Clenbuterol induces lean mass and muscle protein accretion, but attenuates cardiorespiratory fitness and desensitizes muscle β₂‐adrenergic signalling

Abstract

Abstract

The β₂‐adrenergic agonist clenbuterol is widely abused because of its purported fat‐burning actions, muscle accretion properties and performance enhancing effects, and yet it remains unexplored in randomized controlled trials. In the present study, we subjected 11 healthy men (aged 18–40 years) to two 2 week cycles of oral clenbuterol (80 µg day⁻¹) or placebo, separated by a 3 week washout. During each cycle, we assessed body composition, cardiorespiratory fitness, sprint power output, cardiac left ventricular mass and intravascular blood volume. We obtained vastus lateralis muscle biopsies and analysed them for protein content, 3‐hydroxyacyl CoA dehydrogenase (HAD) activity, oxidative phosphorylation complex (OXPHOS) abundance, platelet endothelial cell adhesion molecule (PECAM‐1) abundance and β₂‐adrenergic signalling. Compared to placebo, clenbuterol induced a 0.91 kg lean mass gain (95% confidence interval = 0.02–1.81, P < 0.05) but had no effect on fat mass. Clenbuterol reduced maximal oxygen uptake by 7% (P < 0.001) and exercise capacity by 4% (P < 0.001) but had no effects on sprint power output, left ventricular mass, intravascular blood volume or haemoglobin mass. Clenbuterol increased muscle protein content (P < 0.05) and PECAM‐1 abundance (P < 0.05) but repressed HAD activity (P < 0.01) and OXPHOS complex V abundance (P < 0.05). Clenbuterol markedly activated muscle protein kinase A (P < 0.001) and phosphorylated ribosomal protein S6 (Ser235/236) but this effect declined during the 2 week cycle. Although a 2 week clenbuterol cycle effectively induces lean mass gain and muscle protein accretion, it negatively affects cardiorespiratory fitness, represses muscle oxidative capacity, and induces tolerance in β₂‐adrenergic signalling and ribosomal protein S6 phosphorylation. The adverse effects of clenbuterol along with its muscle anabolic actions justify its prohibition in elite sports.

Key points

• Clenbuterol, a potent β₂‐adrenergic agonist, has purported fat‐burning and muscle accretion properties. However, its purported effects, along with its potential adverse effects on cardiorespiratory fitness, remain unexplored in humans.

• A short 2 week clenbuterol cycle induces lean mass gain and muscle protein accretion in healthy young men.

• Clenbuterol induces β₂‐adrenergic signalling and phosphorylates RpS6^Ser235/236 in skeletal muscle, but this signalling response is attenuated with repeated exposure.

• Clenbuterol negatively affects cardiorespiratory fitness and represses muscle oxidative capacity.

• Clenbuterol does not affect left ventricular mass, intravascular blood volume or haemoglobin mass.

Abstract figure legend This randomized controlled study investigated the effect of a 2 week treatment cycle of either oral clenbuterol or placebo in healthy young men. We found that clenbuterol: (1) increased lean mass but (2) impaired maximal oxygen uptake (${\dot{V}}_{{\mathrm{O}}_2\max }$) and incremental exercise capacity. In addition, we observed that clenbuterol (3) induced marked activation of protein kinase A (PKA) and ribosomal protein S6 kinase (RpS6) in skeletal muscle but that this signalling response was attenuated after 2 weeks. Our findings demonstrate that clenbuterol acts muscle anabolic but negatively affects cardiorespiratory fitness and induces tolerance in β₂‐adrenergic signalling in skeletal muscle of young men.

Introduction

Clenbuterol, a long‐acting and potent β₂‐adrenergic agonist with sympathomimetic actions, is widely abused because of its purported fat‐burning and muscle accretion properties (Milano et al., 2018; Schifano et al., 2018). Although clenbuterol was initially developed as a bronchodilator in the late 1970s for treating asthma (Anderson & Wilkins, 1977), its abuse rose rapidly in the wake of studies demonstrating apparent muscle hypertrophy and fat reductions in clenbuterol‐fed mammals (Dalrymple et al., 1984; Parkins et al., 1989). Accordingly, clenbuterol quickly became one of the most heavily abused substances worldwide (Milano et al., 2018; Spiller et al., 2013).

Surveys show that around 40% of bodybuilders use clenbuterol (Li et al., 2023; Steele et al., 2020) and pharmacovigilance studies reveal a prevalent misuse in the European Union (Milano et al., 2018), predominantly among young men, with several reports of hospitalization and fatal outcomes (Kumari et al., 2023; Spiller et al., 2013). Illicit use of clenbuterol is not only a problem pertaining to the fitness and bodybuilding communities (Prather et al., 1995). The World Anti Doping Agency (WADA) prohibits its use in sports, but it nevertheless accounts for a significant number of adverse analytical findings (AAFs) in doping control (Velasco‐Bejarano et al., 2017).

Despite the prevalent abuse of clenbuterol and claims of it being ‘probably the best fat burner’ (Palumbo, 2022), hailed for its ability to induce leanness (O'Connor, 2024; Palumbo, 2022) and enhance performance (Fragkaki et al., 2013; O'Connor, 2024; Prezelj et al., 2003), randomized controlled trials in healthy humans are sparse. In an open‐label study, clenbuterol (80 µg orally) potentiated protein kinase A (PKA) and mammalian target of rapamycin (i.e. mTOR)‐signalling, and augmented fat oxidation by 39% in young men (Jessen et al., 2020). However, whether this translates into lean mass gain and fat mass reduction with prolonged treatment is unclear. Based on studies in animals, there is a good reason to assume this is the case. Horses treated with clenbuterol gained around 8% lean mass and lost in the excess of 15% fat mass after just 2 weeks of treatment (Kearns et al., 2001). We also know from other substances within the drug class that β₂‐agonists effectively induce muscle hypertrophy in humans (Hostrup & Onslev, 2022; Hostrup et al., 2020). The commonly prescribed β₂‐agonists salbutamol and formoterol, both of which have shorter half‐lives and durations of action than clenbuterol, have been shown to act muscle anabolic in humans (Hostrup, Reitelseder et al., 2018; Jessen et al., 2023, 2021; Lee et al., 2015). Furthermore, terbutaline, a short‐acting β₂‐agonist commonly prescribed in Northern Europe, increased lean mass by around 1.5 kg at the same time as lowering fat mass over the course of only 2–4 weeks (Acheson et al., 1988; Hostrup et al., 2015). Given the widespread off‐label use of clenbuterol and perception of it being particularly effective, this warrants investigation under controlled settings (O'Connor, 2024; Palumbo, 2022).

Prolonged use of clenbuterol can also impose adverse effects. In horses, clenbuterol compromises cardiorespiratory fitness as reflected by impairments in maximal oxygen uptake (${\dot{V}}_{{\mathrm{O}}_2\max }$), cardiac function and exercise capacity (Drake et al., 2013; Kearns & McKeever, 2009). These findings, along with observations of clenbuterol compromising mitochondrial volume and function in rat skeletal muscle (Hoshino et al., 2012; Kitaoka et al., 2019), point towards adverse effects that warrant investigation in humans not only to inform on the potential risks of abuse, but also before expanding the medical indications for clenbuterol beyond that of bronchodilation, such as in muscle wasting conditions as proposed previously (Herrera et al., 2001; Martineau et al., 1993).

Individuals using clenbuterol off‐label to promote leanness often resort to 2–3 week on–off cycles to offset tolerance development (O'Connor, 2024; Prather et al., 1995). Although the concept of tolerance development upon chronic exposure to β₂‐agonists is a well‐described phenomenon (Salpeter et al., 2004; Sato et al., 2015), less is known about its implication for myocellular signalling events important for muscle hypertrophy. Repeated exposure to clenbuterol may lead to attenuated activation of β₂‐adrenergic signalling pathways, including canonical cAMP‐PKA signalling and downstream effectors regulating protein synthesis. In this respect, ribosomal protein S6 (RpS6) appears particularly interesting because RpS6 has been implicated in muscle growth (Ruvinsky et al., 2009), is activated by resistance training in human skeletal muscle (Hodson et al., 2022) and is directly phosphorylated by PKA at a serine 235/236 site (Ser235/236) (Biever et al., 2015).

In the present study, we investigated the effect of a 2 week clenbuterol cycle on body composition, cardiorespiratory fitness, sprint power output, cardiac left ventricular function and intravascular blood volume in healthy young men. In addition, we examined changes in total protein content, oxidative capacity and β₂‐adrenergic signalling in skeletal muscle. We hypothesized that clenbuterol would increase lean mass at the same time as reducing fat mass, as well as enhance sprint power output, but would exert a detrimental effect on cardiorespiratory fitness. Furthermore, we hypothesized that repeated exposure to clenbuterol would induce tolerance in muscle β₂‐adrenergic signalling and downstream phosphorylation of RpS6 (Ser235/236).

Methods

Study design and ethical approval

The study comprised a randomized, double‐blinded, placebo‐controlled, cross‐over trial conducted in the Department of Nutrition, Exercise and Sports (NEXS), University of Copenhagen, Denmark, between November 2021 and December 2022 (Fig. 1). The study adhered to the 2013 Declaration of Helsinki and was approved by the science ethics committee of the Capital region, Denmark (H‐17011319). All participants were informed about possible risks involved and gave their oral and written consent before inclusion. The study was pre‐registered in Clinicaltrials.gov (Trial identifier: NCT03860870) and was part of a larger series of trials examining strategies to detect clenbuterol abuse for doping control purposes and illuminating its physiological effects (Jessen et al., 2020; Solheim et al., 2020).

Figure 1. Study design.

Overview of the study design (A) and experimental trials (B). DXA, dual‐energy X‐ray absorptiometry; CO, carbon monoxide.

The main outcome measure was change in body composition (lean and fat mass) for clenbuterol (80 µg day⁻¹) vs. placebo. Secondary outcomes were change in cardiorespiratory fitness and muscle β₂‐adrenergic signalling related to protein synthesis. Exploratory outcomes included sprint power output, cardiac left ventricular mass, intravascular blood volume, haemoglobin mass, heart rate and blood pressure, as well as total protein content and abundance and maximal activity of oxidative enzymes in skeletal muscle. Sample size was determined for the main outcome measure (change in lean mass) in G*Power, version 3.1.9.3 (Faul et al., 2007). We estimated a mean ± SD difference of +1.4 ± 1.8 kg for clenbuterol vs. placebo based on former β₂‐agonist studies (Acheson et al., 1988; Hostrup et al., 2015). With an α‐level of 0.05 and β‐level of 0.8 for a linear mixed model repeated measures design this required a sample size of 10 participants to complete the study in a cross‐over design.

Participants and eligibility criteria

Inclusion criteria were healthy male volunteers aged 18–40 years with no known contraindications for anabolic drugs (e.g. cancer and cardiac abnormalities), as well as oral and written informed consent. We chose to enrol males only because data from the European Medicines Agency show an overwhelming proportion of abuse in males (female/male ratio = 0.09) (Milano et al., 2018). Exclusion criteria were self‐reported steroid abuse, ongoing use of prescription medicine, resistance training more than twice weekly, heart abnormality or disease deemed by the study responsible medical doctor to infer a risk to participate in the trial.

Assessment of eligibility criteria

We assessed eligibility criteria during an examination with electrocardiography (ECG‐2150; Nihon Kohden, Rosbach, Germany), echocardiography (GE Vivid E9; transducer GE M5Sc, GE Healthcare, Brøndby, Denmark) and blood pressure measurements (M7 Intellisense; OMRON, Kyoto, Japan). Further, participants underwent dual‐energy X‐ray absorptiometry (DXA) to assess body composition followed by a ramp test to task failure on a bike ergometer (Monark LC6; Monark, Vansbro, Sweden) for determination of cardiorespiratory fitness. Thereafter, we familiarized participants to a 6 s sprint test on a peak bike ergometer (Monark Ergomedic 894E Peak Bike; Monark Exercise AB, Vansbro, Sweden).

Randomization, blinding and drug intervention

Participants who fulfilled the eligibility criteria were included in the study and randomized to initiate a 2 week cycle with either clenbuterol (4 × 20 µg day⁻¹ clenbuterol hydrochloride, Spiropent; Boehringer Ingelheim International GmbH, Ingelheim am Rhein, Germany) or placebo (four lactosemonohydrate tablets per day, regional pharmacy of the Capital Region, Denmark). We chose a treatment duration of 2 weeks because previous β₂‐agonist studies show that this is sufficient to induce adaptations (Acheson et al., 1988; Hostrup et al., 2016) and is the commonly reported cycle length in bodybuilders abusing clenbuterol for leanness (Prather et al., 1995). The dose of 80 µg day⁻¹ was based on our previous study on the acute effect of clenbuterol on resting metabolic rate and muscle signalling (Jessen et al., 2020). We introduced a 3 week washout between the two treatment periods, which was deemed adequate based on prior experience with β₂‐agonists and former β₂‐agonist studies in which a 4 week washout was employed (Collomp et al., 2000; Le Panse et al., 2006). During each treatment period, we instructed participants to ingest the study drug as a once‐daily dose in the morning during breakfast. We supervised all ingestions via video monitoring (e.g. Skype/FaceTime) to ensure maximum drug compliance.

Randomization was performed by personnel not involved in enrolment or data acquisition. Participants were randomized using a random number generator to determine whether they would begin with the clenbuterol or placebo treatment period. A pre‐specified allocation list was used to ensure an equal number of participants started with each treatment. Participants received clenbuterol or placebo in identically looking tablets. Allocation sequence was concealed to assessors and participants throughout the study. We instructed participants to maintain daily activity and dietary habits throughout the study. In addition, we asked participants not to donate blood during the intervention.

Before and 3 days after each 2 week treatment period, participants attended an experimental trial to assess changes in outcome measures with the intervention. The 3 day washout period between the last study drug administration of the 2 week treatment period was to minimize acute spillover effects of clenbuterol given it's long half‐life of ~25–35 h (Solheim et al., 2020; Yamamoto et al., 1985; Yang et al., 2015).

Pre‐ and post‐intervention trials

Participants came into the laboratory in the morning after a standardized breakfast. First, we assessed participants’ body composition by a DXA scan followed by echocardiography and assessment of heart rate and blood pressure where participants were instrumented with ECG electrodes to record heart rate from an electrocardiogram (ECG; GE Medical Systems, Chicago, IL, USA) and arterial systolic and diastolic pressures were measured three consecutive times with a electrosphygmomanometer (M7 Intellisense; OMRON), where the average systolic and diastolic pressure were used for data analysis. Then, a muscle biopsy was sampled from the vastus lateralis. Thereafter, participants ingested 80 µg of clenbuterol or placebo and rested in a supine position for 2 h based on the findings of muscle signalling following clenbuterol ingestion from our previous study (Jessen et al., 2020) and when clenbuterol reaches systemic peak concentrations (Yamamoto et al., 1985). Then, we performed an assessment of acute changes in heart rate and blood pressure, and another muscle biopsy was obtained from m. vastus lateralis to assess acute β₂‐adrenergic signalling. Participants then completed an individualized warmup on a bike ergometer (Monark LC6; Monark Exercise AB) that consisted of 3 × 4 min at a workload corresponding to 30%, 50% and 70% of ${\dot{V}}_{{\mathrm{O}}_2\max }$ followed by two short submaximal sprints. Five minutes after the warmup, participants performed a 6 s maximal sprint on a peak bike ergometer (Monark Ergomedic 894E Peak Bike; Monark Exercise AB). Following 10 min of rest, participants completed a bike ergometer ramp test performed to task failure for determination of exercise capacity and ${\dot{V}}_{{\mathrm{O}}_2\max }$ as indicators of cardiorespiratory fitness. After 3 h of rest, intravascular blood volume and haemoglobin mass were measured.

All trials were conducted at the same time of day for each participant to minimize the impact of circadian fluctuations. We instructed participants to standardize their meal and fluid intake on the day of each experimental trial and to avoid alcohol, caffeine, and nicotine for 24 h before testing. We also instructed participants to refrain from vigorous exercise for 48 h leading up to experimental days.

Experimental procedures

DXA

For all participants during each trial, we performed two consecutive DXA scans (Lunar iDXA; GE Healthcare) to account for scan‐to‐scan variation (Zemski et al., 2019) with the mean of the two scans used for analysis. Our in‐lab, scan‐to‐scan, within‐individual coefficient of variation (CV) for lean mass is typically 0.3 ± 0.2% (mean ± SD) for healthy trained individuals. Before the scan, participants rested in a standardized supine position on the scanner bed for 10 min to minimize the influence of body fluid shifts (Berg et al., 1993). The scanner was calibrated daily.

Power output during a 6 s sprint

We determined participants’ peak and mean power output during a 6 s bike ergometer sprint as previously described (Jessen et al., 2021). Participants pedalled slowly against a resistance of 6 N until a cadence of ∼100 rpm was reached. Thereafter, ergometer resistance increased to 0.9 N kg⁻¹ body mass, and participants pedalled as fast as possible for 6 s in a seated position. Power output during the 6 s sprint was recorded using Monark software (Monark Anaerobic Test software, version 3.3). Peak power output was determined as the highest power output recorded over 1 s, whereas mean power output was determined as the average power output during the entire 6 s sprint. Saddle height and handlebar positions were recorded for each participant at the screening visit and replicated on subsequent visits.

Cardiorespiratory fitness during a ramp test

We assessed participants’ cardiorespiratory fitness during a ramp test to task failure. The test started at 150 W and increased continuously at a rate of 30 W min⁻¹. Pulmonary gas exchange was measured during the test breath‐by‐breath with a gas analysing system (Vyntus CPX; Vyaire Medical Gmbh, Hoechberg, Germany). Participants maintained a cadence of ∼90 rpm. Task failure was defined as a drop in cadence below 70 rpm for more than 3 s despite strong verbal encouragement. The highest workload achieved before task failure was recorded (W _max). Following the ramp test, participants rested for 10 min before performing a constant power output test to exhaustion at 110% of W _max, as recommended by Poole & Jones (2017) to validate ${\dot{V}}_{{\mathrm{O}}_2\max }$ assessment. Thus, ${\dot{V}}_{{\mathrm{O}}_2\max }$ was identified as the highest average value recorded during any consecutive 30 s period in either the ramp test or the subsequent constant power output test at 110% of W _max.

Echocardiography

After 10 min of rest, we assessed participants’ left ventricular mass and function by ultrasound using transthoracic echocardiograms with a 2.5 MHz transducer (GE Vivid E9; GE Healthcare) according to current guidelines (Lang et al., 2015). During measurements, participants rested in a supine left lateral position in an air‐conditioned and darkened room maintained at 23°C. A minimum of three consecutive cardiac cycles were collected for each series. Echocardiographic images were analysed using EchoPac software version 203 (GE Healthcare). Left ventricular mass was assessed using the linear method using the cube formula and ejection fraction the Teichholz method (Lang et al., 2015). Stroke volume was evaluated during using the velocity time integral method, with the measurement of the left ventricular outflow tract diameter at diastole recorded at rest and was assumed to remain constant (Lang et al., 2015). Assessments and analysis were performed by the same investigator (MF), who was blinded to the treatment of the participants.

Intravascular blood volume and haemoglobin mass

We determined participants’ intravascular blood volume and haemoglobin mass using the carbon monoxide (CO)‐rebreathing technique (Breenfeldt Andersen et al., 2024). Participants rested in a supine position with legs raised to avoid blood pooling in extremities. A venous blood sample was collected and gently inverted five to eight times before an immediate complete blood count analysis on a Sysmex XN‐450 (Sysmex, Kobe, Japan). Then, four capillary blood samples from a pre‐heated fingertip were collected in 35 µL pre‐heparinized tubes (safeClinitubes; Radiometer, Brønshøj, Denmark) and analysed for percentage carboxyhaemoglobin on an ABL 800 blood gas analyser (Radiometer). Participants were then instructed to exhale completely to measure end‐tidal CO before being connected to a custom designed spirometer (Hans Rudolph, Shawnee, KS, USA). Thereafter, 1.0 mL kg⁻¹ body weight chemically pure (99.997%) CO (CO N47; Air Liquide, Paris, France) was delivered via a 100 mL plastic syringe (Nemoto; Tokyo, Japan) to the spirometer creating a closed system. The system contained 5 L of 100% oxygen, which was rebreathed for 2 min with the applied dosage of CO. Four capillary blood samples were collected and analysed 9 min after the inhalation of CO using the same technique used during baseline sampling. A CO analyser (Draeger, Lübeck, Germany) was used to assess whether a leak in the closed system occurred during the rebreathing period and measure any leftover CO in the spirometer and end‐tidal CO 3 min after the rebreathing period. The haematocrit, haemoglobin concentration and the difference in percent carboxyhaemoglobin were used to calculate total haemoglobin mass and intravascular volumes (Schmidt & Prommer, 2005), which was adjusted for a loss of CO to myoglobin (0.3% min⁻¹) and through ventilation (estimated alveolar ventilation of 5 L min⁻¹). A second CO‐rebreathing procedure was performed immediately after the first with the mean of the two rebreathing procedures used for analysis as described elsewhere (Breenfeldt Andersen et al., 2024). All measurements were performed by the same investigators (ABA and JB) who were blinded to the treatment of the participants.

Muscle biopsies

Muscle biopsies were obtained, under local anaesthesia (lidocaine without epinephrine, xylocaine, 20 mg mL⁻¹; AstraZeneca, Cambridge, UK), through a small incision in the skin at the belly of the vastus lateralis using a Bergström needle (Stille, Stockholm, Sweden) with suction (Bergström, 1975). Immediately after sampling, the biopsy was washed in sterile saline solution (0.9% NaCl; Fresenius Kabi, Uppsala, Sweden), and quickly dissected free of visible blood, connective tissue and fat and then frozen in liquid nitrogen and stored at –80°C. Before biochemical analyses, ∼20 mg of wet weight muscle was freeze‐dried and then dissected free from blood, fat and connective tissue under a microscope with fine forceps and stored at –80°C until analysis.

Total muscle protein content

Around 0.5 mg of dry weight muscle tissue was homogenized (1:400) in a 0.3 mol L⁻¹ phosphate buffer (pH 7.7) with two rounds of 30 s using a TissueLyser II (Qiagen, Valencia, CA, USA). Then, total protein content was determined in triplicate with a BSA kit (Thermo Fisher Scientific, Waltham, MA, USA).

Maximal activity of 3‐hydroxyacyl CoA dehydrogenase

Maximal activity of 3‐hydroxyacyl CoA dehydrogenase (HAD) was determined from ∼0.5 mg of dry weight muscle tissue, before homogenization (1:400) in 0.3 mol L⁻¹ phosphate buffer (pH 7.7) with two rounds of 30 s using a TissueLyser II (Qiagen). Maximal enzyme activity was determined fluorometrically NAD‐NADH coupled reactions (Fluoroscan Ascent; Thermo Fisher Scientific) at 25°C as previously described (Lowry, 2012).

SDS‐PAGE and immunoblotting

Protein abundance and phosphorylation were determined by western blotting as previously described (Thomassen et al., 2016). We assessed β₂‐adrenergic signalling by phosphorylation of PKA substrates and the downstream effector RpS6. We determined the abundance of complexes I–V of the oxidative phosphorylation (OXPHOS) and platelet endothelial cell adhesion molecule (PECAM‐1), also known as cluster of differentiation 31 (CD31), as a marker of capillarization (Wagatsuma & Osawa, 2006). Briefly, ∼1 mg of dry weight muscle tissue was homogenized for 1 min at 30 Hz on a shaking bead‐mill (TissueLyser II; Qiagen) in ice‐cold lysis buffer containing: 10% glycerol, 20 mm Na‐pyrophosphate, 150 mm NaCl, 50 mm HEPES (pH 7.5), 1% NP‐40, 20 mm β‐glycerophosphate, 2 mm Na₃VO₄, 10 mm NaF, 2 mm phenylmethanesulfonyl fluoride, 1 mm EDTA (pH 8), 1 mm EGTA (pH 8) 10 µg mL⁻¹ aprotinin, 10 µg mL⁻¹ leupeptin and 3 mm benzamidine. Samples were rotated end‐over‐end for 30 min at 4°C and centrifuged (18,320  g ) for 20 min at 4°C. The protein concentration of each sample was determined in triplicate with a BSA kit (Thermo Fisher Scientific) and after determination of muscle protein content samples were created in duplicate with 6 × Laemmli buffer (7 mL of 0.5 m Tris‐base, 3 mL of glycerol, 0.93 g of DTT, 1 g of SDS and 1.2 mg of bromophenol blue) and ddH₂O to achieve equal protein concentration. Equal amounts of protein were loaded in wells of pre‐cast 4–15% gels (Bio‐Rad Laboratories, Hercules, CA, USA), except for OXPHOS complex I–V determination, which was on precast 16.5% gels, with all samples for each participant loaded on the same gel. Proteins were then separated according to their molecular weight by SDS‐PAGE and semi‐dry transferred to a PVDF membrane (Millipore A/S, Copenhagen, Denmark). Membranes were blocked for 15 min in either 2% skim milk or 3% BSA in Tris‐buffered saline containing 0.1% Tween 20 before an overnight incubation in primary antibody at 4°C and a subsequent incubation in horseradish peroxidase conjugated secondary antibody at room temperature for 1 h. Bands were visualized with ECL (Millipore A/S) and recorded with a digital camera (ChemiDoc MP Imaging System; Bio‐Rad Laboratories). Bands were quantified using Image Lab, version 6.0 (Bio‐Rad Laboratories) and determined as the total band intensity adjusted for background intensity. The primary antibodies used were: CD31/PECAM‐1 (#AF806; R&D Systems Inc., Minneapolis, MN, USA), OXPHOS (ab110411; Abcam, Cambridge, UK), phospho‐(Ser/Thr) PKA substrate (#9621; Cell Signaling Technology, Danvers, MA, USA), phospho‐RpS6 (ab225676; Abcam) and phospho‐RpS6^Ser235/236 (#4858; Cell Signaling Technology). Secondary antibodies used were horseradish peroxidase‐conjugated rabbit anti‐sheep (dilution 1:5000; P0163; Dako, Glostrup, Denmark), goat anti‐rabbit IgM/IgG (4010‐05; SouthernBiotech, Birmingham, AL, USA) and goat anti‐mouse (dilution 1:5000; P0447; Dako).

Statistical analysis

We used SPSS, version 29 (IBM Corp., Armonk, NY, USA) for statistical analyses. First, we tested data for normality using the Shapiro–Wilk test and Q‐Q plots. Because some immunoblotting data partly violated normality, we log‐transformed these data. Data are presented as the mean ± SD and outcome statistics as delta mean‐change with 95% confidence intervals (CI). To estimate the chronic effects of clenbuterol vs. placebo after each 2 week treatment period, we employed a repeated measures two‐factor mixed linear model with treatment (clenbuterol vs. placebo) and trial (pre vs. post) as fixed factors, participant as a random factor, and a first‐order autoregressive (AR1) covariance structure. The AR1 structure was selected based on the temporal correlation of repeated measurements within each 2 week treatment cycle and expected decay in correlations between treatment cycles. We compared AR1 with compound symmetry and unstructured covariance matrices using Aike information criterion/Bayesian information criterion for the primary outcome, with AR1 providing the best model fit. To assess the acute effects of clenbuterol vs. placebo for each 2 week period a three‐factor mixed linear model that also included time (rest/2 h) was employed. P < 0.05 was considered statistically significant.

Results

Participants, compliance and side effects

Thirteen participants fulfilled the eligibility criteria for inclusion. Two participants withdrew from the study, whereas one participant only completed the first 2 weeks of the study (randomized to clenbuterol). Participant characteristics are presented in Table 1.

Table 1.

Participant characteristics

Characteristic	Value
Age (years)	26 ± 6
Height (cm)	184 ± 7
Weight (kg)	80.3 ± 5.5
Lean body mass (kg)	63.2 ± 5.4
Fat mass (kg)	14.1 ± 2.9
Body fat percentage (%)	18.3 ± 3.7
${\dot{V}}_{{\mathrm{O}}_2\max }$ (mL min−1 kg−1)	52.0 ± 5.0

Data are the mean ± SD (n = 11). ${\dot{V}}_{{\mathrm{O}}_2\max }$, maximal oxygen uptake.

Compared to placebo, clenbuterol increased heart rate 2 h after ingestion (+10 bpm, 95% CI = 6–15 bpm, treatment × time interaction: P < 0.001) (Table 2), but with no significant change throughout the 2 week cycle (+4 bpm, 95% CI = –1 to 8 bpm, treatment × trial interaction: P = 0.082) (Table 2). Likewise, clenbuterol decreased diastolic blood pressure (–4 mmHg, 95% CI = –7 to –1 mmHg, treatment × time interaction: P = 0.007) (Table 2) and mean arterial pressure (MAP) (–5 mmHg, 95% CI = –10 to 0 mmHg, treatment × time interaction: P = 0.043) (Table 2) 2 h following ingestion, but with no significant changes in diastolic blood pressure (+3 mmHg, 95% CI = 0 to 6 mmHg, treatment × trial interaction: P = 0.081) (Table 2) and MAP (+5 mmHg, 95% CI = 0 to 10 mmHg, treatment × trial interaction: P = 0.056) (Table 2) throughout the 2 week cycle. No changes in systolic blood pressure were observed 2 h following clenbuterol ingestion (–7 mmHg, 95% CI = –19 to 6 mmHg, treatment × time interaction: P = 0.270) (Table 2) or during the 2 week clenbuterol cycle (+8 mmHg, 95% CI = –4 to 21 mmHg, treatment × trial interaction: P = 0.153) (Table 2).

Table 2.

Heart rate and blood pressure at rest and 2 h after ingestion of clenbuterol or placebo before (Pre) and after (Post) 2 weeks of treatment

	Clenbuterol (n = 11)				Placebo (n = 10)
	Pre		Post		Pre		Post
	Rest	2 h	Rest	2 h	Rest	2 h	Rest	2 h
HR (bpm)	58 ± 6	64 ± 7 ***	60 ± 5	64 ± 9 ***	62 ± 8	56 ± 8	59 ± 8	55 ± 8
DBP (mmHg)	67 ± 6	67 ± 5 **	67 ± 4	66 ± 5 **	67 ± 6	71 ± 7	65 ± 6	67 ± 6
SBP (mmHg)	121 ± 7	117 ± 36	125 ± 7	124 ± 11	123 ± 8	130 ± 10	121 ± 6	124 ± 8
MAP (mmHg)	85 ± 6	84 ± 12 *	86 ± 5	85 ± 6 *	86 ± 6	91 ± 8	84 ± 6	86 ± 7

Values are the mean ± SD. HR, heart rate. DBP, diastolic blood pressure. SBP, systolic blood pressure. MAP, mean arterial pressure.

^*Different from placebo (treatment × time interaction: P < 0.05).

^**Different from placebo (treatment × time interaction: P < 0.01).

^***Different from placebo (treatment × time interaction: P < 0.001).

We observed no order of treatment effects (i.e. carry‐over effects) for any of the reported outcomes.

Body composition

Clenbuterol increased lean body mass by 0.91 kg (95% CI = 0.02–1.81 kg) compared to placebo (treatment × trial interaction: P = 0.046) (Fig. 2A ). Clenbuterol did not change fat mass (0.00 kg, 95% CI = –0.52 to 0.52 kg; treatment × trial interaction: P = 0.994) (Fig. 2B ) or percentage (–0.20%, 95% CI = –0.79 to 0.39%; treatment × trial interaction: P = 0.501) (Fig. 2C ) compared to placebo. No regional‐specific differences were observed in response to clenbuterol vs. placebo (treatment × trial × region: all P > 0.5) (Fig. 2D–F ) (Table 3).

Figure 2. Changes in body composition in men following clenbuterol or placebo.

Changes in body composition in men following 2 weeks of clenbuterol (CLEN, n = 11) or placebo (PLA, n = 10). Bars represent mean changes in whole body lean mass (A), fat mass (B) and fat percentage (C), as well as regional changes in lean mass (D), fat mass (E) and fat percentage (F). Circles represent data from individual participants.

Table 3.

Body composition before (Pre) and after (Post) 2 weeks with clenbuterol or placebo

	Clenbuterol (n = 11)			Placebo (n = 10)
	Pre	Post	∆‐Change (95% CI)	Pre	Post	∆‐Change (95% CI)
Whole‐body lean mass (kg)	62.7 ± 5.6	63.5 ± 6.0	0.9 (0.3–1.5) *	62.9 ± 6.6	62.9 ± 6.6	0.0 (−0.6 to 0.6)
Leg lean mass (kg)	21.2 ± 1.9	21.5 ± 2.0	0.3 (0.0‐−0.6)	21.4 ± 2.1	21.5 ± 2.1	0.1 (−0.2 to 0.4)
Arm lean mass (kg)	8.2 ± 1.3	8.4 ± 1.3	0.2 (0.1‐−0.3) *	8.3 ± 1.4	8.3 ± 1.5	0.0 (−0.2 to 0.1)
Torso lean mass (kg)	29.5 ± 2.7	29.9 ± 2.9 *	0.4 (0.0‐−0.8)	29.5 ± 3.3	29.5 ± 3.3	0.0 (−0.4 to 0.4)
‍
Whole‐body fat mass (kg)	13.5 ± 3.0	13.2 ± 2.8	−0.3 (−0.6 to 0.0)	13.6 ± 2.9	13.3 ± 2.8	−0.3 (−0.6 to 0.0)
Leg fat mass (kg)	4.6 ± 1.1	4.5 ± 1.1	−0.1 (−0.2 to 0.0)	4.6 ± 1.2	4.5 ± 1.1	−0.2 (−0.3 to 0.0)
Arm fat mass (kg)	1.5 ± 0.4	1.4 ± 0.4	0.0 (−0.1 to 0.0)	1.5 ± 0.4	1.5 ± 0.4	0.0 (0.0 to 0.1)
Torso fat mass (kg)	6.5 ± 1.7	6.3 ± 1.6	−0.2 (−0.5 to 0.0)	6.5 ± 1.6	6.4 ± 1.6	−0.1 (−0.4 to 0.1)
‍
Whole‐body fat percentage (%)	17.7 ± 3.4	17.2 ± 3.4	−0.5 (−0.9 to −0.1)	17.8 ± 3.7	17.5 ± 3.5	−0.3 (−0.7 to 0.1)
Leg fat percentage (%)	17.5 ± 3.5	17.2 ± 3.5	−0.4 (−0.7 to 0.0)	17.7 ± 4.0	17.1 ± 3.7	−0.5 (−0.9 to −0.2)
Arm fat percentage (%)	15.1 ± 3.4	14.7 ± 3.2	−0.5 (−0.9 to 0.0)	15.1 ± 3.7	15.3 ± 3.5	0.2 (−0.2 to 0.6)
Torso fat percentage (%)	18.0 ± 4.0	17.4 ± 4.1	−0.7 (−1.3 to −0.1)	18.2 ± 4.3	17.8 ± 4.3	−0.3 (−1.0 to 0.3)

Values are the mean ± SD or 95% confidence interval (CI).

^*Different from placebo (P < 0.05).

Cardiorespiratory fitness

Clenbuterol decreased cardiorespiratory fitness as reflected by both an absolute (–249 mL min⁻¹, 95% CI = –378 to –120; treatment × trial interaction: P < 0.001) (Fig. 3A ) and relative (–3.6 mL min⁻¹ kg⁻¹, 95% CI = –5.4 to –1.9; treatment × trial interaction: P < 0.001) (Fig. 3B ) decline in ${\dot{V}}_{{\mathrm{O}}_2\max }$ compared to placebo and a lower W_max attained during the ramp test for clenbuterol than placebo (–16 W, 95% CI = –23 to –8; treatment × trial interaction: P < 0.001) (Fig. 3C ). Likewise, the power output that elicited ${\dot{V}}_{{\mathrm{O}}_2\max }$ declined with clenbuterol compared to placebo (–16 W, 95% CI = –27 to –6; treatment × trial interaction: P = 0.004) (Fig. 3D ) (Table 4).

Figure 3. Changes in indicators of cardiorespiratory fitness in men following clenbuterol or placebo.

Changes in indicators of cardiorespiratory fitness in men following 2 weeks of clenbuterol (CLEN, n = 11) or placebo (PLA, n = 10). Bars represent mean changes in absolute (A) and relative (C) ${\dot{V}}_{{\mathrm{O}}_2\max }$, as well as highest power out (W _max) at ${\dot{V}}_{{\mathrm{O}}_2\max }$ (C) and before task failure (D) during a ramp test on a bicycle ergometer. Circles represent data from individual participants.

Table 4.

Indicators of cardiorespiratory fitness before (Pre) and after (Post) 2 weeks with clenbuterol or placebo

	Clenbuterol (n = 11)			Placebo (n = 10)
	Pre	Post	∆‐Change (95% CI)	Pre	Post	∆‐Change (95% CI)
${\dot{V}}_{{\mathrm{O}}_2\max }$ (mL min−1)	4082 ± 637	3973 ± 618	−109 (−192 to −26) ***	3940 ± 678	4080 ± 704	140 (227 to 53)
${\dot{V}}_{{\mathrm{O}}_2\max }$ (mL min−1 kg−1)	51.5 ± 6.1	49.9 ± 6.3*	−1.6 (−2.7 to −0.5) ***	49.6 ± 7.2	51.6 ± 7.6	2.0 (0.9 to 3.2)
W max (W)	353 ± 47	343 ± 48 **	−10 (−16 to −4) ***	349 ± 51	356 ± 51	7 (1 to 13)
Aerobic W max (W)	337 ± 45	326 ± 46	−11 (−18 to −4) **	330 ± 50	335 ± 47	5 (−2 to 12)

Values are the mean ± SD or 95% confidence interval (CI).

^**Different from placebo (P < 0.01).

^***Different from placebo (P < 0.001).

Peak power during maximum sprinting

Peak (–36 W, 95% CI = –84 to 13 W; treatment × trial interaction: P = 0.135) and mean (–21 W, 95% CI = –75 to 33 W; treatment × trial interaction: P = 0.402) power output during the 6 s sprint was not affected by clenbuterol compared to placebo. No differences were observed across or within each treatment cycle (Table 5).

Table 5.

Mean and peak power during 6‐s maximal sprinting before (Pre) and after (Post) 2 weeks with clenbuterol or placebo

	Clenbuterol (n = 10)			Placebo (n = 10)
	Pre	Post	∆‐Change (95% CI)	Pre	Post	∆‐Change (95% CI)
Peak power output (W)	1225 ± 152	1219 ± 157	−6 (−51 to 40)	1204 ± 160	1234 ± 175	30 (−5 to 65)
Mean power output (W)	1094 ± 127	1084 ± 108	−10 (−49 to 29)	1094 ± 141	1105 ± 151	11 (−33 to 54)

Values are the mean ± SD or 95% confidence interval (CI).

Central intravascular blood volume and left ventricle adaptations

Clenbuterol had no apparent effects on haematocrit (treatment × trial interaction: P = 0.527), haemoglobin mass (treatment × trial interaction: P = 0.169), red blood cell volume (treatment × trial interaction: P = 0.785), plasma volume (treatment × trial interaction: P = 0.335) or blood volume (treatment × trial interaction) (P = 0.309) (Table 6). While left ventricular mass and ejection fraction did not change during the intervention for either treatment period, differences between treatments were evident for resting stroke volume and cardiac output (Table 6). Resting stroke volume decreased by 11 mL (95% CI = 0–21, treatment × trial interaction: P = 0.043) with placebo compared to clenbuterol and cardiac output increased by 1.3 L min⁻¹ (95% CI = 0.6–1.9, treatment × trial interaction: P < 0.001) with clenbuterol compared to placebo.

Table 6.

Vascular and cardiac left ventricular measures before (Pre) and after (Post) 2 weeks of treatment with placebo or clenbuterol

Haematology	Clenbuterol (n = 11)			Placebo (n = 10)
	Pre	Post	∆‐Change (95% CI)	Pre	Post	∆‐Change (95% CI)
Haematocrit (%)	41.7 ± 2.6	40.7 ± 2.7	−1.0 (−2.3 to 0.3)	41.8 ± 2.0	41.4 ± 2.6	0.4 (−1.7 to 0.9)
Haemoglobin mass (g)	834 ± 85	829 ± 92	−2 (−15 to 10)	834 ± 99	815 ± 96	−16 (−29 to −4)
RBC volume (mL)	2430 ± 272	2418 ± 309	−15 (−46 to 76)	2433 ± 299	2405 ± 285	−32 (−89 to 26)
Plasma volume (mL)	3391 ± 305	3514 ± 390	121 (−25 to 268)	3402 ± 475	3408 ± 462	14 (−143 to 171)
Blood volume (mL)	5820 ± 499	5932 ± 633	108 (−48 to 264)	5835 ± 740	5813 ± 684	−18 (−181 to 146)
Left ventricle adaptations
Mass (g)	195 ± 41	192 ± 37	−3 (−15 to 9)	188 ± 47	190 ± 41	2 (−10 to 14)
Stroke volume (mL)	93 ± 14	95 ± 11	2 (−5 to 8) *	100 ± 14	92 ± 16	−8 (−15 to ‐2)
Cardiac output (L min−1)	5.4 ± 0.9	5.9 ± 0.9	0.5 (0.1 to 1.0) ***	6.1 ± 1.0	5.4 ± 1.0	−0.7 (−1.2 to −0.2)
Ejection fraction	0.54 ± 0.07	0.56 ± 0.07	0.02 (−0.02 to 0.06)	0.57 ± 0.06	0.56 ± 0.06	−0.01 (−0.06 to 0.03)

Values are the mean ± SD or 95% confidence interval (CI). RBC, red blood cell; LV, left ventricle.

^*Different from placebo (P < 0.05).

^***Different from placebo (P < 0.001).

Total muscle protein content

Clenbuterol increased total muscle protein content compared to placebo (+101 µg mg⁻¹ protein, 95% CI = 5–197 µg protein mg⁻¹; treatment × trial interaction: P = 0.040) (Fig. 4), with an increase in protein content from 446 ± 29 to 519 ± 156 µg mg⁻¹ protein (+74 µg mg⁻¹ protein, 95% CI = 8–139 µg mg⁻¹ protein; P = 0.030) (Fig. 4), whereas no change occurred with placebo (pre: 462 ± 35 µg mg⁻¹ protein; post: 435 ± 38 µg protein mg⁻¹; mean change: –28 µg ∙mg⁻¹ protein, 95% CI = –97 to 42 µg mg⁻¹; P = 0.414) (Fig. 4). The increase in muscle protein content with clenbuterol compared to placebo remained after removal of an outlier in the clenbuterol period (treatment × trial interaction: P = 0.008) (Fig. 4).

Figure 4. Change in muscle protein content in men following clenbuterol or placebo.

Change in muscle protein content (µg mg⁻¹ protein) in men following 2 weeks with clenbuterol (CLEN, n = 11) or placebo (PLA, n = 10). Bars represent mean change. Circles represent data from individual participants.

Muscle oxidative enzymes and PECAM‐1

Clenbuterol repressed HAD activity (–7.5 µmol min⁻¹ g protein⁻¹, 95% CI = –12.4 to –2.7 µmol min⁻¹ g protein⁻¹; treatment × trial interaction: P = 0.004) (Fig. 5A ) but increased CD31/PECAM‐1 abundance compared to placebo (+0.16 a.u., 95% CI = 0.02‐−0.29 a.u.; treatment × trial interaction: P = 0.030) (Fig. 5B ). Clenbuterol also decreased abundance of OXPHOS complex V (–0.06 a.u., 95% CI = –0.11 to –0.01 a.u.; treatment × trial interaction: P = 0.030) (Fig. 5C ) compared to placebo. However, no treatment differences were observed for OXPHOS complex I‐III, as well as mean OXPHOS abundance (Fig. 5C ) (Table 7).

Figure 5. Changes in muscle enzyme activity and protein abundance in men following clenbuterol or placebo.

Changes in muscle enzyme activity and protein abundance in men following 2 weeks with clenbuterol (CLEN, n = 11) or placebo (PLA, n = 10). Bars represent mean changes in 3‐hydroxyacyl CoA dehydrogenase activity (HAD) (A), platelet endothelial cell adhesion molecule abundance (PECAM‐1) (B) and oxidative phosphorylation complexes (OXPHOS) I, II and III, CV, and mean of all complexes (mean) (C). Circles represent data from individual participants.

Table 7.

Skeletal muscle markers of oxidative capacity before (Pre) and after (Post) 2 weeks with clenbuterol or placebo

	Clenbuterol (n = 11)			Placebo (n = 10)
	Pre	Post	∆‐Change (95% CI)	Pre	Post	∆‐Change (95% CI)
HAD (µmol min−1 g protein)	28.3 ± 10.9	23.2 ± 8.7	−5.1 (−8.2 to ‐2.0) **	23.2 ± 5.6	25.7 ± 6.8	2.5 (−0.9 to 5.9)
PECAM‐1 (a.u.)	1.84 ± 0.12	2.04 ± 0.14	0.20 (0.10 to 0.29) *	1.97 ± 0.20	2.01 ± 0.17	0.04 (−0.06 to 0.14)
OXPHOS
Complex I (a.u.)	1.95 ± 0.12	1.89 ± 0.19	−0.06 (−0.12 to 0.00)	1.93 ± 0.12	1.92 ± 0.13	−0.01 (−0.05 to 0.07)
Complex II (a.u.)	1.93 ± 0.15	1.92 ± 0.21	−0.02 (−0.09 to 0.06)	1.94 ± 0.14	1.90 ± 0.16	−0.04 (−0.11 to 0.04)
Complex III (a.u.)	1.96 ± 0.24	1.79 ± 0.40	−0.17 (−0.32 to −0.02)	1.87 ± 0.30	1.85 ± 0.34	−0.02 (−0.17 to 0.14)
Complex V (a.u.)	1.98 ± 0.08	1.95 ± 0.10	−0.04 (−0.08 to 0.00) *	1.95 ± 0.07	1.97 ± 0.06	0.02 (−0.02 to 0.06)
Mean (a.u.)	1.96 ± 0.14	1.89 ± 0.22	−0.07 (−0.14 to 0.00)	1.92 ± 0.15	1.91 ± 0.16	−0.01 (−0.09 to 0.07)

Values are the mean ± SD or 95% confidence interval (CI).

^*Different from placebo (P < 0.05).

^**Different from placebo (P < 0.01).

Muscle β₂‐adrenergic signalling and RpS6‐phosphorylation

Clenbuterol induced pronounced activation of PKA (treatment main effect: P < 0.001) (Fig. 6A ), no change in protein content of RpS6 (treatment × trial interaction: P = 0.662) (Fig. 6B ) and substantial phosphorylation of RpS6^Ser235/236 (treatment main effect: P = 0.043) (Fig. 6C ) compared to placebo. However, for both activation of PKA (treatment × trial interaction: P = 0.033) (Fig. 6A ) and phosphorylation of RpS6 ^Ser235/236 (treatment × trial interaction: P = 0.043) (Fig. 6C ) this effect was greater at the first day of treatment as compared to day 14.

Figure 6. Substrate phosphorylation of PKA and RpS6 content and phosphorylation in muscle following clenbuterol or placebo.

PKA substrate phosphorylation (A) and protein abundance (B) and phosphorylation (C) of RpS6 in men 2 h after ingestion of clenbuterol (CLEN, n = 11) or placebo (PLA, n = 10) at day 1 and day 14 of treatment. Bars are mean and circles represent data from individual participants.

Discussion

Our findings demonstrate that a 2 week cycle with clenbuterol induces lean mass gain and muscle protein accretion but negatively affects cardiorespiratory fitness and represses oxidative enzymes in skeletal muscle of healthy young men. In addition, we show that clenbuterol induces tolerance development in muscle β₂‐adrenergic signalling and downstream RpS6^Ser235/236 phosphorylation.

The full body lean mass gain of 0.91 kg in 2 weeks underscores the potency of clenbuterol to induce leanness in young individuals. Despite a much shorter treatment cycle, the observed gains correspond to the around 1 kg lean mass gains in patients with chronic heart failure treated with 40 µg clenbuterol twice daily for 12 weeks (Kamalakkannan et al., 2008) and even gains typically reported after resistance training programs of 2–3 week duration (Thomas et al., 2019). β₂‐agonists, in general, have proven effective in inducing lean mass gains after only a few weeks of treatment with effects ranging from 0.8 to 1.7 kg depending on dose, duration and type of β₂‐agonist used (Acheson et al., 1988; Hostrup et al., 2015; Jessen et al., 2023, 2018). Thus, lean mass effects are a class‐specific characteristic of β₂‐agonists rather than limited to clenbuterol alone.

The lean mass gain induced by the 2 week clenbuterol cycle was accompanied by a 17% increase in skeletal muscle protein content. Importantly, we performed these measures on freeze‐dried muscle tissue, confirming the increase was not due to water retention. In young and old rats, both salbutamol and clenbuterol have previously been shown to increase gastrocnemius muscle protein content by 20–25% during 3 week treatment periods (Carter & Lynch, 1994). These longitudinal effects align with the stimulating effect of β₂‐agonists on myofibrillar protein synthesis, inducing a favourable protein balance (Hostrup, Reitelseder, et al., 2018; Lee et al., 2015). Therefore, lean mass gains following β₂‐agonist treatment reflect genuine muscle hypertrophy and protein accretion rather than fluid retention or non‐muscle tissue expansion, as also demonstrated at the level of individual muscle fibres (Hostrup et al., 2015; Jessen et al., 2021).

Contrasting to the widespread illicit use of clenbuterol to reduce fat mass (O'Connor, 2024; Prather et al., 1995), we observed no effect of the 2 week clenbuterol cycle on whole body and regional fat mass. Although one could speculate that this reflects the short treatment period, 2–4 weeks of treatment with terbutaline reduced fat mass in the excess of 1 kg and increased lean mass by around 1.5 kg in healthy men (Acheson et al., 1988; Hostrup et al., 2015). Both habitual physical activity and diet could affect these outcomes, which are factors that we did not strictly control in the present study. When used over the course of an endurance training program, salbutamol and terbutaline lowered fat or body mass (Hostrup, Onslev, et al., 2018; Hostrup et al., 2023). Furthermore, given that we did not impose any caloric restriction in the present study, it may be that the participants compensated for a clenbuterol‐induced increase in energy expenditure (Jessen et al., 2020) during the 2 week cycle.

Consistent with the detrimental cardiorespiratory changes observed in clenbuterol‐treated horses (Drake et al., 2013; Kearns & McKeever, 2009), we observed that the 2 week clenbuterol cycle attenuated cardiorespiratory fitness as reflected by reduced ${\dot{V}}_{{\mathrm{O}}_2\max }$ and impairments in exercise capacity of 7% and 4%, respectively. Because ${\dot{V}}_{{\mathrm{O}}_2\max }$ is determined by cardiac output and the arteriovenous oxygen difference, the fact that clenbuterol did not affect left ventricular mass or ejection fraction, nor blood volume and haemoglobin mass, suggests that clenbuterol lowered ${\dot{V}}_{{\mathrm{O}}_2\max }$ predominantly via mechanisms that result in decreased skeletal muscle oxygen extraction or utilization, rather than central mechanisms. However, we observed an increase in PECAM‐1/CD31 with clenbuterol, indicating increased muscle capillarization and hence potential for extraction (Wagatsuma & Osawa, 2006). By contrast, clenbuterol decreased muscle OXPHOS complex V abundance and reduced maximal HAD activity, suggestive of a repression of the capacity for oxidative ATP synthesis and β‐oxidation in the working muscles (Maher et al., 2014). In line with this, rats treated with clenbuterol for 3 weeks exhibit apparent reductions in abundance of oxidative enzymes, including complex IV and HAD (Hoshino et al., 2012). Furthermore, formoterol also impairs ${\dot{V}}_{{\mathrm{O}}_2\max }$ and muscle mitochondrial respiratory capacity, whereas it has no effect on cardiac function and intravascular blood volume (Jessen et al., 2023). Collectively, these observations indicate that β₂‐agonists can compromise cardiorespiratory fitness via peripheral mechanisms pertaining to the oxidative capacity of skeletal muscle. Given the importance of cardiorespiratory fitness in morbidity and all‐cause mortality, even among young individuals (Lang et al., 2024), the lean mass gained during a clenbuterol cycle probably does not outweigh its detrimental effects, especially considering the effect on ${\dot{V}}_{{\mathrm{O}}_2\max }$ after only one 2 week cycle.

Unexpectedly, clenbuterol did not enhance sprint power output during maximal exercise in our study. Other β₂‐agonists, including salbutamol and terbutaline, have been shown to enhance sprinting power output when taken for a period of 3–4 weeks (Hostrup et al., 2015; Le Panse et al., 2005; Martineau et al., 1992; Sanchez et al., 2012). This discrepancy could relate to the short treatment duration of 2 weeks for which muscle gains induced by clenbuterol precede functional adaptations. It is not uncommon that muscle mass accretion induced by anabolic compounds is not accompanied by increases in force when not combined with resistance training (Creutzberg et al., 2003; Ferreira et al., 1998).

We observed that clenbuterol potently activated not only the canonical PKA effector signalling pathway, but also induced a marked phosphorylation of RpS6^Ser235/236, which is a key regulator of translation initiation that is implicated in muscle hypertrophy (Ruvinsky & Meyuhas, 2006; Ruvinsky et al., 2009). Although the first dosing of clenbuterol induced substantial activation of PKA and phosphorylation of RpS6^Ser235/236, this effect was markedly reduced after 2 weeks of treatment indicative of apparent tolerance development to β₂‐adrenergic stimuli in skeletal muscle. Tolerance development is a well‐described phenomenon after repeated exposure to β₂‐agonist within respiratory medicine (Sato et al., 2015), which can occur rapidly after only a few days of treatment (Salpeter et al., 2004). Thus, the muscle hypertrophic actions of clenbuterol probably gradually diminish over time.

We employed both chronic administration of clenbuterol throughout a 2 week cycle to assess longitudinal adaptations, as well as administration of clenbuterol on trial days during the first and last day of administration to assess acute changes in heart rate, blood pressure and muscle signalling in response to repeated exposure. Although this enabled us to assess tolerance development to repeated clenbuterol ingestion, it cannot be excluded that ingestion of clenbuterol on trial days may have affected the performance tests. However, the detrimental effect on cardiorespiratory outcomes observed for clenbuterol after the 2 week period coincides with other β₂‐agonist studies in which ${\dot{V}}_{{\mathrm{O}}_2\max }$ was assessed after a drug washout (Hostrup, Onslev, et al., 2018; Hostrup et al., 2023; Jessen et al., 2023; Lemminger et al., 2019). Likewise, acute intake of β₂‐agonists does not affect ${\dot{V}}_{{\mathrm{O}}_2\max }$ (Elers et al., 2012). Thus, the acute ingestion of clenbuterol on trial days probably did not influence the ${\dot{V}}_{{\mathrm{O}}_2\max }$ measurements.

In conclusion, the present study shows that a 2 week clenbuterol cycle not only induces lean mass gain and muscle protein accretion in young men, but also compromises cardiorespiratory fitness and induces tolerance development of β₂‐adrenergic signalling and RpS6 phosphorylation in skeletal muscle. Clenbuterol did not exert any enhancing effect on sprint power output, but impaired exercise capacity. Given its negative impact on ${\dot{V}}_{{\mathrm{O}}_2\max }$ and muscle oxidative capacity, clenbuterol misuse is not without pitfalls for exercise performance. Its adverse effects on cardiorespiratory fitness along with its muscle mass accretion properties justify the prohibition of clenbuterol in elite sports.

Competing interests

The authors declare that they have no competing interests.

Author contributions

M.H. and S.J. conceived the study. M.H., S.J. and Y.D. designed the study. M.H., L.M., S.J., M.F., K.W., M.P., A.B‐A and J.B. performed the experiments and collected data. M.H., L.M., S.J., M.F., AB‐A and J.B. analysed data. M.H. wrote the first draft. All authors critically revised the manuscript and approved the final version of the manuscript submitted for publication.

Funding

The study was supported by Anti Doping Denmark (principal investigator: M. Hostrup).

Supporting information

Peer Review History

Biographies

Morten Hostrup, MD, PhD, is Associate Professor and Head of the Clinical and Experimental Physiology group at the Department of Nutrition, Exercise and Sports (NEXS), University of Copenhagen. His research focuses on human integrative physiology with an emphasis on metabolism, performance and health. His group conducts translational studies across diverse populations, from elite athletes to patients with chronic conditions.

Lukas Moesgaard is a PhD fellow at NEXS. His research focuses on skeletal muscle hypertrophy and elucidating skeletal muscle adaptations that accompany muscle hypertrophy.

Data availability statement

All data supporting the results are presented in the published paper.