Muscle development in childhood and muscle regeneration in adults are highly regulated processes that are necessary for reaching and maintaining optimal muscle mass and strength throughout life. Muscle repair after injury relies on stem cells, termed satellite cells, whose activity is controlled by complex signals mediated by cell-cell contact, by growth factors, and by hormones, which interact with genetic programs controlled by myogenic transcription factors. Insulin-like growth factors (IGFs) play key roles in muscle development and help coordinate muscle repair after injury, primarily by stimulating the phosphatidylinositol 3-kinase-Akt signaling pathway, and both in vitro and in vivo studies have shown that Akt kinase activity is critical for optimal muscle growth and regeneration. Here we find that of the two Akts expressed in muscle, Akt1 is essential for initiation of differentiation in culture and is required for normal myoblast motility, while Akt2 is dispensable. Although Akt2 deficiency did lead to diminished myotube maturation, as assessed by a decline in myofiber area and in fusion index, either Akt1 or Akt2 could restore these processes toward normal. Thus levels of Akt expression rather than distinct actions of individual Akt species are critical for normal myofiber development during the later stages of muscle differentiation.
- skeletal muscle differentiation
- myofiber fusion
sarcopenia, defined as the physiologically critical loss of skeletal muscle mass and strength during disease or aging (9), has been estimated to affect over 25% of elderly individuals, and to account for medical costs in the billions (9, 32). The health risks from the frailty that accompanies sarcopenia are substantial and include diminished physical activity, increased rates of falls and fractures, and prolonged rehabilitation after illness, which further accentuates the decline in muscle mass and strength (31). Therapy for sarcopenia is inadequate (31, 33), primarily because the biochemical mechanisms responsible for limiting muscle loss have not been elucidated.
Multiple signaling pathways activated by different growth factors and hormones positively and negatively influence muscle growth, regeneration, and metabolism (4, 38). Among these signaling networks, there is substantial evidence that actions of the insulin-like growth factors (IGFs) are crucial for normal muscle development (27, 42) and can enhance muscle growth and repair (2, 6). The IGF family consists of two growth factors, IGF-I and IGF-II, two receptors, IGF-IR and IGF-IIR, and six high-affinity IGF-binding proteins (10). Actions of both IGFs are mediated by the IGF-IR, a ligand-activated tyrosine protein kinase related to the insulin receptor (23). The importance of IGF actions in muscle is supported by many experimental observations. IGF-IR deficiency in mice caused neonatal death secondary to respiratory failure from marked muscle hypoplasia and weakness (19). By contrast, targeted overexpression of IGF-I in mice stimulated increased muscle mass (1), enhanced muscle growth responses to exercise (25), and counteracted symptomatic development of experimental muscular dystrophy (2). In addition, a quantitative trait locus in pigs associated with elevated muscle mass was mapped to a single nucleotide polymorphism in the IGF-II gene (39) that was responsible for boosting IGF-II gene expression in muscle cells by interfering with binding of a transcriptional repressor (5, 20).
Stimulation of the IGF-IR triggers its tyrosine kinase function and initiates protein-protein interactions that activate multiple intracellular signaling pathways (23). While several IGF-regulated signaling networks exert actions in myoblasts, the phosphatidylinositol 3-(PI3)-kinase-Akt pathway appears to be critical for muscle development, growth, and repair (3, 11, 17, 24). Mice globally lacking both Akt1 and Akt2 had a severe muscle deficiency phenotype that resembled loss of the IGF-IR (26). In contrast, transgenic mice expressing a constitutively active Akt in muscle had extensive hypertrophy (17). Akt actions also can prevent muscle atrophy by directly inhibiting FoxO transcription factors by phosphorylation, thereby blocking induction of atrogin-1/MAFbx and MuRF1, E3 ubiquitin ligases that promote muscle protein breakdown (34, 36). Other growth factors also use PI3-kinase and Akt to enhance muscle growth. Wnt7 can promote myofiber hypertrophy through association of its receptor Fzd7 with PI3-kinase and consequent stimulation of Akt (41), and blocking FoxO protein activity can reverse the inhibitory effects of Notch on muscle differentiation (16).
In previous studies using cells with targeted knockdown or genetic knockout, we found that Akt1 was required for muscle differentiation (43, 30). In contrast, Akt2 appeared to play a more limited role in myofiber maturation, as its loss had little effect on muscle gene or protein expression, but the resultant Akt2-deficient myotubes were smaller and thinner than controls (43, 30). Here we have examined the impact of restoration of Akt2 or replacement with Akt1 on myofiber formation and conclude that levels of Akt rather than specific effects of individual Akt isoforms are necessary for normal myotube development during the later stages of muscle differentiation.
MATERIALS AND METHODS
Fetal and newborn calf serum was purchased from Hyclone (Logan, Utah). Horse serum, goat serum, DMEM, PBS, and trypsin/EDTA solution were obtained from Life Technologies (Carlsbad, CA). Other items were purchased from the following suppliers: protease inhibitor tablets complete mini (Roche, Indianapolis, IN); porcine gelatin, paraformaldehyde, and sodium orthovanadate (Sigma, St. Louis, MO); okadaic acid MP (Biomedicals, Solon, OH); Hoechst 33258 nuclear dye (Polysciences, Warrington, PA); BCA protein assay kit (Pierce Biotechnology, Rockford, IL); Immobilon-FL (Millipore, Billerico, MA); AquaBlock tm/EIA/WIB solution (East Coast Biologicals, North Berwick, ME); and R3-IGF-I (GroPep, Adelaide, Australia). Primary antibodies were from the following suppliers: Akt1 (Abcam, Cambridge, UK); MyoD (BD-Pharmigen, San Diego, CA); α-tubulin (Sigma); and Akt2, total Akt, phospho-FoxO1 (Thr24)/FoxO3a (Thr32), phospho-AktSer473, and phospho-AktThr308 (Cell Signaling, Beverly, MA). Myogenin (F5D from W. E. Wright), creatine kinase (CK-JAC from G.E. Morris), myosin heavy chain (MF20 from D.A. Fischman), and troponin-T (CT3 from J. J-C. Lin) were purchased from the Developmental Studies Hybridoma Bank (Iowa City, IA). Secondary antibodies included AlexaFluor 680/594/488-conjugated-goat anti-mouse IgG and anti-rabbit IgG (Life Technologies) and goat-anti-mouse IgG-IR800 (Rockland Immunochemical, Gilbertsville, PA). All other chemicals were reagent grade and were purchased from commercial suppliers.
Recombinant adenoviruses and lentiviruses.
Recombinant lentiviruses were generated to express enhanced green fluorescent protein (EGFP), mouse Akt1, and mouse Akt2. Akt cDNAs were cloned from mouse C2 myoblast RNA and were inserted into the PmeI and EcoRI sites of the lentiviral plasmid pWPXLd (by D. Trono, from Addgene, Cambridge, MA). Lentiviruses were prepared by transfecting pWPXLd with packaging plasmids pMDLg/pRRE, pRSV-Rev, and pMD2g (from D. Trono via Addgene) into HEK293FT cells. Two days later, conditioned medium was collected, and lentiviruses were concentrated by centrifugation in a SW28 rotor at 19,000 rpm for 2 h at 4°C. The pellet was resuspended in PBS with 1% BSA, aliquoted, and stored at −80°C until use. The following recombinant adenoviruses have been described: Ad-MyoD, Ad-shAkt1, and Ad-shAkt2 (43). After amplification in HEK293 cells, they were purified by ultracentrifugation through CsCl density gradients and titered by optical density. Before use, all viruses were diluted in DMEM plus 2% fetal calf serum and filtered through a 0.45-μM Gelman syringe filter.
Cells were incubated at 37°C in humidified air with 5% CO2. C2 myoblasts (45) were grown on gelatin-coated tissue culture dishes in DMEM with 10% heat-inactivated fetal calf serum and 10% newborn calf serum. C3H10T1/2 mouse embryonic fibroblasts (ATCC CCL226) and embryonic fibroblasts from Akt1 and Akt2 double knockout mice (28, 47) were incubated on gelatin-coated tissue culture dishes in DMEM with 10% fetal calf serum. Muscle differentiation was induced with cells at ∼95% of confluent density by replacement of growth medium with DMEM plus 2% horse serum (differentiation medium). For lentiviral infection, cells were seeded in 6-cm diameter dishes at 5 × 104 cells/dish and incubated for 3 h. Lentiviruses were then added in the presence of 6 μg/ml polybrene, and cells were incubated until they reached ∼90% of confluent density, at which point they were seeded for experiments. Over 90% of cells expressed the recombinant protein, and protein expression persisted at comparable levels for more than five additional passages. For adenoviral infection, cells were seeded until 50% of confluent density was reached and were infected with one or two recombinant adenoviruses as described previously (42).
Protein extraction and immunoblotting.
Whole cell protein lysates were prepared as described (42) and were stored at −80°C until use. Protein concentrations were determined with the BCA protein assay kit, and aliquots (15 μg per lane) were separated by SDS-PAGE, transferred to Immobilon-FL membranes, blocked in AquaBlock, and incubated with primary and secondary antibodies. Primary antibodies were used at the following dilutions: anti-myogenin and anti-myosin heavy chain (1:100), anti-phospho-FoxO1/FoxO3a, anti-phospho-Akt, anti-Akt2, anti-total Akt, anti-MyoD (1:500), anti-α-tubulin (1:10,000), all others (1:1,000), and secondary antibodies at 1:5,000. Results were visualized and images captured and quantified using an Odyssey Infrared Imaging System and v3.0 analysis software (LiCoR Biosciences, Lincoln, NB).
Analysis of gene expression.
Whole cell RNA (2 μg), isolated as described previously (21), was reverse-transcribed with the Superscript III first-strand synthesis kit using oligo (dT) primers in a final volume of 20 μl. PCR was performed with 1 μl of cDNA per reaction and previously published primer pairs for mouse MyoD, S17, Akt1, and Akt2 (21, 43). Cycle numbers for PCR were within the linear range for each primer pair and ranged from 20–30. Results were visualized after agarose gel electrophoresis.
Immunocytochemistry and image analysis.
Cells were fixed using 4% paraformaldehyde, followed by methanol:acetone permeabilization and incubation with antibodies as described (42). Primary antibodies were added in blocking buffer containing goat serum for 16 h at 4°C. Secondary antibody was added at 1:2,000 dilution, and Hoechst stain (1:1,000) was added in the dark for 1.5 h. Images were captured using a Nikon DS-Qi1Mc camera attached to a Nikon Eclipse Ti-U inverted microscope using the NIS elements 3.1 software. Image analysis was performed with this software. Results are presented as the means ± SE of at least three experiments, with each experimental point consisting of five randomly captured microscopic fields for each treatment group. Myotube area was calculated by measuring the percentage of troponin-T-positive staining in a microscopic field. The fusion index is defined as the total number of nuclei (excluding mononucleated) in troponin-T-positive cells divided by the total number of nuclei per field. The number of nuclei per myotube was established by counting nuclei within every myotube per microscopic field divided by the number of myotubes in the field, where a myotube is defined as a troponin-T-positive cell with at least two nuclei. Myotube length and width were established by measuring the 5 longest or widest myotubes in each individual microscopic field, with a total of ≥75 myotubes per treatment group being measured. Panoramic images were selected by taking multiple ×100 images in sequence and then utilizing the NIS elements software to stitch images together.
Live cell imaging.
Ad-MyoD 10T1/2 cells were infected with Ad-EGFP, Ad-shAkt1, or Ad-shAkt2 as described previously (42). Once cells reached confluent density, growth medium was replaced with differentiation medium and the plate was placed within the IncuCyte imaging system (Essen Bioscience, Ann Arbor, MI), which resides within a tissue culture incubator. The same microscopic fields (3/well) were imaged at 15-min intervals for 48 h. The resulting data were registered and cells manually tracked using ImageJ imaging software (W. S. Rasband, ImageJ, National Institutes of Health, Bethesda, MD).
Data are presented as the means ± SE. Statistical significance was determined using paired Students t-test, with results presented in the figure legends. Results were considered statistically significant when P ≤ 0.05.
Akt2 deficiency does not inhibit muscle differentiation but reduces myofiber development.
During muscle differentiation, Akt2 mRNA and protein levels increase (13, 15, 40) and Akt phosphorylation and enzymatic activity are enhanced (Fig. 1B and Ref. 42). We previously found that loss of Akt1 blocked initiation of muscle differentiation but that inhibition of Akt2 expression had a more subtle effect in the C2 muscle cell line and in myoblasts derived from mouse embryo fibroblasts (MEFs) from Akt-deficient mice (30). As seen in Fig. 1, in 10T1/2 cells acutely converted to the myoblast lineage by MyoD (Ad-MyoD infection), knockdown of Akt1 by adenovirus-mediated delivery of a short-hairpin interfering RNA (Ad-shAkt1) decreased Akt1 mRNA and protein levels (Fig. 1, B–D), impaired induction of muscle protein expression, and completely inhibited myotube formation (Fig. 1, B, C, and E). In contrast, reduction in Akt2 levels by > 90% by Ad-shAkt2 (Fig. 1, B–D), which caused a > 50% decline in Akt phosphorylation, did not alter the pattern or levels of expression of muscle differentiation-specific proteins and did not prevent myotube formation (Fig. 1, B, C, and E). However, the quality of myotubes lacking Akt2 was different from controls, as judged by a reduction in the number of large multinucleated myofibers, diminished myotube area, and a lowered fusion index (Fig. 1, D and E).
Loss of Akt1 results in reduced myoblast motility.
Further examination of Akt1- and Akt2-deficient myoblasts by dynamic live cell imaging revealed that loss of Akt1 impaired cell motility (Fig. 2 and Supplemental Movies S1–S3; Supplemental Material for this article is available online at the Am J Physiol Cell Physiol website). As seen by composite frame-by-frame cell tracking, myoblast movement was minimal in the absence of Akt1 but was normal with Akt2 deficiency (Fig. 2B). Graphical representation of the results demonstrates that loss of Akt1 resulted in a >80% decline in cell movement during incubation in differentiation medium for 48 h but that knockdown of Akt2 did not reduce motility significantly when compared with controls (Fig. 2C).
Defining the subcellular distribution of Akt1 and Akt2 during muscle differentiation.
We examined the location of Akts within differentiating myoblasts by immunocytochemistry. Although before the onset of differentiation, Akt1 was diffusely distributed throughout the cytoplasm in control cells, as detected with a pan-Akt antibody, minimal Akt phosphorylation was observed, which was confined to <10% of myoblasts (Fig. 3A, left). Although the subcellular distribution of Akt did not change after 48 h of differentiation, the overall intensity increased, reflecting the rise in Akt concentrations secondary to induction of Akt2 expression (Fig. 3A, right, and Ref. 42), and this was accompanied by a dramatic enhancement of Akt phosphorylation (Fig. 3A, right middle). Higher magnification images of phosphorylated Akt revealed protein accumulation in membrane protrusions, including at sites of myofiber fusion (Fig. 3C, arrows). In myoblasts lacking Akt2, the pattern of expression of Akt1 and levels of Akt phosphorylation before onset of differentiation were unchanged from controls (Fig. 3B, left), as anticipated since Akt2 is minimally expressed at this time (Fig. 1, B and C, and Ref. 30). In contrast, after 48 h of differentiation, the intensity of Akt phosphorylation was dramatically reduced in cells with targeted knockdown of Akt2 (Fig. 3B, right middle).
Forced expression of Akt1 or Akt2 in myoblasts lacking Akt2 restores myofiber development.
To determine if Akt2 had specific actions during myofiber formation, we expressed a version of mouse Akt2 that was resistant to knockdown by short-hairpin interfering RNA [using transduction by recombinant lentivirus (LV)] and compared its effects to those obtained with mouse Akt1 or with a negative control, EGFP (Fig. 4A). Both recombinant wild-type Akts could be expressed in cells at comparable levels, could be activated by IGF-I, and were able to phosphorylate substrate proteins to a similar extent (Supplemental Fig. S1). Reintroduction of Akt2 in myoblasts lacking Akt2, or substitution with Akt1, each led to an increase in total Akt levels at all time points and boosted the amount of Akt activated during muscle differentiation to a similar extent, as measured by increased phosphorylation on both threonine 308 and serine 473 (Fig. 4B). However, despite an apparently greater degree of Akt activity compared with controls transduced with LV-EGFP, neither Akt1 nor Akt2 caused a rise in biochemical parameters of muscle differentiation, as indicated by similar levels of myogenin, myosin heavy chain, muscle creatine kinase, and troponin-T over 48 h in differentiation-promoting conditions (Fig. 4, B and C).
Forced Akt expression did have an impact on myotubes, as reintroduction of Akt2 or substitution of Akt1 both led to an increase in myofiber area and fusion index when compared with controls lacking Akt2 (Fig. 4, D and E). Moreover, both Akt1 and Akt2 stimulated a 35% rise in overall myofiber length (Fig. 4E). Similar results were seen with Akt2 in C2 myoblasts (Supplemental Fig. S2).
Overexpression of Akt1 or Akt2 enhances myofiber formation but does not increase muscle protein expression.
We next assessed the effects of overexpression of Akt1 or Akt2 on muscle differentiation. LV-Akt1 or LV-Akt2 was delivered to 10T1/2 cells before their conversion to myoblasts with Ad-MyoD, and differentiation was induced (see experimental scheme in Fig. 5A). Both Akts were readily expressed and led to an upregulation of levels of Akt phosphorylation during differentiation when compared with controls transduced with LV-EGFP (Fig. 5B), but neither Akt1 nor Akt2 caused any increase in abundance of muscle proteins compared with LV-EGFP transduced controls (Fig. 5, B and C). However, overexpression of each Akt did lead to an increase in myofiber formation, as measured both by a rise in myotube area at 24 and 48 h and by an upregulation of the fusion index and the average number of nuclei per myotube (Fig. 5, D and E). Similar effects were observed when Akt2 was overexpressed in C2 myoblasts (Supplemental Fig. S3). Taken together with results in Fig. 4, these observations support the idea that Akt levels rather than the type of Akt determines the rate and extent of myofiber development and maturation.
Multiple in vitro and in vivo studies have shown that Akt kinase activity is essential for normal skeletal muscle development, growth, regeneration, and metabolism (12, 22). In undifferentiated myoblasts, Akt1 is nearly an order of magnitude more abundant than Akt2, and its elimination prevented differentiation from progressing beyond initiation (30, 43), as we confirm here in 10T1/2 mesenchymal precursor cells acutely converted to the myoblast lineage by forced expression of MyoD. Lack of Akt1 also dramatically reduced myoblast motility. In contrast, Akt2 deficiency had a minimal effect on cell motility, or on early or intermediate phases of differentiation, as expression of muscle proteins and initial stages of myocyte fusion proceeded normally, but accumulation of large multinucleated myotubes was impaired. Similar results in other muscle cell lines had led us to hypothesize previously that Akt2 was not necessary for muscle differentiation but rather played a role in myofiber maturation (30, 43). We now find that either Akt1 or Akt2 can restore myotube formation in Akt2-deficient myoblasts, indicating that levels of Akt expression rather than actions of individual Akt species are critical for normal myofiber development.
Several published papers have linked increases in Akt activity with myofiber formation, primarily by showing that overexpression could lead to hypertrophy, although in all prior cases this was achieved through use of modified Akt proteins that bypassed normal regulatory constraints (3, 8, 17, 29, 37). In the most extreme example, a constitutively active Akt targeted to skeletal muscle in transgenic mice led to extensive hypertrophy (17), but analogous results were achieved with direct delivery of similarly active Akt isoforms to individual muscles in mice or rats (3, 8, 29, 37). Although in most previous experiments modified versions of Akt1 were used, Cleasby et al. (8) found that constitutively active Akt1 or Akt2 was able to promote equivalent hypertrophy of rat tibialis anterior muscle. Our studies extend these observations by showing that overexpression of wild-type Akt1 or Akt2 can promote hypertrophy, even though neither Akt was able to enhance the rate or extent of differentiation (Fig. 5), and by finding that each Akt can restore myofiber development toward normal in the absence of Akt2 (Fig. 4).
The effects of Akt-mediated signaling on cellular motility and migratory activity have been evaluated in multiple cell systems, with studies focusing primarily on actions that potentiate cancer cell metastasis (7, 14, 18, 46). In this context, analyses of breast carcinoma cell lines have concluded that overexpression of Akt1 inhibited motility and reduced metastatic potential (14, 18, 46) and have found that its targeted knockdown stimulated both epithelial to mesenchymal cell transition and invasiveness (14, 46). At least two mechanisms have been proposed to explain these observations. Akt1 has been shown to block the Mek-Erk signaling cascade, with Mek inhibition phenocopying effects of Akt1 overexpression (14); Akt1 also has been found to impair the actions of the actin-associated protein palladin (7). Surprisingly, limited studies in noncancer cell systems have reached different conclusions regarding the specificity of individual Akt isoforms on motility. In MEFs from Akt-deficient mice, Akt1 was needed for normal cell migration, and Akt2 functioned as a suppressor of motility (47). Similar results were obtained with endothelial cells from Akt1 knockout mice, where both adhesion to fibronectin and movement were impaired (35). The latter results are similar to our observations in Akt1-deficient myoblasts. To date, the biochemical mechanisms responsible for these potentially opposite functions of Akt1 and Akt2 on motility in normal and cancer cells have not been elucidated.
Our results provide further evidence demonstrating that the functions of Akt1 and Akt2 differ even in normal mesenchymal cell derivatives. As seen in mice with global Akt1 deficiency and in 3T3L1 cells with targeted loss of Akt1, it appears that Akt1 is essential for adipocyte differentiation (26, 44), similar to what we have observed for skeletal muscle differentiation (30, 43). In contrast, previously we have found that Akt2 is required for initiation of osteoblast differentiation from mesenchymal progenitors, including the 10T1/2 cell line and Akt-deficient MEFs and that Akt1 cannot substitute for loss of Akt2 (21). Taken together with the studies outlined above, these observations demonstrate that there are profound differences in biological actions of Akt1 and Akt2 both between normal and cancer cells, and among derivatives of normal cell types, and suggest that cell-type specific modifiers must influence the effects of individual Akts within a background of apparently shared signaling pathways.
In summary, we have identified unique actions of Akt1 in permitting initiation of muscle differentiation and maintaining myoblast motility and have determined that an increase in Akt levels and activity are necessary for the later phases of myofiber maturation. A key unresolved problem is the identification of the critical downstream signaling molecules responsible for the specific biological effects of individual Akts. Understanding the steps in Akt activation has the potential to enhance therapeutic options for muscle disorders and for treatment of sarcopenia during aging and in systemic disease (9, 32).
This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant 5R01-DK-042748–23 (to P. Rotwein).
No conflicts of interest, financial or otherwise, are declared by the author(s).
Author contributions: S.G. and P.R. conception and design of research; S.G. and M.A. performed experiments; S.G., M.A., and P.R. analyzed data; S.G. and P.R. interpreted results of experiments; S.G. prepared figures; S.G. and P.R. drafted, edited and revised manuscript; all authors approved final version of manuscript.
We thank Trey Knollman of Essen Bioscience for the use of IncuCyte imaging system. Mouse embryonic fibroblasts from Akt1/Akt2 double knockout mice were obtained from Dr. Morris Birnbaum, University of Pennsylvania School of Medicine.
- Copyright © 2012 the American Physiological Society