Parkin overexpression protects from ageing‐related loss of muscle mass and strength

Abstract

Key points

• Recent evidence suggests that impaired mitophagy, a process in charge of removing damaged/dysfunctional mitochondria and in part regulated by Parkin, could contribute to the ageing‐related loss of muscle mass and function.

• In the present study, we show that Parkin overexpression attenuates ageing‐related loss of muscle mass and strength and unexpectedly causes hypertrophy in adult skeletal muscles. We also show that Parkin overexpression leads to increases in mitochondrial content and enzymatic activities. Finally, our results show that Parkin overexpression protects from ageing‐related increases in markers of oxidative stress, fibrosis and apoptosis.

• Our findings place Parkin as a potential therapeutic target to attenuate sarcopenia and improve skeletal muscle health and performance.

Abstract

The ageing‐related loss of muscle mass and strength, a process called sarcopenia, is one of the most deleterious hallmarks of ageing. Solid experimental evidence indicates that mitochondrial dysfunctions accumulate with ageing and are critical in the sarcopenic process. Recent findings suggest that mitophagy, the process in charge of the removal of damaged/dysfunctional mitochondria, is altered in aged muscle. Impaired mitophagy represents an attractive mechanism that could contribute to the accumulation of mitochondrial dysfunctions and sarcopenia. To test this hypothesis, we investigated the impact of Parkin overexpression in skeletal muscles of young and old mice. Parkin was overexpressed for 4 months in muscles of young (3 months) and late middle‐aged (18 months) mice using i.m. injections of adeno‐associated viruses. We show that Parkin overexpression increased muscle mass, fibre size and mitochondrial enzyme activities in both young and old muscles. In old mice, Parkin overexpression increased muscle strength, primordial germ cell‐1α content and mitochondrial density. Parkin overexpression also attenuated the ageing‐related increase in 4‐hydroxynonenal content (a marker of oxidative stress) and type I collagen content (a marker of fibrosis), as well as the number of terminal deoxynucleotidyl transferase dUTP nick‐end labelling‐positive myonuclei (a marker of apoptosis). Overall, our results indicate that Parkin overexpression attenuates sarcopenia and unexpectedly causes hypertrophy in adult muscles. They also show that Parkin overexpression leads to increases in mitochondrial content and enzymatic activities. Finally, our results show that Parkin overexpression protects against oxidative stress, fibrosis and apoptosis. These findings highlight that Parkin may be an attractive therapeutic target with respect to attenuating sarcopenia and improving skeletal muscle health and performance.

Key points

• Recent evidence suggests that impaired mitophagy, a process in charge of removing damaged/dysfunctional mitochondria and in part regulated by Parkin, could contribute to the ageing‐related loss of muscle mass and function.

• In the present study, we show that Parkin overexpression attenuates ageing‐related loss of muscle mass and strength and unexpectedly causes hypertrophy in adult skeletal muscles. We also show that Parkin overexpression leads to increases in mitochondrial content and enzymatic activities. Finally, our results show that Parkin overexpression protects from ageing‐related increases in markers of oxidative stress, fibrosis and apoptosis.

• Our findings place Parkin as a potential therapeutic target to attenuate sarcopenia and improve skeletal muscle health and performance.

Introduction

One of the most deleterious hallmarks of ageing is the progressive decline of muscle mass and function, a process called sarcopenia. Sarcopenia can have dramatic consequences for afflicted individuals because it progressively leads to mobility impairment, falls and physical frailty (Janssen et al. 2002; Deschenes, 2004; Janssen et al. 2004a; Kim & Choi, 2013). In addition, sarcopenia represents a major burden for various societies. It is estimated that the prevalence of sarcopenia increases from 14% in 65–70‐year‐olds to over 50% in those above 80 years old (Santilli et al. 2014). The number of people around the world aged over 60 years was estimated at 600 million in the year 2000 and is expected to rise to 1.2 billion by 2025 and to 2 billion by 2050. Even with a very conservative estimate of prevalence, sarcopenia currently affects over 50 million people worldwide and will affect over 200 million in the next 30 years (Santilli et al. 2014). Furthermore, the direct healthcare costs attributable to sarcopenia and its sequela are considerable, estimated at 18.5 billion dollars for 2000 alone in the USA (Janssen et al. 2004b). Developing effective therapeutic interventions to counteract sarcopenia is therefore one of the major challenges facing health research.

There is strong evidence that accumulation of dysfunctional mitochondria plays an important role in the skeletal muscle ageing process. Notably, aged skeletal muscles display impaired mitochondrial energetics (Trounce et al. 1989; Short et al. 2005; Gouspillou et al. 2010; Lanza et al. 2012; Gouspillou et al. 2014a) and increased mitochondrial‐mediated apoptosis (Selman et al. 2003; Dirks & Leeuwenburgh, 2004; Leeuwenburgh et al. 2005; Chabi et al. 2008; Marzetti et al. 2008; Picard et al. 2010; Picard et al. 2011; Gouspillou et al. 2014b). Furthermore, overexpression of a mitochondrial‐targeted catalase (antioxidant enzyme) has been shown to attenuate the effect of ageing on skeletal muscle, thereby supporting a causal role for mitochondrial dysfunction in the sarcopenic process (Umanskaya et al. 2014). In addition, the two most efficient non‐pharmacological strategies to attenuate the effects of ageing on skeletal muscle, calorie‐restriction (Mayhew et al. 1998; Baker et al. 2006; Hepple et al. 2006) and endurance training (Coggan et al. 1992; Song et al. 2006), are known to improve mitochondrial function (Gouspillou & Hepple, 2013).

Although the mechanisms underlying this ageing‐related accumulation of dysfunctional mitochondria are not yet fully understood, recent evidence indicates that reduced mitophagy may play a causal role (O'Leary et al. 2013; Gouspillou et al. 2014b). Mitophagy is the selective removal of dysfunctional mitochondria by autophagosomes. This process is regulated to a large extent by the Parkin–Pink1 pathway, which plays an important role in maintaining mitochondrial quality in skeletal muscles by selectively removing dysfunctional and depolarized mitochondria (Narendra et al. 2008). In the context of muscle ageing, it has been reported that the ratio of Parkin to VDAC (voltage‐dependent anion channel; a protein recruiting Parkin to dysfunctional mitochondria for the initiation of mitophagy) (Sun et al. 2012) is significantly reduced in atrophied muscle of old men (Gouspillou et al. 2014b). In addition, it has been reported that autophagy‐related genes, such as BNIP3, BECLIN‐1, ATG7 and PARK2 (the gene coding for Parkin), are downregulated in inactive older women (Drummond et al. 2014). Interestingly, skeletal muscles from adult Parkin knockout mice display some ageing‐like features, including contractile dysfunction, impaired mitochondrial energetics and a sensitization of the apoptosis‐regulating mitochondrial permeability transition pore (Gouspillou et al. 2018). In line with these findings, it was shown that inhibition of mitophagy by selectively deleting autophagy‐related genes causes mitochondrial dysfunction and muscle atrophy (Masiero et al. 2009; Carnio et al. 2014). By contrast, Parkin overexpression in Drosophila skeletal muscle was shown to increase mitochondrial content and attenuate the accumulation of protein aggregates, a marker of cellular ageing (Rana et al. 2013). Altered mitophagy with ageing therefore represents a potential mechanism that could lead to an accumulation of mitochondrial dysfunctions and, in turn, to sarcopenia.

To test the hypothesis, we overexpressed Parkin, a key regulator of mitophagy, for 4 months in skeletal muscles of young and late middle‐aged mice using i.m. injections of adeno‐associated viruses (AAV). We found that Parkin overexpression in adult skeletal muscle results in muscle hypertrophy and an increase in mitochondrial enzyme activities. We also showed that overexpression of Parkin in aged mice increases muscle mass, muscle strength, mitochondrial content and enzymatic activity, at the same time as lowering markers of oxidative stress, fibrosis and apoptosis. Taken altogether, these results indicate that Parkin is a potential therapeutic target to attenuate sarcopenia and improve skeletal muscle health and performance.

Methods

Ethical approval

The present study was carried out in strict accordance with standards established by the Canadian Council of Animal Care and the guidelines and policies of UQAM. All procedures were approved by the animal ethics committees of UQAM (#CIPA883). Experimental protocols were designed to minimize suffering and the number of animals used in the study. The authors declare that their work complies with the ethical principles under which The Journal of Physiology operates (Grundy, 2015).

Animal procedures and AAV injection

Experiments were conducted on 3‐month‐old (purchased from Jackson Laboratories, Bar Harbor, ME, USA) and 18‐month‐old (obtained through the Quebec Research Network on Aging, Montreal, QC, Canada) male C57BL/6J mice. Three to four mice were housed per cage under a 12:12 h light/dark photocycle at 24 ± 1°C and 50–60% relative humidity with access to standard chow diet and water available ad libitum. All AAV used in the present study were purchased from Vector Labs (Burlingame, CA, USA) and were of Serotype 1, which is a serotype with a proven tropism for skeletal muscle cells. After a 3‐day acclimatization period, an AAV containing a muscle specific promoter (muscle creatine kinase), a sequence coding for the reporter protein green fluorescemt protein (GFP) and a sequence coding for Parkin (details on the AAV construction are available in Fig. 1 A) were injected i.m. (25 μL per site; 2.5 × 10¹¹ gc) into the right tibialis anterior (TA) and gastrocnemius (GAS) muscles. In this construction, the sequences coding for Parkin and GFP were separated by a sequence coding for the autocleavable 2A peptide, allowing the separation of Parkin and GFP once translated. A control AAV containing only the GFP sequence under the control of the MCK promoter was injected into the contralateral leg. Injections were carried out under general anaesthesia using 2% isoflurane. Because the AAVS1 recombination site in wild‐type AAV was deleted in these recombined AAVs, both GFP and Parkin expression comprised episomal expression without integration into the host DNA. Two AAV injections, separated by 2 months, were therefore performed to ensure that Parkin expression remained elevated throughout the duration of the study. The first sets of i.m. injection were performed at 3 and 18 months, with the second set at 5 and 20 months. After 4 months of Parkin and/or GFP overexpression, mice either underwent a protocol to assess muscle contractile function or were anaesthetized with isoflurane and subsequently killed by cervical dislocation. For mice killed by cervical dislocation, the TA and GAS from both legs were removed. The TA and GAS were cut in half; one‐half was mounted for histology as described in (Gouspillou et al. 2014c; Leduc‐Gaudet et al. 2015). Small strips from the white GAS were prepared for analyses by transmission electron microscopy (TEM). The rest of the TA and GAS were frozen in liquid nitrogen and stored –80 °C until use for western blot and quantitative PCR experiments.

Figure 1. Successful overexpression of Parkin in skeletal muscles of young and old mice.

A, details of the AAV construct. B, information on the location and timing of AAV‐GFP and AAV‐Parkin injections. C, evolution of the body weight of young (n = 10) and old (n = 8) mice. D, representative images of GFP fluorescence obtained from cross‐sections of a 3‐month‐old control (no AAV injection), a 7mo‐AAV‐Parkin and a 22mo‐AAV‐Parkin TA. These images show that the AAV transduction efficiency was almost 100% (red scale bar = 100 μm; white scale bars = 1000 μm). E, western blot detection and quantification of Parkin content in the TA of young (n = 10) and old (n = 8) animals injected with either AAV‐Parkin or AAV‐GFP. Parkin content is shown to decline with ageing and injection of AAV‐Parkin results in successful Parkin overexpression. F, fold change in Parkin expression in young and old GAS induced by AAV‐Parkin injection. The graph shows successful Parkin overexpression. G, western blot detection and quantification of the ubiquitin content in the TA of old mice injected with either AAV‐Parkin or AAV‐GFP. Parkin overexpression is shown to result in a significant increase in protein ubiquitination. ^*Statistically significant. [Color figure can be viewed at wileyonlinelibrary.com]

In situ assessment of muscle contractile function

Mice were aanesthetized with an i.p. injection of a ketamine‐xylazine cocktail (ketamine: 130 mg kg^–1; xylazine: 20 mg kg^–1). Anaesthesia was maintained with supplemental doses pf 0.05 mL as needed. The surgical procedure and in situ contractile stimulation protocol were performed as described previously, with minor modifications (Mofarrahi et al. 2015). The distal tendon of the left and right TA muscle was isolated and attached in turn with surgical 4.0 silk to the lever arm of a 305C‐LR servomotor (Aurora Scientific Instruments, Aurora, ON, Canada). The Dynamic Muscle Control and Analysis Software Suite (Aurora Scientific Instruments) was used for collection and data analysis. The partially exposed muscle surface of the TA was kept moist for the contractile stimulation protocol and was directly stimulated with an electrode placed on the belly of the muscle. In situ measurement of the TA with direct stimulation was chosen over sciatic nerve stimulation, thereby removing potential negative effects such as a central contribution and, because blood delivery is intact, eliminating potential problems of isolated muscles (Allen et al. 2008). Optimal muscle length and voltage was progressively adjusted to produce maximal tension. The pulse duration was set to 0.2 ms for all tetanic contractions. Force–frequency relationships curves were determined at muscle optimal length at 10, 30, 50, 70, 100, 120 and 150 Hz, with 1 min intervals between stimulations to avoid fatigue.

Skeletal muscle sample sectioning for histology

Cross‐sections (thickness 10 μm) were cut in a cryostat at −18°C and mounted on lysine coated slides (Superfrost; Thermo Fisher, Waltham, MA, USA) to determine AAV transduction efficiency, muscle fibre size, succinate dehydrogenase (SDH) and cytochrome c oxidase (COX) activities, the proportion of terminal deoxynucleotidyl transferase dUTP nick‐end labelling (TUNEL) positive myonuclei, as well as the content of primordial germ cell (PGC)‐1α, type I collagen and 4‐hydroxynonenal (HNE) using immunohistological procedures described previously (Gouspillou et al. 2014c; Leduc‐Gaudet et al. 2015). These sections were also used to assess transduction efficiency of AAVs.

Assessment of AAV transduction efficiency

Because both AAVs used in the present study contained a sequence coding for GFP, transduction efficiency was assessed by simply examining the proportion of GFP positive fibres on muscle cross‐sections. Accordingly, slides were removed from our –80°C freezer and directly cover slipped using ProlongDiamond (Thermo Fisher) as mounting medium. Slides were then immediately imaged with an Axio 2 imager microscope (Carl Zeiss, Oberkochen, Germany) using a GFP fluorescence filter. A control slide obtained from a TA of a 3‐month‐old control (i.e. no AAV injection) mouse was used to confirm that all image acquisition parameters were not yielding the detection of autofluorescence. Once all of the acquisition parameters were set, with no detectable signal on the control slide, all slides were imaged using the exact same settings.

In situ determination of fibre size

Muscle cross‐sections were immunolabelled for laminin. Briefly, muscle cross‐sections were first allowed to reach room temperature and rehydrated with PBS (pH 7.2) and then blocked with goat serum (10% in PBS). Sections were then incubated with primary rabbit immunoglobulin (Ig)G polyclonal anti‐laminin antibody (L9393; Sigma, St Louis, MO, USA; dilution 1:750) for 1 h at room temperature. Sections were washed three times in PBS before being incubated for 1 h at room temperature with an Alexa Fluor 594 goat anti‐rabbit IgG antibody (A‐11037; Invitrogen, Carlsbad, CA, USA; dilution 1∶100). Sections were then washed three times in PBS and slides were cover slipped using Prolong Gold (P36930; Invitrogen) as mounting medium. Slides were imaged with a fluorescence microscope (Zeiss Axio Imager 2). The average number of fibres analysed is presented in Table 1.

Table 1.

Number of fibres analysed to quantify the impact of ageing and Parkin overexpression on muscle fibre size

	TA	GAS
7mo‐AAV‐GFP	504 ± 135	298 ± 26
7mo‐AAV‐Parkin	452 ± 88	296 ± 33
22mo‐AAV‐GFP	658 ± 157	303 ± 6
22mo‐AAV‐Parkin	614 ± 266	289 ± 29

Data are presented as the mean ± SD.

In situ determination of PGC‐1α, type I collagen and 4‐HNE content

For each sample, muscle cross‐sections were immunolabelled to assess PGC‐1α, type I collagen and 4‐HNE content in situ. Sections were first allowed to reach room temperature and were fixed in acetone at 4°C for 15 min (Gouspillou et al. 2014c), a step causing the loss of GFP fluorescence. Samples were then washed for 5 min in PBS (pH 7.4) at 4°C before being incubated for 15 min in a permeabilization solution (0.1% Triton X‐100 in PBS) at 4°C (Gouspillou et al. 2014c). Slides were then washed three times in PBS before being incubated at 4°C in a blocking solution (10% goat serum in PBS) for 30 min at room temperature (Gouspillou et al. 2014c). Slides were then incubated for 60 min at room temperature with a rabbit IgG polyclonal anti‐PGC‐1α antibody (Ab3242; Abcam, Cambridge, MA, USA; dilution 1∶50), a rabbit polyclonal antibody anti‐type I collagen (Ab34710; Abcam; dilution 1∶50) or a mouse monoclonal anti‐4HNE antibody (#HNE 13‐M; Alpha Diagnostic International, San Antonio, TX, USA; dilution 1 :100) (Gouspillou et al. 2014c). For PGC‐1α immunolabelling, sections were also incubated with a mouse IgG monoclonal anti‐dystrophin antibody (D8168; Sigma; dilution 1:100). Slides were then washed three times in PBS at 4°C before being incubated for 60 min at room temperature with an Alexa Fluor 488 IgG goat anti‐mouse antibody (A‐11001; Invitrogen,; dilution 1:500), an Alexa Fluor 594 goat anti‐mouse IgG, (A‐21235; Invitrogen; dilution 1∶100) and/or an Alexa Fluor 594 goat anti‐rabbit IgG antibody (A‐11037; Invitrogen; dilution 1∶100) (Gouspillou et al. 2014c). Slides were then washed three times in PBS. For PGC‐1α immunolabelling, sections were then incubated for 10 min in a PBS solution containing 4′,6‐diamidino‐2‐phenylindole (DAPI) (D1306; Invitrogen; [300 nm]) at 4 °C and subsequently washed three times in PBS. Sections were finally cover slipped using Prolong Gold (P36930; Invitrogen) as mounting medium (Gouspillou et al. 2014c). Images of cross‐sections were then taken with a fluorescence microscope (Zeiss Axio Imager 2) and analysed using ImageJ (NIH, Bethesda, MD, USA).

In situ determination of SDH and COX activities

Muscle cross‐sections were stained for SDH (complex II of the respiratory chain) and COX (complex IV of the respiratory chain) activity as described previously (Gouspillou et al. 2014c; Leduc‐Gaudet et al. 2015; St‐Jean‐Pelletier et al. 2017). Briefly, sections were first allowed to reach room temperature and were rehydrated with PBS (pH 7.2). Sections were then incubated in a solution containing nitroblue tetrazolium (1.5 mm; 11585029001; Sigma), sodium succinate (130 mm; S2378; Sigma), phenazine methosulphate (0.2 mm; P9625; Sigma) and sodium azide (0.1 mm) for SDH activity or in a solution containing cytochrome c (100 μm; C7752; Sigma) and 3,3′‐diaminobenzidine tetrahydrochloride (4 mm; D5637; Sigma) for 20 min at 37°C. Cross‐sections were then washed three times in PBS and cover‐slipped using an aqueous mounting medium (VectaMount AQ Medium, H‐5501; Vector Labs). All samples were processed at the same time and using the same incubation solution, ensuring that all samples underwent the exact same conditions.

Detection of apoptosis by TUNEL

The presence of 180–200 bp DNA fragments, obtained during genome fragmentation by endonucleases (Kyrylkova et al. 2012), is a robust marker of cellular apoptosis. TUNEL is a detection method used to locate these DNA fragments generated during apoptosis. In the present study, apoptotic myonuclei were detected using a commercial TUNEL kit (catalogue number 11684795910; Roche, Basel, Switzerland) in accordance with the manufacturer's instructions. Sections were also incubated for 1 h at room temperature with a mouse IgG monoclonal anti‐dystrophin antibody (D8168; Sigma; dilution 1:100), followed by a 1 h incubation at room temperature with an Alexa Fluor 647 goat anti‐mouse IgG antibody (A‐21235; Invitrogen; dilution 1:100). Sections were finally incubated for 10 min in a PBS solution containing DAPI (D1306; Invitrogen; [300 nm]) at 4 °C, washed three times in PBS and cover slipped using Prolong Gold (P36930; Invitrogen) as mounting medium. Images were acquired using a fluorescence microscope (Zeiss Axio Imager 2) and analysed using ImageJ (NIH).

Immunoblotting

Approximately 20 mg of each TA was homogenized in 10 volumes of an extraction buffer composed of 50 mm Tris base, 150 mm NaCl, 1% Triton X‐100, 0.5% sodium deoxycolate, 0.1% SDS and 10 μL mL^–1 of a protease inhibitor cocktail (P8340; Sigma). The homogenate was centrifuged at 15,000 g for 15 min at 4°C. Protein content in the supernatant was determined using the Bradford method with BSA as standard.

Aliquots of supernatant were mixed with Laemmli buffer and subsequently boiled at 95°C for 5 min. Equal amounts (20 μg) of proteins were loaded onto 8–12% gels, electrophoresed by SDS‐PAGE and transferred to polyvinylidene fluoride membranes (Transblot Turbo Midi‐size LF PVDF Membrane; Bio‐Rad, Hercules, CA, USA). Membranes were incubated for 1 h at room temperature in a blocking buffer composed of Tris‐buffered saline containing 5% non‐fat dried milk and 0.1% Tween 20 (TBS‐T) and probed either overnight at 4 °C or for 1 h at room temperature with the primary antibodies: mouse monoclonal antiPRK8 (Parkin; sc32282; Santa Cruz Biotechnology, Santa Cruz, CA, USA), mouse monoclonal anti‐S6 ribosomal protein (54D2; #2317; Cell Signaling Technology, Beverly, MA, USA; dilution 1:500), rabbit polyclonal anti‐phospho‐S6 ribosomal protein (Ser240/244) (#2215; Cell Signaling; dilution 1:500), mouse monoclonal anti‐4‐HNE (MAB3249; Bio‐Techne, Minneapolis, MN, USA; dilution 1:500) and a mouse monoclonal anti‐ubiquitin (Ub P4D1; sc‐8017; Santa Cruz Biotechnology) diluted in blocking buffer. Membranes were washed (6 × 5 min) in TBS‐T and subsequently incubated with appropriated HRP‐conjugated secondary antibodies (Ab6721; Abcam; dilution 1:5000; and Ab6728, Abcam; dilution 1:5000) diluted in blocking buffer for 1 h at room temperature. Protein signals were detected using enhanced chemiluminescence substrate (24080; Thermo Fisher), imaged with the ChemiDoc Touch Imaging System (Bio‐Rad).

TEM

The assessment of mitochondrial density and area was performed using TEM as described previously (Picard et al. 2013a; Picard et al. 2013b). Small stripes prepared from white GAS were incubated in 2% glutaraldehyde buffer solution in 0.1 m cacodylate (pH 7.4) and subsequently post‐fixed in 1% osmium tetroxide in 0.1 m cacodylate buffer, dehydrated in increasing concentrations of ethanol and propylene oxide, and embedded in epon. Sections (thickness 1 μm) were stained with toluidine blue to confirm the orientation of the muscle tissue prior the ultrathin sectioning. Ultrathin sections were cut in longitudinal or transverse orientation using an Ultracut S ultramicrotome (Leica Microsystems, Wetzlar, Germany) and mounted on nickel carbon‐formvar coated grids. Uranyl acetate and lead citrate stained sections were imaged using a Philips CM100 electron microscope (FEI, Hillsboro, OR, USA). Digital micrographs were captured using an XR80 CCD digital camera (Advanced Microscopy Techniques, Woburn, MA, USA) at 7900× magnification. Individual intermyofibrillar mitochondria from four old GFP expressing and four old Parkin overexpressing white GAS were manually traced in longitudinal orientations using ImageJ (NIH) to quantify the average area of individual mitochondria (in μm²) and mitochondrial density (i.e. the percentage of the cytoplasmic surface occupied by mitochondria).

Quantitative real‐time PCR

Total RNA was extracted from frozen GAS muscles using a PureLink™ RNA Mini Kit (Invitrogen Canada, Burlington, ON, Canada). Total RNA (2 μg) was reverse transcribed using a Superscript II® Reverse Transcriptase Kit and random primers (Invitrogen Canada). Reactions were performed at 42°C for 50 min and at 90°C for 5 min. Real‐time PCR detection of mRNA expression was performed using a Prism® 7000 Sequence Detection System (Applied Biosystems, Foster City, CA, USA). Real‐time PCR experiments were performed in triplicate. Relative mRNA quantifications of the Park2 gene were determined using the threshold cycle (ΔΔCT) method using the housekeeping gene cyclophilin. The primer sequences for Park2 and cyclophilin mRNA amplification were: mouse Park2: 5′‐TCTTCCAGTGTAACCACCGTC‐3′ and 5′‐GGCAGGGAGTAGCCAAGTT‐3′; mouse cyclophilin: 5′‐GCGTCTCCTTCGAGCTGTTT‐3′ and 5′‐CTGGCACATGAATCCTGGAAC‐3′.

Statistical analysis

The effects of ageing on parameters of interest were assessed by comparing 7mo‐AAV‐GFP and 22mo‐AAV‐GFP using bilateral unpaired t tests, with the exception of the impact of ageing on fibre size distribution, which was tested using two‐way ANOVA. When data for all four groups (i.e. 7mo‐AAV‐GFP, 7mo‐AAV‐Parkin, 22mo‐AAV‐GFP and 22mo‐AAV‐Parkin) were available, the effects of Parkin overexpression on parameters of interest were assessed using two‐way repeated measures ANOVA. Differences between 7mo‐AAV‐GFP and 7mo‐Saline were tested using bilateral unpaired t tests. When data were available for only one group of age (i.e. either 7mo‐AAV‐GFP + 7‐mo‐Parkin or 22mo‐AAV‐GFP + 22mo‐AAV‐Parkin), comparisons involving one variable were performed using bilateral paired t tests. When data were available for only one group of age but multiple comparisons were performed (e.g. force–frequency relationship) differences were tested using two‐way repeated measures ANOVA (to assess the impact of Parkin overexpression). For all ANOVA and t tests, P < 0.5 was considered statistically significant. Correction for multiple comparisons following two‐way ANOVA and two‐way repeated measures ANOVA were performed using the two‐stage step‐up method of Benjamini, Krieger and Yekutieli (Benjamini et al. 2006; q < 0.1 was considered statistically significant). All statistical analyses were performed using Prism, version 7 (GraphPad Software Inc., San Diego, CA, USA).

Results

Successful overexpression of Parkin using i.m. AAV injections

Parkin overexpression in the TA and GAS was achieved via i.m. injections of AAVs expressing mouse PARK2 and GFP under the control of a muscle specific promoter (muscle creatine kinase) (Fig. 1 A and B). Control AAVs expressing GFP were injected into the muscles of the contralateral leg (Fig. 1 B). Our experimental design provides an important advantage in that each animal is its own control. This is particularly relevant to the issue of muscle ageing because ageing‐related changes in nutritional intake, hormonal status, physical activity and systemic inflammation can all influence sarcopenia (Tieland et al. 2018). In young adult mice, Parkin overexpression was initiated at 3 months of age (after the developmental phase) to avoid any potential impact of its overexpression on muscle development. In the old group, Parkin overexpression was initiated at 18 months of age because muscle atrophy is not yet detectable at that time point (Hwee et al. 2014). Animals were studied after 4 months of Parkin and/or GFP overexpression (at the ages of 7 or 22 months, respectively) (Fig. 1 C). As shown in Fig. 1 D, AAV transduction efficiency was almost 100% (all fibres were GFP‐positive) for all samples examined. Injection of AAV‐Parkin significantly increased muscle Parkin mRNA expression and protein content in both young and old mice (Fig. 1 E and F). Importantly, we observed that endogenous Parkin protein content was significantly lower in old as compared to young TA, indicating that muscle Parkin content declines with ageing (Fig. 1 E). In line with the ubiquitin E3 ligase activity of Parkin, immunoblotting for ubiquitin protein conjugates confirmed that Parkin overexpression significantly increases overall muscle protein ubiquitination (Fig. 1 G).

Parkin overexpression triggers hypertrophy in young mice and attenuates ageing‐related loss of muscle mass and strength

To test the hypothesis that Parkin overexpression in old muscle will attenuate sarcopenia, we measured the impact of ageing in the absence and presence of Parkin overexpression on muscle mass and fibre size. In the absence of Parkin overexpression, muscles of old mice demonstrated clear signs of fibre atrophy, including decreased muscle mass and fibre size, compared to muscles of young mice (Fig. 2 A–H). Parkin overexpression in both young and old mice resulted in significant increases in muscle weight (Fig. 2 A and B) and fibre size (Fig. 2 D–J). We should emphasize that prolonged overexpression of GFP (4 months) had no effect on muscle weight (Fig. 2 C). To assess whether the increase in muscle weight and fibre size in response to Parkin overexpression is mediated by an increase in the activity of the Akt‐mTORC1 pathways, we quantified the extent of S6 protein phosphorylation as an index of mTORC1 activation. Figure 2 K shows that Parkin overexpression was associated with significant increases in S6 protein phosphorylation in young but not old skeletal muscles.

Figure 2. AAV‐mediated overexpression of Parkin increases muscle weight and fibre size in young mice and attenuates sarcopenia.

A and B, quantification of the impact of ageing and Parkin overexpression on TA (A) and GAS (B) weights. Ageing is shown to be associated with significant muscle atrophy (7mo‐AAV‐GFP (n = 10) vs. 22mo‐AAV‐GFP). Parkin overexpression results in higher muscle mass in both young and old muscles vs. their GFP expressing counterparts. Also shown in (A) are representative images demonstrating that old muscles injected with AAV‐Parkin are bigger than muscles injected with AAV‐GFP (scale bars = 5 mm). C, quantification of the effect of GFP expression on the weight of the TA in young mice. The graph shows that GFP expression did not alter TA weight. D, representative laminin immunolabelling performed on muscle cross‐sections of young and old GAS muscles expressing either GFP or Parkin. E and F, quantification of the impact of ageing and Parkin overexpression on average gastrocnemius fibre size (E) and minimal Feret diameter (F). The graphs show that ageing is associated with myofibre atrophy, whereas Parkin overexpression results in a higher fibre size in both young (n = 10 AAV‐GFP; n = 9 AAV‐Parkin) and old (n = 8) mice (scale bar = 50 μm). G, quantification of the impact of Parkin on TA fibre size in young and old mice. Parkin overexpressing TA also demonstrated larger fibres vs. their GFP expressing counterparts. H, quantification of the effect of ageing on GAS fibre size distribution between young (n = 9) and old (n = 8) mice. The graph shows a distributional shift to the left in aged muscles, a clear sign of atrophy. I, quantification of the effect of Parkin overexpression on GAS fibre size distribution in young (n = 9) mice. The graph shows a distributional shift to the right in Parkin overexpressing muscle, indicating hypertrophy. J, quantification of the effect of Parkin overexpression on GAS fibre size distribution in old (n = 8) mice. The graph shows a distributional shift to the right in old Parkin overexpressing muscle, clearly showing that old Parkin overexpressing muscles have larger fibres vs. their GFP expressing counterparts. K, western blot detection and quantification of pS6 and S6 content in the TA of young and old GFP expressing or Parkin overexpressing muscles. Also shown in (K) is a quantification of the pS6 to S6 ratio, a marker of the Akt‐mTOR pathway activity. The graph shows that Parkin overexpression increased the pS6 to S6 ratio only in young muscles. L, quantification of the strength/stimulation frequency relationship in old AAV‐GFP (n = 5) and AAV‐GFP‐Parkin (n = 5) injected muscles showing that Parkin expression increased muscle strength for all of the stimulation frequencies we tested. ^*Statistically significant. [Color figure can be viewed at wileyonlinelibrary.com]

To define whether the greater muscle fibre size and muscle mass seen in Parkin overexpressing muscle was translating into functional gain, we next assessed in situ TA strength in old mice. Accordingly, a new batch of old mice was used. Importantly, this second batch of animals allowed us to replicate our original findings on the impact of Parkin overexpression on the TA weight (data not shown). Overexpression of Parkin was associated with a significant increase in force generation at any given stimulation frequency compared to GFP overexpression (Fig. 2 L).

Parkin overexpression increases mitochondrial enzyme activity and mitochondrial content

Because Parkin is well known for its role in mitochondrial quality control processes, we investigated whether the positive impact of Parkin overexpression on muscle mass and function was associated with changes in mitochondrial enzymatic activities and mitochondrial content. Parkin overexpression significantly increased the activities of SDH and COX (complexes II and IV of the electron transfer system) (Fig. 3 A–D). Measurement of muscle mitochondrial density with TEM confirmed that Parkin overexpression in old muscles significantly increased mitochondrial density by 72% compared to GFP overexpression (Fig. 3 E and F). Interestingly, this increase in mitochondrial density was associated with a significant decrease in the average area of individual mitochondria (Fig. 3 G). To determine whether the changes in mitochondrial enzyme functions and mitochondrial density elicited by Parkin overexpression were the result of increased mitochondrial biogenesis, we quantified PGC‐1α expression using immunostaining of muscle cross‐sections (Fig. 3 H). We found that normal ageing is associated with a trend for a decrease in PGC‐1α expression, which is a finding consistent with previous studies (Baker et al. 2006; Chabi et al. 2008; Joseph et al. 2012). Parkin overexpression elicited a significant increase in PGC‐1α expression in old but not young skeletal muscles (Fig. 3 H and I).

Figure 3. Parkin overexpression increases mitochondrial enzyme activity and mitochondrial content.

A and B, representative SDH stains (A) performed on muscle cross‐sections of young and old muscles expressing either GFP or Parkin. Whole cross‐sections stained for SDH are shown in the first row of images, whereas representative images obtained at a higher magnification are shown in the second row (black scale bars = 1000 μm; white scale bars = 100 μm). B, corresponding quantification of SDH stain intensity, which is proportional to SDH activity. The quantification reveals that Parkin overexpression increases SDH activity in both young (n = 10) and old (n = 8) muscles. C and D, representative COX stains (C) performed on muscle cross‐sections of young muscles expressing either GFP or Parkin and corresponding quantification (D) of COX stain intensity, which is proportional to COX activity (scale bar = 100 μm). The quantification reveals that Parkin overexpression increased COX activity in young muscles (n = 7 per group). E–G, representative TEM images (E) of old GFP expressing and Parkin overexpressing muscles (scale bar = 2 μm). From such TEM images, the mitochondrial volume density (F) and the average size of individual mitochondria (G) were quantified. F and G, Parkin overexpression in old muscles increased mitochondrial density (F) (n = 4 per group) at the same time as decreasing the size of individual mitochondria (G) (22mo‐AAV‐GFP: n = 586 mitochondria; 22mo‐AAV‐Parkin: n = 666 mitochondria; unpaired bilateral t test). H and I, representative in situ immunolabelling of PGC‐1a (red), dystrophin (green) and nuclei (blue) (H) performed in young (n = 9) and old (n = 8) muscles expressing GFP or overexpressing Parkin. Whole cross‐sections are shown in the first row of images, whereas representative images obtained at a higher magnification are shown in the second row (red scale bars = 1000 μm; white scale bars = 50 μm). I, corresponding quantification of PGC‐1α immunolabelling intensity showing a trend for decreased PGC‐1α content with ageing and increased PGC‐1α content induced by Parkin overexpression in old mice. ^*Statistically significant. NS, not significant. [Color figure can be viewed at wileyonlinelibrary.com]

Parkin overexpression attenuates ageing‐related oxidative stress

Oxidative stress has been consistently detected in aged skeletal muscle and is considered to contribute to ageing‐related decreases in muscle mass and function (Yan & Sohal, 1998; Capel et al. 2005; Yarian et al. 2005; Chabi et al. 2008; Sohal & Orr, 2012; Sinha‐Hikim et al. 2013; Umanskaya et al. 2014). To assess whether Parkin overexpression may affect the development of oxidative stress in old muscles, we assessed muscle content of 4‐HNE (i.e. a marker of lipid peroxidation) using immunostaining and immunoblotting. Figure 4 shows that normal ageing is associated with a significant increase in muscle 4‐HNE content. Parkin overexpression elicited a significant decrease in 4‐HNE content in old but not young skeletal muscles (Fig. 4). These results indicate that Parkin overexpression attenuates ageing‐related increases in oxidative stress.

Figure 4. Parkin overexpression attenuates ageing‐related increase in oxidative stress.

A, representative in situ immunolabelling of 4‐HNE (a marker of oxidative stress) performed in young (n = 8) and old (n = 8) muscles expressing GFP or overexpressing Parkin. Immunolabelling of whole cross‐sections is shown in the first row of images, whereas representative images obtained at a higher magnification are shown in the second row (red scale bars = 1000 μm; white scale bars = 50 μm). Also shown in (A) is the corresponding quantification of 4‐HNE immunolabelling intensity, with a strong trendency for an increase in 4‐HNE content with ageing and a decrease in 4‐HNE content induced by Parkin overexpression in old mice. B, western blot detection and quantification of 4‐HNE content in the TA of young (n = 5) and old (n = 5) animals injected with either AAV‐Parkin‐GFP or AAV‐GFP. The data and conclusions obtained by 4‐HNE immunolabelling are strengthened. ^*Statistically significant. [Color figure can be viewed at wileyonlinelibrary.com]

Parkin overexpression attenuates ageing‐related increase in myonuclear apoptosis

Apoptosis increases significantly in old skeletal muscles and contributes to the development of sarcopenia (Dirks & Leeuwenburgh, 2004; Leeuwenburgh et al. 2005; Chabi et al. 2008; Gouspillou et al. 2014b). We assessed whether Parkin overexpression alters the extent of apoptosis in ageing skeletal muscles. Apoptotic myonuclei were detected in muscle cross‐sections using the TUNEL approach combined with DAPI staining and dystrophin immunostaining. As can be seen in Fig. 5, normal ageing was associated with a significant increase in the number of TUNEL‐positive myonuclei (Fig. 5). Parkin overexpression significantly decreased the number of TUNEL‐positive myonuclei in old skeletal muscles (Fig. 5).

Figure 5. Parkin overexpression attenuates ageing‐related increase in myonuclear apoptosis.

A, representative TUNEL stain (green; marking apoptotic nuclei) coupled with DAPI staining (blue; marking nuclei) and dystrophin immunolabelling (red) performed in young (n = 9) and old (n = 6) muscles expressing GFP or overexpressing Parkin. B, quantification of the proportion of TUNEL positive myonuclei showing an increase in apoptosis with ageing and an anti‐apoptotic effect of Parkin overexpression (Scale bar = 25 μm). ^*Statistically significant. [Color figure can be viewed at wileyonlinelibrary.com]

Parkin overexpression attenuates deposition of type I collagen in old skeletal muscles

Fibrosis develops in old skeletal muscles and contributes to ageing‐related declines in skeletal muscle function (Parker et al. 2017). The effect of Parkin overexpression on fibrosis inside skeletal muscles was assessed by measuring the content of type I collagen using immunostaining. Normal ageing was associated with a significant increase in muscle type I collagen content (Fig. 6). Parkin overexpression significantly decreased type I collagen content in old but not in young skeletal muscles (Fig. 6). These results indicate that Parkin overexpression attenuates the ageing‐related increase in fibrosis in skeletal muscles.

Figure 6. Parkin overexpression attenuates type I collagen deposition in old mice.

A, representative type I collagen immunolabelling (red) performed on cross‐sections of young (n = 8) and old (n = 7) muscles expressing GFP or overexpressing Parkin (scale bar = 500 μm). B, type I collagen immunolabelling intensity showing an ageing‐related increase in type I collagen content with ageing and a decrease in type I collagen content induced by Parkin overexpression in aged muscles. ^*Statistically significant. [Color figure can be viewed at wileyonlinelibrary.com]

Discussion

Although the aetiology of sarcopenia is extremely complex and only partly understood, the accumulation of mitochondrial dysfunction has been shown to contribute to sarcopenia (Trounce et al. 1989; Selman et al. 2003; Dirks & Leeuwenburgh, 2004; Leeuwenburgh et al. 2005; Short et al. 2005; Chabi et al. 2008; Marzetti et al. 2008; Gouspillou et al. 2010; Picard et al. 2010; Picard et al. 2011; Lanza et al. 2012; Umanskaya et al. 2014; Gouspillou et al. 2014a; Gouspillou et al. 2014b). Recent studies have proposed that mitophagic processes controlled by the Parkin–Pink1 pathway may be impaired in aged muscles and are probably responsible for the ageing‐related accumulation of dysfunctional mitochondria (O'Leary et al. 2013; Rana et al. 2013; Gouspillou et al. 2014b). On the basis of this proposal, we tested the effect of Parkin overexpression in skeletal muscles of late middle‐aged mice on ageing‐related loss of muscle mass and strength. The results obtained demonstrate that 4 months of Parkin overexpression in muscles of late‐middle aged mice significantly increased muscle mass and strength. These findings provide the proof of principle that increasing Parkin expression just before the appearance of sarcopenia, either through pharmaceutical or non‐pharmaceutical means, could represent an effective strategy for attenuating sarcopenia. In this respect, a recent study has shown that supplementing old mice with urolithin A for 27–31 weeks increased their grip strength and endurance (Ryu et al. 2016). Although the exact mechanisms by which Urolithin A exerts its beneficial effects in aged mice was not identified, Park2 (i.e. the gene coding for Parkin) was one of the most upregulated genes upon urolithin A treatment (Ryu et al. 2016).

An intriguing observation of the present study is that Parkin overexpression in muscles of young mice triggered hypertrophy. The mechanisms by which Parkin mediates its hypertrophic effects in adult skeletal muscle remain to be fully clarified; our results suggest that Parkin overexpression elicited an increase in the activity of the Akt‐mTORC1 pathway, as suggested by the significant increase in phosphorylation of S6 that was observed in mice overexpressing Parkin. The Akt‐mTOR pathway is central to the regulation of muscle mass because it promotes protein synthesis and inhibits protein degradation (Schiaffino & Mammucari, 2011). The increase in S6 phosphorylation in response to Parkin overexpression in young muscles is in line with recent literature indicating that Parkin can ubiquitinate mTOR, a post‐translational modification known to increase mTOR activity (Park et al. 2014). The lack of impact of Parkin overexpression on S6 phosphorylation suggests that the mechanisms regulating hypertrophy in adult skeletal muscle differ from those involved in the attenuation of sarcopenia.

Amongst the hallmarks of muscle ageing, the presence of oxidative stress is often reported in aged skeletal muscle and has been suggested as a contributing factor to ageing‐related declines in muscle mass and function (Yan & Sohal, 1998; Capel et al. 2005; Yarian et al. 2005; Chabi et al. 2008; Sohal & Orr, 2012; Sinha‐Hikim et al. 2013; Umanskaya et al. 2014; ). Our results are in line with these observations because we found that aged skeletal muscle demonstrated an increase in 4‐HNE content, a marker of oxidative stress. Importantly, 4 months of Parkin overexpression resulted in a significant decrease in 4‐HNE content, strongly suggesting that Parkin overexpression attenuated the ageing‐related increase in oxidative stress. Although the mechanisms conferring Parkin its protective effects against oxidative stress were not investigated in the present study, it has been shown in p53‐depleted pleural cells that Parkin overexpression can increase the glutathione/glutathione disulphide ratio, suggesting that Parkin plays role in the regulation of antioxidant defense systems (Zhang et al. 2011).

Another hallmark of muscle ageing is the progressive appearance of fibrosis (Parker et al. 2017) and the higher type I collagen content observed in aged muscles in the present study indicates that the old mice developed moderate fibrosis. Such an alteration in muscle connective tissue is assumed to contribute to ageing‐related declines in skeletal muscle function (Parker et al. 2017). Interestingly, our data show that Parkin overexpression attenuated the ageing‐related increase in fibrosis, suggesting that Parkin might play a role in the maintenance of the extracellular matrix. As discussed below, and although the mechanisms underlying this anti‐fibrotic effect of Parkin remain to be clarified, we speculate that this anti‐fibrotic effect might be linked to the impact of Parkin on apoptosis.

Apoptosis is well known to increase with ageing and contribute to sarcopenia (Dirks & Leeuwenburgh, 2004; Leeuwenburgh et al. 2005; Chabi et al. 2008; Gouspillou et al. 2014b). The present study further reinforces the available literature by showing that aged muscles display an increase in the proportion of TUNEL positive myonuclei, a reliable marker of apoptosis. Importantly, we show that Parkin overexpression significantly attenuated the ageing‐related increase in apoptosis. When triggered in skeletal muscle, apoptosis leads to the loss of one or several myonuclei, thereby decreasing the capacity of a muscle fibre for protein synthesis and, ultimately, leading to fibre atrophy (Dupont‐Versteegden, 2006). Prolonged activation of apoptosis can even result in the loss of muscle fibres (Dupont‐Versteegden, 2006). The attenuation of apoptosis triggered by Parkin overexpression therefore probably contributed to the higher muscle mass and strength that we measured in old Parkin overexpressing muscles. We also speculate that this anti‐apoptotic effect of Parkin might have attenuated fibrosis by preventing the loss of myofibres and their replacement by connective tissue. It is also worth noting that our findings add to the available literature showing that Parkin can exert anti‐apoptotic effects (Charan et al. 2014).

Another important finding of the present study is the robust increase in mitochondrial content and enzymatic activities that is a result of Parkin overexpression. These findings are in line with a previous study by Rana et al. (2013) showing that Parkin overexpression in muscles of Drosophila melanogaster resulted in a significant increase in citrate synthase, complex I and complex II enzymatic activities. These findings clearly show that the contribution of Parkin to mitochondrial quality control processes is not limited to mitophagy and extends to mitochondrial biogenesis. Although the underlying mechanisms were not investigated in the present study, the increase in mitochondrial content and enzymatic activities that we observed upon Parkin overexpression was probably mediated by the interaction between Parkin and PARIS. Indeed, it has been recently shown that Parkin can target PARIS, a transcriptional repressor of PGC‐1α, for degradation (Shin et al. 2011) and thereby increase mitochondrial biogenesis. The increase in mitochondrial content and enzymatic activities caused by Parkin overexpression indicates that Parkin can positively impact skeletal muscle energy metabolism.

Declining PGC‐1a expression with ageing has been proposed as a contributing factor to explain ageing‐related loss of muscle mass and function (Dillon et al. 2012). In the present study, we show that Parkin overexpression in old muscle attenuated the ageing‐related decline in PGC‐1α content. Considering the fact that PGC‐1α overexpression has been shown to partly attenuate ageing‐related loss of fibre size (Garcia et al. 2018), it is possible that some of the anti‐ageing effects of Parkin overexpression were mediated by its impact on PGC‐1α content in old muscles.

Recent studies have provided evidence that Parkin is involved in the regulation of mitochondrial dynamics and morphology. Indeed, Parkin can ubiquitinate the pro‐fusion proteins Mfn2 and Mfn1, thereby targeting these proteins for proteosomal degradation (Gegg et al. 2010). The degradation of these pro‐fusion proteins is considered to tip the fusion/fission balance towards mitochondrial fission (Gegg et al. 2010). In line with this view, Parkin overexpression in fly skeletal muscle and in rat hippocampal neurons has been shown to stimulate mitochondrial fragmentation (Yu et al. 2011; Rana et al. 2013). The results of the present further strengthen this literature because we show that Parkin overexpression in old muscle decreased the average area of individual intermyofibrillar mitochondria, therefore suggesting mitochondrial fragmentation. Interestingly, we have recently shown that skeletal muscle ageing is associated with an enlargement of subsarcolemmal mitochondria and an increase in the morphological complexity (increased branching and length) of intermyofibrillar mitochondria (Leduc‐Gaudet et al. 2015). Our data suggest that Parkin overexpression might have attenuated the effects of ageing on mitochondrial morphology and dynamics. Because mitophagy of large mitochondria requires their fission (Twig & Shirihai, 2011), it is tempting to speculate that the smaller mitochondria seen in old Parkin overexpressing muscles might facilitate their recycling when they become damaged/dysfunctional. Combined with the documented positive effect of Parkin overexpression on mitochondrial biogenesis, mitochondrial fragmentation caused by Parkin overexpression might help in the maintenance of a healthy pool of mitochondria throughout ageing.

In conclusion, the present study shows that Parkin overexpression attenuates ageing‐related loss of muscle mass and strength and causes hypertrophy in adult skeletal muscles. In addition, our results show that Parkin overexpression leads to increases in mitochondrial content and enzymatic activities. Finally, our findings indicate that Parkin overexpression attenuates ageing‐related increases in markers of oxidative stress, fibrosis and apoptosis. These results place Parkin as a potential therapeutic target for attenuating sarcopenia and improving skeletal muscle health and performance.

Additional information

Competing interests

The authors declare that they have no competing interests.

Author contributions

GG and SH designed and conceived the study. JPLG, OR and GG collected and analysed the data. All authors interpreted the data and contributed to the writing of the manuscript. All authors approved the final version of the manuscript submitted for publication.

Funding

This work was funded by grants from the Natural Sciences and Engineering Council of Canada (NSERC) awarded to Gilles Gouspillou (#RGPIN‐2014‐04668) and from the Canadian Institute of Health Research (CIHR; MOP‐93760) awarded to Sabah N. A. Hussain. Gilles Gouspillou is also supported by a Chercheur Boursier Junior 1 salary award from the Fonds de Recherche du Québec en Santé (FRQS‐35184). Jean‐Philippe Leduc‐Gaudet is supported by a CIHR Vanier Fellowship. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Biographies

Jean‐Philippe Leduc‐Gaudet obtained a BSc in Kinesiology and completed a MSc in Exercise Physiology at UQAM (Montréal, Canada). During his MSc, his research focused on the impact of nutrition on mitochondrial energetics in skeletal muscle. He is currently a PhD candidate under the supervision of Dr Gilles Gouspillou and Dr Sabah Hussain at the Meakins‐Christie Laboratories affiliated to McGill University (Montréal, Canada). His research now focuses on the investigation of the role played by mitochondrial quality control in muscle health, ageing and diseases.

Olivier Reynaud obtained a BSc in Molecular Biology and completed a MSc in Exercise Physiology at UQAM under the supervision of Dr Gilles Gouspillou. His research interests are in the field of molecular biology, with a particular focus on skeletal muscle physiology and mitochondrial biology in ageing and disease. He now works as a research assistant in Dr Gilles Goupillou's laboratory.