β‐Adrenergic modulation of skeletal muscle contraction: key role of excitation–contraction coupling

Abstract

Our aim is to describe the acute effects of catecholamines/β‐adrenergic agonists on contraction of non‐fatigued skeletal muscle in animals and humans, and explain the mechanisms involved. Adrenaline/β‐agonists (0.1–30 μm) generally augment peak force across animal species (positive inotropic effect) and abbreviate relaxation of slow‐twitch muscles (positive lusitropic effect). A peak force reduction also occurs in slow‐twitch muscles in some conditions. β₂‐Adrenoceptor stimulation activates distinct cyclic AMP‐dependent protein kinases to phosphorylate multiple target proteins. β‐Agonists modulate sarcolemmal processes (increased resting membrane potential and action potential amplitude) via enhanced Na⁺–K⁺ pump and Na⁺–K⁺–2Cl⁻ cotransporter function, but this does not increase force. Myofibrillar Ca²⁺ sensitivity and maximum Ca²⁺‐activated force are unchanged. All force potentiation involves amplified myoplasmic Ca²⁺ transients consequent to increased Ca²⁺ release from sarcoplasmic reticulum (SR). This unequivocally requires phosphorylation of SR Ca²⁺ release channels/ryanodine receptors (RyR1) which sensitize the Ca²⁺‐induced Ca²⁺ release mechanism. Enhanced trans‐sarcolemmal Ca²⁺ influx through phosphorylated voltage‐activated Ca²⁺ channels contributes to force potentiation in diaphragm and amphibian muscle, but not mammalian limb muscle. Phosphorylation of phospholamban increases SR Ca²⁺ pump activity in slow‐twitch fibres but does not augment force; this process accelerates relaxation and may depress force. Greater Ca²⁺ loading of SR may assist force potentiation in fast‐twitch muscle. Some human studies show no significant force potentiation which appears to be related to the β‐agonist concentration used. Indeed high‐dose β‐agonists (∼0.1 μm) enhance SR Ca²⁺‐release rates, maximum voluntary contraction strength and peak Wingate power in trained humans. The combined findings can explain how adrenaline/β‐agonists influence muscle performance during exercise/stress in humans.

Introduction

Catecholamines such as adrenaline (epinephrine) or noradrenaline (norepinephrine), released from the adrenal medulla or sympathetic nerves, can assist exercise performance through well‐known effects such as raised heart rate and myocardial contractility, bronchodilatation, electrolyte regulation, fuel mobilization and blood flow redistribution (Bowman, 1980; Zouhal et al. 2008; Roatta & Farina, 2010; Hostrup et al. 2014 a,b). In addition catecholamines, sympathetic nerve stimulation, or β‐adrenergic agonists can enhance the contractile performance of skeletal muscles of amphibia, animals and humans (for reviews see Bowman, 1980; Williams & Barnes, 1989 a; Roatta & Farina, 2010). Such effects on contractile function include a potentiation of peak force (positive inotropic effect) (Figs 1 A and C, 3 B, and 4 C and D) and a faster relaxation of some muscles (positive lusitropic effect) (Fig. 1 A–C). However, in some situations a depression of peak force can occur (negative inotropic effect) (Fig. 1 B). The magnitude and nature of these adrenergic effects vary depending on such aspects as the fibre‐type composition of the muscle, species, concentration of agent, and the muscle condition, e.g. fatigued (Juel, 1988; Cairns & Dulhunty, 1994; Decorte et al. 2015) or K⁺ depressed (Clausen, 2003; Cairns et al. 2011). These effects all require activation of β₂‐adrenergic receptors, since they manifest with general β‐adrenergic agonists, e.g. isoprenaline (isoproterenol), or selective β₂‐adrenergic agonists such as salbutamol or terbutaline, i.e. short‐acting β₂‐agonists, and are also antagonized with general or selective β₂‐adrenoceptor blockers (Bowman, 1980). With the advent of long‐acting β₂‐agonists, e.g. clenbuterol, some caution is needed about non‐selective effects especially when high concentrations are used (Head & Ha, 2011; McCormick et al. 2010), and their chronic application promotes muscle hypertrophy (Lynch & Ryall, 2008). The focus of the present article is on acute effects of catecholamines and short‐acting β‐agonists on cellular processes and contractile function of non‐fatigued skeletal muscle.

Figure 1. Representative effects of terbutaline on isometric force or power in animal and human muscles .

A, influence of terbutaline (10 μm for 15 min) on maximal tetanic contractions (100 Hz) in isolated rat slow‐twitch soleus fibre bundles. Supramaximal stimulation pulses delivered from parallel plate electrodes, 24°C. Reproduced from Cairns & Dulhunty, 1993 a with permission. B, influence of oral terbutaline (8 mg for 2 h+) compared with control (pre‐ingestion) on unfused tetanic contractions (10 Hz) in human soleus muscle in vivo. Stimulation of soleus muscle at body temperature (G. Crivelli, F. Borrani, R. Capt, G. Gremion & N. A. Maffiuletti, unpublished data). C, influence of inhaled terbutaline (15 mg for ∼20 min, plasma 24 ± 1 ng ml⁻¹) compared with placebo on maximal voluntary contraction for quadriceps muscles in a trained human male. D, influence of inhaled terbutaline (15 mg for ∼30 min, plasma 24 ± 1 ng ml⁻¹) compared with placebo on Wingate cycle power in a trained human male. C and D are reproduced from Hostrup et al. 2014 a with permission.

Figure 3. Representative effects of 10 μm terbutaline on myoplasmic [Ca²⁺] ([Ca²⁺]_i) and tension during tetanic stimulation (100 Hz) of single, intact fast‐twitch flexor digitorum brevis fibres of the mouse in vitro .

[Ca²⁺]_i was measured with the fluorescent Ca²⁺ indicator Indo‐1. ‘Terbutaline’ is the maximal effect obtained after 10 min terbutaline exposure. Parallel plate stimulation electrodes, 22°C. Adapted from Cairns et al. 1993 with permission.

Figure 4. Influence of 10 μm isoprenaline on sarcoplasmic reticulum [Ca²⁺] ([Ca²⁺]_SR) and cell shortening (i.e. fibre deflection) during twitch stimulation of fast‐twitch tibialis anterior fibres of the mouse in situ .

[Ca²⁺]_SR was measured with a cameleon Ca²⁺ sensor as the fluorescence ratio (F(535 nm)/F(450 nm)). Data are means ± SEM (25 individual twitches from 3 fibres). B and D were obtained 10 min after application of isoprenaline compared with before (control), i.e. A and C, respectively. Values are normalized to control. The positive inotropic effect is shown as the 50% increase in peak cell shortening (D). Reproduced from Rudolf et al. 2006 with permission (© 2006 Kim and Roy. Journal of Cell Biology. 173:187–193. doi:10.1083/jcb.200601160).

Several reviews over the last 15 years have described β‐adrenergic effects on selected muscle processes (e.g. Catterall, 2000; Clausen, 2003; Lynch & Ryall, 2008; Capes et al. 2011) yet none have provided a wholistic appraisal of the β‐adrenergic mechanisms influencing muscle contraction since Williams & Barnes (1989 a). Moreover, a resurgence of studies on β‐adrenergic effects on human muscle (Table 1) have yielded inconsistent findings on peak force, in contrast to responses in animal muscles, and these require explanation. Also debate has occurred on whether or not a stress response elicits an adrenergic potentiation of muscle force (Andersson et al. 2012; Roatta & Farina, 2013). Therefore, the aims of the present review are: (1) to provide a comprehensive update on the mechanisms underpinning β‐adrenergic effects on non‐fatigued skeletal muscle contraction in humans and animal models; and (2) to explain how adrenaline/β‐agonists influence muscle performance during exercise or stress in humans.

Table 1.

Effects of catecholamines or β‐adrenergic agonists on skeletal muscle contraction across species

		Peak force
Species	Drug (concentration)	Twitches	Unfused tetani	Maximal tetani	Relaxation rate
Amphibia	Ad, Iso (1–30 μm)	↑30–130% (Refs 1–4)	‐ ‐ ‐ ‐	↔ (Refs 1,4)	↑30–50% (Refs 2,4)
Mammals Fast‐twitch	Ad, NAd, Iso, Salb, Terb (0.1–120 μm)	↑7–51% (Refs 6–12)	↑12–150% (Refs 5,7,8,10)	↑7–20% (Refs 9,10,12)	↔· or ↓7–11% (Refs 5–9)
Slow‐twitch (cat, dog, rabbit, guinea‐pig)	Ad, NAd, Iso, Salb	↓18–26% (Refs 7,8)	↓20–60% (Refs 5,7–9,14)	↔ or ↑3% (Refs 7,8 or 9)	↑33% (Refs 5,7,8)
Slow‐twitch (rat, mouse)	Ad, Iso, Salb, Terb (0.1–10 μm)	↑15–52% (Refs 10,12–15)	↑7–41% (Refs 10,14)	↑7–29% (Refs 10,13)	↑10–41% (Refs 10,14,15)
Humans (isometrics)	Ad, Terb, Salb (6–8 or 15–20 mg)	↔ or ↑11% (Refs 16–19 or 20)	↓16–40% (Refs 16,17,19)	↔ or ↑5–8% (MVC) (Refs 16–19 or 20–22)	↑10–28% (Refs 16–20)
(sprint cycling, isokinetics)	Salb or Terb (4–6 or 15–20 mg)	‐ ‐ ‐ ‐	↑4–14% (Refs 23–26)	↑2–14% (PPO) (Refs 21,23–25)	‐ ‐ ‐ ‐

Values display the range of mean data from non‐fatigued muscles across the studies cited. ‐ ‐ ‐ ‐, no data reported. Drugs, applied as a bolus or continuous exposure, include adrenaline (Ad), noradrenaline (NAd), isoprenaline (Iso), salbutamol (Salb), terbutaline (Terb). Mammalian fast‐twitch muscles: tibialis anterior, extensor digitorum longus, gastrocnemius, diaphragm, sternomastoid, or flexor digitorum brevis. Mammalian slow‐twitch muscles: soleus, plantaris. Human muscles: quadriceps, triceps surae soleus, adductor pollicis. MVC, maximal isometric voluntary contraction, PPO, peak power output. Relaxation rate was determined for twitches or tetani using the maximum slope of force decline, or 1/relaxation time. References: (1) Arreola et al. 1987; (2) Gonzalez‐Serratos et al. 1981; (3) Oota & Nagai, 1977; (4) Williams & Barnes, 1989 b; (5) Al‐Jeboory & Marshall, 1978; (6) Andersson et al. 2012; (7) Bowman, 1980; (8) Bowman & Zaimis, 1958; (9) Bowman & Nott, 1970; (10) Cairns & Dulhunty, 1993 a; (11) Cairns et al. 1993; (12) Kӧssler et al. 1991; (13) McCormick et al. 2010; (14) Ha et al. 1999; (15) Slack et al. 1997; (16) Crivelli et al. 2013; (17) Crivelli & Maffiuletti, 2014; (18) Crivelli et al. 2011; (19) Marsden & Meadows, 1970; (20) Hostrup et al. 2014 a; (21) Hostrup et al. 2014 b; (22) Kalsen et al. 2013; (23) Collomp et al. 2005; (24) Le Panse et al. 2007 (25) Sanchez et al. 2012; (26) van Baak et al. 2000. Note: several studies with humans show non‐significant effects with β‐agonists, presumably due to low doses used, and are not indicated here (see Pluim et al. 2011 for references).

β‐Adrenergic effects on contraction

Peak force

Table 1 depicts the positive findings seen with high concentrations of adrenaline/β‐agonists on contraction of different types of skeletal muscle from various species. Animal studies show an increased peak force of submaximal isometric contractions for amphibian twitch‐muscle in vitro, all mammalian fast‐twitch muscles in situ or in vitro, and slow‐twitch soleus muscles from rats and mice in vitro. The relative twitch potentiation for mammals is less for skeletal muscle, i.e. <50% initial (Table 1) than for cardiac muscle, i.e. 2‐ to 5‐fold increase (Kurihara & Konishi, 1987; Wolska et al. 1996; Li et al. 2000). Furthermore, peak tetanic force is amplified by up to 30% in isolated fast‐ and slow‐twitch muscles of rats and mice at room temperature (Figs 1 A and 3 B). In contrast, negative inotropic effects appear for submaximal isometric contractions of slow‐twitch soleus muscles of larger mammals in situ, and with unfused tetani in humans in vivo (Fig. 1 B). Whilst these contrasting effects seemingly involve fibre‐type or species differences, it is evident that both positive and negative inotropic effects can manifest in the same muscle. For example, prolonged exposure of isolated guinea‐pig soleus muscles to terbutaline (2.3 μm) induced a biphasic effect on peak twitch force with a 13% decline by 5 min which reversed into a 25% potentiation by 30 min (Holmberg & Waldeck, 1980). Also tetani evoked at >40 Hz are potentiated slightly in soleus muscles of the anaesthetized cat whereas unfused tetani are depressed (Bowman & Nott, 1970).

With non‐asthmatic humans, exposure to therapeutic doses of adrenaline/β‐agonists (i.e. 800 μg to 8 mg) often have no effect on the peak twitch force of various muscles (Table 1). However, salbutamol can increase peak force during isokinetic shortening contractions of the quadriceps (Table 1) and induce a rightwards shift of the force–velocity relationship (Sanchez et al. 2012). Many studies show an unchanged maximum isometric voluntary contraction (MVC) strength with adrenaline/β‐agonists (Pluim et al. 2011), although the side‐effects of tremor, nausea or tachycardia (Marsden & Meadows, 1970; van Baak et al. 2000) may influence assessments involving voluntary contractions. However, recent studies with extremely high‐dose terbutaline (15–20 mg, ∼0.1 μm) (Fig. 1 C), or combined β‐agonists (Kalsen et al. 2013) show a 5–8% potentiation of the quadriceps MVC in athletes. Interestingly, β‐agonists also increase peak power during Wingate cycle tests (Fig. 1 D) by augmenting the peak force applied to the pedals (Table 1). Evidently, the β‐adrenergic potentiation of peak force seen in animal muscles can also manifest in human muscles.

Relaxation

Table 1 shows that adrenaline/β‐agonists accelerate twitch relaxation in amphibian muscle, but with mammalian muscle the relaxation effect depends on fibre‐type composition. Twitch and tetanic relaxation are abbreviated in slow‐twitch muscles across all animal species (Fig. 1 A and B), an effect detected with isoprenaline concentrations from 1 nm (Slack et al. 1997). With cardiac muscle the β‐adrenergic acceleration of twitch relaxation (Kurihara & Konishi, 1987; Wolska et al. 1996; Li et al. 2000) is at least double that for skeletal muscle (Table 1). In contrast, fast‐twitch mammalian muscles exhibit an unchanged or slower relaxation (Figs 3 B, and 4 C and D). With human muscles there is a faster relaxation for the quadriceps (Fig. 1 C) and soleus (Table 1), but the adductor pollicis appears unaffected (Marsden & Meadows, 1970). We surmise that the speeding of relaxation with β‐agonists is a feature of slow‐twitch fibres in animal and human muscles.

Mechanisms for β‐adrenergic effects on contraction

The positive inotropic effect for MVC strength in humans could, in principle, involve enhanced motor drive from the central nervous system, but this idea can be excluded since the voluntary activation ratio is unchanged with β‐agonists (Crivelli et al. 2011, 2013; Hostrup et al. 2014 b). Facilitation of either muscle blood flow (Zouhal et al. 2008) or events at the neuromuscular junction (Bowman, 1980) are not required for force potentiation since it transpires in directly stimulated muscle in vitro. Accordingly the mechanism must involve a modulation of muscle cell process(es).

Cell signalling processes

To appreciate the mechanistic studies an understanding of the β₂‐adrenergic signalling pathway is needed (Fig. 2). This involves activation of sarcolemmal β₂‐adrenoceptors, coupled to stimulatory G‐proteins, which in turn activate adenylate cyclase to convert ATP to adenosine 3′,5′‐cyclic monophosphate (cAMP). This second messenger then activates several different cAMP‐dependent protein kinases (PKA) which phosphorylate target proteins to alter cell processes. Certainly, the finding that the β‐adrenergic potentiation of force requires 5–10 min for maximal effects in isolated single fibres (Gonzalez‐Serratos et al. 1981; Cairns et al. 1993) is coherent with formation and involvement of second messengers. A key feature is that myoplasmic cAMP ([cAMP]_i) increases by 2‐ to 3‐fold in mammalian and human muscle with adrenaline/β‐agonists (Al‐Jeboory & Marshall, 1978; Chasiotis et al. 1983; Godinho & Costa, 2003; Rudolf et al. 2006). Notably, maximal [cAMP]_i is attained only with β‐agonist concentrations of 10–50 μm (Al‐Jeboory & Marshall, 1978; Godinho & Costa, 2003), and this surpasses the physiological range (Harmer et al. 2000; Zouhal et al. 2008). The [cAMP]_i reaches a peak in 5 min then declines by 20–60% over time due to extrusion from the cell (Chasiotis et al. 1983; Godinho & Costa, 2003; Rudolf et al. 2006; Duarte et al. 2012). With prolonged β‐agonist exposure the raised interstitial [cAMP] forms adenosine which may antagonize any positive inotropic effects (Duarte et al. 2012). Strong backing for an involvement of [cAMP]_i comes from findings that extracellular application of membrane‐permeable analogues of cAMP, e.g. dibutyryl cAMP (DBcAMP) or 8‐bromo‐cAMP, exert positive inotropic effects in non‐fatigued amphibian (Oota & Nagai, 1977) or rodent muscles (Varagić & Kentera, 1978; Cairns & Dulhunty, 1993 a). Indeed, maximum increases of peak force or abbreviation of relaxation with DBcAMP quantitatively mimic that achieved with terbutaline in rat soleus (Cairns & Dulhunty, 1993 a). Interestingly, the slowly entering DBcAMP provokes an initial twitch depression which reverses into potentiation (Cairns & Dulhunty, 1993 a), as occurs with terbutaline (Holmberg & Waldeck, 1980). Hence the suggestion that lower [cAMP]_i activates a force‐depressing process whilst higher [cAMP]_i activates a force‐potentiating process.

Figure 2. The β2‐adrenergic or cAMP–PKA signalling pathway for skeletal muscle fibres (Lynch & Ryall, 2008; Berdeaux & Stewart, 2012 ), together with several target proteins/processes that may influence contraction .

Common pharmacological tools used to explore this pathway are shown. AKAP, A‐kinase anchoring protein; cAMP, cyclic AMP; DBcAMP, dibutyryl‐cAMP; G_s, stimulatory guanine nucleotide binding protein; GTPγS, non‐hydrolysable analogue of guanosine triphosphate; NKCC, Na⁺–K⁺–2Cl⁻ cotransporter; PKA, cAMP‐dependent protein kinase; RyR1, skeletal muscle ryanodine receptor; SR, sarcoplasmic reticulum; + stimulation; − inhibition. Target proteins are phosphorylated by PKA in intact fibres, membrane fragments or as isolated proteins (functional changes in parentheses). References for protein phosphorylation: (1) Bibert et al. 2008; (2) Wong et al. 2001; (3) Gosmanov & Thomason, 2002; (4) Yang & Barchi, 1990; (5) Emrick et al. 2010; (6) Johnson et al. 2005; (7) Mundina‐Weilenmann et al. 1991; (8) Liu et al. 1997; (9) Slack et al. 1997; (10) Anderson et al. 2012; (11) Hain et al. 1994; (12) Mayrleitner et al. 1995; (13) Reiken et al. 2003; (14) Suko et al. 1993; (15) Soderling et al. 1970.

Cyclic AMP activates several different PKA which phosphorylate specific serine residues on multiple target proteins (some indicated in Fig. 2). Different PKA isoforms are localized in spatially distinct microdomains, such as near the sarcolemma or the transverse (t‐) tubular‐sarcoplasmic reticulum (SR) interface, which allows localized elevations of [cAMP]_i (Rӧder et al. 2009). This compartmentalization is mediated by A‐kinase anchoring proteins (AKAPs) (Ruehr et al. 2003; Johnson et al. 2005; Emrick et al. 2010) and an intact actin cytoskeleton (Johnson et al. 2005; Rӧder et al. 2009), both of which influence the extent of phosphorylation and function of target proteins (Johnson et al. 1994; Ruehr et al. 2003). The cAMP‐dependent phosphorylation sites on target proteins are also influenced by Ca²⁺/calmodulin‐dependent protein kinases and phosphatases (Suko et al. 1993; Emrick et al. 2010). These target proteins exist as macromolecular complexes and incorporate several kinases and phosphatases (Marx et al. 2001; Capes et al. 2011) which could potentially influence the final cellular event. The abundance of phosphorylatable target proteins helps to explain why multiple processes are influenced by β‐agonists, as acknowledged long ago (Bowman, 1980). The next question involves determining which of the many target proteins/processes stimulated via the β‐adrenergic pathway (Fig. 2) specifically accounts for the modulation of contractile performance.

Sarcolemmal processes

Catecholamines/β‐agonists are known to modulate at least two sarcolemmal ion transport processes in a manner that could potentially influence action potentials and force.

Na⁺–K⁺ pump and Na⁺–K⁺–2Cl⁻ cotransporter. The ion transport process best known to be stimulated by β‐agonists in skeletal muscle is the Na⁺–K⁺ pump (i.e. Na⁺–K⁺‐ATPase) (Clausen & Flatman, 1977; Juel, 1988; Clausen, 2003; Juel et al. 2014). Ouabain, an inhibitor of the Na⁺–K⁺ pump, antagonizes the ability of adrenaline, β‐agonists or DBcAMP to stimulate net Na⁺ efflux and K⁺ influx, lower intracellular [Na⁺] ([Na⁺]_i), raise intracellular [K⁺] ([K⁺]_i), and hyperpolarize the sarcolemma (Clausen & Flatman, 1977; McArdle & D'Alonzo, 1981; Clausen, 2003). The PKA‐mediated phosphorylation of phospholemman (FXYD1) (Fig. 2), a regulatory component of the Na⁺–K⁺ pump complex (Bibert et al. 2008; Bers & Despa, 2009), causes stimulation of pump activity by increasing its affinity for intracellular Na⁺ without changing the maximal pump transport (Clausen, 2003; Bibert et al. 2008). Details for this process have recently emerged (Bibert et al. 2008; Bers & Despa, 2009) showing that unphosphorylated FXYD1 inhibits Na⁺–K⁺ pump activity by reducing its affinity for intracellular Na⁺. β‐Adrenergic phosphorylation of FXYD1 subsequently releases this inhibition to increase Na⁺–K⁺ pump activity. Another process stimulated by adrenaline/β‐agonists is the Na⁺–K⁺–2Cl⁻‐cotransporter (NKCC) which contributes about one‐third of the isoprenaline (30 μm)‐induced K⁺ uptake in resting soleus and plantaris muscles of rats (Wong et al. 2001; Gosmanov & Thomason, 2002). This effect requires phosphorylation and activation of extracellular signal‐regulated kinase 1/2 and the mitogen‐activated protein kinase pathway (Gosmanov & Thomason, 2002). Various sarcolemmal ion channel activities may also be modulated (Bers & Despa, 2009). For example, the voltage‐activated Na⁺ channel, which is responsible for the upstroke of the action potential, is phosphorylated (Yang & Barchi, 1990), yet high concentrations of isoprenaline/DBcAMP inhibit maximal Na⁺ currents in rat fast‐twitch fibres (Desaphy et al. 1998).

Resting membrane potential (E _M) and action potentials. Adrenergic agents/DBcAMP hyperpolarize the sarcolemma of non‐fatigued rodent fibres with the resting E _M increasing by 2–9 mV (Clausen & Flatman, 1977; McArdle & D'Alonzo, 1981; Juel, 1988; Cairns & Dulhunty, 1994). This occurs with adrenaline levels starting from 6 nm (Clausen & Flatman, 1977). This hyperpolarization appears to follow stimulation of the Na⁺–K⁺ pump but with a contribution from the NKCC. These processes raise [K⁺]_i and lower interstitial [K⁺]_o to increase the K⁺ equilibrium potential and, along with a greater electrogenic Na⁺–K⁺ pump current (Clausen, 2003), causes an increased resting E _M. There are also reports showing that high concentrations of adrenaline/β‐agonists (0.5–50 μm) exert several positive effects on intracellularly recorded action potentials in rodent muscle. These include the overshoot increasing by 4–11 mV (McArdle & D'Alonzo, 1981; Head & Ha, 2011), the overshoot remaining higher within a brief train of action potentials (Head & Ha, 2011), the waveform broadening (McArdle & D'Alonzo, 1981; Head & Ha, 2011), and conduction velocity increasing (Kӧssler et al. 1991). Such effects are small and inconsistent in different mammalian muscle types when assessed using extracellular recordings of action potentials (M‐wave) (Bowman, 1980). Similarly, β‐agonists are without detectable effect on the surface M‐wave in human muscles (Marsden & Meadows, 1970; Crivelli et al. 2011, 2013; Crivelli & Maffiuletti, 2014). Our conclusion is that the muscle action potential can be influenced in a manner that could increase/maintain force at least in rodent muscles. The β‐agonist‐induced heightening of action potential overshoot occurs independently of changes to resting E _M in non‐fatigued muscle (McArdle & D'Alonzo, 1981) but it may follow the Na⁺–K⁺ pump‐induced decline of [Na⁺]_i since this elevates the Na⁺ equilibrium potential (McArdle & D'Alonzo, 1981; Juel, 1988).

Despite these altered sarcolemmal processes neither the Na⁺–K⁺ pump nor NKCC contribute to the positive inotropic effect in non‐fatigued muscle. Terbutaline potentiates force when Na⁺–K⁺ pumps are blocked with ouabain in rat soleus (Cairns & Dulhunty, 1993 b) and any effect via altered action potential characteristics is discounted given that β‐agonists/DBcAMP augment submaximal K⁺ contractures (Oota & Nagai, 1977; Cairns & Dulhunty, 1993 b). K⁺ contractures involve bypassing action potentials to directly activate the voltage sensors of excitation–contraction (E–C) coupling (García et al. 1990; Avila et al. 2007). In fact, all force potentiation in non‐fatigued muscle is accounted for by processes stimulated beyond the action potential (Cairns & Dulhunty, 1993 b) which must include myofilament processes and/or E–C coupling processes. Nevertheless, β‐adrenergic modulation of sarcolemmal processes is important in other situations. For example, β‐agonists restore force in K⁺‐depressed muscle by stimulating the Na⁺–K⁺ pump to improve sarcolemmal excitability (Clausen, 2003; Cairns et al. 2011), and Na⁺–K⁺ pump stimulation may contribute to improved fatigue‐resistance in vitro (Juel, 1988).

Myofilament processes

Two myofilament proteins are phosphorylated during β‐adrenoceptor activation in cardiac muscle, namely troponin‐I (which reduces myofilament Ca²⁺ sensitivity with less Ca²⁺ binding to troponin‐C), and C‐protein (which increases maximal crossbridge cycling) (Wolska et al. 1996; Gruen et al. 1999; Matsuba et al. 2009), but neither protein is phosphorylated in skeletal muscle (Stull & High, 1977; Gruen et al. 1999; Matsuba et al. 2009), nor is myosin light chain kinase (Manning & Stull, 1982). Functional testing for β‐adrenergic effects on myofilament processes requires investigation of the force–[Ca²⁺] relationship. Such experiments on intact fibres from fast‐twitch or slow‐twitch muscles from rodents show that terbutaline has no effect on myofibrillar Ca²⁺ sensitivity or maximum Ca²⁺‐activated force (reflecting maximum crossbridge activity) (Cairns et al. 1993; Ha et al. 1999). Other experiments on chemically skinned mammalian fibres (where only the contractile apparatus remains) confirm that there is an unchanged Ca²⁺ sensitivity and maximum Ca²⁺‐activated force with cAMP/PKA (Fabiato & Fabiato, 1978; Ha et al. 1999; Matsuba et al. 2009). Hence, in contrast to cardiac muscle, the β‐adrenergic effects on peak force or relaxation do not involve altered myofilament processes in skeletal muscle.

Excitation–contraction coupling processes

The initial finding of a β‐adrenergic effect on intracellular Ca²⁺ transients was with terbutaline (10 μm) in single, intact fast‐twitch fibres from mice (Fig. 3 A) where peak myoplasmic [Ca²⁺] ([Ca²⁺]_i) was augmented for unfused and maximal tetani (Cairns et al. 1993). The magnitude of this effect increased with higher stimulation frequencies. An increased peak tetanic [Ca²⁺]_i was subsequently observed with terbutaline in slow‐twitch fibres from rats (Ha et al. 1999) and in frog twitch fibres (Bruton et al. 1996). Moreover, an increased peak twitch [Ca²⁺]_i occurs in rodent hindlimb fibres with β‐agonists (Ha et al. 1999; Prakash et al. 1999; Andersson et al. 2012) or DBcAMP (Liu et al. 1997; Prakash et al. 1999). Notably, a basal level of phosphorylation had already augmented peak twitch [Ca²⁺]_i and restricted the effects of further phosphorylation with DBcAMP (Liu et al. 1997). This indicates that several phosphorylation processes can influence Ca²⁺ transients. The increase of peak [Ca²⁺]_i in these studies on isolated rodent muscle fibres at room temperature was typically 20–35% initial which is somewhat less than for cardiac muscle where the magnitude depends on the extracellular [Ca²⁺] (Kurihara & Konishi, 1987; Wolska et al. 1996; Li et al. 2000). Concentration–response effects have only been studied in myocytes from neonatal limb muscles where peak twitch [Ca²⁺]_i was increased with salbutamol concentrations over 10–100 nm (Prakash et al. 1999). Furthermore, an elegant study by Rudolf et al. (2006) made important progress by showing directly that isoprenaline (10 μm) facilitated the emptying of Ca²⁺ from the SR of mouse tibialis anterior muscle in situ (Fig. 4 A and B), and this effect was greater with tetanic than twitch stimulation. Thus the increased Ca²⁺ transient amplitude involves greater Ca²⁺ efflux from the SR and can occur at body temperature.

A key finding is that terbutaline simultaneously increases peak [Ca²⁺]_i and force (Fig. 3) in a manner which causes an upwards shift along the force–[Ca²⁺]_i relationship (Cairns et al. 1993; Ha et al. 1999). This establishes that the positive inotropic effect is fully accounted for by a modulation of processes involved with E–C coupling. Based on phosphorylation studies (Fig. 2) we envisioned that the β‐adrenergic effects on Ca²⁺ handling may comprise the following proteins: (1) the voltage‐activated Ca²⁺ channel (i.e. L‐type/Ca_v1.1 channel, dihydropyridine receptor, DHPR) located in t‐tubular membranes; (2) phospholamban (PLB), a protein associated with the SR Ca²⁺ pump (i.e. Ca²⁺‐ATPase, SERCA2); and/or (3) the Ca²⁺ release channel/ryanodine receptor of the terminal cisternae of SR (i.e. RyR1 isoform).

Voltage‐activated Ca ^{2 +}‐channels and trans‐sarcolemmal Ca ^{2 +} influx. β‐Adrenergic modulation of the DHPR may entail greater charge movement – the signal arising from the DHPR as the voltage sensor for E–C coupling (García et al. 1990; Avila et al. 2007) and/or enhanced Ca²⁺ influx to either trigger further Ca²⁺ release from the SR or load the SR with Ca²⁺ for subsequent release. The first possibility is discounted since neither isoprenaline nor PKA increase charge movement (García et al. 1990) but a boosted Ca²⁺ influx seems feasible. Indeed, β‐agonists/PKA amplify slow Ca²⁺ currents in amphibian (Arreola et al. 1987; García et al. 1990) and rodent fibres (García et al. 1990; Johnson et al. 1994, 2005). Moreover, PKA‐induced phosphorylation of the DHPR shifts the voltage dependence of channel activation during repetitive depolarizations to more negative E _M, which considerably magnifies the Ca²⁺ influx (Johnson et al. 1994, 2005). It seems plausible that this phenomenon occurs during tetanic stimulation. Such trans‐sarcolemmal Ca²⁺ influx does not directly increase the Ca²⁺ transient (Rudolf et al. 2006) but may lead to increased Ca²⁺ release from SR. Single DHPR channel recordings divulge that PKA elicits an increased channel open time, a leftwards shift in voltage dependence of activation, a decreased rate of inactivation and an increased channel availability (Mundiña‐Weilenmann et al. 1991). These combined actions adequately explain the amplified macroscopic Ca²⁺ current. Interestingly, it appears that PKA exerts greater facilitation of Ca²⁺ currents in amphibian than in mammalian fibres (García et al. 1990), and stimulates an additional small but fast Ca²⁺ current in amphibian fibres (Arreola et al. 1987). β‐Agonists also elevate resting [Ca²⁺]_i in amphibian fibres (Bruton et al. 1996), which is indicative of Ca²⁺ influx, but not in mammalian fibres (Cairns et al. 1993; Liu et al. 1997; Ha et al. 1999; Prakash et al. 1999). A third proposal is that phosphorylation of the DHPR results in a protein–protein interaction with the RyR1 to potentiate Ca²⁺ release, and this idea has some but not extensive backing (Lu et al. 1995).

An enhanced trans‐sarcolemmal Ca²⁺ influx is crucial for the β‐adrenergic potentiation of cardiac twitches (Kurihara & Konishi, 1987; Li et al. 2000; Bers & Despa, 2009), but its importance for skeletal muscle varies between muscles. First, Ca²⁺ channel blockers antagonize the β‐adrenergic potentiation of force in amphibian fibres (Williams & Barnes, 1989 b) and rat diaphragm (Varagić & Kentera, 1978; Todorović et al. 2006), but not in rat limb muscle (Cairns & Dulhunty, 1993 b). Also, the positive inotropic effect is magnified in the presence of Ca²⁺ ionophores in rat diaphragm (Varagić et al. 1979). Second, the β‐adrenergic force potentiation is abolished or attenuated with Ca²⁺‐free solutions in amphibian fibres (Arreola et al. 1987; Williams & Barnes, 1989 b) or rat diaphragm (Varagić & Kentera, 1978, Varagić et al. 1979). Third, in quiescent muscles, when sarcolemmal DHPR channels remain closed, the force potentiation is abolished in amphibian fibres (Arreola et al. 1987; Williams & Barnes, 1989 b) but is well developed in mouse fast‐twitch fibres (Cairns et al. 1993; Rudolf et al. 2006). Hence a trans‐sarcolemmal Ca²⁺ influx is likely to contribute to force potentiation in diaphragm and amphibian fibres, but not in mammalian limb muscles.

Phospholamban and SR Ca ^{2 +} uptake. A larger Ca²⁺ transient may involve phosphorylation of PLB, which stimulates Ca²⁺ uptake into SR for subsequent release in cardiac muscle (Wolska et al. 1996; Li et al. 2000; Bers & Despa, 2009). However, PLB exists only in the SR of slow‐twitch and not fast‐twitch mammalian muscle (Liu et al. 1997; Vangheluwe et al. 2005) and phosphorylated PLB only increases the Ca²⁺ uptake rate for SR vesicles from slow‐twitch fibres of mammals (Kirchberger & Tada, 1976) and humans (Salviati et al. 1982). Mechanistic work (Song et al. 2004; Bibert et al. 2008; Bers & Despa, 2009) shows that unphosphorylated PLB inhibits the Ca²⁺ pump by lowering its affinity for Ca²⁺. Phosphorylation of PLB then reverses this inhibition to increase Ca²⁺ activation of the pump. Crucial observations relating to contraction are that β‐agonists/DBcAMP potentiate force and increase the amplitude of Ca²⁺ transients in fast‐twitch fibres and soleus fibres knockout for PLB (Cairns et al. 1993; Liu et al. 1997). Also the isoprenaline‐induced twitch potentiation is identical in soleus muscles that are wild‐type, overexpressing PLB (Song et al. 2004), or knockout for PLB (Slack et al. 1997). These combined findings confirm that the β‐adrenergic potentiation of force is not mediated via PLB in skeletal muscle.

An alternative hypothesis is that enhanced Ca²⁺ loading of SR occurs independently of PLB. However, maximal Ca²⁺ loading is unchanged with terbutaline/PKA in skinned slow‐twitch fibres from rats or humans (Salviati et al. 1982; Ha et al. 1999). In contrast, a raised basal [Ca²⁺]_SR has recently been shown with isoprenaline in mouse tibialis anterior muscle which seems to account for much of the increased Ca²⁺ release (Fig. 4 A and B). A similar Ca²⁺‐loading effect was inferred long ago for cat fast‐twitch caudofemoralis fibres that were chemically skinned and exposed to cAMP (Fabiato & Fabiato, 1978), and for split frog fibres (Gonzalez‐Serratos et al. 1981). Increased SR Ca²⁺ loading in fast‐twitch fibres does not involve PLB but may require phosphorylation of a 95 kDa protein, which could be the Ca²⁺ pump itself (Schwartz et al. 1976). Interestingly, phosphorylase kinase increases Ca²⁺ pump activity (Schwartz et al. 1976), which implies that the well‐known β‐adrenergic stimulation of muscle glycogenolysis/glycolysis (Chasiotis et al. 1983; Hostrup et al. 2014 b) may be intricately linked to SR Ca²⁺ loading (Schwartz et al. 1976). A role for this process in the positive inotropic effect in fast‐twitch muscle appears plausible but needs greater experimental support.

The β‐adrenergic phosphorylation of PLB is, nevertheless, functionally important since it is postulated to mediate the faster relaxation in slow‐twitch fibres, as in cardiac muscle (Wolska et al. 1996; Li et al. 2000). β‐Agonist exposure to slow‐twitch or transgenic muscles overexpressing PLB abbreviates relaxation (Cairns & Dulhunty, 1993 a; Ha et al. 1999; Song et al. 2004), whereas this positive lusitropic effect does not happen when PLB is absent (Cairns & Dulhunty, 1993 a; Slack et al. 1997). Furthermore, in soleus fibres with PLB present β‐agonists/DBcAMP accelerate the decline of Ca²⁺ transients (Liu et al. 1997; Ha et al. 1999) and PKA accelerates SR Ca²⁺ uptake (Kirchberger & Tada, 1976). Similar enhanced Ca²⁺ sequestering effects occur in SR from human quadriceps muscle after terbutaline exposure (Hostrup et al. 2014 b). In contrast, the rate of decline of [Ca²⁺]_i is unaltered in mouse fibres lacking PLB (Cairns et al. 1993; Liu et al. 1997). Taken together, it seems that the PKA‐induced phosphorylation of PLB stimulates the Ca²⁺ pump to enhance Ca²⁺ uptake into SR, speeds the decline of the Ca²⁺ transient, and hence abbreviates relaxation in slow‐twitch fibres.

The negative inotropic effect with β‐agonists is most apparent with unfused tetani in slow‐twitch muscles of larger mammals and humans (Fig. 1 B, Table 1), when there is also faster relaxation and reduced fusion between stimuli (Bowman & Zaimis, 1958; Bowman & Nott, 1970; Al‐Jeboory & Marshall, 1978). Hence it is tempting to speculate that the force depression may be due to phosphorylation of PLB and greater SR Ca²⁺ pumping. This effect potentially lowers the amplitude of the Ca²⁺ transient which depends on the rates of both SR Ca²⁺ release and uptake. The mechanical expression of cAMP could well hinge on the relative contributions from phosphorylated PLB (on SR Ca²⁺ pump) and those processes that stimulate SR Ca²⁺ release (Ha et al. 1999). In line with this idea the β‐adrenergic depression of force is often absent in rat and mouse soleus muscles (Table 1), which also have low levels of PLB/Ca²⁺ pumps (Vangheluwe et al. 2005). This feature may permit the β‐adrenergic processes which augment force to dominate.

Ca ^{2 +} release channel and SR Ca ^{2 +} release. The β‐adrenergic facilitation of Ca²⁺ release may involve modulation of the SR Ca²⁺ release channel/ryanodine receptor (i.e. RyR1 isoform), as discussed for cardiac muscle (Capes et al. 2011). This could occur indirectly via provision of ATP, consequent to stimulation of glycogenolysis/glycolysis (Chasiotis et al. 1983; Hostrup et al. 2014 b), since ATP modulates RyR1 channel activity (Sonnleitner et al. 1997). However, the β‐adrenergic force potentiation endures in the presence of iodoactetate, which blocks glycolysis (Cairns & Dulhunty, 1993 b), thus negating this possibility. A scenario, with mounting experimental support, is that PKA phosphorylates and directly modulates the RyR1 channel. First, Meissner (1984) showed that cAMP enhances the rate of Ca²⁺ efflux from isolated SR vesicles from rabbit muscle via the so‐called Ca²⁺‐induced Ca²⁺ release mechanism. Second, forskolin, which increases [cAMP]_i (Fig. 2), stimulates Ca²⁺‐induced Ca²⁺ release in cells when AKAP is colocalized with RyR1 (Ruehr et al. 2003). Third, pretreatment with either caffeine or ryanodine, which act on the RyR1, antagonizes the salbutamol‐induced potentiation of Ca²⁺ transients in myocytes (Prakash et al. 1999). Fourth, Andersson et al. (2012) have now provided water‐tight evidence that the isoprenaline‐induced increase of peak twitch [Ca²⁺]_i in mouse fast‐twitch fibres requires RyR1 phosphorylation. They exchanged the serine phosphorylation site on RyR1 with alanine, which cannot be phosphorylated, with the outcome being that isoprenaline no longer augmented the Ca²⁺ transient. Fifth, key findings on animal muscles have now been extended to human muscles, with the RyR1 being phosphorylated (Marx et al. 2001). Also Hostrup et al. (2014 b) showed that high‐dose terbutaline induced a 15% increase in SR Ca²⁺ release rate in homogenates from human quadriceps muscle. Hence it is undeniable that the β‐adrenergic effects on SR Ca²⁺ handling seen in animals also occur in humans, and it involves modulation of the RyR1.

Details of the mechanism by which PKA regulates skeletal muscle RyR1 function have been developed from studies on isolated RyR1 and single channel currents. Phosphorylation of the RyR1 facilitates Ca²⁺ release (gated by Ca²⁺) by making the RyR1 channel insensitive to block with Mg²⁺ (Hain et al. 1994; Mayrleitner et al. 1995). In addition, PKA enhances RyR1 Ca²⁺ currents that are gated by ATP in the absence of Ca²⁺ (Sonnleitner et al. 1997). The RyR1 macromolecular protein complex incorporates PKA and a regulatory FK‐506 binding protein (FKBP12) which stabilizes the RyR1 channel in its closed state (Reiken et al. 2003). Phosphorylation of the RyR1 stimulates its channel activity (gated by Ca²⁺) by releasing FKBP12 which increases the probability of RyR1 channel opening, whilst decreasing open and closed dwell times (Reiken et al. 2003).

The β‐adrenergic effects via the RyR1 appear intimately linked to force potentiation. Caffeine pretreatment abolishes the β‐adrenergic increase of both peak force (Cairns & Dulhunty, 1993 b) and peak [Ca²⁺]_i (Prakash et al. 1999), which vindicates the RyR1 in these effects. However, it is the well‐conceived and sophisticated experiments of Andersson et al. (2012) which provide unequivocal support for a modulatory role of the RyR1 since both the isoprenaline‐induced phosphorylation of the RyR1 and twitch potentiation were abolished in mouse fast‐twitch muscle when the serine phosphorylation site on the RyR1 was replaced with alanine.

Summary. The β‐adrenergic elevation of peak twitch [Ca²⁺]_i necessitates phosphorylation of the RyR1 to enhance SR Ca²⁺ release, presumably by sensitizing the Ca²⁺‐induced Ca²⁺ release mechanism. This process is also likely to increase peak tetanic [Ca²⁺]_i with contributions from trans‐sarcolemmal Ca²⁺ influx or Ca²⁺ loading of SR independently of PLB also being possible. The location on the sigmoidal force–[Ca²⁺]_i relationship determines the extent of the β‐adrenergic force potentiation. Small increases of peak [Ca²⁺]_i promote larger force increments when on the steep part of this relationship as for unfused tetani, and large increases of peak [Ca²⁺]_i cause smaller force increases when troponin nears saturation with Ca²⁺, as for maximal tetani.

Influence of catecholamines/β‐agonists during exercise or stress

Endogenous catecholamines

In order to understand the role of catecholamines released from the adrenal medulla or sympathetic nerves on human muscle performance it is imperative to know the catecholamine concentrations that muscles are exposed to. Plasma catecholamine levels increase with the intensity and duration of exercise (Zouhal et al. 2008) so that during maximal aerobic exercise the concentration of noradrenaline is 4–5 times greater than for adrenaline, i.e. ∼10 nm versus 2 nm, respectively (Zouhal et al. 2008). With supramaximal exercise these concentrations are exceeded with noradrenaline reaching 10–20 nm and adrenaline 5–10 nm (Harmer et al. 2000; Zouhal et al. 2008). Since both catecholamines bind to β₂‐adrenoceptors, a combined concentration of 30 nm could be regarded as an extreme physiological level for exercise. Given that the fight‐or‐flight or stress response has been equated to an adrenergic response the catecholamine levels in some stressful scenarios were identified. With risky sports, such as rock climbing or high ropes events, the adrenaline/noradrenaline levels reach 2–5 nm (Williams et al. 1978; Bunting & Gibbons, 2001) and slightly lower levels are achieved during laboratory stress testing (Bunting & Gibbons, 2001). Such catecholamine concentrations during exercise or stress would probably evoke submaximal [cAMP]_i (Al‐Jeboory & Marshall, 1978; Godinho & Costa, 2003) and therefore cause submaximal stimulation of glycogenolysis, the Na⁺–K⁺ pump or SR Ca²⁺ pump (Clausen & Flatman, 1977; Chasiotis et al. 1983; Slack et al. 1997). This information concurs with the view of Roatta & Farina (2013) that stress would not induce an increase of force, since the highest catecholamine levels achieved during stress (i.e. 5 nm) are unlikely to potentiate Ca²⁺ transients (Prakash et al. 1999) and evoke positive inotropic effects. However, the highest catecholamine concentrations during exercise (i.e. 30 nm) may facilitate some Ca²⁺ release (Prakash et al. 1999) to increase force. This information also implies that other endogenous physiological agents which elevate [cAMP]_i, such as calcitonin gene‐related peptide (CGRP) (Clausen, 2003; Avila et al. 2007; Rӧder et al. 2009), may act synergistically with catecholamines to improve muscle function, and this idea warrants further study.

Exogenous β‐agonists

Several studies with β‐agonists in humans show increased muscle force/power (Fig. 1 C and D), yet other studies are without significant effect (see Pluim et al. 2011). These inconsistencies can be explained if the β‐agonist concentrations are too low to permit significant positive inotropic effects. Certainly, a common therapeutic dose (8 mg) yields median plasma concentrations of 11–16 ng ml⁻¹, i.e. 60–90 nm (Elers et al. 2012), and with high individual variability (Elers et al. 2012). Plasma concentrations also generally peak 60–180 min after ingestion (Elers et al. 2012), which is when [cAMP]_i has presumably fallen due to extrusion from muscle (Chasiotis et al. 1983; Godinho & Costa, 2003; Rudolf et al. 2006) and this may attenuate β‐adrenergic effects. Hostrup and colleagues (2014 a,b) have now essentially resolved this issue by showing that very high‐dose terbutaline (15–20 mg, plasma concentration 17–22 ng ml, 0.08–0.1 μm), whether administered orally or nasally, does significantly augment quadriceps MVC strength. Although this 0.1 μm is much lower than the 1–10 μm used to potentiate SR Ca²⁺ release and force in animals (Table 1), it appears sufficient to augment Ca²⁺ release (Prakash et al. 1999) and actually does increase the SR Ca²⁺ release rate in humans (Hostrup et al. 2014 b). In addition, the endogenous phosphorylation of RyR1 (Hain et al. 1994; Liu et al. 1997) may limit the degree of further phosphorylation with exogenous β‐agonists (Liu et al. 1997) to mask or limit improvements to performance. During exercise any effect of β‐agonists will necessarily be superimposed on those of circulating catecholamines, and a basal phosphorylation of the RyR1 is also likely to occur with CGRP or Ca²⁺–calmodulin (Suko et al. 1993; Mayrleitner et al. 1995). Indeed CGRP can increase SR [Ca²⁺]_i release in some conditions (Avila et al. 2007). Greater understanding of the processes contributing to basal phosphorylation of human muscle in vivo is needed.

Conclusions and perspectives

β‐Agonists exert positive inotropic effects in human and animal skeletal muscle, but only when the concentrations are high enough. The cAMP–PKA pathway with lower [cAMP]_i stimulates muscle glycogenolysis, the Na⁺–K⁺ pump and SR Ca²⁺ pump, during exercise and stressful situations. The cellular machinery and mechanisms which enhance SR Ca²⁺ release to increase force/power are present in human muscle, but need to be activated with higher [cAMP]_i as occurs with very high β‐agonist concentrations. The β‐adrenergic processes which regulate SR Ca²⁺ handling are still not fully understood. More experiments and knowledge are needed on concentration–response relationships for many E–C coupling processes, including RyR1 function, Ca²⁺ loading of SR, trans‐sarcolemmal Ca²⁺ currents, and DHPR–RyR1 interactions. Studies on β‐adrenergic effects in human muscle tissue in vitro may provide enlightenment. Finally, experiments are needed on myoplasmic Ca²⁺ transients in relation to the negative inotropic effect seen in some slow‐twitch muscles.

Additional information

Competing interests

The authors have no conflict of interest that relates to the content of this article.

Funding

No sources of funding were used in order to prepare this article.

Biographies

Simeon Cairns (with cap) of the Auckland University of Technology in New Zealand, gained his PhD from the Australian National University under the supervision of Angela Dulhunty and Peter Gage. His research background in muscle physiology revolves around regulation of force production and mechanisms of fatigue from the single fibre level up to whole‐body exercise and sport performance. He is a former New Zealand table tennis representative and Olympic coach.

Fabio Borrani (right) works at the University of Lausanne in Switzerland. He obtained his PhD at the University of Montpellier I (France) in exercise physiology. His research is articulated around the field of exercise physiology and biomechanics. A main focus is to understand the mechanisms underlying fatigue in all its forms during exercise in humans.