16 12 2012
The Basic Biochemical Principles of Skeletal Muscle Hypertrophy (Part 1)
The growth of adult skeletal muscle is a complex process that requires nutrients, growth factors and exercise, with all three components being of equal relevance to achieve maximal success. In the following article we are going to have an EXTREMELY in-depth look at the physiological and biochemical processes involved. Because this knowledge is quite relevant to many other articles on this site, I’m going to attempt to write this article at two speeds, providing a fully referenced and in-depth look with the necessary images of one aspect in each section, but begin each section with a short intro and end that section with a brief cliff notes outline of the key points, no fluff involved. This way you will get a complete picture of the process, but for those who find this information too complex (or just not interesting enough), they can just remember the take-home message.
Note: I’ve decided to rename this article part 1, because halfway through it is fast turning into a veritable novel, and I still want to provide in-depth discussion of mTORC1’s role in autophagy, the Akt/GSKβ3 and Akt/Foxo pathways, the role and importance of β-catenin, prime catabolic pathways NFκβ, TNFα and myostatin and alternative signaling pathways of growth factors including PKC and MAPK pathways, which all together should provide fodder for another article of similar length before I manage to discuss all the processes involved in skeletal muscle hypertrophy.
mTORC1 – the key signal integrator for cellular hypertrophy
Cellular growth and proliferation are the primary processes uniquely governed by the mTOR complex 1 (mTORC1). Because we will be limiting ourselves to differentiated (mature) skeletal muscle cells, which cannot proliferate (multiply), it is safe to say the primary role of mTORC1 is to control regeneration and growth of the muscle. It does this by integrating a number of signals from exercise, energy availability, growth factors, nutrient abundance and other stressors. Each one of these factors is rate-limiting and thus required for the optimal, maximal functioning of mTORC1, and thus skeletal muscle hypertrophy (muscle growth).
In most cells, the mechanistic target of rapamycin (mTOR – previously also named mammalian target of rapamycin) is the core protein of two distinct complexes: mTORC1 and mTORC2. Both complexes contain mTOR and the GβL protein, but differ in regard to other proteins involved, primarily the catalytic subunits. mTORC1 contains raptor (regulatory associated protein of TOR) and mTORC2 contains rictor (rapamycin-insensitive companion of TOR). As the name suggests mTORC2 is not sensitive to inhibition by rapamycin, which may seem strange for a complex contain the TARGET of rapamycin, but one needs to remember that not only was the mTORC2 unkknown when mTOR was named, the name is still not entirely outdated now, because rapamycin binds mTOR regardless, but its mechanism of inhibition is decreasing its association with raptor, so while mTORC2 will still bind rapamycin, it’s simply not impacted by it, since it does not contain raptor.
mTORC2 plays distinct roles in the growth factor signaling pathway (discussed below) and its functions remain to be fully elucidated, since the complex is a relatively new discovery. Because mTORC2 is not essential to skeletal muscle hypertrophy(35), and for simplicity’s sake, we will not be discussing it at length in this article. The focus of the remainder of this section will be on mTORC1.
mTORC1, once activated, proceeds to initiate a complex but well orchestrated assembly of the eIF3 translation complex, explained in the box below. The eIF3 complex then becomes fully active at the ribosome, the place where mRNA transcripts from the DNA are translated to proteins. If you are unclear how this happens, clearer explanations can be found on Wikipedia and any basic biochemistry course book, but in short, DNA consists of a series of 4 different nucleotides that are divided in codons of 3. DNA is double-stranded and is divided by a transcription complex that reads out and copies one strand to a strand of mRNA (messenger RNA), and then zips up the DNA again, so the DNA remains intact. The mRNA copy is then moved to the ribosome where it is read by Initiation factor complexes like eIF3, and each codon forming a specific 3-nucleotide sequence is matched with one specific amino acid so that the ribosome forms a chain of amino acids. This chain is a primary protein structure. Because of the interactions with the amino acids amongst each other, as well as certain modifications, the protein proceeds to fold up to a secondary, tertiary and quaternary structure resulting in a 3D structure with a certain functionality. These proteins make up nearly everything in our body, from the structural actin and myosin that makes up our muscle, to hormones like insulin and IGF-1 (I used these examples because they are both most likely to make sense to a sports enthusiast) all the way across to literally everything discussed in this article (mTOR, Akt, PDK and all the acronyms that may or may not still sound like a foreign language to you at this point). So to make a long story short, activated mTOR will proceed to cause the assembly of the building blocks for growth. In muscle it has been shown to specifically stimulate the manufacture of, among other things, the myosin Heavy chain (myHC) protein needed to make more contractile units, and thus a stronger and bigger muscle.
Method of mTORC1 activation
mTORC1 once activated near the lysosome, will phosphorylate two proteins : S6 Kinase 1 (S6K1, phosphorylated on position Thr389) and 4E-binding Protein 1 (4E-BP1, phosphorylated on positions Thr37/46, Thr65 and Thr70). In the case of S6K1, this causes the dissociation between it and the actual eIF3 (eukaryotic initiation factor 3) protein. This free up both S6K1 and eIF3 for further interaction. eIF3 now binds to the remainder of mTORC1 (mTOR + raptor) and S6K1 is further stimulated by PDK1 (discussed in the section on growth factor pathways) which results in its phosphorylation (it is after all a Kinase) of both eukaryotic initiation factor 4B (eIF4B, phosphorylated on Ser422) and the 70kDA ribosomal protein S6 (70pS6, phosphorylated on 5 residues). pS6 is part of the small ribosomal subunit (40S) and permits its interaction with eIF4B, which is needed to recruit 40S to the eIF3 complex to begin translation initiation.
In the mean time, 4E-BP1 is dissociated from the eukaryotic initiation factor 4E (eIF4E) which permits it and eIF4G to bind the eIF3 complex. eIF4E is needed to unravel the mRNA so it can be read and translated by the complex. This whole process is called translation initiation and it is the major bottleneck in creating proteins. Many signals stimulate transcription (mRNA readouts of DNA) and an abundance of mRNA leads to increased translation, but never in a 1:1 ratio. Only a small fraction, or even none in some cases, of the mRNA yields actual protein. Only the convergence of anabolic signaling through signal integrators like mTORC1 can lead to a higher percentage of proteins being produced.
Cliff Notes: mTORC1 integrates anabolic signals like abundance of nutrients, energy and oxygen as well as growth factors (insulin, IGF-1) to initiate the manufacture of proteins needed to grow muscle. It is both sufficient and necessary for skeletal muscle hypertrophy. This section detailed very specifically how mTORC1 did this. All following sections will focus on the signals needed to activate mTORC1 and their necessity.
Growth factors – signals of whole body energy and nutrient sufficiency
Growth factors are extracellular signaling proteins that bind and activate a membrane bound receptor on the cell surface, which initiates a cascade of reactions inside the cell. The most important growth factors in skeletal muscle are insulin, which acts as a long range hormonal signal to signify whole-body energy abundance, and Insulin-like Growth Factor 1 (IGF-1) and its isoforms. Although IGF-1 is also present in the body as an endocrine hormone, like insulin, its systemic levels under physiological conditions are never sufficient to help stimulate skeletal muscle hypertrophy, but IGF-1 is also secreted as an autocrine/paracrine (secreted by the cell itself or a neighboring cell) factor in muscle tissue itself. Exercise also leads to the appearance of a splice variant of IGF-1, namely IGF-1c or Mechano-Growth Factor (MGF). Growth Factor signaling is primarily a signal of local and systemic energy and nutrient sufficiency or abundance, and primarily lifts the inhibition of mTORC1 signaling caused by nutrient or energy deprivation.
When Growth factors bind to their respective receptors, in this case the insulin receptor and the IGF-1 receptor (IR and IGF-1R), It causes a conformational change of that receptor which phosphorylates the insulin receptor substrates (IRS1 and IRS2), these in turn signal and activate phosphatidyl Inositol 3 Kinase (PI3K) which catalyzes the formation of PI(3,4,5)P3 (PIP3 for short) from phosphatidylinositol. PIP3 binds and recruits two other proteins to the inner cell membrane, namely Pyruvate dehydrogenase Kinase 1 (PDK1) and Akt (also known as Protein Kinase B or PKB). As you recall if you read more than the cliff notes from the previous paragraphs, PDK1 is also required for the final phosphorylation of S6K and the activity of the eIF3 complex (see box 1), so you begin to understand how Growth Factors are an integral part of mTORC1 signaling. PDK1 is also required to shut down key enzymes in the process of fatty acid beta-oxidation (fat burning) because it signifies an abundance of glucose as a primary substrate (and thus energy abundance) which will give you first insights into why it is nearly impossible to gain muscle and lose fat at the same time. The co-localization of PDK1 and Akt leads to the phosphorylation and activation of Akt (on position Ser473) by PDK1. Akt then works on various levels to support growth. These include the inhibition of protein degradation and apoptosis (cell death) by inhibiting Foxo and its downstream targets, both promoting transcription and increasing nutrient uptake by the cell by inhibiting GSK3β and relieving inhibitory signals of mTORC1 through its negative regulation of TSC2. The other functions are all very interesting and complex interactions that contribute to the overall growth of the cell, and may form the topic of a future article if so desired, but the focus of this article is mTORC1, so this section will primarily focus on that aspect.
The Tuberous Sclerosis complex (TSC) consists of two distinct proteins, tuberin (TSC1) and hamartin (TSC2). Together they form an active dimer whose primary purpose is in fact to limit mTORC1 signaling. It does this through GTPase activity. GTPases are small proteins that change their activity depending on whether they are bound to guanosine tri-phosphate (GTP) or guanidine di-phosphate (GDP). GTPase activity splits a phosphate from GTP to yield GDP and change the activity of the GTPase protein, while other factors can result in GTP loading by replacing the GDP with GTP. One such protein is the small GTPase Rheb (Ras homolog enriched in brain) which is an essential component of mTORC1 (1). However to yield active mTORC1 Rheb needs to be GTP loaded. Increased TSC2 activity will maintain Rheb in a GTP-bound and therefor inactive state (9). Phosphorylated Akt can in turn phosphorylate TSC2 on 2 to 5 residues, rendering it inactive (6,7,8). The effect does not seem to inhibit the GTPase activity, so likely it either prevents the formation of the active complex, or sequesters it from Rheb. Other factors like energy or oxygen deprivation will increase TSC2 activity, but those will be discussed later on, at length. Aside from TSC2, Akt also inhibits the actions of the 40kDA proline-rich Akt Substrate (PRAS40), a small protein that associates with and inhibits mTORC1 activity (10). It is proposed that growth factor signaling, through the PI3K/Akt pathway, limits the inhibitions of TSC2 and PRAS40 imposed by lack of nutrients, oxygen and energy, is not only permissive, but required for optimal mTORC1 signaling (11). Recent evidence suggests that both actions are especially crucial in relation to signaling to 4E-BP1 (43), and that PRAS40 may actually occupy the binding place of 4E-BP1.
In this story it is crucial to briefly sketch the importance of the two key growth factors. Insulin is the prime indicator of systemic nutrient abundance. As soon as a meal is ingested signals are already sent to increase insulin secretion from the pancreas. Insulin is an endocrine factor, only secreted from the pancreas, which travels the body to signal to peripheral tissues like adipose tissue and skeletal muscle that there is an influx of energy coming. This signal not only primes the cell for uptake of nutrients about to flood the circulation, it also switches off many of the negative signals imposed on growth, which is dependent on sufficient local energy. What is crucial to remember however is that physiological amounts of insulin (the fluctuation between fasting and the maximal amount that can be secreted by the pancreas by feeding) don’t have a dose-dependent signal, but rather function like an on/off switch, where a minimal amount of insulin is required for nutrients to be able to stimulate muscle protein synthesis (12), as well as inhibit muscle protein breakdown, but higher doses beyond that will not have any further effect (13). This is in sharp contrast with high supra-physiological doses of insulin, which we know to be very anabolic. IGF-1, despite a different receptor, works largely via the same mechanism. But while IGF-1 is systemically distributed, its systemic amounts never reach sufficiency to initiate muscle protein synthesis. However localized IGF-1, typically increased by contractile activity (exercise of any kind) or muscle damage, can reach much higher concentrations around the cell and neighboring cells to reach doses and activation far beyond that of systemic insulin and are therefore considered to be much more anabolic. Under the influence of contraction and muscle damage the muscle cell releases IGF-1a, similar to systemic IGF-1, as well as MGF (or IGF-1c, a truncated form of IGF-1 less susceptible to inhibiting actions of binding proteins) and IGF-2, a more potent activator of IGF-1R. The functional operations of IGF-1 on muscle can be quite complicated, and will no doubt form the focus of a future article, as it is beyond the scope of this one, but I did want to add that IGF-1 isoforms, while absolutely necessary for muscle growth, are not dose-dependently required for growth either. Much higher concentrations are reached in oxidative fibers and muscle injury than in glycolytic fibers, both conditions that actually limit growth. The localized contraction induced increase in growth factors however is a prime mechanism to maintain muscle size in periods of energy deprivation, as they can override the lack of nutrient sufficiency and maintain the anabolic machinery, preserving muscle protein when the bother would preferably catabolize it for energy.
Cliff Notes: The growth factors stimulate uptake of nutrients and signal energy sufficiency, lifting certain inhibitory factors on mTORC1 signaling, which allows nutrient translocation and stimulation of mTORC1. Only ample whole-body and localized cellular energy and resistance exercise can stimulate the growth factor release needed to allow mTORC1 action, even though they do not promote the activity of mTORC1 by themselves.
Nutrients – Building blocks, localization signals and activators or mTORC1
When we speak of nutrients in this context we primarily mean essential amino acids, and with a particular role for the branched chain amino acid leucine as a signaling molecule. Leucine seems to be essential for mTORC1 signaling, but ultimately it only increases protein synthesis when provided alongside a complete complement of the other amino acids. Here, at least in relation to mTORC1 activation, growth factors play a mere permissive role in releasing the inhibitions placed on signaling, leucine seems to be the key direct activator of mTORC1. In fact, in the absence of essential amino acids, growth factors are completely incapable of stimulating mTORC1 activity (17,18).
The full extent of how leucine signals mTORC1 related protein synthesis still remains to be fully elucidated, but a lot of headway has been made in the last couple of years, and more and more research is popping up left and right (over half the information here I literally only read in the last month), so expect this particular section to be updated in future months and years as more becomes clear about this complex and intriguing pathway. The process obviously begins with the uptake of leucine into the cell. Now we know leucine is a potent stimulator of insulin release in the pancreas (14), but even in cell cultures high content of leucine in the culture leads to increased leucine uptake. This is dependent on the membrane-bound receptor SLC7A5/SLC3A2 (I swear to you I don’t make these names up), which requires the expulsion of glutamine from the cell to increase uptake of leucine into the cell. This means the cell has to be glutamine loaded prior to the addition of essential amino acids for leucine to do its thing. Indeed, glutamine starvation prevents leucine uptake into the cell (15) and this denotes the importance of glutamine in the anabolic signaling cascade. This is likely why glutamine is the most abundant free amino acid in the body, and the largest storage of glutamine is skeletal muscle tissue (16). That is also likely why it is the most abundant dietary amino acid by far (I recently calculated that nearly 20% or a whopping 45g of the amino acids I get from dietary protein are glutamine), although, at least in muscle tissue, glutamine is considered a non-essential amino acid because muscle cells have the capacity to synthesize glutamine from leucine.
Once inside the cell nutrients have two distinct roles in activating mTORC1. The first, almost certainly to be exclusive to leucine, is the localization of mTORC1, normally present in the cytoplasm, to the lysosome(23), where its companion protein, the GTPase Rheb (discussed earlier) is located. In order to achieve this leucine stimulates the activation of another set of small GTPase proteins, named RagA through RagD. RagA and RagB, as well as RagC and RagD, share a striking similarity and interchangeability, so we will refer to them as RagA/B and RagC/D. These two types can form 4 different heterodimers with each other (AC, AD, BC, BD) (19,20)) and each of them are capable of initiating mTORC1 activation by translocation to the lysosome when activated (21,22). As previously explained with Rheb, GTPase proteins become active or inactive depending on whether they are GDP or GTP loaded. RagA/B is GDP-loaded and RagC/D GTP-loaded in the absence of amino acids, and thus inactive. In the presence of leucine the loading state of these GTPases is altered. These heterodimers are then capable of binding mTORC1 and together with the Ragulator complex, they will dock it to the lysosome where Rheb can be integrated. The Ragulator complex is a trimeric (three part) protein complex consisting of LAMTOR1 through 3 (up until very recently these were termed with the very non-descript names p14,p18 and MP1, but now renamed to LAMTOR, which stands for Lysosomal adaptor and MAPK and mTOR activator).
The exact mechanisms by which leucine activates the Rag heterodimers however remains to be fully explained, although two very interesting studies offer novel perspectives. One of them has another crucial role for glutamine in the leucine-dependent activation of RagB (24), namely the stimulation of glutaminolysis. Glutaminolysis is the two-step process of converting glutamine to α-ketoglutarate (αKG). First the enzyme glutaminase (GLS) deaminates glutamine into glutamate, and glutamate is then transformed to αKG by glutamate dehydrogenase (GDH). αKG has many uses in the cell ranging from being a substrate for the Krebs cycle to being a precursor for nucleotides and other amino acids. Now it seems it also has a function in loading at the very least RagB, and possibly other Rag GTPases. Leucine has been shown to allosterically promote the activation of GDH leading to amplified αKG output. The exact mechanism by which αKG loads RagB currently remains a mystery however.
The other study proposes (25) a role for leucyl-tRNA. tRNA’s sense their matching amino acid and at the ribosome promote the ligation to mRNA for the process of mRNA transcription to protein. In this case leucyl-tRNA, obviously a sensor for leucine, directly interacts with at least RagD and has been shown to be critical for mTORC1 signaling. These two studies together already demonstrate the possibility of the formation of an active RagB-RagD complex, sufficient to stimulate mTORC1 translocation to the lysosome.
Another study of note, perhaps, was the one by Zoncu et al. (26), demonstrating a more general amino acid signaling complex that senses lysosomal amino acid content and can signal mTORC1 through the Vacuolar H+-ATPase. To sketch a clearer picture, this ATPase is a transmembrane protein of the lysosome and signals internal lysosomal amino acid content to the mTORC1 complex on the outside of the lysosomal membrane. No Rag-loading has been established for this mechanism, and given the requirement for mTORC1 localization near the lysosome for this, its role in translocation is questionable. This may however be one way in which sufficiency of other amino acids is sensed, which is also a requirement for muscle protein synthesis, since leucine alone does not lead to increased MPS.
Nutrients also work in another way to amplify the signal of assembled mTORC1. A role for the class III PI3K human vacuolar protein sorting 34 (hVPS34) in amino acid sensing is probably the most long-standing theorem for amino acid mediated mTORC1 activation (28,29). It was initially proposed that it was activated through a Calmodulin dependent pathway (27), but future studies contradicted that (28), and sketched a role for co-activation by hVPS15 instead that was later confirmed in several studies. It also seemed highly illogical that a prime component of fast-to-slow fiber type switching (the change from a glycolytic, hypertrophy prone muscle cell to a more oxidative fiber that remains small and boosts endurance capacity) would be involved in a mechanism that was also important in hypertrophy. Although how exactly leucine activates hVPS34 will prove to be the topic of many a future study, three studies together (30,36,37) seem to at least provide a tentative picture of how hVPS34 works downstream towards mTORC1. The first (30) links hVPS to phosphatidic acid (PA) signaling, which had previously been shown to be an important signaling molecule in the mitogenic (growth factor) pathway of mTORC1 activation (31,32) as well as in the contraction-induced pathway of mTORC1 activation (33). PA can be created in a plethora of ways, some of which we will discuss later on when we discuss contraction mediated signaling, but the primary way is through the enzymes Phospholipase D1 and D2 (PLD1 and PLD2), who yield PA from the phospholipid phosphatidylcholine (PC). PA appears to be a direct activator of mTORC1 activity by binding mTOR itself on the FRB domain that is also the site of action for the inhibitor rapamycin, after which mTOR is named. The interaction is highly dependent on Rheb (40), although amino acids do not alter the GTP-loading state of Rheb (38,39). Although PLD2 is perfectly capable of stimulating mTORC1 activity, it seems that PLD1 is prime candidate for induction by mitogens (growth factors) and amino acids. The second study (36) was the first to note the involvement of the small GTPase RalA. At the time I didn’t really pay much attention to the study because of a number of incongruities it contained, but the third study (37) seemed to sort of tie it all together by pointing out that RalA is constitutively bound to PLD1, and that it was a necessary component downstream of hVPS34. Now the end result of the interaction of hVPS34 with phosphatidylinositiol is PI3P (as opposed to primarily PIP3 for normal PI3K). Both studies duly pointed out that PLD1 has a PX domain that can bind PI3P, and while the first (30)postulated this was a means of activation, the third (37) pointed out that RalA was a more likely activator since it can attracts and binds ARF6, another component previously linked to mTOR signaling (41), and known to activate PLD1 directly. The first however also noticed that PLD1 was translocated to the lysosome, where it could be targeted to mTORC1 and Rheb. Together, these studies sketch a picture where hVPS34 in concert with hVPS15 mediate the production of PI3P, which binds PLD1/RalA, targets it to the lysosome and activates Ral1, which in turn attracts and binds ARF6, which then activated PLD1 yielding PA production in close proximity to mTORC1. This interaction seems to confer insensitivity to (and thus seems to replace to some extent) growth factor signaling, demonstrating the ubiquitous role of PA in activating mTORC1. Of note is also that the hVPS34 pathways, unlike the Rag-Ragulator pathway which is only sensitive to leucine, is sensitive to nutrients in general, as it was activated by amino acids at large as well as glucose, but not insulin (37), which could perhaps tie in with the Zoncu study (26) demonstrating general amino acid signaling from inside the lysosome via an inside out complex involving H+-VATPase, but the exact way in which nutrients signal to hVPS34 will be fodder for many a future study.
Amino acids also promote mTORC1 activity by promoting the stability between raptor and mTOR (42) by modulation of the enzyme Inositol Polyphosphate Multikinase (IPMK). Although IPMK is well known as an inositol and lipid kinase, in similar fashion to PI3K and hVPS34, its kinase-activity does not appear to be required for this effect. Rather it seems to directly bind both raptor and mTOR, stabilizing the link of the complex. Blocking IPMK does apparently not block either translocation (via Rag-Ragulator) or activation (via hVPS34), but does reduce mTORC1 activity by up to 60%, which does not exclude the fact that it could work downstream of either cascade. Stabilized binding could be a function of phosphatidic acid stimulation or lysosomal amino acid signaling. In all likelihood a whole host of factors remain to be discovered and elucidated in the near future in regards to both pathways. For instance MAP3K4 is also frequently mentioned as being a possible effector of amino acid related mTORC1 signaling, but that is largely based on a link between MAP3K4 and RagA in Drosophilla melanogaster (fruitfly) (34), but the same research group failed to establish a similar interaction between MAP3K4 and RagA-C in human cells, so it would be unwise as of yet to include it in the list of known regulators of the amino acid – mTORC1 interaction.
Cliff Notes: Leucine, in the presence of a full complement of amino acids, enters the glutamine-loaded muscle cell where it signals to the Rag GTPases, who translocate the mTORC1 to the lysosome where its key co-activator Rheb is located, and the target of the growth factor pathway. Secondly nutrient abundance in general (amino acids and glucose, signals translocation of PLD1 to the lysosome as well, and activates it, resulting in a localized increase in Phosphatidic acid which binds mTOR as a ligand and activates it, possibly by activating Rheb in similar fashion to growth factors. Many factors of the amino acid sensing mechanism, its impact on, and method of mTORC1 signaling still remain to be fully explained, but it is a hot topic in research currently.
Contraction – Amplification of activation and Initiation of local growth factor signaling
Next to growth factors, the third pillar of mTORC1 mediated skeletal muscle hypertrophy is resistance exercise. Relatively little is known about how exercise signals growth. It seemingly increases growth factors, but can also signal to mTORC1 via a phosphatidic mediated pathway in the timeframe that preceeds the appearance of local growth factors (IGF-1,IGF-2, MGF) (33). One interesting possibility proposed by the authors is the increased availability of PLD1, since PLD1 is bound in an inactive state to α-actinin, the main component of the Z-disk, which makes up the demarcation line between sequential sarcomeres (contractile units in series and parallel), and the Z-disk is extremely sensitive to contractile perturbations. A similar proposition exists for STARS-mediated Serum Response Factor (SRF) activation, which can also play various roles in hypertrophy through signaling to downstream factors. STARS is also bound to α-actinine. I could, and may expand this section in the near future (although I think I will focus on part 2 first) with various tidbits we do know about the relation between contraction and mTORC1 signaling, but more important in grasping the whole of this, is the role of resistance exercise.
The model that Phillips et al. (44) propose sketches it best. Feeding induces protein synthesis, while the period in between feedings leads to proteolysis (loss of protein). Each feeding compensates the protein loss in the periods between feedings, however there is a limit to the stimulation that can be attained by nutrients and systemic growth factors alone, which should be obvious to anyone, since no one ever grew muscle sitting in his couch and eating protein. Resistance exercise seems to increase muscle protein synthesis in the same manner, via mTORC1, but through distinct and separate pathways that show a high degree of synergism with feeding, and especially protein ingestion. Because unlike the transient nature of nutrient related signaling, the increase in muscle protein synthesis imparted by resistance exercise can last up to 48h (45), it both increases the amount of protein synthesis during times of feeding and increases the amount of protein lost during fasting to some extent for the entire timespan, leading to a drastic increase in protein accumulation, and thus skeletal muscle hypertrophy.
AMPK – Cellular energy sensor and negative regulator of muscle hypertrophy
Next to mTORC1 there is another key player in muscle metabolism, namely the AMP-activated protein Kinase (AMPK). The most rudimentary description of AMPK is that of a cellular energy sensor that responds to the AMP:ATP ratio. When energy, in the form of ATP, is used, it decreases the ATP level and increases the AMP level of a cell, and conversely increases the AMP:ATP ratio.
Adenosine tri-phosphate or ATP is the primary cellular energy currency. It consists of an adenosine nucleotide backbone with a string of three phosphates attached (as the name would suggest). Energy in the cell is released by splitting the last high energy phosphate bond, yielding adenosine di-phosphate (ADP) and a phosphate molecule. When ADP levels in the cell rise they stimulate the conversion of ADP to ATP + AMP (adenosine monophosphate) via adenylate kinase (also called myokinase) by transferring one phosphate from one ADP to another. Alternatively high muscle creatine levels lead to the creation of creatine phosphate, which is also capable of donating a phosphate to ADP to create ATP. For more information click here.
AMPK sets in motion a mechanism that promotes energy conservation by way of promoting more economic fuel use, namely fatty acid oxidation, in liver and muscle, while releasing fatty acids from the fat tissue to fuel this process. The two distinct ways in which it does this in skeletal muscle tissue is through mitochondrial biogenesis (the creation of more mitochondria, the cellular energy plants that produce ATP through the Krebs cycle) and by limiting energy expenditure by non-essential systems. Non-essential systems in this case would be skeletal muscle hypertrophy. In an energy deprived state AMPK is there to prevent that energy is wasted to increase muscle size while it is needed for cellular survival. AMPK is capable of increasing TSC2/hamartin (see growth factor pathway above) phosphorylation (2) which allows it to form a complex with TSC1/tuberin and thereby bind and sequester the protein Rheb (Ras Homolog enriched in brain), a necessary component of the mTOR complex 1 (1). It also directly phosphorylates mTOR (2) itself, and raptor (3). In this fashion a high cellular AMPK level leads to inhibition of protein synthesis and skeletal muscle hypertrophy.
AMPK is composed of one catalytic α-subunit and 2 regulatory β and γ-subunits, of which there are several isoforms. The α-subunit has two known isoforms, the α1 and the α2. Research into how the expression of the two isoforms is regulated is scarce since it’s a relatively new discovery, but a clearer picture is beginning to form. It seems it is predominantly the α1 isoform that limits muscle growth, while the α2 isoform is responsible for mitochondrial biogenesis. The picture I’m about to sketch still has to be confirmed through more studies, but it appears that the α2 isoform is a more local sensor that is more acutely affected by the cell’s own energy status, while the α1 isoform is affected more by whole body energy stress and signals that indicate the inability to generate energy. As such the two are usually expressed at the same time, but their ratio will differ. Deletion of AMPK α1 indeed leads to an increase in muscle protein synthesis, while a deletion in AMPKα2 seems to slightly reduce it. This is in the first place because the two isoforms tend to compensate for one another. When upstream promoters of AMPKα2 are eliminated, AMPKα1 increases (4), which makes sense since an inability to induce mitochondrial biogenesis would lead to a reduced capacity to generate energy. Whole body energy depletion would have a similar effect, since there is no fuel at hand, so mitochondrial biogenesis would have no point, and energy as a whole is best conserved. Exercise on the other hand seems to induce a higher α2:α1 ratio because the exercise is reducing local ATP but whole body energy can sustain activity if the cell is to switch to a more economic fuel source, namely fat. This is likely, although I’m going out on a limb here, why resistance exercise promotes hypertrophy. Like any exercise it promotes local energy depletion, in fact we know the metabolic stress related to the time under tension is a crucial factor in signaling hypertrophy, but because high-intensity, unlike low-intensity, exercise induces a state of transient hypoxia (oxygen shortage) through muscle swelling and venous blood flow restriction, that would make it hard to utilize fat as a fuel source (since that requires oxygen for oxidation) and signal an increase in muscle size in order to store more glycogen, the only fuel source it is capable of using to sustain force output in the absence of oxygen.
Method of AMPK activation
As an addendum for the science-geeks among you, a brief overview of how AMPK is activated. AMPK is primarily activated through phosphorylation by upstream kinases at the conserved Thr172 position, causing a 20-fold increase in activity. An increase in activity of those kinases (LKB1 and CaMKK being the main ones) will therefor lead to AMPK activation but to varying degrees. The γ-subunit of AMPK can bind nucleotides like adenosine. In an energy-deficient state that means it will be binding the abundant AMP which increases the susceptibility to kinases and decreases the susceptibility to phosphatases (which would dephosphorylate Thr172) causing as much as a 1000-fold overall increase in activity. However when energy supply is sufficient, the γ-subunit will bind mostly ATP, which eliminates phosphatase inhibition and suppresses the activity of AMPK.
The relationship between AMPK and mTORC1 isn’t unilateral however. Although AMPK’s chokehold on mTORC1 is stronger than the other way around, increased mTORC1 activity also has an inhibitory action on AMPK (5), and especially the α2 isoform, because it is trying to hog the nutrients for muscle growth instead of mitochondrial biogenesis. If you recall from the discussion on the growth factor pathways, TSC2 inhibition is a very potent inducer of skeletal muscle growth, and as you can probably deduce, so is inhibition of potent activators of TSC2 like AMPK. As a result we will want to permanently keep AMPKα1 on the low side by way of sufficient whole body energy supply through a hypercaloric diet (read: eating more than maintenance) and frequent feeding periods in order to sustain and support exercise-induced mTORC1 activation. We will actually want to induce AMPKα2 with resistance exercise for brief periods of time, preferably with high intensity exercise that increases local hypoxia in an equally brief and transient way, but we’ll also want to avoid AMPKα2 staying high, something we achieve through pre- and postworkout feeding which will deliver nutrients to the muscle during and after the workout to supply amino acids for muscle protein synthesis and glucose for glycogen and ATP replenishment. This transient nature of brief, hypoxic induction of α2 will actually be a key signal in inducing hypertrophy, but keep in mind that prolonged activation in a hypoxic environment will cause a shift toward AMPKα1, and that prolonged activation in a normoxic environment will cause a shift toward mitochondrial biogenesis rather than increase contractile protein synthesis (and over sufficient time a muscle fiber-type shift from glycolytic to oxidative – keeping in mind that oxidative fibers are purposely kept small because of their unusually high protein turnover and energy use). However, since strategies to increase mTORC1 will already blunt AMPKα2 response to some degree this is more or less a self-evident factor.
Cliff Notes: AMPK is the primary signal of a lack of energy. It has two isoforms named after its catalytic subunits, α1 and α2, that are usually co-expressed, but the relative amount of each is dependent on the type of energy stress. When α2 is more abundant it is typically a sign of localized cellular energy stress that will initiate changes to commence using systemic energy, like fatty acids. This isoform is primarily responsible for initiating changes that promote endurance performance under lack of local energy, but abundance of oxygen and sufficiency of systemic energy. When supply of systemic oxygen or energy is impaired however, the α1 isoform becomes abundant, and its only purpose is to limit energy waste (and thus muscle growth) by inhibiting mTORC1. AMPK as a whole is preferably kept low throughout the day and especially prior to and immediately after a workout. However when we workout AMPK will increase under energy stress, and as the workout progresses a shift will occur in the ratio of α2:α1 ratio in favor of α1 due to an initial localized energy stress and an eventual deprivation of localized oxygen (anaerobic exercise) . This is however the very signal that will incite our muscle to grow in order to sustain longer and heavier workouts by not only increasing contractile force, but higher capacity to store anaerobic energy (glycogen)
REDD1/2 – Oxygen stress sensor
Although hypoxia (lack of oxygen) is primarily signaled by LKB1/AMPK pathways directly to mTORC1 (46), in some cell types this signaling is dependent on the induction of the protein REDD1, induced by the protein HIF-1α (Hypoxia inducible factor 1 alpha). REDD1 seems to directly act on TSC2 upstream of mTORC1, and bind to the inhibitory protein “14-3-3” that negatively regulates TSC2. Binding of REDD1 to 14-3-3 reduces its binding to TSC2, which enhances the negative effect of TSC2 on mTORC1 (47). Skeletal muscle appears to fall under the latter category, which offers a dissociation between AMPK signaling at large, and signaling by hypoxia specifically. This allows AMPK to be induced and regulate a switch from local energy stores (glycogen) to systemic energy stores (fatty acids) for energy provision, with an additional signal to signal a state of hypoxia that would prevent use of systemic energy, and thus signal prevention of increases in protein synthesis. REDD1 and 2 signaling can also occur in other circumstances, such as prolonged inactivity (48). What is particularly interesting is that REDD proteins work exclusively upstream of mTORC1, but in a manner that allows it to control the downstream effects of it. REDD proteins do not interfere with the effect of nutrients on mTORC1 phosphorylation, but it does affect PDK1 signaling, which results in faulty S6K1 signaling and thus reduced protein synthesis.
Cliff Notes: This short paragraph was merely to introduce the HIF1α/REDD1 signaling pathway as a distinct pathway signaling reductions in muscle protein synthesis in response to oxygen deprivation of the muscle and stress the importance of oxygen provision in the long term in obtaining and sustaining optimal muscle size.
- Inoki K, Guan KLTuberous sclerosis complex, implication from a rare genetic disease to common cancer treatment. Hum Mol Genet. 2009 Apr 15;18(R1):R94-100.
- Cheng PY, Lee YM, Law KK, Lin CW, Yen MH. The involvement of AMP-activated protein kinases in the anti-inflammatory effect of nicotine in vivo and in vitro. Biochem Pharmacol. 2007 Dec 15;74(12):1758-65.
- Gwinn DM, Shackelford DB, Egan DF, Mihaylova MM, Mery A, Vasquez DS, Turk BE, Shaw RJ. AMPK phosphorylation of raptor mediates a metabolic checkpoint. Mol Cell. 2008 Apr 25;30(2):214-26
- McGee SL, Mustard KJ, Hardie DG, Baar K.Normal hypertrophy accompanied by phosphoryation and activation of AMP-activated protein kinase alpha1 following overload in LKB1 knockout mice.J Physiol. 2008 Mar 15;586(6):1731-41.
- Mounier R, Lantier L, Leclerc J, Sotiropoulos A, Foretz M, Viollet B.Antagonistic control of muscle cell size by AMPK and mTORC1.Cell Cycle. 2011 Aug 15;10(16):2640-6.
- Inoki K, Li Y, Zhu T, Wu J, Guan KL.TSC2 is phosphorylated and inhibited by Akt and suppresses mTOR signalling.Nat Cell Biol. 2002 Sep;4(9):648-57.
- Manning BD, Tee AR, Logsdon MN, Blenis J, Cantley LC.Identification of the tuberous sclerosis complex-2 tumor suppressor gene product tuberin as a target of the phosphoinositide 3-kinase/akt pathway.Mol Cell. 2002 Jul;10(1):151-62.
- Dennis PB, Jaeschke A, Saitoh M, Fowler B, Kozma SC, Thomas G.Mammalian TOR: a homeostatic ATP sensor.Science. 2001 Nov 2;294(5544):1102-5.
- Tee AR, Fingar DC, Manning BD, Kwiatkowski DJ, Cantley LC, Blenis J.Tuberous sclerosis complex-1 and -2 gene products function together to inhibit mammalian target of rapamycin (mTOR)-mediated downstream signaling.Proc Natl Acad Sci U S A. 2002 Oct 15;99(21):13571-6.
- Vander Haar E, Lee SI, Bandhakavi S, Griffin TJ, Kim DH.Insulin signalling to mTOR mediated by the Akt/PKB substrate PRAS40.Nat Cell Biol. 2007 Mar;9(3):316-23.
- Sancak Y, Thoreen CC, Peterson TR, Lindquist RA, Kang SA, Spooner E, Carr SA, Sabatini DM.PRAS40 is an insulin-regulated inhibitor of the mTORC1 protein kinase.Mol Cell. 2007 Mar 23;25(6):903-15.
- Greenhaff PL, Karagounis LG, Peirce N, Simpson EJ, Hazell M, Layfield R, Wackerhage H, Smith K, Atherton P, Selby A, Rennie MJ.Disassociation between the effects of amino acids and insulin on signaling, ubiquitin ligases, and protein turnover in human muscle.Am J Physiol Endocrinol Metab. 2008 Sep;295(3):E595-604.
- Staples AW, Burd NA, West DW, Currie KD, Atherton PJ, Moore DR, Rennie MJ, Macdonald MJ, Baker SK, Phillips SM.Carbohydrate does not augment exercise-induced protein accretion versus protein alone.Med Sci Sports Exerc. 2011 Jul;43(7):1154-61.
- Yang J, Dolinger M, Ritaccio G, Mazurkiewicz J, Conti D, Zhu X, Huang Y.Leucine stimulates insulin secretion via down-regulation of surface expression of adrenergic α2A receptor through the mTOR (mammalian target of rapamycin) pathway: implication in new-onset diabetes in renal transplantation.J Biol Chem. 2012 Jul 13;287(29):24795-806.
- Nicklin P, Bergman P, Zhang B, Triantafellow E, Wang H, Nyfeler B, Yang H, Hild M, Kung C, Wilson C, Myer VE, MacKeigan JP, Porter JA, Wang YK, Cantley LC, Finan PM, Murphy LO.Bidirectional transport of amino acids regulates mTOR and autophagy.Cell. 2009 Feb 6;136(3):521-34.
- Mittendorfer B, Volpi E, Wolfe RR.Whole body and skeletal muscle glutamine metabolism in healthy subjects.Am J Physiol Endocrinol Metab. 2001 Feb;280(2):E323-33.
- Arsham AM, Neufeld TP.Thinking globally and acting locally with TOR.Curr Opin Cell Biol. 2006 Dec;18(6):589-97.
- Jacinto E, Lorberg A.TOR regulation of AGC kinases in yeast and mammals.Biochem J. 2008 Feb 15;410(1):19-37.
- Gong R, Li L, Liu Y, Wang P, Yang H, Wang L, Cheng J, Guan KL, Xu Y.Crystal structure of the Gtr1p-Gtr2p complex reveals new insights into the amino acid-induced TORC1 activation.Genes Dev. 2011 Aug 15;25(16):1668-73.
- Sekiguchi T, Hirose E, Nakashima N, Ii M, Nishimoto T.Novel G proteins, Rag C and Rag D, interact with GTP-binding proteins, Rag A and Rag B.J Biol Chem. 2001 Mar 9;276(10):7246-57.
- Dubouloz F, Deloche O, Wanke V, Cameroni E, De Virgilio C.The TOR and EGO protein complexes orchestrate microautophagy in yeast.Mol Cell. 2005 Jul 1;19(1):15-26.
- Sancak Y, Peterson TR, Shaul YD, Lindquist RA, Thoreen CC, Bar-Peled L, Sabatini DM.The Rag GTPases bind raptor and mediate amino acid signaling to mTORC1.Science. 2008 Jun 13;320(5882):1496-501.
- Sancak Y, Bar-Peled L, Zoncu R, Markhard AL, Nada S, Sabatini DM.Ragulator-Rag complex targets mTORC1 to the lysosomal surface and is necessary for its activation by amino acids.Cell. 2010 Apr 16;141(2):290-303.
- Durán RV, Oppliger W, Robitaille AM, Heiserich L, Skendaj R, Gottlieb E, Hall MN.Glutaminolysis activates Rag-mTORC1 signaling.Mol Cell. 2012 Aug 10;47(3):349-58.
- Han JM, Jeong SJ, Park MC, Kim G, Kwon NH, Kim HK, Ha SH, Ryu SH, Kim S.Leucyl-tRNA synthetase is an intracellular leucine sensor for the mTORC1-signaling pathway.Cell. 2012 Apr 13;149(2):410-24.
- Zoncu R, Bar-Peled L, Efeyan A, Wang S, Sancak Y, Sabatini DM.mTORC1 senses lysosomal amino acids through an inside-out mechanism that requires the vacuolar H(+)-ATPase.Science. 2011 Nov 4;334(6056):678-83.
- Gulati P, Gaspers LD, Dann SG, Joaquin M, Nobukuni T, Natt F, Kozma SC, Thomas AP, Thomas G.Amino acids activate mTOR complex 1 via Ca2+/CaM signaling to hVps34.Cell Metab. 2008 May;7(5):456-65. doi: 10.1016/j.cmet.2008.03.002.
- Yan Y, Flinn RJ, Wu H, Schnur RS, Backer JM.hVps15, but not Ca2+/CaM, is required for the activity and regulation of hVps34 in mammalian cells.Biochem J. 2009 Feb 1;417(3):747-55.
- Byfield MP, Murray JT, Backer JM.hVps34 is a nutrient-regulated lipid kinase required for activation of p70 S6 kinase.J Biol Chem. 2005 Sep 23;280(38):33076-82.
- Yoon MS, Du G, Backer JM, Frohman MA, Chen J.Class III PI-3-kinase activates phospholipase D in an amino acid-sensing mTORC1 pathway.J Cell Biol. 2011 Oct 31;195(3):435-47.
- Foster DA.Regulation of mTOR by phosphatidic acid?Cancer Res. 2007 Jan 1;67(1):1-4.
- Sun Y, Chen J.mTOR signaling: PLD takes center stage.Cell Cycle. 2008 Oct;7(20):3118-23.
- O’Neil TK, Duffy LR, Frey JW, Hornberger TA.The role of phosphoinositide 3-kinase and phosphatidic acid in the regulation of mammalian target of rapamycin following eccentric contractions.J Physiol. 2009 Jul 15;587(Pt 14):3691-701.
- Yan L, Mieulet V, Burgess D, Findlay GM, Sully K, Procter J, Goris J, Janssens V, Morrice NA, Lamb RF.PP2A T61 epsilon is an inhibitor of MAP4K3 in nutrient signaling to mTOR.Mol Cell. 2010 Mar 12;37(5):633-42.
- Bentzinger CF, Romanino K, Cloëtta D, Lin S, Mascarenhas JB, Oliveri F, Xia J, Casanova E, Costa CF, Brink M, Zorzato F, Hall MN, Rüegg MA.Skeletal muscle-specific ablation of raptor, but not of rictor, causes metabolic changes and results in muscle dystrophy.Cell Metab. 2008 Nov;8(5):411-24.
- Maehama T, Tanaka M, Nishina H, Murakami M, Kanaho Y, Hanada K.RalA functions as an indispensable signal mediator for the nutrient-sensing system.J Biol Chem. 2008 Dec 12;283(50):35053-9.
- Xu L, Salloum D, Medlin PS, Saqcena M, Yellen P, Perrella B, Foster DA.Phospholipase D mediates nutrient input to mammalian target of rapamycin complex 1 (mTORC1).J Biol Chem. 2011 Jul 22;286(29):25477-86.
- Smith EM, Finn SG, Tee AR, Browne GJ, Proud CG.The tuberous sclerosis protein TSC2 is not required for the regulation of the mammalian target of rapamycin by amino acids and certain cellular stresses.J Biol Chem. 2005 May 13;280(19):18717-27.
- Long X, Ortiz-Vega S, Lin Y, Avruch J. Rheb binding to mammalian target of rapamycin (mTOR) is regulated by amino acid sufficiency. J Biol Chem. 2005 Jun 24;280(25):23433-6.
- Sun Y, Fang Y, Yoon MS, Zhang C, Roccio M, Zwartkruis FJ, Armstrong M, Brown HA, Chen J.Phospholipase D1 is an effector of Rheb in the mTOR pathway.Proc Natl Acad Sci U S A. 2008 Jun 17;105(24):8286-91.
- Ma WN, Park SY, Han JS.Role of phospholipase D1 in glucose-induced insulin secretion in pancreatic Beta cells.Exp Mol Med. 2010 Jun 30;42(6):456-64.
- Kim S, Kim SF, Maag D, Maxwell MJ, Resnick AC, Juluri KR, Chakraborty A, Koldobskiy MA, Cha SH, Barrow R, Snowman AM, Snyder SH. Amino acid signaling to mTOR mediated by inositol polyphosphate multikinase. Cell Metab. 2011 Feb 2;13(2):215-21.
- Sato T, Nakashima A, Guo L, Tamanoi F.Specific activation of mTORC1 by Rheb G-protein in vitro involves enhanced recruitment of its substrate protein.J Biol Chem. 2009 May 8;284(19):12783-91.
- Phillips SM, Tang JE, Moore DR. The role of milk- and soy-based protein in support of muscle protein synthesis and muscle protein accretion in young and elderly persons. J Am Coll Nutr. 2009 Aug;28(4):343-54.
- Phillips SM, Tipton KD, Aarsland A, Wolf SE, Wolfe RR. Mixed muscle protein synthesis and breakdown after resistance exercise in humans. Am J Physiol. 1997 Jul;273(1 Pt 1):E99-107.
- Wolff NC, Vega-Rubin-de-Celis S, Xie XJ, Castrillon DH, Kabbani W, Brugarolas J.Cell-type-dependent regulation of mTORC1 by REDD1 and the tumor suppressors TSC1/TSC2 and LKB1 in response to hypoxia.Mol Cell Biol. 2011 May;31(9):1870-84.
- DeYoung MP, Horak P, Sofer A, Sgroi D, Ellisen LW.Hypoxia regulates TSC1/2-mTOR signaling and tumor suppression through REDD1-mediated 14-3-3 shuttling.Genes Dev. 2008 Jan 15;22(2):239-51.
- Kelleher AR, Kimball SR, Dennis MD, Schilder RJ, Jefferson LS.The mTORC1 signaling repressors REDD1/2 are rapidly induced and activation of p70S6K1 by leucine is defective in skeletal muscle of an immobilized rat hindlimb.Am J Physiol Endocrinol Metab. 2012 Nov 27