Beta₂‐adrenoceptor agonist salbutamol increases protein turnover rates and alters signalling in skeletal muscle after resistance exercise in young men

Abstract

Key points

• Animal models have shown that beta₂‐adrenoceptor stimulation increases protein synthesis and attenuates breakdown processes in skeletal muscle. Thus, the beta₂‐adrenoceptor is a potential target in the treatment of disuse‐, disease‐ and age‐related muscle atrophy.

• In the present study, we show that a few days of oral treatment with the commonly prescribed beta₂‐adrenoceptor agonist, salbutamol, increased skeletal muscle protein synthesis and breakdown during the first 5 h after resistance exercise in young men.

• Salbutamol also counteracted a negative net protein balance in skeletal muscle after resistance exercise.

• Changes in protein turnover rates induced by salbutamol were associated with protein kinase A‐signalling, activation of Akt2 and modulation of mRNA levels of growth‐regulating proteins in skeletal muscle.

• These findings indicate that protein turnover rates can be augmented by beta₂‐adrenoceptor agonist treatment during recovery from resistance exercise in humans.

Abstract

The effect of beta₂‐adrenoceptor stimulation on skeletal muscle protein turnover and intracellular signalling is insufficiently explored in humans, particularly in association with exercise. In a randomized, placebo‐controlled, cross‐over study investigating 12 trained men, the effects of beta₂‐agonist (6 × 4 mg oral salbutamol) on protein turnover rates, intracellular signalling and mRNA response in skeletal muscle were investigated 0.5–5 h after quadriceps resistance exercise. Each trial was preceded by a 4‐day lead‐in treatment period. Leg protein turnover rates were assessed by infusion of [¹³C₆]‐phenylalanine and sampling of arterial and venous blood, as well as vastus lateralis muscle biopsies 0.5 and 5 h after exercise. Furthermore, myofibrillar fractional synthesis rate, intracellular signalling and mRNA response were measured in muscle biopsies. The mean (95% confidence interval) myofibrillar fractional synthesis rate was higher for salbutamol than placebo [0.079 (95% CI, 0.064 to 0.093) vs. 0.066 (95% CI, 0.056 to 0.075%) × h⁻¹] (P < 0.05). Mean net leg phenylalanine balance 0.5–5 h after exercise was higher for salbutamol than placebo [3.6 (95% CI, 1.0 to 6.2 nmol) × min⁻¹ × 100 g_{Leg Lean Mass} ⁻¹] (P < 0.01). Phosphorylation of Akt2, cAMP response element binding protein and PKA substrate 0.5 and 5 h after exercise, as well as phosphorylation of eEF2 5 h after exercise, was higher (P < 0.05) for salbutamol than placebo. Calpain‐1, Forkhead box protein O1, myostatin and Smad3 mRNA content was higher (P < 0.01) for salbutamol than placebo 0.5 h after exercise, as well as Forkhead box protein O1 and myostatin mRNA content 5 h after exercise, whereas ActivinRIIB mRNA content was lower (P < 0.01) for salbutamol 5 h after exercise. These observations suggest that beta₂‐agonist increases protein turnover rates in skeletal muscle after resistance exercise in humans, with concomitant cAMP/PKA and Akt2 signalling, as well as modulation of mRNA response of growth‐regulating proteins.

Key points

• Animal models have shown that beta₂‐adrenoceptor stimulation increases protein synthesis and attenuates breakdown processes in skeletal muscle. Thus, the beta₂‐adrenoceptor is a potential target in the treatment of disuse‐, disease‐ and age‐related muscle atrophy.

• In the present study, we show that a few days of oral treatment with the commonly prescribed beta₂‐adrenoceptor agonist, salbutamol, increased skeletal muscle protein synthesis and breakdown during the first 5 h after resistance exercise in young men.

• Salbutamol also counteracted a negative net protein balance in skeletal muscle after resistance exercise.

• Changes in protein turnover rates induced by salbutamol were associated with protein kinase A‐signalling, activation of Akt2 and modulation of mRNA levels of growth‐regulating proteins in skeletal muscle.

• These findings indicate that protein turnover rates can be augmented by beta₂‐adrenoceptor agonist treatment during recovery from resistance exercise in humans.

Introduction

Skeletal muscle encompasses ∼40% of body mass in lean individuals, making it the largest organ of the human body (Zurlo et al. 1990). Loss of muscle mass (muscle atrophy) can have critical consequences for contractile function and exercise capacity (Ryall et al. 2007) and may reduce quality of life and life expectancy (McLeod et al. 2016). During muscle atrophic conditions, muscle protein breakdown exceeds synthesis, resulting in a negative protein balance (Goldspink & Goldspink, 1977). Strategies that reduce muscle protein breakdown and/or increase synthesis are therefore an important area of research. Although resistance exercise training effectively promotes muscle hypertrophy and decelerates disuse‐ and age‐related muscle atrophy and weakness (Macaluso & De Vito, 2004), pharmacological compounds may be used to augment the response to exercise (Kumar et al. 2009; Egan & Zierath, 2013).

Skeletal muscle beta₂‐adrenoceptors are among potential therapeutic targets that have attracted interest in treatment of muscle atrophy and weakness (Lynch & Ryall, 2008; Joassard et al. 2013). Beta₂‐adrenoceptors are the most predominant subtype of adrenoceptors in skeletal muscle (Williams et al. 1984; Jensen et al. 2002), where they serve a crucial role in the adrenergic fight‐or‐flight response (Emrick et al. 2010; Andersson et al. 2012; Hostrup et al. 2014b b). Furthermore, beta₂‐adrenoceptors play a role in regulation of muscle protein turnover. Beta₂‐adrenoceptor knockout mice display lower muscle mass than their wild‐type peers (Hinkle et al. 2002) and beta₂‐adrenoceptor stimulation with selective agonists (beta₂‐agonists) increases muscle mass in mammals (Lynch & Ryall, 2008), including humans (Hostrup et al. 2015; Jessen et al. 2018). In rodents, beta₂‐agonists have also been shown to accelerate muscle recovery from injury (Beitzel et al. 2004; Church et al. 2014) and to reverse muscle atrophy associated with ageing (Ryall et al. 2007), cancer cachexia (Busquets et al. 2004) and muscular dystrophies (Harcourt et al. 2007; Gehrig et al. 2010). Moreover, studies in humans have shown that a few weeks of beta₂‐agonist treatment enhances muscle strength (Martineau et al. 1992; Hostrup et al. 2015, 2016) and preserves muscle function during disuse conditions when combined with resistance exercise (Caruso et al. 2004, 2005). Accordingly, beta₂‐agonists have been proposed as pharmacotherapy to prevent muscle atrophy and loss of muscle function in muscle wasting conditions (Signorile et al. 1995; Lynch & Ryall, 2008; Atherton & Szewczyk, 2011; Joassard et al. 2013) and to augment muscle adaptations to exercise training (Caruso et al. 2004, 2005; Hostrup et al. 2018; Jessen et al. 2018).

Beta₂‐agonists are widely used because of their application as first‐line treatment of the bronchoconstriction associated with asthma, exercise‐induced bronchoconstriction (Price et al. 2014) and chronic obstructive pulmonary disease (Barnes, 2005). Although beta₂‐agonists have been marketed for decades, information on the effect of these substances on skeletal muscle protein turnover in humans is lacking. Contradictory findings exist in that Robinson et al. (2010) found no effect of the non‐selective beta‐agonist isoproterenol on whole‐body and muscle protein synthesis, whereas a recent study showed that 7 days of treatment with selective beta₂‐agonist formoterol increased whole‐body protein synthesis (Lee et al. 2015). In mice, however, Koopman et al. (2010) observed that beta₂‐agonist only increased muscle protein synthesis after consecutive days of treatment and not after the first day. Thus, the therapeutic application of beta₂‐agonists in humans may involve several days of treatment before a net positive protein balance incurs (Koopman et al. 2010; Atherton & Szewczyk, 2011). Nevertheless, despite the advances made, no studies have investigated the potential of consecutive days of beta₂‐agonist treatment to improve muscle protein balance after exercise in humans.

In rodents, the muscle hypertrophic effect of beta₂‐agonist is mediated by increased protein synthesis (Maltin et al. 1989; Hesketh et al. 1992) and/or reduced breakdown (Busquets et al. 2004; Yimlamai et al. 2005), resulting in an overall net positive protein balance. The mechanisms underlying the growth‐promoting actions of beta₂‐agonists, however, are not entirely clear, although they involve modulation of various signalling pathways and gene programs that regulate muscle protein synthesis and proteolysis in rodents (Spurlock et al. 2006; Pearen et al. 2009; Koopman et al. 2010). Beta₂‐adrenergic signalling induces cAMP‐dependent activation of protein kinase A (PKA), Epac and extracellular signal‐regulated kinases 1/2 (Shi et al. 2007; Ohnuki et al. 2014), which have a wide range of downstream targets that regulate ribosomal translation processes and transcription of growth‐modulating genes (Spurlock et al. 2006; Pearen et al. 2009), including cAMP response element binding protein (CREB) (Hinkle et al. 2002), Akt, mammalian target of rapamycin (mTOR) and mitogen‐activated protein kinase (MAPK) (Kline et al. 2007; Koopman et al. 2010). In addition, several regulators of protein synthesis may be modulated by beta₂‐adrenergic signalling, including Akt‐effector Forkhead box protein O1 (FoxO1), a regulator of the atrophy‐related genes atrogin and MurF (Bodine & Baehr, 2014), and eEF2, a regulator of translational elongation. However, the myocellular signalling and mRNA response to beta₂‐agonists in relation to muscle protein synthesis and breakdown after exercise remain unexplored in humans.

The present study aimed to investigate the effect of 5 days of beta₂‐adrenoceptor stimulation with the selective beta₂‐agonist salbutamol on protein turnover of skeletal muscle following resistance exercise in young men. Secondary purposes were to elucidate associated changes in intracellular signalling and mRNA content of selected canonical beta₂‐adrenergic targets in skeletal muscle. We hypothesized that beta₂‐agonist treatment would increase protein synthesis and reduce breakdown, resulting in an overall net positive protein balance compared to placebo during recovery from resistance exercise.

Methods

Human subjects and ethics

Thirteen healthy trained young men volunteered to participate in the present study. Before inclusion in the study, subjects underwent a medical examination where resting blood pressure, heart rate and ECG of the subjects were measured. Furthermore, body composition was measured by dual‐energy X‐ray absorptiometry (Lunar DPX‐IQ, GE Healthcare, Chalfont St Giles, UK). Inclusion criteria were age 18–40 years and an active life‐style, defined as more than 3 h of physical activity per week. Exclusion criteria were smoking, chronic disease, allergy towards medication and the use of beta₂‐agonist or other prescription medication. Subjects were informed about risks and discomforts related to the different tests and procedures of the study. Each subject provided their written and oral informed consent prior to inclusion in the study. The study was approved by the Committee on Health Research Ethics of the Capital Region of Denmark (H‐1‐2012‐119) and performed in accordance with the standards set by the Declaration of Helsinki. The study was registered in ClinicalTrials.gov (NCT02551276).

Of the 13 subjects who were screened, 12 were included in the study and completed it (Fig. 1). The characteristics of the 12 subjects who completed the study are presented in Table 1.

Figure 1. Flow diagram.

Flow diagram providing details of the present study.

Table 1.

Subject characteristics (n = 12)

Age (years)	23.4 ± 3.8
Height (cm)	181.3 ± 5.6
Body mass (kg)	74.4 ± 9.2
Lean body mass (kg)	61.0 ± 5.4
Leg lean mass (kg)	17.8 ± 1.8

Values are the mean ± SD.

Study design

The study was designed as a randomized, double‐blinded, placebo‐controlled, cross‐over study. During two identical trials, subjects received either oral salbutamol or placebo. Each trial was preceded by a 4‐day lead‐in period with oral salbutamol (4 × 4 mg × day⁻¹) or placebo treatment because animal studies have shown that the effect of beta₂‐agonist on protein synthesis is evident after a few days of treatment (Koopman et al. 2010). The two trials were separated by 3–6 weeks to minimize potential confounding carry‐over effects of salbutamol (Le Panse et al., 2005). Prior to the first experimental trial, subjects met at the laboratory for two familiarizations to the resistance exercise protocol of the experimental trials.

Experimental protocol

An overview of the experimental protocol is illustrated in Fig. 2. After the 4 days of lead‐in treatment, subjects met in the morning after an overnight fast and received either oral salbutamol (6 × 4 mg) or placebo (same treatment as during lead‐in) with a standardized light meal low on protein and fat consisting of white bread with jam (energy: 369 kcal; protein: 12 g; carbohydrate: 67 g; fat: 3 g) and 400 mL of water. Subjects then rested in a bed in the supine position and catheters were inserted: one in the dorsal hand vein for tracer infusion, one in the brachial artery and one in the femoral vein during local anaesthesia (lidocaine without epinephrine, Xylocaine; AstraZeneca, Cambridge, UK) for arterial and venous blood sampling. A primed, continuous infusion of stable amino acid isotope [¹³C₆]‐phenylalanine (l‐phenylalanine, ring‐13C6, 99%, CLM‐1055‐MPT; Cambridge Isotope Laboratories, Inc., Tewksbury, MA, USA) was used for measurement of amino acid kinetics across the limb and incorporation of labelled phenylalanine into muscle. [¹³C₆]‐phenylalanine was dissolved in isotonic saline (0.9%) using a sterile procedure, filtered through disposable, sterile, non‐pyrogenic filters with 0.2 μm pore size (Minisart; Sartorius Stedim Biotech, Aubagne, France) and kept at 5°C until infusion. The priming dose of 8 μmol × kg⁻¹ lean body mass (LBM) labelled phenylalanine was dissolved in 20 mL saline and infused at once (1 min). The continuous infusion rate of labelled phenylalanine was 7 μmol × kg_LBM ⁻¹ × h⁻¹ dissolved in saline and infused with a constant rate throughout the trial.

Figure 2. Overview of the experimental protocol.

In a randomized placebo‐controlled, double‐blinded, cross‐over design, the study participants conducted two experimental trials (salbutamol vs. placebo) that were separated by 3–6 weeks. Filled circles indicate when arterial and venous blood samples were drawn. Filled triangles indicate when muscle biopsies of the vastus lateralis were collected. EX, resistance exercise.

After 90 min of [¹³C₆]‐phenylalanine infusion (i.e. to reach tracer steady‐state), subjects moved to a knee extensor resistance exercise model. Subjects then performed two sets of 10 repetition knee extensor exercise at an intensity corresponding to 50% of three repetition maximum, followed by eight sets of 12 repetitions of knee‐extensor exercise at an intensity corresponding to 12 repetition maximum (75 ± 11 kg) (mean ± SD) with 2 min of recovery between each set. If subjects failed to perform 12 repetitions in a given set, load was decreased for the following set. The mean load performed during the final set was 69 ± 12 kg (mean ± SD). Intensity and recovery time were duplicated for each subject during the two trials. After exercise, subjects remained inactive in a supine position for 5 h. Biopsies were obtained from the vastus lateralis muscle 0.5 and 5 h after resistance exercise. Brachial arterial and femoral venous blood samples were drawn in EDTA tubes (9 mL) prior to exercise as well as 0.5, 1, 2, 3, 4 and 5 h following exercise. Blood samples were kept at 5°C for 30 min before centrifugation at 5°C and 3200 g for 10 min, after which plasma was collected and stored at –80°C until analyses. Furthermore, 1 mL of arterial and venous blood were sampled in heparin tubes at same time points for determination of haematocrit using an ABL 800 flex (Radiometer, Copenhagen, Denmark). Prior to exercise as well as 0.5, 1, 2, 3, 4 and 5 h following exercise, femoral arterial blood flow was measured with ultrasound Doppler (Vivid E9; GE Healthcare, Brøndbyvester, Denmark) equipped with a linear probe operating at an imaging frequency of 8 MHz and Doppler frequency of 3.1 MHz, as described previously (Nyberg et al. 2014).

Subjects were asked to refrain from caffeine, nicotine and alcohol 24 h before each trial, as well as from exercise 48 h before each trial.

Study drugs

Salbutamol (Ventolin, 4 mg tablets; GlaxoSmithKline, London, UK) and identically looking placebo (lactose monohydrate/starch) were delivered by the hospital pharmacy of Copenhagen. Beta₂‐adrenoceptors were stimulated with the highly selective beta₂‐agonist salbutamol (Baker, 2010), which has a duration of action of 6–8 h and a plasma elimination half‐life of 3–4 h (Rosen et al. 1986; Jacobson et al. 2015). Salbutamol concentrations peak systemically 1 h 30 min to 3 h after oral administration (Hostrup et al. 2014a). The dose administered during the lead‐in period (16 mg × day⁻¹) was based on studies showing significant effect of daily treatment with oral salbutamol (16 mg × day⁻¹) on muscle strength (Martineau et al. 1992; Caruso et al. 2004). The increase in dose to 24 mg of oral salbutamol during the experimental day was because of potential desensitization of the beta₂‐adrenoceptors during the lead‐in treatment period. Drugs were administered in a double‐blinded manner. Randomization was conducted in SPSS, version 24 (IBM Corp., Armonk, NY, USA) by personnel who did not take part in any of the experimental procedures or data analyses. To ensure a drug compliance of 100% during the 4‐day lead‐in period, subjects met at the laboratory in the morning or noon and ingested the study drugs during supervision. Eight of the 12 subjects experienced common side effects of salbutamol during the first 2 days of treatment, including tremors (n = 7) and palpitations (n = 6).

Dual‐energy X‐ray absorptiometry

Subjects laid in the scanner in supine position undressed for 20 min before the scan. To reduce variation, two scans at medium speed were performed in accordance with the manufacturer's guidelines. The scanner was calibrated before the scan using daily calibration procedures (Lunar ‘System Quality Assurance’). All scans were conducted by the same hospital technician.

Muscle biopsies

Muscle biopsies were obtained from the vastus lateralis using a 4 mm Bergström biopsy needle (Stille, Stockholm, Sweden) with suction (Bergström, 1975). Before biopsies were sampled, two incisions were made in the skin at the belly of the vastus lateralis muscle during local anaesthesia (2 mL of lidocaine without epinephrine, Xylocaine 20 mg × mL⁻¹; Astra Zeneca). After sampling, the muscle biopsy was cleaned from visible blood, connective tissue and fat and immediately frozen in liquid nitrogen. Biopsies were stored in cryo tubes at –80°C until analyses.

Leg and muscle protein turnover rates

The influence of beta₂‐agonist on muscle protein turnover and myofibrillar protein fractional synthesis rates (FSR) were measured 0.5 to 5 h during recovery from exercise by infusion of stable isotope‐labelled phenylalanine, collection of arteriovenous blood samples and muscle biopsies from the vastus lateralis muscle, and measurement of femoral blood flow (Fig. 2).

Femoral arteriovenous plasma phenylalanine enrichment and concentration were measured using 400 μL of plasma with a known amount of [U‐¹³C₉]‐phenylalanine added as internal standard. Samples were derivatized using N‐methyl‐N‐(tert‐butyldimethylsilyl) trifluoroacetamide + 1% tert‐butyl‐dimethylchlorosilane (Regis Technologies, Morton Grove, IL, USA) and analysed on a triple‐stage quadrupole‐mass spectrometer via gas chromatography‐tandem mass spectrometry (GC‐MS/MS) (TSQ Quantum; Thermo Scientific, San Jose, CA, USA), as described previously (Holm et al. 2014).

Muscle specimens of ∼20 mg wet weight were homogenized using a FastPrep 120A‐230 homogenizer (Thermo Savant, Holbrook, NY, USA) in 1.5 mL of ice‐cold Milli‐Q saline water (Merck‐Millipore, Burlington, MA, USA) and after a spin (5500 g for 10 min at 5°C), the supernatant containing the muscle free amino acids was transferred to a new vial. Muscle free phenylalanine enrichment was then measured by GC‐MS/MS in the same way as described for the plasma phenylalanine enrichment. The pellet from the spin was added a Tris‐buffer (pH 7.4, containing 2 mm EGTA chelating agent, 0.5% Triton‐X100 and 0.25 m sucrose), homogenized once again, left 3 h at 5°C and spun (800 g for 20 min at 5°C) to pellet all structural proteins. Subsequently, myofibrillar proteins were dissolved in a 0.7 m KCl and 0.1 m Na₄P₂O₇‐buffer and, after overnight incubation at 5°C and a subsequent spin (1600 g for 20 min at 5°C), the supernatant containing the myofibrillar proteins were transferred to other vials. The myofibrillar proteins were denatured by adding 2.3 volumes of ethanol. After 2 h of incubation at 5°C, vials were spun (1600 g for 20 min at 5°C) to pellet the myofibrillar proteins. After a wash with 70% ethanol, proteins were hydrolysed in 6 m HCl at 110°C overnight, and the ratio of ¹³CO₂ and ¹²CO₂ from N‐acetyl n‐propyl (NAP)‐derivatized phenylalanine was analysed using a GC‐combustion‐isotope ratio mass spectrometer (Finnigan Delta Plus, Bremen, Germany), as described previously (Holm et al. 2014).

Calculations of phenylalanine kinetics across the leg and muscle, as well as muscle protein turnover parameters, were based on two‐ and three‐pool modelling and myofibrillar FSR on a direct incorporation model (Wolfe & Chinkes, 2005; Smith et al. 2015). All calculations are based on phenylalanine enrichment as mole percent excess (MPE) or atomic percent excess, and phenylalanine concentration is the total concentration (i.e. unlabelled and labelled phenylalanine). Plasma flow, derived from blood flow and haematocrit, were used in the calculations. All phenylalanine kinetic values are expressed as nmol × min⁻¹ × 100 g leg lean mass (LLM)⁻¹. LLM was derived from the dual‐energy X‐ray absorptiometry scan. Models and calculations applied in the present study are in accordance with those described previously (Biolo et al. 1995; Wolfe & Chinkes, 2005; Smith et al. 2015):

Two‐ and three‐pool models shared calculations

$$\begin{array}{c}\hfill \mathrm{Leg}\mathrm{plasma}\mathrm{flow}(\mathrm{LPF})\\ & & =\mathrm{leg}\mathrm{blood}\mathrm{flow}\times (1-\mathrm{haematocrit})\hfill \end{array}$$

$$\mathrm{Delivery}\mathrm{to}\mathrm{the}\mathrm{leg}(F_{\mathrm{in}})=C_{\mathrm{A}}\times \mathrm{LPF}$$

$$\mathrm{Output}\mathrm{from}\mathrm{the}\mathrm{leg}(F_{\mathrm{out}})=C_{\mathrm{V}}\times \mathrm{LPF}$$

$$\mathrm{Leg}\mathrm{net}\mathrm{balance}(\mathrm{leg}\mathrm{NB})=(C_{\mathrm{A}}-C_{\mathrm{V}})\times \mathrm{LPF}$$

where C _A and C _V are arterial and venous phenylalanine concentration, respectively.

Two‐pool model calculations

$$\begin{array}{c}\hfill \mathrm{Rate}\mathrm{of}\mathrm{disappearance}\mathrm{in}\mathrm{the}\mathrm{leg}(\mathrm{leg}{R}_{\mathrm{d}})\\ & & =(C_{\mathrm{A}}\times E_{\mathrm{A}}-C_{\mathrm{V}}\times E_{\mathrm{V}})\times \mathrm{LPF}/E_{\mathrm{A}}\hfill \end{array}$$

$$\mathrm{Rate}\mathrm{of}\mathrm{appearance}\mathrm{in}\mathrm{the}\mathrm{leg}(\mathrm{leg}{R}_{\mathrm{a}})=\mathrm{leg}{R}_{\mathrm{d}}-\mathrm{leg}\mathrm{NB}$$

where E _A and E _V are arterial and venous phenylalanine enrichment, respectively.

Three‐pool model calculations

$$\begin{array}{c}\hfill \mathrm{Muscle}\mathrm{inward}\mathrm{transport}(F_{\mathrm{M},\mathrm{A}})\\ & & =((C_{\mathrm{V}}\times ((E_{\mathrm{M}}-E_{\mathrm{V}})/(E_{\mathrm{A}}-E_{\mathrm{M}})))+C_{\mathrm{A}})\times \mathrm{LPF}\hfill \end{array}$$

$$\begin{array}{c}\hfill \mathrm{Muscle}\mathrm{outward}\mathrm{transport}(F_{\mathrm{V},\mathrm{M}})\\ & & =((C_{\mathrm{V}}\times ((E_{\mathrm{M}}-E_{\mathrm{V}})/(E_{\mathrm{A}}-E_{\mathrm{M}})))+C_{\mathrm{V}})\times \mathrm{LPF}\hfill \end{array}$$

$$\mathrm{Arteriovenous}\mathrm{shunting}(F_{\mathrm{V},\mathrm{A}})=F_{\mathrm{in}}-F_{\mathrm{M},\mathrm{A}}$$

$$\begin{array}{c}\hfill \mathrm{Muscle}\mathrm{protein}\mathrm{breakdown}\mathrm{estimate}(F_{\mathrm{M},\mathrm{O}})\\ & & =F_{\mathrm{M},\mathrm{A}}\times (E_{\mathrm{A}}/E_{\mathrm{M}}-1)\hfill \end{array}$$

$$\begin{array}{c}\hfill \mathrm{Muscle}\mathrm{protein}\mathrm{synthesis}\mathrm{estimate}(F_{\mathrm{O},\mathrm{M}})\\ & & =F_{\mathrm{M},\mathrm{O}}+\mathrm{leg}\mathrm{NB}\hfill \end{array}$$

where E _M is muscle intracellular phenylalanine enrichment.

Direct incorporation model calculations for myofibrillar protein synthesis

$$\begin{array}{c}\hfill \mathrm{Fractional}\mathrm{synthesis}\mathrm{rate}(\mathrm{FSR})\\ & & =((E_{\mathrm{P}2}-E_{\mathrm{P}1})/(E_{\mathrm{precursor}}\times (T_2-T_1)))\times 100\hfill \end{array}$$

where E _P1 and E _P2 are the myofibrillar product enrichment at time point 0.5 and 5 h, respectively, E _precursor is the muscle free phenylalanine enrichment (or venous or arterial plasma phenylalanine enrichment), and T ₁ and T ₂ are the specific time points at 0.5 and 5 h, respectively. Myofibrillar FSR is expressed as % × h⁻¹.

Protein phosphorylation in muscle homogenate lysates

Protein phosphorylation was determined by western blotting, as described previously (Thomassen et al. 2016). In short, ∼1.5 mg freeze dried muscle tissue was homogenized (Qiagen Tissuelyser II; Retsch GmbH, Haan, Germany) in a fresh batch of buffer containing (in mm): 10% glycerol, 20 Na‐pyrophosphate, 150 NaCl, 50 Hepes (pH 7.5), 1% NP‐40, 20 β‐glycerophosphate, 2 Na₃VO₄, 10 NaF, 2 phenylmethylsulphonyl fluoride, 1 EDTA (pH 8), 1 EGTA (pH 8), 10 μg × mL⁻¹ Aprotinin, 10 μg × mL⁻¹ Leupeptin and 3 benzamidine. Samples were rotated end‐over‐end for 1 h at 4°C and centrifuged at 18,320 g for 20 min at 4°C to exclude non‐dissolved structures and the supernatant (lysate) was used for further analyses. Total protein concentration in each sample was determined by a BSA standard kit (Thermo Fisher Scientific, Hvidovre, Denmark) and samples were mixed with 6 × Laemmli buffer (7 mL of 0.5 m Tris‐base, 3 mL of glycerol, 0.93 g of dithiothreitol, 1 g of SDS and 1.2 mg of bromophenol blue) and ddH₂0 to reach equal protein concentration before protein content was determined by western blotting.

An equal amount of total protein was loaded in each well of precast gels (Bio‐Rad, Hercules, CA, USA). All samples from each subject were loaded on the same gel with a mixed human muscle standard lysate loaded in two different wells used for normalization. Analysis of phosphorylated proteins and corresponding total protein were performed on separate gels. Proteins were separated according to their molecular weight by SDS page gel electrophoresis and semi‐dry transferred to a polyvinylidene fluoride membrane (Bio‐Rad). The membranes were blocked in either 2% skimmed milk or 3% BSA in Tris‐buffered saline including 0.1% Tween‐20 before overnight incubation in primary antibody at 4°C and a subsequent 1 h incubation in HRP conjugated secondary antibody at room temperature. The bands were visualized with ECL (Merck‐Millipore) and recorded with a digital camera (ChemiDoc MP Imaging System; Bio‐Rad). Densitometry quantification of the western blot band intensity was performed using Image Lab, version 4.0 (Bio‐Rad) and determined as the total band intensity adjusted for background intensity. Primary antibodies used are presented in Table 2. Primary antibodies were optimized by use of mixed human muscle standard lysates. Two mixed study samples containing tissue from biopsies were used to ensure that the protein amount loaded would result in band signal intensities localized on the steep and linear part of a standard curve. Secondary antibodies used were HRP conjugated rabbit anti‐sheep (P‐0163), goat anti‐mouse (P‐0447; Dako, Glostrup, Denmark) and goat anti‐rabbit IgM/IgG (4010‐05; SouthernBiotech, Birmingham, AL, USA).

Table 2.

Primary antibodies used for western blotting

Target protein	Manufacturer	Number	Molecular weight (kDa)
4E‐BP1	Cell Signaling	9452	15–20
p‐4E‐BP1Thr37/46	Cell Signaling	2855	15–20
Akt2	Cell Signaling	3063	60
p‐Akt2Ser473	Cell Signaling	9271	60
CREB	Cell Signaling	9197	43
p‐CREBSer129/Ser133	Abcam	ab10564	37–43
eEF2	Abcam	ab130187	95
p‐eEF2Thr56	Cell Signaling	2331	95
mTOR	Cell Signaling	2972	289
p‐mTORSer2448	Cell Signaling	2971	289
p38MAPK	Cell Signaling	9212	37–43
p‐p38MAPKThr180/Tyr182	Cell Signaling	9211	37–43
p70S6K	Cell Signaling	2708	70
p‐p70S6KThr389	Cell Signaling	9234	70

Suppliers: Abcam, Cambridge, MA, USA; Cell Signaling, Beverly, MA, USA.

RNA isolation, reverse transcription, and real‐time PCR

The method for RNA isolation, reverse transcription and real‐time PCR has been described previously (Pilegaard et al. 2000; Brandt et al. 2016). Total RNA was isolated from ∼5 mg wet weight muscle tissue using a modified guanidinium thiocyanate–phenol–chloroform extraction method from Chomczynski and Sacchi (1987) as described by Pilegaard et al. (2000), except for the use of a TissueLyser (TissueLyser II; Qiagen, Valencia, CA, USA) for homogenization. Superscript II RNase H‐ and Oligo dT (Invitrogen, Carlsbad, CA, USA) were used to reverse transcribe mRNA to cDNA (Pilegaard et al. 2000). Quantification of cDNA as a measure of mRNA content of a given gene was performed by real‐time PCR using an ABI 7900 sequence‐detection system (Applied Biosystems, Foster City, CA, USA). Probes and primers were either self‐designed (Table 3) or pre‐developed gene expression assays (ActivinRIIB Hs00609603_m1, calpain‐1 Hs00559804_m1, Smad3 Hs00969210_m1) (Applied Biosystems). Self‐designed probes and 5′‐6‐carboxyfluorescein (FAM)/3′‐6‐carboxy‐N,N,N′,N′‐tetramethylrhodamine labelled TaqMan probes were designed from human specific databases from ensemble (www.ensembl.org/homo_sapiens/info/index) using Primer Express, version 3.0 (Applied Biosystems) and were obtained from TAG Copenhagen (Copenhagen, Denmark).

Table 3.

Primers used for real‐time PCR

Target gene	Sense primer	Antisense primer	TaqMan probe
Atrogin	5'‐GATGTTACCCAAGGAAAGAGCAGTAT‐3'	5'‐ACGGATGGTCAGTGCCCTT‐3'	5'‐CCCTTCAGCTCTGCAAACACTGTCACAT‐3'
FoxO1	5'‐ACCGAACAGGATGATCTTGGA‐3'	5'‐CCATCTGCCGCAAAGATGGCCTCTA‐3'	5'‐CCATCTGCCGCAAAGATGGCCTCTA‐3'
MurF	5'‐GGAGCCACCTTCCTCTTGACT‐3'	5'‐CTCAAAGCCCTGCTCTGTCTTC‐3'	5'‐AACTCATCAAAAGCATTGTGGAAGCTTCCAA‐3'
Myostatin	5'‐ACCAGGAGAAGATGGGCTGAA‐3'	5'‐GTCAAGACCAAAATCCCTTCTGGA‐3'	5'‐CCGTTTTTAGAGGTCAAGGTAACAGACACACCA‐3'
PGC‐1α	5'‐CAAGCCAAACCAACAACTTTATCTCT‐3'	5'‐CACACTTAAGGTGCGTTCAATAGTC‐3'	5'‐AGTCACCAAATGACCCCAAGGGTTCC‐3'

Real‐time PCR was performed in triplicates in a total reaction volume of 10 μL using Universal Mastermix with UNG (Applied Biosystems). The obtained cycle threshold values reflecting the initial content of the specific transcript in the samples were converted to a relative amount by using standard curves constructed from serial dilution of a pooled sample made from all samples. Target mRNA content was normalized to single‐stranded DNA content in each sample determined by using OliGreen reagent (Molecular Probes, Leiden, The Netherlands) as described previously (Lundby et al. 2005).

Plasma concentrations of salbutamol

Plasma concentrations of salbutamol were measured by ultra high performance liquid chromatography (UPLC)‐MS/MS using deuterated internal standard based on methods described previously (Jacobson et al. 2015). In brief, calibration samples were prepared using unlabelled salbutamol in drug free plasma over a concentration range of 2–200 ng × mL⁻¹ and internal standard salbutamol‐D3 (3‐hydroxymethyl‐D2, α‐D1; Medical Isotopes, Inc., Pelham, NH, USA) was added to each plasma sample (200 μL) or calibration sample equivalent to 20 ng × mL⁻¹. Ammonia solution (200 μL, pH 9) was then added to each sample and vortex mixed before the addition of 1000 μL of HPLC grade ethyl acetate. This was vortex mixed for 1 min and then centrifuged at 15,000 g for 5 min.  The organic supernatant was then transferred to a glass autosampler vial, from which the solvent was evaporated under nitrogen at 40°C. The residue was reconstituted using 100 μL of methanol and vortex mixed prior to analysis via UPLC‐MS/MS consisting of a Waters Acquity® H‐class UPLC system (Waters Corporation, Milford, MA, USA) with chromatography performed using an Astec® CHIROBIOTIC™ T2 chiral column (4.6 × 250 mm × 5 μm particles) (Sigma‐Aldrich) coupled to a Waters Xevo® triple quadrupole mass spectrometer (Waters Corporation) with analyses undertaken using multiple reaction monitoring with conditions as described previously (Jacobson et al. 2015). Assay performance data were within acceptance criteria, with accuracy and precision (% relative SD; n = 5 at 5 ng × mL⁻¹) both less than 5% and calibration r ² > 0.9998. Total salbutamol levels were calculated from the sum of individual enantiomers.

Statistical analysis

Statistical analyses were performed in SPSS, version 24 (IBM Corp., Armonk, US). Sample size was determined for the primary outcome measure (myofibrillar FSR) and was estimated from the effect of beta₂‐agonist treatment on protein synthesis in animals (Koopman et al. 2010) and between‐subject SD from resistance exercise studies in humans (Kumar et al., 2009). Data were tested for normality using the Shapiro–Wilks test and Q‐Q plots. Variables that violated normality were log‐transformed (i.e. phosphorylation‐ratio and mRNA level data). To estimate differences between treatments, two‐tailed linear mixed modelling was used with treatment as a fixed effect and a random effect for subjects. In addition, age and LBM were included in the model as time invariant covariates because they may confound the effect of beta₂‐agonist (White & Leenen, 1994; Cheymol, 2000). Area under the phenylalanine leg net balance–time curve (AUC) was analysed using the trapezoidal rule with inclusion of baseline net balance as a covariate in the mixed model. For mRNA content, technical replicates were nested within the fixed effects (Acharya & Zhu, 2011). In the case of repeated measures, sampling point was included in the model as a fixed effect for a full factorial design. Within‐sampling point P values were adjusted using the Bonferroni method. Data are presented as the mean with the 95% confidence interval (CI), unless otherwise stated, and exact P values (unless <0.01 or >0.50) to represent probability for treatment fixed effects.

Results

Plasma concentrations of salbutamol

Arterial plasma concentrations of salbutamol were 46.5 (95% CI, 39.1 to 53.8) and 52.0 (95% CI, 42.8 to 61.3) ng × mL⁻¹ 0.5 and 5 h after exercise (2 h 30 min and 7 h after drug administration, respectively). No salbutamol was detected in the blood during the placebo trial.

Circulating phenylalanine availability

Average arterial plasma [¹³C₆]‐phenylalanine enrichment was lower (P < 0.01) for salbutamol than placebo, whereas femoral venous enrichment was higher (P < 0.01) for salbutamol than placebo (Fig. 3 A and B). Average arterial and femoral venous plasma concentrations of phenylalanine were lower (P < 0.01) for salbutamol than placebo (Fig. 3 C and D).

Figure 3. Circulating phenylalanine availability for salbutamol (black) and placebo (white) before (–0.5 h) and 0.5–5 h following resistance exercise.

A, arterial and venous [¹³C₆]‐phenylalanine enrichment. B, arteriovenous difference in [¹³C₆]‐phenylalanine enrichment. C, femoral arterial and venous phenylalanine concentration. D, arteriovenous difference in phenylalanine concentration. Values are the mean (n = 12). Error bars represent upper or lower bound of the 95% CI. ^**Treatment difference (P < 0.01) at same point. ^$$Overall treatment main effect (P < 0.01).

Leg phenylalanine kinetics based on the two‐pool model

Average femoral arterial plasma flow was more than two‐fold higher (P < 0.01) for salbutamol than placebo (Fig. 4 A). No differences were observed in [¹³C₆]‐phenylalanine leg net balance between salbutamol and placebo before exercise (Fig. 4 B). In the period 0.5–5 h after exercise, mean phenylalanine leg net balance was 0.6 (95% CI, −2.4 to 3.7) and –3.0 (95% CI, −0.7 to −5.2) nmol × min⁻¹ × 100 g _LLM ⁻¹ for salbutamol and placebo, respectively (P < 0.01) (Fig. 4 B). Phenylalanine leg net balance AUC was higher (P = 0.05) for salbutamol than placebo (Fig. 4 C). Average rate of disappearance and appearance of [¹³C₆]‐phenylalanine was higher (P < 0.01) for salbutamol than placebo (Fig. 4 D and E).

Figure 4. Leg phenylalanine kinetics based on the two‐pool model for salbutamol (black) and placebo (white) before (–0.5 h) and 0.5–5 h following resistance exercise.

A, femoral arterial plasma flow. B, leg net phenylalanine balance curve. C, area under the leg net phenylalanine balance–time curve. C and D, rate of disappearance. E, rate of appearance. Values are the mean (n = 12). Error bars represent upper or lower bound of the 95% CI.^**Treatment difference (P < 0.01) at same point. ^$$Overall treatment main effect (P < 0.01).

Muscle protein synthesis rate and leg phenylalanine kinetics based on 3‐pool model

Myofibrillar FSR was determined as the incorporation of tracer into myofibrillar proteins using the i.m. tracer enrichment (MPE) as precursor [salbutamol 0.5 h: 7.0 (95% CI, 6.4 to 7.6%) and 5 h: 7.6 (95% CI, 7.0 to 8.2%) and placebo 0.5 h: 7.6 (95% CI, 6.9 to 8.3%) and 5 h: 7.5 (95% CI, 6.8 to 8.2%)]. Myofibrillar FSR was 0.013 (95% CI, 0.001 to 0.025%) × h⁻¹ higher (P = 0.03) for salbutamol than placebo (Fig. 5 A). LBM was a significant negative confounder (P = 0.03) of the salbutamol‐induced change in myofibrillar FSR, whereas age did not confound the response (P > 0.50). The effect of salbutamol on myofibrillar FSR was moderate (Cohen's d = 0.78).

Figure 5. Muscle protein synthesis rate and leg phenylalanine kinetics based on the three‐pool model for salbutamol (black) and placebo (white) 0.5–5 h following resistance exercise.

A, myofibrillar FSR. B, average protein synthesis (F _O,M) (three‐pool model). C, average protein breakdown (F _M,O) (three‐pool model). Values are the mean (n = 12). Error bars represent upper or lower bound of the 95% CI. ^**Treatment difference (P < 0.01).

Based on the three‐pool model, salbutamol had higher estimates of protein synthesis (F _O,M) (P < 0.01) and breakdown (F _M,O) (P < 0.01) than placebo (Fig. 5 B and C).

Inward and outward muscle transmembrane transport of phenylalanine was not significantly different between treatments 0.5 h after exercise, although it was higher (P < 0.01) for salbutamol than placebo 5 h after exercise (Table 4). Arteriovenous shunting was more than two‐fold higher (P < 0.01) for salbutamol than placebo 0.5 and 5 h after exercise (Table 4).

Table 4.

Selected phenylalanine kinetics parameters based on the three‐pool model

	Placebo		Salbutamol		Tests of fixed effects
	0.5 h	5 h	0.5 h	5 h	Treatment	Time	Treatment by Time
Inward muscle transmembrane transport (PHE nmol × min−1 × 100 g LLM−1)	57	41	72	78	<0.01	0.58	0.43
	(40 to 73)	(32 to 49)	(39 to 105)	(62 to 94)**
Outward muscle transmembrane transport (PHE nmol × min−1 × 100 g LLM−1)	59	45	69	79	0.03	0.79	0.21
	(45 to 72)	(36 to 53)	(37 to 101)	(58 to 99)**
Arteriovenous shunting (PHE nmol × min−1 × 100 g LLM−1)	166	83	324	209	<0.01	<0.01	0.87
	(107 to 224)	(54 to 112)	(249 to 399)**	(131 to 286)**

Values are the mean (95% CI) (n = 12). ^**Statistically significant difference (P ≤ 0.01) compared to placebo at same time point.

Muscle signalling

PKA substrate intensity (P = 0.01) and phosphorylation of Akt2 (P = 0.02) and CREB (P < 0.01) were higher for salbutamol than placebo 0.5 h after exercise, whereas no relevant differences were observed between the treatments in phosphorylation of 4E‐BP1 (P = 0.23), eEF2 (P = 0.17), MAPK (P > 0.50), mTOR (P = 0.17) and p70S6K (P = 0.39) (Fig. 6 A). PKA substrate intensity (P < 0.01) and phosphorylation of Akt2 (P < 0.01), CREB (P < 0.01) and eEF2 (P = 0.01) were higher for salbutamol than placebo 5 h following exercise, whereas phosphorylation of 4E‐BP1 (P = 0.38), MAPK (P > 0.50), mTOR (P > 0.50) and p70S6K (P = 0.13) was not different between treatments (Fig. 6 B).

Figure 6. Phosphorylation‐ratio induced by salbutamol (SAL) compared to placebo (PLA).

Phosphorylation‐ratio induced by SAL compared to PLA in biopsies sampled from the vastus lateralis muscle 0.5 (A) and 5 h (B) after resistance exercise. Values are the mean log‐change (n = 12). Error bars represent upper or lower bound of the 95% CI. ^*Treatment difference (P < 0.05). ^**Treatment difference (P < 0.01). C, representative blots for salbutamol (S) and placebo (P).

mRNA content

The mRNA content of calpain‐1 (P < 0.01), FoxO1 (P < 0.01), myostatin (P < 0.01) and Smad3 (P < 0.01) was higher for salbutamol than placebo 0.5 h after exercise, whereas no significant changes were induced by salbutamol in content of ActivinRIIB (P = 0.25), atrogin (P > 0.50), MuRF (P > 0.50) and PGC‐1α (P = 0.33) compared to placebo (Fig. 7 A). The mRNA content of ActivinRIIB was lower (P < 0.01) for salbutamol than placebo 5 h after exercise, whereas FoxO1 (P < 0.01) and myostatin (P < 0.01) mRNA content was higher for salbutamol than placebo (Fig. 7 B). No treatment differences were observed in the mRNA content of atrogin (P = 0.48), calpain‐1 (P > 0.50), PGC‐1α (P = 0.09), MuRF (P = 0.24) and Smad3 (P > 0.50) 5 h after exercise (Fig. 7 B).

Figure 7. mRNA response induced by salbutamol (SAL) compared to placebo (PLA).

mRNA response induced by SAL compared to PLA in biopsies sampled from the vastus lateralis muscle 0.5 (A) and 5 h (B) after resistance exercise. Values are the mean log‐change (n = 12). Error bars represent upper or lower bound of the 95% CI. ^**Treatment difference (P < 0.01).

Discussion

In the present study, we have reported the beta₂‐adrenergically‐induced changes in protein turnover rates and associated changes in intracellular signalling and mRNA content in skeletal muscle after resistance exercise in trained young men. The most important findings are that beta₂‐adrenergic stimulation with the commonly prescribed selective beta₂‐agonist, salbutamol, increased myofibrillar FSR and protein turnover rates, thus favouring an improved net protein balance in skeletal muscle following resistance exercise. Changes in protein turnover induced by salbutamol were associated with PKA‐signalling, activation of Akt2 and modulation of mRNA response of growth‐regulating proteins in skeletal muscle.

Although beta₂‐agonists have been marketed for more than 50 years, the present study is first to show that a few days of beta₂‐agonist treatment increases myofibrillar FSR and leg protein turnover rates, resulting in an improved leg net protein balance after resistance exercise in humans. The higher myofibrillar FSR induced by salbutamol was in agreement with our working hypothesis and consistent with reported observations in rodents (Maltin et al. 1989; Hesketh et al. 1992; Koopman et al. 2010). Contrary to our observations, isoproterenol was shown to have no effect on whole‐body and muscle protein synthesis in young men (Robinson et al. 2010). The type of beta₂‐agonist and dosing regimen applied may explain this discrepancy. In the present study, we administered salbutamol, which has superior selectivity for the beta₂‐adrenoceptor than isoproterenol (Baker, 2010). Furthermore, we chose to administer salbutamol in supratherapeutic doses and daily for four days prior to the experiments because studies in rodents have shown that the stimulatory effect of beta₂‐agonist on anabolism is dose‐dependent and that the beta₂‐agonist‐induced increase in protein synthesis requires consecutive days of treatment (Koopman et al. 2010). Consistent with this, Lee et al. (2015) observed that 7 days of oral treatment with formoterol increased whole‐body protein synthesis. In addition, we investigated the effect of beta₂‐agonist on 5 h of protein turnover following resistance exercise and not during resting conditions as in the previous human studies (Robinson et al. 2010; Lee et al. 2015). Accordingly, the present study indicates that a few days of beta₂‐agonist treatment increase protein turnover rates in the first 5 h period following resistance exercise, which is in agreement with the augmenting effect of daily salbutamol treatment on muscle adaptations to resistance training observed in previous studies (Caruso et al. 2004, 2005).

We observed that LBM confounded the effect of salbutamol on myofibrillar FSR, whereas no relevant confounding effect was observed for body mass (Pearson's r = –0.096, P = 0.32; data not shown). Distribution of drugs, including beta₂‐agonists, is influenced by body composition (Cheymol, 2000), and LBM has been shown to be a superior predictor of the response to drugs than body mass (Morgan & Bray, 1994; Han et al. 2007). The influence of LBM on the response to salbutamol is probably related to the distribution kinetics of salbutamol, exhibiting extensive disposition in skeletal muscle (Jacobson et al. 2014). The beta‐adrenoceptor cardiac response has also been shown to decline with age (White & Leenen, 1994), although we observed no relevant impact of subject age on the salbutamol‐induced change in myofibrillar FSR. This may be explained by the relatively low heterogeneity in the present study population (range 19–32 years) and the different target tissue (cardiac vs. skeletal muscle). Nevertheless, the present observations suggest that LBM may be taken into consideration when investigating effects of beta₂‐agonists. In this context, it has been speculated that the effect of beta₂‐agonist on exercise performance and muscle excitation–contraction coupling depends on the training level of the subject (van Baak et al. 2004; Decorte et al. 2013).

Although studies in rodents have indicated that the hypertrophic effect of beta₂‐agonist may involve both attenuation of muscle proteolytic processes (Busquets et al. 2004; Yimlamai et al. 2005) and an increase in protein synthesis (Maltin et al. 1989; Hesketh et al. 1992), we observed that salbutamol markedly increased leg protein turnover rates by almost doubling the rate of protein breakdown and synthesis during the 5 h period following exercise. Given that the resistance exercise undertaken was matched between the salbutamol and placebo trial, the higher rate of protein breakdown induced by salbutamol is related to factors other than the total work performed. Although a putative mechanism could be the pronounced increase in arterial femoral plasma flow (Biolo et al. 1997) induced by salbutamol, we observed no apparent association between femoral plasma flow and myofibrillar FSR (r = 0.11, P = 0.60) or leg net balance AUC (r = 0.05, P = 0.80). Furthermore, despite a higher femoral plasma flow and lower arterial plasma concentration of phenylalanine, the arteriovenous phenylalanine difference was more positive for salbutamol than placebo, in which there was a net release of phenylalanine in the 0.5–5 h following exercise. As such, our observations suggest that the greater protein turnover rates for salbutamol is related to myocellular mechanisms.

Despite the increase in rate of protein breakdown, salbutamol counteracted a net negative protein balance following resistance exercise, which was evident in the placebo condition. The negative protein balance observed for placebo is consistent with previous studies, where the balance is negative in the post‐absorptive state following resistance exercise (Biolo et al. 1995; Phillips et al. 1997). In this context, it is important to emphasize that subjects in the present study consumed a standardized low‐protein meal 2 h prior to the resistance exercise to provide some energy to be available for the exercise session, at the same time as being sufficiently low in protein to affect metabolism at the post‐exercise measurements. Studies have shown that there is a graded response of muscle FSR to dietary protein or amino acid infusion (Bohé et al. 2003; Moore et al. 2009), and whether the effect of beta₂‐agonist on protein turnover rates would have been different in completely fasting conditions or in conditions where subjects had consumed higher amounts of essential amino acids remains to be determined. Nonetheless, the beneficial effect of salbutamol in enhancing net balance to more positive levels than with placebo underpins the efficacy of beta₂‐agonist in stimulating muscle anabolism.

We observed that salbutamol induced significant beta₂‐adrenergic signalling in skeletal muscle 0.5 and 5 h following resistance exercise, as indicated by a higher phosphorylation of PKA substrates and downstream activation of cAMP/PKA‐dependent target CREB. In rodents, the growth‐promoting mechanisms of beta₂‐adrenergic signalling involves Akt, mTOR and MAPK pathways (Kline et al. 2007; Sato et al. 2013), which are predominant in the regulation of translation initiation (Goodman, 2014) and cell proliferation and differentiation (Pearson et al. 2001). Despite the induced PKA‐signalling and higher phosphorylation of Akt2 with salbutamol 0.5 and 5 h following exercise, we observed no changes in phosphorylation of mTOR^Ser2448 and downstream effectors of translation initiation, p70S6K and 4E‐BP1, or in phosphorylation of p38‐MAPK. Although the latter observations may appear to be inconsistent with reports in rodents (Sato et al. 2013), beta₂‐adrenergic signalling may be muscle fibre‐type specific (Gonçalves et al. 2012) and some studies report no effect of beta₂‐adrenergic stimulation on p38‐MAPK phosphorylation (Kim et al. 2013). Reports in mice also indicate that beta₂‐adrenergic stimulation does not phosphorylate mTOR^Ser2448, although it may phosphorylate mTOR^Ser2481 (Sato et al. 2014), which could be a possible explanation of the observed increase in phosphorylation of Akt^Ser473 (Copp et al. 2010). Indeed, mTOR^Ser2448 may not be a target of Akt (Figueiredo et al. 2017). We also observed that salbutamol increased the phosphorylation of eEF2, which acts to reduce ribosomal elongation activity (Ryazanov et al. 1988). Although this may appear to be unexpected considering the higher protein synthesis rate and phosphorylation of Akt2 with salbutamol, studies have shown that cAMP‐PKA‐dependent signalling induces phosphorylation of eEF2 and inhibition of peptide elongation in vitro (Redpath & Proud, 1993).

Aside from the induced changes in signalling, we observed that salbutamol modulated mRNA levels of ActivinRIIB, calpain‐1, FoxO1, myostatin and Smad3 following exercise. Most noteworthy was the upregulation of mRNA levels of the negative regulator of growth, myostatin. Although increased mRNA levels of myostatin may appear counterintuitive given the anabolic properties of beta₂‐agonists, Abo et al. (2012) observed that hypertrophy induced by beta₂‐agonist was associated with increased protein levels of myostatin in rats. Importantly, we also observed that salbutamol induced a downregulation of the mRNA level of the receptor target of myostatin, ActivinRIIB, 5 h after exercise. Thus, a potential upregulation of myostatin induced by beta₂‐agonist may be counteracted by a concurrent downregulation of ActivinRIIB, which is consistent with that observed in rat tibialis anterior muscle following beta₂‐agonist treatment (Pearen et al. 2009). In addition, we observed that salbutamol upregulated mRNA levels of Smad3 and FoxO1. Given that Smad3‐null mice display loss of satellite cells and muscle atrophy (Ge et al. 2011), it may be speculated that a beta₂‐agonist‐induced upregulation of Smad3 plays a role in growth‐promotion. FoxO1, a regulator of the atrophy‐related genes atrogin and MuRF (Bodine & Baehr, 2014), is among the targets that are regulated by the ActivinRIIB‐myostatin system and Akt signalling. However, despite significant Akt activation and upregulation of FoxO1 mRNA levels with salbutamol, we observed no changes in the mRNA level of atrogin and MuRF with salbutamol compared to placebo. Furthermore, although a potential effect of CREB activation is increased transcription of PGC‐1α, we observed no effect of salbutamol on mRNA levels of PGC‐1α compared to placebo. The latter observation is consistent with observations in rats, where beta₂‐agonist treatment with clenbuterol did not necessarily affect PGC‐1α mRNA levels (Kim et al. 2013; Shimamoto et al. 2017). We also observed that the calpain1 mRNA content was increased by salbutamol 0.5 h following exercise, which potentially may have contributed to Ca²⁺‐dependent proteolysis and thus the higher protein breakdown for salbutamol than placebo. The observation that salbutamol increased calpain mRNA levels is consistent with rodent studies, where beta₂‐agonist treatment with formoterol increased calpain mRNA levels (Koopman et al. 2010).

The effect of beta₂‐agonist on gene transcription and signalling possibly depends on timing of sampling and the biological samples (e.g. cells vs. tissue), as well as on type and dose of beta₂‐agonist used (Baker, 2010; Wannenes et al. 2012). For example, although clenbuterol repressed mRNA levels of atrogin and MuRF in C2C12 muscle cell lines (Wannenes et al. 2012), no effect was found in rat soleus muscle after 3 days of treatment with clenbuterol (Gonçalves et al. 2012). Furthermore, unlike clenbuterol, salbutamol was shown to have no apparent effect on mRNA levels of atrogin and MuRF in C2C12 muscle cell lines (Wannenes et al. 2012). Based on the present study, along with studies in rodents, changes in muscle protein turnover induced by beta₂‐agonists are possibly multifactorial, involving complex regulation of gene transcription and ribosomal translation (Pearen et al. 2009; Koopman et al. 2010).

In summary, the present observations show that selective activation of beta₂‐adrenoceptors with salbutamol increases myofibrillar FSR and protein turnover rates in skeletal muscle following resistance exercise in trained young men. Furthermore, our observations indicate that LBM confounds the salbutamol‐induced change in myofibrillar FSR. The effect of salbutamol in protein turnover rates was associated with pronounced PKA‐signalling and phosphorylation of CREB and Akt2, as well as a concurrent mRNA response for growth‐regulating genes, including ActivinRII, FoxO1 and myostatin.

Methodological considerations

We observed that the arterial phenylalanine enrichment rose from ∼12% to 13.5% MPE during the period of which the tracer measures were performed and therefore whether isotopic steady‐state was achieved in the present study could be considered. Nonetheless, we observed a constant venous enrichment of ∼11.5% MPE and no difference in the i.m. enrichment at 0.5 and 5 h in recovery from exercise (0.079 ± 0.014 and 0.082 ± 0.013, respectively, P = 0.32) (mean ± SD), demonstrating that close to the actual site of protein turnover, tracer enrichments were not changing significantly. In clinical trials where homeostasis may be affected by drugs, exercise or other factors, minor fluctuations in circulating tracer enrichments may also be expected. In the present study, as well as in some other protocols (Rahbek et al. 2014; Mikkelsen et al. 2015), we applied a rather high tracer infusion rate (7 μmol × kg LBM⁻¹ × h⁻¹) compared to 3.6 μmol × kg whole body weight⁻¹ × h⁻¹ used in other studies (Wilkinson et al. 2015; Wall et al. 2016). Our rationale for this infusion rate was to improve analytical sensitivity to allow detection of expectedly small intervention differences. However, with the precision of modern mass spectrometers, the relative high tracer infusion rate was probably not necessary and it is recommended to use a lower infusion rate to limit costs and reduce potential impact of the tracer on metabolism. It was shown, however, that the myofibrillar FSR was unaffected by flooding with 1665 mg phenylalanine (>10,000 μmol), increasing the blood (and most probably also intracellular) concentrations several‐fold (Holm et al. 2014). For comparison, the infusion in the present study equalled a total amount of ∼500–600 mg phenylalanine over a 7 h time period, which probably did not affect phenylalanine metabolism and the muscle protein synthesis rate. In addition, in the present study, we used different tracer principles to investigate the effect of salbutamol vs. placebo on protein turnover rates. Although some variability was observed within the different estimates, consistency appeared across the findings when evaluated over the entire post‐exercise period and, for the primary outcome measure (myofibrillar FSR determined by the direct incorporation method), we observed consistent findings with the three‐pool tracer dilution estimate of protein synthesis rate (F _O,M) (Pearson's r = 0.52, P = 0.009). It should also be highlighted that the effect of beta₂‐agonist vs. placebo on protein turnover rates observed in the present study was in a postprandial setting where the subjects also performed exercise. Therefore, any interpretation of the sole beta₂‐adrenergic effect based on the present study should be made with caution because nutritional intake and exercise may confound the effect of beta₂‐agonist.

Translational perspectives

The present study adds to animal studies showing that beta₂‐agonist can alter protein turnover in skeletal muscle following resistance exercise in humans. The practical implications of beta₂‐agonist‐induced changes in protein turnover rates remain to be determined. Although studies in rodents have provided support of beta₂‐agonists as treatment of muscle atrophy, concerns were raised because of concurrent adverse ventricular remodelling and collagen filtration (Gregorevic et al. 2005; Burniston et al. 2007). However, given the markedly lower relative doses prescribed to humans, such effects are possibly not a major concern. Furthermore, the most commonly used beta₂‐agonists in humans, such as salbutamol and formoterol, have superior selectivity for the beta₂‐adrenoceptor compared to clenbuterol and fenoterol (Baker, 2010), thus reducing or avoiding potential adverse activation of cardiac beta₁‐adrenoceptors. Recent human studies also show that beta₂‐agonists may hold some promise as anabolic agents with few minor side effects (Hostrup et al. 2015; Lee et al. 2015; Jessen et al. 2018). The beta₂‐agonist‐induced increase in protein turnover may also have implications for proteome signature remodelling of various components in skeletal muscle. Indeed, beta₂‐agonist treatment has been shown to modulate proteome signature adaptations to endurance training in humans (Hostrup et al. 2018). Furthermore, given that remodelling and recycling of myocellular proteins are important adaptive processes with respect to stress and exercise (Camera et al. 2017), it may be that beta₂‐agonists augment post‐exercise recovery processes after resistance exercise. The observation that supratherapeutic oral doses of salbutamol increase protein turnover rates in association with resistance exercise provides support for the anti‐doping regulatory restrictions regarding the supratherapeutic use of beta₂‐agonists in competitive sport.

Additional information

Competing interests

The authors declare that they have no competing interests.

Author contributions

MH designed the study and participated in the human experiments. LH and SR contributed to the conception of design and phenylalanine analyses. AK, JB, JE, MK, MN, SJ and VB participated in the human experiments. MT performed the immunoblotting. CMK and HP performed the RNA isolation, reverse transcription and real‐time PCR. GAJ developed the salbutamol assay and performed the analysis of salbutamol in plasma. All authors contributed to the analysis and interpretation of data and the drafting of the manuscript. All authors approved the final version of the manuscript submitted for publication.

Funding

The study was supported by grants from the Danish Ministry of Culture and the World Anti‐Doping Agency.

Biography

Morten Hostrup is Associate Professor in the Section of Integrative Physiology, Department of Nutrition, Exercise and Sports, University of Copenhagen. He obtained his PhD degree in integrative physiology from the University of Copenhagen in 2015, covering the effects of beta₂‐adrenoceptor agonists in relation to skeletal muscle physiology and exercise performance. His current research areas are the biomedical basis of exercise performance and experimental pharmacology in humans, with an emphasis on the adrenergic system, metabolism, ion transport and skeletal muscle remodelling induced by exercise.