The hypertrophy of skeletal muscle in response to mechanical loading is among the most obvious examples of the powerful regulatory effects of contractile activity in this tissue (4,8,21). In recent years, several excellent studies have established an important role for Akt (also referred to as protein kinase B (PKB)) activation in the hypertrophic response of skeletal muscle (6,17,28,35). Activation or inhibition of a number of downstream targets of Akt, including mammalian target of rapamycin (mTOR), glycogen synthase kinase-3 (GSK3) and 70-kD S6 protein kinase (p70S6k), is observed in protocols that induce muscle hypertrophy (Fig. 1). Moreover, decreases in the activation of these pathways are associated with conditions that result in muscle atrophy (11,17).
Although muscle contraction has clearly been identified as a mechanism for Akt activation (33,34), the extent to which different mechanical stimuli activate Akt is not clear (24). Furthermore, there are a number of noncontractile mechanisms at work during muscle contraction that might influence Akt phosphorylation, such as membrane depolarization and calcium release and reuptake. The concept of a possible role for these processes in muscle adaptation to activity is supported by the observations that electrical activity regulates a number of processes in muscle cells (12,14). Indeed, recent work has demonstrated that at least one signaling factor, mitogen-activated protein kinase (MAPK) p38, responds to muscle activation independently of force production (13,31).
The present study examined the effects of several of the mechanical and neuromuscular factors associated with contractile activity on Akt phosphorylation in skeletal muscle. By manipulating the force-frequency relationship (FFR) and the length-tension relationship, as described previously (31), it was possible to assess the effect of a twofold change in stimulation frequency within a physiological range (15 vs 30 Hz) while maintaining the same active contractile force. In addition, the changes in muscle length induced differences in passive tension, allowing examination of that parameter. Finally, by delivering 30-Hz stimulation at optimal length, the effect of increased active tension was evaluated. Akt has also been shown to be insulin responsive, and it has been suggested that it plays an important role in the metabolic adaptation of skeletal muscle. We therefore determined muscle glycogen content to assess whether depletion of glycogen influenced any observed activation of Akt.
The initial hypothesis was that Akt phosphorylation would be related to active contractile tension. It was further predicted that greater activation frequency would result in greater Akt phosphorylation within protocols producing comparable contractile forces, as has been previously observed with p38 under similar experimental conditions. In addition, because it has recently been demonstrated that passive tension does not induce Akt phosphorylation (24), no effect of passive tension within protocols producing comparable contractile forces was predicted.
Animal use was approved by the University of Maryland institutional animal care and use committee. Male Sprague-Dawley rats (10-12 wk old, Charles River Laboratories, Wilmington, MA) were randomly distributed into one of four groups (N = 5 per group): 15-Hz stimulation at optimal length (15 Hzopt), 30-Hz stimulation at optimal length (30 Hzopt), 30-Hz stimulation at suboptimal length (30 Hzsub), and 30-Hz stimulation at supraoptimal length (30 Hzsupra). Additional animals were used in experiments to determine the effects of the surgical procedure, independent of stimulation (N = 4).
Surgical procedure and stimulation protocols.
Animals were anesthetized with intraperitoneal ketamine/xylazine (40:10 mg·kg−1 body mass, respectively). The surgical procedure for in situ stimulation was the same as that previously used (31). Briefly, the distal tendon of the tibialis anterior muscle (TA) was released and secured in a custom-made metal clamp attached to a load cell (FT03; Grass Instruments, Warwick, RI) mounted to a micromanipulator (Kite Manipulator; World Precision Instruments, Sarasota, FL) to allow adjustment of muscle length and proper alignment of the muscle. The tibia was stabilized with a 16-gauge needle, and the peroneal nerve was dissected free through a small incision and clamped with a subminiature electrode (Harvard Apparatus, Holliston, MA) that was used to stimulate the TA with supramaximal (120% of the stimulation intensity that produced a maximal twitch response at optimal length) stimulation pulses (200 μs), delivered by a Grass S48 stimulator in series with an isolation unit (model PSIU6; Grass Instruments). The signal from the load cell was amplified (model P122; Grass Instruments) and acquired using commercial acquisition software (PolyVIEW version 2.1; Grass Instruments). Throughout the experiments, the TA was protected from cooling by a heat lamp and from dehydration by mineral oil.
Once the animal was securely positioned for muscle stimulation, muscle length was adjusted to find the length that produced maximum twitch force (l 0). For the 15-Hzopt and 30-Hzopt groups, all subsequent testing took place at l 0. For the 30-Hzsub group, the muscle was stimulated at l 0 with a 330-ms, 15-Hz train, and the force was recorded. The muscle length was then shortened incrementally until a 330-ms, 30-Hz train produced the same force that the 15-train produced at l 0. The same procedure was followed for the 30-Hzsupra group, except that the muscle was incrementally lengthened rather than shortened (Fig. 2). Finally, it should be noted that although all of the protocols in the present study involved supramaximal stimulation intensities, the activation frequencies used in the present study produced submaximal contractions, because of the influence of the force-frequency relationship.
Once the experimental length was determined, the muscles were rested for 2 min, after which they underwent a modified Burke stimulation protocol (10), consisting of 2 min of stimulation at a rate of one train per second. All trains were 330 ms in duration, and the within-train frequency was 30 Hz for all groups, except the 15-Hzopt group, which received 15-Hz trains. After the stimulation, muscles were rapidly dissected free, blotted dry, weighed, divided, and snap-frozen in liquid nitrogen. The nonstimulated TA muscles were processed in the same manner to serve as controls. Once the muscles were removed, the still-anesthetized animals were euthanized with an overdose of anesthetic (100 mg·kg−1), followed by open aortic resection. To confirm that there was no effect of the surgical procedure on Akt phosphorylation, four animals were subjected to the muscle isolation procedure, attached to the force transducer, set at l 0 and held for the duration of the stimulation protocol, but not actually stimulated (sham). Muscles from this sham group were compared with the respective contralateral muscles, which were simply dissected and processed, as were the controls for the experimental groups.
Muscle samples were brought to 4°C, weighed, and maintained at 0-4°C throughout the protein extraction. Each portion was minced on cold glass and then homogenized with a motorized pellet pestle (Kontes, Vineland, NJ) in 10 mM NaPO4, pH 7.2, 2 mM EDTA, 10 mM NaN3, 120 mM NaCl, 1% NP-40, plus several protease inhibitors. Homogenates were incubated on ice for 1 h with occasional vortexing, then centrifuged at 14,000g for 30 min. Total protein concentrations were estimated (26), using a standard curve of twice-recrystallized bovine serum albumin (BSA). Samples were stored at −80°C.
SDS-PAGE and immunoblotting.
Muscle homogenates were diluted with 2× Laemmli sample buffer, boiled for 5 min, and loaded (40 μg of total protein per lane) on discontinuous (5% stacking, 10% separating) SDS gels for electrophoresis. Immunoblotting was performed (40), using PVDF membranes (Millipore Corp., Bedford, MA) with the gels transferred at 4°C. The membranes were dried overnight before use, then blocked for 2 h in 150 mM NaCl, 10 mM Tris (pH 7.5), saline buffer (TBS) containing 3% (wt/vol) bovine serum albumin, 5% (wt/vol) nonfat dry milk, and 50 mM NaF (13). After blocking, membranes were incubated overnight at 4°C with polyclonal antibodies (Cell Signaling Technologies) to the phosphorylated (Thr 308; #9275) and total (#9272) forms of Akt. Although phosphorylation of Akt occurs at both the Thr 308 and Ser 473 sites, the Thr 308 response has been shown to be more rapid in fast skeletal muscle, similar to the TA in the present study (34). Because of the short time course of the stimulation protocols used here, it was thought observing a detectable change would be more likely at Thr 308. Although activation of both sites is required for full activation of Akt, it does not seem that phosphorylation of one site is more critical to the hypertrophy response than the other (28). The blots were rinsed and incubated for 2 h at room temperature with an antirabbit IgG, HRP-conjugated secondary antibody (Sigma, St. Louis, MO). After subsequent rinsing, the blots were developed in TMB solution (KP Laboratories, Gaithersburg, MD), dried overnight, scanned densitometrically (23), and analyzed using Scion Image (Scion Corporation, Frederick, MD). The optical density of the phospho-Akt (pAkt) bands was normalized to the optical density of the pan-Akt bands for the corresponding sample, and the stimulated values were compared with the control values, as previously described (39). We also examined the phosphorylation of two downstream targets of Akt, p70S6K and the beta-subunit of glycogen synthase kinase-3 (GSK3β). In addition, we examined the phosphorylation of adenosine monophosphate kinase-activated protein kinase (AMPK), which is believed to be activated by metabolic changes rather than tension (3). Assays for the total and phosphorylated forms of the protein were performed in the same manner as for Akt, except that different specific antibodies were used (Cell Signaling Technologies, p70S6K: total, #9202; phosphorylated (Thr389), #9205; GSK3β: total, #9332; phosphorylated (Ser9), #9336, AMPKα: total, #2532; phosphorylated (Thr172), #2531). Primary antibody dilutions were made in 1% bovine serum albumin/TBS with 0.05% Tween (TBS-T) at 1:2,000. Secondary dilutions were at 1:10,000.
Portions of the stimulated and control muscles were lyophilized and homogenized, as done previously (31), and glycogen content was determined spectrophotometrically from the absorbance at 490 nm. The difference in glycogen content between the stimulated and control muscles was used as an index of metabolic demand. Although other metabolite assays might have been used, glycogen seemed an appropriate marker, because at least one study has shown that Akt phosphorylation is proportional to glycogen depletion during exercise (32), and other markers such as phosphocreatine or ATP may have recovered too rapidly.
Animal and muscle characteristics, muscle force and fatigue measurements, and glycogen depletion were analyzed for the main effect of the experimental protocol, using one-way analyses of variance (ANOVA). Where significant main effects were observed, post hoc comparisons were performed, using the Student-Newman-Keuls test. Differences in the control/stimulated ratios of the total and phosphorylated forms of the different signaling proteins were analyzed, using mixed-model, two-way, repeated-measures ANOVA. Where a significant main effect of phosphorylation state was noted, post hoc testing was conducted via paired t-tests. For a significant main effect of the experimental protocol or a significant protocol × phosphorylation interaction, post hoc comparisons were performed, using Dunnett's test for comparisons against the sham condition. Significance was established at α ≤ 0.05 for all analyses.
Animal and muscle characteristics.
No differences in body weight, muscle wet weight, or resting muscle glycogen were present across the experimental groups (Table 1).
Muscle force and fatigue.
As previously reported (31), the systematic alterations of muscle length and stimulation frequency achieved the comparable initial forces for the 15-Hzopt, 30-Hzsub, and 30-Hzsupra protocols, whereas the 30-Hzopt group had higher initial peak forces than the other three (Table 2). In addition to differences in active tension, there were clear differences in passive tension associated with the different protocols (resting tension, Table 2), with the 30-Hzsupra protocol exhibiting markedly higher passive tension than the others. Despite the high passive tension in the 30-Hzsupra protocol, the strain (~17%) was comparable with or less than those used in previous studies of passive stretch on MAPK (9,20,27).
Muscle fatigue was assessed as the decline in peak force during the 2-min stimulation protocols (Table 2). There was a significant effect of protocol on the decline in force (F = 5.9, P = 0.01). During the course of the fatigue protocols, the total FTI was greatest in the 30-Hzopt group, followed by the 15-Hzopt, and then the 30-Hzsupra and 30-Hzsub groups, which did not differ from each other in terms of FTI (Table 2).
The two-way repeated-measures ANOVA of the Akt-stimulated/control ratio of optical density revealed a significant effect of phosphorylation state (F = 16.40, P < 0.01) and significant protocol × phosphorylation state interaction (F = 3.82, P = 0.02). Post hoc testing found no differences between the total and phospho-Akt-stimulated/control ratios in muscles subjected to the sham, and the 30-Hzsub protocols exhibited no change in Akt phosphorylation relative to control. In contrast, the phospho-Akt ratio was significantly greater than the total ratio in muscles that received the 15-Hzopt, 30-Hzopt, and 30-Hzsupra protocols (all P < 0.05). Dunnett's test further revealed that the phospho-Akt ratio of the 30-Hzopt protocol, which produced the greatest FTI and the greatest initial peak force, was significantly greater than that of the sham condition (Fig. 3).
Downstream target activation.
Similar to what was observed for the Akt response, there were significant effects of phosphorylation state (F = 5.31, P = 0.03) and a significant protocol × phosphorylation state interaction (F = 3.74, P = 0.02) on the GSK3β-stimulated/control ratios. The phosphorylation responses were, however, not so robust as those observed for Akt, and only the 30-Hzopt protocol induced a significant change in phosphorylation (Fig. 4). Assays of p70S6K activation revealed no appreciable phosphorylation in any of the conditions (data not shown).
A significant main effect of phosphorylation state (F = 31.29, P < 0.01), but only a trend for an effect of protocol (F = 2.81, P = 0.06), and no protocol × phosphorylation state interaction, were found. All of the stimulation protocols induced AMPK phosphorylation, the degree of which was significant for the 15-Hzopt, 30-Hzsupra, and 30-Hzopt protocols (P = 0.02, 0.04, and 0.03, respectively) but not for the 30-Hzsub (P = 0.10) protocol (Fig. 5). No increase in phosphorylation was detected in the sham condition.
All of the protocols produced marked and significant declines in glycogen (all P < 0.02). There were, however, no significant differences in depletion across groups (Table 2).
The principal finding of the present study was that phosphorylation of Akt in skeletal muscle immediately after contraction was more dependent on active contractile force than passive tension. The protocol producing the greatest contractile tension produced greater changes in Akt phosphorylation than any of the other protocols. It is unlikely that glycogen depletion played any role in the observed differences in pAkt, because all four protocols reduced glycogen to a similar extent. In contrast to our hypothesis, stimulation train frequency was not related to observed differences in Akt phosphorylation, suggesting that neither the number nor the delivery rate of activation pulses influence Akt activation. Interestingly, it seems that increasing passive tension during contractile activation, within stimulation protocols producing comparable active tension, can positively affect the degree of Akt activation. This observation suggests a possible secondary role for passive tension, which may interact with active tension to activate the Akt signaling pathway.
The 30-Hzopt protocol produced the greatest initial forces, whereas the other three protocols produced comparable initial forces, as was the intent of the combined perturbations of stimulation frequency and muscle length. The observation that pAkt was greatest after the 30-Hzopt protocol suggests that Akt activation was more sensitive to active contractile tension than any of the other factors assessed. This is plausible because Akt phosphorylation is believed to play a key role in muscle hypertrophy, and increasing force of muscle contraction is known to increase hypertrophy (8,21). Interestingly, it seems that peak contractile force was a more important factor in Akt phosphorylation than was total FTI in the present study. The 15-Hzopt protocol produced a significantly greater total FTI than did the 30-Hzsupra protocol, but the level of pAkt was higher after the 30-Hzsupra protocol, although the difference did not achieve statistical significance. Phosphorylation of GSK3β was less robust than the Akt response, with only the 30-Hzopt protocol exhibiting a significant change relative to control. This protocol did, however, also produce the greatest degree of Akt phosphorylation. These data are consistent with previous findings in the literature that also demonstrate lesser degrees of GSK3β phosphorylation relative to Akt phosphorylation (34,37). Another downstream target of Akt, p70S6K, exhibited no observable changes in phosphorylation with any of the protocols. This result is most likely attributable to the fact that muscles were harvested and frozen immediately after the contractions. The p70S6K response is delayed relative to that of Akt, although it is more long lasting and typically requires several hours to observe (5,29).
Phosphorylation of AMPK was also increased after stimulation, but, in contrast to Akt and GSK3β, no significant protocol × phosphorylation state interaction was present. This result is consistent with previous work suggesting that AMPK is responsive to metabolic changes, including glycogen depletion (35), because all four stimulation protocols in the present study produced significant glycogen depletion. It is worth noting, however, that AMPK phosphorylation produced by the 30-Hzsub protocol did not achieve statistical significance (Fig. 5), despite significant reductions in glycogen (Table 2). AMPK activation has been associated with stimulation protocols designed to mimic endurance exercise, but not with protocols mimicking strength training exercise (3). The protocol in the present study, however, is perhaps most similar to sprint training, which has been shown to increase both muscle mass and fast-slow myosin heavy-chain transition (18), phenomena associated with activation of Akt and AMPK, respectively (3). Similar activation of both Akt and AMPK has been reported after functional overload, despite the fact that AMPK hyperphosphorylation has been related to a decrease in protein synthesis, which would be expected to blunt muscle hypertrophy (38).
The highest degree of Akt phosphorylation observed here (an approximately threefold change) is less than that reported in several other studies (24,33,34). However, it is worth noting that the bulk of these studies have used forced lengthening (eccentric) contraction protocols. Such contractions produce markedly greater forces than maximal isometric tetanic contractions. If tension is indeed the main stimulus for Akt activation, it would be no surprise that forced lengthening protocols induce greater levels of pAkt. Perhaps muscle fiber injury contributes to the greater Akt activation with forced lengthening contractions. Such contractions are often associated with muscle damage (24,25), whereas the protocol used in the present study is not (31).
Using a comparable experimental paradigm, it has been demonstrated that increasing stimulation frequency can increase phosphorylation of p38 independently of force production (31), but no such effect was observed for Akt in the present study. The three 30-Hz protocols produced markedly different levels of pAkt, and when the three protocols producing similar initial forces (15 Hzopt, 30 Hzsupra, and 30 Hzsub) were compared, it was clear that frequency was not a factor (Fig. 3). In fact, the 30-Hzsub protocol was no different from the sham protocol in the extent of Akt phosphorylation observed. If anything, passive tension (or total tension, because initial active forces were comparable) may have played a role in the differences in Akt activation among the 15-Hzopt, 30-Hzsupra, and 30-Hzsub protocols. The levels of pAkt observed among these three protocols increased as the passive tension associated with the protocol increased, suggesting a secondary role for passive tension, possibly mediated by stretch-activated channels in skeletal muscle (2,37). This interaction of passive and active tension may account for the large degree of Akt phosphorylation observed after forced lengthening contractions, as noted above. Consistent with the role of Akt activation in hypertrophy, forced lengthening contractions are frequently (15,19,29), although not always (1), shown to produce greater hypertrophy than other modes of exercise.
These results have potential clinical applications for strength training/rehabilitation protocols. Although speculative, the data suggest that hypertrophy/strengthening responses to submaximal contractions might be enhanced by superimposing such contractions (either voluntary or stimulated) on a stretched muscle. Because it has previously been shown that this sort of contraction protocol does not induce muscle membrane damage (31), in contrast to forced lengthening exercise (24,25), it would offer the distinct advantage of a reduced risk of injury while still stimulating hypertrophy-related signaling pathways. The lack of Akt phosphorylation observed when the muscle was activated at the shortened length in the 30-Hzsub protocol may also have clinical implications. It is well known that muscles immobilized in shortened positions are more prone to atrophy than those immobilized in lengthened positions (16). The present data suggest that stimulating shortened muscles to contract during periods of immobilization may be insufficient to stave off muscle atrophy.
Some investigators have argued that fatigue itself is a factor in inducing a strength training response, possibly because of the changes in pH, high-energy phosphates, or resting intramuscular calcium that typically accompany fatigue (2,30,36). The present study does not address this question, although the lack of differences in glycogen depletion across protocols would seem to rule out that factor. The results do suggest, however, that if fatigue contributes to hypertrophy, it is unlikely to do so via a pAkt-mediated mechanism. The degree of fatigue in the present study was clearly unrelated to Akt phosphorylation, because the 30-Hzsub protocol produced substantially greater fatigue than did the 30-Hzopt protocol, yet it showed no appreciable Akt phosphorylation, whereas the 30-Hzopt protocol showed a more than threefold change.
The results of the present study suggest that at least one factor involved in muscle contraction (stimulation frequency) does not contribute to the activation of a signaling pathway believed to be involved in the hypertrophic response of skeletal muscle, the Akt/PKB pathway. Instead, the data suggest that a combination of active and passive muscle tension are key factors, with active tension playing the greater role. These findings have some clinical implications for the design of training and rehabilitation protocols aimed at increasing muscle mass and strength. Further research is needed to clarify the contribution of mechanical stimuli to both muscle hypertrophy and to the activation of the various signaling pathways that may contribute to it.
The author wishes to thank Alaina Nutwell for her assistance in data collection. Further gratitude is due to Drs. Richard Lovering, Stylianos Scordilis, and Espen Spangenburg for critical reviews and helpful suggestions related to earlier drafts of this manuscript. Financial support for the project was provided through the Office of the Dean of the University of Maryland School of Medicine.
1. Adams GR, Cheng DC, Haddad F, Baldwin KM. Skeletal muscle hypertrophy in response to isometric, lengthening and shortening training bouts of equivalent duration. J Appl Physiol.
2. Allen DG. Skeletal muscle function: role of ionic changes in fatigue, damage and disease. J Clin Exp Pharmacol Physiol.
3. Atherton PJ, Babraj JA, Smith K, Singh J, Rennie MJ, Wackerhage H. Selective activation of AMPK-PGC-1 or PKB-TSC2-mTOR signaling can explain specific adaptive responses to endurance or resistance training-like electrical muscle stimulation. FASEB J.
4. Baar K. Training for endurance and strength: lessons from cell signaling. Med Sci Sports Exerc.
5. Baar K, Esser K. Phosphorylation of p70s6k correlates with increased skeletal muscle mass following resistance exercise. Am J Physiol (Cell Physiol.)
6. Bodine SC. mTOR signaling and the molecular adaptation to resistance exercise. Med Sci Sports Exerc.
7. Bodine SC, Stitt TN, Gonzalez M, et al. Akt/mTOR pathway is a crucial regulator of skeletal muscle hypertrophy and can prevent muscle atrophy in vivo. Nat Cell Biol.
8. Booth FW, Thomason DB. Molecular and cellular adaptation of muscle in response to exercise: perspectives of various models. Physiol Rev.
9. Boppart MD, Hirshman MF, Sakamoto K, Fielding RA, Goodyear LJ. Static stretch increases c-jun NH2-terminal kinase activity and p38 phosphorylation in rat skeletal muscle. Am J Physiol (Cell Physiol.)
10. Burke RE, Levine DN, Tsairis P, Zajac FE. Physiological types and histochemical profiles in motor units of the cat gastrocnemius. J Physiol (Lond.)
11. Childs TE, Spangenburg EE, Vyas DR, Booth FW. Temporal alterations in protein signaling cascades during recovery from muscle atrophy. Am J Physiol (Cell Physiol.)
12. De Deyne PG. Formation of sarcomeres in developing myotubes: role of mechanical stretch and contractile activation. Am J Physiol (Cell Physiol.)
13. Dentel JN, Blanchard SG, Ankrapp DP, McCabe LR, Wiseman RW. Inhibition of cross-bridge formation has no effect on contraction-associated phosphorylation of p38 MAPK in mouse skeletal muscle. Am J Physiol (Cell Physiol.)
14. Eftimie R, Brenner HR, Buonanno A. Myogenin and MyoD join a family of skeletal muscle genes regulated by electrical activity. Proc Natl Acad Sci USA
15. Farthing JP, Chilibeck PD. The effects of eccentric and concentric training at different velocities on muscle hypertrophy. Eur J Appl Physiol.
16. Fournier M, Roy RR, Perham H, Simard CP, Edgerton VR. Is limb immobilization a model of muscle disuse? Exp Neurol.
17. Glass DJ. Skeletal muscle hypertrophy and atrophy signaling pathways. Int J Biochem Cell Biol.
18. Harridge SDR, Bottinelli R, Canepari M, et al. Sprint training, in vitro and in vivo muscle function and myosin heavy chain expression. J Appl Physiol.
19. Higbie EJ, Cureton KJ, Warren GL, Prior BM. Effects of concentric and eccentric training on muscle strength, cross-sectional area, and neural activation. J Appl Physiol.
20. Hornberger TA, Stuppard R, Conley KE, et al. Mechanical stimuli regulate rapamycin-sensitive signalling by a phosphoinositide 3-kinase-, protein kinase B-and growth factor-independent mechanism. Biochem J.
21. Ishihara A, Roy RR, Ohira Y, Ibata Y, Edgerton VR. Hypertrophy of rat plantaris muscle fibers after voluntary running with increasing loads. J Appl Physiol.
22. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature
23. Liu Y, Mayer S, Opitz-Gress A, et al. Human skeletal muscle HSP70 response to training in highly trained rowers. J Appl Physiol.
24. Lockhart NC, Baar K, Mazzeo RS, Brooks SV. Activation of Akt as a potential mediator of adaptations that reduce muscle injury. Med Sci Sports Exerc.
25. Lovering RM., De Deyne PG. Contractile function, sarcolemma integrity and the loss of dystrophin after skeletal muscle eccentric contraction-induced injury. Am J Phyisol (Cell Physiol.)
26. Lowry OH, Rosenberg NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem.
27. Martineau LC, Gardiner PF. Insight into skeletal muscle mechanotransduction: MAPK activation is quantitatively related to tension. J Appl Physiol.
28. Nader GA. Molecular determinants of skeletal muscle mass: getting the "AKT" together. Int J Biochem Cell Biol.
29. Nader GA, Esser KA. Intracellular signaling specificity in skeletal muscle in response to different modes of exercise. J Appl Physiol.
30. Rooney KJ, Herbert RD, Balnave RJ. Fatigue contributes to the strength training stimulus. Med Sci Sports Exerc.
31. Russ DW, Lovering RM. Influence of activation frequency on cellular signalling pathways during fatiguing contractions in rat skeletal muscle. Exp Physiol.
32. Sakamoto K, Arnolds DEW, Ekberg I, Thorell A, Goodyear LJ. Exercise regulates Akt and glycogen synthase kinase-3 in human skeletal muscle. Biochem Biophys Res Comm.
33. Sakamoto K, Aschenbach WG, Hirshman MF, Goodyear LJ. Akt signaling in skeletal muscle: regulation by exercise and passive stretch. Am J Physiol (Endocrinol Metab.)
34. Sakamoto K, Hirshman MF, Aschenbach WG, Goodyear LJ. Contraction regulation of Akt in rat skeletal muscle. J Biol Chem.
35. Sakamoto K, Goodyear LJ. Invited review: intracellular signaling in contracting skeletal muscle. J Appl Physiol.
36. Schott J, McCully KK, Rutherford OM. The role of metabolites in strength training. II. Short versus long isometric contractions. Eur J Appl Physiol Occup Physiol.
37. Spangenburg EE, McBride TA. Inhibition of stretch-activated channels during eccentric muscle contraction attenuates p70S6K activation. J Appl Physiol.
38. Thompson DM, Gordon SE. Diminished overload-induced hypertrophy in aged fast-twitch skeletal muscle is associated with AMPK hyperphosphorylation. J Appl Physiol.
39. Thompson HS, Maynard EB, Morales ER, Scordilis SP. Exercise-induced HSP27, HSP70 and MAPK responses in human skeletal muscle. Acta Physiol Scand.
40. Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc Natl Acad Sci U S A