<<

bb10i54.qxd 10/5/00 9:13 am Page 543

Brief Communication 543

The retinoblastoma-like p130 is involved in the determination of reserve cells in differentiating myoblasts Gilles Carnac*, LLuis Fajas†, Aurore L’honoré*, Claude Sardet†, Ned J.C. Lamb* and Anne Fernandez*

During skeletal muscle differentiation, a subset of reserve cells, which remain adherent to culture dishes myoblasts remains quiescent and undifferentiated but (Figure 1a). Western blot analysis of pRb, p107 and p130 retains the capacity to self-renew and give rise to in proliferating myoblasts, myotubes and reserve cells differentiating myoblasts [1–3]: this sub-population of showed that pRb was found in both hyper- and hypo- muscle cells was recently termed ‘reserve cells’ [3]. In phosphorylated forms in growing myoblasts (Figure 1a). It order to characterise genes that can regulate the ratio was also present in reserve cells and in myotubes, but only between reserve cells and differentiating myoblasts, we in a hypo-phosphorylated form. p107 was present at high examined members of the retinoblastoma tumor levels in proliferating myoblasts and its expression suppressor family — Rb, p107 and p130 — an important dropped to undetectable levels as differentiation pro- family of negative regulators of transcription ceeded. p130 was expressed at low levels in myoblasts and factors and progression [4]. Although pRb and at increased levels in differentiated cells but clearly, this p107 positively regulate differentiation p130 upregulation mainly occurred in reserve cells. As we [5–7], the role of p130 in muscle cells remains unknown. previously reported, Myf-5 was detectable only in reserve We show here that p130 (protein and mRNA), but cells, whereas MyoD became restricted to myotubes as neither pRb nor p107, preferentially accumulates during differentiation proceeded. Troponin T, a marker of differ- muscle differentiation in reserve cells. Also, p130 is the entiation, was present only in myotubes, demonstrating major Rb-family protein present in E2F complexes in that reserve cells and myotubes had been correctly sepa- this sub-population of cells. Although forced expression rated (Figure 1a). To determine whether up-regulation of of either p130 or pRb in mouse C2 myoblasts efficiently p130 protein was accompanied by an increase in p130 blocked cell cycle progression, only p130 inhibited the mRNA, we performed a northern blot analysis. The level differentiation program. Furthermore, muscle cells of p130 mRNA was low in proliferating myoblasts and dra- overexpressing p130 had reduced levels of the muscle- matically increased (3–4-fold) in reserve cells (Figure 1b). promoting factor MyoD. In addition, p130 repressed the Myf-5 mRNA levels were the same in proliferative and transactivation capacity of MyoD, an effect abolished by reserve cells and it was totally absent from myotubes. co-transfection of pRb. Thus, we propose that p130, by MLC1A, a marker of differentiation, was mainly restricted blocking cell cycle progression and differentiation, to the myotube fraction (Figure 1b). Therefore, the could be part of a specific pathway that defines a pool changes in the pool of p130 protein appeared to parallel of reserve cells during terminal differentiation. the fluctuations in p130 mRNA level, suggesting that

Addresses: *Cell Biology Unit, IGH UPR 1142, 141 rue de la Figure 1 Cardonille, 34396 Montpellier cedex 5, France. †Institut de Génétique Moléculaire, UMR 5535, CNRS, 1919 route de mende, 34293 Montpellier cedex 5, France.

Correspondence: Anne Fernandez E-mail: [email protected]

Received: 13 January 2000 Revised: 29 February 2000 Accepted: 13 March 2000

Published: 20 April 2000

Current Biology 2000, 10:543–546

0960-9822/00/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved. Up-regulation of p130 expression during muscle differentiation is mainly restricted to reserve cells. C2.7 cells were grown for 2 days in Results and discussion proliferative conditions and then switched to low serum-containing p130 expression is mainly confined to reserve cells medium for 4 days to induce the differentiation program. (a) Western We previously showed [1] that a short and mild trypsinisa- blot analysis; (b) northern blot analysis. P, proliferative C2 myoblasts; R, reserve cells; M, myotubes; Trop.T, troponin T; Tub, tubulin. Tubulin tion of cultured differentiated myoblasts removes most and S26 are loading controls. myotubes, leaving only undifferentiated residual cells, the bb10i54.qxd 10/5/00 9:13 am Page 544

544 Current Biology Vol 10 No 9

Figure 2 reserve cells (Figure 2a), indicating that the different sub- populations of cells had been correctly isolated. These two subpopulations contained one major E2F-like DNA- binding activity (Figure 2b), which was highly represented in reserve cells. The binding specificity of this complex was confirmed using mutated E2F-binding oligonucleotides. In reserve cells and to a lesser extent in myotubes, addition of an anti-p130 antibody resulted in conversion of the upper complex to a supershifted complex (Figure 2c), whereas pRb or p107 antibodies did not supershift the complex, as previously reported [9,10]. Taken together, our results indi- cate that p130 is the main retinoblastoma-family linked to the E2F complex in reserve cells.

p130 prevents cell cycle progression and differentiation of mouse C2 myoblasts To determine whether p130 upregulation is crucial for keeping reserve cells in an undifferentiated quiescent stage, we overexpressed p130 in myoblasts (which contain p130 is the prominent component of the E2F complex in reserve cells. EMSA, used to detect that bind to (a) the MLC1a CArG box low levels of endogenous p130) and examined the effect of and (b,c) the E2F-binding site from the adenovirus E2 , was such forced expression on cell proliferation and differentia- performed essentially as described in [21]. (c) For EMSA supershift tion by co-immunofluorescence analysis. For comparison, experiments, antibodies were added to the reaction as indicated. the same experiment was carried out with pRb. C2 WT, wild-type DNA; mut, mutant DNA; P, proliferating myoblasts; R, reserve cells; M, myotubes. myoblasts were transiently transfected with plasmids encoding HA-tagged constructs of p130 or pRb, or encod- ing β-galactosidase driven by the cytomegalovirus pro- p130 accumulation in the pool of reserve cells is most moter (CMV–β-gal) as a control, and were pulse-labelled likely due to transcriptional activation of the p130 gene. with bromo-deoxyuridine (BrdU) to monitor cells undergo- ing S phase. As shown in Figure 3a and Table 1, most cells p130 is associated with E2F in reserve cells expressing either pRb or p130 failed to incorporate BrdU, The E2F plays a critical role in cell showing that they were blocked before S phase entry. In growth control and is a potential target of Rb-related pro- contrast, cells overexpressing β-gal frequently showed co- teins [4]. To determine whether p130 is associated with immunostaining for BrdU, indicating that transfected E2F protein in reserve cells, we performed a standard elec- DNA had no major effect on muscle cell proliferation. trophoretic mobility shift assay (EMSA) to detect com- plexes with E2F-like properties in reserve and myotube To evaluate the impact of p130 on early phases of differ- cell extracts. Specific binding of a protein complex to the entiation, we transiently overexpressed HA–p130 and MLC1A CArG box [8] was detected in myotubes but not in HA–pRb in C2 myoblasts. Cells were then placed in low

Figure 3

Overexpression of p130 blocks both muscle cell proliferation and differentiation. Mouse C2 myoblasts were transfected with CMV–β-gal, CMV–HA–p130 or HA–pRb expressed from the plasmid pECE. (a) At 24 h after transfection, BrdU was added to C2 cultures for 24 h and cells were then processed for co-immunofluorescence analysis using antibodies directed against BrdU to detect DNA synthesis (left panels) and β-gal or HA tag to detect overexpressed proteins (middle panels). Cell nuclei were revealed by staining for DNA with Hoechst (right panels). (b) At 6 h after transfection, cells were switched to low serum-containing differentiation; middle panels). Cell nuclei differentiation; filled arrows indicate staining medium for 48 h and co-immunofluorescence were revealed by staining for DNA with with anti-HA but not anti-, i.e. analysis was then performed using antibodies Hoechst (right panels). Open arrows indicate blocked proliferation or differentiation. The to HA (left panels) or myogenin (to probe for co-staining, i.e. no inhibition of proliferation or scale bar represents 10µm. bb10i54.qxd 10/5/00 9:13 am Page 545

Brief Communication 545

Table 1 Figure 4

Effect of forced expression of p130 and Rb on C2 myoblast proliferation and differentiation.

Cells (%) positive for: BrdU Myogenin MyoD

β-gal 70 (n = 120) n.d. n.d. HA–Rb 10 (n = 128) 90 (n = 220) 90 (n = 165) HA–p130 10 (n = 133) 5 (n = 150) 5 (n = 138)

Summary of quantification for BrdU incorporation, myogenin and MyoD expression in C2 cells overexpressing CMV–β-gal, CMV–HA–p130 or pECE–HA–pRb (see Figures 3,4a). n.d., not determined.

serum-containing medium to induce differentiation. Over- expression of pRb did not significantly inhibit the process of differentiation, as 90% of HA-positive cells expressed the differentiation marker myogenin (Figure 3b, lower panels). In contrast, over 90% of HA–p130 positive cells were negative for myogenin staining, indicating that p130 clearly inhibited differentiation (Figure 3b, lower panels). Therefore, even though forced expression of pRb and p130 both blocked muscle cell proliferation with similar Overexpression of p130 inhibits MyoD expression and protein activity. (a) C2 myoblasts were transfected with either CMV–HA–p130 or efficiency, only p130 actively prevented their differentia- pECE–HA–pRb. At 6 h after transfection, cells were switched to tion (see Table 1). low-serum-containing medium for 48 h. Cells were then processed for co-immunofluorescence analysis using antibodies directed against HA p130 inhibits MyoD expression and its ability to (left panels) and MyoD (middle panels). Nuclei were revealed by staining for with Hoechst (right panels). Open arrows indicate co-staining, i.e. no transactivate muscle genes inhibition of proliferation or differentiation; filled arrows indicate staining In many myogenic cell lines, the capacity of cells to differen- with anti-HA but not anti-myogenin, i.e. blocked proliferation or tiate appears to be linked to the level of MyoD expression differentiation. The scale bar represents 10 µm. (b) 10T1/2 cells were [11]. To test whether p130 can regulate the expression of tranfected with MCK–luc together with plasmids expressing the indicated MyoD, C2 cells were transiently transfected with HA–p130 proteins or a control vector (PEMSV). Cells were grown for 48 h in differentiation medium and then harvested for the luciferase assay. Values or HA–pRb and then analysed by immunofluorescence plotted for experiments 1 and 2 are mean values of duplicate dishes. analysis for MyoD and HA tag expression. Figure 4a shows that ectopic expression of p130 strongly reduced MyoD protein levels: only 10% of HA–p130-positive cells expressed Stem-cell-like subpopulation of muscle precursors in vitro MyoD (upper panels). In contrast, 90% of pRb expressing In this report, we have identified a new marker expressed cells presented normal levels of MyoD protein (Figure 4a, in reserve cells, the Rb-related protein p130, which, by lower panels; see also Table 1). Because MyoD can activate inhibiting both cell proliferation and differentiation, may its own expression [12], we asked whether p130 prevented participate in the establishment of reserve-cell properties. MyoD expression by interfering with its transcriptional activ- Another gene previously shown to be specifically expressed ity. To assess potential regulation of MyoD transactivation in reserve cells was Myf-5, a member of MyoD gene family capacities by p130, we used a standard transcriptional assay. [1–3]. It is puzzling to detect a muscle-promoting factor in A construct containing the promoter of muscle creatine cells that failed to differentiate unless the role of Myf-5 is kinase driving luciferase expression (MCK–luc) was used as to maintain muscle commitment in such cells. Recently, a reporter gene for MyoD activity. When transfected into Dominov et al. [13] reported that Bcl-2, an apoptosis- mouse 10T1/2 fibroblasts, the MCK promoter had a low inhibiting protein, is expressed by myoblasts and persists in level of activity that increased up to threefold upon co-trans- reserve cells. It appears to participate in clonal expansion of fection of MyoD (Figure 4b). Co-expression of p130, but not myogenic cells by protecting them from apoptosis. In the pRb, resulted in a marked inhibition of MyoD-dependent literature, there have been a number of reports of induction activation of MCK–luc. Interestingly, overexpression of pRb of proteins in mature skeletal muscle that paradoxically together with p130 relieved this p130-dependent inhibition induce cell cycle arrest but block differentiation [14,15]. In of MyoD activity. Therefore, we postulate that inhibition on light of our results, we postulate that these inhibitory pro- MyoD and protein activity by p130 is teins might be specifically induced in reserve cells and, in responsible for the negative control imposed by p130 on conjunction with Myf-5, Bcl-2 and p130, could function as a muscle cell differentiation. binary switch that would determine whether cells progress bb10i54.qxd 10/5/00 9:13 am Page 546

546 Current Biology Vol 10 No 9

into the differentiation pathway or commit to the stem-cell- Acknowledgements like subpopulation of muscle precursors. We are grateful to Anne Bonnieu and Marie Vandromme for many helpful discussions and critical reading of the manuscript. This work was supported by grants from Association Francaise contre les Myopathies (A.F.M.) to A.F. p130, the cell cycle and differentiation and Association de la Recherche contre le Cancer (no. 9484) to N.J.C.L. We and others [1–3] showed that, upon entry into differen- tiation, MyoD expression is restricted to myoblasts commit- References ted to differentiate and therefore absent from reserve cells. 1. Kitzmann M, Carnac G, Vandromme M, Primig M, Lamb NJC, Fernandez, A: The muscle regulatory factors MyoD and Myf-5 Yoshida et al. [3] suggested that downregulation of MyoD is undergo distinct cell cycle-specific expression in muscle cells. a causal event in the formation of reserve cells. Therefore, J Cell Biol 1998, 142:1447-1459. 2. Lindon C, Montarras D, Pinset C: Cell cycle-regulated expression of high levels of p130 in reserve cells might be required to the muscle determination factor Myf-5 in proliferating myoblasts. directly block MyoD expression and activity. Alternatively, J Cell Biol 1998, 140:111-118. 3. Yoshida N, Yoshida S, Koishi K, Masuda K, Nabeshima Y: Cell p130 might block myoblasts in a phase of the cell cycle that heterogeneity upon myogenic differentiation: down-regulation of is not permissive for MyoD activity. Consistent with this MyoD and Myf-5 generates “reserve cells”. J Cell Sci 1998, hypothesis, we have previously shown that MyoD is absent 111:769-779. 4. Mulligan G, Jacks T: The retinoblastoma gene family: cousins with from G0-synchronised myoblasts [1]. The observation that overlapping interests. Trends Genet 1998, 14:223-228. in non-muscle cells, the formation of an E2F–p130 complex 5. Gu W, Schneider JW, Condorelly G: Interaction of myogenic factors and the mediates muscle cell is unique to cells in G0 stage [16] reinforces our previous commitment and differentiation. Cell 1993, 72:309-324. hypothesis [1] that myoblasts exit into differentiation at a 6. Schneider JW, GU W, Zhu L, Mahdavi V, Nadal-Ginard B: Reversal point in G1 distinct from this G0-quiescent state typical of of terminal differentiation mediated by p107 in Rb–/– muscle cells. Science 1994, 264:1467-1471. reserve cells. Unlike p130, pRb enhances MyoD activity 7. Zacksenhaus E, Jiang Z, Ghung D, Marth JD, Phillips RA, Gallie BL: and, as shown here, relieves p130 inhibition of MyoD activ- pRb controls proliferation, differentiation, and death of skeletal muscle cells and other lineages during embryogenesis. Genes ity. The biochemical basis of this Rb/p130 functional oppo- Dev 1996, 10:3051-3064. sition remains unclear, because Rb appears to promote 8. Catala F, Wanner R, Barton P, Cohen A, Wright W, Buckingham M: muscle differentiation through various mechanisms includ- A skeletal muscle specific regulated by factor binding to E and CArG boxes is present in the promoter of the mouse ing direct physical interaction with MyoD [5], cooperation myosin light-chain 1A gene. Mol Cell Biol 1995, 15:4585-4596. with MyoD to induce MEF2A transcriptional activity [17] 9. Corbeil HB, Whyte P, Branton PE: Characterisation of transcription and/or modulation of Ras activity [18]. factor E2F complexes during muscle and neuronal differentiation. Oncogene 1995, 11:909-920. 10. Shin EK, Shin A, Paudling C, Schaffhausen, Yee AS: Multiple Given that increased p130 expression is incompatible with changes in E2F function and regulation occur upon muscle differentiation. Mol Cell Biol 1995, 15:2252-2262. differentiation, it is puzzling to detect significant levels of 11. Weintraub, H: The MyoD family and : redundancy, p130 protein in myotubes complexed to E2F. However, we networks, and thresholds. Cell 1993, 75:1241-1244. postulate that this fraction of protein can be inactivated as 12. Thayer MJ, Tapscott SJ, Davis RL, Wright WE, Lassar AB, Weintraub H: Positive autoregulation of the myogenic differentiation proceeds and/or the myogenic inhibitory determination gene MyoD1. Cell 1989, 58:241-248. effect of p130 is distinct from its ability to bind E2F. This 13. Dominov J, Dunn JJ, Miller JB: Bcl-2 expression identifies an early stage of myogenesis and promotes clonal expansion of muscle last hypothesis is reinforced by recent observations realised cells. J Cell Biol 1998, 142:537-542. on pRb. Using various Rb mutant proteins, Sellers et al. [19] 14. Shih HH, Tevosian SG, Yee AS: Regulation of differentiation by found that binding of pRb to E2F is required for pRb to HBP1, a target of the retinoblastoma protein. Mol Cell Biol 1998, 18:4732-4743. induce acute G1/S block but not for it to promote induction 15. Datta B, Min W, Burma S, Lengyel P: Increase in p202 expression of differentiation. This would imply that the differentia- during skeletal muscle differentiation: inhibition of MyoD protein expression and activity of p202. Mol Cell Biol 1998, 18:1074-1083. tion-blocking activity of p130 lies in a region of p130 dis- 16. Smith EJ, Leone G, DeGregori J, Jakoi L, Nevins JR: The tinct from its pocket domain (a region highly homologous accumulation of an E2F-p130 transcriptional repressor between Rb and p130 and involved in binding to E2F). distinguishes a G0 cell state from a G1 cell state. Mol Cell Biol 1996, 16:6965-6976. 17. Novitch BG, Spicer DB, Kim PS, Cheung WL, Lassar AB: pRb is Mice lacking the p130 gene showed varying degrees of required for -dependent gene expression as well as cell- cycle arrest during skeletal muscle differentiation. Curr Biol 1999, tissue disorganisation and, notably, a reduced number of 9:449-459. skeletal muscle cells [20]. This phenotype is difficult to 18. Lee KY, Ladha MH, McMahon C, Ewen ME: The retinoblastoma protein is linked to the activation of Ras. Mol Cell Biol 1999, interpret, as tissues adjacent to myotome known to have a 19:7724-7732. positive influence on muscle cell behaviour, such as the 19. Sellers WR, Novitch BG, Miyake S, Heith A, Otterson GA, Kaye FJ, neural tube, are also deeply affected. In the light of our et al.: Stable binding to E2F is not required for the retinoblastoma protein to activate transcription, promote differentiation, and results, however, it appears consistent that inactivation of suppress tumor . Genes Dev 1998, 12:95-106. p130, by reducing the pool of reserve cells, will lead to 20. Lecouter JE, Kablar B, Whyte PFM, Ying C, Rudnicki M: Strain- dependent embryonic lethality in mice lacking the fewer skeletal muscle fibres. retinoblastoma-related p130 gene. Development 1998, 125:4669-4679. 21. Le Cam L, Polanowska J, Fabbrizio E, Olivier M, Philips A, Eaton E, Supplementary material et al.: Timing of E gene expression depends on the Supplementary material including additional methodological details is regulated association of a bipartite repressor element with a available at http://current-biology.com/supmat/supmatin.htm. novel E2F complex. EMBO J 1999, 18:1878-1890.