Introduction

Aging is associated with the reduction of all health-related fitness components, in particular with such cardio-respiratory components as VO2max, and with changes in body composition (Mazzeo et al. 1998). Moreover, an increase in fat mass and a decrease in fat-free mass are associated with impaired glucose and lipid metabolism which is a condition characterised by an elevated risk of cardiovascular and metabolic diseases. Physical inactivity, on the other hand, is an independent risk factor for chronic diseases and a strong mortality predictor (Pedersen and Saltin 2006).

An active lifestyle and exercise training reduce the risk of cardiovascular and metabolic diseases (Kim et al. 2007; Ostergard et al. 2006). The beneficial effects of structured exercise sport on cardiovascular and health-related fitness are reviewed in Krustrup et al. (2010a, b). Football is an intermittent exercise that involves multiple anaerobic actions of high-intensity interspersed with periods of low-intensity recovery. It is also the most popular team sport in the world (Krustrup et al. 2010a, b). Football and recreational soccer training improves cardio-respiratory fitness and the oxidative component of muscle fibres in healthy and unhealthy individuals (Helge et al. 2014; Krustrup et al. 2010a, b; Randers et al. 2012; Schmidt et al. 2013). Despite growing evidence that football training exerts a broad spectrum of beneficial effects on cardiovascular, metabolic and musculoskeletal health variables, only one study has analysed the molecular effects of long-term (1 year) football training on muscle metabolism in healthy subjects (Alfieri et al. 2015). A recent study of the effects of lifelong football training on cardiovascular fitness and health-related body composition parameters showed that elderly subjects (veteran football players; VPG) had better cardiovascular fitness and a healthier body composition than untrained age-matched controls (control group, CG) (Schmidt et al. 2015).

Starting from these experimental findings, we assessed the effects of lifelong football training in VPG vs CG on the muscle expression of key markers involved in (1) mitochondrial biogenesis and oxidative metabolism; (2) DNA repair promotion and senescence suppression, which are two pathways involved in healthy longevity.

Methods

Subjects and sample collections

Twenty healthy male volunteers aged 65–71 years were enrolled at the University of Copenhagen and divided into two groups: 10 veteran football players (VPG; 68.2 ± 3.0 years) who in the last 10 years had trained one session per week (1.5 ± 0.6 h/session) and played in 26 ± 12 football matches (2 × 35 min) per year in local football clubs in Copenhagen; 19 active healthy untrained men (CG; 66.7 ± 1.3 years).

The inclusion criterion for study entry was the ability of volunteers to perform the battery of tests. The exclusion criterion was a history or symptoms of cardiovascular disease or cancer, type 2 diabetes, hypertension, nephropathy or musculoskeletal complaints that were considered to preclude testing. Clinical, biochemical and fitness parameters were evaluated as reported previously (Schmidt et al. 2015). Blood samples were withdrawn from a cubital vein, between 7 and 9 a.m., under standardised conditions after an overnight fast. Total cholesterol, fasting blood glucose, HDL cholesterol, LDL cholesterol and triglycerides were determined with an automated analysers (Cobas Fara, Roche, Neuilly sur Seine, France). All subjects were informed about any potential discomforts or risks related to the experimental study protocol and gave their written informed consent to the study. The study was conducted according to the Declaration of Helsinki and was approved by the local ethics committee of the University of Copenhagen; H-1-2011-013. ClinicalTrials.gov Identifier: NCT01530035.

Body composition was assessed between 7 and 9 a.m. under standardised conditions after an overnight fast using a whole-body dual-energy X-ray absorptiometry scan (Prodigy Advance, Lunar Corporation, Madison, Wisconsin, USA); the effective radiation dose was 0.6 mSv/scan. Height, total body weight (kg), total lean body mass (kg), total fat percentage (fat%) and bone mass (kg) were recorded and calculated using Encore 2006 software (Lunar Corporation).

A biopsy was obtained from the vastus lateralis muscle under local anaesthesia using the Bergstrom technique as previously described (Schmidt et al. 2015) and immediately frozen in liquid nitrogen. Maximal oxygen uptake—VO2max—(Oxycon Pro, VIASYS Healthcare, Hoechberg, Germany) and resting heart rate (RHR; bpm) (Polar Team System, Polar electro Oy, Kempele, Finland) were measured during a standardised cycling protocol performed on an electronically braked ergometer bike (Monark E839, Varberg, Sweden). The subjects started exercising at a cadence of 80 rpm with a workload of 40 W, and the workload was increased by 20 W every 2 min until volitional fatigue. Oxygen uptake was measured continuously and VO2max determined as the highest value achieved in 30 s. Diastolic and systolic blood pressure (BP, mmHg) were also measured.

RNA extraction and RTqPCR

Total RNA was extracted from the muscle biopsies using an miRNeasy® kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. The RNA integrity number (RIN) of samples was assessed using a Bio-Rad Experion automated electrophoresis station (Hercules, CA, USA) before cDNA synthesis. For reverse transcription-PCR (RT-PCR) analysis, 0.5 µg of total RNA was used in the reaction with SuperScript reverse transcriptase (Life Technologies). The resulting cDNA was analysed by real-time quantitative PCR (RTqPCR) performed with the iQ5-iCycler Optical System (Bio-Rad, Hercules, CA, USA). IQ SYBR Bio-Rad protocol (100 mM KCl, 40 mM Tris–HCl pH 8.4, 0.4 mM dNTPs, iTaq DNA polymerase 50 U/ml, 6 mM MgCl2, SYBR Green I, 20 nM fluorescein, stabiliser) was applied according to the manufacturer’s instructions. Reaction mixtures were incubated at 95 °C for 30 s, followed by 2 cycles at 95 °C for 30 s and 95 °C for 3 min and by 40 cycles at 95 °C for 15 s and 60 °C for 1 min. Finally, 80 cycles were run starting at 55 °C and increasing the temperature by 5 °C every 10 s up to 95 °C. Fluorescence signals were measured during the elongation step. All measurements were performed in duplicate. The target mRNA expression levels were normalised to the levels of the peptidylprolyl isomerase A (cyclophilin A) (PPIA) and polymerase (RNA) II (DNA directed) polypeptide A (Pol2A) gene using the 2− ΔΔCT method (Liss 2002). The oligonucleotide primer sequences used in RTqPCR are reported in Supporting Information (Table 1S).

Table 1 Clinical–biochemical characteristics of subjects enrolled in this study
Full size table

Quantification of the mitochondrial DNA to nuclear DNA (mitDNA/nucDNA) ratio

Approximately 10 mg of muscle biopsies was freeze-dried. Genomic DNA was extracted using DNeasy Blood & Tissue columns (Qiagen, Valencia, CA, USA) according to the manufacturer’s instructions. Total DNA concentration was quantified spectrophotometrically (NanoDrop). The mitDNA/nucDNA ratio was determined by quantifying the expression of mitochondrial DNA and nuclear DNA by real-time PCR (Bio-Rad,) as previously described (Nishioka et al. 2004). Purified total DNA (1 ng for mitochondrial DNA amplification and 10 ng for nuclear DNA amplification) was amplified in a 25-µL PCR containing SYBR Green Master Mix (Bio-Rad) and 100 nM of each primer. RTqPCR for the detection of mitDNA and nucDNA was performed as two separate reactions, but within the same plate run for each sample. All samples were run in duplicate for each DNA target. The primers were human specific and designed to target mitDNA (forward: CTCGCACGGACTACAACC; reverse: TGGGCGATTGATGAAAAG) or melanocortin (forward: CGCCGTGGACCGCTACATCT; reverse: GCACGGCCATGAGCACCAG). Serial dilutions of pooled DNA were run for each DNA target to generate a standard curve allowing assessment of RTqPCR amplification efficiency and quantification of mitDNA and nucDNA in relative copy number based on respective threshold cycle (Ct) values.

Citrate synthase activity

Citrate synthase activity, which is a widely used marker of mitochondrial mass and cellular oxidative capacity, was determined as reported elsewhere (Faloona and Srere 1969; Srere and Brooks 1969) with minor modifications. Activity assays were conducted in a final volume of 1 mL containing 0.1 M tris–HCl pH 8, 1 mM DTNB, 0.3 mM acetyl CoA and 0.5 mM oxaloacetate. Mitochondria were prepared from muscle biopsies using a Qproteome Mitochondria Isolation kit (Qiagen) in accordance with the manufacturer’s instructions. Mitochondrial samples ranged between 5 and 10 µg. Assays were performed at room temperature in triplicate using a 50BIO UV–Visible spectrophotometer (Varian). Change in absorbance at 412 nmol/L was measured every 30 s for 3 min. The data were analysed using Student’s t test. The level of significance was set at p < 0.05.

Western blotting

The muscle biopsies were mechanically pulverised, solved in RIPA buffer (150 mM NaCl, 1% NP40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulphate, 50 mM Tris pH 8.0) and centrifuged at 14,000 rpm for 20 min. Supernatants were withdrawn and aliquots (1 µL) used for protein determination using a Bio-Rad Protein Assay kit (Bio-Rad). Protein extracts (each 40 µg) were resolved with 8–12% polyacrylamide gels. High molecular weight myosin α and β heavy chain protein isoforms (MyHC-α and MyHC-β) were separated on 4–20% precast gradient polyacrylamide gels (Bio-Rad), transferred to the Hybond ECL nitrocellulose membrane (GE Healthcare) and checked by Ponceau S staining to verify equal loading. The membranes were immunoblotted using antibodies against adenosine monophosphate-activated protein kinase (AMPKα), AKT serine/threonine kinase 1 (Akt), p42-44 mitogen-activated protein kinase (MAPK or Erk1/2), Forkhead Box M1 (FoxM1)polyclonal antibodies (1:1000; Cell Signaling, Technology), monoclonal mouse antibodies anti-MyHC-α and anti-MyHC-β (Abcam, 1:500), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (1:1000; Santa-Cruz Biotechnology Inc.). Blots were incubated with appropriate horseradish peroxidase-conjugated secondary antibody and target proteins were visualised by ECL detection (GE Healthcare). Densitometric measurements were carried out using Quantity One software (Bio-Rad) as reported elsewhere (Spaziani et al. 2014). GAPDH protein was used to estimate the total amount of loaded proteins. Results were normalised as a percentage of the mean of controls in each membrane.

Statistical analysis

Data were reported as means ± SEM. Relative mRNA expression was reported as relative quantitation (RQ) values, calculated as 2−ΔΔCt, where ΔCt is calculated as Cttarget gene − Cthousekeeping genes (PPIA and PolR2A mRNA expression) and analysed with Student’s t test. Relative protein abundance of Erk 1/2, FoxM1 and Akt was calculated with respect to GAPDH protein abundance and analysed with Student’s t test. All variables were tested for normal distribution by the Kolmogorov–Smirnov test. Pearson’s correlation was used to determine the correlation coefficient between molecular biomarkers and clinical variables. Multiple linear regression analysis was performed to determine the association between variables after correction for BMI and age. Differences were considered statistically significant at p < 0.05.

Results

Effect of lifelong football training on body composition, VO2max, and on the glucose and lipid profile

The clinical–biochemical characteristics of subjects enrolled in this study are reported in Table 1. The VPG had a healthier body composition than the CG. In fact, they had a lower body weight (78.2 ± 6.5 vs 91.2 ± 11.2 kg; p < 0.01), less body fat% (22.6 ± 8.0 vs 31.4 ± 5.0; p < 0.01), a lower BMI (24.4 ± 1.6 vs 28.8 ± 4.0 kg m−2; p < 0.01), and a higher VO2max by body weight (34.7 ± 3.8 vs 27.3 ± 4.0 ml/min/kg; p < 0.001). Moreover, as shown in Table 1, the glucose and lipid profile was healthier in the VPG than in the CG. In fact, we found lower levels of fasting blood glucose, total cholesterol, LDL cholesterol and triglycerides, higher HDL cholesterol values, and a lower resting heart rate in the VPG than in the CG.

Lifelong football training induces upregulation of key markers involved in muscle mitochondrial biogenesis and oxidative metabolism

We measured mRNA and protein expression levels of the two catalytic isoforms of AMPK (AMPKα1 and α2) that regulate the muscle intermediate metabolism. As shown in Fig. 1a, b, compared with the CG, the VPG expressed significantly higher amounts of α1 and α2 messengers (p < 0.05 and p < 0.001, respectively) and of proteins (p < 0.05). The analysis of the expression of key messengers and proteins involved in muscle mitochondrial biogenesis and oxidative metabolism revealed increased expression of PGC1α (p < 0.01), NAMPT (p < 0.01) and TFAM (p < 0.01) messengers in VPG muscle compared to CG (Fig. 1c). We also found a significant increase in citrate synthase activity (p < 0.05), overexpression of the MyHC-β isoform (p < 0.05) and a positive expression trend of the MyHC-α isoform in VPG compared to CG muscle (Fig. 1d, e). Notably the levels of FNDC5 and mTOR messengers were higher in VPG than in CG, although the difference was not significant (Fig. 1f). Similarly, the mitDNA/nucDNA ratio in muscle did not differ significantly between VPG and CG (Fig. 1g), which suggests that increased oxidative metabolism was not due to a greater number of mitochondria in VPG muscle. The apparent discrepancy between the similar number of mitochondria in VPG and CG muscles (Fig. 1g) and the increased oxidative activity in VPG muscle is probably due to the higher number of type 1 fibres (as witnessed by the higher MyHC-β expression) rather than to the increase of number of mitochondria per cell. To investigate the effects of lifelong football training on angiogenesis and muscle capillarisation, we measured the expression of the key messengers eNOS and VEGF in VPG and CG muscle. The expression of both mRNAs was higher in VPG than in CG, although the difference was not significant (Fig. 1h).

Fig. 1
figure 1

Effects of lifelong football training on the expression of markers involved in the muscle metabolism pathway. a Skeletal muscle mRNA and b protein expression of AMPKα1 and α2; c mRNA expression of PGC1α, NAMPT and TFAM; d citrate synthase activity; e MyHC-α and MyHC-β protein expression; f mRNA expression of FNDC5 and mTOR; g mitDNA/nucDNA ratio; h mRNA expression of eNOS and VEGF. Messenger and protein expression, citrate synthase activity and mitDNA/nucDNA ratio were determined in muscle biopsies from 10 CG subjects (black bars) and 10 VPG subjects (grey bars). An arbitrary value of 1 was assigned to the level of expression of each messenger in CG. GAPDH served as loading control. Representative blots are reported for each protein analysed. Data were expressed as relative percentage of protein expression vs CG. a–h Data represent the means (±SEM) of 3 different experiments and were analysed using Student’s t test. Differences were considered significant at *p < 0.05, **p < 0.01 and ***p < 0.001, respectively, compared to CG

Full size image

Lifelong football training affects the muscle expression of key markers involved in DNA repair and senescence suppression

To assess the effects of lifelong football training on healthy longevity, we measured the expression of key proteins involved in the muscle DNA repair promotion and senescence suppression pathways, namely Erk1/2, Akt and FoxM1 in muscle biopsies of the VPG and the CG. All four proteins were significantly higher (p < 0.05) in the VPG than in the CG (Fig. 2a–c). This suggests that lifelong football training has positive effects on these pathways and so contribute to healthy longevity.

Fig. 2
figure 2

Effects of lifelong football training on the expression of key markers involved in DNA repair promotion and senescence suppression pathways. a–c Protein expression levels of Erk1/2, Akt and FoxM1 in muscle biopsies from 10 controls (black bars) and 10 VPG subjects (grey bars) analysed by western blotting. GAPDH served as loading control. Representative blots are reported for each protein analysed. Data were expressed as percentage of CG. Data were compared using Student’s t test and represent the means (±SEM) of 3 different experiments. Differences were considered significant at *p < 0.05 vs CG

Full size image

Correlation of the study variables

Pearson’s correlation analysis showed correlations between most of the molecular markers and clinical and biochemical variables examined (Table 2). In particular, fat% was negatively associated with Erk1/2 (r −0.52; p = 0.04) and Akt (r −0.58; p = 0.02) protein expression and with VO2max (r −0.71; p < 0.001). Conversely, VO2max was positively associated with citrate synthase activity (r 0.49; p = 0.03), TFAM (r 0.55; p = 0.01), AMPKα2 (r 0.59; p = 0.01) and PGC1α (r 0.56; p = 0.01) mRNA expression and with Erk1/2 (r 0.66; p = 0.01) protein expression. At BMI- and age-adjusted multiple regression analysis, fat% was independently associated with Akt protein expression (Table 3; p = 0.01) and VO2max was independently associated with TFAM (p = 0.001) mRNA and with Erk1/2 (p < 0.001) protein expression (Table 4).

Table 2 Pearson’s correlation matrix of the variables studied
Full size table
Table 3 Multiple regression analysis showing the molecular biomarkers independently associated with fat percentage
Full size table
Table 4 Multiple regression analysis showing the molecular biomarkers independently associated with VO2max
Full size table

Discussion

To our knowledge, this is the first study to investigate the effects of lifelong football training on the muscle expression profile of molecular markers (genes and proteins) involved in healthy longevity. Skeletal muscle is a plastic tissue, and regular exercise is the most effective tool with which to counteract muscle atrophy pathways (Arany 2008). Exercise induces an increase in several molecular stress signals in skeletal muscle, a process that activates AMPK isoforms. AMPK regulates cellular energy status during exercise: AMPK activation induces downstream upregulation of molecular markers correlated to mitochondrial biogenesis and oxidative metabolism in muscle (PGC1α, NAMPT and TFAM), and to the molecular targets controlling cellular respiration, mitochondrial fusion and transcription (Mounier et al. 2015). AMPK expression levels were lower in muscle from elderly vs young rats, and was associated with decreased mitochondrial content (Reznick et al. 2007) and with a lower mitochondrial oxidative capacity (Lantier et al. 2014), which collectively contribute to an age-related decline in metabolic active muscle mass.

We previously demonstrated that long-term football training (64 weeks) increased the muscle expression of AMPKα1 and α2 messengers in trained vs untrained healthy adults (Alfieri et al. 2015). Here, we provide evidence that lifelong football training counteracts the reduction in muscle mitochondrial oxidative capacity, which is a hallmark of the aging process. In fact, the expression of the AMPKα1 and α2 isoforms was higher in VPG muscle than in controls. Also the downstream markers of the muscle oxidative metabolism pathway, i.e. isoforms PGC1α, TFAM, NAMPT and MyHC-β, were significantly higher in muscle from veterans vs untrained healthy elderly subjects. In addition, mitochondrial citrate synthase enzyme activity, a key marker of muscle oxidative capacity (Gerhart-Hines et al. 2007; Short et al. 2003) was higher in VPG muscle than in control muscle (see Fig. 3). The sustained increase in PGC1α and TFAM expression was closely correlated with reduction of fat mass percentage and to a healthier body composition and metabolic profile (lipid and glucose) in VPG but not in controls. This suggests that lifelong football training is a tool with which to induce and sustain muscle oxidative capacity in the elderly. Several studies reported that concurrent endurance and resistance exercise contributes to the regulation of the expression of two molecular markers (AMPK and PGC1α) and that this increases muscle oxidative capacity (Egan et al. 2010; Skovgaard et al. 2016). Furthermore, increased PGC1α expression in muscle is the initial event in the regulation of oxidative metabolism and exercise intensity influences its expression (Nordsborg et al. 2010). Most of the above-cited studies focused on the acute effects of training in young or middle-aged subjects. Here we demonstrate that lifelong football training, which has been proven to be a combination of endurance training and strength training (Krustrup et al. 2010a, b), induces similar adaptations in oxidative metabolism in elderly muscle also in a chronic condition. However, we cannot rule out the possibility that similar adaptation can be obtained in elderly subjects trained at the same metabolic intensity for a minimum of 6 months.

Fig. 3
figure 3

Schematic representation of the effects of lifelong football training on molecular pathways related to longevity. Long-term football training induces upregulation of key markers involved in the mitochondrial biogenesis, oxidative metabolism pathways (AMPK, PGC1α, NAMPT and TFAM) and in DNA repair promotion and senescence suppression (Akt, ERK1/2 and FoxM1) in muscle from VPG. Blue increased expression. Grey non-analysed molecules. Pink mildly increased expression. Green arrows positive input. Red arrows negative input

Full size image

Our results confirm the previously reported improved cardio-respiratory fitness in the VPG vs the CG (Schmidt et al. 2015). In particular, the expression of protein MyHC-β, the peculiar isoform of slow, oxidative skeletal muscle fibres (Paoli et al. 2013), was higher in the VPG than in controls; this finding is correlated to the increased VO2max observed in VPG.

Nitric oxide reduction bioavailability is an hallmark of endothelial dysfunction, which in turn is an independent risk factor for adverse cardiac events in the elderly (Nyberg et al. 2012). Lifelong endurance training counteracts the age-related decline in nitric oxide bioavailability (Rubinshtein et al. 2010). Interestingly, we found increased levels of the angiogenesis and capillarisation markers eNOS and VEGF messengers in VPG muscle.

We also investigated the effects of lifelong football training on skeletal muscle aging. Erk1/2 and Akt are two of the most widely studied longevity markers. Erk controls muscle cell proliferation and differentiation thereby playing an important role in the myogenic response to exercise (Carlson et al. 2001; Parkington et al. 2004). Thus far, the effect of Akt on suppression of muscle senescence has been controversial. Indeed, increased Akt expression in adult mouse muscle models was correlated with growth and hypertrophy (Bodine et al. 2001; Pallafacchina et al. 2002), whereas Akt expression did not differ between muscle from elderly active and muscle from sedentary subjects (Sandri et al. 2013). In the present study, Erk1/2 and Akt expression was higher in VPG muscle than in muscle from active, untrained elderly subjects. This finding suggests that there is an adaptive response of skeletal muscle to long-term exercise training that counteracts the consequences of senescence (see Fig. 3). In line with our findings, Frosig et al. (2007) reported that Akt expression was higher in muscle from trained elderly subjects that in untrained elderly subjects. Moreover, AMPK and Akt, respectively, negatively and positively regulated mTOR expression (Gwinn et al. 2008; Manning et al. 2002), which is a master regulator of cell growth and longevity (Harrison et al. 2009; Katewa and Kapahi 2011). In fact, mTOR forms two different protein complexes, mTORC1 and mTORC2. Inhibition of mTORC1 has been reported to extend lifespan and delay aging in various organisms (Sharples et al. 2015), whereas inhibition of mTORC2 negatively affects mammalian health and longevity (Arriola Apelo and Lamming 2016). In mouse models, mTOR expression was higher in the skeletal muscle of aging mice than in young (Lamming et al. 2012; Reznick et al. 2007). The role of mTOR in human muscle aging and sarcopenia is not completely understood. In fact, no differences were identified in mTOR expression in muscle among young, active and sedentary old subjects due to interindividual variability (Sandri et al. 2013). Accordingly, the muscle expression of mTOR messenger did not significantly differ between VPG subjects and our active untrained controls.

The FoxM1 transcription factor regulates a broad range of biological functions, including cell proliferation, cell cycle progression, renewal, differentiation, cell migration, and angiogenesis. FoxM1 overexpression has been reported in liver, prostate, brain, breast and colon cancers (Zona et al. 2014). It is also involved in senescence suppression and in the DNA damage repair process, being particularly active in case of age-dependent accumulation of DNA damage (Behrens et al. 2014). Reduced FoxM1 expression has been associated with an age-related decline in cell proliferation, whereas increased FoxM1 expression has been implicated in the promotion of stem cell renewal, DNA repair and senescence suppression (Bella et al. 2014).

This is the first study to investigate the effects of lifelong training on the expression of molecular markers involved in DNA repair and senescence suppression in skeletal muscle. Here we demonstrate that lifelong football training positively affects the expression of checkpoint markers involved in the DNA repair and senescence suppression pathways, including FoxM1 (see Fig. 3). A limit of this study is the small number of participants and muscle biopsies analysed.

Conclusions

Lifelong football training is a tool with which to modulate expression levels of key messengers involved in muscle oxidative metabolism, DNA repair and senescence suppression. Our results confirm the improvements in body composition and cardiovascular or microvascular function and the healthier glucose and lipid profile found in VPG subjects (Schmidt et al. 2015). All these factors counteract the risk of metabolic diseases and negative cardiovascular events and promote healthy longevity. The current study provides further evidence that lifelong football training can be useful to counteract metabolic decline in the elderly.