Cell Biology International 33 (2009) 1020e1025 www.elsevier.com/locate/cellbi
Short communication
Defective myogenic differentiation of human rhabdomyosarcoma cells is characterized by sialidase Neu2 loss of expression Elena Stoppani, Stefania Rossi, Sergio Marchesini, Augusto Preti, Alessandro Fanzani* Department of Biomedical Sciences and Biotechnology, Unit of Biochemistry, University of Brescia, viale Europa 11, 25123 Brescia, Italy Received 13 March 2009; revised 30 April 2009; accepted 3 June 2009
Abstract Sialidase Neu2 is a glycohydrolytic enzyme whose tissue distribution has been detected principally in differentiated skeletal muscle. In this study we show that Neu2 expression is absent in different embryonal and alveolar human tumor rhabdomyosarcoma (RMS) cells, which are genetically committed myoblasts characterized by delayed differentiation. Forced myogenic differentiation of an embryonal RMS cell line, as obtained via pharmacological and genetic p38 activation or via follistatin overexpression, was characterized by Neu2 loss of expression despite the significant rise of different muscle-specific markers, suggesting therefore that the defective myogenic program of RMS cells is accompanied by Neu2 suppression. Ó 2009 International Federation for Cell Biology. Published by Elsevier Ltd. All rights reserved. Keywords: Neu2; Sialidases; Rhabdomyosarcoma; p38; Follistatin
1. Introduction Cytosolic sialidase Neu2 belongs to a family of glycohydrolytic enzymes (EC 3.2.1.18) whose activity catalyzes the removal of terminal sialic acids from glycoconjugates (Miyagi et al., 1993; Monti et al., 2002; Tringali et al., 2004). The tissue distribution of Neu2 has been principally restricted to skeletal muscle (Miyagi et al., 1993; Monti et al., 1999), and several in vitro studies suggest that Neu2 expression and enzymatic activity normally increase during myogenic differentiation (Fanzani et al., 2003, 2006, 2008; Sato and Miyagi, 1996). Human rhabdomyosarcoma (RMS) is the most common pediatric soft tissue sarcoma that arises from mesenchymal precursors with the potential to differentiate into skeletal muscle cells (Tonin et al., 1991), but fail to undergo
proper differentiation because of the presence of chromosomal aberrations (Galili et al., 1993; Loh et al., 1992). We have investigated Neu2 expression in different alveolar and embryonal RMS cell subtypes by means of RT-PCR analysis, gene reporter and enzymatic assays; additionally, Neu2 expression has been investigated in embryonal RD cells pharmacologically and genetically manipulated to restore the p38 pathway (Puri et al., 2000) or blunt the myostatin signaling (Langley et al., 2004; Ricaud et al., 2003), in order to force significantly the myogenic program. 2. Materials and methods All reagents were from SigmaeAldrich, if not otherwise indicated. 2.1. Cell cultures and treatments
Abbreviations: bp, base pairs; Cav-3, Caveolin-3; DMEM, Dulbecco’s modified Eagle’s Medium; FBS, fetal bovine serum; HS, horse serum; MKK6, MAP kinase kinase 6; MyHC, Myosin heavy chain; PBS, phosphate buffered solution; TPA, 12-O-tetradecanoylphorbol-13-acetate. * Corresponding author. Tel.: þ39 030 3717568; fax: þ39 030 3701157. E-mail address:
[email protected] (A. Fanzani).
Human embryonal RD cells were purchased from European collection of cell cultures (ECACC). Human RMS and murine C2C12 cells were cultured under standard conditions at 37 C and 5% CO2 in humidified air incubator in growth medium (GM)
1065-6995/$ - see front matter Ó 2009 International Federation for Cell Biology. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.cellbi.2009.06.005
E. Stoppani et al. / Cell Biology International 33 (2009) 1020e1025
consisting of DMEM high glucose supplemented with 10% FBS and 100 mg/ml penicillinestreptomycin. To induce differentiation, confluent cells were cultured every day with differentiation medium (DM) consisting of DMEM high glucose supplemented with 2% HS. Rat L6MLC/IGF1 myoblasts were cultured essentially as described elsewhere (Musaro` and Rosenthal, 1999). TPA treatment was used at the indicated doses when the cells reached 80% confluence in GM, without replacing the medium over the entire time-course. 2.2. Sialidase assay Sialidase was assayed by fluorometry, as described by Fanzani et al. (2008). A pH curve was set up for each experiment, usually between 5 and 6.5 points, to determine the optimum of the sialidase activity assayed in the cytosolic fraction. 2.3. RT-PCR analysis Total RNA was obtained by Tri-reagent extraction, digested with DNase (DNA-free, Ambion), and reverse-transcribed (2 mg) in the presence of 400 units of MMLV-RT (Promega). The following primers (250 nM) were used: mouse Neu2 forward 50 -CGAGCCAGCAAGACGGATGAG-30 and reverse amplify 50 -GGCTCTACAAGCTTACTCACTACCCGG-30 a 679 bp long fragment (33 cycles); rat Neu2 forward 5’-CCG TCCAGGACCTCACAGAG-30 and reverse 50 -TCACTGAGCA CCATGTACTG-30 amplify a 727 bp long fragment (33 cycles); human Neu2 forward 50 -CCTGCAGAAGGAGAGCGTGTT-30 and reverse 50 -GGTGAAGTTTCCGGTAGGCGTA-30 amplify a 550 bp long fragment (45 cycles); human Cav-3, forward 50 -ACCCCAAGAACATTAACGAG-3’ and reverse: 50 -TGCA GAAGGTGCGGATGCAG-30 amplify a 310 bp long fragment (38 cycles); human MyHC forward 50 -GGCAGAGAAGAC AGGTG AGCCTCAG-30 and reverse: 50 -CCTCATCT GGCTTTAGCACCGTAGC-30 amplify a 571 bp long fragment (38 cycles); human follistatin forward 50 -CTCTTCAAGTGG ATGATTTTC-30 and reverse: 50 -ACAGTAGGCATTATTG GTCTG-30 amplify a 344 bp long fragment (27 cycles). Gene expression levels were normalized to tubulin mRNA expression. 2.4. Plasmid transfection Transfections were carried out with Lipofectamine 2000 reagent (Invitrogen), according to manufacturer’s instructions. RD cells stably transfected with the pBabe vector harbouring the short human form of follistatin were obtained after 20 days selection in presence of 2 mg/ml puromycin. 2.5. Luciferase assay Luciferase activity was measured using the Promega Dual luciferase assay system, after transfection of the cells with a mix consisting of pGL2-Basic-Neu2 promoter (1410 bp long) vector and pRL-TK-Renilla luciferase control vector. Data were corrected for transfection efficiency by measuring
1021
the Renilla luciferase activity, according to manufacturer’s instructions. 2.6. Immunoblotting analysis Protein concentration was obtained by bicinchoninic acid assay (Pierce). Samples were separated by SDS-PAGE under reducing conditions and transferred to nitrocellulose membranes. To detect Cav-3, immunoblots were made with a diluted 1:1000 mouse monoclonal antibody against N-terminal portion of Cav-3 (clone 26, BD Transduction Laboratories), using Triton-insoluble membrane fractions. To detect MyHC, immunoblots were done with a diluted 1:1000 mouse monoclonal antibody (Hybridoma Bank, University of Iowa), using total cell lysates from harvested cells at 4 C in RIPA lysis buffer (1% Nonidet P40, 0.5% sodium deoxycholate, 0.1% SDS in 50 mM NaCl, 20 mM TriseHCl pH 7.6) containing a mix of protease inhibitors. Blots were incubated with secondary antibodies conjugated with horseradish peroxidase (Chemicon) and revealed by enhanced chemiluminescence (Chemicon). To normalize protein expression, an anti-tubulin antibody was used. 2.7. Immunofluorescence microscopy Myoblasts cultured on laminin (Roche) coated glass coverslips were fixed with ice-cold methanol, incubated in a humid atmosphere with a diluted 1:100 mouse monoclonal MyHC antibody (Hybridoma Bank, University of Iowa), followed by a diluted 1:500 anti-mouse Cy3 conjugated secondary antibody (Jackson Immunoresearch), washed and mounted with PBS/ Glicerol (1:9 v/v). Fluorescent staining was observed under an Axiovert S100 microscope (Zeiss). Pictures were taken with a digital camera (SensiCam) using the Image-Pro Plus software version 6.2. 3. Results 3.1. Analysis of the sialidase Neu2 expression levels in different rhabdomyosarcoma cell lines The expression of sialidase Neu2 in human tumor rhabdomyosarcoma (RMS) cells committed to myoblast lineage that fail to differentiate properly was investigated. To this purpose, we used a commercially available embryonal RD cell line, as well as 2 clones termed embryonal RD/18 and alveolar RD/12 (Lollini et al., 1991) to compare with the murine C2C12 and L6MLC/IGF1 myoblasts, which express sialidase Neu2 during differentiation (Fanzani et al., 2003; Musaro` and Rosenthal, 1999). After treatment with differentiating medium (DM) for up to day 5, C2C12 and L6MLC/IGF1 myoblasts underwent formation of numerous myotubes, whereas RMS cells had a delayed ability to fuse (Fig. 1A, phase contrast pictures), as represented in the case of embryonal RD cells (comparable behavior was also observed for RD/18 and RD/12 cells, not shown). Under the same conditions, immunofluorescence images clearly showed the presence of multinucleated myotubes positive for MyHC (MyHCþ) in C2C12 and L6MLC/IGF1
1022
E. Stoppani et al. / Cell Biology International 33 (2009) 1020e1025
Fig. 1. Analysis of Neu2 transcriptional and enzymatic levels in human RMS cells. A, Phase contrast pictures depicting the morphology of C2C12, L6MLC/IGF1 and RD cells exposed to DM up to day 5. In the same conditions, immunofluorescence (IF) was performed to detect the expression of MyHC. Bars ¼ 80 mM. B, SqRT-PCR was performed to detect Neu2 transcript in murine C2C12 and L6MLC/IGF1 myoblasts or in human RD, RD/18 and RD/12 cells exposed to DM up to day 5. Total RNAs are shown as control loading. Densitometric analysis is calculated as the ratio between the Neu2 product amplification and the respective 28S RNA band. C, 72 h after the Neu2 promoter transfection, the luciferase reporter activity was measured in cell lysates obtained from C2C12, L6MLC/IGF1 and human RMS cells. D, Neu2 enzymatic activity was assayed in the cytosolic fractions of C2C12, L6MLC/IGF1 and RMS cells exposed to DM up to day 5. The values are the average of 3 independent experiments.
myoblasts (Fig. 1A), a parameter representative of myoblast differentiation; in contrast, the few RD cells which developed as MyHCþ were characterized by reduced size compared to murine myotubes. Neu2 mRNA levels were measured by semi-quantitative RT-PCR analysis (sqRT-PCR) in human RMS cell lines and compared to those obtained in murine C2C12 and L6MLC/ IGF1 myoblasts (Fanzani et al., 2003, 2006; Musaro` and Rosenthal, 1999) after DM treatment up to day 5. Neu2 expression was easily detectable in C2C12 and L6MLC/IGF1 cells by 33 rounds of amplification (Fig. 1B), whereas the human Neu2 transcript levels were barely detectable after 45 rounds, as further quantified by densitometric band analysis, thereby
indicating that Neu2 transcript is poorly expressed in RMS cells. A gene reporter assay, as performed after transient transfection of a 1410 bp long Neu2 fragment promoter (Fanzani et al., 2003), showed that the Neu2 luciferase reporter activity was very low in RMS cells exposed to DM up to 72 h, whereas it was elevated in C2C12 and L6MLC/IGF1 myoblasts (Fig. 1C), confirming that the Neu2 transcriptional levels are suppressed in RMS cells. Subsequently, the enzymatic activity of sialidase Neu2 was tested using a fluorometric assay as currently the most reliable method of detecting Neu2 expression, given the absence of commercially available antibodies that recognize efficiently the endogenous protein. After DM treatment for up to day 5,
Fig. 2. Analysis of Neu2 transcriptional levels in embryonal RD cells undergoing forced myoblast differentiation. SqRT-PCR analysis was performed to detect Neu2, Cav-3 and MyHC transcript levels at day 4 in control and RD cells exposed to TPA treatment (A), or transfected with MKK6EE form (C) and follistatin (E). Tubulin expression was used to normalize the transcript content. Subsequently, immunoblotting analysis was performed to detect Cav-3 and MyHC expression in control and RD cells treated with TPA (0.1 and 0.5 mM) (B), or transfected with MKK6EE (D) and follistatin (F), respectively, up to day 4. Additionally, the phosphorylation of p38 was verified in control and RD cells transiently transfected with MKK6EE along the time-course differentiation (D).
1024
E. Stoppani et al. / Cell Biology International 33 (2009) 1020e1025
Neu2 activity was normally detectable in C2C12 and L6MLC/ IGF1 myoblasts, whereas it was practically absent in all the tested RMS cell lines (Fig. 1D). This preliminary data suggest that the defective differentiation of RMS cells is accompanied by loss of expression of sialidase Neu2. Additionally, the results prompted us to investigate whether Neu2 expression might be at least partially restored upon triggering forced differentiation in RMS cells. 3.2. Analysis of the sialidase Neu2 expression in embryonal RD cells pharmacologically and genetically manipulated to undergo terminal differentiation Hereafter we employed different strategies to force the myogenic conversion of embryonal RD cells in order to analyze Neu2 expression. The first approach was to reactivate the p38 pathway, either via administration of the permeable molecule TPA (Bouche´ et al., 1993; Mauro et al., 2002), or via transient transfection of the constitutively activated MKK6EE kinase form (Puri et al., 2000). After TPA administration or MKK6EE transfection, the differentiation was monitored through the expression of Cav-3 and MyHC, which was significantly increased in terms of transcription (Fig. 2A and C) and protein levels (Fig. 2B and D) after 4 days compared to untreated cells. As shown in the case of MKK6EE transfection, increased differentiation was dependent on the reactivation of p38 phosphorylation along the time-course of differentiation (Fig. 2D). However, Neu2 transcript levels were unaffected after both TPA and MKK6EE expression in RD cells (Fig. 2A and C). The second approach was to overexpress a short human follistatin form (Iezzi et al., 2004) in order to blunt the excessive myostatin signaling present in RD cells (Langley et al., 2004; Ricaud et al., 2003). Stably transfected RD cells (namely RD/ follistatin) displayed elevated levels of follistatin transcript, followed by increased Cav-3 and MyHC transcript (Fig. 2E) and protein expression (Fig. 2F) after 4 days compared to untransfected cells. Despite the observed increase of these markers, Neu2 transcript levels also remained unaffected in RD/follistatin cells (Fig. 2E). Phase contrast pictures document that TPA treatment or MKK6EE and follistatin overexpression increased the RD differentiation by formation of numerous myotubes, as depicted after 5 days (Fig. 3A); accordingly, immunofluorescence images (Fig. 3A) and graphic quantification of the MyHCþ myotubes (Fig. 3B) confirmed the increased differentiation of pharmacologically and genetically manipulated RD cells compared to untreated cells. However, the sialidase assay showed that the Neu2 enzymatic activity was suppressed in RD cells even after TPA treatment or MKK6EE and follistatin transfection, whereas it was normal in murine myoblasts exposed to DM up to day 5 (Fig. 3C). 4. Discussion The expression of sialidase Neu2 usually increases together with muscle-specific markers during in vitro myoblast differentiation and hypertrophy (Fanzani et al., 2003, 2006, 2008; Sato and Miyagi, 1996), suggesting that this enzyme plays a role
Fig. 3. Loss of Neu2 enzymatic activity is observed in embryonal RD cells undergoing forced differentiation. A, Phase contrast pictures depicting the morphology of embryonal RD cells exposed to DM, treated with TPA or alternatively transfected with MKK6EE form and follistatin (day 5). In the same conditions, immunofluorescence (IF) was performed to detect the MyHCþ myotubes. Bars ¼ 80 mM. B, The graph represents the average number of MyHCþ myotubes counted per field in RD cells exposed to DM, treated with TPA, and transfected with MKK6EE or follistatin up to day 5. C, Sialidase assay showing that Neu2 enzymatic activity is suppressed in RD cells undergoing forced differentiation, as obtained after TPA treatment or MKK6EE and follistatin overexpression (day 5); in contrast, murine C2C12 and L6MLC/IGF1 myoblasts exposed to DM exhibited high Neu2 enzymatic activity (day 5). The values are the average of 3 independent experiments.
in myogenesis. The present data suggest that the delayed myogenic differentiation detected in different RMS cell lines is characterized by loss of Neu2 at both transcriptional and enzymatic levels, as corroborated by RT-PCR, gene reporter and enzymatic analyses. Given this evidence, our assumption was that the RMS cell lines fail to express Neu2 due to lack of a proper myogenic program. For this reason, we subsequently exploited pharmacological and genetical manipulations to restore, at least partially, the myoblast differentiation in embryonal RD cells in order to verify whether a rise of Neu2
E. Stoppani et al. / Cell Biology International 33 (2009) 1020e1025
expression might occur. The strategies represent the most reliable methods of enhancing RD cell differentiation, as obtained by either reactivating p38 pathway (Puri et al., 2000), or interrupting myostatin signaling (Ricaud et al., 2003). However, the data suggest that the genetic phenotype of RMS cells leads to the complete suppression of sialidase Neu2 because the activation of p38 pathway or the attenuation of myostatin signalling are necessary to enhance RMS cell differentiation, but not sufficient to upregulate Neu2 expression. Under these conditions, Neu2 expression was never augmented in terms of transcription rate, although a significant recovery of Cav-3 and MyHC was obtained in RD cells. As a consequence, Neu2 enzymatic activity was always undetectable in RD cells, suggesting that the lack of Neu2 regulation might be first dependent on transcription deficiency. Another point for consideration concerns the size of myotubes; the number of myotubes in RD cells was significantly improved with the different stimuli in this study, but the average size was always significantly reduced compared to C2C12 or L6MLC/IGF1 myotubes. This suggests that a remarkable ability to elongate has been gained by RD cells upon treatment, but the process of myotube accretion remains severely impaired. Of note, the highest transcript and enzymatic Neu2 levels were detectable in myoblasts along the late differentiation step and during hypertrophy, when myotubes increase in size via further cell or myofiber recruitment, as demonstrated by several in vitro models (Fanzani et al., 2003, 2006, 2008). Therefore, it is conceivable that lack of Neu2 activity in RD nascent myotubes might be a consequence of delayed growth of myotube size. Finally, it is of note that Neu2 loss of expression was detected in other tumor cell types, such as the murine B16 melanoma and the leukemic K562 cells (Tokuyama et al., 1997; Tringali et al., 2007), and that Neu2 restoration by stable transfection decreases the proliferation rate (Tokuyama et al., 1997) or triggers apoptosis (Tringali et al., 2007), suggesting that loss of Neu2 function might correlate with cancer cell progression and metastatic invasiveness. In line with this hypothesis, Neu2 loss of expression might exacerbate the defective myogenic differentiation of RMS cells. Acknowledgments We thank Pierluigi Lollini (University of Bologna, Italy) for kindly providing RD/18 and RD/12 cells, Antonio Musaro` (University of Rome, Italy) for L6MLC/IGF1 myoblasts, Lorenzo Puri (Dulbecco Telethon Institute, Rome, Italy) for pCDNA/MKK6EE vector, and Vittorio Sartorelli (National Institutes of Health, Bethesda, USA) for pBabe/follistatin vector. This work was supported by grants from 60% MIUR to A.F. and to S.M. and by grants from ‘‘Amici per il cuore’’. References Bouche´ M, Senni MI, Grossi AM, Zappelli F, Polimeni M, Arnold HH, et al. TPA-induced differentiation of human rhabdomyosarcoma cells: expression of the myogenic regulatory factors. Exp Cell Res 1993;208:209e17.
1025
Fanzani A, Giuliani R, Colombo F, Zizioli D, Presta M, Preti A, et al. Overexpression of cytosolic sialidase Neu2 induces myoblast differentiation in C2C12 cells. FEBS Lett 2003;547:183e8. Fanzani A, Colombo F, Giuliani R, Preti A, Marchesini S. Insulin-like growth factor 1 signaling regulates cytosolic sialidase Neu2 expression during myoblast differentiation and hypertrophy. FEBS J 2006;273: 3709e21. Fanzani A, Giuliani R, Colombo F, Rossi S, Stoppani E, Martinet W, et al. The enzymatic activity of sialidase Neu2 is inversely regulated during in vitro myoblast hypertrophy and atrophy. Biochem Biophys Res Commun 2008; 370:376e81. Galili N, Davis RJ, Fredericks WJ, Mukhopadhyay S, Rauscher FJ, Emanuel BS, et al. Fusion of a fork head domain gene to PAX3 in the solid tumour alveolar rhabdomyosarcoma. Nat Genet 1993;5:230e5. Iezzi S, Di Padova M, Serra C, Caretti G, Simone C, Maklan E, et al. Deacetylase inhibitors increase muscle cell size by promoting myoblast recruitment and fusion through induction of follistatin. Dev Cell 2004;6: 673e84. Langley B, Thomas M, McFarlane C, Gilmour S, Sharma M, Kambadur R. Myostatin inhibits rhabdomyosarcoma cell proliferation through an Rbindependent pathway. Oncogene 2004;23:524e34. Loh EWJ, Scrable HJ, Livanos E, Arboleda MJ, Cavenee WK, Oshimura M, et al. Human chromosome 11 contains two different growth suppresser genes for embryonal rhabdomyosarcoma. Proc Natl Acad Sci U S A 1992; 89:1755e9. Lollini PL, De Giovanni C, Landuzzi L, Nicoletti G, Scotlandi K, Nanni P. Reduced metastatic ability of in vitro differentiated human rhabdomyosarcoma cells. Invasion Metastasis 1991;11:116e24. Mauro A, Ciccarelli C, De Cesaris P, Scoglio A, Bouche´ M, Molinaro M, et al. PKCalpha-mediated ERK, JNK and p38 activation regulates the myogenic program in human rhabdomyosarcoma cells. J Cell Sci 2002; 115:3587e99. Miyagi T, Konno K, Emori Y, Kawasaki H, Suzuki K, Yasui A, et al. Molecular cloning and expression of cDNA encoding rat skeletal muscle cytosolic sialidase. J Biol Chem 1993;268:26435e40. Monti E, Preti A, Rossi E, Ballabio A, Borsani G. Cloning and characterization of NEU2, a human gene homologous to rodent soluble sialidases. Genomics 1999;57:137e43. Monti E, Preti A, Venerando B, Borsani G. Recent development in mammalian sialidase molecular biology. Neurochem Res 2002;27:649e63. Review. Musaro` A, Rosenthal N. Maturation of the myogenic program is induced by postmitotic expression of insulin-like growth factor I. Mol Cell Biol 1999; 19:3115e24. Puri PL, Wu Z, Zhang P, Wood LD, Bhakta KS, Han J, et al. Induction of terminal differentiation by constitutive activation of p38 MAP kinase in human rhabdomyosarcoma cells. Genes Dev 2000;14:574e84. Ricaud S, Vernus B, Duclos M, Bernardi H, Ritvos O, Carnac G, et al. Inhibition of autocrine secretion of myostatin enhances terminal differentiation in human rhabdomyosarcoma cells. Oncogene 2003;22:8221e32. Sato K, Miyagi T. Involvement of an endogenous sialidase in skeletal muscle cell differentiation. Biochem Biophys Res Commun 1996;221: 826e30. Tokuyama S, Moriya S, Taniguchi S, Yasui A, Miyazaki J, Orikasa S, et al. Suppression of pulmonary metastasis in murine B16 melanoma cells by transfection of a sialidase cDNA. Int J Cancer 1997;73:410e5. Tonin PN, Scrable H, Shimada H, Cavenee WK. Muscle-specific gene expression in rhabdomyosarcomas and stages of human fetal skeletal muscle development. Cancer Res. 1991;51:5100e6. Tringali C, Papini N, Fusi P, Croci G, Borsani G, Preti A, et al. Properties of recombinant human cytosolic sialidase HsNEU2. The enzyme hydrolyzes monomerically dispersed GM1 ganglioside molecules. J Biol Chem 2004; 279:3169e79. Tringali C, Lupo B, Anastasia L, Papini N, Monti E, Bresciani R, et al. Expression of sialidase Neu2 in leukemic K562 cells induces apoptosis by impairing Bcr-Abl/Src kinases signaling. J Biol Chem 2007;282: 14364e72.