The relationship between ATP and skeletal muscle fatigue


There has been no universal mechanism found to explain skeletal muscle fatigue. With the onset of intense exercise, muscle force has been shown to fall, which is often illustrated as a sign of fatigue. As ATP supplies the immediate energy to the muscle cell, the relationship between ATP and fatigue has been a topic of interest among researchers. During muscle contraction the majority of ATP is used for cross bridge cycling, ion pumps and transporters, and chemical signalling cascades. A fall in [ATP] would result in the development of rigor and goes against the viability of the cell. It is proposed that whole muscle [ATP] does not fall with repeated contraction, though there may be differences in [ATP] between fiber types with fatigue. Rather, fatigue is a protective mechanism by the muscle to preserve [ATP] accomplished by down regulating cellular processes that use ATP for energy.

Key-words : fatigue, ATP concentration, exercise, skeletal muscle


Introduction to fatigue

MacIntosh et al (2006) define skeletal muscle fatigue as a contractile response that is less than that which is expected or anticipated for a given stimulation. Fatigue is generally recorded experimentally as a decline in muscle force. Muscular fatigue differs from muscle injury; both result in a reversible decline in performance, but fatigue does not involve structural damage to the muscle fiber (Allen et al. 2008). Figure 1 shows the decline in force associated with fatigue in a mouse flexor digitorum brevis muscle, stimulated at 70Hz for 350ms every 4s for 2min (Allen et al. 2008). It is seen that in the initial thirty to forty seconds of stimulation force decreases 10-20% of its initial value (Allen et al. 2008).

Fatigue can be of central or peripheral origin (MacIntosh and Rassier, 2000). Central fatigue occurs when there is neural down regulation within the central nervous system, resulting in a decline in muscle force even though the muscle is capable of more force (MacIntosh and Rassier, 2000). This type of fatigue occurs within the central nervous system, from the initiation of a movement in the brain, down the spinal cord, and ends at the alpha motoneuron. Peripheral fatigue occurs between the neuromuscular junction and the muscle fibers themselves. This encompasses changes in motor unit action potentials, ion concentrations and ion pump functioning, as well as altered cross bridge kinetics that are observed with fatigue (MacIntosh et al. 2006). This paper will focus on consequences of peripheral fatigue.

Whether or not [ATP] changes with peripheral fatigue and whether or not changes in [ATP] are relevant to force production, have been controversial topics among researchers. ATP is used in excitation contraction coupling, by ion pumps and transporters, and for chemical signaling cascades (MacIntosh et al. 2012). A decrease in ATP would, for example, result in reduced force production and altered cross bridge mechanics (Fitts, 1994). While there have been reports of [ATP] falling to critical levels, the majority of research suggests that overall muscle [ATP] does not fall more than 10-20% of normal values (Fitts, 1994; Allen et al, 2008; MacIntosh et al. 2012). This paper will focus on [ATP] changes with peripheral fatigue, and suggests the following:

  1. The average [ATP] within a muscle does not change

  2. Metabolic properties differ between fast and slow twitch fibers, affecting ATP use and subsequent changes in concentration

  3. Fatigue may be a protective mechanism to preserve [ATP]

Let’s first look at what exactly ATP is, and how it is used and replenished within a muscle fiber.

Adenosine triphosphate (ATP)

ATP is the energy currency of biological cells. The hydrolysis of ATP into its constituents releases a certain amount of energy that the muscle cell can use for work by the following reaction:

ATP <=> ADP + Pi + E

Where ADP is adenosine diphospate, Pi is inorganic phosphate, and E is the energy released. The amount of energy available for release by the hydrolysis of ATP depends on the energy status of the cell. This can be measured roughly by a cells energy charge using a ratio of the reactants and products (MacIntosh et al. 2006):


An increase in [ADP] and [Pi] has a feedback effect on the cell, stimulating metabolism to synthesize ATP (MacIntosh et al. 2006).

Maintaining ATP

There are three main energy systems that function to maintain the levels of ATP within skeletal muscle at rest or during exercise: the immediate energy system (anaerobic), short term (anaerobic glycolytic), and long-term (aerobic) energy systems. These three methods are in place as a means for the cell to maintain the rate of ATP synthesis with that of its use (Hochcka and Matheson, 1992). At rest, there is about 4-7mM of endogenous ATP stored in a cell (Hochchka and Matheson, 1992). Oxidative metabolism maintains [ATP] at rest. With the onset of intense exercise, rates of ATP use can increase upwards of 100 fold (Hochchka and Matheson, 1992).

In skeletal muscle, three main ATPases contribute to the large increased rate of ATP use with isometric tetanus (Homsher, 1987). Actomyosin ATPase is located on the myosin globular head and uses approximately 65-80% of available ATP during contraction for cross bridge cycling. Ca2+ ATPase (SERCA) sits on the sarcoplasmic reticulum and uses ATP to enable Ca2+ release and uptake during excitation-contraction coupling. This uses about 20-35% of available ATP. The last ~10% of ATP is used by the Na+/K+ pump, which works to restore the resting membrane potential following an action potential (Homsher, 1987).

With intense exercise, oxidative metabolism alone cannot supply enough ATP to meet the needs of these molecular motors and ion pumps (Fitts, 1994). Without a faster means of ATP synthesis, [ATP] would deplete within seconds. This would be detrimental to the cell; absolute depletion of ATP is seen in cell death (MacIntosh et al, 2011). The immediate and short-term energy systems exist to meet the increased ATP demand.

The immediate energy system is a rapid anaerobic method of restoring ATP, using PCr and the enzyme creatine kinase (CK) to phosphorylate ADP (MacIntosh et al. 2006).

ADP + PCr <=> ATP + Cr


Two ADP can also be used to make an ATP with the enzyme myoadenylate kinase (MK) (MacIntosh et al. 2006). This reaction results in one ATP and one AMP (adenosine monophosphate).

2ADP <=> ATP + AMP


The net reaction of the short-term energy system breaks down glucose into lactate and hydrogen ions while phosphorylating ATP (Allen et al. 2008). This system has multiple enzymes that tend to be bound in close proximity to each other and are located in areas that have an increased use of ATP (MacIntosh et al. 2006). The by-products of glycolysis and the immediate energy system – pyruvate, nicotinamide adenine dinucleotide (NADH), and creatine – stimulate oxidative metabolism (MacIntosh, 2006).

1. The average [ATP] within a muscle does not change

These systems all exist to maintain and return the cell to homeostasis. Since ATP binds to myosin globular heads releasing them from actin binding sites to terminate cross bridge formations, a lack of free ATP would stop cross bridge cycling. If global muscle [ATP] did fall below critical levels with fatigue, the muscle would lose its ability to relax (Fitts, 1994). This would lead to the development of rigor. A state of rigor is not observed in exercise.

Using phosphorus-31 magnetic resonance spectroscopy (31P-MRS) whole muscles have been observed in-vivo and show marginal declines in [ATP] with severe fatigue, under aerobic conditions. Figure 2 shows a 31P-MRS spectrum obtained by Dawson et al. (1978). The amplitude of the peak represents the amount of metabolite present. As can be seen, the peak labeled ‘Adenosine P’ (ATP) exhibits barely any change in amplitude over the course of an hour with cyclic stimulation. Argov et al. (2000) found similar results in human muscle. This provides support for the first suggestion made by this paper, stating that average [ATP] within a muscle does not change with fatigue.

It should be noted that these results were obtained under aerobic conditions. With submaximal activation producing contractions of less than 50% maximal force, blood flow to the intact muscle is maintained (Allen et al. 2008). The by-products of the immediate and short-term energy systems will induce oxidative metabolism. With maximal activation (contractions >50% maximum force) blood flow is occluded (Allen et al. 2008). This inhibits oxidative metabolism; once the substrates for the immediate and short-term energy systems are used up (after about 2 minutes), ATP concentrations will begin to fall. To avoid such a catastrophe, intact muscles have been shown to reduce their firing rate during a maximal voluntary contraction (Bigland-Ritchie et al 1983). It has also been suggested that intact motor units cyclically activate in order to reduce anaerobic fatigue of a single motor unit (MacIntosh et al. 2006). By directly stimulating a motor unit or muscle in-situ/in-vivo at a constant frequency, there is no reduced neural firing frequency or cyclic activation which would occur in an intact muscle. This could lead to an unrealistic amount of ATP use and consequentially an overestimation of ATP depletion with fatigue.

To stimulate anaerobic conditions as seen with maximal contractions, a pressure cuff is placed proximal to the muscle being studied. The muscle is then exercised at a high intensity or stimulated at a high frequency. De Haan (1990) looked at rat gastrocnemius muscle, with three groups exercised over 10s with blocked circulation. All groups were stimulated over 10s at 120Hz with group A performing ten contractions, group B twenty-five contractions, and group C forty contractions. ATP was found to fall to 65% of resting values for group A and to 35% of resting values for group B and C. A study looking at human quadriceps stimulated under anaerobic conditions at 20Hz for 1.6s with 1.6s rest 64 times (totaling 204.8s) found [ATP] reached 56.6% of resting levels (Spriet et al. 1987). Soderlund and Hultman (1990) used a similar protocol to Spriet et al. (1987), stimulating quadriceps muscle at 20Hz for 1.6s with 1.6s rest, 52 times. They found whole muscle ATP to decrease by 42% with occluded blood flow. By ‘simulating’ anaerobic conditions it can be seen that ATP depletion is much greater than reported under aerobic conditions, and the percentage of depletion depends on the number of contractions. More contractions require more cross bridge cycling, which uses more ATP. By blocking oxygen perfusion, at rest or with exercise, metabolites may accumulate and minimize the energy provided by ATP hydrolysis (Dawson et al. 1978).

2. Metabolic properties differ between slow and fast twitch fibers

Slow twitch motor units tend to be used at low workloads because they have a low threshold to neural stimulation (Fitts, 1994). Maximal efforts will recruit high-threshold fast twitch motor units (Fitts, 1994). This leads to different metabolic properties between slow twitch and fast twitch motor units.

Muscle fibers differ in their calcium and ATP handling, and oxidative capacities (Allen et al. 2008). These factors affect the fiber’s contraction velocity and fatigue resistance (Allen et al. 2008), and will indirectly affect their recruitment during exercise. Fast twitch fibers tend to have a higher SR and myofibrillar ATPase activity (Fitts, 1994). Therefore these fibers have a corresponding high contraction velocities and short isometric twitch contractions (Fitts, 1994). Fast anaerobic twitch fibers also have a poor oxidative capacity because of their small number of mitochondria, making them more fatigable than slow twitch aerobic fibers (Fitts, 1994).

Using maximal cycling over 25s, it was found that the fastest fibers (type IIX) show the greatest depletion of ATP, decreasing as much as 80% from resting values once PCr fell to 11% (Karatzaferi et al. 2001). [ATP] have been shown to decrease slightly more in fast twitch fibers compared against slow twitch during high intensity exercise (Jansson et al, 1987;Soderlund and Hultman, 1990), in fact, with aerobic exercise [ATP] may rise in highly oxidative slow twitch fibers (Fitts, 1994). The depletion in fast twitch fibers is likely due to their limited oxidative capacity (and therefore limited energy supply), and to a lesser extent, the anaerobic environment created during high intensity exercise. Karatzaferi et al (2001) found that low levels of ATP in type IIX fibers occurred when power was at 50% of its initial value. While single fiber ATP depletion is much greater than that which occurs across the whole muscle, an increased prevalence of [ATP] depletion in fast twitch fibers could have important performance implications for power athletes performing high intensity work.

3. ATP and fatigue

Since ATP is an integral part of excitation contraction coupling, it can be postulated that the decline in force seen with fatigue might be a protective mechanism to preserve [ATP] during repeated muscle activation. This could be accomplished by a regulating ATP hydrolysis rates or regulating Ca2+ handling, and might be initiated by localized depletions of ATP within the cell.

a) Reduced rate of ATP hydrolysis

The initiation of high intensity exercise activates the immediate energy system, hydrolyzing down PCr to replenish ATP. [PCr] depletes to near zero within 30s of exercise (Spriet et al.). Rather than causing a decrease in cellular ATP, this results in an increase in free ADP (Allen et al. 2008). A rise in [ADP] changes the energy charge of the cell, decreasing the availability of free energy from the hydrolysis of ATP (Dawson et al, 1978). Figure 3(b) shows force as a function of affinity for ATP hydrolysis, illustrating a marginal decrease in affinity with a corresponding fall in force. The amount of force produced per cross-bridge is constant, so a decline in force is likely the result of a reduced rate of ATP hydrolysis (Dawson et al. 1978).

b)Regulation of cellular processes that use ATP

To avoid the metabolic catastrophe that would accompany a complete depletion of ATP, it has been suggested that the cell regulates processes using ATP during excitation contraction coupling (MacIntosh et al. 2012). Actomyosin ATPase and Ca2+ ATPase together account for roughly 90% of ATP use during muscle contraction and are active when intracellular concentrations of Ca2+ rise (MacIntosh et al. 2012). By down regulating Ca2+ use and release, ATP hydrolysis can be minimized and [ATP] can be preserved.

c) Localized depletion of ATP

Metabolite concentrations vary across the cell; in particular certain areas of the cell have higher rates of ATP use than others (Allen et al. 2008). While metabolite gradients across the cell are small, areas where it is difficult for ATP to diffuse could result in large changes and localized depletions. One such area is the triad region, consisting of one transverse tubule communicating with two SR’s separated by a small amount of cytoplasm (Allen et al. 2008). The transverse tubule contains roughly half of the cell’s Na+/K+ pumps as well as other ATPases, and the SR has a dense concentration of Ca2+ ATPases (Allen et al. 2008). With muscle activation, the rate of ATP use in this area increases exponentially and ATP is rapidly replenished by glycolysis (Allen et al. 2008). An area such as this could have drastically different metabolic concentrations than the rest of the cell during muscle contraction (Allen et al. 2008). The triads intimate relationship with the muscle activation and ATP use allow it to quickly recognize an increase in cellular activity (Allen et al. 2008). This could have important implications for cellular monitoring of muscle activity. For example, because of the triads close relation to the SR, which stores and releases Ca2+, an increase in ATP activity may act as a signal. This could result in a down regulation of Ca2+ release at the onset of muscle contraction. As mentioned above (section 3b), this is one of the proposed mechanisms by which the cell preserves ATP during muscle activation.


The goal of this paper was to illustrate the relationship between skeletal muscle fatigue and ATP. During fatigue, whole muscle [ATP] remains relatively high, though [ATP] may vary between individual muscle fiber types. Local depletions of ATP may act as a signaling mechanism within the cell to down regulate calcium release and subsequent ATP use. This down regulation is likely one factor resulting in the decrease in muscle force seen in fatigue, instead of a decrease in [ATP]. ATP use is closely regulated and fatigue is likely a protective mechanism to prevent ATP depletion.


Allen, D.G., Lamb, G.D., and Westerblad, H. 2008. Skeletal muscle fatigue: cellular mechanisms. Physiol. Rev. 88 : 287-332. doi:10.1152/physrev.00015.2007.

Argov, Z, Lofberg, M. and Arnold, D.L. 2000. Insights into muscle diseases gained by phosphorous magnetic resonance spectroscopy. Muscle Nerve. 23 : 1316-1334

Bigland-Ritchie, B., Johansson, R., Lippold. O.C., Smith, S., and Woods, J.J. 1983. Changes in motoneuron firing rates during sustained maximal voluntary contractions. J. Physiol. 340 : 335-346.

Dawson, M.J., Gadian, D.G. and Wilkie, D.R. 1978. Muscular fatigue investigated by phosphorus nuclear magnetic resonance. Nature, 274 : 861-866

Fitts, R.H. 1974. Cellular mechanisms of muscle fatigue. Physiol. Rev. 74 : 49-94

de Haan, A. 1990. High-energy phosphates and fatigue during repeated dynamic contractions of rat muscle. Experimental Physiology. 75 : 851-854

Hochachka, P.W. and Matheson, G.O. 1992. Regulating ATP turnover rates over broad dynamic work ranges in skeletal muscles. J. Appl. Physiol. 73 : 1697-1703

Homsher, E. 1987. Muscle enthalpy production and its relationship to actomyosin ATPase. Ann. Rev. Physiol. 49 : 673-690.

Jansson, E., Dudley, G.A., Norman, A. and Tesch, P.A. 1987. ATP and IMP in single human muscle fibers after high intensity exercise. Clin Physiol. 7 : 337-345. doi:10.1111/j.1475-097X.1987.tb00177.x

Karatzaferi, C., de Haan A., Ferguson, R.A., van Mechelen, W. and Sargeant, A.J. 2001. Phosphocreatine and ATP content in human single muscle fibers before and after maximum dynamic exercise. Pflugers Arch. 442 : 467-474. doi:10.1007/s004240100552

MacIntosh, B.R., Gardiner, P., and McComas, A. 2006. Skeletal Muscle Form and Function. Human Kinetics.

MacIntosh, B.R., Holash, R.J., and Renaud, J.M. 2012. Skeletal muscle fatigue – regulation of excitation-contraction coupling to avoid metabolic catastrophe. Journal of Cell Science. 125 : 2105-2114. doi: 10.1242/jcs.093674.

MacIntosh, B.R. and Rassier D.E. 2002. What is fatigue? Can. J. Appl. Physiol. 27: 42-56

Macintosh, B.R. & Shahi, M.R.S. 2011. A peripheral governor regulates muscle contraction. Appl. Physiol. Nutr. Metab. 36 : 1-11. doi:10.1139/H10-073.

Soderlund, K. & Hultman, E. 1990. ATP content in single fibers from human skeletal muscle after electrical stimulation and during recovery. Acta. Physiol. Scand. 139 : 459-466 ISSN 0001-677.

Spriet, L.L., Soderlund, K., Bergstrom, M., and Hultman, E. 1987. Anaerobic energy release in skeletal muscle during electrical stimulation in men. J. Appl. Physiol. 62 : 611-615



  • Facebook - White Circle
  • Twitter - White Circle
  • LinkedIn - White Circle
This site was designed with the
website builder. Create your website today.
Start Now