at a wavelength of 488 nm.

STAU1 transgenic mouse model

STAU1 transgenic mice (MCK-Staufen1-HA, Tg-6898) were generated in our laboratory as described previously [37].

Statistical analysis

Unpaired two-tailed Student’s t-tests were used to calculate significances between control and experimental conditions. mRNA half-lives (t_1/2) were calculated by non-linear regression analysis using GraphPad Prism 5.02 software.

Results

STAU1 is highly expressed in ARMS cells and its downregulation promotes apoptosis

Quantitative analysis of STAU1 mRNA and protein levels in normal human skeletal myoblasts (SKMC, HSMM, HSMM-C2, and HSMM-C3) and ARMS cell lines (RH30 and RH41) revealed a marked increase in its expression in the ARMS cells (Fig. 1a-b). To investigate the significance of elevated STAU1 expression in ARMS cells, we generated inducible STAU1-knockdown ARMS cell lines using a mixture of three independent STAU1-targeting lentiviral shRNA constructs. Subsequent Western blot and RT-qPCR analyses revealed a knockdown efficacy of ~ 60–70 % (p < 0.01) in both cell lines (Fig. 1c-d). Next, we examined the pro-apoptotic effect of STAU1 silencing on ARMS cells using an Annexin V-based apoptosis assay. Flow cytometry of stained cells indicated that the percentage of apoptotic cells significantly increased after STAU1 downregulation in both RH30 (p < 0.05) and RH41 (p < 0.01) cell lines (Fig. 1e). This effect, together with the elevated level of cleaved-caspase7, indicates activation of apoptosis in STAU1-depleted cells (Fig. 1f).

Fig. 1.

STAU1 is highly expressed in ARMS cell lines and its downregulation promotes apoptosis. a Western blot analysis of STAU1 in normal human skeletal muscle cells (NHSMCs) and ARMS cells. Quantification of STAU1 protein levels normalized to β-actin are presented as ratios of the control (n = 3). b RT-qPCR analysis of STAU1 in NHSMC and ARMS cells. Data were normalized to GAPDH (n = 3). c ARMS cells were infected with Doxycycline-Inducible LV-pTRIPZ or LV-shSTAU1 lentiviral particles followed by puromycin selection. Inducible control and STAU1-knockdown cell lines were treated with 20 µg/ml Doxycycline for 96 h. Western blot quantification of STAU1 protein levels in RH30 and RH41 cells; β-actin served as a control for protein loading. Data were normalized to β-actin and are presented as a ratio of the pTRIPZ control (n = 3). The linear range for quantitative Western blot detection was determined by loading different amounts of protein (5 µg, 10 and 20 µg) on the gel. The band intensity (the first three bands) was proportional to the amount of target protein (β-actin). d RT-qPCR analysis of STAU1 in RH30 and RH41 cells. Data were normalized to GAPDH (n = 3). e Representative dot plots for Annexin V/Live-dead Violet staining of pTRIPZ- and shSTAU1-expressing ARMS cells. Quantification of dead cells is presented as percentage of control cells (n = 3). f Representative Western blot of total-caspase7 (T-CASP7) and cleaved-caspase7 (C-CASP7) in ARMS cells. The ratio of C-CASP7/T-CASP7 is presented in the corresponding bar graph (n = 3). Data are mean ± SD; t-test was performed versus control (pTRIPZ) *** p ≤ 0.001; ** p ≤ 0.01; * p ≤ 0.05

STAU1 silencing attenuates autophagy degradation in ARMS cells

To examine autophagy, Western blot analyses of microtubule-associated protein 1 light chain 3 (LC3) A/B and SQSTM1/p62 (referred to as p62) showed a ~ 50 % (RH30: p < 0.01 and RH41: p < 0.05) reduction in the LC3II/I ratio associated with an accumulation of p62 protein, suggesting autophagy inhibition in STAU1-depleted cells. This effect was accompanied by decreases in BECN1, ATG16L1 and Bnip3 protein levels (Fig. 2a). Moreover, a marked reduction in autophagy-related gene (ATG) mRNA levels was observed in STAU1-depleted cells (Supplementary Fig. S1A). Similarly, autophagosome and autolysosome staining with fluorescent green dye suggested a lower accumulation (p ˂ 0.05) of autophagic vesicles in STAU1-knockdown cells (Fig. 2b).

Fig. 2.

STAU1 silencing downregulates autophagy in ARMS cell lines. a Representative Western blots of autophagy-related proteins (LC3A/B, p62, ATG16L1, BECN1 and Bnip3) in RH30 and RH41 cells. β-actin served as a control for protein loading (Note: for RH41 cells, the Bnip3 blot was re-probed with ATG16L1 antibody, so these two blots share the same loading control). Quantification of autophagy proteins, normalized to β-actin, is presented as ratio of the control (n = 3). The linear range for quantitative Western blot detection was determined by loading different amounts of protein (5 µg, 10 and 20 µg) on the gel. All analyses were performed after 96 h of pTRIPZ- and shSTAU1-expressing ARMS cells treated with 20 µg/ml Doxycycline. b Representative histogram of autophagy green staining in RH30 and RH41 cells. Quantification of the mean of fluorescent intensity is presented as ratio of the control cells. c Western blot analysis of LC3A/B in RH30 cells after 4 h treatment with 50, 100 and 200 nM Bafilomycin A1 (BafA1). d Western blot analysis of LC3A/B in RH30 cells after 4 h treatment with 100, 200 and 300 µM Chloroquine (CQ). Quantification of LC3A/BII normalized to β-actin is presented as a ratio of the non-treated control (n = 3). Data are mean ± SD; t-test was performed versus pTRIPZ or non-treated control *** p ≤ 0.001; ** p ≤ 0.01; * p ≤ 0.05

To substantiate evidence for the impact of STAU1 on autophagy, we focused on RH30 cells, the most conventional in vitro ARMS model. We examined the fusion between autophagosomes and lysosomes by co-staining endogenous LC3A/B and lysosome-associated membrane protein 1 (LAMP1) in these cells. We found a ~ 50 % (p ˂ 0.001) reduction in the level of co-staining between LC3A/B and LAMP1 in STAU1-depleted cells, suggesting that the fusion between autophagosomes and lysosomes was impaired (Supplementary Fig. S1B).

Finally, autophagic flux was assessed as the accumulation of undegraded LC3II-positive vesicles after blocking autophagosome degradation with the autophagy inhibitors Bafilomycin A1 (BafA1) and Chloroquine (CQ). As expected, LC3II was found to be highly accumulated (p < 0.001 or p < 0.01) in pTRIPZ control cells compared to STAU1-knockdown cells after treatment with the respective autophagy inhibitors (Fig. 2c-d). Together, these data indicate that autophagy is reduced in STAU1-depleted cells.

STAU1 directly interacts with BECN1 and ATG16L1 mRNAs and increases their stability

Given the available knowledge about the involvement of RBPs in maintaining the stability of ATG mRNAs [38–40] and our data on reduced expression levels of BECN1 and ATG16L1 upon STAU1 depletion (Fig. 2a, Supplementary S1A), we next focused on the impact of STAU1 on the stability of these transcripts. STAU1 binding to BECN1 and ATG16L1 mRNAs was examined by RNA immunoprecipitation (RIP) using a STAU1-specific antibody and IgG as a control (Fig. 3a). mRNA analyses of the RIP samples revealed binding of STAU1 to BECN1, ATG16L1 and c-Myc (positive control) mRNAs [23], but not to β-actin mRNA (negative control) (Fig. 3b-c). These findings are consistent with a previous report showing the presence of STAU1-binding sites in the 3’UTR of ATG mRNAs [41].

Fig. 3.

STAU1 directly binds to BECN1 and AG16L1 mRNAs and controls their stability. a RNA co-IP assays were performed with control immunoglobulin G (IgG) or rabbit polyclonal antibody directed against STAU1. The precipitated protein was analyzed by Western blotting with STAU1-specific antibody and the precipitated RNA from the same experiment was analyzed by RT-qPCR with primers for endogenous mRNAs. Fold precipitation relative to β-actin mRNA was calculated (not shown). Representative Western blot of precipitated STAU1 protein. b Representative agarose gel of the RT-qPCR analysis of co-precipitated β-actin, c-Myc, BECN1 and ATG16L1 mRNAs. c RT-qPCR analysis of co-precipitated c-Myc, BECN1 and ATG16L1 mRNAs normalized to GAPDH (n = 3). d RH30 cells were treated with 10 µg/ml Actinomycin D for 2, 4 and 6 h after which RT-qPCR was performed to determine the percentage of remaining mRNA of β-actin (control gene), BECN1 and ATG16L1. The mRNA half-lives (t_1/2) were calculated with 95 % confidence intervals. Data are mean ± SD; t-test was performed versus IgG control *** p ≤ 0.001; ** p ≤ 0.01; * p ≤ 0.05

To next determine the effect of STAU1 binding on the stability of BECN1 and ATG16L1 mRNAs, we performed stability assays. RT-qPCR analyses of BECN1 (p < 0.009) and ATG16L1 (p < 0.01) mRNAs after 2-, 4- and 6-hours treatment with 10 µg/ml Actinomycin D indicated that the percentage of remaining mRNAs was significantly reduced in STAU1-depleted cells compared to controls while the levels of β-actin mRNA remained unchanged (Fig. 3d). More specifically, the observed ~ 50 % (p < 0.01) and 75 % (p < 0.01) drop in the half-lives (t_1/2) of the BECN1 and ATG16L1 mRNAs, respectively, indicates that the direct interaction of STAU1 to these mRNAs increases their stability leading to autophagy activation.

STAU1 silencing downregulates the JNK signaling pathway via increasing the DUSP8 level which leads to autophagy inhibition

To further characterize the mechanism underlying the control of autophagy by STAU1, we examined two main autophagy regulatory pathways, i.e., the mTOR and JNK pathways [27, 42, 43]. The mTOR pathway was examined by Western blot analysis of mTOR, phospho-mTOR (p-mTOR), AKT and phospho-AKT (p-AKT) expression. No changes were observed in the levels of these proteins in STAU1-depleted cells (Fig. 4a), suggesting mTOR-independent regulation of autophagy. Given the role of JNK signaling in autophagy [42, 44], we next determined the impact of STAU1 depletion on JNK. An observed marked decrease in phospho-JNK (p-JNK) without an effect on total JNK expression indicated post-translational downregulation of JNK (Fig. 4b). More specifically, the ratio of p-JNK/total JNK decreased by approximately 80 % (p < 0.01) in STAU1-depleted cells. Of relevance, we noted a similar effect in RH41 cells. Specifically, STAU1 downregulation caused a significant reduction in the phosphorylation of JNK with no impact on total JNK levels in these cells, thereby resulting in a decrease (p < 0.01) in the p-JNK/total JNK ratio (Supplementary Fig. S2A). Conversely, STAU1 silencing did not affect the levels of total mTOR, p-mTOR, AKT and p-AKT (Supplementary Fig. S2B-C).

Fig. 4.

STAU1 silencing downregulates the JNK signaling pathway via increasing DUSP8 expression. (a) Representative Western blot of p-mTOR, mTOR, p-AKT and AKT in RH30 cells, showing no change in the mTOR signaling pathway in shSTAU1-expressing cells. Quantification of proteins, normalized to β-actin, is presented as ratio of the pTRIPZ control (n = 3). The analysis was performed after 96 h of pTRIPZ- and shSTAU1-expressing RH30 cells treated with 20 µg/ml Doxycycline. b Western blot analyses of p-JNK and Total-JNK in pTRIPZ- and shSTAU1-expressing RH30 cells indicate a statistically significant reduction in the ratio of p-JNK/JNK, suggesting JNK pathway downregulation. Quantification of p-JNK and JNK, normalized to β-actin, is presented as ratio of the pTRIPZ control, and the pJNK/JNK ratio was calculated accordingly (n = 3). c pTRIPZ-expressing cells were treated with 50 µM JNK inhibitor, SP600125 (SP), and shSTAU1-expressing cells were treated with 15 ng/ml JNK activator, Anisomycin (Ans.) for 24 h, after which the protein level of p-JNK was determined. d Western blotting of LC3II/I and BECN1 after treatment with 50 µM SP600125 and 15 ng/ml Anisomycin. e Western blot and RT-qPCR analyses of DUSP8 in RH30 cells, showing an increased DUSP8 level in STAU1-depleted cells. Quantifications of protein and mRNA normalized to β-actin and GAPDH, respectively, are presented as ratio of the pTRIPZ control (n = 3). f STAU1-depleted RH30 cells were transfected for 72 h with three different siRNAs targeting DUSP8. Representative Western blot analyses of DUSP8, p-JNK and JNK indicate a reduced DUSP8 expression and an upregulated JNK pathway in STAU1-/DUSP8-depleted cells. Quantification of Western blot analyses of DUSP8 and p-JNK/JNK indicates that DUSP8 silencing leads to upregulation of the JNK pathway in STAU1-depleted cells. g Western blotting of LC3A/B and p62 in RH30 cells, 96 h after transfection with si-scrambled (control) or si-DUSP8 showing activated autophagy in STAU1-/DUSP8-depleted cells. Corresponding statistical significances are presented as a ratio of control. Data are mean ± SD; t-test was performed versus control *** p ≤ 0.001; ** p ≤ 0.01; * p ≤ 0.05

To gain direct insight into the relevance of JNK signaling to autophagy regulation in RH30 cells, we treated control and STAU1-depleted cells with 50 µM of the JNK inhibitor SP600125 or 50 ng/ml of Anisomycin (a JNK activator), respectively. JNK inhibition or activation was confirmed by a substantial reduction (p < 0.01) in phosphorylated JNK (Fig. 4c). Subsequent Western blotting of LC3A/B and BECN1 revealed that JNK inhibition mimicked the effect of STAU1 silencing on autophagy. More specifically, we noted a ~ 60 % decrease in the LC3II/I ratio and a marked reduction in BECN1 protein levels (p < 0.001), indicative of autophagy inhibition in RH30 cells treated with this JNK inhibitor (Fig. 4d). In parallel, an elevated LC3II/I ratio (p < 0.001), associated with a ~ 50 % increase in BECN1 protein levels (Fig. 4d) in Anisomycin-treated cells, indicated that activation of JNK is sufficient to restore autophagy in RH30 cells despite reduced levels of STAU1.

In addition, autophagosome-lysosome fusion was assessed via co-staining of endogenous LC3A/B and LAMP1 in RH30 cells after treatment with SP600125 or Anisomycin. In agreement with the findings above, these experiments revealed a ~ 30 % reduction in the colocalization of LC3A/B with LAMP1 in SP600125-treated cells, suggesting autophagy inhibition after downregulation of JNK. Conversely, we observed a marked accumulation (p < 0.001) of colocalized LC3A/B with LAMP1 in STAU1-depleted cells after Anisomycin treatment indicating autophagy activation (Supplementary Fig. S3A-B).

Next, we determined how STAU1 downregulation attenuates JNK phosphorylation by focusing on post-translational events known to regulate JNK. We found that STAU1 depletion positively affected mRNA and protein expression of a JNK-specific phosphatase, Dual Specificity Phosphatase 8 (DUSP8). More specifically, RT-qPCR and Western blot analyses of DUSP8 revealed a ~ 2-fold increase in both DUSP8 mRNA (p < 0.01) and protein (p ˂ 0.001) levels after STAU1 silencing (Fig. 4e). To confirm that elevated DUSP8 expression in STAU1-depleted cells causes JNK inactivation and autophagy inhibition, we knocked down DUSP8 expression in STAU1-silenced cells using three siRNAs (Fig. 4f). Our results indicated that DUSP8 silencing restored phosphorylation of JNK in STAU1-depleted cells, as shown by a significant (p < 0.01) increase in the p-JNK/total JNK ratio (Fig. 4f). Moreover, enhanced autophagy in DUSP8-/STAU1-depleted cells was detected by an increased LC3II/I ratio and reduced p62 levels (Fig. 4g). Together, these data indicate that elevated DUSP8 levels in STAU1-depleted cells account for the attenuation of JNK signaling and autophagy inhibition.

STAU1 downregulation promotes mTOR-dependent autophagy increase in mouse myoblasts

To investigate the impact of STAU1 silencing on autophagy in non-transformed muscle cells, we knocked down STAU1 in C2C12 cells using a mixture of three independent STAU1-targeting lentiviral-pLKO.1 constructs. Subsequent Western blot and RT-qPCR assays revealed a 80–90 % knockdown efficiency (Fig. 5a-b). Next, using STAU1-depleted C2C12 cells, we assessed the impact of STAU1 levels on autophagy in non-transformed myoblasts. Western blot analyses of LC3A/B and p62 revealed a positive impact of STAU1 downregulation on autophagy in C2C12 cells. In striking contrast to ARMS cells, STAU1 silencing induced autophagy in C2C12 cells, as shown by a ~ 16-fold increase (p < 0.001) in the LC3II/I ratio and a marked decrease (p < 0.01) in the p62 protein level (Fig. 5c). In contrast to ARMS cells, no change (p > 0.05) in the p-JNK/JNK ratio was seen in STAU1-depleted C2C12 cells (Fig. 5d). However, in STAU1-knockdown C2C12 cells, a ~ 75 % reduction (p < 0.001) in the p-mTOR and total mTOR levels was observed (Fig. 5d). These data suggest that mTOR inhibition in STAU1-depleted C2C12 cells leads to autophagy activation. The decrease in the protein levels of phosphorylated and total mTOR was independent of the mTOR mRNA level, which remained unchanged (Fig. 5e). Next, in the presence of endogenous levels of STAU1, we activated autophagy in C2C12 cells by serum starvation or rapamycin (a mTOR inhibitor) treatment. We observed a significant reduction in the p-mTOR/total mTOR ratio, which was associated with an increase in the LC3II/I ratio and elevated p62 degradation (Fig. 5f), suggesting mTOR-dependent regulation of autophagy in C2C12 cells.

Fig. 5.

STAU1 silencing in normal mouse myoblasts upregulates autophagy in a mTOR-dependent manner. a STAU1 was silenced in C2C12 cells using a mixture of thee independent pLKO-shSTAU1 constructs. Western blot analysis of STAU1 in C2C12 cells. β-actin served as a control for protein loading. Quantification of STAU1, normalized to β-actin, is presented as ratio of the control (n = 3). b RT-qPCR analysis of STAU1 in C2C12 cells. Data were normalized to GAPDH (n = 3). c Representative Western blot analyses of LC3A/B and p62 in C2C12 cells, suggesting a reduction in autophagy after STAU1 silencing. β-actin served as a control for protein loading. Quantification of LC3A/B and p62, normalized to β-actin, is presented as ratio of the control (n = 3). In panels a and c, the linear range for quantitative Western blot detection was determined by loading different amounts of protein (5 µg, 10 and 20 µg) on the gel. d Western blotting of p-JNK, JNK, p-mTOR and mTOR in C2C12 cells showing activation of the mTOR pathway, but without effect on JNK signaling in STAU1-depleted cells. Quantification of proteins, normalized to β-actin, is presented as ratio of the pLKO control (n = 3). e RT-qPCR analysis of mTOR mRNA in control and STAU1-depleted C2C12 cells; data were normalized to β-actin (n = 3). f Protein levels of p-mTOR, mTOR, LC3A/B and p62 in C2C12 cells after 24 h treatment with rapamycin (50 and 100 nM) or serum starvation suggesting a mTOR-dependent regulation of autophagy in these cells. The ratio of LC3II/LC3I, as well as fold change of mTOR, p-mTOR and p62 proteins normalized to β-actin is presented as a bar graph (n = 3). Data are mean ± SD; t-test was performed versus non-treated control, *** p ≤ 0.001; ** p ≤ 0.01; * p ≤ 0.05; ns, p > 0.05

In a separate experiment, we generated transient green fluorescent protein (GFP)- and STAU1-overexpressing C2C12 cells using lentiviral pCDH-GFP or pCDH-STAU1⁵⁵-HA₃ (containing mouse STAU1⁵⁵ ORF) particles, respectively. Subsequent Western blot analysis of STAU1 with haemagglutinin (HA) tag showed elevated STAU1⁵⁵-HA₃ levels 72 h post-transduction (Fig. 6a). Moreover, a ~ 50 % decrease (p < 0.001) in the LC3II/I ratio accompanied by a significant (p < 0.05) p62 accumulation in STAU1-overexpressing cells (Fig. 6a) indicated that exogenous overexpression of STAU1 in C2C12 cells leads to autophagy inhibition. Further analysis revealed activation of the mTOR pathway in STAU1-overexpressing cells without affecting JNK (Fig. 6b). Collectively, these results indicate that in non-transformed myoblasts, STAU1 negatively controls autophagy by modulation of the mTOR pathway.

Fig. 6.

STAU1 overexpression in normal mouse myoblasts and muscle tissues downregulates autophagy. a C2C12 cells were transfected with lentiviral particles carrying pCDH-GFP and pCDH-STAU1-HA₃ constructs. Cells were analyzed for the protein level of HA-tag, STAU1, and autophagy markers, LC3A/B and p62. Western blot data were quantified and normalized to β-actin, and data are presented as a ratio of GFP-expressing control cells (n = 3). Data are mean ± SD; t-test was performed versus control (pCDH-GFP). b Western blot analyses of p-JNK, total JNK, p-mTOR and total mTOR indicating no alteration in JNK signaling, but upregulation of the mTOR pathway in STAU1-overexpressing C2C12 cells (n = 4); data are mean ± SD; t-test was performed versus control (pCDH-GFP). c A STAU1 transgenic mouse model was previously developed in our lab. TA muscle samples of 4 weeks-old wildtype (n = 3) and transgenic mice (n = 3) were used for mRNA and protein extraction. Western blot and RT-qPCR analyses of STAU1 in wildtype (WT) and STAU1 transgenic (TG) mice. Quantification of STAU1, normalized to GAPDH, is presented as ratio of the WT control (n = 3). d Representative Western blots of autophagy markers (LC3A/B and p62). Bar graph represents statistical significance of LC3II/I and p62 levels in TG compared to WT. Data were normalized to GAPDH or β-actin and are presented as ratio of the WT control (n = 3). Data are mean ± SD; t-test was performed versus control (WT), *** p ≤ 0.001; ** p ≤ 0.01; * p ≤ 0.05; ns, p > 0.05

STAU1 transgenic mice show autophagy inhibition in muscle tissues

To confirm our findings in vivo, we examined autophagy in tibialis anterior (TA) muscles obtained from a STAU1 transgenic (TG) mouse model. The skeletal muscle-specific STAU1 transgenic mouse model (MCK-Staufen1-HA or MCK-Stau1) was developed in our lab as previously described [37]. TA muscle samples were obtained from 4-week-old wildtype (WT) and STAU1 transgenic mice and examined for STAU1 expression levels by RT-qPCR and Western blotting (Fig. 6c). We found that increasing STAU1 levels in vivo caused significant reductions in autophagy compared to muscles from wildtype mice. Specifically, a more than 2-fold accumulation of LC3II (p < 0.004) and p62 (p < 0.01) protein levels was observed in transgenic muscles, indicating a reduction in autophagosome-lysosome fusion (Fig. 6d). These findings in skeletal muscle tissue confirm the negative regulation of autophagy by STAU1 in vivo.

STAU1 upregulation inhibits autophagy in non‐transformed human skeletal muscle cells

We further examined the impact of STAU1 overexpression on autophagy in two immortalized human skeletal muscle cell lines, HSMM-C2 and HSMM-C3 [34]. For this, cells were transduced with lentiviral vectors carrying pCDH-GFP or pCDH-STAU1⁵⁵-HA₃ (containing human STAU1 ORF) constructs. The efficacy of STAU1 overexpression was evaluated by Western blot analysis of HA and STAU1 72 h post-transduction (Fig. 7a). Autophagy inhibition was determined via LC3II/I and p62 measurements in STAU1-overexpressing cells. Specifically, Western blotting of LC3A/B and p62 revealed a ~ 50 % decrease (p < 0.001) in the LC3II/I ratio, which was accompanied by a significant (p < 0.05) accumulation of p62 protein in both HSMM-C2 and HSMM-C3 STAU1-overexpressing cells (Fig. 7b). As expected, STAU1 overexpression had no impact on the p-JNK and JNK protein levels, nor on the p-JNK/JNK ratio (Fig. 7c). However, in accordance with our data obtained with C2C12 cells, downregulation of autophagy in STAU1-overexpressing cells was accompanied by mTOR upregulation, indicated by a ~ 2-fold increase (p < 0.01) in the total mTOR and p-mTOR levels (Fig. 7d), independent of changes in mTOR mRNAs (Fig. 7e). Collectively, these data indicate that STAU1 negatively regulates autophagy in non-transformed human skeletal muscle cells by activating the mTOR pathway.

Fig. 7.

STAU1 upregulation in normal human skeletal muscle cells reduces autophagy in a mTOR-dependent manner. a Immortalized human skeletal muscle cell lines (HSMM-C2 and HSMM-C3) were transiently transduced with LV-pCDH-GFP and LV-pCDH-STAU1-HA₃ lentiviral particles. Seventy-two hours post-transduction, proteins were extracted. Western blot analyses of HA-tag and STAU1 confirms STAU1 overexpression; quantification of HA-tag and STAU1 (n = 3). b Western blot analysis of LC3A/B and p62 proteins shows autophagy inhibition after STAU1 upregulation; quantifications of LC3A/B and p62 normalized to β-actin are presented as ratio of the GFP control (n = 3). c Western blot analyses of p-JNK and total JNK in control and STAU1-overexpressing cells. (Note: in panels a and c, the JNK blot was re-probed with HA antibody, so these two blots share the same loading control). Quantifications of p-JNK, total JNK, and ratio of p-JNK/JNK normalized to β-actin are presented as ratio of the GFP control (n = 3). d Representative Western blots of mTOR and p-mTOR, normalized to β-actin, indicating an activation of the mTOR pathway in STAU1-overexpressing cells (n = 4). (Note: here, the mTOR blot was re-probed with p-mTOR antibody, so these two blots share the same loading control). e RT-qPCR analysis of mTOR mRNA in control and STAU1-overexpressing cells; data were normalized to β-actin (n = 3). Data are mean ± SD, t-test was performed versus control (pCDH-GFP) *** p ≤ 0.001; ** p ≤ 0.01; * p ≤ 0.05

Discussion

Based on our previous findings on the crucial role of STAU1 in myogenic differentiation and neuromuscular diseases [24, 37, 45], we recently developed a new line of investigation focused on the role of STAU1 in RMS [20]. Although our previous work revealed a key role of STAU1 in the progression of ARMS [20], the mechanisms and signaling events by which STAU1 exerts its impact in ARMS remained to be examined. In the current study, we investigated the role of STAU1 in the regulation of autophagy as a pro-survival mechanism in both non-transformed and malignant (ARMS) muscle cells. Our findings highlight a key role of STAU1 in ARMS development via the control of autophagy and downstream signaling events.

It is well established that induction of autophagy is essential for cell survival as well as cancer progression and metastasis [46]. Autophagy control is complex and, recently, the impact of RBPs in the regulation of autophagy has emerged [47]. More specifically, it has been shown that a group of RBPs is responsible for the control of different aspects of ATG mRNA metabolism such as maintaining their stability (via DDX6 and ELAVL1/HuR), modulating their translation (via CPEB1 and ELAVL4/HuD) and regulating splicing (via TARDBP) [38–40, 47–49]. In the present study we, therefore, investigated the involvement of the RBP STAU1 in autophagy and post-transcriptional regulation of the autophagy-related genes BECN1 and ATG16L1. Although the significance of autophagy for the survival of ARMS cells has been previously shown [32], we uncovered a novel role of STAU1 in autophagy control in ARMS cells. We found that interaction of STAU1 with BECN1 and ATG16L1 mRNAs enhanced their stability leading to increased levels of the respective proteins. Concordantly, we found that knockdown of STAU1 caused attenuation of lipidation of LC3A/B by BECN1 and ATG16L1 and inhibition of autophagy.

It is well-known that the mechanisms underlying autophagy may involve various signaling pathways including the mTOR [50, 51], JNK [44, 50], Ras/PKA [52] and HIF1α1 [53] pathways. Here, we focused on a putative involvement of the mTOR and JNK pathways in the STAU1-mediated regulation of autophagy [42]. Our findings in ARMS cells indicate an involvement of JNK signaling in autophagy control. Several studies have reported JNK-mediated regulation of autophagy in different cancers [44, 54]. Similarly, we found that in the presence of STAU1, JNK is responsible for autophagy control in ARMS cells. These findings suggest a novel role of STAU1 in the regulation of JNK signaling in human cancer. JNK activity is known to be regulated by several mechanisms that involve activation and/or inactivation of various upstream factors [55]. For instance, two MAP kinase kinases (MKK4 and MKK7) have been found to be responsible for the direct phosphorylation and activation of JNK, while DUSP8 is known for inhibiting JNK via dephosphorylation [56, 57]. Available evidence on a direct interaction of STAU1 to mRNAs of other members of the DUSP family prompted us to investigate the impact of STAU1 on DUSP8 expression [41]. We found that STAU1 silencing increased DUSP8 expression, leading to downregulation of JNK signaling and autophagy inhibition.

Due to the fundamental role of autophagy in maintaining homeostasis [58], its inhibition has been found to exert deleterious effects on both healthy and malignant cells [46, 59]. To prevent adverse effects of STAU1 targeting in normal cells, it is critical to determine whether STAU1 silencing inhibits autophagy and induces apoptosis in normal muscle cells in the same way as in ARMS cells. To examine this and to be able to propose STAU1 as a cancer-specific target for ARMS treatment, we determined the impact of STAU1 on autophagy and cell death in non-transformed muscle cells. We found that, contrary to ARMS cells, STAU1 downregulation in non-transformed myoblasts activated autophagy in a mTOR-dependent manner, without affecting cell death, suggesting that STAU1 silencing selectively sensitizes cancer cells to apoptotic cell death through inhibition of autophagy.

Our findings on the tumorigenic effect of elevated autophagy in ARMS cells, expressing high levels of STAU1, are in agreement with previous work examining the role of autophagy in the progression of sarcomas. A growing body of evidence suggests that the therapeutic potential of autophagy inhibitors alone or in combination with anticancer drugs in the treatment of different types of sarcomas including osteosarcoma, fibrosarcoma, chondrosarcoma and RMS [60], might be highly relevant. It has been found, for instance, that autophagy inhibition with CQ, BafA1, 3-MA or siRNAs targeting LC3 and BECN-1, reduced the proliferation of murine fibrosarcoma L929 cells while increasing apoptosis [61, 62]. Moreover, autophagy blockade has been reported to enhance chemotherapy-induced cell death in sarcoma cell cultures. In this context, it was found that CQ treatment of osteosarcoma cells increased their chemosensitivity to cisplatin through downregulation of p62 protein [63]. Likewise, in chemo-resistant chondrosarcoma cell lines (SW1353, HCS-2/8 and Hs819T), treatment with autophagy inhibitors (i.e., CQ, BafA1 and 3-MA) blocked drug-induced autophagy and enhanced cytotoxic effects of chemotherapeutics [64, 65]. Similar observations have been reported in animal models of different sarcomas [66]. For example, in a xenograft mouse model of sarcoma, a combined treatment of X-ray irradiation and CQ enhanced endoplasmic reticulum (ER) stress-induced tumour cell death and led to decreased tumour sizes [67].

Of relevance, several other studies have investigated the impact of autophagy inhibition on RMS progression and chemosensitivity. It has been reported that in RMS cells silencing of ATG7 or autophagy inhibition by BafA1 treatment reduced their growth [68]. In addition, a combination of CQ and Bortezomib or 17-DMAG (anticancer drugs), was shown to enhance drug-induced apoptosis in the RMS cell lines RH30 and RD [69]. Together, these findings support our current results on the anticancer impact of autophagy inhibition on ARMS. Therefore, we propose STAU1 targeting as a novel approach for autophagy inhibition to improve ARMS treatment.

Our data also revealed a differential regulation of autophagy in non-transformed and malignant muscle cells via distinct signaling pathways. In brief, STAU1 increased the protein level of mTOR, possibly by direct activation of its translation [70], in non-transformed muscle cells leading to the activation of the mTOR pathway and downregulation of autophagy. However, in malignant ARMS cells, STAU1 expression activates autophagy via both direct and indirect events by stabilizing ATG mRNAs and activating the JNK pathway, respectively (Fig. 8). These observations indicate that through different mechanisms, STAU1 oppositely regulates autophagy in normal and malignant myoblasts. The observed differential impacts of STAU1 on autophagy in malignant versus non-transformed cells may relate to its regulatory effect on distinct signaling pathways in cancer cells. Several pathways are altered in cancer cells to activate cancer-protection mechanisms under various environmental stresses. As a result, dysregulated pathways will not be subjected to normal control mechanisms. Given this, the observed difference in STAU1-mediated control of signaling pathways in malignant cells versus non-transformed cells may be due to various genetic and epigenetic changes that affect the mechanisms of autophagy control in cancer cells [71]. For instance, alteration in the redox balance leading to elevated reactive oxygen species (ROS) levels is one of the common hallmarks of cancer progression [72]. Along these lines, a role of ROS in activation of the JNK signaling pathway is a well-known phenomenon [73] and a crucial role of JNK signaling in the activation of various cancer mechanisms has been suggested [74]. In this context, JNK-dependent regulation of autophagy has been reported in different types of sarcoma including osteosarcoma [75], fibrosarcoma [76] and RMS [77]. More specifically, in response to cellular stress or drug treatment, elevated ROS has been found to lead to activation of the JNK pathway and to increased autophagy in sarcomas [60]. Although an essential role of the mTOR pathway alone in cancer progression and treatment is clear [78], signaling cross-talk between mTOR and other cancer pathways including the JNK pathway, is considered critical for cancer progression and treatment [79]. In this regard, previous studies reported that mTOR regulation by the activated JNK pathway is essential for tumour enhancement in intestinal cancer [80]. Given these findings, the observed role of the JNK pathway in the control of autophagy in ARMS may relate to a reciprocal relationship between autophagy-regulatory pathways that favor cancer cell survival.

Fig. 8.

Schematic presentation of STAU1-mediated regulation of autophagy in non-transformed and malignant myoblasts. Both endogenous and exogenous expression of STAU1 in non-transformed myoblast causes accumulation of mTOR and p-mTOR proteins leading to activation of the mTOR pathway. mTOR activation results in autophagy inhibition. However, in malignant myoblasts, i.e., ARMS cell lines, upregulated STAU1 directly binds to BECN1 and ATG16L1 mRNA and increases their stability, leading to autophagy activation. In parallel, STAU1 causes indirect activation of the JNK pathway by reducing the expression of DUSP8. Reduced DUSP8 inhibits dephosphorylation of p-JNK, which sustains its activation. Activated JNK further phosphorylates c-Jun, thereby increasing the transcription of autophagy-related genes and promoting autophagy in ARMS cells

Since highly metastatic cancer cells show a poorly differentiated phenotype, alterations in gene expression compared to non-transformed cells may drive cancer growth and survival. Therefore, the observed difference in the activated survival pathways between ARMS cells and non-transformed skeletal muscle cells may be linked to variations in the differentiation status of the studied cell lines. Also, it is known that the RH30 and RH41 cell lines are poorly differentiated and harbor mutations of the p53 tumor suppressor gene [81], a key player in stem cell differentiation [82]. Therefore, the presence of mutant p53 in ARMS cells in conjunction with a poorly differentiated phenotype may contribute to the observed variations. We have, however, cultured non-transformed skeletal muscle cells in a growth medium rather than a specialized differentiation medium in order to minimize the impact of cellular differentiation. Identification of the exact mechanisms underlying the differential activation of signaling pathways in malignant and non-cancerous cells will require further investigation. We also provide evidence that in ARMS cells pharmacological inhibition of autophagy by chloroquine does not affect the protein level of STAU1, while in non-transformed myoblasts, chloroquine treatment leads to the accumulation of STAU1. These data are concordant with a previous study that suggested autophagy as a mechanism for STAU1 degradation in non-transformed human cells [45]. This contrast in the impact of autophagy on STAU1 levels in non-transformed versus cancer cells may further explain the differential control of autophagy by STAU1 in these cells.

Based on the observed negative impact of STAU1 depletion on the survival of ARMS cells, STAU1 targeting may have a potential for ARMS treatment. However, due to multiple factors including its genetic complexity, heterogeneity and anti-drug resistance development, combinatorial therapies are recommended to enhance the effectiveness. A growing number of studies focuses on the identification of novel anti-cancer therapeutic targets in order to develop combinatorial therapies using conventional chemotherapeutics and modulators of newly identified targets. Such approaches may increase the efficacy of chemotherapeutic agents while reducing chemoresistance and improving treatment outcomes [83]. In the current context, further study is required to investigate the effect of STAU1 depletion on the effectiveness of chemotherapeutics, thereby evaluating its potential for ARMS treatment.

In summary, we provide evidence that STAU1 regulates autophagy in ARMS and non-transformed muscle cells in opposite manners involving activation of different signaling pathways. Our findings suggest that STAU1 targeting may serve as a novel therapeutic strategy to inhibit autophagy and promote ARMS cell death without autophagy impairment in normal cells.

Abbreviations

STAU1

Staufen1

ARMS

Alveolar Rhabdomyosarcoma

BECN1

Beclin-1

DUSP8

Dual specificity phosphatase 8

ERMS

Embryonal Rhabdomyosarcoma

RBPs

RNA-binding proteins

GAPDH

3-phosphate dehydrogenase

LC3

Microtubule-associated protein 1 light chain 3

BafA1

Bafilomycin A1

CQ

Chloroquine

RIP

RNA immunoprecipitation

p-mTOR

phospho-mTOR

p-AKT

phospho-AKT

p-JNK

phospho-JNK

TA

Tibialis Anterior

TG

Transgenic

WT

Wildtype

Author contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Shekoufeh Almasi. The first draft of the manuscript was written by Shekoufeh Almasi and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

This study was supported by grants from the Cancer Research Society (CRS).

Data availability

The data supporting the findings are available within the article and its supplementary materials. Additional supporting data and material are available from the corresponding author upon reasonable request.

Code availability

Not applicable.

Declarations

Conflict of interest

The authors declare no conflicts of interests. 

Ethics approval

All animal experimental protocols were approved by the University of Ottawa Institutional Animal Care Committee and were in accordance with the Canadian Council of Animal Care guidelines.

Consent to participate

Not applicable. 

Consent for publication

Not applicable.