Myostatin Regulates Cardiomyocyte Growth Through Modulation of Akt Signaling
Myostatin is a highly conserved, potent negative regulator of skeletal muscle hypertrophy in many species, from rodents to humans, although its mechanisms of action are incompletely understood. Transcript profiling of hearts from a genetic model of cardiac hypertrophy revealed dramatic upregulation of myostatin, not previously recognized to play a role in the heart. Here we show that myostatin abrogates the cardiomyocyte growth response to phenylephrine in vitro through inhibition of p38 and the serine-threonine kinase Akt, a critical determinant of cell size in many species from drosophila to mammals. Evaluation of male myostatin-null mice revealed that their cardiomyocytes and hearts overall were slightly smaller at baseline than littermate controls but exhibited more exuberant growth in response to chronic phenylephrine infusion. The increased cardiac growth in myostatin-null mice corresponded with increased p38 phosphorylation and Akt activation in vivo after phenylephrine treatment. Together, these data demonstrate that myostatin is dynamically regulated in the heart and acts more broadly than previously appreciated to regulate growth of multiple types of striated muscle.
Akt is a serine-threonine kinase whose role in controlling cell growth has been conserved from Drosophila to mammalian species.1–4 To study the effects of chronic Akt activation in the heart, we generated 2 transgenic murine lines with cardiac-specific expression of activated Akt.1 Both lines develop substantial cardiac hypertrophy characterized by an increase in cardiomyocyte size with preserved cardiac function,1 without the “fetal” transcriptional profile characteristic of pathological cardiac hypertrophy.5 Together these data suggest constitutive Akt activation in the heart induces an exaggerated growth response, consistent with its role in other species.2,3 One of the transgenic lines generated exhibited X-linked inheritance and, in the hearts of female mice, the expected chimeric transgene expression caused by X inactivation.5 As expected, transgene-expressing cardiomyocytes from these mice were larger than littermate control cardiomyocytes. However, cardiomyocytes not expressing the transgene following chromosomal inactivation were notably smaller than control cardiomyocytes,1 raising the possibility that a negative regulator of cardiomyocyte growth may be induced, perhaps as a counter-regulatory response to the overall increase in heart size.
To identify possible inhibitors of cardiomyocyte growth, we performed transcript profiling of Akt-transgenic hearts in comparison to transgene-negative littermates.5 The transcript most highly upregulated in both lines (65- and 18-fold)5 was myostatin (MSTN), a highly conserved transforming growth factor (TGF)-β family member and potent negative regulator of skeletal muscle growth. Although expression of MSTN in the heart has been previously reported,6 a functional role for myostatin in the heart has not been appreciated.7 MSTN was not directly induced by Akt activation in cardiomyocytes,5 suggesting induction occurs as an indirect consequence, perhaps in response to the dramatic cardiac hypertrophy manifest in these mice.
Although targeted deletion of MSTN in mice produces impressive skeletal muscle hypertrophy and resistance to diabetes, the responsible signaling mechanisms have not been fully delineated. A detailed analysis of hearts from MSTN−/− mice has not been reported.
To explore the role of MSTN in cardiomyocyte growth, we examined the effects of cardiomyocyte expression of MSTN or the inhibitory pro-domain (dnMSTN) in vitro on the response to hypertrophic stimuli. We found that MSTN regulates cardiomyocyte growth in a stimulus-specific manner while inhibiting p38 and Akt phosphorylation. Studies in MSTN−/− mice suggest these findings have in vivo relevance as well. Together these data demonstrate that MSTN regulates not only skeletal but also cardiac muscle growth. The clinical relevance of these findings has recently been underscored by the discovery of MSTN mutations in people,8,9 as well as interest in inhibiting MSTN in skeletal muscle diseases.10
Materials and Methods
Recombinant Adenoviruses Expressing Full-Length and Truncated Forms of MSTN
Mouse cDNA encoding MSTN and truncated forms of MSTN (dnMSTN) were prepared from total heart cDNA by PCR. Recombinant adenoviruses (Ad.MSTN and Ad.dnMSTN) expressing cytomegalovirus (CMV)-driven green fluorescent protein (GFP) and MSTN or dnMSTN were generated by homologous recombination. Adenovirus expressing GFP (Ad.GFP), myristoylated Akt (Ad.myr-Akt), and dnAkt (Akt-AA) have been described previously.11 Inactive mutant (dual phosphorylation site TGY changed to AGF) p38α (DNp38α) and activated MKK3b (MKK3bE) were kind gifts from Dr Yibin Wang (University of California, Los Angeles).12 Animals were handled in strict accordance to the guidelines of the Massachusetts General Hospital Subcommittee on Research Animal Care.
Primary cultures of neonatal rat ventricular cardiomyocytes were prepared from Sprague-Dawley neonates as previously described.13
MSTN−/− mice7 were kindly provided by Dr Se Jin Lee (Johns Hopkins University, Baltimore, Md). Mice were backcrossed to C57B6 for ≥6 generations, and littermate controls were used in all data presented.
Ex Vivo MSTN Assay
Hearts from Akt transgenics and littermates were harvested and washed in PBS. Atria were removed and hearts were minced. Resulting tissue was incubated (48 hours, 37°C) in 3.5 mL of DMEM containing 10% FCS. Media were removed and centrifuged (2000g, 15 minutes), and supernatants were collected for analysis.
Cardiomyocyte proteins were lysed as described,13 after stimulation (30 minutes, 100 μmol/L phenylephrine [PE] [Sigma], 1 nmol/L leukemia inhibitory factor [LIF] [Chemicon], or 10 nmol/L insulin-like growth factor-I [IGF-I] [Chemicon]). SB239063 (10 μmol/L; Calbiochem) was added 1 hour before stimulation to inhibit p38. Protein from 8- to 12-week-old mouse hearts was obtained after atria removed. After concentration determination by the Bradford method (Bio-Rad), proteins (30 μg) were separated by SDS-PAGE on 12% gels and transferred to nitrocellulose membranes (Bio-Rad) by semidry transfer. Blots were incubated with anti-Akt, anti-phospho-Akt (473), anti-phospho-GSK3(9/21), anti-Erk, anti-phospho-Erk, anti-Jnk, anti-phospho-Jnk, anti-p38, or anti-phospho-p38 (1:1000; Cell Signaling), overnight at 4°C and subsequently incubated with horseradish peroxidase (HRP)-conjugated secondary antibody (1:5000; DAKO), and detected by chemiluminescence (Cell Signaling).
Measurement of Protein Synthesis
Media were changed to serum-free DMEM containing [3H]-leucine (1 μCi/mL), and cells were stimulated with PE (100 μmol/L; Sigma), LIF (1 nmol/L; Chemicon), or IGF-I (10 nmol/L; Chemicon) for 24 hours before washing with PBS, H2O, harvesting in 0.25 N NaOH, and determining [3H]-leucine incorporation by scintillation counting.
Media were changed to serum-free DMEM with or without PE (100 μmol/L, 24 hours) as indicated. Cells were washed with PBS, trypsinized, resuspended in fresh medium, and counted with a hemocytometer.
Akt Kinase Assay
Akt was immunoprecipitated from heart lysates with anti-Akt antibody and kinase activity was measured according to the manufacturer’s instructions (Cell Signaling).
Echocardiography was performed on nonanesthetized mice using a 13L high-frequency linear (10 MHz) transducer (VingMed 5, GE Medical Services) with depth set at 1 cm and 236 frames per second for 2-D images. M-mode images used for measurements were taken at the papillary muscle level.
Adult Cardiomyocyte Measurement
Adult mouse CMs from MSTN+/+ and MSTN−/− were isolated, fixed on slides, and measured as described previously.1
Acute Phenylephrine Infusion
Under isoflurane anesthesia, 6- to 8-week-old mice were injected (via inferior vena cava) with 50 μL of 2 mmol/L PE dissolved in sterile 0.9% NaCl containing 0.02% ascorbic acid. After 3 minutes, mice were euthanized and hearts were excised, washed in cold PBS, and atria were removed and snap frozen in liquid nitrogen for analysis.
Chronic Phenylephrine Infusion
Under isoflurane anesthesia, Alzet microosmotic pumps (DURECT Corp, model no. 1002) were implanted subcutaneously in 6- to 8-week-old age-matched MSTN−/− and MSTN+/+ mice dorsally to deliver PE for 14 days (75 mg/kg per day; dissolved in sterile 0.9% NaCl/0.02% ascorbic acid). Vehicle-treated, gender/age-matched mice served as controls.
Data are represented as mean±SEM and compared by 2-tailed Student’s t test or ANOVA where appropriate. The null hypothesis was rejected for P<0.05.
An expanded Materials and Methods section can be found in the online data supplement, available at http://circres.ahajournals.org.
Because commercially available antibodies failed to reliably detect MSTN protein by Western blotting in either heart or skeletal muscle, we established an ELISA sensitive to nanomolar concentrations of MSTN (data not shown). We used this ELISA to determine whether the observed transcriptional changes resulted in an increase in MSTN secretion from isolated hearts. We found a 3.1-fold increase in MSTN secretion from Akt transgenic hearts compared with littermate transgene-negative controls (P<0.001). Interrogation of a publicly available microarray database14 revealed that cardiac MSTN mRNA also increased after 2 weeks of intensive exercise (4-fold, P<0.004) and in mice with cardiac expression of constitutively active (but not dominant negative) phosphoinositide 3-kinase (PI3K) (2.3-fold, P<0.04), the upstream activator of Akt. Thus, MSTN expression is dynamically regulated in the heart, especially under hypertrophic conditions that would be expected to activate Akt.
To examine the biological effects of MSTN in cardiomyocytes, we generated recombinant adenoviral vectors encoding either full-length MSTN (Ad.MSTN) or the amino-terminal propeptide previously demonstrated to inhibit the actions of MSTN15 (Ad.dnMSTN). Both viruses mediated expression of the appropriate transgene in cardiomyocytes (data not shown).
To model the hypertrophic response, we stimulated cardiomyocytes in vitro with the α1-adrenergic agonist PE.16 After PE stimulation, both cell size and protein synthesis increased in control virus-infected cardiomyocytes (Figure 1A, 1C, and 1D). Whereas MSTN expression did not alter size or protein synthesis in unstimulated cardiomyocytes, it completely abrogated the increase in both seen after PE stimulation (Figure 1A, 1C, and 1D). Coexpression of dnMSTN in MSTN-infected cells restored the growth response to PE (Figure 1A, last panel), suggesting dnMSTN was able to antagonize the effects of the expressed MSTN. Moreover, in unstimulated cardiomyocytes, dnMSTN expression increased both cell size and protein synthesis to levels comparable to that seen after PE stimulation (Figure 1B through 1D). dnMSTN also enhanced sarcomere prominence (Figure 1B), another characteristic of cardiomyocyte hypertrophy. Neither MSTN nor dnMSTN altered cardiomyocyte number 24 hours after treatment, confirming that changes in protein synthesis were caused by cell growth rather than proliferation (Figure 1E).
Although MSTN has not been postulated to act through Akt signaling, our observations in transgenic Akt mice suggested this might be the case. In chimeric Akt hearts, cardiomyocytes not expressing the transgene were smaller than those from nontransgenic littermates, whereas cardiomyocytes expressing myr-Akt were comparable to those from nonchimeric transgenic mice,1 suggesting Akt may lie downstream of the effects of MSTN. We examined this hypothesis in neonatal cardiomyocytes. Control cells treated with PE demonstrated phosphorylation of both Akt (P<0.001) and glycogen synthase kinase 3β (GSK3β), a downstream target relevant to hypertrophy.17,18 In contrast, MSTN-expressing cardiomyocytes manifested a dramatic inhibition of PE-stimulated phosphorylation of both Akt (P<0.01) and GSK3β (Figure 2A and 2C), with corresponding changes in Akt kinase activity (data not shown). Consistent with this, expression of dnMSTN increased Akt activation and phosphorylation (P<0.05) (Figure 2B and 2C).
Although PE has been reported to activate Akt,19 the functional relevance of Akt in PE-mediated hypertrophy has not been demonstrated. Therefore, we examined the effects of dnAkt or control viruses on PE-induced cardiomyocyte hypertrophy. dnAkt inhibited PE-induced increases in both protein synthesis and cell size (Figure 3A and 3B; P<0.001). Thus, Akt activation appears necessary for PE-induced cardiomyocyte hypertrophy, and MSTN inhibition of Akt could contribute to its inhibition of the growth response to PE.
To test this, we coexpressed either activated or dnAkt with MSTN or dnMSTN expression, respectively. Expression of activated Akt substantially restored PE-stimulated cell growth in MSTN-expressing cardiomyocytes (Figure 3C; P<0.001 versus GFP and MSTN+PE). Conversely, Akt inhibition in dnMSTN-expressing cardiomyocytes inhibited the cell growth seen with dnMSTN alone (Figure 3C; P<0.001 versus dnMSTN; P=NS versus GFP). Together these data demonstrate that MSTN modulates cardiomyocyte growth in vitro, at least in part, through inhibition of Akt activation.
To further explore the mechanisms involved, we stimulated cardiomyocytes with LIF and IGF-I, hypertrophic agonists that activate Akt through distinct mechanisms, and examined the ability of MSTN to inhibit hypertrophy and Akt activation. Although the increase in protein synthesis induced by LIF (P<0.05) appeared slightly reduced by MSTN infection, this was not statistically significant compared with LIF-stimulated, GFP-infected cardiomyocytes (Figure 4A). Similarly, MSTN expression appeared to slightly but not significantly reduce LIF-induced Akt phosphorylation (Figure 4C). In contrast, MSTN did not inhibit IGF-I-induced protein synthesis at all (Figure 4B), which remained significantly increased above baseline (P<0.01). Similarly, MSTN did not inhibit IGF-I-induced Akt phosphorylation (Figure 4D; P<0.01). Thus, MSTN-mediated inhibition of Akt paralleled its effects on hypertrophy and appears to be stimulus specific, presumably because of differences in the upstream pathways mediating Akt activation after PE as compared with LIF or IGF-I.
We next examined the effects of MSTN on activation of mitogen-activated protein (MAP) kinases (extracellular signal-regulated kinase [Erk], c-Jun N-terminal kinase [JNK], and p38) both because of crosstalk between these pathways and Akt20,21 as well as their independent regulation of cardiomyocyte growth.22–24 The pattern of Erk phosphorylation in response to PE, LIF, and IGF-I was not significantly altered by MSTN expression (Figure 5A). None of the agonists induced substantial JNK phosphorylation (Figure 5B). JNK phosphorylation appeared modestly suppressed in MSTN-infected cardiomyocytes, although neither this difference nor the baseline induction of JNK phosphorylation was statistically significant (Figure 5B). In contrast, PE induced significant phosphorylation of p38 (Figure 5C; P<0.01), which was completely inhibited by MSTN expression. LIF-mediated p38 phosphorylation was only partially attenuated by MSTN. IGF-I did not increase p38 phosphorylation, and MSTN expression did not alter this. These data demonstrate that, in addition to inhibiting Akt activation, MSTN inhibits PE-mediated p38 phosphorylation. Consistent with this, MSTN also inhibited PE-induced atrial natriuretic factor (ANF) expression, known to be increased by PE and p3822 but not by Akt (Figure 5D; P<0.01).
To examine whether MSTN-mediated inhibition of p38 was connected to its inhibition of Akt, we treated cardiomyocytes with 10 μmol/L SB239063, a p38-specific inhibitor. We found that treatment with SB significantly inhibited PE-induced Akt phosphorylation in GFP- or MSTN-infected cardiomyocytes (P<0.01 versus GFP+PE; Figure 6A). Moreover, adenoviral expression of DNp38α blocked PE-induced phosphorylation of both p38 and Akt (Figure 6B). Conversely, p38 activation with constitutively activated MKK3bE induced Akt phosphorylation, even in MSTN-expressing cardiomyocytes, suggesting that MSTN acts upstream of MKK and not directly on p38 or Akt (Figure 6B). dnMSTN infection alone did not induce p38 phosphorylation. This could reflect the involvement of other pathways in the effects MSTN or interaction of dnMSTN with other TGF-β family members. In addition, it seems likely that p38 phosphorylation seen in whole-cell lysates is an imperfect reflection of the specific subcellular p38 that may be important in these signaling events.
To see whether MSTN regulates cardiomyocyte growth in vivo, we examined hearts from MSTN−/− mice.7 Surprisingly, hearts from 8-week-old male MSTN−/− mice were modestly (9.6%) smaller than those from wild-type littermates, although this difference did not achieve statistical significance (P<0.06). Cardiomyocytes from MSTN−/− mice were significantly smaller than those from wild-type littermates (P<0.001). Baseline cardiac function and chamber dimensions as assessed by echocardiography in 8- to 10-week-old mice were not different in MSTN−/− mice compared with littermates (Table). To determine whether the in vivo response to PE mirrored the changes observed in vitro, we infused PE via miniosmotic pump for 2 weeks. Given the dramatic effect of MSTN genotype on body habitus and composition, we compared the heart weight in each genotype to age-matched, vehicle-treated controls at the end of this period. MSTN-deficient mice exhibited greater cardiac growth in response to PE infusion (Figure 7A; P<0.05). Interestingly, this difference was most dramatic for male MSTN−/− mice (14% versus 2.4%) (Figure 7A; P<0.05 versus wild-type littermates). After PE infusion, male MSTN−/− hearts were comparable in weight to age-matched wild-type littermates (124.0±5.6 versus 123.1±6.8 mg) and significantly larger than vehicle-treated, age-matched MSTN−/− controls (108.8±4.0 mg). PE did not induce substantial cardiac growth in female mice independent of genotype (Figure 7A; P=NS).
Whereas there was no difference in baseline Akt expression or activity in MSTN−/− hearts (data not shown), acute infusion of PE resulted in greater Akt activation in hearts from MSTN−/− mice (Figure 7B: 5.4±1.6-fold; P<0.04). This Akt activation was accompanied by an increase in PE-stimulated p38 phosphorylation (Figure 7C; 4.5±1.0-fold; P<0.05). Together, these data suggest that endogenous MSTN also plays a role modulating cardiac p38 phosphorylation, Akt activation, and growth in vivo.
We found that MSTN, a potent negative regulator of skeletal muscle growth, is dynamically regulated in the heart and regulates cardiomyocyte growth in vitro. These data suggest that MSTN may play a broader role than previously appreciated, regulating growth of striated muscle cells of both skeletal and cardiac lineages. Moreover, the unanticipated finding that MSTN modulates activity of Akt, a critical determinant of cell size in many systems, as well as p38 phosphorylation, may have implications for our understanding of the actions of MSTN in other settings.
The observation that cardiomyocytes in chimeric Akt-transgenic hearts that did not express the transgene were smaller than wild-type cardiomyocytes and that MSTN was induced in these hearts, raised the possibility that MSTN regulates cardiomyocyte growth through Akt. Induction of MSTN mRNA in transgenic Akt mice was confirmed by quantitative RT-PCR,5 and an increase in cardiac secretion was documented by ELISA. Interrogation of a publicly available microarray database14 revealed that cardiac MSTN mRNA is also increased by 2 weeks of intensive exercise and in mice with cardiac activation of PI3K, the upstream activator of Akt. Nevertheless, whether the reduction in nontransgenic cardiomyocyte size seen in chimeric Akt hearts can be fully attributed to MSTN is currently unknown.
The effects of MSTN expression on cardiomyocyte growth were examined first in vitro. MSTN expression completely suppressed the increase in cell size and protein synthesis induced by the α-adrenergic agonist PE. Conversely, expression of the inhibitory propeptide dnMSTN increased cardiomyocyte size and protein synthesis to a degree comparable to that seen with PE stimulation alone. These effects correlated with modulation of Akt activation as well as downstream target phosphorylation (GSK3β) and the functional importance of this was demonstrated by the ability of appropriate activating or inhibitory Akt constructs to significantly reverse the effects of MSTN or dnMSTN expression, respectively. Thus we suggest a previously unappreciated mechanism by which MSTN regulates cell size is through modulation of Akt activity. Although these observations were initially made in cardiomyocytes, they likely also apply to skeletal muscle, where baseline MSTN expression is higher than in heart.
Interestingly, MSTN did not suppress cardiomyocyte growth or Akt phosphorylation induced by LIF or IGF-I. These observations suggested MSTN modulates PE-induced cardiomyocyte growth through effects on upstream pathways different from those used by LIF or IGF-I. An examination of MAP kinase signaling revealed that MSTN also inhibited p38 phosphorylation. Previous work in other cell types has demonstrated p38 can function upstream of Akt.20,21,25 Moreover, p38 activation is both necessary and sufficient for PE-induced cardiomyocyte hypertrophy,22 whereas LIF-mediated hypertrophy depends less on p38 activation.23 IGF-I does not activate p38 and activates Akt through p38-independent mechanisms. In cardiomyocytes, p38 inhibition with a pharmacological inhibitor or DNp38α blocked PE-induced Akt phosphorylation. Expression of activated MKK3b, which preferentially activates p38α, rescued Akt phosphorylation in PE-stimulated, MSTN-expressing cardiomyocytes. Although causality is difficult to definitively establish, taken together these data suggest that MSTN inhibits PE-induced cardiomyocyte growth and Akt phosphorylation by inhibiting p38α activation upstream of MKK3. Nevertheless, we recognize that other pathways likely also contribute.
There are several possible explanations for the lack of p38 phosphorylation after dnMSTN expression. First, this may indicate the involvement of pathways beyond those considered here. Because we do not believe that MSTN interacts directly with p38 or Akt, this could still be consistent with our overall model. Second, it seems likely that p38 phosphorylation in whole-cell lysates is an imperfect reflection of the specific subcellular compartments important in this context. Finally, although dnMSTN effectively inhibits MSTN, it is possible that dnMSTN interacts with other homologous members of the TGF-β superfamily, 1 of which was recently reported to modulate cardiac Akt signaling,26 or other activin type II receptor ligands postulated to regulate cell growth in conjunction with MSTN.27 Thus additional effects of dnMSTN may obscure effects on p38. For all of these reasons, we chose to focus additional efforts on the MSTN knockout model, which is not confounded in these ways. Importantly, studies in MSTN knockout mice revealed enhanced phosphorylation of p38 after PE treatment in association with increased Akt activation and an exaggerated growth response thus reinforcing our overall model.
The role of p38 in cardiomyocyte hypertrophy is complex. In vitro activation of p38 in neonatal rat cardiomyocytes appears to be prohypertrophic,22,28 whereas the in vivo situation is less clear. Transgenic overexpression of DNp38α and DNMKK3b result in cardiac hypertrophy,29 suggesting an antihypertrophic role for p38 in vivo. However, cardiac overexpression of TGF-β–activated kinase-1 (TAK1) induces p38 activation and cardiac hypertrophy,30 and expression of MAP kinase phosphatase-1 (MKP-1) reduces p38 activity while mitigating hypertrophy after pressure overload.31 Thus the functional role of p38 in cardiac hypertrophy may be context dependent, consistent with the observation that MKK-independent pathways can also activate p38 with divergent effects.12,32
It remains unclear why 8 week-old male MSTN−/− hearts were slightly but paradoxically smaller at baseline. However, a comparable reduction in heart size was also seen in another line of MSTN−/− mice.33 This could reflect a distinct role for MSTN during development or secondary effects of systemic MSTN deletion. Similarly, in vivo inactivation of modulatory calcineurin-interacting protein (MCIP), which inhibits cardiomyocyte hypertrophy, results in smaller hearts at baseline.34
Interestingly, we found that male MSTN−/− mice had enhanced cardiac growth after PE infusion in vivo, whereas females did not. This is consistent with prior work demonstrating that α1-adrenergic signaling is critical for physiological cardiac growth in male but not female mice.35 Moreover, in a model of adrenergic-induced cardiac hypertrophy and death, female mice were relatively resistant, apparently because of an estrogen-inducible increase in MAP kinase phosphatase-1 (MKP-1) counteracting adrenergic-mediated phosphorylation of p38.36 Thus, the lack of an effect in female MSTN−/− could reflect gender-specific modulation of p38 as well as its involvement in both MSTN and adrenergic signaling. In addition, it is also possible that MSTN has p38-independent, gender-specific effects. Of note, muscle-specific transgenic MSTN overexpression decreases muscle mass in male but not female mice,37 and MSTN has been reported to inhibit expression of ARA-70, an androgen receptor cofactor.38
These findings may have clinical implications for ongoing efforts to inhibit MSTN in skeletal muscle diseases10 or as an unapproved means for enhanced athletic performance.39 The experiments described here do not accurately model inhibition initiated in an adult. However, these data do suggest that clinical studies of MSTN inhibitors should consider potential cardiac effects and raise reasonable hypotheses as to what those effects might be. For example, MSTN inhibition might lead to enhanced cardiac growth or even cardiac hypertrophy, and men may be particularly susceptible to this. If the signaling changes observed in these studies are conserved, treatment with MSTN inhibitors might enhance Akt activation. We would predict that this effect would be cardioprotective over the short term40 but could lead to deleterious effects over time.11 All of these considerations may also be relevant to people with inactivating MSTN mutations.8 However, we must emphasize that there is currently no evidence from clinical studies that MSTN mutation or inhibition causes cardiac complications in humans. The current work should simply prompt careful and longitudinal evaluation of the cardiac status of such patients.
In summary, we have found that MSTN, a potent negative regulator of skeletal muscle growth, is dynamically regulated in the heart and acts to modulate cardiomyocyte growth in a stimulus-specific manner. Inhibition by MSTN of the growth response to the α1-adrenergic stimulus, PE, appears mediated through inhibition of p38α and Akt. Together these studies suggest MSTN acts more broadly than initially postulated to regulate growth of cardiac, as well as skeletal, striated muscle, which may have implications for the design of clinical trials with MSTN inhibitors. Moreover, MSTN modulation of Akt and p38 signaling could provide insight into its mechanism of action in other systems.
We thank Dr Se Jin Lee (Johns Hopkins University, Baltimore, Md) for MSTN−/− mice, Dr Yibin Wang (University of California, Los Angeles) for the DNp38 and MKK3bE adenoviruses, and Dr Serafima Zaltsman for expert management of our mouse colony.
Sources of Funding
Supported by the NIH (HL-59521 and HL-61557 [to A.R.]; 5T32HL07208–25 [to M.R.M.]; HL-04250 [to T.M.]), an American Heart Association Postdoctoral Fellowship (to M.R.M.), the Wellcome Trust (to S.A.C.), and the British Heart Foundation (PG/04/053/17031 [to G.B. and G.M.]).
Original received February 21, 2005; resubmission received December 12, 2005; revised resubmission received May 17, 2006; accepted May 26, 2006.
Matsui T, Li L, Wu JC, Cook SA, Nagoshi T, Picard MH, Liao R, Rosenzweig A. Phenotypic spectrum caused by transgenic overexpression of activated Akt in the heart. J Biol Chem. 2002; 277: 22896–22901.
Bodine SC, Stitt TN, Gonzalez M, Kline WO, Stover GL, Bauerlein R, Zlotchenko E, Scrimgeour A, Lawrence JC, Glass DJ, Yancopoulos GD. Akt/mTOR pathway is a crucial regulator of skeletal muscle hypertrophy and can prevent muscle atrophy in vivo. Nat Cell Biol. 2001; 3: 1014–1019.
Cook SA, Matsui T, Li L, Rosenzweig A. Transcriptional effects of chronic Akt activation in the heart. J Biol Chem. 2002; 277: 22528–22533.
Nagoshi T, Matsui T, Aoyama T, Leri A, Anversa P, Li L, Ogawa W, del Monte F, Gwathmey JK, Grazette L, Hemmings BA, Kass DA, Champion HC, Rosenzweig A. PI3K rescues the detrimental effects of chronic Akt activation in the heart during ischemia/reperfusion injury. J Clin Invest. 2005; 115: 2128–2138.
Lu G, Kang YJ, Han J, Herschman HR, Stefani E, Wang Y. TAB-1 modulates intracellular localization of p38 MAP kinase and downstream signaling. J Biol Chem. 2006; 281: 6087–6095.
Matsui T, Li L, del Monte F, Fukui Y, Franke TF, Hajjar RJ, Rosenzweig A. Adenoviral gene transfer of activated PI 3-kinase and Akt inhibits apoptosis of hypoxic cardiomyocytes in vitro. Circulation. 1999; 100: 2373–2379.
Genomics of Cardiovascular Development, Adaptation, and Remodeling. NHLBI Program for Genomic Applications, Harvard Medical School. Available at: http://www.cardiogenomics.org.
Hill JJ, Davies MV, Pearson AA, Wang JH, Hewick RM, Wolfman NM, Qiu Y. The myostatin propeptide and the follistatin-related gene are inhibitory binding proteins of myostatin in normal serum. J Biol Chem. 2002; 277: 40735–40741.
Simpson PC, Long CS, Waspe LE, Henrich CJ, Ordahl CP. Transcription of early developmental isogenes in cardiac myocyte hypertrophy. J Mol Cell Cardiol. 1989; 21 (suppl 5): 79–89.
Haq S, Choukroun G, Kang ZB, Ranu H, Matsui T, Rosenzweig A, Molkentin JD, Alessandrini A, Woodgett J, Hajjar R, Michael A, Force T. Glycogen synthase kinase-3beta is a negative regulator of cardiomyocyte hypertrophy. J Cell Biol. 2000; 151: 117–130.
Antos CL, McKinsey TA, Frey N, Kutschke W, McAnally J, Shelton JM, Richardson JA, Hill JA, Olson EN. Activated glycogen synthase-3 beta suppresses cardiac hypertrophy in vivo. Proc Natl Acad Sci U S A. 2002; 99: 907–912.
Rane MJ, Coxon PY, Powell DW, Webster R, Klein JB, Pierce W, Ping P, McLeish KR. p38 Kinase-dependent MAPKAPK-2 activation functions as 3-phosphoinositide-dependent kinase-2 for Akt in human neutrophils. J Biol Chem. 2001; 276: 3517–3523.
Horowitz JC, Lee DY, Waghray M, Keshamouni VG, Thomas PE, Zhang H, Cui Z, Thannickal VJ. Activation of the pro-survival phosphatidylinositol 3-kinase/AKT pathway by transforming growth factor-beta1 in mesenchymal cells is mediated by p38 MAPK-dependent induction of an autocrine growth factor. J Biol Chem. 2004; 279: 1359–1367.
Zechner D, Thuerauf DJ, Hanford DS, McDonough PM, Glembotski CC. A role for the p38 mitogen-activated protein kinase pathway in myocardial cell growth, sarcomeric organization, and cardiac-specific gene expression. J Cell Biol. 1997; 139: 115–127.
Ng DC, Long CS, Bogoyevitch MA. A role for the extracellular signal-regulated kinase and p38 mitogen-activated protein kinases in interleukin-1 beta-stimulated delayed signal transducer and activator of transcription 3 activation, atrial natriuretic factor expression, and cardiac myocyte morphology. J Biol Chem. 2001; 276: 29490–29498.
Kempf T, Eden M, Strelau J, Naguib M, Willenbockel C, Tongers J, Heineke J, Kotlarz D, Xu J, Molkentin JD, Niessen HW, Drexler H, Wollert KC. The transforming growth factor-beta superfamily member growth-differentiation factor-15 protects the heart from ischemia/reperfusion injury. Circ Res. 2006; 98: 351–360.
Lee SJ, Reed LA, Davies MV, Girgenrath S, Goad ME, Tomkinson KN, Wright JF, Barker C, Ehrmantraut G, Holmstrom J, Trowell B, Gertz B, Jiang MS, Sebald SM, Matzuk M, Li E, Liang LF, Quattlebaum E, Stotish RL, Wolfman NM. Regulation of muscle growth by multiple ligands signaling through activin type II receptors. Proc Natl Acad Sci U S A. 2005; 102: 18117–18122.
Wang Y, Huang S, Sah VP, Ross J Jr, Brown JH, Han J, Chien KR. Cardiac muscle cell hypertrophy and apoptosis induced by distinct members of the p38 mitogen-activated protein kinase family. J Biol Chem. 1998; 273: 2161–2168.
Braz JC, Bueno OF, Liang Q, Wilkins BJ, Dai YS, Parsons S, Braunwart J, Glascock BJ, Klevitsky R, Kimball TF, Hewett TE, Molkentin JD. Targeted inhibition of p38 MAPK promotes hypertrophic cardiomyopathy through upregulation of calcineurin-NFAT signaling. J Clin Invest. 2003; 111: 1475–1486.
Bueno OF, De Windt LJ, Lim HW, Tymitz KM, Witt SA, Kimball TR, Molkentin JD. The dual-specificity phosphatase MKP-1 limits the cardiac hypertrophic response in vitro and in vivo. Circ Res. 2001; 88: 88–96.
Ge B, Gram H, Di Padova F, Huang B, New L, Ulevitch RJ, Luo Y, Han J. MAPKK-independent activation of p38alpha mediated by TAB1-dependent autophosphorylation of p38alpha. Science. 2002; 295: 1291–1294.
Vega RB, Rothermel BA, Weinheimer CJ, Kovacs A, Naseem RH, Bassel-Duby R, Williams RS, Olson EN. Dual roles of modulatory calcineurin-interacting protein 1 in cardiac hypertrophy. Proc Natl Acad Sci U S A. 2003; 100: 669–674.
O’Connell TD, Ishizaka S, Nakamura A, Swigart PM, Rodrigo MC, Simpson GL, Cotecchia S, Rokosh DG, Grossman W, Foster E, Simpson PC. The alpha(1A/C)- and alpha(1B)-adrenergic receptors are required for physiological cardiac hypertrophy in the double-knockout mouse. J Clin Invest. 2003; 111: 1783–1791.
Dash R, Schmidt AG, Pathak A, Gerst MJ, Biniakiewicz D, Kadambi VJ, Hoit BD, Abraham WT, Kranias EG. Differential regulation of p38 mitogen-activated protein kinase mediates gender-dependent catecholamine-induced hypertrophy. Cardiovasc Res. 2003; 57: 704–714.
Reisz-Porszasz S, Bhasin S, Artaza JN, Shen R, Sinha-Hikim I, Hogue A, Fielder TJ, Gonzalez-Cadavid NF. Lower skeletal muscle mass in male transgenic mice with muscle-specific overexpression of myostatin. Am J Physiol Endocrinol Metab. 2003; 285: E876–E888.
Sweeney HL. Gene doping. Sci Am. 2004; 291: 62–69.
Matsui T, Tao J, del Monte F, Lee KH, Li L, Picard M, Force TL, Franke TF, Hajjar RJ, Rosenzweig A. Akt activation preserves cardiac function and prevents injury after transient cardiac ischemia in vivo. Circulation. 2001; 104: 330–335.