Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You and Your Team.

Learn More →

Absence of γ-sarcoglycan alters the response of p70S6 kinase to mechanical perturbation in murine skeletal muscle

Absence of γ-sarcoglycan alters the response of p70S6 kinase to mechanical perturbation in murine... Background: The dystrophin glycoprotein complex (DGC) is located at the sarcolemma of muscle fibers, providing structural integrity. Mutations in and loss of DGC proteins cause a spectrum of muscular dystrophies. When only the sarcoglycan subcomplex is absent, muscles display severe myofiber degeneration, but little susceptibility to contractile damage, suggesting that disease occurs not by structural deficits but through aberrant signaling, namely, loss of normal mechanotransduction signaling through the sarcoglycan complex. We extended our previous studies on mechanosensitive, γ-sarcoglycan-dependent ERK1/2 phosphorylation, to determine whether additional pathways are altered with the loss of γ-sarcoglycan. Methods: We examined mechanotransduction in the presence and absence of γ-sarcoglycan, using C2C12 -/- myotubes, and primary cultures and isolated muscles from C57Bl/6 (C57) and γ-sarcoglycan-null (γ-SG )mice. All were subjected to cyclic passive stretch. Signaling protein phosphorylation was determined by immunoblotting of lysates from stretched and non-stretched samples. Calcium dependence was assessed by maintaining muscles in calcium-free or tetracaine-supplemented Ringer’s solution. Dependence on mTOR was determined by stretching isolated muscles in the presence or absence of rapamycin. Results: C2C12 myotube stretch caused a robust increase in P-p70S6K, but decreased P-FAK and P-ERK2. Neither Akt nor ERK1 were responsive to passive stretch. Similar but non-significant trends were observed in C57 primary -/- cultures in response to stretch, and γ-SG cultures displayed no p70S6K response. In contrast, in isolated muscles, -/- p70S6K was mechanically responsive. Basal p70S6K activation was elevated in muscles of γ-SG mice, in a -/- calcium-independent manner. p70S6K activation increased with stretch in both C57 and γ-SG isolated muscles, -/- and was sustained in γ-SG muscles, unlike the transient response in C57 muscles. Rapamycin treatment blocked all of p70S6K activation in stretched C57 muscles, and reduced downstream S6RP phosphorylation. However, even -/- though rapamycin treatment decreased p70S6K activation in stretched γ-SG muscles, S6RP phosphorylation remained elevated. Conclusions: p70S6K is an important component of γ-sarcoglycan-dependent mechanotransduction in skeletal muscle. Our results suggest that loss of γ-sarcoglycan uncouples the response of p70S6K to stretch and implies that γ-sarcoglycan is important for inactivation of this pathway. Overall, we assert that altered load-sensing mechanisms exist in muscular dystrophies where the sarcoglycans are absent. Keywords: Sarcoglycan, Sarcoglycanopathies, Limb girdle muscular dystrophy, Mechanotransduction, Mechano-sensing, Load-sensing, p70S6K, S6K, p70S6 kinase, ERK1/2 * Correspondence: erbarton@dental.upenn.edu Department of Anatomy and Cell Biology, School of Dental Medicine, University of Pennsylvania, Philadelphia, PA, USA Pennsylvania Muscle Institute, University of Pennsylvania, Philadelphia, PA, USA Full list of author information is available at the end of the article © 2014 Moorwood et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Moorwood et al. Skeletal Muscle 2014, 4:13 Page 2 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 Background the muscular dystrophies results from abnormal levels and The dystrophin glycoprotein complex (DGC) is found at activity of the mechanosensitive TRP channels and/or mis- the sarcolemma of skeletal, cardiac, and smooth muscle regulation of store operated calcium entry via the STIM1 cells, where it physically links the extracellular matrix and Orai1 channels [24-29]. Direct disruption of the sarco- (ECM) with the intracellular cytoskeleton, providing struc- lemma, for which there is evidence in SG-null animal tural support [1-3]. Mutations in DGC components cause models [7,9,30,31], could also contribute to loss of calcium different types of muscular dystrophy; for example, muta- homeostasis. Furthermore, several strategies to improve 2+ tions in dystrophin cause Duchenne muscular dystrophy Ca handling are known to counteract the pathology as- (DMD), while mutations in α-, β-, γ-, or δ-sarcoglycan sociated with the muscular dystrophies [25,32-34]. There- (SG) cause limb girdle muscular dystrophy (LGMD) [2-4]. fore, identification of mechanosensitive signaling that is When dystrophin is mutated in DMD or in the mdx attributable to the SG complex rather than other pro- mouse model of DMD, the entire DGC is substantially cesses occurring during mechanical perturbation has been reduced at the sarcolemma. In contrast, when any one of challenging. the SGs is mutated, in LGMD or any of the SG knock-out One pathway of interest involves p70S6K, which is ca- mice, the other three SGs are either absent or reduced at nonically activated in response to mitogens via the phos- the sarcolemma, but the rest of the DGC remains, includ- phoinositide 3-kinase (PI3K) pathway (reviewed in [35]) ing the link formed by dystrophin and dystroglycan be- and is known to respond to mechanical load [36]. Acti- tween the ECM and the cytoskeleton [2,3,5,6]. Unlike the vation of p70S6K involves a hierarchical series of phos- skeletal muscles of the mdx mouse, muscles of the γ-SG phorylation events, beginning with phosphorylation of -/- knock-out (γ-SG ) mouse display no mechanical fragility, multiple sites in the C-terminal autoinhibitory domain, at least until 4 months of age, as shown by a minimal loss followed by mammalian target of rapamycin (mTOR)- of force-generating capacity following a series of eccentric dependent phosphorylation of sites in the linker region, -/- contractions (ECCs) [7,8]. In spite of this, the γ-SG which allows for full activation of the kinase via phos- mouse exhibits a severe dystrophic phenotype on histo- phorylation of threonine 229 (T229) in the catalytic do- logical examination, with extensive myofiber degeneration main by phosphoinositide-dependent kinase 1 (PDK1) and regeneration, fibrosis, and disruption of sarcolemmal (reviewed in [37]). Although phosphorylation of T229 is integrity, similar to the mdx mouse [9]. The lack of required for p70S6K activation, phosphorylation of T389 mechanical fragility suggests that aberrant signaling in the linker region has been found to correlate most may contribute to the muscle degeneration seen in the closely with in vivo activity [38], and can be used as a -/- γ-SG mouse. Indeed, our previous studies demon- measure of kinase activation. p70S6K has a multitude of strated that localization of the SG complex to the sarco- downstream targets, with roles in protein synthesis, lemma and phosphorylation of the tyrosine 6 residue of growth, proliferation, survival, and more [35], including γ-SG are essential for normal signaling by extracellular S6 ribosomal protein (S6RP), which closely correlates signal-regulated kinases 1 and 2 (ERK1/2), in response with protein translation rates [39]. to ECCs [10,11]. Based upon these data, we have In the current study, we examined ERK1/2, Akt, focal asserted that the SG complex acts as a mechanosensor adhesion kinase (FAK), and p70S6K responses to passive -/- in skeletal muscle because of its position within the stretch in C57 and γ-SG skeletal muscle to further eluci- DGC, the modifications that occur to γ-SG with mech- date the importance of the SG complex for mechanotrans- anical perturbation, and the necessity of the complex duction. While differences in ERK1/2 phosphorylation -/- for normal signaling. between C57 and γ-SG muscle were calcium-dependent, A complication of using ECCs to invoke signal transduc- differences in p70S6K activation were independent of cal- tion is that there are multiple processes in play. Not only is cium. In addition, the p70S6K response to stretch in both there externally applied tension on the sarcolemma and primary myofiber cultures and isolated extensor digitorum DGC as a result of lengthening, but also active contraction longus (EDL) muscles was differentially regulated in the 2+ resulting in internally applied tension and Ca flux within absence of γ-SG. Specifically, experiments in isolated the fibers and across the sarcolemma, all of which could muscles suggest that γ-SG is required for inactivation of contribute to alterations in signaling in the absence of p70S6K. The findings increase our understanding of the γ-SG. Calcium is known to be involved in many mechano- contribution of aberrant load-sensing to the pathology of sensitive signaling pathways (reviewed in [12,13]) and aber- muscular dystrophies where the SG complex is deficient. rant calcium regulation is a feature of SG-deficient muscle [14-22]. Indeed, an exaggerated ERK1/2 response to mech- Methods anical stimulation that is dependent on extracellular cal- Animals -/- cium has been demonstrated in the mdx mouse diaphragm Adult C57BL/6 (C57) and γ-SG-null (γ-SG )mice were [23]. Studies suggest that aberrant calcium regulation in used. For ex vivo protocols, mice were aged 8 to 16 weeks. Moorwood et al. Skeletal Muscle 2014, 4:13 Page 3 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 -/- The γ-SG mouse lacks γ-SG due to gene targeting, below. Unless otherwise indicated, all cell culture reagents resulting in an additional loss of β-and δ-SG and a de- were purchased from Gibco. crease of α-SG [9]. All experiments were approved by the University of Pennsylvania Institutional Animal Care and Myotube stretching protocol Use Committee. C2C12 myotubes were stretched using an apparatus that produces isotropic two-dimensional strain of cells in vitro, as described previously [40]. Briefly, myotubes were sub- C2C12 myotube culture jected to 10% strain, 40 times per minute, for 30 min, at Flexible silicone membranes (Specialty Manufacturing, 37°C in a humidified atmosphere of 5% CO in air. Control Inc.) were stretched across the bottom of custom cylinders myotubes were treated identically but were not stretched. which acted as a culture chamber. The membranes were Lysates were harvested immediately as described below. held in place using an O-ring as described previously [40]. Primary myotubes were stretched using the protocol Membranes were then coated with a thin layer of 2 mg/mL described for C2C12 myotubes. Stretched myotubes were GFR Matrigel (BD #354230). C2C12 myoblasts (3.5 × 10 / then incubated without stretch for a further 1, 2, or 4 h at cylinder) were seeded onto the membranes and maintained 37°C before harvesting lysates for immunoblotting as at 5% CO at 37°C in growth media (10% FBS, 100 U described below. Control (non-stretched) myotubes were penicillin, 100 μg streptomycin, 100 μg/mL gentamycin harvested immediately after the end of the stretching in DMEM) for approximately 24 h until 70% to 80% protocol used for the stretched myotubes. confluent, then switched to differentiation media (2% HS, 100 U penicillin, 100 μg streptomycin, 100 μg/mL Isolated muscle stretching protocol Gentamycin in DMEM). Myoblasts were allowed to Mice were anaesthetized using ketamine and xylazine. EDL differentiate into multinucleated myotubes for 5 days, muscles were dissected and placed in an organ bath during which media was changed every other day before containing oxygenated high-glucose (25 mM) DMEM with stretching as described below. HEPES (25 mM) (Life Technologies), at room temperature. For rapamycin sensitivity experiments, high-glucose DMEM Primary myoblast culture was supplemented with 150 nM rapamycin (Sigma) or ve- Mice were euthanized using CO inhalation. Flexor digi- hicle only (0.1% DMSO). Muscles were adjusted to 9.3 mN torum brevis (FDB) muscles were dissected and incu- of resting tension, approximately equivalent to optimal bated with 2 mg/mL collagenase I (Sigma), 10% FBS in length, based on our previous experiments [42]. After a Tyrode’s solution (125 mM NaCl, 5 mM KCl, 1 mM 10-min equilibration period, the length of the muscle was CaCl , 1 mM MgCl , 1 mM KH PO , 20 mM HEPES measured using calipers, and the muscle was subjected to 2 2 2 4 (all from Fisher), 5.5 mM glucose (Sigma), pH 7.4) for a stretching protocol of 15% strain (held for 100 ms with 90 min at 37°C, with shaking, as previously described ramp times of 50 ms), 20 times per minute, for either 30 [41]. Muscles were washed in 10% FBS in Tyrode’s solu- or 90 min, using an in vitro muscle test system (Aurora tion and clumps of fibers were liberated by pipetting up Scientific). Muscles were snap-frozen immediately follow- and down in 10% FBS, 100 U penicillin, 100 μg strepto- ing the end of the stretch protocol. Contralateral muscles mycin in Tyrode’s solution using a wide-mouthed glass were used as controls and were adjusted to the same length pipette. Clumps of fibers were transferred to a second as the stretched muscles, then incubated in oxygenated dish with the same solution and pipetted up and down high-glucose DMEM with HEPES at room temperature for again. Single fibers were transferred to a third dish be- the equivalent length of time, before snap-freezing. fore plating on silicone membranes coated with Matrigel (Becton Dickinson; 2 mg/mL diluted in DMEM). Growth Calcium dependence experiments media (20% FBS, 10 ng/mL mouse basic fibroblast growth EDL muscles were dissected as described above and factor (MP Bio), 100 U penicillin, 100 μgstreptomycin, placed in an organ bath contained oxygenated Ringer’s 1 μg/mL Fungizone, 100 μg/mL Gentamycin in Ham’s solution (119 mM NaCl, 4.74 mM KCl, 2.54 mM CaCl , F-10 media) was carefully added and cultures were incu- 1.18 mM KH PO , 1.18 mM MgSO , 25 mM HEPES, 2 4 4 bated without disturbance for 3 days at 37°C. Media was 2.75 mM glucose), calcium-free Ringer’s solution (CaCl then changed every 2 days. After 7 to 10 days, when myo- replaced with 2.5 mM MgCl ) or Ringer’s solution sup- blast cultures were 70% to 80% confluent, media was plemented with 100 μM tetracaine. They were incubated switched to differentiation media (10% HS, 0.5% chicken for 30 min before snap-freezing in liquid nitrogen. embryo extract, 100 U penicillin, 100 μg streptomycin, 1 μg/mL Fungizone, 100 μg/mL Gentamycin in DMEM) Immunoprecipitation and myoblasts were allowed to differentiate into multinu- Immunoprecipitation experiments were carried out using cleated myotubes for 5 days before stretching as described the Pierce Classic IP kit (Thermo Scientific). Muscle lysates Moorwood et al. Skeletal Muscle 2014, 4:13 Page 4 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 containing 100 μg total protein was immunoprecipitated Statistics with anti-P-tyrosine (Cell Signaling #9411 1:100) overnight Comparisons between non-stretched and stretched C2C12 at 4°C with end-over-end mixing. Samples were purified cells (Figure1)weredonebyunpairedTtest.Comparisons -/- using Protein A/G Plus Agarose beads (Roche) for 1 h between C57 and γ-SG primary cultures across time at 4°C with end-over-end mixing. The immune complex (Figure2)weredonebytwo-way ANOVA with Tukey’s was eluted with non-reducing sample buffer and boiled multiple comparisons test. Comparisons between C57 and -/- at 100°C for 5 min before being applied to a SDS-PAGE γ-SG muscles in calcium experiments (Figure 3) were gel, transferred, and immunoblotted as described below. done by unpaired T test. Comparisons between C57 and -/- γ-SG muscles with and without stretch for each time Immunoblotting point (Figure 4) or with and without rapamycin (Figure 5) C2C12 and primary myotubes were washed with ice-cold were done by two-way ANOVA with Tukey’smultiple PBS before lysing in 100 to 200 μLofRIPAbuffer(50 mM comparisons test. HEPES pH 7.5, 150 mM NaCl, 5 mM EDTA, 1 mM EGTA, 15 mM p-nitrophenyl phosphate disodium hexahydral, 1% Results NP-40, 0.1% SDS, 1% deoxycholate, 0.025% sodium azide) p70S6K, but not ERK1/2 or Akt responds to passive with protease and phosphatase inhibitor cocktails (Sigma). stretch in vitro Lysates were incubated on ice for 30 min, centrifuged at Studies using whole muscles from animal models of the 16,000 rcf for 20 min at 4°C and the supernatants retained. dystrophies are made more complex by the presence of EDL muscles were ground using a dry ice-cooled pestle multiple cell types, as well as pathological processes such and mortar, and lysed in 200 μL of RIPA buffer with prote- as fibrosis. Therefore, we initially investigated mechano- ase and phosphatase inhibitors. Lysates were incubated on transduction signaling in passively stretched C2C12 ice for 1 h, vortexing half-way through, centrifuged at myotubes. We found that passive stretching in vitro did 16,000 rcf for 20 min at 4°C and the supernatants retained. not alter phosphorylation of ERK1 or Akt and that ERK2 Protein content was determined using a Bradford method and FAK phosphorylation decreased following stretch protein assay kit (Bio-Rad). Lysates (30 μg total protein for (Figure 1A,C-F,H-I). However, we found an increase in myotube cultures, 90 μg total protein for EDL muscles) p70S6K phosphorylation at T389, which reflects activity, were separated by SDS-PAGE on Tris-HCl polyacrylamide in response to passive stretching of C2C12 myotubes gels (Bio-Rad) and transferred to PVDF membranes. (Figure 1B,G). Therefore, this in vitro model reflected Membranes were blocked in 5% milk in TTBS (10 mM some, but not all, stretch responses observed in muscle Tris, 150 mM NaCl, 0.1% v/v Tween-20, pH 8), with 2% in vivo [43,44], and highlighted p70S6K as a pathway of BSA added for some antibodies, then probed with interest. The lack of Akt phosphorylation suggests that antibodies to the following: phospho (P)-p70S6K (T389) p70S6K phosphorylation occurred through an Akt- (Cell Signaling #9205 1:200 for primary cultures, Cell independent pathway, while the lack of FAK phosphoryl- Signaling #9234 1:250 for isolated muscles), P-p70S6K ation supports an integrin-independent mechanism. (T421/S424) (Cell Signaling #9204 1:1,000), P-S6RP (Cell Signaling #2211 1:2,000), P-ERK1/2 (Cell Signaling #9101 Differential p70S6K stretch response occurs in C57 and -/- 1:500), total (T)-ERK1/2 (Cell Signaling #9107 1:1,000), γ-SG primary cultures P-Akt (Cell Signaling #9271 1:300), T-Akt (Cell Sig- Having established pathways of interest in vitro using the naling #2920 1:2,000), P-FAK (Millipore 05-1140, 1:500), C2C12 cell line, we used primary myoblast cultures from -/- T-FAK (Millipore 06-543, 1:1,000), γ-SG (Novocastra C57 and γ-SG mice (Figure 2A) to investigate the VP-G803 1:300), glyceraldehyde 3-phosphate dehydrogenase changes in mechanotransduction signaling associated (GAPDH) (Santa Cruz sc-32233 1:5,000), and tubulin solely with the loss of the SG complex in myofibers. C57 -/- (Sigma T5168 1:20,000). Band intensities were quantified and γ-SG cultures were stretched for 30 min as de- using ImageQuant TL, 1D gel analysis (C2C12 and myo- scribed above and lysates were harvested 1, 2, or 4 h after tube cultures, rapamycin sensitivity experiments) or ImageJ stretching ended, to allow observation of the signaling (NIH) (all other isolated muscles). P-p70S6K and P-S6RP time course. Immunoblotting analysis showed that there were normalized to either GAPDH or tubulin, P-FAK was was no difference in basal ERK1/2 phosphorylation -/- normalized to either T-FAK or GAPDH, P-Akt was normal- between C57 and γ-SG myotubes and little change in ized to T-Akt, and P-ERK1/2 was normalized to T-ERK1/2. ERK1/2 phosphorylation in response to stretch (Figure 2B, D,E). Therefore, similar to the C2C12 cells, primary Microscopy cultures did not reflect the ERK1/2 phosphorylation -/- Images of primary cultures were taken using a Leica DM responses found previously in C57 and γ-SG mice RBE microscope and Leica DFC300 CCD camera, using in vivo [10,11]. Neither P-Akt nor P-FAK displayed -/- OpenLab software (Perkin Elmer). significant differences between C57 and γ-SG cultures Moorwood et al. Skeletal Muscle 2014, 4:13 Page 5 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 Figure 1 p70S6K responds to stretch in C2C12 cells. C2C12 myotubes were cultured on silicone membranes and subjected to passive stretching for 30 min. (A-D) Representative immunoblots for P-ERK1/2, total T-ERK1/2, P-p70S6K (T389 site), tubulin, P-Akt, T-Akt, P-FAK, and T-FAK in non-stretched (NS) and stretched (S) C2C12 cells. (E-I) Quantification of activation levels. P-ERK1 and 2 were normalized to T-ERK1 and 2, respectively, P-p70S6K was normalized to tubulin, P-Akt was normalized to T-Akt, and P-FAK was normalized to T-FAK. n = 5-6 wells of C2C12 cells per group. Bars represent mean ± standard error. * Significantly different from non-stretched myotubes by unpaired T test. NS, non-stretched; S, stretched. (Figure 2B,F,G). Further, these proteins did not show basal phosphorylation state of the upstream pathways any prolonged response to passive stretch in the (Figure 3D; Normal). primary cultures, consistent with the lack of an acute Heightened mechanosensitive signaling could arise through positive stretch response in C2C12 cells. Examination of increased flux of ions across the sarcolemma, particu- 2+ p70S6K revealed a trend towards elevated T389 phos- larly Ca [17]. This could occur either through en- -/- phorylation at baseline in γ-SG myotubes, compared hanced activity of channels, such as the TRP family of -/- to C57 myotubes. Activation of p70S6K in γ-SG myo- cation channels (reviewed in [28,29]), or via membrane tubes upon stretch was not apparent (Figure 2B,C), but ruptures. To determine the calcium dependence of the ob- C57 myotubes displayed a trend for increased activation served differences in p70S6K and ERK1/2 phosphorylation -/- at 2 and 4 h after stretching (Figure 2B,C). Therefore, in γ-SG muscles, we incubated muscles in calcium-free while the primary cultures reflect some of the responses Ringers solution, retaining the same ionic strength. 2+ found in C2C12 myotubes, the experimental system is Absence of extracellular Ca did not alter the relative -/- too variable to draw firm conclusions regarding mech- difference in P-p70S6K between γ-SG and C57 muscles anical signaling pathways associated with γ-SG. (Figure 3C). However, the increased P-ERK1/2 and -/- P-S6RP found in γ-SG muscles in normal Ringer’ssolu- -/- Elevated P-p70S6K in γ-SG muscles is calcium tion was abrogated when there was no calcium in the independent bathing solution (Figure 3A,B,D). Thus, only p70S6K Our experiments in myotubes, together with our previ- phosphorylation appeared to be calcium independent. Be- ous studies in isolated muscles [10,11], suggested that cause intracellular calcium stores can also alter the intra- ERK1/2 phosphorylation changes require active contrac- cellular calcium concentration, particularly during muscle tion in addition to stretch, whereas p70S6K responds to activation, tetracaine was used to inhibit sarcoplasmic 2+ stretch alone. To test this hypothesis in vivo, we ex- reticulum release of Ca through the ryanodine receptors. tended our analysis to examine the response of p70S6K Verification of this inhibition was established in a separate to passive stretch of isolated muscles. P-p70S6k was experiment, through measuring tetanic force generation -/- elevated approximately 1.7-fold in resting γ-SG EDL by EDL muscles before and after addition of tetracaine. muscles incubated in normal oxygenated Ringer’s solu- After 15 min incubation with tetracaine, force production tion (Figure 3C; Normal). As in our previous studies, was virtually eliminated (force was 297 mN prior to there was a 3- and 1.5-fold increase of P-ERK1 and addition of tetracaine, 2.4 mN after 15 min incubation -/- P-ERK2, respectively, in resting γ-SG EDL muscles with tetracaine and not detectable after 20 min incubation (Figure 3A-B; Normal). Because both of these pathways with tetracaine). Again, relative P-p70S6K levels between converge to phosphorylate S6RP, we compared the the two muscle groups were not altered by tetracaine 2+ phosphorylation state of this protein in muscles from (Figure 3C). In contrast, blockade of SR Ca release -/- -/- both genotypes. γ-SG EDL muscles exhibited an 8-fold reduced P-ERK1 levels in γ-SG muscles relative to C57 increase in P-S6RP, which was consistent with the higher muscles, even though there was no alteration in P-ERK2 Moorwood et al. Skeletal Muscle 2014, 4:13 Page 6 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 -/- -/- Figure 2 Differential p70S6K stretch response in C57 and γ-SG primary cultures. Primary myotubes from C57 and γ-SG FDB fibers were cultured on silicone membranes and subjected to passive stretching for 30 min. Lysates were harvested 1, 2, or 4 h after stretch. (A) Representative images of i, satellite cells migrating from an FDB fiber and ii, differentiated myotubes. (B) Representative immunoblot for P-p70S6K (T389 site), P- and T-ERK1/2, P-FAK, P-Akt, T-Akt, and GAPDH in non-stretched (C) and stretched (1 h, 2 h, 4 h) primary cultures. (C-G) Quantification of activation levels. Legend in C applies to all graphs. P-ERK1 and 2 were normalized to T-ERK1 and 2, respectively, P-Akt was normalized to T-Akt and P-p70S6K and P-FAK were normalized to GAPDH. n = 3 (p70S6K) or 4 (all other proteins) independent sets of primary cultures per genotype. Bars represent mean ± standard error. Statistical significance was tested by two-way ANOVA. (Figure 3A,B). Phosphorylation of S6RP remained elevated γ-SG-dependent mechanotransduction in vivo. A passive -/- in γ-SG muscles in the presence of tetracaine, but to a stretching protocol comprised of a 15% strain, 20 times lesser extent than in normal Ringer’s solution (Figure 3D). per min, for 30 min in high-glucose DMEM was sufficient Taken together, both extracellular and intracellular cal- to cause increased γ-SG phosphorylation in the EDL, as is cium contribute to the heightened P-ERK1 levels in the case for eccentric contraction of the EDL (Figure 4A; -/- -/- γ-SG muscles, whereas the increase in basal P-p70S6K [10]). We stretched C57 and γ-SG EDL muscles for -/- in γ-SG muscles is not dependent on either extracellular either 30 or 90 min and immediately snap-froze them in or intracellular calcium. liquid nitrogen. We then used immunoblotting to meas- ure phosphorylation of p70S6K at T389, as above, and also Differential p70S6K stretch response occurs in isolated at T421/S424 in the auto-inhibitory domain. Basal phos- -/- C57 and γ-SG muscles phorylation of p70S6K at T389 showed a trend to be in- -/- Having established that P-p70S6K changes at rest did not creased in unstretched γ-SG muscles compared to C57 depend on calcium, we pursued the role of p70S6K in muscles (Figure 4B,C). However, basal phosphorylation at Moorwood et al. Skeletal Muscle 2014, 4:13 Page 7 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 -/- -/- Figure 3 Elevated p70S6K in γ-SG muscles is independent of calcium. EDL muscles from C57 and γ-SG mice were maintained in normal oxygenated Ringer’s solution, calcium-free oxygenated Ringer’s solution or oxygenated Ringer’s solution supplemented with tetracaine for 30 min. (A-D) Representative immunoblots and quantification for P-ERK1 (A),P-ERK2 (B), P-p70S6K (T389 site; C), and P-S6RP (D).Legend in A applies to all graphs. P-ERK 1 and 2 were normalized to T-ERK 1 and 2, respectively; P-p70S6K and P-S6RP were normalized to tubulin. Independent immunoblots -/- were performed for each condition and γ-SG activation levels were normalized to C57 activation levels in each case. n = 3 muscles per genotype and condition. Bars represent mean ± standard error. All datasets were tested by unpaired T test. * Significantly different from C57 by unpaired T test. -/- T421/S424 was not different between C57 and γ-SG activation of p70S6K in response to passive stretch, imply- muscles (Figure 4D). After 30 min of stretch, phosphoryl- ing that γ-SG plays a role in p70S6K inactivation. ation of p70S6K at T389 and T421/S424 was increased to -/- a similar degree in stretched C57 and γ-SG EDL mus- Stretch response of p70S6K T389, but not S6RP, is -/- cles, compared to non-stretched controls. However, after rapamycin-sensitive in γ-SG muscles 90 min of stretch, phosphorylation at T389 and T421/ Because mTOR is a key mediator of p70S6K activation, S424 had decreased in stretched C57 muscles and was we examined the effect of the mTOR inhibitor rapamycin -/- close to non-stretched levels. In contrast, phosphorylation on stretch responses in isolated C57 and γ-SG muscles. at T389 was further increased, and T421/S424 phosphor- EDL muscles were subjected to cyclic stretch for 90 min, -/- ylation remained elevated, in stretched γ-SG muscles, as described above. Unlike in C2C12 cells and primary compared to non-stretched controls (Figure 4B-D). For cultures, P-Akt showed a trend to increase on stretching, -/- T389 phosphorylation after 90 min of stretch, genotype, which was statistically significant in γ-SG muscles. stretch, and the interaction between them were all statisti- As anticipated, P-Akt was unaffected by rapamycin cally significant, by two-way ANOVA. T421/S424 showed (Figure 5A,B). Rapamycin treatment completely blocked the a similar trend to T389; however, while the effect of increase in p70S6K T389 phosphorylation after passive stretch was statistically significant, the difference between stretch of C57 muscles, consistent with previous studies -/- genotypes was not. Thus, in contrast to primary cultures, [36,37]. In γ-SG muscles, rapamycin abrogated most -/- γ-SG muscles exhibited heightened and prolonged p70S6K T389 phosphorylation, but residual phosphorylation Moorwood et al. Skeletal Muscle 2014, 4:13 Page 8 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 remained in stretched muscles (Figure 5A,C). T421/S424 showed a trend to increase in response to stretch in both -/- C57 and γ-SG muscles. Surprisingly, rapamycin blunted the p70S6K T421/S424 stretch response in C57 muscles, which is inconsistent with previous studies [36]. However, -/- the T421/S424 response to stretch persisted in γ-SG muscles in the presence of rapamycin (Figure 5A,D). S6RP phosphorylation increased in response to stretch -/- in both C57 and γ-SG muscles. Interestingly, while rapamycin blocked stretch-induced phosphorylation of S6RP in C57 muscles, phosphorylation in response to -/- stretch was preserved in γ-SG muscles (Figure 5A,E). Taken together, these results suggest either that the level -/- of active p70S6K remaining in γ-SG musclesissuffi- cient to phosphorylate S6RP regardless of rapamycin or that an alternate pathway bypasses p70S6K to phos- phorylate S6RP in muscles lacking γ-SG. Discussion Skeletal muscle has a remarkable ability to adapt to changes in workload. Almost all muscle properties can be modulated, such as muscle fiber size, contractile properties and metabolism. Changes in patterns of gene expression as well as shifts in the balance between pro- tein synthesis and degradation are required to complete the adaptational response. Identification of major path- ways that directly regulate gene expression and protein synthesis/degradation demonstrate that multiple inputs (mechanical, chemical, and metabolic) can converge on final common pathways for muscle growth and adaptation (reviewed in [45]). However, sorting out the contribution of the wide variety of inputs on muscle adaptation has been more difficult. In our previous work, we used an eccentric contraction protocol to in- vestigate the dependence of ERK1/2 mechano-sensing on phosphorylation of γ-SG. However, this protocol alters multiple factors, including externally applied tension, internally generated tension and changes in extracellular and intracellular calcium fluxes, all of Figure 4 Differential p70S6K stretch response in isolated C57 -/- -/- and γ-SG muscles. EDL muscles from C57 and γ-SG mice were which potentially have effects on mechanosensitive maintained in oxygenated high glucose DMEM and subjected to signaling pathways. In the present study, we used a passive stretching for 30 or 90 min. (A) Immunoblot of γ-SG following passive stretching protocol to isolate the effects of ex- immunoprecipitation with anti-P-Tyr or lysate only, showing γ-SG ternally applied tension in the absence of active contrac- phosphorylation in response to 30 min of stretch. (B) Representative tion, in order to examine the downstream signaling in immunoblots of P-p70S6K (T389 and T421/S424 sites) and tubulin. (C, D) Quantification of P-p70S6K T389 (C) and T421/S424 more detail. (D), normalized to tubulin. Samples for each time point were run Passive stretching protocols can be performed in both on separate immunoblots and normalized to C57 NS. n = 3-5 cell cultures and whole muscle preparations, and the muscles per genotype, condition, and time point. Bars represent reductionist approach of utilizing cultures can eliminate mean ± standard error. All datasets were tested by two-way ANOVA. some of the physiological complexities associated with Stretch was statistically significant for T421/S424 after 30 min of stretch; genotype, stretch, and the interaction between them were intact or isolated muscles. As such, our initial experi- statistically significant T389 after 90 min of stretch, by two-way ments using C2C12 cells were key to identifying p70S6K ANOVA. *Significantly different from C57 S at 90 min by two-way as being activated in response to stretch, in contrast to ANOVA with Tukey’s multiple comparisons test. NS, non-stretched; the lack of response by ERK1/2, Akt, or FAK. Primary S, stretched. -/- myotubes generated from C57 or γ-SG mice had the Moorwood et al. Skeletal Muscle 2014, 4:13 Page 9 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 Figure 5 Stretch response of p70S6K T389, but not S6RP, is -/- rapamycin-sensitive in γ-SG muscles. EDL muscles from C57 -/- and γ-SG mice were maintained in oxygenated high glucose DMEM supplemented with or without rapamycin and subjected to passive stretching for 90 min. (A) Representative immunoblots of P-Akt, T-Akt, P-p70S6K (T389 and T421/S424 sites), P-S6RP, and tubulin. Left panels DMEM alone; right panels DMEM + rapamycin. (B-E) Quantification of P-Akt (B), P-p70S6K T389 (C), P-p70S6K T421/ S424 (D), and P-S6RP (E). Legend in B applies to all graphs. P-Akt was normalized to T-Akt; all other proteins were normalized to tubulin. n = 2-3 muscles per genotype and condition. Bars represent mean ± standard error. All datasets were tested by two-way ANOVA. -/- For P-Akt in γ-SG muscles, stretch was significant. For P-p70S6K -/- T389 in γ-SG muscles, stretch, rapamycin treatment, and the -/- interaction between them were all significant. For P-S6RP in γ-SG -/- muscles, stretch was significant. ‡Significantly different to NS γ-SG -/- control and †significantly different to S γ-SG without rapamycin by two-way ANOVA with Tukey’s multiple comparisons test. NS, non-stretched; S, stretched. distinct advantage of efficient germline elimination of γ-SG combined with an in vitro culture system. Even though these experiments displayed trends towards differ- ential activation of p70S6K after stretch, the inherent variability of the preparation impaired identification of signaling patterns that were dependent upon either stretch or γ-SG. Thus, we returned to isolated muscles from C57 -/- and γ-SG mice to investigate γ-SG-dependent mechano- transduction pathways. Using this model, we observed a -/- modest increase of P-p70S6K in γ-SG muscles at rest that was independent of intra- or extracellular calcium, and a prolonged activation of p70S6K following stretch. These results support a role for γ-SG in particular, or the SG complex in general, in mechanical signal transduction, where the loss of this protein leads to an increase in activation, and deficit in deactivation, or a combination of both. Given the dependence of our findings on the experi- mental platform utilized, future directions will include verification of the results in an even more intact system, such as in situ muscle preparations or whole animals. An intriguing explanation for our in vivo results is that γ-SG is required for dephosphorylation and deactivation of p70S6K. There is considerable evidence that p70S6K is directly dephosphorylated by protein phosphatase 2A (PP2A), independently of mTOR [46-49]. The phosphat- ase PHLPP has also been shown to target p70S6K [50]. γ-SG may mediate the activation of these phosphatases in response to sustained mechanical stimulation. Alter- natively, γ-SG may regulate pathways that deactivate p70S6K indirectly. For example, the phosphatase SHP-2 can cause mTOR-dependent dephosphorylation of p70S6K [51,52]. Further studies will be required to define the inacti- vation pathway disrupted by γ-SG loss. Passive stretch eliminates the contribution of active con- traction or SR calcium fluxes, but does not eliminate the Moorwood et al. Skeletal Muscle 2014, 4:13 Page 10 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 2+ effects of extracellular Ca fluxes through mechanically phosphorylation of the p70S6K T421/S424 autoinhibitory sensitive channels. Further, as previously shown, passive domain sites by ERK1/2. 2+ stretch causes greater Ca influx into myotubes lacking Our experiments with rapamycin showed that, for both -/- members of the SG complex [17], raising the possibility C57 and γ-SG muscles, phosphorylation of p70S6K at that the mechanical signal transduction pathways we eval- T389 is mTOR-dependent, consistent with previous stud- uated previously may be modulated not only by the SG ies [36,37]. The T421/S424 autoinhibitory domain sites complex, but also by additional channels in the sarco- were phosphorylated in response to stretch in both C57 -/- lemma. To address this, we examined the contribution and γ-SG muscles, which was different to our initial of calcium to the elevated p70S6K and ERK1/2 activity experiment in stretched isolated muscles, where the C57 -/- found in γ-SG muscles. We found that the elevation of response had diminished by 90 min of stretching. How- P-ERK1/2 in the absence of γ-SG was dependent on both ever, in the presence of rapamycin, this response was internal and external sources of calcium. In contrast, the not present. This is surprising given that the T421/S424 -/- difference in basal P-p70S6K between C57 and γ-SG sites are not thought to be targeted by mTOR. Previous muscles was not calcium dependent. This suggests that studies have shown these sites to be rapamycin-insensitive while ERK1/2 activation may lie downstream of the cal- [36], but recent evidence suggests a modest mTOR de- -/- cium misregulation that occurs in SG-deficient muscle, pendence [61]. In γ-SG muscles, rapamycin had no ef- changes in p70S6K activation may be a more direct conse- fect on T421/S424 phosphorylation. Further experiments quence of the absence of the SG complex. This is of great are needed to fully understand the role of mTOR on phos- interest given that γ-SG has been shown to be important phorylation of the p70S6K autoinhibitory domain in C57 -/- for mechanotransduction, but the downstream signaling and γ-SG skeletal muscle. Interestingly, the stretch- pathways are uncharacterized [10,11]. Furthermore, p70S6K induced phosphorylation of S6RP was rapamycin sensitive -/- has been implicated in mechanotransduction in skeletal in C57 muscles, but not in γ-SG muscles. This suggests muscle, but the upstream initiation signals are not that an alternative pathway can bring about S6RP phos- -/- known [36,53-55]. However, it should be noted that phorylation in γ-SG muscles when p70S6K is not acti- SG-deficient muscle undergoes substantial degeneration vated. One possibility is that S6RP is phosphorylated by and subsequent regeneration, which may also explain p90 ribosomal S6 kinase, which is activated by ERK; this is -/- the elevated basal p70S6K, which is transiently increased consistent with the increase in basal ERK1/2 in γ-SG during regeneration [56]. muscles, and the over-response of ERK2 on mechanical Our study found that the pattern of differential p70S6K stimulation by eccentric contraction [10]. -/- phosphorylation in response to stretch in γ-SG muscles We did not observe a strong Akt response to passive -/- was similar both for phosphorylation of T389, which cor- stretch, or any difference between C57 and γSG , im- relates with kinase activity, and for T421/S424, two of the plying that mTOR and/or p70S6K were being activated four phosphorylation sites in the autoinhibitory domain. through Akt-independent pathways. This is consistent Phosphorylation of T389 is mTOR-dependent, while with previous studies showing that Akt does not respond phosphorylation of the autoinhibitory domain is carried to mechanical stimulation in skeletal muscle, and that out by proline-directed kinases. Furthermore, it is thought p70S6K phosphorylation in response to stretch is inde- that phosphorylation of the autoinhibitory domain is ne- pendent of PI3K [62]. We also did not see increased cessary for phosphorylation of T389 [37]. The correlation phosphorylation of FAK in response to passive stretch in between phosphorylation of T421/S424 and T389 in our C2C12 cells or primary myotubes. Although integrins can isolated muscle model therefore suggests that phosphoryl- participate in mechanotransduction, it appears that our ation of the autoinhibitory domain was the rate-limiting cyclic passive stretch protocols did not cause integrin step for p70S6K activation, an intriguing prospect given activation. Further studies will be needed to elucidate that the autoinhibitory domain may be targeted by ERK1/ the details of crosstalk between SG-dependent and 2 [57]. Therefore, a future hypothesis to test is that differ- integrin-dependent signaling pathways, as well as the ential p70S6K activation is a downstream consequence role of calcium in these signaling cascades. -/- of differential ERK1/2 activation in γ-SG muscle. This Based on our findings, we position γSG as a mechano- 2+ would implicate Ca as an indirect modulator of p70S6K sensor, schematized in Figure 6, that is important for tran- -/- activity, since the increase in P-ERK1/2 in γ-SG muscle sient ERK1/2 activation during active contractions, as 2+ is dependent upon heightened Ca flux. It is worth noting well as modulation of p70S6K activation during passive that recent work by others has shown that stretch- stretch. Because passive stretch does not appear to in- induced activation of mTOR and p70S6K at T389 is inde- crease P-FAK or P-Akt, γSG is likely to regulate p70S6K pendent of ERK1/2 [54], which can regulate mTOR via through other pathways. These may include regulation tuberous sclerosis proteins 1 and 2 and Raptor [58-60]. of ERK1/2, which can promote p70S6K activation indir- However, this pathway is separate from the putative direct ectly via mTOR or directly by phosphorylation of the Moorwood et al. Skeletal Muscle 2014, 4:13 Page 11 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 [63,64] and overexpression of integrin α7 can improve the dystrophic phenotype through increased survival signaling via p70S6K [65], suggesting that p70S6K inhibition would not be advantageous. However, treatment of mdx mice with the mTOR inhibitor rapamycin improves the dys- trophic phenotype [66]. It is also interesting to note that p70S6K is inhibited by glucocorticoids, which are used in the treatment of DMD and LGMD [67]. Our results begin to provide mechanistic insight into how mechanical signaling is disrupted and altered in the absence of γ-SG. In addition to increasing our under- standing of the normal function of the SG complex, there is potential to provide more refined targets that could be beneficial to patients either in isolation or in combination with other therapeutic approaches. Conclusions We have identified p70S6K as part of a novel SG- Figure 6 Relevant signaling pathways and relationship to γ-SG. dependent mechanosensitive signaling pathway in skeletal Schematic of signaling pathways measured or discussed in this muscle. Our results suggest that γ-SG is required for the manuscript. Dotted lines indicate possible relationships to γ-SG. inactivation of p70S6K following its activation in response Arrowheads indicate an activating relationship, while blunt ends to mechanical stimulation. These studies provide new indicate a repressing relationship. Dashed line indicates priming, insights into the normal function of the SG complex, and rather than full activation. the mechanisms by which its deficiency in some forms of muscular dystrophy may contribute to pathology. autoinhibitory domain, and/or phosphatases such as Competing interests PP2A that dephosphorylate p70S6K. Loss of γSG un- The authors declare that they have no competing interests. couples the response to stretch, which may contribute to muscle pathology. Authors’ contributions CM carried out the calcium dependence experiments and the stretching The stability of the SG complex is directly affected in experiments in isolated muscles, and drafted the manuscript. AP carried out several LGMDs and in DMD, and a significant part of the the passive stretching experiments in primary myotube cultures. JS carried pathology in these diseases appears to be inappropriate out the passive stretching experiments in C2C12 myotubes, the immunoprecipitation of phosphorylated γ-SG, and the immunoblotting for load sensing. The first step in therapeutic development is rapamycin sensitivity experiments. BK participated in the passive stretching identifying and understanding the target, but little is cur- experiments in primary cultures. EJM developed the apparatus used for rently known about the role of the SG complex in load stretching of myotube cultures. ERB conceived of the study, participated in its design, coordination, and interpretation of the results, and helped draft sensing. Therefore, understanding the functions that are the manuscript. All authors read and approved the final manuscript. disrupted and the pathways that are involved in mechano- transduction involving the SG complex will help in the de- Acknowledgements sign of therapies for LGMDs and DMD. While restoration The authors thank Min Liu and Tian Zuozhen of the Wellstone Muscular Dystrophy Physiological Assessment Core for technical assistance with the of a completely normal SG complex either through gene rapamycin experiments. This work was supported by grants from NASA correction or protein replacement would also normalize (NNX09AH44G) and NIH (U54 AR052646) to ERB. CM was supported by the mechanical signal transduction, this may not be possible Paul D Wellstone Muscular Dystrophy Cooperative Research Center Training Grant (U54 AR052646). AP was supported by NNX09AH44G. JS is supported for all mutations responsible for DMD and LGMD. It is by the Pennsylvania Muscle Institute Training Grant (NRSA T32-AR053461). BK known that localization of the SG complex is not the sole was supported by a School of Dental Medicine Summer Research Fellowship. criterion for appropriate signaling [11]. Hence, other EJM was supported by NNX09AH44G and P50 DK52620. The funding bodies had no role in the study design, the collection, analysis, and interpretation of proteins may be necessary to correct signaling even when data, the writing of the manuscript, or the decision to submit the manuscript the complex is restored, and downstream pathways may for publication. emerge as more feasible therapeutic targets. We do not Author details know whether the enhanced basal and stretch-responsive Department of Anatomy and Cell Biology, School of Dental Medicine, -/- activation of p70S6K in γ-SG muscle contributes to University of Pennsylvania, Philadelphia, PA, USA. Pennsylvania Muscle pathology or compensates for it. Likewise, it is not clear Institute, University of Pennsylvania, Philadelphia, PA, USA. Current address: Department of Physiology, Medical School, National and Kapodistrian whether inhibition of p70S6K would have a beneficial or University of Athens, Goudi, Athens, Greece. Current address: Department of a detrimental effect in dystrophic muscle. Muscle specific Dermatology and Cutaneous Biology, Jefferson University College of gene targeting of mTORC1 components induces myopathy Medicine, Philadelphia, PA, USA. Moorwood et al. Skeletal Muscle 2014, 4:13 Page 12 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 Received: 25 March 2014 Accepted: 29 May 2014 24. Iwata Y, Katanosaka Y, Arai Y, Komamura K, Miyatake K, Shigekawa M: Anovel Published: 1 July 2014 mechanism of myocyte degeneration involving the Ca2 + -permeable growth factor–regulated channel. J Cell Biol 2003, 161:957–967. 25. Iwata Y, Katanosaka Y, Arai Y, Shigekawa M, Wakabayashi S: Dominant-negative References inhibition of Ca2+ influx via TRPV2 ameliorates muscular dystrophy in 1. Ervasti JM, Sonnemann KJ: Biology of the striated muscle dystrophin-glycoprotein animal models. Hum Mol Genet 2009, 18:824–834. complex. Int Rev Cytol 2008, 265:191–225. 26. Millay DP, Goonasekera SA, Sargent MA, Maillet M, Aronow BJ, Molkentin JD: 2. Emery AE, Muntoni F: Duchenne Muscular Dystrophy. Oxford: Oxford Calcium influx is sufficient to induce muscular dystrophy through a University Press; 2003. TRPC-dependent mechanism. Proc Natl Acad Sci U S A 2009, 106:19023–19028. 3. Engel AG, Franzini-Armstrong C: Myology: Basic and Clinical. 3rd edition. 27. Goonasekera SA, Davis J, Kwong JQ, Accornero F, Wei-LaPierre L, Sargent New York: McGraw-Hill, Inc; 2004. MA, Dirksen RT, Molkentin JD: Enhanced Ca2+ influx from STIM1–Orai1 4. Durbeej M, Campbell KP: Muscular dystrophies involving the dystrophin- induces muscle pathology in mouse models of muscular dystrophy. glycoprotein complex: an overview of current mouse models. Hum Mol Genet 2014. [Epub ahead of print]. Curr Opin Genet Dev 2002, 12:349–361. 28. Brinkmeier H: TRP channels in skeletal muscle: gene expression, function 5. Blake DJ, Weir A, Newey SE, Davies KE: Function and genetics of dystrophin and implications for disease. Adv Exp Med Biol 2011, 704:749–758. and dystrophin-related proteins in muscle. Physiol Rev 2002, 82:291–329. 29. Gailly P: TRP channels in normal and dystrophic skeletal muscle. 6. Davies KE, Nowak KJ: Molecular mechanisms of muscular dystrophies: Curr Opin Pharmacol 2012, 12:326–334. old and new players. Nat Rev 2006, 7:762–773. 30. Araishi K, Sasaoka T, Imamura M, Noguchi S, Hama H, Wakabayashi E, 7. Hack AA, Cordier L, Shoturma DI, Lam MY, Sweeney HL, McNally EM: Yoshida M, Hori T, Ozawa E: Loss of the sarcoglycan complex and Muscle degeneration without mechanical injury in sarcoglycan sarcospan leads to muscular dystrophy in β-sarcoglycan-deficient mice. deficiency. Proc Natl Acad Sci U S A 1999, 96:10723–10728. Hum Mol Genet 1999, 8:1589–1598. 8. Petrof BJ, Shrager JB, Stedman HH, Kelly AM, Sweeney HL: Dystrophin 31. Duclos F, Straub V, Moore SA, Venzke DP, Hrstka RF, Crosbie RH, Durbeej M, protects the sarcolemma from stresses developed during muscle Lebakken CS, Ettinger AJ, van der Meulen J, Holt KH, Lim LE, Sanes JR, contraction. Proc Natl Acad Sci 1993, 90:3710–3714. Davidson BL, Faulkner JA, Williamson R, Campbell KP: Progressive 9. Hack AA, Ly CT, Jiang F, Clendenin CJ, Sigrist KS, Wollmann RL, McNally EM: muscular dystrophy in α-sarcoglycan-deficient mice. JCellBiol 1998, Gamma-sarcoglycan deficiency leads to muscle membrane defects and 142:1461–1471. apoptosis independent of dystrophin. J Cell Biol 1998, 142:1279–1287. 32. Goonasekera SA, Lam CK, Millay DP, Sargent MA, Hajjar RJ, Kranias EG, 10. Barton ER: Impact of sarcoglycan complex on mechanical signal transduction Molkentin JD: Mitigation of muscular dystrophy in mice by SERCA in murine skeletal muscle. Am J Physiol Cell Physiol 2006, 290:C411–C419. overexpression in skeletal muscle. J Clin Invest 2011, 121:1044–1052. 11. Barton ER: Restoration of gamma-sarcoglycan localization and mechanical 33. Morine KJ, Sleeper MM, Barton ER, Sweeney HL: Overexpression of SERCA1a signal transduction are independent in murine skeletal muscle. J Biol Chem in the mdx diaphragm reduces susceptibility to contraction-induced 2010, 285:17263–17270. damage. Hum Gene Ther 2010, 21:1735–1739. 12. Benavides Damm T, Egli M: Calcium’s role in mechanotransduction during 34. Parsons SA, Millay DP, Sargent MA, Naya FJ, McNally EM, Sweeney HL, muscle development. Cell Physiol Biochem 2014, 33:249–272. Molkentin JD: Genetic disruption of calcineurin improves skeletal muscle 13. Burkholder TJ: Mechanotransduction in skeletal muscle. Front Biosci 2007, pathology and cardiac disease in a mouse model of limb-girdle muscular 12:174–191. dystrophy. J Biol Chem 2007, 282:10068–10078. 14. Andersson D, Meli A, Reiken S, Betzenhauser M, Umanskaya A, Shiomi T, 35. Fenton TR, Gout IT: Functions and regulation of the 70 kDa ribosomal D’Armiento J, Marks A: Leaky ryanodine receptors in beta-sarcoglycan S6 kinases. Int J Biochem Cell Biol 2011, 43:47–59. deficient mice: a potential common defect in muscular dystrophy. 36. Hornberger TA,Stuppard R, Conley KE,FedeleMJ, Fiorotto ML,Chin ER, EsserKA: Skeletal Muscle 2012, 2:9. Mechanical stimuli regulate rapamycin-sensitive signalling by a 15. Iwata Y, Katanosaka Y, Shijun Z, Kobayashi Y, Hanada H, Shigekawa M, phosphoinositide 3-kinase-, protein kinase B- and growth factor-independent Wakabayashi S: Protective effects of Ca2+ handling drugs against mechanism. Biochem J 2004, 380:795–804. abnormal Ca2+ homeostasis and cell damage in myopathic skeletal 37. Dufner A, Thomas G: Ribosomal S6 kinase signaling and the control of muscle cells. Biochem Pharmacol 2005, 70:740–751. translation. Exp Cell Res 1999, 253:100–109. 16. Nakamura TY, Iwata Y, Sampaolesi M, Hanada H, Saito N, Artman M, Coetzee WA, 38. Pearson RB, Dennis PB, Han JW, Williamson NA, Kozma SC, Wettenhall RE, Shigekawa M: Stretch-activated cation channels in skeletal muscle Thomas G: The principal target of rapamycin-induced p70s6k inactivation myotubes from sarcoglycan-deficient hamsters. Am J Physiol Cell Physiol is a novel phosphorylation site within a conserved hydrophobic domain. 2001, 281:C690–C699. EMBO J 1995, 14:5279–5287. 17. Sampaolesi M, Yoshida T, Iwata Y, Hanada H, Shigekawa M: Stretch-induced 39. Nielsen PJ, Manchester KL, Towbin H, Gordon J, Thomas G: The phosphorylation cell damage in sarcoglycan-deficient myotubes. Pflugers Arch 2001, of ribosomal protein S6 in rat tissues following cycloheximide injection, in 442:161–170. diabetes, and after denervation of diaphragm. A simple immunological 18. Solares-Perez A, Alvarez R, Crosbie RH, Vega-Moreno J, Medina-Monares J, determination of the extent of S6 phosphorylationonprotein blots. Estrada FJ, Ortega A, Coral-Vazquez R: Altered calcium pump and secondary JBiolChem 1982, 257:12316–12321. deficiency of gamma-sarcoglycan and microspan in sarcoplasmic reticulum 40. He Y, Macarak EJ, Korostoff JM, Howard PS: Compression and tension: membranes isolated from delta-sarcoglycan knockout mice. Cell Calcium differential effects on matrix accumulation by periodontal ligament 2010, 48:28–36. fibroblasts in vitro. Connect Tissue Res 2004, 45:28–39. 19. Solares-Perez A, Sanchez JA, Zentella-Dehesa A, Garcia MC, Coral-Vazquez 41. Barton ER, Park S, James JK, Makarewich CA, Philippou A, Eletto D, Lei H, RM: Intracellular Ca2+ transients in delta-sarcoglycan knockout mouse Brisson B, Ostrovsky O, Li Z, Argon Y: Deletion of muscle GRP94 impairs skeletal muscle. Biochim Biophys Acta 2010, 1800:373–379. both muscle and body growth by inhibiting local IGF production. 20. Hassoni C: Calcium homeostasis and ultrastructural studies in a patient FASEB J 2012, 26:3691–3702. with limb girdle muscular dystrophy type 2C. Neuropathol Appl Neurobiol 42. Moorwood C, Liu M, Tian Z, Barton ER: Isometric and eccentric force 1999, 25:244–253. generation assessment of skeletal muscles isolated from murine models 21. Fraysse B, Nagi SM, Boher B, Ragot H, Lainé J, Salmon A, Fiszman MY, Toussaint of muscular dystrophies. J Vis Exp 2013, 71:e50036. M, Fromes Y: Ca2+ overload and mitochondrial permeability transition pore 43. Hornberger TA, Mateja RD, Chin ER, Andrews JL, Esser KA: Aging does not activation in living δ-sarcoglycan-deficient cardiomyocytes. Am J Physiol Cell alter the mechanosensitivity of the p38, p70S6k, and JNK2 signaling Physiol 2010, 299:C706–C713. pathways in skeletal muscle. J Appl Physiol (1985) 2005, 98:1562–1566. 22. Lipskaia L, Pinet C, Fromes Y, Hatem S, Cantaloube I, Coulombe A, Lompré A-M: 44. Martineau LC, Gardiner PF: Insight into skeletal muscle Mutation of δ-sarcoglycan is associated with Ca2 + -dependent vascular mechanotransduction: MAPK activation is quantitatively related to remodeling in the Syrian hamster. Am J Pathol 2007, 171:162–171. tension. J Appl Physiol (1985) 2001, 91:693–702. 23. Kumar A, Khandelwal N, Malya R, Reid MB, Boriek AM: Loss of dystrophin causes aberrant mechanotransduction in skeletal muscle fibers. FASEB J 45. Glass DJ: Signalling pathways that mediate skeletal muscle hypertrophy 2004, 18:102–113. and atrophy. Nat Cell Biol 2003, 5:87–90. Moorwood et al. Skeletal Muscle 2014, 4:13 Page 13 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 46. Hahn K, Miranda M, Francis VA, Vendrell J, Zorzano A, Teleman AA: PP2A 65. Boppart MD, Burkin DJ, Kaufman SJ: Activation of AKT signaling promotes regulatory subunit PP2A-B’ counteracts S6K phosphorylation. Cell Metab cell growth and survival in α7β1 integrin-mediated alleviation of muscular 2010, 11:438–444. dystrophy. Biochim Biophys Acta Mol Basis Dis 2011, 1812:439–446. 47. Peterson RT, Desai BN, Hardwick JS, Schreiber SL: Protein phosphatase 2A 66. Eghtesad S, Jhunjhunwala S, Little SR, Clemens PR: Effect of rapamycin on interacts with the 70-kDa S6 kinase and is activated by inhibition of immunity induced by vector-mediated dystrophin expression in mdx FKBP12-rapamycinassociated protein. Proc Natl Acad Sci U S A 1999, skeletal muscle. Sci Rep 2012, 2:399. 96:4438–4442. 67. Shah OJ, Kimball SR, Jefferson LS: Glucocorticoids abate p70S6k and eIF4E 48. Petritsch C, Beug H, Balmain A, Oft M: TGF-beta inhibits p70 S6 kinase via function in L6 skeletal myoblasts. Am J Physiol Endocrinol Metabol 2000, protein phosphatase 2A to induce G(1) arrest. Genes Dev 2000, 14:3093–3101. 279:E74–E82. 49. Cho DH, Choi YJ, Jo SA, Ryou J, Kim JY, Chung J, Jo I: Troglitazone acutely inhibits protein synthesis in endothelial cells via a novel mechanism doi:10.1186/2044-5040-4-13 involving protein phosphatase 2A-dependent p70 S6 kinase inhibition. Cite this article as: Moorwood et al.: Absence of γ-sarcoglycan alters the Am J Physiol Cell Physiol 2006, 291:C317–C326. response of p70S6 kinase to mechanical perturbation in murine skeletal muscle. Skeletal Muscle 2014 4:13. 50. Liu J, Stevens PD, Li X, Schmidt MD, Gao T: PHLPP-mediated dephosphorylation of S6K1 inhibits protein translation and cell growth. Mol Cell Biol 2011, 31:4917–4927. 51. Ketroussi F, Giuliani M, Bahri R, Azzarone B, Charpentier B, Durrbach A: Lymphocyte cell-cycle inhibition by HLA-G is mediated by phosphatase SHP-2 and acts on the mTOR pathway. PLoS One 2011, 6:e22776. 52. Zito CI, Qin H, Blenis J, Bennett AM: SHP-2 regulates cell growth by controlling the mTOR/S6 kinase 1 pathway. JBiol Chem 2007, 282:6946–6953. 53. Hornberger TA, Chu WK, Mak YW, Hsiung JW, Huang SA, Chien S: The role of phospholipase D and phosphatidic acid in the mechanical activation of mTOR signaling in skeletal muscle. Proc Natl Acad Sci U S A 2006, 103:4741–4746. 54. You JS, Frey JW, Hornberger TA: Mechanical stimulation induces mTOR signaling via an ERK-independent mechanism: implications for a direct activation of mTOR by phosphatidic acid. PLoS One 2012, 7:e47258. 55. You JS, Lincoln HC, Kim CR, Frey JW, Goodman CA, Zhong XP, Hornberger TA: The role of diacylglycerol kinase zeta and phosphatidic acid in the mechanical activation of mammalian target of rapamycin (mTOR) signaling and skeletal muscle hypertrophy. J Biol Chem 2014, 289:1551–1563. 56. Richard-Bulteau H, Serrurier B, Crassous B, Banzet S, Peinnequin A, Bigard X, Koulmann N: Recovery of skeletal muscle mass after extensive injury: positive effects of increased contractile activity. Am J Physiol Cell Physiol 2008, 294:C467–C476. 57. Mukhopadhyay NK, Price DJ, Kyriakis JM, Pelech S, Sanghera J, Avruch J: An array of insulin-activated, proline-directed serine/threonine protein kinases phosphorylate the p70 S6 kinase. JBiol Chem 1992, 267:3325–3335. 58. Carriere A, Romeo Y, Acosta-Jaquez HA, Moreau J, Bonneil E, Thibault P, Fingar DC, Roux PP: ERK1/2 Phosphorylate Raptor to Promote Ras-dependent Activation of mTOR Complex 1 (mTORC1). JBiol Chem 2011, 286:567–577. 59. Ma L, Chen Z, Erdjument-Bromage H, Tempst P, Pandolfi PP: Phosphorylation and Functional Inactivation of TSC2 by Erk: Implications for Tuberous Sclerosisand Cancer Pathogenesis. Cell 2005, 121:179–193. 60. Roux PP, Ballif BA, Anjum R, Gygi SP, Blenis J: Tumor-promoting phorbol esters and activated Ras inactivate the tuberous sclerosis tumor suppressor complex via p90 ribosomal S6 kinase. Proc Natl Acad Sci U S A 2004, 101:13489–13494. 61. Axelrod MJ, Gordon V, Mendez RE, Leimgruber SS, Conaway MR, Sharlow ER, Jameson MJ, Gioeli DG, Weber MJ: p70S6 kinase is a critical node that integrates HER-family and PI3 kinase signaling networks. Cell Signal 2014, 26:1627–1635. 62. Hornberger TA, Chien S: Mechanical stimuli and nutrients regulate rapamycin-sensitive signaling through distinct mechanisms in skeletal muscle. J Cell Biochem 2006, 97:1207–1216. 63. Bentzinger CF, Romanino K, Cloëtta D, Lin S, Mascarenhas JB, Oliveri F, Xia J, Submit your next manuscript to BioMed Central Casanova E, Costa CF, Brink M, Zorzato F, Hall MN, Ruegg MA: Skeletal and take full advantage of: muscle-specific ablation of raptor, but not of rictor, causes metabolic changes and results in muscle dystrophy. Cell Metab 2008, 8:411–424. • Convenient online submission 64. Risson V, Mazelin L, Roceri M, Sanchez H, Moncollin V, Corneloup C, • Thorough peer review Richard-Bulteau H, Vignaud A, Baas D, Defour A, Freyssenet D, Tanti JF, Le-Marchand-Brustel Y, Ferrier B, Conjard-Duplany Romanino K, Bauche S, • No space constraints or color figure charges Hantai D, Mueller M, Kozma SC, Thomas G, Ruegg MA, Ferry A, Pende M, • Immediate publication on acceptance Bigard X, Koulmann N, Schaeffer L, Gangloff YG: Muscle inactivation of mTOR causes metabolic and dystrophin defects leading to severe • Inclusion in PubMed, CAS, Scopus and Google Scholar myopathy. J Cell Biol 2009, 187:859–874. • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Skeletal Muscle Springer Journals

Absence of γ-sarcoglycan alters the response of p70S6 kinase to mechanical perturbation in murine skeletal muscle

Loading next page...
 
/lp/springer-journals/absence-of-sarcoglycan-alters-the-response-of-p70s6-kinase-to-lbUOliaJqs
Publisher
Springer Journals
Copyright
Copyright © 2014 by Moorwood et al.; licensee BioMed Central Ltd.
Subject
Life Sciences; Cell Biology; Developmental Biology; Biochemistry, general; Systems Biology; Biotechnology
eISSN
2044-5040
DOI
10.1186/2044-5040-4-13
pmid
25024843
Publisher site
See Article on Publisher Site

Abstract

Background: The dystrophin glycoprotein complex (DGC) is located at the sarcolemma of muscle fibers, providing structural integrity. Mutations in and loss of DGC proteins cause a spectrum of muscular dystrophies. When only the sarcoglycan subcomplex is absent, muscles display severe myofiber degeneration, but little susceptibility to contractile damage, suggesting that disease occurs not by structural deficits but through aberrant signaling, namely, loss of normal mechanotransduction signaling through the sarcoglycan complex. We extended our previous studies on mechanosensitive, γ-sarcoglycan-dependent ERK1/2 phosphorylation, to determine whether additional pathways are altered with the loss of γ-sarcoglycan. Methods: We examined mechanotransduction in the presence and absence of γ-sarcoglycan, using C2C12 -/- myotubes, and primary cultures and isolated muscles from C57Bl/6 (C57) and γ-sarcoglycan-null (γ-SG )mice. All were subjected to cyclic passive stretch. Signaling protein phosphorylation was determined by immunoblotting of lysates from stretched and non-stretched samples. Calcium dependence was assessed by maintaining muscles in calcium-free or tetracaine-supplemented Ringer’s solution. Dependence on mTOR was determined by stretching isolated muscles in the presence or absence of rapamycin. Results: C2C12 myotube stretch caused a robust increase in P-p70S6K, but decreased P-FAK and P-ERK2. Neither Akt nor ERK1 were responsive to passive stretch. Similar but non-significant trends were observed in C57 primary -/- cultures in response to stretch, and γ-SG cultures displayed no p70S6K response. In contrast, in isolated muscles, -/- p70S6K was mechanically responsive. Basal p70S6K activation was elevated in muscles of γ-SG mice, in a -/- calcium-independent manner. p70S6K activation increased with stretch in both C57 and γ-SG isolated muscles, -/- and was sustained in γ-SG muscles, unlike the transient response in C57 muscles. Rapamycin treatment blocked all of p70S6K activation in stretched C57 muscles, and reduced downstream S6RP phosphorylation. However, even -/- though rapamycin treatment decreased p70S6K activation in stretched γ-SG muscles, S6RP phosphorylation remained elevated. Conclusions: p70S6K is an important component of γ-sarcoglycan-dependent mechanotransduction in skeletal muscle. Our results suggest that loss of γ-sarcoglycan uncouples the response of p70S6K to stretch and implies that γ-sarcoglycan is important for inactivation of this pathway. Overall, we assert that altered load-sensing mechanisms exist in muscular dystrophies where the sarcoglycans are absent. Keywords: Sarcoglycan, Sarcoglycanopathies, Limb girdle muscular dystrophy, Mechanotransduction, Mechano-sensing, Load-sensing, p70S6K, S6K, p70S6 kinase, ERK1/2 * Correspondence: erbarton@dental.upenn.edu Department of Anatomy and Cell Biology, School of Dental Medicine, University of Pennsylvania, Philadelphia, PA, USA Pennsylvania Muscle Institute, University of Pennsylvania, Philadelphia, PA, USA Full list of author information is available at the end of the article © 2014 Moorwood et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Moorwood et al. Skeletal Muscle 2014, 4:13 Page 2 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 Background the muscular dystrophies results from abnormal levels and The dystrophin glycoprotein complex (DGC) is found at activity of the mechanosensitive TRP channels and/or mis- the sarcolemma of skeletal, cardiac, and smooth muscle regulation of store operated calcium entry via the STIM1 cells, where it physically links the extracellular matrix and Orai1 channels [24-29]. Direct disruption of the sarco- (ECM) with the intracellular cytoskeleton, providing struc- lemma, for which there is evidence in SG-null animal tural support [1-3]. Mutations in DGC components cause models [7,9,30,31], could also contribute to loss of calcium different types of muscular dystrophy; for example, muta- homeostasis. Furthermore, several strategies to improve 2+ tions in dystrophin cause Duchenne muscular dystrophy Ca handling are known to counteract the pathology as- (DMD), while mutations in α-, β-, γ-, or δ-sarcoglycan sociated with the muscular dystrophies [25,32-34]. There- (SG) cause limb girdle muscular dystrophy (LGMD) [2-4]. fore, identification of mechanosensitive signaling that is When dystrophin is mutated in DMD or in the mdx attributable to the SG complex rather than other pro- mouse model of DMD, the entire DGC is substantially cesses occurring during mechanical perturbation has been reduced at the sarcolemma. In contrast, when any one of challenging. the SGs is mutated, in LGMD or any of the SG knock-out One pathway of interest involves p70S6K, which is ca- mice, the other three SGs are either absent or reduced at nonically activated in response to mitogens via the phos- the sarcolemma, but the rest of the DGC remains, includ- phoinositide 3-kinase (PI3K) pathway (reviewed in [35]) ing the link formed by dystrophin and dystroglycan be- and is known to respond to mechanical load [36]. Acti- tween the ECM and the cytoskeleton [2,3,5,6]. Unlike the vation of p70S6K involves a hierarchical series of phos- skeletal muscles of the mdx mouse, muscles of the γ-SG phorylation events, beginning with phosphorylation of -/- knock-out (γ-SG ) mouse display no mechanical fragility, multiple sites in the C-terminal autoinhibitory domain, at least until 4 months of age, as shown by a minimal loss followed by mammalian target of rapamycin (mTOR)- of force-generating capacity following a series of eccentric dependent phosphorylation of sites in the linker region, -/- contractions (ECCs) [7,8]. In spite of this, the γ-SG which allows for full activation of the kinase via phos- mouse exhibits a severe dystrophic phenotype on histo- phorylation of threonine 229 (T229) in the catalytic do- logical examination, with extensive myofiber degeneration main by phosphoinositide-dependent kinase 1 (PDK1) and regeneration, fibrosis, and disruption of sarcolemmal (reviewed in [37]). Although phosphorylation of T229 is integrity, similar to the mdx mouse [9]. The lack of required for p70S6K activation, phosphorylation of T389 mechanical fragility suggests that aberrant signaling in the linker region has been found to correlate most may contribute to the muscle degeneration seen in the closely with in vivo activity [38], and can be used as a -/- γ-SG mouse. Indeed, our previous studies demon- measure of kinase activation. p70S6K has a multitude of strated that localization of the SG complex to the sarco- downstream targets, with roles in protein synthesis, lemma and phosphorylation of the tyrosine 6 residue of growth, proliferation, survival, and more [35], including γ-SG are essential for normal signaling by extracellular S6 ribosomal protein (S6RP), which closely correlates signal-regulated kinases 1 and 2 (ERK1/2), in response with protein translation rates [39]. to ECCs [10,11]. Based upon these data, we have In the current study, we examined ERK1/2, Akt, focal asserted that the SG complex acts as a mechanosensor adhesion kinase (FAK), and p70S6K responses to passive -/- in skeletal muscle because of its position within the stretch in C57 and γ-SG skeletal muscle to further eluci- DGC, the modifications that occur to γ-SG with mech- date the importance of the SG complex for mechanotrans- anical perturbation, and the necessity of the complex duction. While differences in ERK1/2 phosphorylation -/- for normal signaling. between C57 and γ-SG muscle were calcium-dependent, A complication of using ECCs to invoke signal transduc- differences in p70S6K activation were independent of cal- tion is that there are multiple processes in play. Not only is cium. In addition, the p70S6K response to stretch in both there externally applied tension on the sarcolemma and primary myofiber cultures and isolated extensor digitorum DGC as a result of lengthening, but also active contraction longus (EDL) muscles was differentially regulated in the 2+ resulting in internally applied tension and Ca flux within absence of γ-SG. Specifically, experiments in isolated the fibers and across the sarcolemma, all of which could muscles suggest that γ-SG is required for inactivation of contribute to alterations in signaling in the absence of p70S6K. The findings increase our understanding of the γ-SG. Calcium is known to be involved in many mechano- contribution of aberrant load-sensing to the pathology of sensitive signaling pathways (reviewed in [12,13]) and aber- muscular dystrophies where the SG complex is deficient. rant calcium regulation is a feature of SG-deficient muscle [14-22]. Indeed, an exaggerated ERK1/2 response to mech- Methods anical stimulation that is dependent on extracellular cal- Animals -/- cium has been demonstrated in the mdx mouse diaphragm Adult C57BL/6 (C57) and γ-SG-null (γ-SG )mice were [23]. Studies suggest that aberrant calcium regulation in used. For ex vivo protocols, mice were aged 8 to 16 weeks. Moorwood et al. Skeletal Muscle 2014, 4:13 Page 3 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 -/- The γ-SG mouse lacks γ-SG due to gene targeting, below. Unless otherwise indicated, all cell culture reagents resulting in an additional loss of β-and δ-SG and a de- were purchased from Gibco. crease of α-SG [9]. All experiments were approved by the University of Pennsylvania Institutional Animal Care and Myotube stretching protocol Use Committee. C2C12 myotubes were stretched using an apparatus that produces isotropic two-dimensional strain of cells in vitro, as described previously [40]. Briefly, myotubes were sub- C2C12 myotube culture jected to 10% strain, 40 times per minute, for 30 min, at Flexible silicone membranes (Specialty Manufacturing, 37°C in a humidified atmosphere of 5% CO in air. Control Inc.) were stretched across the bottom of custom cylinders myotubes were treated identically but were not stretched. which acted as a culture chamber. The membranes were Lysates were harvested immediately as described below. held in place using an O-ring as described previously [40]. Primary myotubes were stretched using the protocol Membranes were then coated with a thin layer of 2 mg/mL described for C2C12 myotubes. Stretched myotubes were GFR Matrigel (BD #354230). C2C12 myoblasts (3.5 × 10 / then incubated without stretch for a further 1, 2, or 4 h at cylinder) were seeded onto the membranes and maintained 37°C before harvesting lysates for immunoblotting as at 5% CO at 37°C in growth media (10% FBS, 100 U described below. Control (non-stretched) myotubes were penicillin, 100 μg streptomycin, 100 μg/mL gentamycin harvested immediately after the end of the stretching in DMEM) for approximately 24 h until 70% to 80% protocol used for the stretched myotubes. confluent, then switched to differentiation media (2% HS, 100 U penicillin, 100 μg streptomycin, 100 μg/mL Isolated muscle stretching protocol Gentamycin in DMEM). Myoblasts were allowed to Mice were anaesthetized using ketamine and xylazine. EDL differentiate into multinucleated myotubes for 5 days, muscles were dissected and placed in an organ bath during which media was changed every other day before containing oxygenated high-glucose (25 mM) DMEM with stretching as described below. HEPES (25 mM) (Life Technologies), at room temperature. For rapamycin sensitivity experiments, high-glucose DMEM Primary myoblast culture was supplemented with 150 nM rapamycin (Sigma) or ve- Mice were euthanized using CO inhalation. Flexor digi- hicle only (0.1% DMSO). Muscles were adjusted to 9.3 mN torum brevis (FDB) muscles were dissected and incu- of resting tension, approximately equivalent to optimal bated with 2 mg/mL collagenase I (Sigma), 10% FBS in length, based on our previous experiments [42]. After a Tyrode’s solution (125 mM NaCl, 5 mM KCl, 1 mM 10-min equilibration period, the length of the muscle was CaCl , 1 mM MgCl , 1 mM KH PO , 20 mM HEPES measured using calipers, and the muscle was subjected to 2 2 2 4 (all from Fisher), 5.5 mM glucose (Sigma), pH 7.4) for a stretching protocol of 15% strain (held for 100 ms with 90 min at 37°C, with shaking, as previously described ramp times of 50 ms), 20 times per minute, for either 30 [41]. Muscles were washed in 10% FBS in Tyrode’s solu- or 90 min, using an in vitro muscle test system (Aurora tion and clumps of fibers were liberated by pipetting up Scientific). Muscles were snap-frozen immediately follow- and down in 10% FBS, 100 U penicillin, 100 μg strepto- ing the end of the stretch protocol. Contralateral muscles mycin in Tyrode’s solution using a wide-mouthed glass were used as controls and were adjusted to the same length pipette. Clumps of fibers were transferred to a second as the stretched muscles, then incubated in oxygenated dish with the same solution and pipetted up and down high-glucose DMEM with HEPES at room temperature for again. Single fibers were transferred to a third dish be- the equivalent length of time, before snap-freezing. fore plating on silicone membranes coated with Matrigel (Becton Dickinson; 2 mg/mL diluted in DMEM). Growth Calcium dependence experiments media (20% FBS, 10 ng/mL mouse basic fibroblast growth EDL muscles were dissected as described above and factor (MP Bio), 100 U penicillin, 100 μgstreptomycin, placed in an organ bath contained oxygenated Ringer’s 1 μg/mL Fungizone, 100 μg/mL Gentamycin in Ham’s solution (119 mM NaCl, 4.74 mM KCl, 2.54 mM CaCl , F-10 media) was carefully added and cultures were incu- 1.18 mM KH PO , 1.18 mM MgSO , 25 mM HEPES, 2 4 4 bated without disturbance for 3 days at 37°C. Media was 2.75 mM glucose), calcium-free Ringer’s solution (CaCl then changed every 2 days. After 7 to 10 days, when myo- replaced with 2.5 mM MgCl ) or Ringer’s solution sup- blast cultures were 70% to 80% confluent, media was plemented with 100 μM tetracaine. They were incubated switched to differentiation media (10% HS, 0.5% chicken for 30 min before snap-freezing in liquid nitrogen. embryo extract, 100 U penicillin, 100 μg streptomycin, 1 μg/mL Fungizone, 100 μg/mL Gentamycin in DMEM) Immunoprecipitation and myoblasts were allowed to differentiate into multinu- Immunoprecipitation experiments were carried out using cleated myotubes for 5 days before stretching as described the Pierce Classic IP kit (Thermo Scientific). Muscle lysates Moorwood et al. Skeletal Muscle 2014, 4:13 Page 4 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 containing 100 μg total protein was immunoprecipitated Statistics with anti-P-tyrosine (Cell Signaling #9411 1:100) overnight Comparisons between non-stretched and stretched C2C12 at 4°C with end-over-end mixing. Samples were purified cells (Figure1)weredonebyunpairedTtest.Comparisons -/- using Protein A/G Plus Agarose beads (Roche) for 1 h between C57 and γ-SG primary cultures across time at 4°C with end-over-end mixing. The immune complex (Figure2)weredonebytwo-way ANOVA with Tukey’s was eluted with non-reducing sample buffer and boiled multiple comparisons test. Comparisons between C57 and -/- at 100°C for 5 min before being applied to a SDS-PAGE γ-SG muscles in calcium experiments (Figure 3) were gel, transferred, and immunoblotted as described below. done by unpaired T test. Comparisons between C57 and -/- γ-SG muscles with and without stretch for each time Immunoblotting point (Figure 4) or with and without rapamycin (Figure 5) C2C12 and primary myotubes were washed with ice-cold were done by two-way ANOVA with Tukey’smultiple PBS before lysing in 100 to 200 μLofRIPAbuffer(50 mM comparisons test. HEPES pH 7.5, 150 mM NaCl, 5 mM EDTA, 1 mM EGTA, 15 mM p-nitrophenyl phosphate disodium hexahydral, 1% Results NP-40, 0.1% SDS, 1% deoxycholate, 0.025% sodium azide) p70S6K, but not ERK1/2 or Akt responds to passive with protease and phosphatase inhibitor cocktails (Sigma). stretch in vitro Lysates were incubated on ice for 30 min, centrifuged at Studies using whole muscles from animal models of the 16,000 rcf for 20 min at 4°C and the supernatants retained. dystrophies are made more complex by the presence of EDL muscles were ground using a dry ice-cooled pestle multiple cell types, as well as pathological processes such and mortar, and lysed in 200 μL of RIPA buffer with prote- as fibrosis. Therefore, we initially investigated mechano- ase and phosphatase inhibitors. Lysates were incubated on transduction signaling in passively stretched C2C12 ice for 1 h, vortexing half-way through, centrifuged at myotubes. We found that passive stretching in vitro did 16,000 rcf for 20 min at 4°C and the supernatants retained. not alter phosphorylation of ERK1 or Akt and that ERK2 Protein content was determined using a Bradford method and FAK phosphorylation decreased following stretch protein assay kit (Bio-Rad). Lysates (30 μg total protein for (Figure 1A,C-F,H-I). However, we found an increase in myotube cultures, 90 μg total protein for EDL muscles) p70S6K phosphorylation at T389, which reflects activity, were separated by SDS-PAGE on Tris-HCl polyacrylamide in response to passive stretching of C2C12 myotubes gels (Bio-Rad) and transferred to PVDF membranes. (Figure 1B,G). Therefore, this in vitro model reflected Membranes were blocked in 5% milk in TTBS (10 mM some, but not all, stretch responses observed in muscle Tris, 150 mM NaCl, 0.1% v/v Tween-20, pH 8), with 2% in vivo [43,44], and highlighted p70S6K as a pathway of BSA added for some antibodies, then probed with interest. The lack of Akt phosphorylation suggests that antibodies to the following: phospho (P)-p70S6K (T389) p70S6K phosphorylation occurred through an Akt- (Cell Signaling #9205 1:200 for primary cultures, Cell independent pathway, while the lack of FAK phosphoryl- Signaling #9234 1:250 for isolated muscles), P-p70S6K ation supports an integrin-independent mechanism. (T421/S424) (Cell Signaling #9204 1:1,000), P-S6RP (Cell Signaling #2211 1:2,000), P-ERK1/2 (Cell Signaling #9101 Differential p70S6K stretch response occurs in C57 and -/- 1:500), total (T)-ERK1/2 (Cell Signaling #9107 1:1,000), γ-SG primary cultures P-Akt (Cell Signaling #9271 1:300), T-Akt (Cell Sig- Having established pathways of interest in vitro using the naling #2920 1:2,000), P-FAK (Millipore 05-1140, 1:500), C2C12 cell line, we used primary myoblast cultures from -/- T-FAK (Millipore 06-543, 1:1,000), γ-SG (Novocastra C57 and γ-SG mice (Figure 2A) to investigate the VP-G803 1:300), glyceraldehyde 3-phosphate dehydrogenase changes in mechanotransduction signaling associated (GAPDH) (Santa Cruz sc-32233 1:5,000), and tubulin solely with the loss of the SG complex in myofibers. C57 -/- (Sigma T5168 1:20,000). Band intensities were quantified and γ-SG cultures were stretched for 30 min as de- using ImageQuant TL, 1D gel analysis (C2C12 and myo- scribed above and lysates were harvested 1, 2, or 4 h after tube cultures, rapamycin sensitivity experiments) or ImageJ stretching ended, to allow observation of the signaling (NIH) (all other isolated muscles). P-p70S6K and P-S6RP time course. Immunoblotting analysis showed that there were normalized to either GAPDH or tubulin, P-FAK was was no difference in basal ERK1/2 phosphorylation -/- normalized to either T-FAK or GAPDH, P-Akt was normal- between C57 and γ-SG myotubes and little change in ized to T-Akt, and P-ERK1/2 was normalized to T-ERK1/2. ERK1/2 phosphorylation in response to stretch (Figure 2B, D,E). Therefore, similar to the C2C12 cells, primary Microscopy cultures did not reflect the ERK1/2 phosphorylation -/- Images of primary cultures were taken using a Leica DM responses found previously in C57 and γ-SG mice RBE microscope and Leica DFC300 CCD camera, using in vivo [10,11]. Neither P-Akt nor P-FAK displayed -/- OpenLab software (Perkin Elmer). significant differences between C57 and γ-SG cultures Moorwood et al. Skeletal Muscle 2014, 4:13 Page 5 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 Figure 1 p70S6K responds to stretch in C2C12 cells. C2C12 myotubes were cultured on silicone membranes and subjected to passive stretching for 30 min. (A-D) Representative immunoblots for P-ERK1/2, total T-ERK1/2, P-p70S6K (T389 site), tubulin, P-Akt, T-Akt, P-FAK, and T-FAK in non-stretched (NS) and stretched (S) C2C12 cells. (E-I) Quantification of activation levels. P-ERK1 and 2 were normalized to T-ERK1 and 2, respectively, P-p70S6K was normalized to tubulin, P-Akt was normalized to T-Akt, and P-FAK was normalized to T-FAK. n = 5-6 wells of C2C12 cells per group. Bars represent mean ± standard error. * Significantly different from non-stretched myotubes by unpaired T test. NS, non-stretched; S, stretched. (Figure 2B,F,G). Further, these proteins did not show basal phosphorylation state of the upstream pathways any prolonged response to passive stretch in the (Figure 3D; Normal). primary cultures, consistent with the lack of an acute Heightened mechanosensitive signaling could arise through positive stretch response in C2C12 cells. Examination of increased flux of ions across the sarcolemma, particu- 2+ p70S6K revealed a trend towards elevated T389 phos- larly Ca [17]. This could occur either through en- -/- phorylation at baseline in γ-SG myotubes, compared hanced activity of channels, such as the TRP family of -/- to C57 myotubes. Activation of p70S6K in γ-SG myo- cation channels (reviewed in [28,29]), or via membrane tubes upon stretch was not apparent (Figure 2B,C), but ruptures. To determine the calcium dependence of the ob- C57 myotubes displayed a trend for increased activation served differences in p70S6K and ERK1/2 phosphorylation -/- at 2 and 4 h after stretching (Figure 2B,C). Therefore, in γ-SG muscles, we incubated muscles in calcium-free while the primary cultures reflect some of the responses Ringers solution, retaining the same ionic strength. 2+ found in C2C12 myotubes, the experimental system is Absence of extracellular Ca did not alter the relative -/- too variable to draw firm conclusions regarding mech- difference in P-p70S6K between γ-SG and C57 muscles anical signaling pathways associated with γ-SG. (Figure 3C). However, the increased P-ERK1/2 and -/- P-S6RP found in γ-SG muscles in normal Ringer’ssolu- -/- Elevated P-p70S6K in γ-SG muscles is calcium tion was abrogated when there was no calcium in the independent bathing solution (Figure 3A,B,D). Thus, only p70S6K Our experiments in myotubes, together with our previ- phosphorylation appeared to be calcium independent. Be- ous studies in isolated muscles [10,11], suggested that cause intracellular calcium stores can also alter the intra- ERK1/2 phosphorylation changes require active contrac- cellular calcium concentration, particularly during muscle tion in addition to stretch, whereas p70S6K responds to activation, tetracaine was used to inhibit sarcoplasmic 2+ stretch alone. To test this hypothesis in vivo, we ex- reticulum release of Ca through the ryanodine receptors. tended our analysis to examine the response of p70S6K Verification of this inhibition was established in a separate to passive stretch of isolated muscles. P-p70S6k was experiment, through measuring tetanic force generation -/- elevated approximately 1.7-fold in resting γ-SG EDL by EDL muscles before and after addition of tetracaine. muscles incubated in normal oxygenated Ringer’s solu- After 15 min incubation with tetracaine, force production tion (Figure 3C; Normal). As in our previous studies, was virtually eliminated (force was 297 mN prior to there was a 3- and 1.5-fold increase of P-ERK1 and addition of tetracaine, 2.4 mN after 15 min incubation -/- P-ERK2, respectively, in resting γ-SG EDL muscles with tetracaine and not detectable after 20 min incubation (Figure 3A-B; Normal). Because both of these pathways with tetracaine). Again, relative P-p70S6K levels between converge to phosphorylate S6RP, we compared the the two muscle groups were not altered by tetracaine 2+ phosphorylation state of this protein in muscles from (Figure 3C). In contrast, blockade of SR Ca release -/- -/- both genotypes. γ-SG EDL muscles exhibited an 8-fold reduced P-ERK1 levels in γ-SG muscles relative to C57 increase in P-S6RP, which was consistent with the higher muscles, even though there was no alteration in P-ERK2 Moorwood et al. Skeletal Muscle 2014, 4:13 Page 6 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 -/- -/- Figure 2 Differential p70S6K stretch response in C57 and γ-SG primary cultures. Primary myotubes from C57 and γ-SG FDB fibers were cultured on silicone membranes and subjected to passive stretching for 30 min. Lysates were harvested 1, 2, or 4 h after stretch. (A) Representative images of i, satellite cells migrating from an FDB fiber and ii, differentiated myotubes. (B) Representative immunoblot for P-p70S6K (T389 site), P- and T-ERK1/2, P-FAK, P-Akt, T-Akt, and GAPDH in non-stretched (C) and stretched (1 h, 2 h, 4 h) primary cultures. (C-G) Quantification of activation levels. Legend in C applies to all graphs. P-ERK1 and 2 were normalized to T-ERK1 and 2, respectively, P-Akt was normalized to T-Akt and P-p70S6K and P-FAK were normalized to GAPDH. n = 3 (p70S6K) or 4 (all other proteins) independent sets of primary cultures per genotype. Bars represent mean ± standard error. Statistical significance was tested by two-way ANOVA. (Figure 3A,B). Phosphorylation of S6RP remained elevated γ-SG-dependent mechanotransduction in vivo. A passive -/- in γ-SG muscles in the presence of tetracaine, but to a stretching protocol comprised of a 15% strain, 20 times lesser extent than in normal Ringer’s solution (Figure 3D). per min, for 30 min in high-glucose DMEM was sufficient Taken together, both extracellular and intracellular cal- to cause increased γ-SG phosphorylation in the EDL, as is cium contribute to the heightened P-ERK1 levels in the case for eccentric contraction of the EDL (Figure 4A; -/- -/- γ-SG muscles, whereas the increase in basal P-p70S6K [10]). We stretched C57 and γ-SG EDL muscles for -/- in γ-SG muscles is not dependent on either extracellular either 30 or 90 min and immediately snap-froze them in or intracellular calcium. liquid nitrogen. We then used immunoblotting to meas- ure phosphorylation of p70S6K at T389, as above, and also Differential p70S6K stretch response occurs in isolated at T421/S424 in the auto-inhibitory domain. Basal phos- -/- C57 and γ-SG muscles phorylation of p70S6K at T389 showed a trend to be in- -/- Having established that P-p70S6K changes at rest did not creased in unstretched γ-SG muscles compared to C57 depend on calcium, we pursued the role of p70S6K in muscles (Figure 4B,C). However, basal phosphorylation at Moorwood et al. Skeletal Muscle 2014, 4:13 Page 7 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 -/- -/- Figure 3 Elevated p70S6K in γ-SG muscles is independent of calcium. EDL muscles from C57 and γ-SG mice were maintained in normal oxygenated Ringer’s solution, calcium-free oxygenated Ringer’s solution or oxygenated Ringer’s solution supplemented with tetracaine for 30 min. (A-D) Representative immunoblots and quantification for P-ERK1 (A),P-ERK2 (B), P-p70S6K (T389 site; C), and P-S6RP (D).Legend in A applies to all graphs. P-ERK 1 and 2 were normalized to T-ERK 1 and 2, respectively; P-p70S6K and P-S6RP were normalized to tubulin. Independent immunoblots -/- were performed for each condition and γ-SG activation levels were normalized to C57 activation levels in each case. n = 3 muscles per genotype and condition. Bars represent mean ± standard error. All datasets were tested by unpaired T test. * Significantly different from C57 by unpaired T test. -/- T421/S424 was not different between C57 and γ-SG activation of p70S6K in response to passive stretch, imply- muscles (Figure 4D). After 30 min of stretch, phosphoryl- ing that γ-SG plays a role in p70S6K inactivation. ation of p70S6K at T389 and T421/S424 was increased to -/- a similar degree in stretched C57 and γ-SG EDL mus- Stretch response of p70S6K T389, but not S6RP, is -/- cles, compared to non-stretched controls. However, after rapamycin-sensitive in γ-SG muscles 90 min of stretch, phosphorylation at T389 and T421/ Because mTOR is a key mediator of p70S6K activation, S424 had decreased in stretched C57 muscles and was we examined the effect of the mTOR inhibitor rapamycin -/- close to non-stretched levels. In contrast, phosphorylation on stretch responses in isolated C57 and γ-SG muscles. at T389 was further increased, and T421/S424 phosphor- EDL muscles were subjected to cyclic stretch for 90 min, -/- ylation remained elevated, in stretched γ-SG muscles, as described above. Unlike in C2C12 cells and primary compared to non-stretched controls (Figure 4B-D). For cultures, P-Akt showed a trend to increase on stretching, -/- T389 phosphorylation after 90 min of stretch, genotype, which was statistically significant in γ-SG muscles. stretch, and the interaction between them were all statisti- As anticipated, P-Akt was unaffected by rapamycin cally significant, by two-way ANOVA. T421/S424 showed (Figure 5A,B). Rapamycin treatment completely blocked the a similar trend to T389; however, while the effect of increase in p70S6K T389 phosphorylation after passive stretch was statistically significant, the difference between stretch of C57 muscles, consistent with previous studies -/- genotypes was not. Thus, in contrast to primary cultures, [36,37]. In γ-SG muscles, rapamycin abrogated most -/- γ-SG muscles exhibited heightened and prolonged p70S6K T389 phosphorylation, but residual phosphorylation Moorwood et al. Skeletal Muscle 2014, 4:13 Page 8 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 remained in stretched muscles (Figure 5A,C). T421/S424 showed a trend to increase in response to stretch in both -/- C57 and γ-SG muscles. Surprisingly, rapamycin blunted the p70S6K T421/S424 stretch response in C57 muscles, which is inconsistent with previous studies [36]. However, -/- the T421/S424 response to stretch persisted in γ-SG muscles in the presence of rapamycin (Figure 5A,D). S6RP phosphorylation increased in response to stretch -/- in both C57 and γ-SG muscles. Interestingly, while rapamycin blocked stretch-induced phosphorylation of S6RP in C57 muscles, phosphorylation in response to -/- stretch was preserved in γ-SG muscles (Figure 5A,E). Taken together, these results suggest either that the level -/- of active p70S6K remaining in γ-SG musclesissuffi- cient to phosphorylate S6RP regardless of rapamycin or that an alternate pathway bypasses p70S6K to phos- phorylate S6RP in muscles lacking γ-SG. Discussion Skeletal muscle has a remarkable ability to adapt to changes in workload. Almost all muscle properties can be modulated, such as muscle fiber size, contractile properties and metabolism. Changes in patterns of gene expression as well as shifts in the balance between pro- tein synthesis and degradation are required to complete the adaptational response. Identification of major path- ways that directly regulate gene expression and protein synthesis/degradation demonstrate that multiple inputs (mechanical, chemical, and metabolic) can converge on final common pathways for muscle growth and adaptation (reviewed in [45]). However, sorting out the contribution of the wide variety of inputs on muscle adaptation has been more difficult. In our previous work, we used an eccentric contraction protocol to in- vestigate the dependence of ERK1/2 mechano-sensing on phosphorylation of γ-SG. However, this protocol alters multiple factors, including externally applied tension, internally generated tension and changes in extracellular and intracellular calcium fluxes, all of Figure 4 Differential p70S6K stretch response in isolated C57 -/- -/- and γ-SG muscles. EDL muscles from C57 and γ-SG mice were which potentially have effects on mechanosensitive maintained in oxygenated high glucose DMEM and subjected to signaling pathways. In the present study, we used a passive stretching for 30 or 90 min. (A) Immunoblot of γ-SG following passive stretching protocol to isolate the effects of ex- immunoprecipitation with anti-P-Tyr or lysate only, showing γ-SG ternally applied tension in the absence of active contrac- phosphorylation in response to 30 min of stretch. (B) Representative tion, in order to examine the downstream signaling in immunoblots of P-p70S6K (T389 and T421/S424 sites) and tubulin. (C, D) Quantification of P-p70S6K T389 (C) and T421/S424 more detail. (D), normalized to tubulin. Samples for each time point were run Passive stretching protocols can be performed in both on separate immunoblots and normalized to C57 NS. n = 3-5 cell cultures and whole muscle preparations, and the muscles per genotype, condition, and time point. Bars represent reductionist approach of utilizing cultures can eliminate mean ± standard error. All datasets were tested by two-way ANOVA. some of the physiological complexities associated with Stretch was statistically significant for T421/S424 after 30 min of stretch; genotype, stretch, and the interaction between them were intact or isolated muscles. As such, our initial experi- statistically significant T389 after 90 min of stretch, by two-way ments using C2C12 cells were key to identifying p70S6K ANOVA. *Significantly different from C57 S at 90 min by two-way as being activated in response to stretch, in contrast to ANOVA with Tukey’s multiple comparisons test. NS, non-stretched; the lack of response by ERK1/2, Akt, or FAK. Primary S, stretched. -/- myotubes generated from C57 or γ-SG mice had the Moorwood et al. Skeletal Muscle 2014, 4:13 Page 9 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 Figure 5 Stretch response of p70S6K T389, but not S6RP, is -/- rapamycin-sensitive in γ-SG muscles. EDL muscles from C57 -/- and γ-SG mice were maintained in oxygenated high glucose DMEM supplemented with or without rapamycin and subjected to passive stretching for 90 min. (A) Representative immunoblots of P-Akt, T-Akt, P-p70S6K (T389 and T421/S424 sites), P-S6RP, and tubulin. Left panels DMEM alone; right panels DMEM + rapamycin. (B-E) Quantification of P-Akt (B), P-p70S6K T389 (C), P-p70S6K T421/ S424 (D), and P-S6RP (E). Legend in B applies to all graphs. P-Akt was normalized to T-Akt; all other proteins were normalized to tubulin. n = 2-3 muscles per genotype and condition. Bars represent mean ± standard error. All datasets were tested by two-way ANOVA. -/- For P-Akt in γ-SG muscles, stretch was significant. For P-p70S6K -/- T389 in γ-SG muscles, stretch, rapamycin treatment, and the -/- interaction between them were all significant. For P-S6RP in γ-SG -/- muscles, stretch was significant. ‡Significantly different to NS γ-SG -/- control and †significantly different to S γ-SG without rapamycin by two-way ANOVA with Tukey’s multiple comparisons test. NS, non-stretched; S, stretched. distinct advantage of efficient germline elimination of γ-SG combined with an in vitro culture system. Even though these experiments displayed trends towards differ- ential activation of p70S6K after stretch, the inherent variability of the preparation impaired identification of signaling patterns that were dependent upon either stretch or γ-SG. Thus, we returned to isolated muscles from C57 -/- and γ-SG mice to investigate γ-SG-dependent mechano- transduction pathways. Using this model, we observed a -/- modest increase of P-p70S6K in γ-SG muscles at rest that was independent of intra- or extracellular calcium, and a prolonged activation of p70S6K following stretch. These results support a role for γ-SG in particular, or the SG complex in general, in mechanical signal transduction, where the loss of this protein leads to an increase in activation, and deficit in deactivation, or a combination of both. Given the dependence of our findings on the experi- mental platform utilized, future directions will include verification of the results in an even more intact system, such as in situ muscle preparations or whole animals. An intriguing explanation for our in vivo results is that γ-SG is required for dephosphorylation and deactivation of p70S6K. There is considerable evidence that p70S6K is directly dephosphorylated by protein phosphatase 2A (PP2A), independently of mTOR [46-49]. The phosphat- ase PHLPP has also been shown to target p70S6K [50]. γ-SG may mediate the activation of these phosphatases in response to sustained mechanical stimulation. Alter- natively, γ-SG may regulate pathways that deactivate p70S6K indirectly. For example, the phosphatase SHP-2 can cause mTOR-dependent dephosphorylation of p70S6K [51,52]. Further studies will be required to define the inacti- vation pathway disrupted by γ-SG loss. Passive stretch eliminates the contribution of active con- traction or SR calcium fluxes, but does not eliminate the Moorwood et al. Skeletal Muscle 2014, 4:13 Page 10 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 2+ effects of extracellular Ca fluxes through mechanically phosphorylation of the p70S6K T421/S424 autoinhibitory sensitive channels. Further, as previously shown, passive domain sites by ERK1/2. 2+ stretch causes greater Ca influx into myotubes lacking Our experiments with rapamycin showed that, for both -/- members of the SG complex [17], raising the possibility C57 and γ-SG muscles, phosphorylation of p70S6K at that the mechanical signal transduction pathways we eval- T389 is mTOR-dependent, consistent with previous stud- uated previously may be modulated not only by the SG ies [36,37]. The T421/S424 autoinhibitory domain sites complex, but also by additional channels in the sarco- were phosphorylated in response to stretch in both C57 -/- lemma. To address this, we examined the contribution and γ-SG muscles, which was different to our initial of calcium to the elevated p70S6K and ERK1/2 activity experiment in stretched isolated muscles, where the C57 -/- found in γ-SG muscles. We found that the elevation of response had diminished by 90 min of stretching. How- P-ERK1/2 in the absence of γ-SG was dependent on both ever, in the presence of rapamycin, this response was internal and external sources of calcium. In contrast, the not present. This is surprising given that the T421/S424 -/- difference in basal P-p70S6K between C57 and γ-SG sites are not thought to be targeted by mTOR. Previous muscles was not calcium dependent. This suggests that studies have shown these sites to be rapamycin-insensitive while ERK1/2 activation may lie downstream of the cal- [36], but recent evidence suggests a modest mTOR de- -/- cium misregulation that occurs in SG-deficient muscle, pendence [61]. In γ-SG muscles, rapamycin had no ef- changes in p70S6K activation may be a more direct conse- fect on T421/S424 phosphorylation. Further experiments quence of the absence of the SG complex. This is of great are needed to fully understand the role of mTOR on phos- interest given that γ-SG has been shown to be important phorylation of the p70S6K autoinhibitory domain in C57 -/- for mechanotransduction, but the downstream signaling and γ-SG skeletal muscle. Interestingly, the stretch- pathways are uncharacterized [10,11]. Furthermore, p70S6K induced phosphorylation of S6RP was rapamycin sensitive -/- has been implicated in mechanotransduction in skeletal in C57 muscles, but not in γ-SG muscles. This suggests muscle, but the upstream initiation signals are not that an alternative pathway can bring about S6RP phos- -/- known [36,53-55]. However, it should be noted that phorylation in γ-SG muscles when p70S6K is not acti- SG-deficient muscle undergoes substantial degeneration vated. One possibility is that S6RP is phosphorylated by and subsequent regeneration, which may also explain p90 ribosomal S6 kinase, which is activated by ERK; this is -/- the elevated basal p70S6K, which is transiently increased consistent with the increase in basal ERK1/2 in γ-SG during regeneration [56]. muscles, and the over-response of ERK2 on mechanical Our study found that the pattern of differential p70S6K stimulation by eccentric contraction [10]. -/- phosphorylation in response to stretch in γ-SG muscles We did not observe a strong Akt response to passive -/- was similar both for phosphorylation of T389, which cor- stretch, or any difference between C57 and γSG , im- relates with kinase activity, and for T421/S424, two of the plying that mTOR and/or p70S6K were being activated four phosphorylation sites in the autoinhibitory domain. through Akt-independent pathways. This is consistent Phosphorylation of T389 is mTOR-dependent, while with previous studies showing that Akt does not respond phosphorylation of the autoinhibitory domain is carried to mechanical stimulation in skeletal muscle, and that out by proline-directed kinases. Furthermore, it is thought p70S6K phosphorylation in response to stretch is inde- that phosphorylation of the autoinhibitory domain is ne- pendent of PI3K [62]. We also did not see increased cessary for phosphorylation of T389 [37]. The correlation phosphorylation of FAK in response to passive stretch in between phosphorylation of T421/S424 and T389 in our C2C12 cells or primary myotubes. Although integrins can isolated muscle model therefore suggests that phosphoryl- participate in mechanotransduction, it appears that our ation of the autoinhibitory domain was the rate-limiting cyclic passive stretch protocols did not cause integrin step for p70S6K activation, an intriguing prospect given activation. Further studies will be needed to elucidate that the autoinhibitory domain may be targeted by ERK1/ the details of crosstalk between SG-dependent and 2 [57]. Therefore, a future hypothesis to test is that differ- integrin-dependent signaling pathways, as well as the ential p70S6K activation is a downstream consequence role of calcium in these signaling cascades. -/- of differential ERK1/2 activation in γ-SG muscle. This Based on our findings, we position γSG as a mechano- 2+ would implicate Ca as an indirect modulator of p70S6K sensor, schematized in Figure 6, that is important for tran- -/- activity, since the increase in P-ERK1/2 in γ-SG muscle sient ERK1/2 activation during active contractions, as 2+ is dependent upon heightened Ca flux. It is worth noting well as modulation of p70S6K activation during passive that recent work by others has shown that stretch- stretch. Because passive stretch does not appear to in- induced activation of mTOR and p70S6K at T389 is inde- crease P-FAK or P-Akt, γSG is likely to regulate p70S6K pendent of ERK1/2 [54], which can regulate mTOR via through other pathways. These may include regulation tuberous sclerosis proteins 1 and 2 and Raptor [58-60]. of ERK1/2, which can promote p70S6K activation indir- However, this pathway is separate from the putative direct ectly via mTOR or directly by phosphorylation of the Moorwood et al. Skeletal Muscle 2014, 4:13 Page 11 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 [63,64] and overexpression of integrin α7 can improve the dystrophic phenotype through increased survival signaling via p70S6K [65], suggesting that p70S6K inhibition would not be advantageous. However, treatment of mdx mice with the mTOR inhibitor rapamycin improves the dys- trophic phenotype [66]. It is also interesting to note that p70S6K is inhibited by glucocorticoids, which are used in the treatment of DMD and LGMD [67]. Our results begin to provide mechanistic insight into how mechanical signaling is disrupted and altered in the absence of γ-SG. In addition to increasing our under- standing of the normal function of the SG complex, there is potential to provide more refined targets that could be beneficial to patients either in isolation or in combination with other therapeutic approaches. Conclusions We have identified p70S6K as part of a novel SG- Figure 6 Relevant signaling pathways and relationship to γ-SG. dependent mechanosensitive signaling pathway in skeletal Schematic of signaling pathways measured or discussed in this muscle. Our results suggest that γ-SG is required for the manuscript. Dotted lines indicate possible relationships to γ-SG. inactivation of p70S6K following its activation in response Arrowheads indicate an activating relationship, while blunt ends to mechanical stimulation. These studies provide new indicate a repressing relationship. Dashed line indicates priming, insights into the normal function of the SG complex, and rather than full activation. the mechanisms by which its deficiency in some forms of muscular dystrophy may contribute to pathology. autoinhibitory domain, and/or phosphatases such as Competing interests PP2A that dephosphorylate p70S6K. Loss of γSG un- The authors declare that they have no competing interests. couples the response to stretch, which may contribute to muscle pathology. Authors’ contributions CM carried out the calcium dependence experiments and the stretching The stability of the SG complex is directly affected in experiments in isolated muscles, and drafted the manuscript. AP carried out several LGMDs and in DMD, and a significant part of the the passive stretching experiments in primary myotube cultures. JS carried pathology in these diseases appears to be inappropriate out the passive stretching experiments in C2C12 myotubes, the immunoprecipitation of phosphorylated γ-SG, and the immunoblotting for load sensing. The first step in therapeutic development is rapamycin sensitivity experiments. BK participated in the passive stretching identifying and understanding the target, but little is cur- experiments in primary cultures. EJM developed the apparatus used for rently known about the role of the SG complex in load stretching of myotube cultures. ERB conceived of the study, participated in its design, coordination, and interpretation of the results, and helped draft sensing. Therefore, understanding the functions that are the manuscript. All authors read and approved the final manuscript. disrupted and the pathways that are involved in mechano- transduction involving the SG complex will help in the de- Acknowledgements sign of therapies for LGMDs and DMD. While restoration The authors thank Min Liu and Tian Zuozhen of the Wellstone Muscular Dystrophy Physiological Assessment Core for technical assistance with the of a completely normal SG complex either through gene rapamycin experiments. This work was supported by grants from NASA correction or protein replacement would also normalize (NNX09AH44G) and NIH (U54 AR052646) to ERB. CM was supported by the mechanical signal transduction, this may not be possible Paul D Wellstone Muscular Dystrophy Cooperative Research Center Training Grant (U54 AR052646). AP was supported by NNX09AH44G. JS is supported for all mutations responsible for DMD and LGMD. It is by the Pennsylvania Muscle Institute Training Grant (NRSA T32-AR053461). BK known that localization of the SG complex is not the sole was supported by a School of Dental Medicine Summer Research Fellowship. criterion for appropriate signaling [11]. Hence, other EJM was supported by NNX09AH44G and P50 DK52620. The funding bodies had no role in the study design, the collection, analysis, and interpretation of proteins may be necessary to correct signaling even when data, the writing of the manuscript, or the decision to submit the manuscript the complex is restored, and downstream pathways may for publication. emerge as more feasible therapeutic targets. We do not Author details know whether the enhanced basal and stretch-responsive Department of Anatomy and Cell Biology, School of Dental Medicine, -/- activation of p70S6K in γ-SG muscle contributes to University of Pennsylvania, Philadelphia, PA, USA. Pennsylvania Muscle pathology or compensates for it. Likewise, it is not clear Institute, University of Pennsylvania, Philadelphia, PA, USA. Current address: Department of Physiology, Medical School, National and Kapodistrian whether inhibition of p70S6K would have a beneficial or University of Athens, Goudi, Athens, Greece. Current address: Department of a detrimental effect in dystrophic muscle. Muscle specific Dermatology and Cutaneous Biology, Jefferson University College of gene targeting of mTORC1 components induces myopathy Medicine, Philadelphia, PA, USA. Moorwood et al. Skeletal Muscle 2014, 4:13 Page 12 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 Received: 25 March 2014 Accepted: 29 May 2014 24. Iwata Y, Katanosaka Y, Arai Y, Komamura K, Miyatake K, Shigekawa M: Anovel Published: 1 July 2014 mechanism of myocyte degeneration involving the Ca2 + -permeable growth factor–regulated channel. J Cell Biol 2003, 161:957–967. 25. Iwata Y, Katanosaka Y, Arai Y, Shigekawa M, Wakabayashi S: Dominant-negative References inhibition of Ca2+ influx via TRPV2 ameliorates muscular dystrophy in 1. Ervasti JM, Sonnemann KJ: Biology of the striated muscle dystrophin-glycoprotein animal models. Hum Mol Genet 2009, 18:824–834. complex. Int Rev Cytol 2008, 265:191–225. 26. Millay DP, Goonasekera SA, Sargent MA, Maillet M, Aronow BJ, Molkentin JD: 2. Emery AE, Muntoni F: Duchenne Muscular Dystrophy. Oxford: Oxford Calcium influx is sufficient to induce muscular dystrophy through a University Press; 2003. TRPC-dependent mechanism. Proc Natl Acad Sci U S A 2009, 106:19023–19028. 3. Engel AG, Franzini-Armstrong C: Myology: Basic and Clinical. 3rd edition. 27. Goonasekera SA, Davis J, Kwong JQ, Accornero F, Wei-LaPierre L, Sargent New York: McGraw-Hill, Inc; 2004. MA, Dirksen RT, Molkentin JD: Enhanced Ca2+ influx from STIM1–Orai1 4. Durbeej M, Campbell KP: Muscular dystrophies involving the dystrophin- induces muscle pathology in mouse models of muscular dystrophy. glycoprotein complex: an overview of current mouse models. Hum Mol Genet 2014. [Epub ahead of print]. Curr Opin Genet Dev 2002, 12:349–361. 28. Brinkmeier H: TRP channels in skeletal muscle: gene expression, function 5. Blake DJ, Weir A, Newey SE, Davies KE: Function and genetics of dystrophin and implications for disease. Adv Exp Med Biol 2011, 704:749–758. and dystrophin-related proteins in muscle. Physiol Rev 2002, 82:291–329. 29. Gailly P: TRP channels in normal and dystrophic skeletal muscle. 6. Davies KE, Nowak KJ: Molecular mechanisms of muscular dystrophies: Curr Opin Pharmacol 2012, 12:326–334. old and new players. Nat Rev 2006, 7:762–773. 30. Araishi K, Sasaoka T, Imamura M, Noguchi S, Hama H, Wakabayashi E, 7. Hack AA, Cordier L, Shoturma DI, Lam MY, Sweeney HL, McNally EM: Yoshida M, Hori T, Ozawa E: Loss of the sarcoglycan complex and Muscle degeneration without mechanical injury in sarcoglycan sarcospan leads to muscular dystrophy in β-sarcoglycan-deficient mice. deficiency. Proc Natl Acad Sci U S A 1999, 96:10723–10728. Hum Mol Genet 1999, 8:1589–1598. 8. Petrof BJ, Shrager JB, Stedman HH, Kelly AM, Sweeney HL: Dystrophin 31. Duclos F, Straub V, Moore SA, Venzke DP, Hrstka RF, Crosbie RH, Durbeej M, protects the sarcolemma from stresses developed during muscle Lebakken CS, Ettinger AJ, van der Meulen J, Holt KH, Lim LE, Sanes JR, contraction. Proc Natl Acad Sci 1993, 90:3710–3714. Davidson BL, Faulkner JA, Williamson R, Campbell KP: Progressive 9. Hack AA, Ly CT, Jiang F, Clendenin CJ, Sigrist KS, Wollmann RL, McNally EM: muscular dystrophy in α-sarcoglycan-deficient mice. JCellBiol 1998, Gamma-sarcoglycan deficiency leads to muscle membrane defects and 142:1461–1471. apoptosis independent of dystrophin. J Cell Biol 1998, 142:1279–1287. 32. Goonasekera SA, Lam CK, Millay DP, Sargent MA, Hajjar RJ, Kranias EG, 10. Barton ER: Impact of sarcoglycan complex on mechanical signal transduction Molkentin JD: Mitigation of muscular dystrophy in mice by SERCA in murine skeletal muscle. Am J Physiol Cell Physiol 2006, 290:C411–C419. overexpression in skeletal muscle. J Clin Invest 2011, 121:1044–1052. 11. Barton ER: Restoration of gamma-sarcoglycan localization and mechanical 33. Morine KJ, Sleeper MM, Barton ER, Sweeney HL: Overexpression of SERCA1a signal transduction are independent in murine skeletal muscle. J Biol Chem in the mdx diaphragm reduces susceptibility to contraction-induced 2010, 285:17263–17270. damage. Hum Gene Ther 2010, 21:1735–1739. 12. Benavides Damm T, Egli M: Calcium’s role in mechanotransduction during 34. Parsons SA, Millay DP, Sargent MA, Naya FJ, McNally EM, Sweeney HL, muscle development. Cell Physiol Biochem 2014, 33:249–272. Molkentin JD: Genetic disruption of calcineurin improves skeletal muscle 13. Burkholder TJ: Mechanotransduction in skeletal muscle. Front Biosci 2007, pathology and cardiac disease in a mouse model of limb-girdle muscular 12:174–191. dystrophy. J Biol Chem 2007, 282:10068–10078. 14. Andersson D, Meli A, Reiken S, Betzenhauser M, Umanskaya A, Shiomi T, 35. Fenton TR, Gout IT: Functions and regulation of the 70 kDa ribosomal D’Armiento J, Marks A: Leaky ryanodine receptors in beta-sarcoglycan S6 kinases. Int J Biochem Cell Biol 2011, 43:47–59. deficient mice: a potential common defect in muscular dystrophy. 36. Hornberger TA,Stuppard R, Conley KE,FedeleMJ, Fiorotto ML,Chin ER, EsserKA: Skeletal Muscle 2012, 2:9. Mechanical stimuli regulate rapamycin-sensitive signalling by a 15. Iwata Y, Katanosaka Y, Shijun Z, Kobayashi Y, Hanada H, Shigekawa M, phosphoinositide 3-kinase-, protein kinase B- and growth factor-independent Wakabayashi S: Protective effects of Ca2+ handling drugs against mechanism. Biochem J 2004, 380:795–804. abnormal Ca2+ homeostasis and cell damage in myopathic skeletal 37. Dufner A, Thomas G: Ribosomal S6 kinase signaling and the control of muscle cells. Biochem Pharmacol 2005, 70:740–751. translation. Exp Cell Res 1999, 253:100–109. 16. Nakamura TY, Iwata Y, Sampaolesi M, Hanada H, Saito N, Artman M, Coetzee WA, 38. Pearson RB, Dennis PB, Han JW, Williamson NA, Kozma SC, Wettenhall RE, Shigekawa M: Stretch-activated cation channels in skeletal muscle Thomas G: The principal target of rapamycin-induced p70s6k inactivation myotubes from sarcoglycan-deficient hamsters. Am J Physiol Cell Physiol is a novel phosphorylation site within a conserved hydrophobic domain. 2001, 281:C690–C699. EMBO J 1995, 14:5279–5287. 17. Sampaolesi M, Yoshida T, Iwata Y, Hanada H, Shigekawa M: Stretch-induced 39. Nielsen PJ, Manchester KL, Towbin H, Gordon J, Thomas G: The phosphorylation cell damage in sarcoglycan-deficient myotubes. Pflugers Arch 2001, of ribosomal protein S6 in rat tissues following cycloheximide injection, in 442:161–170. diabetes, and after denervation of diaphragm. A simple immunological 18. Solares-Perez A, Alvarez R, Crosbie RH, Vega-Moreno J, Medina-Monares J, determination of the extent of S6 phosphorylationonprotein blots. Estrada FJ, Ortega A, Coral-Vazquez R: Altered calcium pump and secondary JBiolChem 1982, 257:12316–12321. deficiency of gamma-sarcoglycan and microspan in sarcoplasmic reticulum 40. He Y, Macarak EJ, Korostoff JM, Howard PS: Compression and tension: membranes isolated from delta-sarcoglycan knockout mice. Cell Calcium differential effects on matrix accumulation by periodontal ligament 2010, 48:28–36. fibroblasts in vitro. Connect Tissue Res 2004, 45:28–39. 19. Solares-Perez A, Sanchez JA, Zentella-Dehesa A, Garcia MC, Coral-Vazquez 41. Barton ER, Park S, James JK, Makarewich CA, Philippou A, Eletto D, Lei H, RM: Intracellular Ca2+ transients in delta-sarcoglycan knockout mouse Brisson B, Ostrovsky O, Li Z, Argon Y: Deletion of muscle GRP94 impairs skeletal muscle. Biochim Biophys Acta 2010, 1800:373–379. both muscle and body growth by inhibiting local IGF production. 20. Hassoni C: Calcium homeostasis and ultrastructural studies in a patient FASEB J 2012, 26:3691–3702. with limb girdle muscular dystrophy type 2C. Neuropathol Appl Neurobiol 42. Moorwood C, Liu M, Tian Z, Barton ER: Isometric and eccentric force 1999, 25:244–253. generation assessment of skeletal muscles isolated from murine models 21. Fraysse B, Nagi SM, Boher B, Ragot H, Lainé J, Salmon A, Fiszman MY, Toussaint of muscular dystrophies. J Vis Exp 2013, 71:e50036. M, Fromes Y: Ca2+ overload and mitochondrial permeability transition pore 43. Hornberger TA, Mateja RD, Chin ER, Andrews JL, Esser KA: Aging does not activation in living δ-sarcoglycan-deficient cardiomyocytes. Am J Physiol Cell alter the mechanosensitivity of the p38, p70S6k, and JNK2 signaling Physiol 2010, 299:C706–C713. pathways in skeletal muscle. J Appl Physiol (1985) 2005, 98:1562–1566. 22. Lipskaia L, Pinet C, Fromes Y, Hatem S, Cantaloube I, Coulombe A, Lompré A-M: 44. Martineau LC, Gardiner PF: Insight into skeletal muscle Mutation of δ-sarcoglycan is associated with Ca2 + -dependent vascular mechanotransduction: MAPK activation is quantitatively related to remodeling in the Syrian hamster. Am J Pathol 2007, 171:162–171. tension. J Appl Physiol (1985) 2001, 91:693–702. 23. Kumar A, Khandelwal N, Malya R, Reid MB, Boriek AM: Loss of dystrophin causes aberrant mechanotransduction in skeletal muscle fibers. FASEB J 45. Glass DJ: Signalling pathways that mediate skeletal muscle hypertrophy 2004, 18:102–113. and atrophy. Nat Cell Biol 2003, 5:87–90. Moorwood et al. Skeletal Muscle 2014, 4:13 Page 13 of 13 http://www.skeletalmusclejournal.com/content/4/1/13 46. Hahn K, Miranda M, Francis VA, Vendrell J, Zorzano A, Teleman AA: PP2A 65. Boppart MD, Burkin DJ, Kaufman SJ: Activation of AKT signaling promotes regulatory subunit PP2A-B’ counteracts S6K phosphorylation. Cell Metab cell growth and survival in α7β1 integrin-mediated alleviation of muscular 2010, 11:438–444. dystrophy. Biochim Biophys Acta Mol Basis Dis 2011, 1812:439–446. 47. Peterson RT, Desai BN, Hardwick JS, Schreiber SL: Protein phosphatase 2A 66. Eghtesad S, Jhunjhunwala S, Little SR, Clemens PR: Effect of rapamycin on interacts with the 70-kDa S6 kinase and is activated by inhibition of immunity induced by vector-mediated dystrophin expression in mdx FKBP12-rapamycinassociated protein. Proc Natl Acad Sci U S A 1999, skeletal muscle. Sci Rep 2012, 2:399. 96:4438–4442. 67. Shah OJ, Kimball SR, Jefferson LS: Glucocorticoids abate p70S6k and eIF4E 48. Petritsch C, Beug H, Balmain A, Oft M: TGF-beta inhibits p70 S6 kinase via function in L6 skeletal myoblasts. Am J Physiol Endocrinol Metabol 2000, protein phosphatase 2A to induce G(1) arrest. Genes Dev 2000, 14:3093–3101. 279:E74–E82. 49. Cho DH, Choi YJ, Jo SA, Ryou J, Kim JY, Chung J, Jo I: Troglitazone acutely inhibits protein synthesis in endothelial cells via a novel mechanism doi:10.1186/2044-5040-4-13 involving protein phosphatase 2A-dependent p70 S6 kinase inhibition. Cite this article as: Moorwood et al.: Absence of γ-sarcoglycan alters the Am J Physiol Cell Physiol 2006, 291:C317–C326. response of p70S6 kinase to mechanical perturbation in murine skeletal muscle. Skeletal Muscle 2014 4:13. 50. Liu J, Stevens PD, Li X, Schmidt MD, Gao T: PHLPP-mediated dephosphorylation of S6K1 inhibits protein translation and cell growth. Mol Cell Biol 2011, 31:4917–4927. 51. Ketroussi F, Giuliani M, Bahri R, Azzarone B, Charpentier B, Durrbach A: Lymphocyte cell-cycle inhibition by HLA-G is mediated by phosphatase SHP-2 and acts on the mTOR pathway. PLoS One 2011, 6:e22776. 52. Zito CI, Qin H, Blenis J, Bennett AM: SHP-2 regulates cell growth by controlling the mTOR/S6 kinase 1 pathway. JBiol Chem 2007, 282:6946–6953. 53. Hornberger TA, Chu WK, Mak YW, Hsiung JW, Huang SA, Chien S: The role of phospholipase D and phosphatidic acid in the mechanical activation of mTOR signaling in skeletal muscle. Proc Natl Acad Sci U S A 2006, 103:4741–4746. 54. You JS, Frey JW, Hornberger TA: Mechanical stimulation induces mTOR signaling via an ERK-independent mechanism: implications for a direct activation of mTOR by phosphatidic acid. PLoS One 2012, 7:e47258. 55. You JS, Lincoln HC, Kim CR, Frey JW, Goodman CA, Zhong XP, Hornberger TA: The role of diacylglycerol kinase zeta and phosphatidic acid in the mechanical activation of mammalian target of rapamycin (mTOR) signaling and skeletal muscle hypertrophy. J Biol Chem 2014, 289:1551–1563. 56. Richard-Bulteau H, Serrurier B, Crassous B, Banzet S, Peinnequin A, Bigard X, Koulmann N: Recovery of skeletal muscle mass after extensive injury: positive effects of increased contractile activity. Am J Physiol Cell Physiol 2008, 294:C467–C476. 57. Mukhopadhyay NK, Price DJ, Kyriakis JM, Pelech S, Sanghera J, Avruch J: An array of insulin-activated, proline-directed serine/threonine protein kinases phosphorylate the p70 S6 kinase. JBiol Chem 1992, 267:3325–3335. 58. Carriere A, Romeo Y, Acosta-Jaquez HA, Moreau J, Bonneil E, Thibault P, Fingar DC, Roux PP: ERK1/2 Phosphorylate Raptor to Promote Ras-dependent Activation of mTOR Complex 1 (mTORC1). JBiol Chem 2011, 286:567–577. 59. Ma L, Chen Z, Erdjument-Bromage H, Tempst P, Pandolfi PP: Phosphorylation and Functional Inactivation of TSC2 by Erk: Implications for Tuberous Sclerosisand Cancer Pathogenesis. Cell 2005, 121:179–193. 60. Roux PP, Ballif BA, Anjum R, Gygi SP, Blenis J: Tumor-promoting phorbol esters and activated Ras inactivate the tuberous sclerosis tumor suppressor complex via p90 ribosomal S6 kinase. Proc Natl Acad Sci U S A 2004, 101:13489–13494. 61. Axelrod MJ, Gordon V, Mendez RE, Leimgruber SS, Conaway MR, Sharlow ER, Jameson MJ, Gioeli DG, Weber MJ: p70S6 kinase is a critical node that integrates HER-family and PI3 kinase signaling networks. Cell Signal 2014, 26:1627–1635. 62. Hornberger TA, Chien S: Mechanical stimuli and nutrients regulate rapamycin-sensitive signaling through distinct mechanisms in skeletal muscle. J Cell Biochem 2006, 97:1207–1216. 63. Bentzinger CF, Romanino K, Cloëtta D, Lin S, Mascarenhas JB, Oliveri F, Xia J, Submit your next manuscript to BioMed Central Casanova E, Costa CF, Brink M, Zorzato F, Hall MN, Ruegg MA: Skeletal and take full advantage of: muscle-specific ablation of raptor, but not of rictor, causes metabolic changes and results in muscle dystrophy. Cell Metab 2008, 8:411–424. • Convenient online submission 64. Risson V, Mazelin L, Roceri M, Sanchez H, Moncollin V, Corneloup C, • Thorough peer review Richard-Bulteau H, Vignaud A, Baas D, Defour A, Freyssenet D, Tanti JF, Le-Marchand-Brustel Y, Ferrier B, Conjard-Duplany Romanino K, Bauche S, • No space constraints or color figure charges Hantai D, Mueller M, Kozma SC, Thomas G, Ruegg MA, Ferry A, Pende M, • Immediate publication on acceptance Bigard X, Koulmann N, Schaeffer L, Gangloff YG: Muscle inactivation of mTOR causes metabolic and dystrophin defects leading to severe • Inclusion in PubMed, CAS, Scopus and Google Scholar myopathy. J Cell Biol 2009, 187:859–874. • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit

Journal

Skeletal MuscleSpringer Journals

Published: Jul 1, 2014

References