- Top of page
The mammalian target of rapamycin (mTOR) and AMP-activated protein kinase (AMPK) are important nutrient- and energy-sensing and signalling proteins in skeletal muscle. AMPK activation decreases muscle protein synthesis by inhibiting mTOR signalling to regulatory proteins associated with translation initiation and elongation. On the other hand, essential amino acids (leucine in particular) and insulin stimulate mTOR signalling and protein synthesis. We hypothesized that anabolic nutrients would be sensed by both AMPK and mTOR, resulting in an acute and potent stimulation of human skeletal muscle protein synthesis via enhanced translation initiation and elongation.
We measured muscle protein synthesis and mTOR-associated upstream and downstream signalling proteins in young male subjects (n= 14) using stable isotopic and immunoblotting techniques. Following a first muscle biopsy, subjects in the ‘Nutrition’ group ingested a leucine-enriched essential amino acid–carbohydrate mixture (EAC). Subjects in the Control group did not consume nutrients. A second biopsy was obtained 1 h later. Ingestion of EAC significantly increased muscle protein synthesis, modestly reduced AMPK phosphorylation, and increased Akt/PKB (protein kinase B) and mTOR phosphorylation (P < 0.05). mTOR signalling to its downstream effectors (S6 kinase 1 (S6K1) and 4E-binding protein 1 (4E-BP1) phosphorylation status) was also increased (P < 0.05). In addition, eukaryotic elongation factor 2 (eEF2) phosphorylation was significantly reduced (P < 0.05). Protein synthesis and cell signalling (phosphorylation status) was unchanged in the control group (P > 0.05).
We conclude that anabolic nutrients alter the phosphorylation status of both AMPK- and mTOR-associated signalling proteins in human muscle, in association with an increase in protein synthesis not only via enhanced translation initiation but also through signalling promoting translation elongation.
Mammalian cells contain highly conserved nutrient- and energy-sensing pathways which are vital to cell survival and growth. The AMP-activated protein kinase (AMPK) is a major cell energy-sensing protein which, when activated during energetic stress such as hypoxia, glucose starvation, or physical exercise, stimulates catabolic pathways and inhibits anabolic pathways in an effort to supply ATP for cell survival (see Hardie, 2005). We and others have recently shown that a major anabolic pathway, muscle protein synthesis, is inhibited when AMPK is activated during energetic stress (Bolster et al. 2002; Dreyer et al. 2006; Williamson et al. 2006a). This inhibition may be explained, in part, by the ability of AMPK to act as a negative upstream regulator of the mammalian target of rapamycin (mTOR) pathway (Inoki et al. 2003; Williamson et al. 2006b).
There has been a large amount of work in cells and rodents that has examined the effect of insulin and nutrients (primarily amino acids) in the regulation of protein synthesis (see Avruch et al. 2005, 2006; Kimball & Jefferson, 2006; Proud, 2006). Specifically, insulin stimulates protein synthesis by activating the insulin-signalling pathway leading to an increase in phosphatidylinositol 3-kinase (PI3K) and protein kinase B (Akt/PKB) activity. Akt phosphorylates and inhibits the tuberous sclerosis complex (TSC2), which increases mTOR kinase activity (Avruch et al. 2006; Proud, 2006). mTOR signalling to its downstream effectors ribosomal S6 kinase 1 (S6K1) and 4E-binding protein 1 (4E-BP1) is an important mechanism for stimulating translation initiation (Wang & Proud, 2006). Amino acids apparently do not influence TSC2 or Akt/PKB, and can be sensed by a novel class 3 PI3K within cells, which enhances phosphorylation and activation of mTOR by an unknown mechanism (Byfield et al. 2005; Nobukuni et al. 2005). However, recent work has also shown that S6K1 can phosphorylate eukaryotic elongation factor 2 (eEF2) kinase, leading to a reduced phosphorylation/activation of eEF2 and a stimulation of translation elongation (Wang et al. 2001).
Within the past few years, the work in cells and rodents has been translated into human studies. It has now been shown that a 2 h hyperinsulinaemic–euglycaemic clamp (Hillier et al. 2000; Greiwe et al. 2001) or a 6 h hyperinsulinaemic infusion of insulin (Liu et al. 2004) can increase the phosphorylation status of S6K1. In addition, a 2 h infusion of leucine (Greiwe et al. 2001) and a 6 h infusion of branched-chain amino acids (Liu et al. 2001) or a mixture of essential and non-essential amino acids increases the phosphorylation status of both S6K1 and 4E-BP1 (Liu et al. 2002). However, muscle protein synthesis was not directly measured in the studies mentioned above. More recently, an oral ingestion of 10 g of essential amino acids was found to increase both mTOR and S6K1 phosphorylation in association with an increase in muscle protein synthesis in human subjects (Cuthbertson et al. 2005).
We have recently shown in human subjects that an increase in insulin concentration to prandial levels can directly stimulate muscle protein synthesis as long as blood amino acid concentrations are maintained at the baseline level (Bell et al. 2005; Fujita et al. 2006). In addition, when both glucose and amino acids are provided, the muscle protein synthesis response is increased to a greater extent than with either insulin or amino acids alone (Rasmussen et al. 2000; Volpi et al. 2000; see Rasmussen & Phillips, 2003).
Therefore, we hypothesized that an increase in essential amino acid (leucine in particular) and glucose/insulin availability would be sensed by both AMPK and mTOR in human muscle, which would enhance Akt/mTOR signalling to key regulators of translation initiation and elongation, and induce a potent and rapid increase in the rate of muscle protein synthesis.
- Top of page
The primary and novel finding from our study is that an acute increase in nutrient and insulin availability stimulates muscle mTOR signalling to key regulators of not only translation initiation (S6K1 and 4E-BP1) but also translation elongation (eEF2), which potently and rapidly stimulates muscle protein synthesis (i.e. within 1 h). Although we also report a decrease in AMPKα phosphorylation in response to increased nutrient availability (which should enhance mTOR activation), it is readily apparent that the combination of enhanced essential amino acid (most likely due to leucine) and insulin availability increases muscle protein synthesis by enhancing phosphorylation of Akt/PKB, mTOR, 4E-BP1, and especially S6K1. Also of interest was the finding that eEF2 phosphorylation was significantly reduced, indicating that nutrients and insulin also promote translation elongation concurrently with translation initiation.
Our data support the notion that an increase in amino acid availability (leucine in particular) within the muscle, rather than insulin, is the major regulator of muscle protein synthesis. This is because although Akt/PKB phosphorylation was significantly increased following the increase in insulin availability, TSC2 phosphorylation was unchanged. Upstream regulation of mTOR by insulin via Akt/PKB phosphorylation, and inhibition of TSC1–TSC2 has been proposed as an important mechanism for the effect of insulin on increasing mTOR signalling. Insulin activation of mTOR also promotes the phosphorylation of S6K1, an important player in the regulation of translation initiation and cell growth (Baar & Esser, 1999). Since we did not see a significant increase in TSC2 phosphorylation at its Akt/PKB phosphorylation site (Thr 1462), and since the effect of insulin on the promotion of translation initiation appears to be mTOR dependent (i.e. Akt/PKB phosphorylation of TSC2 upstream of mTOR), our data suggest that insulin is playing a supportive and permissive role in the regulation of human muscle protein synthesis. However, we cannot rule out the possibility that Akt/PKB directly phosphorylated mTOR in our study, and future studies need to examine whether the anabolic effect of insulin on human muscle protein synthesis occurs independently of Akt/PKB signalling to TSC2. In addition to promoting translation initiation, insulin can also promote translation elongation by inhibiting eEF2 kinase activity (Redpath et al. 1996). More recently it has also been shown that S6K1 can phosphorylate and inactivate eEF2 kinase in order to promote elongation (Wang et al. 2001). Other phosphorylation sites on eEF2 kinase have also been found to be dependent on mTOR; however, it is not known which kinases, besides S6K1, are responsible for the mTOR-dependent inhibition of eEF2 kinase (Knebel et al. 2001; Browne & Proud, 2004). We also found that eEF2 phosphorylation was reduced when the muscle cells were exposed to essential amino acids, glucose, and insulin (indicating that elongation was also increased). Interestingly, an increase in IRS-1 serine phosphorylation by either mTOR or S6K1 is associated with an inhibition of insulin signalling and can serve as a negative feedback mechanism in muscle cells (Um et al. 2006). However, the modest increase in IRS-1 serine phosphorylation in the Nutrition group appears to have been insufficient for preventing Akt phosphorylation. Although glucose uptake decreased at 60 min following nutrient ingestion in association with the modest (P= 0.08) increase in IRS-1 serine phosphorylation, our data cannot directly determine whether the reduction in glucose uptake was due to an increase in IRS-1 serine phosphorylation.
On the other hand, our data strongly suggest that amino acid availability is the primary regulator of mTOR signalling and muscle protein synthesis in human skeletal muscle. The present findings are in agreement with our recent work showing that insulin is unable to stimulate muscle protein synthesis in human subjects when amino acid availability is reduced (Bell et al. 2005; Fujita et al. 2006). Furthermore, our data are also supported by previous work indicating that amino acid signalling to mTOR is not dependent on TSC2 (Byfield et al. 2005; Nobukuni et al. 2005). The phosphorylation of mTOR at Ser 2448, 4E-BP1 at Thr 37/46 and S6K1 at Thr 389 increased within 1 h following ingestion of essential amino acids and glucose. The largest increases in phosphorylation occurred with mTOR and S6K1, with the increase in S6K1 phosphorylation being extraordinarily large. Therefore, amino acid (probably leucine) stimulation of mTOR signalling to S6K1 and 4E-BP1 appears to be an important regulator of translation initiation and protein synthesis in human muscle. Elongation also appears to be regulated by amino acid stimulation of mTOR signalling, since we also report a very significant reduction in eEF2 phosphorylation at Thr 56. Although we did not directly measure eEF2 kinase activity or phosphorylation, it is likely that the inhibition of eEF2 activity, and thus decreased eEF2 phosphorylation, in our study occurred because of the large increase in S6K1 phosphorylation (and presumably a large increase in kinase activity). However, we cannot rule out that the decrease in eEF2 phosphorylation was, in part, attributable to the reduction in AMPK activity (Horman et al. 2002).
Our data also support the notion that the combination of amino acids, nutrient energy, and insulin enable a large (if not maximal) protein synthetic response (Anthony et al. 2002). Our previous work has shown that when glucose is added to an amino acid mixture, the stimulation of muscle protein synthesis is larger than with amino acids alone (see Rasmussen & Phillips, 2003). More recently, we have confirmed that insulin can indeed increase muscle protein synthesis in humans (Fujita et al. 2006; Rasmussen et al. 2006); however, the increase in muscle protein synthesis is much smaller than what is seen when amino acid availability is increased (Volpi et al. 2000). Furthermore, if amino acid availability is reduced, the ability of insulin to stimulate protein synthesis is inhibited in human skeletal muscle (Bell et al. 2005). This suggests that insulin is playing an important role in the regulation of human muscle protein synthesis, although amino acid availability is obviously a much more potent stimulator of protein synthesis.
The role of nutrient energy status within the muscle cell must also be considered, since the cellular processes of translation initiation and elongation are energetically expensive. A major cellular energy sensor within human muscle cells is AMPK (see Hardie, 2005). We and others have shown that muscle protein synthesis is inhibited in association with reduced mTOR signalling when AMPK activity and/or energy expenditure is increased (Bolster et al. 2002; Dreyer et al. 2006; Williamson et al. 2006a, 2006b). On the other hand, an increase in nutrient availability has also been shown to reduce AMPK activity (Kraegen et al. 2005), and an increase in ATP availability within cells appears to be sensed by mTOR (Dennis et al. 2001). Our data show that phosphorylation of the catalytic α subunit of AMPK is modestly reduced following the ingestion of essential amino acids and glucose. Therefore, the reduction in AMPK activity during nutrient ingestion may help to enhance protein synthesis by removing the inhibition of TSC2 on mTOR and/or by removing the negative regulation of eEF2. AMPK can phosphorylate TSC2 at Ser 1345 and Thr 1227, which inhibits mTOR activity (Inoki et al. 2003). However, we were unable to obtain the antibodies for the AMPK phosphorylation sites on TSC2 to confirm this. Therefore, future studies are needed to determine the role of both AMPK and insulin in the regulation of TSC2 in human muscle.
We also measured the overall rate of muscle protein synthesis via the direct incorporation of labelled phenylalanine into skeletal muscle over time. Our data show that the ingestion of a relatively small amount of amino acids enriched in leucine in combination with glucose causes a rapid and very large increase in muscle protein synthesis within 1 h. This increase in protein synthesis coincided with the increase in phosphorylation status of Akt/PKB, mTOR, 4E-BP1, and S6K1. In addition, we also report a decrease in the phosphorylation status of AMPK and eEF2. Therefore, it appears that the acute increase in muscle protein synthesis induced by the ingestion of nutrients and insulin was due to enhanced translation initiation and elongation resulting from the activation of mTOR signalling. Using compartmental modelling, we also determined the kinetics of phenylalanine during the experiment. As shown in Table 1, the net balance of phenylalanine across the leg became positive (an index of net muscle anabolism). Furthermore, the overall rates of amino acid transport into and out of the muscle tissue, intracellular amino acid availability, and utilization for muscle protein synthesis significantly increased. These amino acid and protein kinetics findings, interpreted in light of the concomitant cell signalling results, provide strong evidence of a prominent role of mTOR signalling in stimulating muscle protein synthesis in response to nutrients via enhanced translation initiation and elongation in human muscle. However, future work is still required (at both the basic and translational level) to determine the cellular mechanisms for how amino acids activate mTOR, how mTOR regulates different downstream targets, the role of muscle protein breakdown, and to determine how each component of the mTOR- and insulin-signalling pathways contribute to enhancing muscle protein synthesis.
We conclude that nutrients (essential amino acids and glucose) inhibit AMPK and activate mTOR signalling in human skeletal muscle in association with an increase in protein synthesis. The increase in protein synthesis appears to be due to not only enhanced translation initiation but also signalling promoting translation elongation. Future studies are required to determine if the provision of highly anabolic nutrients (such as those utilized in this experiment) may be useful in counteracting a variety of human muscle-wasting conditions such as those induced by cancer, ageing, AIDS, physical inactivity, muscular dystrophy, bed rest, and stroke.