When the energy balance is negative, as during dieting for weight loss or in several conditions in patients, ∼25% to 40% of the total reduction in weight is due to loss of skeletal muscle [1
]. In critical patients, the loss of muscle mass is associated with unfavorable clinical outcomes [3
]. In obese and overweight, an excessive loss of fat-free mass (FFM) may also be detrimental since skeletal muscle accounts for a significant fraction of the resting metabolic rate, and is essential for exercising and metabolic health [5
]. Accordingly, there is considerable interest in minimizing the loss of FFM in people dieting and in admitted patients with energy deficit [6
]. Exercise and increasing the ratio of protein-to-carbohydrate in the diet attenuate the loss of FFM during mild-to-moderate energy deficits when dieting [7
], but the underlying molecular mechanisms remain largely unknown. In patients with septic shock, a mild exercise program halved the loss of muscle mass and was associated with an attenuation of an excessive activation of autophagy [6
]. However, multiple administration of drugs, including insulin, limits the interpretations of these results [6
]. Moreover, it remains unknown what is the optimal amount of exercise needed to prevent the loss of muscle mass caused by a negative energy balance.
A negative energy balance elicits an overall increase in whole-body proteolysis, amino acid oxidation, and nitrogen excretion [8
], accompanied by a reduction in muscle protein synthesis [11
]. Compared to muscle protein synthesis, little attention has been given to the molecular mechanisms driving the muscle proteolytic response to energy deficit and its interaction with exercise [6
], and no study to date has addressed the molecular regulation of protein degradation during extreme energy deficits in human skeletal muscle.
Protein degradation in skeletal muscle primarily depends on the activation of the two major proteolytic pathways, the ubiquitin-proteasome system (UPS) and the autophagy-lysosome system (ALS). These two systems are coordinated by the members of the forkhead box O family of transcription factors (FoxOs), particularly by FoxO1and FoxO3a [15
The UPS requires ubiquitin ligase enzymes (E3s) to catalyze the attachment of polyubiquitin chains to the protein substrate, which then is targeted to the 26S proteasome for degradation. Among the E3s, the gene expression of muscle ring finger 1 (MuRF1) and muscle atrophy F box (MAFbx or Atrogin 1) is upregulated in animal and human models of muscle atrophy [16
], although human data are scarce and less consistent [3
]. One of the proteins targeted by MAFbx for subsequent proteasome degradation during protein breakdown is the eukaryotic translation initiation factor 3 subunit f (eIF3f), which plays a pivotal role in the regulation of muscle mass [19
]. Protein degradation through the ALS is required for muscle mass maintenance [20
] and a physiological adaptation to regular exercise [21
]. Adenosine monophosphate-activated protein kinase (AMPK) activates autophagy and is necessary for exercise-mediated autophagy induction [24
]. The formation of the autophagosome is elicited by the unc-51-like kinase 1 (ULK1), which is activated and inhibited by AMPK and mTOR, respectively [26
]. ULK1 activates Beclin 1 through its phosphorylation on Ser15, resulting in Beclin 1-Vps34 (vacuolar sorting protein 34) complex activation, a step necessary for complete autophagy induction [27
]. Nucleation and expansion of the autophagosome require microtubule-associated protein 1 light chain 3 beta I (LC3BI) lipidation to LC3BII, which is recruited to the autophagosomal membrane [28
]. Ubiquitinated proteins and LC3BII bind to the sequestosome 1 (p62/SQSTM1) for degradation together with the autophagosome content following fusion with the lysosome [28
Studies on short-term energy deficit have found little or no change in gene and protein expression markers of protein breakdown [11
], and the scarce evidence from direct measurements of muscle protein breakdown (MPB) have reported conflicting results [12
]. Particularly in overweight populations, it has been stated that the loss of muscle mass during a negative energy balance is caused more by a reduction in protein synthesis rather than an increase in protein degradation [13
]. The high energy demand during severe energy deficits elicited by the combination of prolonged exercise and caloric restriction could evoke a greater activation of skeletal muscle protein degradation, but experimental data on humans are not available.
Skeletal muscle ALS and UPS signaling pathways could be upregulated after ultra-distance races, although results are inconclusive regarding the key signaling markers and the role played by the E3s [31
]. However, it remains unknown how these signaling pathways are regulated in muscles of the upper extremities in humans, i.e., not involved in the exercise.
In humans, the mechanisms regulating protein degradation are influenced by feeding, and essential amino acids (EAA), leucine in particular, and have been shown to reduce skeletal muscle MuRF1 and MAFbx mRNA expression following acute endurance exercise [33
] and resistance exercise [34
]. Moreover, insulin and amino acids downregulate autophagosome formation in human skeletal muscle [35
]. Nevertheless, it remains unknown how different levels of contractile activity and provision of amino acids modulate the UPS and ALS responses during severe energy deficits and the role of both systems on the loss of muscle mass during severe energy deficits.
Therefore, the current investigation was designed to characterize the mechanisms by which exercise attenuates the loss of muscle mass during a severe energy deficit of ~5500 kcal/d for four days (Figure 1
). The effect of local contractile activity on UPS and ALS signaling pathways was examined employing an intra-subject experimental design in which biopsies were taken from three different muscles subjected to different exercise volumes. Our subjects ingested a very low-calorie diet containing exclusively whey protein or sucrose, while performing 45 min of single-arm cranking (moderate volume) and eight hours of walking daily (high volume), with muscles in the contralateral arm serving as non-exercised controls. We hypothesized that exercise would attenuate the expression of markers of UPS and ALS during severe energy deficit in a dose-dependent fashion. Secondarily, we also hypothesized that compared to carbohydrates, the ingestion of whey protein alone would inhibit to a greater extent the proteolytic response.
2. Materials and Methods
A thorough characterization of our study population and experimental procedures can be found in the five previous published articles derived from the experiments carried out in this investigation [37
]. A summary of our participants’ background characteristics is shown in Table 1
. All participants received full oral and written information on the potential risks of taking part in the study and gave written informed consent prior to the start. The study was approved by the Regional Ethical Review Board of Umeå University (Umeå, Sweden).
2.1. Experimental Protocol
The protocol comprised the following three different experimental phases: A baseline phase (PRE), a phase of caloric restriction combined with exercise for four days (CRE), and a phase of controlled diet with reduced levels of exercise (CD) (Figure 1
). The sample size was calculated to disclose any significant difference ≥1.5-fold larger than the coefficient of variation (which was <10% in most cases) between the mean values for any individual variable, with a significance level of p
< 0.05 and statistical power of 0.8. Assessment of body composition by dual-energy X-ray absorptiometry (Lunar iDXA, GE Healthcare, Madison, WI, USA), extraction of 20 mL blood samples (in the supine position) and three muscle biopsies (one from each deltoid muscle, posterior portion, and one from the middle portion of the vastus lateralis) were obtained following a 12 h overnight fast during PRE. The biopsies following the CRE and CD phases were taken in the morning (i.e., 08:00 a.m.) on the next day after the end of the corresponding phase following a 12 h overnight fast (Figure 1
Participants were randomly assigned to ingest a very low-calorie diet (0.8 g/kg body weight/day) consisting solely of sucrose (n
= 7) or whey protein (n
= 8) (Syntrax Nectar, Syntrax Innovations, Scott City, MO, USA) during caloric restriction phase (CRE). On each CRE day, participants performed 45 min of one-arm cranking (at 15% of maximal intensity), followed by eight hours of walking. The deltoid muscles were chosen as representative of upper limb musculature because their fiber type composition is similar to that of vastus lateralis [42
], and both muscles adapt similarly to prolonged low-intensity endurance training [43
]. It has also been reported that, despite a substantially higher proportion of type II fibers in the triceps brachii as compared to vastus lateralis, both muscles adapt to endurance training in a similar manner [44
]. The whey protein solution also contained Na+
(308 mg/L) and K+
(370 mg/L), as did the sucrose solution (160 and 100 mg/L, respectively). Either solution was dissolved in 1.5 L containing minerals and split in three intakes of 0.5 L in the morning (just preceding arm-cranking), and, subsequently, at midday and 8 PM (at the end of the walk). Throughout the walks, groups were allowed to drink a hypotonic rehydrating solution containing Na+
(160 mg/L), Cl−
(200 mg/L), K+
(100 mg/L), citrate (700 mg/L), and sucrose (3g/L) ad libitum
During the three-day CD phase, all participants daily ate three standardized meals reproducing their regular energy intake (as assessed by weighing all food ingested during the 7-day pre-test period), and their physical activity was restricted to a maximum of 10,000 steps. This phase was designed to allow replenishment of body water and stabilization of body weight.
2.2. Assessment of Physical Activity, Nutrition, and Body Composition
Daily energy expenditure due to physical activity was analyzed using the short version of the International Physical Activity Questionnaire [45
]. During the 7-day PRE phase, the participants kept a dietary record, and their food intake was analyzed (Dietist XP, Kost & Näringsdata, Bromma, Sweden). During the CD phase, all participants were provided a diet with the same energy content as that recorded during PRE, and the food ingested was weighed. Energy intake was also calculated employing the Dietist XP program. During the PRE phase, the participants ingested a mixed diet (18.5%, 41.6%, and 39.9% protein, fat, and carbohydrate, respectively, in the sucrose group; and 19.5%, 38.3%, and 42.2%, protein, fat and carbohydrate, respectively, in the whey protein group) for a total energy intake of 2256 ± 513 and 2086 ± 450 kcal/d (means ± SD), for the sucrose and whey protein groups, respectively.
The energy deficit was calculated from the reduction in the energy content of the body using the changes in body composition, assigning energy values of 9416.8 and 884.3 kcal/kg to fat and lean mass, respectively [40
2.3. Hormonal and Biochemical Analyses
After a 12 h overnight fast, 20 mL blood samples were drawn from an antecubital vein directly into Vacutainer tubes (REF: 368499; 368498; BD Vacutainer, Stockholm, Sweden). Some samples were collected in tubes containing EDTA and centrifuged for 5 min at 2000 g and 4 °C, to obtain plasma; while others were centrifuged for 10 min at 2000 g and 4 °C to prepare serum. All of these samples were aliquoted on tubes precooled on ice water and rapidly stored at −80 °C until analyzed.
The concentrations in serum of glucose, insulin, leptin, cortisol, total testosterone, free testosterone, and plasma amino acids were determined as previously reported [38
]. HOMA index was calculated as the fasting plasma concentration of insulin (μU/mL) × the corresponding concentration of glucose (mmol/L)/22.5.
2.4. Biopsy Sampling
Three muscle biopsies were taken from the middle portion of each deltoid muscle and vastus lateralis using Bergstrom’s technique with suction, as described elsewhere [37
]. After disinfection of the skin, 1 mL to 2 mL local anesthetic (Lidocaine 2%) was injected into the skin and subcutaneous tissue, taking care not to penetrate below the superficial fascia. After that, a 6 mm to 7 mm incision was made, and the biopsy Bergstrom-type needle inserted. The muscle sample (~100 mg) was dissected free of any debris and fat tissue present and immediately frozen in liquid nitrogen and stored at −80 °C until further analysis.
2.5. Protein Extraction and Western Blotting
Extracts of muscle protein were prepared as previously described [48
], and total protein content quantified using the bicinchoninic acid assay [49
]. Briefly, 30 mg of muscle was homogenized in urea lysis buffer (6 M urea, 1% SDS and 1X cOmplete protease inhibitor and phosphatases PhosphoStop 1X) and the lysate then centrifuged for 12 min at 25,200 g
at 16 °C. The resulting supernatant containing the protein fraction, was diluted with electrophoresis loading buffer (62.50 mM Tris-HCl, pH 6.8, 2.3% SDS, 10% glycerol, 5% β-mercaptoethanol, and bromophenol blue).
The optimal antibody concentration and the total protein amount to be loaded was first determined by loading a gradient of protein extracts at concentrations between 15 and 35 μg. The linear relation between total protein concentration loaded and quantitative band intensity was calculated. After confirming linearity in this range, equal amounts for the same protein determination (30 to 35 μg) of each sample were electrophoresed with SDS-PAGE and transferred to Immun-Blot PVDF membranes for protein blotting (Bio-Rad Laboratories, Hercules, CA, USA). To compensate for variability between gels, the nine samples from each subject and a control sample in quadruplicate (human muscle) were equally loaded onto the same gel. When the coefficient of variation for the control samples was >20% the blot was repeated. The sample protein bands were normalized to the mean value of the control sample.
Membranes were blocked for one hour in 4% bovine serum albumin in Tris-buffered saline containing 0.1% Tween 20 (TBS-T) (BSA-blocking buffer) and incubated overnight at 4 °C with primary antibodies. In the case of the protein markers whose phosphorylation was assessed, the membranes were initially incubated with the phospho-specific antibodies and, after detection, the antibodies were stripped off using a buffer containing Tris-HCL 1 M pH 6.7, SDS 20%, β-mercaptoethanol 14.3 M and H2O and subsequently re-blocked and re-incubated with antibodies against the total protein. Antibodies were diluted in 4% BSA-blocking buffer or 5% Blotto blocking buffer. After incubation with primary antibodies, the membranes were incubated with an HRP-conjugated antibody (diluted 1:5000 in Blotto blocking buffer in all instances) and subsequent chemiluminescent visualization with Clarity™ Western ECL Substrate (Bio-Rad Laboratories (Hemel Hempstead, Hertfordshire, UK) using the ChemiDoc Touch Imaging System (Bio-Rad Laboratories, Hercules, CA, USA) until saturation of the signal. Finally, band densitometric data were quantified in the exposition immediately prior to saturation of the signal with Image Lab© software (Bio-Rad Laboratories, Hercules, CA, USA). To control for differences in loading and transfer efficiency, membranes were stained with Reactive Brown and quantified. In order to verify equal loading, the blots were stripped and probed with a monoclonal mouse anti-α-tubulin antibody. No significant changes were observed in Reactive Brown total protein amount nor in α-tubulin protein levels (data not shown).
The complete protease inhibitor cocktail and PhosSTOP phosphatases inhibitor cocktail were obtained from Roche Diagnostics (Mannheim, Germany; #04693116001 and #04906845001, respectively). The Immun-Blot PVDF membranes, the Immun-StarTM Western CTM and the Protein Plus Precision All Blue Standards were from Bio-Rad Laboratories (Hemel Hempstead Hertfordshire, UK). The corresponding catalog numbers of primary antibodies were as follows: anti-phospho-AMPKα (Ser172), no. 2531, anti-phospho-Beclin 1 (Ser15), no.8466, anti-phospho-FoxO1 (Ser256), no. 9461, anti-phospho-FoxO3 (Ser253), no. 9466, anti-AMPKα, no. 2532; and anti-FoxO1, no. 9454, were purchased from Cell Signaling Technology (Danvers, MA, USA). Anti-eIF3f, no. ab74568; Anti-LC3B, no. ab48394; anti-MuRF1, no. ab77577; anti-p62/SQSTM1, no. ab56416; and anti-phospho-ULK1 (Ser555), no. ab229537 were purchased from Abcam (Cambridge, UK). Anti-MAFbx, no. sc166806 and anti- FKHRL1, no. sc11351 were purchased from Santa Cruz Biotechnology (Dallas, TX, USA), and anti-Atg1/ULK1, no. A7481 and anti-α-tubulin, no. T5168 was obtained from Sigma-Aldrich (St. Louis, MO, USA). The secondary HRP-conjugated goat anti-rabbit (no. sc2030 and sc2004), chicken anti-rabbit (sc516087) and goat anti-mouse (sc2031) antibodies were from Santa Cruz Biotechnology, and the goat anti-rabbit (no. 111-035-144) was purchased from Jackson ImmunoResearch (West Grove, PA, USA).
2.7. Statistical Analysis
The normality of data was checked using the Shapiro–Wilks tests, and when necessary, data were transformed logarithmically before analysis. A three-way repeated measures ANOVA with time (PRE, CRE, and CD) and muscle (control deltoid, trained deltoid, and vastus lateralis) as within-subject factors, and the two different diets (sucrose versus whey protein) as a between-subjects factor was performed as the primary statistical test. The same type of three-way repeated measures ANOVAs were also conducted with two levels for muscle to distinguish between upper and lower extremity muscle adaptations (the average of both deltoid muscles vs. vastus lateralis). Likewise, to compare the responses of the exercised and non-exercise muscles, the same type of three-way repeated measures ANOVAs was carried out with two levels for the muscle factor defined as exercised vs. non-exercised muscles (i.e., the non-exercised deltoid vs. the average of the exercised deltoid and vastus lateralis). To study the overall signaling response in all muscles, the average of the three biopsies was calculated, and a two-way repeated measures ANOVA was run with time (PRE, CRE, and CD) as a within-subject factor and the two different diets (sucrose versus whey protein) as the between-subjects factor. The Mauchly’s test of sphericity was run before the ANOVA. In the case of violation of the sphericity assumption, the degrees of freedom were adjusted according to the Huynh and Feldt test. When a significant main effect or interaction was observed, pairwise comparisons at specific time points were adjusted for multiple comparisons using the Holm–Bonferroni procedure.
Pearson’s correlation analysis was used to examine the associations between changes in signaling and hormones or amino acids using all subjects (i.e., without a separate analysis for the sucrose and whey protein groups). For this purpose, a mean signaling response was calculated for each subject (mean of the three biopsies) and the study phase. The changes from PRE to CRE (n = 15) and PRE to CD (n = 15) were calculated, and these changes were tested for association with the corresponding changes in hormones or amino acids. Additionally, the same approach was used for the average of both deltoid muscles, to represent the mean changes in signaling of the arms. The changes in signaling in the vastus lateralis were also tested independently since this muscle was submitted to an extremely high volume of exercise. Additional correlations were performed combining the changes from PRE to CRE, and from PRE to CD (n = 30).
Results are expressed as mean ± standard deviations (SD) unless otherwise stated. Statistical significance was set at p < 0.05. All statistical analyses were performed using SPSS v.15.0 for Windows (SPSS Inc., Chicago, IL, USA).