Myostatin and NF-B Regulate Skeletal Myogenesis Through Distinct Signaling Pathways

  • Published on
    06-Jul-2016

  • View
    217

  • Download
    1

Transcript

  • 202

    Myostatin and NF-B Regulate Skeletal MyogenesisThrough Distinct Signaling Pathways

    Nadine Bakkara, Henning Wackerhageb, and Denis C. Guttridgea,c

    a Division of Human Cancer Genetics, Department of Molecular Virology, Immunology and Medi-cal Genetics, Molecular Cellular and Developmental Biology Graduate Program, The Ohio StateUniversity, Columbus, Ohio, USAb Division of Molecular Physiology, University of Dundee, UKc The Arthur G. James Comprehensive Cancer Center, The Ohio State University, Columbus,Ohio, USA

    Myostatin (Mstn) is a potent negative regulator of skeletal development shown to inhibit myoblastproliferation by impinging on cell cycle and suppressing the synthesis of MyoD. Moreover, Mstncauses muscle wasting and its expression is linked with several conditions of muscle loss, mainlydystrophy and cachexia. NF-B is a transcription factor that is constitutively active in proliferatingmyoblasts and also plays a role in cell growth control and skeletal muscle differentiation. NF-Binhibits myogenesis by promoting myoblast growth and inducing loss of MyoD, and NF-B activityis required in states of muscle wasting. However, the extracellular factors that regulate NF-Bactivity to modulate myogenesis are currently not known. Given the similarities in Mstn and NF-Bactivities in muscle cells, we investigated the possibility that Mstn-induced regulation of myogenesismay signal via NF-B. Using a variety of assays to monitor for NF-B activity, we found that Mstnsignaling does not activate NF-B in differentiating C2C12 myoblasts, nor is the constitutive activityof NF-B required for Mstn-mediated inhibition of myogenesis. Likewise, in pre-differentiated my-otubes, Mstn signaling induces only a modest activation of NF-B DNA binding activity. We alsoinvestigated whether NF-B inhibition of myogenesis may occur through the regulation of Mstn.However, activation of NF-B by TNF or IL-1 failed to induce Mstn expression. These resultsthus highlight the distinctive differences by which Mstn and NF-B signal to regulate myogenesis,a finding which broadens our understanding of how these pathways function in both developmentand disease.

    Keywords: NF-B; myostatin; myogenesis; cachexia; wasting

    IntroductionNF-B is a ubiquitously expressed transcription factor in-volved in many cellular processes that regulate immune re-sponses, cellular proliferation, differentiation, and cell sur-vival [1]. The mammalian NF-B family consists of five mem-bers: NF-B1 (p50), NF-B2 (p52), RelA (p65), c-Rel andRelB [2]. All of these proteins share a common REL hom-ology domain that is responsible for dimerization, DNA bind-ing, and nuclear localization. However, these Rel proteinsdiffer in their C-terminus such that only p65, c-Rel and RelB

    DOI: 10.1002/sita.200400039

    Correspondence: Denis C. Guttridge, Division of HumanCancer Genetics, 420 W. 12th Avenue, The Ohio State Uni-versity College of Medicine, Columbus, Ohio, 43210. Phone:+1 614 688-3137, Fax: +1 614 688-4006, e-mail: guttridge-1@medctr.osu.edu

    Signal Transduction 2005, 4, 202210 www.signaltrans.de 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim

    possess a transactivation domain. NF-B proteins can existas homo- or heterodimers, the most commonly found tran-scriptional activator being the p65/p50 heterodimer. In un-stimulated mammalian cells, NF-B is predominantly foundin the cytoplasm in an inactive state, bound to its inhibitoryprotein (IB). Various stimuli such as inflammatory cyto-kines, bacterial products, double stranded RNA, irradiation,reactive oxygen species or growth factors lead to the acti-vation of the I-kappaB kinase (IKK) complex which causesthe phosphorylation and subsequent degradation of IB,thus allowing NF-B to translocate to the nucleus where itbinds to its cognate DNA sequence and stimulates gene ex-pression [3].

    Aside from its more accepted role as a regulator of innateimmunity, accumulating evidence suggest that NF-B is alsoinvolved in the differentiation and maintenance of skeletalmuscle [4]. The dynamics of skeletal myogenesis is con-trolled by members of the myogenic basic helix-loop-helix

  • Absence of NF-B and myostatin crosstalk in myogenesis 203

    (bHLH) transcription factors MyoD, Myf5, myogenin andMRF4 [5], as well as the myogenic enhancing factorsMEF2A, B, C, and D [6]. These myogenic transcription fac-tors are responsible for the early myogenic commitment(MyoD, Myf5) and/or later downstream differentiation eventsinvolving cell cycle arrest, fusion, and expression of contrac-tile genes (MyoD, myogenin, MEF2C, MRF4) [5]. In contrastto most mammalian cells, proliferating C2C12 and primarymurine myoblasts contain constitutive nuclear NF-B activity[7]. Mechanistically, NF-B inhibits myogenic differentiationby promoting cell growth through the transcriptional regu-lation of cyclin D1 [7], and activation of NF-B in responseto inflammatory cytokines inhibits myogenesis [8] by promot-ing the degradation of MyoD mRNA [8, 9]. However, evi-dence for a pro-myogenic role of NF-B also exists [1012].Constitutive activation of NF-B in myotubes has also beenshown to promote myofiber dysfunction through the loss ofMyoD and myosin heavy chain, a scenario associated withskeletal muscle wasting or cachexia [9, 1316]. More formalproof that NF-B is a regulator of wasting was recently pro-vided through the generation of a transgenic mouse ex-pressing a constitutively active form of IKK in skeletal muscle[17]. These mice were found to exhibit severe muscle atro-phy, a hallmark feature of cachexia. Importantly, cachexiawas blocked when mice were crossed with a secondtransgenic line expressing the transdominant IBSR inhibi-tor of NF-B in muscle [17]. IKK-driven muscle wasting wasalso shown to be partly dependent on E3 ubiquitin ligasefunction, but evidence from another wasting model demon-strated that the Foxo-1 and Foxo-3 regulation of E3 ligaseexpression is not mediated through NF-B transcriptionalactivity [18]. Abrogation of muscle atrophy from a disusemodel was also recently demonstrated using mice lackingthe p50 subunit of NF-B [19]. Furthermore, constitutive NF-B activity has been associated with chronic muscle wastingdisease states such as in muscular dystrophies and inflam-matory myopathies [20, 21]. Taken together, these findingssupport the role of NF-B as a pivotal regulator of myogenicdifferentiation and muscle wasting.

    Myostatin (Mstn, also known as growth and differentiationfactor-8, GDF-8) is a member of the transforming growth fac-tor- (TGF) family of growth and differentiation factors thatalso functions as a potent regulator of skeletal myogenesis[22, 23]. Furthermore, deletion of the Mstn gene in both miceand cattle showed dramatic increases in muscle mass andbody weight due to muscle fiber hyperplasia and hypertro-phy thus establishing its role as a negative regulator ofmuscle development. Strikingly, a similar phenotype was re-cently described in a child exhibiting loss-of-function in bothMstn alleles, making this the first case where a human mu-tation mimics the animal phenotype [24]. Mstn is also cap-able of inducing skeletal muscle degeneration. Severalgroups have observed upregulation of Mstn levels in hu-mans with conditions of muscle loss, resulting from HIV-in-

    Signal Transduction 2005, 4, 202210 www.signaltrans.de 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim

    fection, and disuse atrophy [2527]. Likewise, blockade ofMstn resulted in an increase in muscle mass, size and abso-lute muscle strength in both normal and dystrophic mice[2830]. Consistent with these findings, systemically admin-istered Mstn induced muscle and fat loss, in an analogousfashion to what is commonly observed in cachectic patients[31]. Further investigations into the mechanism of action ofMstn revealed that this cytokine is capable of inhibiting myo-blast proliferation and DNA synthesis, and arrests musclecells in the G1 phase of the cell cycle [32]. This arrest isaccompanied by increases in p21 and p53 levels and ac-cumulation of the hypophosphorylated form of retinoblas-toma (Rb) protein [3335]. In addition to its effects onmuscle precursor cell proliferation, Mstn can inhibit myogen-esis by suppressing the synthesis of the transcription factorsMyoD, Myf5 and myogenin [33, 34, 36] and/or phosphorylat-ing MyoD leading to a reduction in its DNA binding activity[36]. Furthermore, Mstn signaling was found to stabilizeSmad2/3 phosphorylation, induce Smad7 expression and in-crease Smad3-MyoD association, which in all cases inhibitsMyoD transactivation function [36, 37].Given the similarities in activities of NF-B and Mstn in con-trolling myogenesis, we considered the possibility that thesetwo signaling pathways may function through crosstalk. Pre-cedence exists for such crosstalk at least between TGFand NF-B regulating cellular and immune processes. Forexample, TGF and NF-B signaling synergize to stimulatetype VII collagen gene expression [38]. In addition, TGF-mediated epithelial to mesenchymal transition in a breastcancer progression model was recently shown to depend onIKK/NF-B signaling [39]. Furthermore, although not for-mally tested, an NF-B DNA consensus binding site wasidentified in the Mstn promoter [40], suggestive of a positivefeedback loop. Studies also demonstrate that TGF and NF-B signaling may function in an antagonistic fashion. TGFmay inhibit NF-B pro-survival activity [41], while NF-B canblock TGF-mediated pro-apoptosis by stimulating the ex-pression of Smad-7 that inhibits TGF- receptor signaling[42]. Since crosstalk between Mstn and NF-B has not beenestablished, we decided to explore the possibility that suchcrosstalk may be involved to regulate myogenesis potentiallythrough a common target like MyoD.

    Material and methodsCell culture and plasmidsC2C12 myoblasts were cultured and differentiated as pre-viously described [7] except where noted when insulin con-centrations was adjusted from 0 to 10 g/mL. Murine Mstnand IL1 were purchased from R & D Research (Minnea-polis, MN), while murine TNF was obtained from RocheIndustries (Indianapolis, IN). Reporter plasmids, 3x-B-Luc,Tn-I-Luc, as well as plasmids expressing MyoD, p65, andIBSR were used as previously described [7]. Myogenin

  • 204 N. Bakkar, H. Wackerhage, D. C. Guttridge

    expression plasmid was provided by A.Lassar (Harvard Uni-versity, Boston, MA) and Gal4-Luc reporter and the CMV-p65TA1 expression plasmids (containing the transactivationdomain of human p65 from amino acids 521-551) were gen-erously provided by A. Baldwin (University of North Carolina,Chapel Hill, NC).

    EMSA and transfectionsPreparation of nuclear extracts from C2C12 myocytes andEMSA analysis was performed as previously described [7].Quantification of NF-B DNA binding activity was performedwith NIH image. For NF-B analysis in whole muscles, nu-clear extracts from gastrocnemius muscles were preparedas described [20]. For transfections, cells were plated in trip-licate in 12 well dishes overnight. The next day cells weretransfected with either reporter plasmids at 250 ng/well, orexpression plasmids at 50 ng/well or at concentrationswhere indicated. Transfection efficiencies were normalizedwith CMV-LacZ added at 250 ng/well. Results were reportedas meanstandard deviation (SD).

    RT-PCRTotal RNA was isolated with Trizol as recommended by themanufacturer (Invitrogen CA). Semi-quantitative RT-PCRwas performed as a two-step RT-PCR with 2 g of RNAusing the following primers: for Mstn, forward primer 5 TTGGCT CAA ACA GCC TGA ATC 3and reverse primer 5 AAAATC GAC CGT GAG GGG GTA 3; for IB, forward primer5 CTG ATG TCA ACG CTC AGG AGC 3, and reverseprimer 5 CCT CTG TGA ATT CTG ACT CCG TG; forGAPDH, forward primer 5 GGA GAT TGT TGC CAT CAACGA CC 3, and reverse primer 5 GGT CAT GAG CCC TTCCAC AAT GC 3.

    AnimalsAnimals were housed in the animal facility at the Ohio StateUniversity Comprehensive Cancer Center under conven-tional conditions including constant temperature and hu-midity, and fed a standard diet. Treatment of mice was inaccordance to institutional guidelines for Animal Care andUse Committee.

    ResultsTo initiate this study we first asked whether Mstn was cap-able of activating NF-B in differentiating C2C12 cells. Elec-trophoretic mobility shift assays (EMSA) displayed the typi-cal double banding pattern of NF-B, which by supershiftanalysis in myoblasts was previously identified to be the p50/p50 homodimer and p50/p65 heterodimer complexes [7].NF-B activation is very sensitive to extracellular signals,

    Signal Transduction 2005, 4, 202210 www.signaltrans.de 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim

    and our results showed that simply switching myoblasts fromgrowth medium (GM) to differentiation medium (DM) for 30min in the absence of Mstn, was sufficient to induce a transi-ent activation of NF-B (Fig. 1A, compares lanes 1 to 2).However, similar to most NF-B activators, this increase inDNA binding activity is lost within 1 hour, most likely re-flecting the resynthesis of IB. Addition of Mstn at increas-ing concentrations (0100 ng/mL) had modest effects onthis activity, which returned back to basal level by 1 h. Like-wise, addition of Mstn over a 24h timecourse caused only aminor induction of NF-B as compared to myoblasts treatedwith DM alone (Fig. 1B). Since insulin is known to activateNF-B in differentiating myoblasts [43], it was possible thatinsulin present in the differentiation media was itself inducingNF-B at a level that would conceal any additional activationby Mstn. To address this point myoblasts were differentiatedfor 30 min in medium lacking or containing Mstn with in-creasing concentrations of insulin. Contrary to previous find-ings [43], our EMSA results showed that insulin alone hadno effect on NF-B activity, and only at the highest concen-trations of insulin (10 g/mL) did NF-B appear to be mod-estly induced by Mstn (Fig. 1C). These results indicate thatthe lack of any substantial activation of NF-B by Mstn is notdue to an insulin masking effect. Furthermore, addition ofMstn to differentiating C2C12 myoblasts inhibited myotubeformation (Fig. 1D) and specifically reduced MyoD, but notmyogenin transactivation function in 10T1/2 fibroblasts (Fig.1E), confirming that the lack of significant induction of NF-B was not related to Mstn inactivity. Together these resultssuggest that Mstn does not induce NF-B nuclear translo-cation or DNA binding activity in myoblasts undergoing dif-ferentiation.

    Given that Ras and Akt/protein kinase B [44] have beenshown to induce NF-B transcriptional activity without in-creasing NF-B nuclear translocation or DNA binding, weasked whether Mstn could potentially regulate NF-B activityvia a similar mechanism. C2C12 myoblasts were thereforetransfected with an NF-B responsive reporter plasmid andcells were subsequently induced to differentiate in the ab-sence or presence of Mstn. In comparison to TNF, whichis known to stimulate NF-B transcriptional activity, no suchactivation was observed in the presence of Mstn in C2C12myoblasts (Fig. 2A). To confirm this result, transfectionswere repeated with a p65 expression plasmid containingonly the carboxyl transactivation domain fused to DNA bind-ing domain of GAL4. Again, while TNF induced p65 trans-activation, similar activation was not observed upon Mstntreatment (Fig. 2B).Since NF-B is constitutively active in proliferating myoblastnuclei [7], we postulated that it is perhaps this portion of NF-B that may be regulated by Mstn to maintain myoblasts inan undifferentiated state. To test this hypothesis, transfec-tions were performed in C2C12 myoblasts with a MyoD re-sponsive reporter in the presence or absence of the trans-

  • Absence of NF-B and myostatin crosstalk in myogenesis 205

    Fig. 1. Mstn causes minor activation ofNF-B DNA binding activity in differen-tiating myoblasts. A. ProliferatingC2C12 cells were induced to differenti-ate (DM) in the absence or presence ofdifferent concentrations of Mstn (0, 10,50, or 100 ng/mL) for 30 min or 1 h.Nuclear extracts were prepared andEMSA was performed, representativeof four independent experiments. B.Myoblasts were induced to differentiatein the absence or presence of 20 ng/mL of Mstn for up to 24 h. Nuclear ex-tracts were prepared at the indicated ti-mes and EMSA was performed. C.C2C12 myoblasts were induced to dif-ferentiate under increasing insulin con-centrations (0, 0.1, 1, or 10 g/mL), inthe presence or absence of Mstn (25ng/mL) for 30 min. Nuclear extractswere then prepared for EMSA. D.C2C12 myoblasts were cultured in dif-ferentiation media in the absence orpresence of Mstn (50 ng/mL) for 72 h.E. 10T1/2 fibroblasts were co-trans-fected with 0.25 g of a luciferase re-porter fused to the troponin I enhancerelement (TnI-luc) and 0.05 g CMV-MyoD, or 0.05 g CMV-Myogenin. Nextday, cells were left untreated (control)or treated with 25 ng/mL Mstn underdifferentiation conditions and har-vested after 48 h.

    dominant inhibitor of NF-B, the IB super-repressor(IBSR), which functions to inhibit basal NF-B activity [7].As reported, Mstn treatment decreased MyoD transactiv-ation [33, 34, 36], yet this regulation was relatively unalteredin cells devoid of NF-B activity due to the expression ofIBSR (Fig. 3A). Similar results were obtained when trans-fections were performed in 10T1/2 fibroblasts where myog-enesis was driven by a MyoD expression plasmid (Fig. 3B).These results argue that basal nuclear activity of NF-B isnot utilized by Mstn to inhibit muscle differentiation.

    Similar to Mstn, constitutive activation of NF-B in myotubesis associated with muscle wasting [9, 17, 19]. TNF inducedactivation of NF-B has also been shown to be higher inmyotubes compared to proliferating myoblasts, suggestingthat mechanisms regulating NF-B activity in myotubes aredistinct from those in myoblasts [16]. Based on this evidencewe considered the possibility that Mstn may be capable ofactivating NF-B in differentiated myocytes. EMSA resultsfirst showed that unlike the activation of NF-B that occurredwhen myoblasts were switched from GM to DM (Fig. 1A), inpre-differentiated myotubes, the switch from 3-day culturedDM to fresh DM (0 ng/mL Mstn) did not induce NF-B activity

    Signal Transduction 2005, 4, 202210 www.signaltrans.de 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim

    (Fig. 4A). Secondly, treatment of myotubes with increasingconcentrations of Mstn for 30 min was seen to cause a mod-est, but reproducible transient increase in NF-B activity.Similar transient increase in NF-B DNA binding activity hasbeen reported with TGF treatment of mature B cells andhepatocytes [45]. By 1 h however, only myotubes treatedwith the highest dose of Mstn (100 ng/mL) retained NF-Bbinding activity, but even at this dose, activity could not besustained (Fig. 4B). In addition, and consistent with this tran-sient increase, we were unable to detect any significant dif-ference in NF-B transcriptional activity in myotubes follow-ing Mstn treatment for 6 h (Fig. 4C), demonstrating thatMstn-induced transient activation of NF-B DNA binding ac-tivity is not sufficient to stimulate NF-B-dependent tran-scription.

    To examine whether the inability to detect Mstn stimulationof NF-B may be related somehow to the in vitro C2C12myogenesis culture system, NF-B activity was also moni-tored in muscles from mice treated with Mstn. Comparedto the expected activation of NF-B in response to TNFtreatment in gastrocnemius muscles, no such stimulationcould be detected by Mstn injections (Fig. 4D). Taken to-

  • 206 N. Bakkar, H. Wackerhage, D. C. Guttridge

    Fig. 2. Mstn does not regulate NF-B transcriptional activity.A. C2C12 myoblasts were transfected with 0.25 g of anNF-B responsive reporter (MHC-3xB-Luc) and the nextday cells were induced to differentiate for up to 24h in DMalone (control) or media containing either 5 ng/mL TNF or20 ng/mL Mstn. Cell lysates were collected 6 and 24 h laterand luciferase activity was measured. B. C2C12 myoblastswere co-transfected with 0.15 g of a Gal4-Luc reporter pla-smid with 0.01 g of a CMV-Gal4-p65TA1 expression pla-smid. Next day similar treatments were applied as in (A) andat indicated times, extracts were prepared and luciferase ac-tivity was measured.

    gether, these data argue that Mstn regulation of myogenesisand muscle turnover does not function through NF-B sig-naling.

    Although our results demonstrate that Mstn inhibition ofmuscle differentiation is NF-B-independent, we consideredthe possibility that NF-B may itself regulate myogenesisthrough Mstn. Interestingly, the Mstn promoter has been re-ported to contain an NF-B consensus DNA binding site[40]. Therefore to examine this point, we treated pre-differ-entiated C2C12 myotubes with TNF or IL-1 to induce NF-B activity and monitored Mstn gene expression by semi-quantitative RT-PCR. IB, a known NF-B target gene, wasreadily induced by these cytokines, and TNF treatment dis-played NF-B biphasic regulation as previously demon-strated (Fig. 5A and 5B) [16]. However, little if any increase

    Signal Transduction 2005, 4, 202210 www.signaltrans.de 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim

    Fig. 3. Endogenous activity of NF-B is not required for Mstninhibition of myogenesis. A. C2C12 myoblasts were co-transfected with 0.25 g of the Tn I-Luc reporter plasmid withor without 0.25 g of the CMV-IBSR expression plasmid.24 h later, cells were induced to differentiate in the absence(control) or presence of 20 ng/mL Mstn. Cell lysates wereprepared and luciferase activity was determined. DNA wasstandardized by the addition of Bluescript plasmid (Strata-gene Inc., La Jolla, CA). B. 10T1/2 fibroblasts were co-trans-fected with 0.25 g Tn I-Luc, 0.05 g CMV-MyoD in the pre-sence or absence of 0.1 g of the CMV-IBSR expressionplasmid. Cells were subsequently treated with Mstn underdifferentiation conditions and luciferase activity was determi-ned 24 h later.

    in Mstn expression was observed by these inducers, nor didit appear to matter whether NF-B activation occurred inmyoblasts or myotubes. We conclude from these data thatthe ability of NF-B to negatively regulate myogenesis isalso independent of Mstn.

    DiscussionMstn and NF-B play major roles regulating skeletal musclehomeostasis. Mstn, signaling through Smad 2/3 and Smad7

  • Absence of NF-B and myostatin crosstalk in myogenesis 207

    Fig. 4. Mstn causes a transient increase in NF-&B DNA bin-ding activity in myotubes. A. C2C12 myoblasts were allowedto differentiate for 3 days, and then myotubes were treatedwith increasing doses of Mstn (0, 10, 50, and 100 ng/mL) infresh DM for 0.5 and 1 h. The DM lane refers to 3-day olddifferentiation media. Nuclear extracts were prepared andEMSA was performed, representative of three independentexperiments. B. Myotubes were treated either in DM aloneor DM containing 100 ng/mL Mstn for various time pointsand nuclear extracts were prepared for EMSA analysis. C.C2C12 cells were differentiated for 48 h, and then treatedwith either 5 ng/mL TNF or 50 ng/mL Mstn in DM for 6 h.Lysates were prepared to assay for luciferase readout. D.Gastrocnemius mouse muscles were injected with eitherPBS, TNF (0.5 g) or Mstn (1 g). At indicated times,muscles were harvested and nuclear extracts prepared forNF-B EMSA.

    [37], and NF-B regulated by several cytokine signals, inhibitmyogenesis through similar mechanisms: they regulatecomponents of the cell cycle machinery, and inhibit the syn-thesis/function of MyoD, a master regulator of myogenesis.Both of these signaling pathways have also been implicatedin skeletal muscle wasting [46]. Given these similarities, theNF-B binding site in the myostatin promoter [40], and the

    Signal Transduction 2005, 4, 202210 www.signaltrans.de 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim

    literature supporting crosstalk between TGF- and NF-Bsignaling [38, 39, 42, 45], we hypothesized that Mstn maybe an upstream regulator of NF-B in C2C12 skeletal myog-enesis. Our results show that these two pathways, thoughtargeting the same downstream effectors do not cooperate:Mstn does not induce NF-B activity in differentiating myo-blasts or differentiated myotubes, nor does it utilize the nu-clear basal B pool to mediate its effect. We also showedthat Mstn expression is not regulated by NF-B trans-criptional activity. In addition, although overexpression ofSmad2/3 and Smad3/4 inhibited myogenesis, these down-stream effectors of Mstn nevertheless failed to activate anNF-B responsive reporter, thus further confirming the ab-sence of crosstalk by these two signalling pathways in myo-blasts. It is noteworthy that in a recent finding, differentialeffects of Mstn on myogenesis were observed depending onwhether this factor was added to cells as a recombinant pro-tein, as was done in this current study, or overexpressed incells by gene transfection [47]. Consistent with these find-ings, expression of MyoD by Mstn was differentially regu-lated in cases of overexpression versus exogenous treat-ment [34, 36]. Given these considerations, it remains pos-sible that NF-B activity may be regulated by the overexpr-ession of Mstn in myoblasts. However, we are less likely tofavor this scenario given the nature of non-specific effectsoften associated with overexpression systems.

    This study thus underscores the selectivity of signaling dur-ing myogenesis, and the importance of having distinct andseparate signaling pathways that could be activated in re-sponse to different upstream signals. Mstn and NF-B, likeother signaling molecules AP-1 [48] and delta Notch [49], doconverge on MyoD, a master switch of myogenesis, to regu-late skeletal differentiation. Mstn was reported to downregul-ate MyoD expression, and inhibit its activity through in-creased Smad3/MyoD binding [36]. NF-B on the other handhas been shown to induce loss of MyoD in differentiatingC2C12 myocytes by a post-transcriptional mechanism [9].However, although both pathways converge on MyoD, theirupstream regulators are distinct. Mstn expression is regu-lated by MyoD itself [50], glucocorticoids [40], or follistatin[51], while NF-B regulation in skeletal myogenesis is lesswell characterized. p38/MAPK and Akt signaling have beenshown to activate NF-B [43], however the significance ofthis activation in myogenesis remains to be completely de-fined. It would appear from our current results that, eventhough Mstn is considered to be a major mediator of muscleintegrity, this activity does not depend on the IKK/NF-Bpathway. Such information will be helpful for better under-standing how these signaling pathways function to regulatemyogenesis potentially in both pre- and post-natal develop-ment. From a clinical standpoint, recognizing that Mstn andNF-B function by distinct pathways may also influence howthese signaling effectors can be targeted for treatment ofvarious skeletal muscle disorders.

  • 208 N. Bakkar, H. Wackerhage, D. C. Guttridge

    Fig. 5. TNF activation of NF-B doesnot induce Mstn expression in C2C12myotubes. A. Myoblasts were differen-tiated for 48h and then treated withTNF (5 ng/mL). At indicated timepoints RNA was extracted and RT-PCRwas performed to detect Mstn or IBexpression. B. Myotubes were treatedwith 10 ng/mL IL1- for the indicatedtime points, RNA was prepared andRT-PCR performed as in (A).

    AcknowledgmentsWe would like to thank J. Yu and P. J. Atherton for input andtechnical assistance and members of the Guttridge lab forhelpful discussions during the course of this study. The workwas supported by NIH grants CA097953 and CA098466.

    References[1] Baldwin, A. S. (2001) Control of oncogenesis and cancer

    therapy resistance by the transcription factor NF-kappaB.J. Clin. Invest. 107(3): 241246.

    [2] Ghosh, S., May, M. J., Kopp, E. B. (1998) NF-kappa B andRel proteins: evolutionarily conserved mediators of immuneresponses. Annu. Rev. Immunol. 16: 225260.

    [3] Karin, M., Ben-Neriah, Y. (2000) Phosphorylation meetsubiquitination: the control of NF-[kappa]B activity. Annu.Rev. Immunol. 18: 621663.

    [4] Guttridge, D. C. (2004) Signaling pathways weigh in on de-cisions to make or break skeletal muscle. Curr. Opin. Clin.Nutr. Metab. Care. 7(4): 443450.

    [5] Sabourin, L. A., Rudnicki, M. A. (2000) The molecular regu-lation of myogenesis. Clin. Genet. 57(1): 1625.

    [6] Naya, F. S., Olson, E. (1999) MEF2: a transcriptional targetfor signaling pathways controlling skeletal muscle growthand differentiation. Curr. Opin. Cell Biol. 11(6): 683688.

    [7] Guttridge, D. C., Albanese, C., Reuther, J. Y., Pestell, R.G., Baldwin Jr., A. S. (1999) NF-kappaB controls cell growthand differentiation through transcriptional regulation of cy-clin D1. Mol. Cell. Biol. 19(8): 57855799.

    [8] Langen, R. C., Schols, A. M., Kelders, M. C., Wouters, E.F., Janssen-Heininger, Y. M. (2001) Inflammatory cytokinesinhibit myogenic differentiation through activation of nuclearfactor-kappaB. FASEB J. 15(7): 11691180.

    [9] Guttridge, D. C., Mayo, M. W., Madrid, L. V., Wang, C.-Y., Baldwin Jr., A. S. (2000) NF-kB-Induced Loss of MyoDMessenger RNA: Possible Role in Muscle Decay and Ca-chexia. Science 289(5488): 23632366.

    [10] Lee, K. H., Kim, D. G., Shin, N. Y., Song, W. K., Kwon, H.,Chung, C. H., Kang, M. S. (1997) NF-kappaB-dependent

    Signal Transduction 2005, 4, 202210 www.signaltrans.de 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim

    expression of nitric oxide synthase is required for mem-brane fusion of chick embryonic myoblasts. Biochem. J.324 (Pt 1): 237242.

    [11] Kim, S. S., Rhee, S., Lee, K. H., Kim, J. H., Kim, H. S.,Kang, M. S., Chung, C. H. (1998) Inhibitors of the protea-some block the myogenic differentiation of rat L6 myoblasts.FEBS Lett. 433(1-2): 4750.

    [12] Kaliman, P., Canicio, J., Testar, X., Palacin, M., Zorzano, A.(1999) Insulin-like growth factor-II, phosphatidylinositol 3-kinase, nuclear factor-kappaB and inducible nitric-oxidesynthase define a common myogenic signaling pathway. J.Biol. Chem. 274(25): 1743717444.

    [13] Li, Y. P., Schwartz, R. J., Waddell, I. D., Holloway, B. R.,Reid, M. B. (1998) Skeletal muscle myocytes undergo pro-tein loss and reactive oxygen-mediated NF-kappaB acti-vation in response to tumor necrosis factor alpha. FASEBJ. 12(10): 871880.

    [14] Li, Y. P, Reid, M. B. (2000) NF-kappaB mediates the proteinloss induced by TNF-alpha in differentiated skeletal musclemyotubes. Am. J. Physiol. Regul. Integr. Comp. Physiol.279(4): R1165011670.

    [15] Acharyya, S., Ladner, K. J., Nelsen, L. L., Damrauer, J.,Reiser, P. J., Swoap, S., Guttridge, D. C. (2004) Cancercachexia is regulated by selective targeting of skeletalmuscle gene products. J. Clin. Invest. 114(3): 370378.

    [16] Ladner, K. J., Caligiuri, M. A., Guttridge, D. C. (2003) Tumornecrosis factor-regulated biphasic activation of NF-kappa Bis required for cytokine-induced loss of skeletal musclegene products. J. Biol. Chem. 278(4): 22942303.

    [17] Cai, D., Frantz, J. D., Tawa, N. E., Jr., Melendez, P. A., Oh,B.-C., Lidov, W. R., Hasselgren, P. O., Frontera, W. R., Lee,J., Glass, D. J., Shoelson, S. E. (2004) IKKbeta/NF-kB acti-vation causes severe muscle wasting in mice. Cell 119:285298.

    [18] Sandri, M., Sandri, C., Gilbert, A., Skurk, C., Calabria, E.,Picard, A., Walsh, K., Schiaffino, S., Lecker, S. H., Gold-berg, A. L. (2004) Foxo transcription factors induce the atro-phy-related ubiquitin ligase atrogin-1 and cause skeletalmuscle atrophy. Cell 117(3): 399412.

  • Absence of NF-B and myostatin crosstalk in myogenesis 209

    [19] Hunter, R. B., Kandarian, S. C. (2004) Disruption of eitherthe Nfkb1 or the Bcl3 gene inhibits skeletal muscle atrophy.J. Clin. Invest. 114(10): 15041511.

    [20] Kumar, A., Boriek, A. M. (2003) Mechanical stress activatesthe nuclear factor-kappaB pathway in skeletal muscle fib-ers: a possible role in Duchenne muscular dystrophy. FA-SEB J. 17(3): 386396.

    [21] Monici, M. C., Aguennouz, M., Mazzeo, A., Messina, C.,Vita, G. (2003) Activation of nuclear factor-kappaB in in-flammatory myopathies and Duchenne muscular dystrophy.Neurology 60(6): 993997.

    [22] Lee, S.-J. (2004) Regulation of Muscle Mass by Myostatin.Ann. Rev. Cell Dev. Biol. 20: 6186.

    [23] Gonzalez-Cadavid, N. F, Bhasin, S. (2004) Role of myosta-tin in metabolism. Curr. Opin. Clin. Nutr. Metab. Care.7(4): 451457.

    [24] Schuelke, M., Wagner, K. R., Stolz, L. E., Hubner, C., Rie-bel, T., Komen, W., Braun, T., Tobin, J. F., Lee, S. J. (2004)Myostatin mutation associated with gross muscle hypertro-phy in a child. N. Engl. J. Med. 350(26): 26822688.

    [25] Reardon, K. A., Davis, J., Kapsa, R. M., Choong, P., Byrne,E. (2001) Myostatin, insulin-like growth factor-1, and leuke-mia inhibitory factor mRNAs are upregulated in chronic hu-man disuse muscle atrophy. Muscle Nerve. 24(7):893899.

    [26] Zachwieja, J. J., Smith, S. R., Sinha-Hikim, I., Gonzalez-Cadavid, N., Bhasin, S. (1999) Plasma myostatin-immuno-reactive protein is increased after prolonged bed rest withlow-dose T3 administration. J. Grav. Physiol. 6(2): 1115.

    [27] Gonzalez-Cadavid, N. F., Taylor, W. E., Yarasheski, K.,Sinha-Hikim, I., Ma, K., Ezzat, S., Shen, R., Lalani, R., Asa,S., Mamita, M., Nair, G., Arver, S., Bhasin, S. (1998) Organ-ization of the human myostatin gene and expression inhealthy men and HIV-infected men with muscle wasting.Proc. Natl. Acad Sci. U.S.A. 95(25): 1493814943.

    [28] Whittemore, L. A., Song, K., Li, X., Aghajanian, J., Davies,M., Girgenrath, S., Hill, J. J., Jalenak, M., Kelley, P., Knight,A., Maylor, R., OHara, D., Pearson, A., Quazi, A., Ryerson,S., Tan, X. Y., Tomkinson, K. N., Veldman, G. M., Widom,A., Wright, J. F., Wudyka, S., Zhao, L., Wolfman, N. M.(2003) Inhibition of myostatin in adult mice increases skel-etal muscle mass and strength. Biochem. Biophys. Res.Commun. 300(4): 965971.

    [29] Wagner, K. R., McPherron, A. C., Winik, N., Lee, S. J.(2002) Loss of myostatin attenuates severity of musculardystrophy in mdx mice. Ann. Neurol. 52(6): 832836.

    [30] Bogdanovich, S., Krag, T. O., Barton, E. R., Morris, L. D.,Whittemore, L. A., Ahima, R. S., Khurana, T. S. (2002)Functional improvement of dystrophic muscle by myostatinblockade. Nature 420(6914): 4184121.

    [31] Zimmers, T. A., Davies, M. V., Koniaris, L. G., Haynes, P.,Esquela, A. F., Tomkinson, K. N., McPherron, A. C., Wolf-man, N. M., Lee, S. J. (2002) Induction of cachexia in miceby systemically administered myostatin. Science 296(5572): 14861488.

    Signal Transduction 2005, 4, 202210 www.signaltrans.de 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim

    [32] Thomas, M., Langley, B., Berry, C., Sharma, M., Kirk, S.,Bass, J., Kambadur, R. (2000) Myostatin, a negative regu-lator of muscle growth, functions by inhibiting myoblast pro-liferation. J. Biol. Chem. 275(51): 4023540243.

    [33] Rios, R., Carneiro, I., Arce, V. M., Devesa, J. (2002) Myos-tatin is an inhibitor of myogenic differentiation. Am. J. Phy-siol. Cell Physiol. 282(5): C993C999.

    [34] Joulia, D., Bernardi, H., Garandel, V., Rabenoelina, F.,Vernus, B., Cabello, G. (2003) Mechanisms involved in theinhibition of myoblast proliferation and differentiation bymyostatin. Exp. Cell Res. 286(2): 263275.

    [35] McCroskery, S., Thomas, M., Maxwell, L., Sharma, M.,Kambadur, R. (2003) Myostatin negatively regulates satel-lite cell activation and self-renewal. J. Cell Biol. 162(6):11351147.

    [36] Langley, B., Thomas, M., Bishop, A., Sharma, M., Gilmour,S., Kambadur, R. (2002) Myostatin inhibits myoblast differ-entiation by down-regulating MyoD expression. J Biol.Chem. 277(51): 4983149840.

    [37] Zhu, X., Topouzis, S., Liang, L. F., Stotish, R. L. (2004)Myostatin signaling through Smad2, Smad3 and Smad4 isregulated by the inhibitory Smad7 by a negative feedbackmechanism. Cytokine 26(6): 262272.

    [38] Kon, A., Vindevoghel, L., Kouba, D. J., Fujimura, Y., Uitto,J., Mauviel, A. (1999) Cooperation between SMAD and NF-kappaB in growth factor regulated type VII collagen geneexpression. Oncogene 18(10): 18371844.

    [39] Huber, M. A., Azoitei, N., Baumann, B., Grunert, S.,Sommer, A., Pehamberger, H., Kraut, N., Beug, H., Wirth, T.(2004) NF-kappaB is essential for epithelial-mesenchymaltransition and metastasis in a model of breast cancer pro-gression. J. Clin. Invest. 114(4): 569581.

    [40] Ma, K., Mallidis, C., Artaza, J., Taylor, W., Gonzalez-Cada-vid, N., Bhasin, S. (2001) Characterization of 5-regulatoryregion of human myostatin gene: regulation by dexa-methasone in vitro. Am. J. Physiol. Endocrinol. Metab.281(6): E1128E1136.

    [41] Arsura, M., Wu, M., Sonenshein, G. E. (1996) TGF beta 1inhibits NF-kappa B/Rel activity inducing apoptosis of Bcells: transcriptional activation of I kappa B alpha. Immunity5(1): 3140.

    [42] Bitzer, M., von Gersdorff, G., Liang, D., Dominguez-Ros-ales, A., Beg, A. A., Rojkind, M., Bottinger, E. P. (2000) Amechanism of suppression of TGF-beta/SMAD signaling byNF-kappa B/RelA. Genes Dev. 14(2): 187197.

    [43] Conejo, R., Valverde, A. M., Benito, M., Lorenzo, M. (2001)Insulin produces myogenesis in C2C12 myoblasts by in-duction of NF-kappaB and downregulation of AP-1 activi-ties. J. Cell. Physiol. 186(1): 8294.

    [44] Madrid, L. V., Wang, C. Y., Guttridge, D. C., Schottelius, A.J., Baldwin Jr., A. S., Mayo, M.W. (2000) Akt suppressesapoptosis by stimulating the transactivation potential of theRelA/p65 subunit of NF-kappaB. Mol. Cell. Biol. 20(5):16261638.

  • 210 N. Bakkar, H. Wackerhage, D. C. Guttridge

    [45] Arsura, M., Panta, G. R., Bilyeu, J. D., Cavin, L. G., Sovak,M. A., Oliver, A. A., Factor, V., Heuchel, R., Mercurio, F.,Thorgeirsson, S. S., Sonenshein, G. E. (2003) Transientactivation of NF-kappaB through a TAK1/IKK kinase path-way by TGF-beta1 inhibits AP-1/SMAD signaling andapoptosis: implications in liver tumor formation. Oncogene22(3): 412425.

    [46] McKinnell, I. W., Rudnicki, M. A. (2004) Molecular mecha-nisms of muscle atrophy. Cell 119(7): 907910.

    [47] Rios, R., Fernandez-Nocelos, S., Carneiro, I., Arce, V. M.,Devesa, J. (2004) Differential response to exogenous andendogenous myostatin in myoblasts suggests that myosta-tin acts as an autocrine factor in vivo. Endocrinology145(6): 27952803.

    [48] Bengal, E., Ransone, L., Scharfmann, R., Dwarki, V. J.,Tapscott, S. J., Weintraub, H., Verma, I. M. (1992) Func-

    Signal Transduction 2005, 4, 202210 www.signaltrans.de 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim

    tional antagonism between c-Jun and MyoD proteins: a di-rect physical association. Cell. 68(3): 507519.

    [49] Hirsinger, E., Malapert, P., Dubrulle, J., Delfini, M. C., Dup-rez, D., Henrique, D., Ish-Horowicz, D., Pourquie, O. (2001)Notch signalling acts in postmitotic avian myogenic cells tocontrol MyoD activation. Development 128(1): 107116.

    [50] Spiller, M. P., Kambadur, R., Jeanplong, F., Thomas, M., Mar-tyn, J. K., Bass, J. J., Sharma, M. (2002) The myostatin geneis a downstream target gene of basic helix-loop-helix tran-scription factor MyoD. Mol. Cell. Biol. 22(20): 70667082.

    [51] Kocamis, H., Gulmez, N., Aslan, S., Nazli, M. (2004) Follis-tatin alters myostatin gene expression in C2C12 musclecells. Acta Vet. Hung. 52(2): 135141.

    [Received: December 8, 2004; accepted: May 30, 2005]

Recommended

View more >