RNAi-mediated HuR Depletion Leads to the Inhibition of Muscle Cell Differentiation
2003; Elsevier BV; Volume: 278; Issue: 47 Linguagem: Inglês
10.1074/jbc.m308889200
ISSN1083-351X
AutoresKate van der Giessen, Sergio Di Marco, Eveline Clair, Imed‐Eddine Gallouzi,
Tópico(s)RNA and protein synthesis mechanisms
ResumoThe formation of muscle fibers involves the sequential expression of many proteins that regulate key steps during myoblast-to-myotube transition. MyoD, myogenin, and the cyclin-dependent kinase inhibitor p21cip1 are major players in the initiation and maintenance of the differentiated state of mouse embryonic muscle cells (C2C12). The messenger RNAs encoding these three proteins contain typical AU-rich elements (AREs) in their 3′-untranslated regions (3′-UTRs), which are known to affect the half-life of many short-lived mRNAs. HuR, an RNA-binding protein that regulates both the stability and cellular movement of ARE-containing mRNAs, interacts and stabilizes the p21cip1 message under UV stress in human RKO colorectal carcinoma cells. Here, by the use of gel shift experiments and immunoprecipitation followed by reverse transcription-PCR analysis, we show that HuR interacts with MyoD, myogenin, and p21cip1 mRNAs through specific sequences in their 3′-UTRs. To demonstrate the implication of endogenous HuR in myogenesis, we knocked down its expression in myoblasts using RNA interference and observed a significant reduction of HuR expression, associated with complete inhibition of myogenesis. Moreover, the expression of MyoD and myogenin mRNAs, as well as proteins, is significantly reduced in the HuR knockdown C2C12 cells. We were able to completely re-establish the myogenic process of these defective cells by introducing back HuR protein conjugated to a cell-permeable peptide. Finally, HuR accumulates in the cytoplasm during myogenesis. Thus, our results clearly demonstrated that endogenous HuR plays a crucial role in muscle differentiation by regulating the expression and/or the nuclear export of ARE-containing mRNAs that are essential for this process. The formation of muscle fibers involves the sequential expression of many proteins that regulate key steps during myoblast-to-myotube transition. MyoD, myogenin, and the cyclin-dependent kinase inhibitor p21cip1 are major players in the initiation and maintenance of the differentiated state of mouse embryonic muscle cells (C2C12). The messenger RNAs encoding these three proteins contain typical AU-rich elements (AREs) in their 3′-untranslated regions (3′-UTRs), which are known to affect the half-life of many short-lived mRNAs. HuR, an RNA-binding protein that regulates both the stability and cellular movement of ARE-containing mRNAs, interacts and stabilizes the p21cip1 message under UV stress in human RKO colorectal carcinoma cells. Here, by the use of gel shift experiments and immunoprecipitation followed by reverse transcription-PCR analysis, we show that HuR interacts with MyoD, myogenin, and p21cip1 mRNAs through specific sequences in their 3′-UTRs. To demonstrate the implication of endogenous HuR in myogenesis, we knocked down its expression in myoblasts using RNA interference and observed a significant reduction of HuR expression, associated with complete inhibition of myogenesis. Moreover, the expression of MyoD and myogenin mRNAs, as well as proteins, is significantly reduced in the HuR knockdown C2C12 cells. We were able to completely re-establish the myogenic process of these defective cells by introducing back HuR protein conjugated to a cell-permeable peptide. Finally, HuR accumulates in the cytoplasm during myogenesis. Thus, our results clearly demonstrated that endogenous HuR plays a crucial role in muscle differentiation by regulating the expression and/or the nuclear export of ARE-containing mRNAs that are essential for this process. The first step of skeletal muscle formation (also named myogenesis) is the cell cycle arrest of myoblasts, which leads to cell fusion and the formation of multinucleated myotubes. This transition is controlled by muscle-specific transcriptional regulators that respond to external signals to couple myogenesis to the development and growth of the organism (1Polesskaya A. Rudnicki M.A. Dev. Cell. 2002; 3: 757-758Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar). MyoD, Myf-5, myogenin, and MRF4 represent the main myogenic transcription factors and belong to the MRF (myogenic regulatory factor) family. MyoD and Myf-5 are expressed in proliferating myoblasts, whereas myogenin expression is induced only upon muscle differentiation (2Sabourin L.A. Rudnicki M.A. Clin. Genet. 2000; 57: 16-25Crossref PubMed Scopus (545) Google Scholar). MyoD activates the cyclin-dependent kinase inhibitor p21cip1 (3Guo K. Wang J. Andres V. Smith R.C. Walsh K. Mol. Cell Biol. 1995; 15: 3823-3829Crossref PubMed Scopus (359) Google Scholar), which, along with p57 (another Cip1/Kip1 family member), leads to cell cycle arrest (4Reynaud E.G. Pelpel K. Guillier M. Leibovitch M.P. Leibovitch S.A. Mol. Cell Biol. 1999; 19: 7621-7629Crossref PubMed Scopus (87) Google Scholar). It is widely accepted that p21cip1 induction follows myogenin expression during the myoblast-to-myotube transition and that a high level of p21cip1 is required for myotube maintenance (5Odelberg S.J. Kollhoff A. Keating M.T. Cell. 2000; 103: 1099-1109Abstract Full Text Full Text PDF PubMed Scopus (370) Google Scholar). C2C12 myoblasts (a mouse embryonic muscle cell line) provide a powerful model to study skeletal muscle differentiation (6Yamaguchi A. Semin. Cell Biol. 1995; 6: 165-173Crossref PubMed Scopus (140) Google Scholar). They overexpress the myogenic factors described above and fuse to form muscle fibers when induced to differentiate. This overexpression is due not only to an increase in the transcription of their corresponding genes, but also to the stabilization of their mRNAs (7Guttridge D.C. Mayo M.W. Madrid L.V. Wang C.Y. Baldwin Jr., A.S. Science. 2000; 289: 2363-2366Crossref PubMed Scopus (740) Google Scholar, 8Bisbal C. Silhol M. Laubenthal H. Kaluza T. Carnac G. Milligan L. Le Roy F. Salehzada T. Mol. Cell Biol. 2000; 20: 4959-4969Crossref PubMed Scopus (76) Google Scholar). Therefore, preventing the degradation of mRNAs that encode key factors of muscle cell differentiation is crucial for maintaining the integrity of the whole process. However, the mechanisms that affect the turnover and the cellular movement of these messages remain elusive. MyoD, myogenin, and p21cip1 mRNAs each harbor an AU-rich element (ARE) 1The abbreviations used are: AREAU-rich elementUTRuntranslated regionAUBPAU-binding proteinELAVembryonic lethal, abnormal visionAPantennapediaGSTglutathione S-transferaseDMEMDulbecco's modified Eagle's mediumsiRNAsmall interference RNARTreverse transcriptionHCHuR-associated complexIREiron-responsive elementDAPI4′,6-diamidino-2-phenylindolehnRNPheterogeneous nuclear ribonucleoproteinRNAiRNA interference.1The abbreviations used are: AREAU-rich elementUTRuntranslated regionAUBPAU-binding proteinELAVembryonic lethal, abnormal visionAPantennapediaGSTglutathione S-transferaseDMEMDulbecco's modified Eagle's mediumsiRNAsmall interference RNARTreverse transcriptionHCHuR-associated complexIREiron-responsive elementDAPI4′,6-diamidino-2-phenylindolehnRNPheterogeneous nuclear ribonucleoproteinRNAiRNA interference. in their 3′-untranslated regions (3′-UTRs) (9Figueroa A. Cuadrado A. Fan J. Atasoy U. Muscat G.E. Munoz-Canoves P. Gorospe M. Munoz A. Mol. Cell Biol. 2003; 23: 4991-5004Crossref PubMed Scopus (154) Google Scholar). These AREs contain one or two AU3A sequences that trigger the rapid degradation of short-lived mRNAs, such as many oncogenes, lymphokines, and cytokines (10Shaw G. Kamen R. Cell. 1986; 46: 659-667Abstract Full Text PDF PubMed Scopus (3103) Google Scholar, 11Jacobson A. Peltz S.W. Annu. Rev. Biochem. 1996; 65: 693-739Crossref PubMed Scopus (575) Google Scholar). AREs do not always target mRNAs for degradation (12Bakheet T. Frevel M. Williams B.R. Greer W. Khabar K.S. Nucleic Acids Res. 2001; 29: 246-254Crossref PubMed Scopus (340) Google Scholar) but may also serve as an anchor for some of the shuttling AU-binding proteins (AUBPs) that monitor the nuclear export of ARE-containing mRNAs (13Brennan C.M. Steitz J.A. Cell. Mol. Life Sci. 2001; 58: 266-277Crossref PubMed Scopus (870) Google Scholar, 14Gallouzi I.E. Steitz J.A. Science. 2001; 294: 1895-1901Crossref PubMed Scopus (233) Google Scholar). Therefore, it is conceivable that the half-lives and cellular movement of MyoD, myogenin, and p21cip1 mRNAs may be regulated through their AREs in association with AUBPs. AU-rich element untranslated region AU-binding protein embryonic lethal, abnormal vision antennapedia glutathione S-transferase Dulbecco's modified Eagle's medium small interference RNA reverse transcription HuR-associated complex iron-responsive element 4′,6-diamidino-2-phenylindole heterogeneous nuclear ribonucleoprotein RNA interference. AU-rich element untranslated region AU-binding protein embryonic lethal, abnormal vision antennapedia glutathione S-transferase Dulbecco's modified Eagle's medium small interference RNA reverse transcription HuR-associated complex iron-responsive element 4′,6-diamidino-2-phenylindole heterogeneous nuclear ribonucleoprotein RNA interference. HuR is a well-studied RNA-binding protein and AUBP that specifically interacts and stabilizes AU3A-containing mRNAs (15Fan X.C. Steitz J.A. EMBO J. 1998; 17: 3448-3460Crossref PubMed Scopus (740) Google Scholar, 16Myer V.E. Fan X.C. Steitz J.A. EMBO J. 1997; 16: 2130-2139Crossref PubMed Scopus (279) Google Scholar, 17Levy N.S. Chung S. Furneaux H. Levy A.P. J. Biol. Chem. 1998; 273: 6417-6423Abstract Full Text Full Text PDF PubMed Scopus (569) Google Scholar, 18Ma W.J. Cheng S. Campbell C. Wright A. Furneaux H. J. Biol. Chem. 1996; 271: 8144-8151Abstract Full Text Full Text PDF PubMed Scopus (566) Google Scholar, 19Peng S.S. Chen C.Y. Xu N. Shyu A.B. EMBO J. 1998; 17: 3461-3470Crossref PubMed Scopus (650) Google Scholar). It is a 32-kDa protein that belongs to the Hu/ELAV family (embryonic lethal, abnormal vision (20Campos A.R. Grossman D. White K. J. Neurogenet. 1985; 2: 197-218Crossref PubMed Scopus (161) Google Scholar, 21Robinow S. Campos A.R. Yao K.M. White K. Science. 1988; 242: 1570-1572Crossref PubMed Scopus (248) Google Scholar)) of RNA-binding proteins. Unlike the other three ELAV proteins, HuB, HuC, and HuD (which are expressed primarily in the brain), HuR is ubiquitously expressed (22Good P.J. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 4557-4561Crossref PubMed Scopus (271) Google Scholar) and functions as an adaptor protein for the nuclear export of many ARE-containing mRNAs (14Gallouzi I.E. Steitz J.A. Science. 2001; 294: 1895-1901Crossref PubMed Scopus (233) Google Scholar, 23Gallouzi I.E. Brennan C.M. Steitz J.A. RNA (N. Y.). 2001; 7: 1348-1361Crossref PubMed Scopus (136) Google Scholar). The elav locus was first discovered as a gene that is essential for the development and maintenance of the nervous system of Drosophila (20Campos A.R. Grossman D. White K. J. Neurogenet. 1985; 2: 197-218Crossref PubMed Scopus (161) Google Scholar, 21Robinow S. Campos A.R. Yao K.M. White K. Science. 1988; 242: 1570-1572Crossref PubMed Scopus (248) Google Scholar). Recently, it has been shown that the ELAV human counterparts, HuB, HuC, and HuD (22Good P.J. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 4557-4561Crossref PubMed Scopus (271) Google Scholar), are involved in mouse embryonic development (24Okano H.J. Darnell R.B. J. Neurosci. 1997; 17: 3024-3037Crossref PubMed Google Scholar) and the differentiation of a variety of embryonic cell lines (25Wakamatsu Y. Weston J.A. Development. 1997; 124: 3449-3460Crossref PubMed Google Scholar, 26Jain R.G. Andrews L.G. McGowan K.M. Pekala P.H. Keene J.D. Mol. Cell Biol. 1997; 17: 954-962Crossref PubMed Scopus (182) Google Scholar, 27Antic D. Lu N. Keene J.D. Genes Dev. 1999; 13: 449-461Crossref PubMed Scopus (176) Google Scholar, 28Akamatsu W. Okano H.J. Osumi N. Inoue T. Nakamura S. Sakakibara S. Miura M. Matsuo N. Darnell R.B. Okano H. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 9885-9890Crossref PubMed Scopus (205) Google Scholar, 29Kasashima K. Terashima K. Yamamoto K. Sakashita E. Sakamoto H. Genes Cells. 1999; 4: 667-683Crossref PubMed Scopus (97) Google Scholar). The first demonstration that an Hu/ELAV protein is involved in differentiation came from the work of Jain and coworkers (26Jain R.G. Andrews L.G. McGowan K.M. Pekala P.H. Keene J.D. Mol. Cell Biol. 1997; 17: 954-962Crossref PubMed Scopus (182) Google Scholar). They showed that the ectopic overexpression of HuB in the preadipocyte cell line 3T3 L1 leads to the enhancement of adipocyte differentiation and the stabilization, as well as the translation induction of the glucose transporter 1 mRNA. Furthermore, HuB induces neurite formation of human embryonic tetracarcinoma (hNT2) cells by stabilizing and/or regulating the translation of the neurofilamin mRNA (27Antic D. Lu N. Keene J.D. Genes Dev. 1999; 13: 449-461Crossref PubMed Scopus (176) Google Scholar). Together, these observations suggest that the tissue-specific ELAV proteins, through their ability to regulate the stability and/or the translation of many mRNAs, play a crucial role in cell differentiation (30Keene J.D. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 5-7Crossref PubMed Scopus (257) Google Scholar). Therefore, it is possible that HuR, which is ubiquitously expressed, could have similar effects on the differentiation of other embryonic cell lines. Establishing this role for endogenous HuR would be unexpected and of high interest, because these regulatory functions are usually associated with tissue-specific proteins. However, there are many examples in the literature suggesting that many RNA-binding proteins may have different functions in different tissues (31Weighardt F. Biamonti G. Riva S. Bioessays. 1996; 18: 747-756Crossref PubMed Scopus (180) Google Scholar, 32Siomi H. Dreyfuss G. Curr. Opin. Genet. Dev. 1997; 7: 345-353Crossref PubMed Scopus (231) Google Scholar, 33Singh O.P. Indian J. Biochem. Biophys. 2001; 38: 129-134PubMed Google Scholar, 34Lambermon M.H. Fu Y. Kirk D.A. Dupasquier M. Filipowicz W. Lorkovic Z.J. Mol. Cell Biol. 2002; 22: 4346-4357Crossref PubMed Scopus (62) Google Scholar, 35Reed R. Hurt E. Cell. 2002; 108: 523-531Abstract Full Text Full Text PDF PubMed Scopus (331) Google Scholar, 36Dreyfuss G. Kim V.N. Kataoka N. Nat. Rev. Mol. Cell Biol. 2002; 3: 195-205Crossref PubMed Scopus (1096) Google Scholar). These different effects are due to a variation in their expression or in their regulation, through protein-protein associations or post-translational modifications, which will modulate the association with their end targets (such as an mRNA). The stabilizing function of HuR in mammalian cells was only demonstrated in experiments whereby cells were co-transfected with the recombinant HuR protein as well as an artificial ARE-containing message (15Fan X.C. Steitz J.A. EMBO J. 1998; 17: 3448-3460Crossref PubMed Scopus (740) Google Scholar, 17Levy N.S. Chung S. Furneaux H. Levy A.P. J. Biol. Chem. 1998; 273: 6417-6423Abstract Full Text Full Text PDF PubMed Scopus (569) Google Scholar, 19Peng S.S. Chen C.Y. Xu N. Shyu A.B. EMBO J. 1998; 17: 3461-3470Crossref PubMed Scopus (650) Google Scholar). Thus, the question of whether endogenous HuR exercises the same function in vivo remains unanswered. Recently, it was observed that the overexpression of HuR in myoblasts correlates with the stabilization of MyoD, myogenin, and p21cip1 messages, as well as with the acceleration of muscle cell differentiation (9Figueroa A. Cuadrado A. Fan J. Atasoy U. Muscat G.E. Munoz-Canoves P. Gorospe M. Munoz A. Mol. Cell Biol. 2003; 23: 4991-5004Crossref PubMed Scopus (154) Google Scholar). Because this effect was observed only upon the overexpression of exogenous HuR, we still do not know whether the observed stabilization and differentiation stimulation reflect the true function of endogenous HuR during myogenesis. Therefore, the best approach to define the implication of HuR in this process is to disturb its expression in myoblasts by RNA interference, then stimulate muscle differentiation and monitor the effect on myotube formation. HuR shuttles between the nuclear and cytoplasmic compartments by virtue of its HNS (HuR nucleocytoplasmic shuttling domain (37Fan X.C. Steitz J.A. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 15293-15298Crossref PubMed Scopus (395) Google Scholar)). We recently used a novel approach for delivering protein domains to the interior of mammalian cells to delineate the shuttling pathways of HuR (14Gallouzi I.E. Steitz J.A. Science. 2001; 294: 1895-1901Crossref PubMed Scopus (233) Google Scholar). Specifically, the HNS and the CRM1-dependent nuclear export signal (38Fukuda M. Asano S. Nakamura T. Adachi M. Yoshida M. Yanagida M. Nishida E. Nature. 1997; 390: 308-311Crossref PubMed Scopus (1015) Google Scholar) domains were fused to the cell-permeable peptide antennapedia (AP). AP facilitates the cellular uptake of such chimeras into mammalian cells via a non-endocytic, non-degradative pathway with high efficiency (>90%). Using this approach, we demonstrated that, unlike other known shuttling proteins, HuR utilizes two pathways to export its mRNA targets from the nucleus to the cytoplasm (14Gallouzi I.E. Steitz J.A. Science. 2001; 294: 1895-1901Crossref PubMed Scopus (233) Google Scholar). However, these findings (14Gallouzi I.E. Steitz J.A. Science. 2001; 294: 1895-1901Crossref PubMed Scopus (233) Google Scholar) did not address whether the stabilizing function of HuR and/or its nucleocytoplasmic movement may affect vital processes such as cell growth and differentiation. To address these issues, C2C12 myoblasts were used to define the role of HuR during cell differentiation. Our results demonstrate that HuR plays an essential role in myogenesis and regulates the expression of mRNAs that encode key myogenic factors. Plasmid Construction and Protein Purification—The AP PCR product was produced by self-amplification using AP-N-sense (5′-ggg gac aag ttt gta caa aaa agc agg ctt cga agg aga tag aac cat gcg tca aat taa gat ttg gtt cca gaa ccg tcg cat g-3′) and AP-N-antisense (5′-ggg gac cac ttt gta caa gaa agc tgg gtc cat atg aag ctt aga tat ctc gag tgc ggc cgc ttt ctt cca ttt cat gcg acg gtt ctg-3′) oligonucleotides. To create the AP-C entry vector, a binding protein recombination reaction between pDONR1 (Invitrogen) and the AP PCR product was performed using the Gateway system (Invitrogen). PCR amplification of HuR was performed using the GST-HuR plasmid as a template (39Brennan C.M. Gallouzi I.E. Steitz J.A. J. Cell Biol. 2000; 151: 1-14Crossref PubMed Scopus (312) Google Scholar) with the primers HuR-TAG-For (5′-ccc ctc gag tct aat ggt tat gaa gac cac-3′) and AP-HuR-Rev (5′-ggg aag ctt tta ttt gtg gga ctt gtt gg-3′). The AP-HuR-GST plasmid was created by recombination between the AP-HuR-TAG entry vector and the pDEST15 plasmid (Invitrogen). The XhoI/HindIII fragment of the HuR PCR product was inserted into the XhoI/HindIII sites of the AP-N entry vector to produce the AP-HuR entry vector. The GST-AP plasmid was generated by recombination reaction between the AP-C entry vector and pDEST15 (Invitrogen). The fusion proteins were purified as described (39Brennan C.M. Gallouzi I.E. Steitz J.A. J. Cell Biol. 2000; 151: 1-14Crossref PubMed Scopus (312) Google Scholar) with the following modifications. The proteins were eluted from the glutathione-agarose beads with three applications of 500-μl glutathione elution buffer (10 mm for the first elution, and 20 mm for the second and third elutions). Proteins were then dialyzed overnight against phosphate-buffered saline at 4 °C. The 20 mm glutathione eluates were the most pure (as determined by SDS-PAGE) and were used in all experiments. Cell Culture and Transfection—C2C12 cells (ATCC) were grown and maintained in Dulbecco's modified eagle medium (DMEM, Invitrogen) containing 20% fetal bovine serum (Invitrogen), penicillin/streptomycin, and l-glutamine, following the manufacturer's directions (Invitrogen). Differentiation was induced immediately upon 100% confluency on plates previously coated with 0.1% gelatin. To induce differentiation, growth media were replaced with differentiation media containing DMEM, 2% horse serum, penicillin/streptomycin antibiotics, and 10 μg/ml insulin (Invitrogen), 10 μg/ml transferrin (Invitrogen), and 50 mm HEPES, pH 7.4 (Invitrogen). The transfection of siRNA into C2C12 cells was performed in six-well plates using twice the number of wells that were eventually needed. The first transfection with HuSi-1, HuSi-C, or mock (transfection reagents only, no oligonucleotide) was performed when cells were 20-30% confluent. At 24 h post-transfection, when cells were 50-60% confluent, the procedure was repeated. At 6-8 h after the second transfection, two wells (with the same siRNA treatment) were combined into one by trypsinizing one plate and moving the cells from each well to corresponding wells on the second plate. For all transfections, LipofectAMINE Plus (Invitrogen) was used, following the manufacturer's protocol. siRNA duplexes were ordered from Dharmacon. The siRNA duplex (0.12 μm) was used for each well of a six-well plate. Differentiation was induced 2 days after the second transfection. For rescue of siRNA-treated cells, 25 nm purified protein was added to the C2C12 growth media. This protein/media solution was added to the cells twice, for 2 days, following the second siRNA transfection. Differentiation was induced the day after the protein was added for the second time. Immunoblotting, Immunofluorescence, and Preparation of Cell Extracts—Total cell extracts, as well as nuclear and cytoplasmic fractions, were prepared as described (40Gallouzi I.E. Parker F. Chebli K. Maurier F. Labourier E. Barlat I. Capony J.P. Tocque B. Tazi J. Mol. Cell Biol. 1998; 18: 3956-3965Crossref PubMed Scopus (167) Google Scholar, 41Di Marco S. Hel Z. Lachance C. Furneaux H. Radzioch D. Nucleic Acids Res. 2001; 29: 863871Crossref Google Scholar). Western blotting was performed as described previously (39Brennan C.M. Gallouzi I.E. Steitz J.A. J. Cell Biol. 2000; 151: 1-14Crossref PubMed Scopus (312) Google Scholar). The blots were probed variously with antibodies to HuR (42Gallouzi I.E. Brennan C.M. Stenberg M.G. Swanson M.S. Eversole A. Maizels N. Steitz J.A. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 3073-3078Crossref PubMed Scopus (272) Google Scholar), hnRNP A1 (kindly provided by Dr. G. Dreyfuss, University of Pennsylvania School of Medicine, Philadelphia), myogenin, MyoD, tubulin (Santa Cruz Biotechnology), myoglobin (DAKO), MF-20 (Developmental Studies Hybridoma Bank), and β-actin (Sigma). Immunofluorescence was performed as previously described (37Fan X.C. Steitz J.A. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 15293-15298Crossref PubMed Scopus (395) Google Scholar). A polyclonal antibody to myoglobin was used at a 1:500 dilution in 1% goat serum/phosphate-buffered saline, and was incubated simultaneously with a monoclonal antibody to HuR (1:1500). Myoglobin was detected using a rhodamine-labeled secondary antibody, and HuR using an fluorescein isothiocyanate secondary antibody, both at a 1:500 dilution. Hoechst stain, 1:1500, was also incubated with the secondary antibodies. A Zeiss Axiovision 3.1 microscope was used to observe the cells using a 40× oil objective, and an Axiocam HR (Zeiss) digital camera was used for immunofluorescence photography. Northern (RNA) Blot Analysis—Northern blot analysis was performed as described (43Wojciechowski W. DeSanctis J. Skamene E. Radzioch D. J. Immunol. 1999; 163: 2688-2696PubMed Google Scholar) using 7 μg of total RNA prepared using TRIzol reagent (Invitrogen). After transferring to a Hybond-N membrane (Amersham Biosciences) and UV-cross-linking, the blot was hybridized with murine p21cip1, MyoD, or myogenin cDNA probes generated by random primer labeling (Roche Applied Science) according to the manufacturer's instructions (44Rios R. Carneiro I. Arce V.M. Devesa J. Biochem. Biophys. Res. Commun. 2001; 280: 561-566Crossref PubMed Scopus (127) Google Scholar). PCR-amplified fragments from MyoD, myogenin, and p21cip1 cDNAs (kindly supplied by Dr. A. Lassar, at Harvard Medical School) were used to generate labeled probes. After hybridization, the membranes were washed and subsequently exposed on BioMax films. Preparation of mRNA (mRNP) Complexes and Analysis with RTPCR—Immunoprecipitation and RNA preparation were performed as previously described (45Tenenbaum S.A. Lager P.J. Carson C.C. Keene J.D. Methods. 2002; 26: 191-198Crossref PubMed Scopus (234) Google Scholar) using antibodies to HuR (3A2) and MF-20, and cell extracts were prepared from differentiating C2C12 cells. mRNP lysate (400 μl) was added to beads (protein-A-Sepharose) for each immunoprecipitation. Specific messages associated with HuR were defined using RT-PCR as described below. Purified RNA was resuspended in 10 μl of water, and 1 μl was reverse-transcribed using the Thermoscript RT-PCR system (Invitrogen) according to the manufacturer's protocol in 20 μl of final volume. Subsequently, 2 μl of cDNA was PCR-amplified using platinum TaqDNA polymerase (Invitrogen) for 22 cycles using MyoD, myogenin, or p21cip1 cDNA-specific primers. The sequences of the primers, as well as the PCR conditions, were described previously (44Rios R. Carneiro I. Arce V.M. Devesa J. Biochem. Biophys. Res. Commun. 2001; 280: 561-566Crossref PubMed Scopus (127) Google Scholar). Preparation of RNA Transcripts, RNA Binding Assay, and Gel Shift Assay—The MyoD, myogenin, and p21cip1 cRNA probes were produced by in vitro transcription as previously described (40Gallouzi I.E. Parker F. Chebli K. Maurier F. Labourier E. Barlat I. Capony J.P. Tocque B. Tazi J. Mol. Cell Biol. 1998; 18: 3956-3965Crossref PubMed Scopus (167) Google Scholar). The p21cip1 cRNA probe was transcribed from a synthesized oligonucleotide subcloned into pGEM 3Zf(+) vector (Promega), which was linearized with BamH1 prior to the reaction. The myogenin cRNA probe was generated from a synthetic oligonucleotide (spanning nucleotides 1244-1279 of myogenin 3′-UTR) fused to a T7 promoter (5′-gaattgtaatacgactcactatagggcga-3′). The MyoD cRNA probe was generated by PCR amplification of a region of the MyoD 3′-UTR spanning nucleotides 1301-1408 using a forward primer fused to the T7 promoter as well as a MyoD cDNA expression vector as the template. The RNA binding assay and the HuR supershift assays were performed as previously described (41Di Marco S. Hel Z. Lachance C. Furneaux H. Radzioch D. Nucleic Acids Res. 2001; 29: 863871Crossref Google Scholar). HuR Expression during C2C12 Differentiation—To investigate HuR function in regulating gene expression during muscle differentiation, mouse embryonic myoblasts (C2C12, also termed C2, myoblasts) were used as a system model. Myotubes and differentiation markers such as the myosin heavy chain (MF-20) and myoglobin become detectable in these cells ∼48-72 h after serum starvation (46Bader D. Masaki T. Fischman D.A. J. Cell Biol. 1982; 95: 763-770Crossref PubMed Scopus (794) Google Scholar, 47Torgan C.E. Daniels M.P. Mol. Biol. Cell. 2001; 12: 1499-1508Crossref PubMed Scopus (45) Google Scholar). HuR expression was first monitored during myogenesis (Fig. 1). Western blot analysis using total extracts from differentiating myoblasts and a monoclonal antibody against HuR demonstrated that HuR expression levels did not change throughout myogenesis and remained high in mature myotubes (Fig. 1, B, lanes 4 and 5, and A, lanes d3 and d5). Both light microscopy and Western blots showed that the C2C12 cells differentiated normally and expressed normal levels of myoglobin and MF-20 (Fig. 1, A and B) as previously described (48Yan Z. Serrano A.L. Schiaffino S. Bassel-Duby R. Williams R.S. J. Biol. Chem. 2001; 276: 17361-17366Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar). These results show that HuR protein levels do not change during the myoblast-to-myotube transition. HuR Associates with MyoD, Myogenin, and p21cip1 Messages during C2C12 Differentiation through Specific Sequences Located in Their 3′-UTR—Nucleotide sequence analysis of MyoD, myogenin, and p21cip1 indicated that each mRNA contains an AU/GU-rich sequences in its 3′-UTR (Ref. 9Figueroa A. Cuadrado A. Fan J. Atasoy U. Muscat G.E. Munoz-Canoves P. Gorospe M. Munoz A. Mol. Cell Biol. 2003; 23: 4991-5004Crossref PubMed Scopus (154) Google Scholar and Fig. 2A). Moreover, it was recently shown that HuR associates with the 3′-UTR of myogenin and p21cip1 during C2C12 differentiation (9Figueroa A. Cuadrado A. Fan J. Atasoy U. Muscat G.E. Munoz-Canoves P. Gorospe M. Munoz A. Mol. Cell Biol. 2003; 23: 4991-5004Crossref PubMed Scopus (154) Google Scholar). However, the association with the MyoD 3′-UTR, as well as the precise sequences that interact with HuR, remain unknown. To delineate the exact HuR binding sites in these messages, we performed RNA gel shift assays, using radiolabeled probes for the 3′-UTR of p21cip1, MyoD, and myogenin, on nuclear and cytoplasmic extracts prepared from differentiating C2C12 cells (Fig. 2B). The HuR antibody was incubated with the extracts in the presence of p21cip1 (Fig. 2A, c), MyoD (Fig. 2A, a), or myogenin (Fig. 2A, b) cRNA probes. With the p21cip1 cRNA probe (Fig. 2B, lanes 20-27), a complex was evident in both the cytoplasmic (Cc) (Fig. 2B, lanes 21-23) and the nuclear (data not shown) extracts. When the HuR antibody was added to these extracts, a shift of an HuR-associated complex (HC) was observed throughout differentiation (Fig. 2B, lanes 24-26). This shift was absent in both nuclear and cytoplasmic extracts incubated with an antibody to p38 (Fig. 2B, lane 27). HuR therefore interacts in vitro with the p21cip1 ARE in both the nuclear (data not shown) and cytoplasmic compartments during C2C12 differentiation. The same result was obtained with the myogenin cRNA probe (spanning nucleotides 1244-1279
Referência(s)