Original Article Low aerobic capacity in McArdle disease: A role for mitochondrial network impairment? M. Villarreal-Salazar 1, 2, 11, A. Santalla 3, 11, A. Real-Martínez 1, 2, G. Nogales-Gadea 4, P.L. Valenzuela 5, C. Fiuza-Luces 5, A.L. Andreu 6, J.C. Rodríguez-Aguilera 3, 10, M.A. Martín 2, 7, J. Arenas 7, J. Vissing 8, A. Lucia 9, T.O. Krag 8, **, T. Pinós 1, 2, * ABSTRACT Background: McArdle disease is caused by myophosphorylase deficiency and results in complete inability for muscle glycogen breakdown. A hallmark of this condition is muscle oxidation impairment (e.g., low peak oxygen uptake (VO2peak)), a phenomenon traditionally attributed to reduced glycolytic flux and Krebs cycle anaplerosis. Here we hypothesized an additional role for muscle mitochondrial network alterations associated with massive intracellular glycogen accumulation. Methods: We analyzed in depth mitochondrial characteristics-content, biogenesis, ultrastructure-and network integrity in skeletal-muscle from McArdle/control mice and two patients. We also determined VO2peak in patients (both sexes, N ¼ 145) and healthy controls (N ¼ 133). Results: Besides corroborating very poor VO2peak values in patients and impairment in muscle glycolytic flux, we found that, in McArdle muscle: (a) damaged fibers are likely those with a higher mitochondrial and glycogen content, which show major disruption of the three main cytoskeleton components-actin microfilaments, microtubules and intermediate filaments-thereby contributing to mitochondrial network disruption in skeletal muscle fibers; (b) there was an altered subcellular localization of mitochondrial fission/fusion proteins and of the sarcoplasmic reticulum protein calsequestrin-with subsequent alteration in mitochondrial dynamics/function; impairment in mitochondrial content/biogenesis; and (c) several OXPHOS-related complex proteins/activities were also affected. Conclusions: In McArdle disease, severe muscle oxidative capacity impairment could also be explained by a disruption of the mitochondrial network, at least in those fibers with a higher capacity for glycogen accumulation. Our findings might pave the way for future research addressing the potential involvement of mitochondrial network alterations in the pathophysiology of other glycogenoses. Ó 2022 The Author(s). Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Keywords McArdle disease; Skeletal muscle; Glycogen; Aerobic capacity; Cytoskeleton and mitochondrial network 1. BACKGROUND glycosidic bond, patients’ muscle fibers are unable to obtain energy from intracellular glycogen stores, with subsequent impairment in McArdle disease, also known as glycogen storage disease (GSD) type V glycolytic flux. Of note, however, the latter is not fully impeded since (OMIMÒ database number 232600, ORPHA:368) or ‘myophosphor- fibers can metabolize blood-borne glucose because the metabolic ylase deficiency’ is an autosomal recessive disorder caused by path- block occurs upstream of blood glucose uptake. ogenic mutations in the gene (PYGM) encoding the skeletal muscle A hallmark of McArdle disease is impairment of muscle aerobic isoform of glycogen phosphorylase (EC 2.4.1.1), commonly known as metabolism, as reflected by very low levels of peak oxygen uptake ‘myophosphorylase’. This condition is one of the most prevalent GSDs, (VO2peak) in most patients, usually below 50% of age and gender- affecting 1:50,000 to 1:200,000 people [1,2]. Because myophos- predicted values [3,4]. This is an important consideration given the phorylase catalyzes the rate-limiting step in muscle glycogen meta- strong prognostic value of VO2peak as an indicator of cardiovascular bolism by releasing glucose-1-phosphate from the terminal alpha-1,4- mortality and cardiometabolic health in general [5], with the American 1 Mitochondrial and Neuromuscular Disorders Unit, Vall d’Hebron Institut de Recerca, Universitat Autònoma de Barcelona, Barcelona, Spain 2Centro de Investigación Bio- médica en Red de Enfermedades Raras (CIBERER), Madrid, Spain 3Universidad Pablo de Olavide, Sevilla, Spain 4Grup de Recerca en Malalties Neuromusculars i Neu- ropediàtriques, Department of Neurosciences, Institut d’Investigacio en Ciencies de la Salut Germans Trias i Pujol i Campus Can Ruti, Universitat Autònoma de Barcelona, Badalona, Spain 5Physical Activity and Health Research Group (‘PaHerg’), Research Institute of the Hospital 12 de Octubre (‘imas12’), Madrid, Spain 6EATRIS, European Infrastructure for Translational Medicine, Amsterdam, Netherlands 7Mitochondrial and Neuromuscular Diseases Laboratory, 12 de Octubre Hospital Research Institute (iþ12), Madrid, Spain 8Copenhagen Neuromuscular Center, Department of Neurology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark 9Faculty of Sport Sciences, European University, Madrid, Spain 10Centro Andaluz de Biología del Desarrollo, Universidad Pablo de Olavide, Sevilla, Spain 11 Both authors contributed equally to this work. *Corresponding author. Departament de Patologia Mitocondrial i Neuromuscular, Hospital Universitari Vall d’Hebron. Institut de Recerca (VHIR), Barcelona, Spain. Tel.: þ34934894057. E-mail: tomas.pinos@vhir.org (T. Pinós). **Corresponding author. Tel.: þ4535457614. E-mail: thomas.krag@regionh.dk (T.O. Krag). Received October 7, 2022  Revision received November 14, 2022  Accepted November 24, 2022  Available online 28 November 2022 MOLECULAR METABOLISM 66 (2022) 101648 Ó 2022 The Author(s). Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 1 www.molecularmetabolism.com Original Article Heart Association recently advocating for the routine assessment of 2. MATERIAL AND METHODS this measure as a clinical vital sign [6]. In this effect, recent data from the European registry of patients with McArdle disease (EUROMAC; 2.1. Ethical approval http://www.euromacregistry.eu) indicate an overall unhealthy car- All experimental procedures in mice were approved by the relevant diometabolic profile for these individuals, with two-thirds showing high institutional review board (vall d’Hebron hospital, barcelona, Spain; body mass index values and w12% having cardiovascular disease protocol number 58/17 CEEA; 35/04/08) and were conducted in (8% with coronary artery disease) despite the relatively young age of accordance with the European Convention for the Protection of the cohort (median age 46 years) [7]. The finding of a low VO2peak in Vertebrate Animals used for Experimental and Other Scientific Pur- patients has been replicated in the genetically manipulated mouse poses (ETS1 2 3) and Spanish laws (32/2007 and R.D. 1201/2005). (knock-in) model of McArdle disease that we previously gen- The study in humans adhered the ethics guidelines of the Declaration eratedharboring the commonest Pygm genotype (p.R50*/p.R50*) of Helsinki, and was approved by the Ethics Committee of Hospital 12 that causes the disorder in Caucasiansand which muscle phenotype de octubre, madrid, Spain (reference 16/081). All participants were mimics that of affected patients, thereby representing an accurate informed about the study procedures and provided written informed study model [7,8]. Indeed, the aerobic fitness of McArdle mice is also consent much lower compared to wild-type (WT) controls [9]. At the mechanistic level, the low VO2peak of patients with McArdle 2.2. Patients’ muscle biopsies disease has been traditionally associated with a reduced flux of sub- Biopsies were taken with a Bergstrøm needle from two patients, pa- strates through the glycolytic pathway, thereby impairing oxidative tient 1 a male age 61 years with proximal weakness and patient 2 one metabolismas reflected for instance by slow oxygen uptake kinetics female age 34 years, both homozygous for the pathogenic during the transition from rest to exercise [10], or slow rate of phos- p.50* variant in the PYGM gene. phocreatine recovery following exertion [11]. Indeed, the production of pyruvate, a molecule that plays an anaplerotic role in the Krebs cycle, is 2.3. Studies in mice reduced in patients [12], with intravenous glucose administration Previously developed p. R50*/p.R50  mice, back-crossed for 10 increasing VO2peak (by about 20%) and muscle oxidation [13]. On the generations to C57/6 J background, were used in this study [19,22]. other hand, not only impaired glycolytic flux, but also functional Cohorts of 10 wild-type (WT) and McArdle mice (male and female, impairment of oxidative metabolism in muscle fibersas reflected by a aged between 8 and 20 weeks) were studied. All mice were sacrificed slow recovery of muscle ADP after brief ischemic exercise exer- by cervical dislocation immediately before dissection of the hind-limb cisehas been described in McArdle patients [14]. In this context, other muscles (i.e. gastrocnemius, quadriceps, and tibialis anterior [TA]). potential variables related to the structure/function of the cell organelles that are responsible for aerobic energy provision, mitochondria, could 2.4. Histochemistry explainat least partlythe low VO2peak that characterizes this disease, Dissected muscles were flash frozen in isopentane cooled in liquid with previous data indeed showing that the muscle levels of citrate nitrogen and stored at 80  C until analysis. Twelve mm cryosections synthase (CS), a key mitochondrial enzyme that reflects the function of were stained with hematoxylin and eosin (H&E) for general histo- this organelle, is lower in McArdle mice than in WT controls [15]. pathological evaluation as previously described [19]. Glycogen content Mitochondria form a branched tubular network, with a balance be- was analyzed with periodic acideSchiff (PAS) staining by sequentially tween the fusion and fission of these organelles needed not only to incubating the sections with periodic acid (Fisher Scientific; Hampton, ensure integrity of this network but also maintenance of normal NH) (0.5%) for 5 min, water wash, Schiff ‘s solution (Merck-Millipore; levels of mitochondrial membrane potential and respiration [16]. On Burlington, MA) for 15 min, water for 1 min, alcohol-xylol dehydration, the other hand, in mammals, VO2peak is associated with skeletal and DPX mounting (SigmaeAldrich; St. Louis, MO). Myophosphorylase muscle mitochondrial respiration [17] and is linearly related with total activity staining was performed by incubating muscle sections for mitochondrial volume and surface area [18]. In this context, it might 45 min with a solution containing 1% glucose-1-phosphate, 0.2% be hypothesized that the massive intracellular glycogen deposits AMP, and 0.02% glycogen in 0.1 M sodium acetate buffer, pH 5.6. associated with McArdle diseasei.e., 27-fold higher in McArdle Sections were washed with water, Lugol’s iodine was applied for 3 min than in WT micedue to the inherited block in glycogenolysis could and samples were mounted with Aquatex. Staining for succinyl de- alter the normal cytoskeleton structure of skeletal muscle fibers and hydrogenase (SDH) was done as by incubating cryosections in 0.1 M thus their mitochondrial tubular network [19]. In turn, because IIA PBS and 0.1 M succinate with 1 mg/ml nitro blue tetrazolium (all and IIX fibers are more structurally affected (i.e., ‘degenerated’) than Merck-Millipore) for 2 h at 37  C followed by water wash, alcohol type I and IIB fibers in McArdle mice [20,21], the mitochondrial dehydration and mounting with Pertex (Sakura, Alphen aan den Rijn, tubular network could also be differently affected based on the The Netherlands). Cytochrome c oxidase (COX) activity staining was relevant muscle phenotype (i.e., with a predominance of either IIA or performed in cryosections as previously described [23]. Staining for IIX fibers, respectively). myosin heavy chain type II (fast) was done by incubating cryosections The aim of this study was to determine whether mitochondrial network with antibody (Leica Biosystems, Nussloch, Germany) for an hour at integrity and function is altered in skeletal muscle fibers in the context room temperature, followed by washing and incubating in secondary of McArdle disease. To this purpose, we used the mouse model of this antibody and development with 3,30 -diaminobenzidine (DAB) accord- condition as well as skeletal muscle biopsies from patients. If mito- ing to manufacturer’s instructions (Agilent, Santa Clara, CA). chondrial network integrity and function were actually altered, this would help to explain, at least partly, the major impairment in muscle 2.5. Immunofluorescence oxidative metabolism that characterizes this disease. In addition, we Formalin fixed paraffin-embedded quadriceps and TA muscles from aimed to provide corroborating, strong evidence for an impaired WT and McArdle mice and OCT embedded frozen quadriceps biopsies VO2peakwell below normative valuesin the longest series of pa- from patients were used. Muscle samples were blocked in buffer (3% tients studied to date. fetal calf serum in PBS) prior to staining. For the studies of 2 MOLECULAR METABOLISM 66 (2022) 101648 Ó 2022 The Author(s). Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). www.molecularmetabolism.com cytoskeleton, mitochondrial network and dynamics, and fiber type, NCBI Reference Sequence: NC_005089.1) 50 -GAGGTGATGTTTTTGG- sections were stained with the relevant antibodies (Supplementary TAAACAGGCGGGGT-30 . The PCR temperature cycling conditions were File 1). Finally, sections were stained with 40 ,6-diamidino-2- as follows: initial denaturation at 92  C for 2 min; 30 cycles of phenylin- dole (DAPI) nucleic acid stain reactive (Invitrogen; Carls- denaturation at 92  C for 2 min, and annealing/elongation at 68  C for bad, CA), for 5 min and mounted with ProLongTM Gold Antifade re- 12 min. The final cycle was followed by extension at 68  C for 15 min. agent (Molecular probes; Eugene, OR). Images were taken using a The PCR product was separated in a 0.8% agarose gel stained with Leica DM-IRB inverted microscope (Leica Microsystems; Wetzlar, ethidium bromide (Sigma) and visualized with the Gel Doc Germany) and a Zeiss LSM980 fully-spectral confocal microscope with XR þ System (Bio-Rad; Hercules, CA). AiryScan 2 confocal super-resolution detector (Zeiss Microscopy; Jena, Germany). Super-resolution images were analyzed with the Zeiss-ZEN 2.9. mRNA levels software (Zeiss Microscopy). Total RNA was obtained from quadriceps and TA muscles from WT and McArdle mice as previously described [25] following the manufac- 2.6. Electron microscopy turer’s instructions (TRIzol, Invitrogen). RNA was treated with DNase I, To visualize the ultrastructure of mitochondria, we used electron mi- amplification grade (Invitrogen) to eliminate any DNA trace. Comple- croscopy. Briefly, a McArdle mouse was perfused with 2% EM-grade mentary DNA was synthesized from RNA using the high-capacity glutaraldehyde in 0.05 M phosphate buffer (SigmaeAldrich), the TA complementary DNA archive kit (Applied Biosystems, Foster City, was dissected, and a piece was cut from the muscle. From patients, a CA), which uses random primers. We used real-time PCR, with Taq- small piece of muscle biopsy was incubated in the same buffer as the Man fluorogenic probes in a 7900 real-time PCR System (Applied mouse was perfused with for 48 h at 4 . Muscle pieces from mice and Biosystems) to assess mRNA levels of the following genes in quadri- patients were then rinsed in 0.15 M sodium cacodylate buffer (Sigmae ceps and TA muscles: peroxisome proliferator activated receptor alpha Aldrich) for 30 min, and postfixed in 2% osmium tetraoxide (Poly- (Ppara, Mm00440939_m1) and beta (Pparb, Mm00803184_m1), sciences Inc., Warrington, PA) in 0.05 M potassium ferrocyanide (K4Fe peroxisome proliferative activated receptor, gamma, coactivator 1 [CN]6) (SigmaeAldrich) overnight at 4  C, rinsed for 3  10 min in alpha (Mm01208835_m1) and beta (Mm00504730_m1), estrogen distilled water, stained with 1% uranyl acetate in distilled water for 1 h related receptor, alpha (Mm00433143_m1), mitochondrially encoded before 3  10 min rinse in water before graded dehydration in ethanol, NADH:ubiquinone oxidoreductase core subunit 4 (mtND4 [forward followed by graded infiltration with propylene oxide and Epon (TAAB primer: 50 TGCATCAATCATAATCCAAACTCCATGA 30 ; reverse primer: 50 Laboratories Equipment Ltd., Aldermaston, UK). The sample was GGCAGAATAGGAGTGATGATGTGA 30 : probe FAM-50 CCGACATCA embedded in 100% Epon overnight followed by polymerization at TTACCGGGTTTTCCTCTTG 30 -MGB]) and 16 S RRNA (mtRNR2) [forward 60  C for 24 h. After postfixation, the fibers were rinsed twice in 0.1 M primer: 50 aatggttcgtttgttcaacgatt 30 ; reverse primer: 50 agaaacc- sodium cacodylate buffer at 4  C, dehydrated through a graded series gacctggattgctc 30 ; probe: FAM-50 aagtcctacgtgatctgagtt 30 -MGB]). of alcohol at 4 to 20  C, infiltrated with graded mixtures of propylene Results were normalized to peptidylprolyl isomerase A (Ppia) gene oxide and Epon at 20  C, and embedded in 100% Epon at 30  C. mRNA levels (probe Mm02342430_g1) and quantified using 7900 SDS Ultrathin sections were cut from the transversely oriented muscle using v2.4.1 software (Applied Biosystems). a Leica EM UC7 ultramicrotome (Leica Microsystems) and contrasted with uranyl acetate and lead citrate (Leica Microsystems). Sections 2.10. Western blots were visualized in a CM100 transmission electron microscope (Philips; Muscle samples from gastrocnemius, quadriceps, and TA muscles Amsterdam, The Netherlands), fitted with a 4-Mpixel Veleta camera from WT and McArdle mice were homogenized using Pellet pestles (Olympus Soft Imaging Solutions GmbH; Munster, Germany). Image Cordless motor (SigmaeAldrich) in cold homogenization buffer (Trise acquisition was done using iTEM software (Olympus Soft Imaging Hcl 20 mM, NaCl 150 mM and Triton 100 1%) and centrifuged Solutions GmbH). (10,000 g) for 10 min at 4  C. The samples were placed in boiling water for 3 min, before 30 mg of protein was applied to each lane. 2.7. Mitochondrial DNA copy number Unspecific binding sites on the blots were blocked by incubation in 5% Total DNA was extracted from TA and quadriceps muscles from WT low-fat dried milk powder in a phosphate buffered saline. Membranes and McArdle mice using the QIAamp DNAMini Kit (Qiagen; Hilden, were incubated in primary antibodies overnight at 4  C Germany). Mitochondrial DNA (mtDNA) copy number was determined (Supplementary File 1) and in secondary antibodies for 3 h at room by real-time PCR, as previously described [24], and expressed as the temperature. Ponceau S and Amido Black staining solutions (both from ratio between mitochondrially-encoded 16 S RRNA (mtRNR2) gene SigmaeAldrich) were used as a loading control for all the membranes. copy number and nuclear Angiogenin-1 (Ang) gene copy number, with Membranes were developed with Immobilon Western Chem- Ang being a nuclear single-copy gene. Results are expressed as the iluminiscent HRP Substrate (Merck-Millipore) and images obtained mtDNA/nuclear DNA (mtRNR2/Ang) copy number ratio. The following with a LICOR OdysseyÒ Fc Imaging System (LICOR Biosciences; Taqman fluorogenic probes were used: (i) mtRNR2 gene (forward Lincoln, NE) and quantified with Image StudioÔ Lite software, version primer: 50 AATGGTTCGTTTGTTCAACGATT 30 ; reverse primer: 50 5.2 (LICOR Biosciences). AGAAACCGACCTGGATTGCTC 30 ; probe: FAM-50 AAGTCCTACGT- GATCTGAGTT 30 -MGB); (ii) Ang (Mm00833184_s1). 2.11. Mitochondrial oxidative phosphorylation system 150 mg of gastrocnemius, quadriceps and TA muscle samples from 2.8. Mitochondrial DNA deletion WT and McArdle mice were resuspended in 200 ml mannitol buffer, pH To determine the presence of mtDNA deletions, a long-polymerase 7.2 (mannitol 225 mM, sucrose 75 mM, TriseHCl 10 mM and EDTA chain reaction (PCR) was performed using the following primers: for- 0.1 mM), and sonicated during 5 s at moderate intensity to obtain a ward primer (nucleotide position 2473e2505 of the NCBI Reference homogenous solution. Protein content was determined using Coo- Sequence: NC_005089.1) 50 -GGTTCGTTTGTTCAACGATTAAAGTCC- massie Plus protein assay reagent (Pierce; Rockford, IL), adjusting the TACGTG-30 ; and reverse primer (nucleotide position 1953e1924 of the solution with mannitol buffer to a final protein concentration of 2 mg/ MOLECULAR METABOLISM 66 (2022) 101648 Ó 2022 The Author(s). Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 3 www.molecularmetabolism.com Original Article ml. Mitochondrial respiratory complex (C) I, II, and IV, and CS activities check whether the analyzed data followed a normal distribution. The were determined as previously described [26]. mean VO2peak values of patients and controls (men and women separately) were compared with the Student’s t test for unpaired 2.12. Mass spectrometry data. The level of significance was set at 0.05 for all statistical 150 mg of quadriceps tissue from WT and McArdle mice were ho- analyses. mogenized using mortar and pestle and resuspended in 10 mM K2HPO4 (100 mL) and 10 volumes of chilling (20  C) 60% 3. RESULTS methanol were added. Then samples were sonicated three times for 20 s with 1 min incubation on ice between cycles. Sonicated samples 3.1. Structural damage, glycogen levels and mitochondrial content were incubated on ice for 10 extra min and centrifuged at full speed In order to elucidate whether mitochondrial structure and function is (17,000 g) in a refrigerated (4  C) microcentrifuge. Supernatant was affected in the different fiber types from the skeletal muscle of filtered using nylon 0.2 mm microfilters and dryer under nitrogen McArdle mice, we first performed an in-depth histopathologic anal- stream to preserve oxygen-free atmosphere. Dried samples were ysis. H&E staining showed muscle fibers in disarray, of varying sizes, conserved in ultra-freezer (80  C) until further use. For NAD/NADH as well as the presence of large intra-fiber voids and some centrally- determination dried samples were reconstituted in 50 mL mobile phase nucleated fibers (Figure 1A i-ii) [20,21]. Also characteristic was the solvent A. UPLC separation was carried out using a gradient of intense positive PAS staining, reflecting large glycogen accumulation, acetonitrile (solvent A 5% acetonitrile in 10 mM ammonium carbonate, with some fibers accumulating more glycogen than others (Figure 1A solvent B 95% acetonitrile in 10 mM ammonium carbonate). The two- iii-iv, yellow and black arrows). Amylase treatment reduced glycogen step gradient consisted in a smooth ramp (initial 5%B raising to 60%B content to normal levels (Figure 1A v), indicating that the accumulated in min 7.5) followed by a 2 min ramp to wash all retained analytes (min glycogen had normal structure. Myophosphorylase staining in WT 9.5, 100%B). Washing conditions were sustained for 1 min (min 10.5) mice showed different intensities depending on fiber type, being more and initial conditions were restored (min 13.5). Column was re- intense in smaller fibers (i.e., usually more oxidative, Figure 1A vi, equilibrated for 5 additional min before next injection cycle. Separa- yellow and black arrows). As expected, myophosphorylase staining tion was achieved using an I-Class Waters UPLC equipped with Acquity was null in the muscle fibers of McArdle mice (Figure 1A vi). This BEH Amide 1.7 mm 2.1  50mm column (Waters, cat#186004800). differential degree of myophosphorylase staining in WT suggests Nucleotides were detected using a single-quadrupole Waters QDA ESI differential glycogen disposal among fiber types. Interestingly, those MS (probe 500  C, 800 V ionization needle, 15 V collision energy, SIR fibers with higher SDH and COX activity staining as well as cyto- channels 664 and 666 m/z). Analytical standards for NADþ/NADH chrome c oxidase subunit II (COX II) and voltage-dependent anion were used for UV-Vis spectral confirmation and elution time (spiked channel (VDAC) immunodetection were the ones showing the most samples). affected morphology in McArdle mice, with larger intra-fiber voids For pyruvate, lactate and glucose-6-phsophate MS SIR channels were (Figure 1A vii-x). Additionally, SDH activity and PAS staining in set to 111, 90 and 260 m/z respectively (probe 500  C, 800 V ioni- muscle consecutive sections showed that fibers with larger glycogen zation needle, 20 V collision energy). Analytical standards of these depots were those with higher SDH activity (Figure 1A xi-xii). metabolites were used for MS confirmation and elution time (spiked Furthermore, immunofluorescence staining against COX II and samples). VDAC mitochondrial proteins showed differential distribution of sub- cellular mitochondria in WT and McArdle mice (Supplementary File 2.13. Peak oxygen uptake in patients 3A). Using specific antibodies against the different fibers types we The patients and their controls were assessed from January 2011 to observed that type I and IIX/IIB fibers were less structurally damaged December 2018 in the exercise physiology laboratory of the Uni- than IIA, IIA/IIX and IIX fibers in the skeletal muscle of McArdle mice versidad Europea de Madrid (UEM). Inclusion criteria for both patients (Figure 1A xiii-xv). In addition, double immunostainings of COX II with and controls were being free of any major cardiorespiratory disease or myosin heavy chain (MHC) type I, IIA or IIB showed that IIA fibers had severe condition contraindicating exercise. In patients (all identified in higher COX II staining intensity than type I and IIB fibers (Figure 1A the Spanish registry of patients [27]), diagnosis of McArdle disease xvi-xviii), with higher VDAC staining observed in IIA and IIX fibers was ascertained by the identification of a documented pathogenic (Supplementary File 3B). Finally, triple fiber type immunostaining mutation in both allelesdwhether in homozygosity or compound and PAS staining in consecutive muscle sections showed that IIA and heterozygositydof the PYGM gene (Supplementary File 2). All the IIX fibers accumulated more glycogen than IIB fibers (Figure 1A xix tests were performed using the same cycle-ergometer (Ergoselect and xx). 220 KL, Ergoline GmbH; Bitz, Germany) and metabolic cart (Vmax 29C, Overall, the aforementioned results would support that the most Sensormedics Corp.; Yorba Linda, CA), The maximum test was pre- degenerated muscle fibers in the McArdle mouse model are those with ceded by a 5-min free-wheel pedaling warm-up, after which the higher glycogen and mitochondrial contentthat is, type IIA and IIX fi- workload was increased following a ramp protocol (10 [patients] or bers. Therefore, the next step was to assess whether mitochondrial 20 W/min [control]s) until volitional exhaustion (9). The VO2peak was integrity was conserved in these highly affected muscle fibers. In this determined as the highest VO2 value (20-second average) recorded effect, electron microscope images showed the presence of abnor- during the tests. All the tests were supervised by the same researchers mally large mitochondria that had lostwhether partially or total- (AL, AS, CFL). lytheir cristae (with rupture and discontinuation of the inner and outer membranes) or were thoroughly degenerated (Figure 1B). The 2.14. Statistical analyses most structurally damaged mitochondria were those surrounded by In McArdle mice, to analyze the relevant genotype effect in normally glycogen particles (Figure 1B). Besides the altered morphology, distributed values, a Student t-test was applied, whereas the non- mitochondria had also lost their normal regular distribution in the parametric ManneWhitney U-test was used when values were not intermyofibrillar space along both the A band and Z line normally distributed. The ShapiroeWilk normally test was used to (Supplementary File 3C). 4 MOLECULAR METABOLISM 66 (2022) 101648 Ó 2022 The Author(s). Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). www.molecularmetabolism.com Figure 1: Histopathological analysis of the skeletal muscle from the McArdle (p.R50*/p.R50* Pygm) mouse model. (A) Histochemical, immunohistochemical and immunofluorescence analysis of the skeletal muscle from McArdle mice; i) and ii) H&E staining in the quadriceps from wild-type and McArdle mice; iii) and iv) PAS staining in the gastrocnemius from wild-type and McArdle mice; v) PAS-Diastase staining in the gastrocnemius from McArdle mice; vi) Myophosphorylase activity staining in the soleus from wild- type and McArdle mice; vii and viii) SDH and COX activity staining, respectively, in the gastrocnemius from McArdle mice; ix) and x) COXII and VDAC immunofluorescence, respectively, in the quadriceps from McArdle mice; xi) and xii) SDH activity and PAS staining, respectively, in gastrocnemius serial sections from McArdle mice; white asterisks show those fibers with high SDH activity and PAS staining; xiii) immunofluorescence showing type I (green) and IIA (red); xiv) type IIA (red) and IIX (green); and xv) type IIX (green) and IIB (red) muscle fibers in the gastrocnemius from McArdle mice; xvi) immunofluorescence staining for COXII (red) and type I fibers (green) in the gastrocnemius from McArdle mice, with yellow asterisks marking the same fibers with and without type I fiber staining; xvii) immunofluorescence staining for COXII (green) and type IIA fibers (red) in the gastrocnemius from McArdle mice, with yellow asterisks marking the same fibers with and without type IIA fiber staining; xviii) immunofluorescence staining for COXII (green) and type IIB fibers (red) in the gastrocnemius from McArdle mice, with yellow asterisks marking the same fibers with and without type IIB fiber staining; xix) immunofluorescence staining showing type IIA (pink), IIX (green), IIB (orange) muscle fibers, and DAPI (the nuclei, blue) in the tibialis anterior muscle of McArdle mice; and xx) consecutive serial section from (xix) stained with PAS; for xix) and xx) yellow asterisks mark muscle fibers with more intense PAS staining while white asterisks mark muscle fibers presenting less intense PAS staining. All scale bars correspond to 75 mm. (B) Electronic microscopy analysis of the ultrastructure integrity of mitochondria in the skeletal muscle of McArdle mice (TA). Z- line mitochondria are indicated with black arrows, while A-band mitochondria are shown with a yellow arrow. Scale bars correspond to 1 or 2 mm as indicates. Abbreviations; QUAD: quadriceps; GAST: gastrocnemius; SOL: soleus; TA: tibialis anterior; WT: wild-type; McA: McArdle; Gly: glycogen. Thereafter we aimed to assess whether the observed loss of mito- 3.2. Mitochondrial network chondrial integrity in those fibers with higher glycogen content was VDAC fluorescent staining showed massive alteration of the mito- associated with mitochondrial network disruption, with the results chondrial network both along the A band and across the Z line in shown below. McArdle mouse muscles (Figure 2A). Additionally, fragmented MOLECULAR METABOLISM 66 (2022) 101648 Ó 2022 The Author(s). Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 5 www.molecularmetabolism.com Original Article Figure 2: Cytoskeleton and mitochondrial network integrity analysis in the skeletal muscle from the McArdle (p.R50*/p.R50 3 Pygm) mouse model. (A) VDAC immunofluorescence analysis of mitochondrial network. Scale bars correspond to 25 mm. (B) Mitochondrial morphology analysis showing different mitochondrial sizes and shapes in McArdle mice; squared regions are magnified and shown in detail. Scale bars correspond to 5 mm. (C) Immunofluorescence analysis of the cytoskeleton and mitochondrial network integrity in the skeletal muscle from wild-type and McArdle mice; wild-type (i-ii) and McArdle mice (iii-iv) immunofluorescence staining of VDAC (green) and aeactinin (red); loss of aeactinin staining is clearly visible in (iii-iv), while a complete disruption of the mitochondrial network and cytoskeleton is shown in the square box in (iv). Wild-type (v-vii) and McArdle mice (viii-ix) immunofluorescence staining of VDAC (green) and aeactin (red); irregular A-band length is visible in (viii), while disrupted actin microfilaments and/or non-polymerized aeactin monomers are shown in (ix). Wild-type (x-xi) and McArdle mice (xii-xiii) immunofluorescence staining of VDAC (green) and beactin (red); a complete disruption of the mitochondrial network and beactin cytoskeleton structures is observed in (xii-xiii); the square box region shows isolated mitochondria and actin monomers. Wild-type (xiv-xv) and McArdle mice (xvi-xviii) immunofluorescence staining of VDAC (green) and a-tubulin (red); subsarcolemmal, as well as longitudinal and transversal interaction of microtubules with mitochondria is observed in (xiv-xv), while microtubules surrounding fiber voids and central nucleated fibers, aggregated in thick tubulin bundles or completely disrupted and/or non-polymerized are shown in (xvi-xviii). Wild-type (xix-xx) and McArdle mice (xxi-xxii) immunofluorescence staining of VDAC (red) and bII- tubulin (green); bII-tubulin aggregates in fiber voids (xxi) or surrounding centrally nucleated fibers (xxii) are observed, while irregularly dispersed bII-tubulin staining non co- localizing with mitochondria is observed in the square box (xxi). Wild-type (xxiii-xxiv) and McArdle mice (xxv-xxvi) immunofluorescence staining of VDAC (red) and plectin (green); regions surrounding fiber voids with no plectin staining are shown in (xxv-xxvi); completely disrupted mitochondrial network with the presence of isolated compact mitochondria can be seen in the square box in (xxv). All scale bars correspond to 25 mm. (D) Western blot analysis of cytoskeleton protein levels in the quadriceps from wild-type and McArdle mice (N ¼ 7 (3 male) in both groups). As all the data followed a normal distribution according to the ShapiroeWilk normally test, the Student’s t-test was applied in all cases. Asterisk correspond to p < 0.005 in the statistic test. Abbreviations: VDAC: voltage-dependent anion channel; QUAD: quadriceps; TA: tibialis anterior; WT: wild-type; McA: McArdle; Gapdh: Glyceraldehyde-3-phosphate dehydrogenase; Amido: Amido black staining. 6 MOLECULAR METABOLISM 66 (2022) 101648 Ó 2022 The Author(s). Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). www.molecularmetabolism.com mitochondrial structures as well as isolated mitochondria showed a network structures, to its isolate localization within the sarcoplasm wide variety of shapes and size associated with network disruption. (Supplementary File 5A), and/or co-localizing with isolated (and Among these, “compact”, elongated, branched as well as c-shaped enlarged) mitochondria (Figure 3A), suggesting the occurrence of or “donut” like mitochondriadwith this type of alterations previously mitochondrial fission events in fiber regions with a highly unstructured shown to be associated to cellular stress [28,29]dwere commonly mitochondrial network; interestingly, fission-like events were also observed close to large fiber voids (glycogen depots) (Figure 2B). observed (Supplementary File 5A and B; white arrowheads), where a-tubulin immunofluorescence staining suggested that microtubules 3.3. Cytoskeleton and mitochondrial network in McArdle mice might help Drp1 to constrain mitochondria to drive scission In a subsequent step we aimed to determine whether mitochondrial (Supplementary File 5B; white arrowheads). Further analysis in network disruption was associated with disarray of the cystoskeleton McArdle mice revealed that Fis1 co-localized with VDAC staining in structure, as connections between the cytoskeleton and mitochondria longitudinal mitochondrial aggregates (Figure 3A), or as in the case of are very important for the positioning of mitochondria within the cell Drp1, in fission-like events in fragmented mitochondrial structures [30]. In this regard, different cytoskeletal proteins comprising inter- (Supplementary File 5C, white arrowheads). In the case of fusion mediate filaments (plectin and desmin), microtubules (a and bII- proteins, Mfn2 staining was more pronounced in mitochondria located tubulin), actin microfilaments (a and b-actin) as well as the sarco- in close proximity to fiber damage, central nucleated fibers, and large mere (a-actinin-3) were analyzed by immunofluorescence, along glycogen depots (Figure 3A), while Opa1 staining co-localized with with the mitochondrial protein VDAC. In WT mice, the different sar- isolated and dispersed mitochondria, as well as nearby regions of large comeric and cytoskeleton proteins showed a normal disposition in all glycogen depots and central nucleated fibers (Figure 3A). Interestingly, muscle fibers, and VDAC staining indicated the presence of an intact the staining of Mfn2 and Opa1 in McArdle mouse muscle did not mitochondrial network throughout the entire longitudinal and trans- entirely co-localize with that of mitochondria, especially in regions of versal axis of muscle fibers (Figure 2C). By contrast, in the skeletal fiber damage and regeneration (Figure 3A). muscle fibers from McArdle mice the cytoskeleton was especially As Mfn2 has also been described to be localized in endoplasmic (ER)/ disrupted in regions with large fiber voidsdincluding myofibrils in sarcoplasmic reticulum (SR) membranes, we next wanted to assess disarray presenting loss of a-actinin-3 staining, actin microfilaments whether the SR network was also disrupted in the skeletal muscle of with irregular A-band length, regions enriched with non-polymerized McArdle mice and found that calsequestrin staining showed enlarged a, b-actin as well as a-tubulin monomers (Figure 2C and SR distribution in longitudinal bundles in comparison to WT animals Supplementary File 4A), aggregation of thick tubulin bundles nearby (Figure 3A). On the other hand, no significant changes in the quantity of fiber voids (Figure 2C and Supplementary File 4A), irregularly the fission proteins Drp1 and mitochondrial dynamics protein of 49 kDa dispersed bII-tubulin throughout the sarcolemma (sometimes distant (Mid49), as well as of the fusion-related protein Opa1, were found in from any mitochondrion) forming clusters within fiber voids or ag- the muscle of McArdle mice; in turn, these animals showed lower gregates surrounding central nucleated fibers, and areas devoid of levels of Mfn2 fusion protein levels, suggesting an impairment in plectin and desmin staining (Figure 2C and Supplemnetary File 4B), mitochondrial fusion (Figure 3B). along with the loss of transverse and longitudinal mitochondrial network. Interestingly, mitochondria and microtubules were inter- 3.5. Mitochondrial content and biogenesis twined forming a helicoid-like structure (Supplementary File 4A, As mitochondrial network and dynamics have been directly related to white arrowheads) that in the case of McArdle mice could help to mitochondrial function, homeostasis and quality control, we next generate tension in order to maintain together the disrupted mito- examined mitochondrial content and biogenesis in the skeletal muscle chondrial network fragments. Additionally, in the skeletal muscle of McArdle mice [28,33]. Remarkably, lower mtDNA levels were found fibers of McArdle mice some microtubules crossed through the holes compared with WT animals (Figure 4A), albeit with no evidence of of the “donut”-like mitochondria (Supplementary File 4A, yellow mtDNA deletions (Figure 4A). Additionally, lower levels of mitochondrial arrowheads). transcription factor A (Tfam) and nuclear respiratory factor (Nrf1) Western Blot protein quantification showed that the cytoskeleton protein levels, as well as lower mtND4, Pgc1a, Ppara, Pparb transcript disruption in the muscle fibers of McArdle mice was accompanied by levels were also observed in the TA muscle of McArdle mice lower a-actinin-3 and a-actin protein levels, and a trend to higher a- (Figure 4BeE), suggesting an impaired mitochondrial biogenesis. tubulin protein levels (Figure 2D), possibly in an attempt to reorganize Remarkably lower levels of VDAC protein levels were also found in the the entire cytoskeleton structure. gastrocnemius and quadriceps muscle of McArdle mice (Figure 4F), thereby suggesting an impairment in mitochondrial content. On the 3.4. Mitochondrial fission/fusion proteins other hand, the protein levels of nuclear factor erythroid 2-related It is known that disruption of the cytoskeleton structure and the factor 2 (Nrf2) were significantly higher in the TA of McArdle mice mitochondrial network may alter mitochondrial fusion/fission dynamics with respect WT mice (Figure 4E). Nrf2 is a master regulator of the [31,32]. Thus, we analyzed the fission-related proteins dynamin- cellular redox homeostasis [34], with increased levels of this molecule related protein 1 (Drp1) and mitochondrial fission 1 protein (Fis1), likely reflecting an attempt to counterbalance mitochondrial ROS and the fusion proteins mitofusin 2 (Mfn 2) and optic atrophy 1 (Opa 1) production derived from mitochondrial network disruption and by immunofluorescence. In WT mouse muscle, both fission proteins dysfunction. were found close to both Z-disc associated and subsarcolemmal mitochondria (Figure 3A and Supplementary File 5A and C) whilst the 3.6. OXPHOS-related proteins fusion proteins Mfn 2 and Opa 1 were localized at the level of Z-disc in Lower levels of several OXPHOS-related proteins were observed in the intermyofibrillar mitochondria, subsarcolemmal mitochondria, and in skeletal muscle of McArdle mice compared to WT, as reflected both by the mitochondria associated with the A-band (Figure 3A and lower Ndufa9 (complex I; CI) and CoxII (complex IV; CIV) protein levels Supplementary File 5D). By contrast, in McArdle mouse muscle, Drp1 in the quadriceps muscle, and lower complex III Uqcrc2 protein levels transitioned from Z-disc mitochondria in semi-preserved mitochondrial in the gastrocnemius muscle, respectively (Figure 5A). In order to MOLECULAR METABOLISM 66 (2022) 101648 Ó 2022 The Author(s). Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 7 www.molecularmetabolism.com Original Article Figure 3: Altered subcellular localization of fission/fussion proteins in the skeletal muscle from the McArdle (p.R50*/p.R50 3 Pygm) mouse model. (A) Immuno- fluorescence analysis of mitochondrial fission/fusion proteins in the skeletal muscle from McArdle mice. Wild-type (i-ii) and McArdle mice (iii-iv) immunofluorescence staining of VDAC (green) and Drp1 (red); Drp1 is observed in close proximity to mitochondria within the network in (i-ii), while Drp1 staining co-localizes with enlarged and isolated mitochondria in (iii-iv); square boxes show magnification of the Drp1-VDAC staining co-localization, while 3D Surface reconstruction images from the square box regions using the Zen-Blue software are also shown. Wild-type (v-vi) and McArdle mice (vii-ix) immunofluorescence staining of VDAC (red) and Fis1 (green); as with Drp1, Fis1 are found in close proximity of mitochondria within the network in (v-vi), while is found co-localizing with VDAC staining in longitudinal mitochondria aggregates in (vii-ix); square regions show magnification of the VDAC-Fis1 co-localization in the aggregates. Wild-type (x-xi) and McArdle mice (xii-xiv) immunofluorescence staining of VDAC (red) and Mfn2 (green); square boxes in (x-xi) show magnification of Mfn2 staining in close proximity of mitochondria in the Z-disc, while in (xii-xiv) show magnification of Mfn2 staining located in close proximity to fiber damage, central nucleated fibers, and large glycogen depots. Wild-type (xv-xvi) and McArdle mice (xvii-xix) immunofluorescence staining of VDAC (green) and Opa1 (red); in (xv-xvi) square boxes show co-localization of Opa1 in Z-disc mitochondria, while in (xvii-xix) square boxes show Opa1 co-localization with dispersed and isolated mitochondria. Wild-type (xx-xxi) and McArdle mice (xxii-xxiv) immunofluorescence staining of VDAC (green) and calsequestrin (red); enlarged calsequestrin distribution in longitudinal bundles in comparison to wild-type mice is observed in (xxii-xxiv). All scale bars correspond to 25 mm. (B) Western blot analysis of mitochondrial fusion and fission protein levels in the quadriceps from wild-type and McArdle mice (N ¼ 6e7 (3 male) in both groups for the different variables). For Drp1 analysis, the Student’s t-test was applied as data followed a normal distribution according to the ShapiroeWilk normally test, whereas the ManneWhitney U-test was used for Mid49, Mfn2 and Opa1. Asterisk correspond to p < 0.005 in the statistic test. Abbreviations: Drp1: Dynamin-related protein 1; Fis1: mitochondrial fission 1 protein; Mfn2: Mitofusin 2; Opa 1: optic atrophy protein 1; Calseq: calsequestrin; Mid49: mitochondrial dynamic protein of 49 kDa; QUAD: quadriceps; TA: tibialis anterior; WT: wild-type; McA: McArdle; Gapdh: Glyceraldehyde-3-phosphate dehydrogenase; Amido: Amido black staining. 8 MOLECULAR METABOLISM 66 (2022) 101648 Ó 2022 The Author(s). Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). www.molecularmetabolism.com Figure 4: Analysis of mitochondrial biogenesis and content. (A) Analysis of mitochondrial DNA content (mtDNA) depletion in the tibialis anterior [N ¼ 7 wild-type (71% male) and 5 McArdle mice (40% male)] and quadriceps [N ¼ 10 wild-type (70% male) and 8 McArdle mice (63% male)] and deletions (in quadriceps) using qPCR and long PCR, respectively. (B) Western blot analysis of Tfam protein relative levels in tibialis anterior [N ¼ 10 wild-type (50% male) and 7 McArdle mice (43% male)] and quadriceps [N ¼ 7 wild-type (3 males) and 7 McArdle mice (3 males)]. (C) qPCR analysis of mteNd4 and mt-Rnr2 mRNA relative levels in tibialis anterior [N ¼ 5e6 wild-type (3 males) and 5 McArdle mice (3 males)] and quadriceps [N ¼ 5 wild-type (3 males) and 5 McArdle mice (3 males)]. (D) qPCR analysis of Pgc1a, Pgc1b, Ppara, Pparb and Esrra mRNA relative levels in the tibialis anterior [N ¼ 4e5 wild-type (2e3 males) and 4e5 McArdle mice (2e3 males)] and quadriceps [N ¼ 5e6 wild-type (50e60% male) and 4e5 McArdle mice (60e75% male)]. (E, F) Western blot analysis of Ppard, Nrf1, Nrf2, Vdac and CS protein relative levels in tibialis anterior [N ¼ 6e7 wild-type (43e67% male) and 5e7 McArdle mice (29e40% male) and quadriceps (N ¼ 7 (3 male) in both groups for the different proteins). When data followed a normal distribution (i.e. mtDNA quadriceps, Tfam, Ppard, Nrf1, Nrf2 quadriceps, VDAC and CS quadriceps protein levels) the Student’s t-test was applied, whereas the non-parametric ManneWhitney U-test was used for those (mtDNA tibialis anterior, mteNd4, mt-Rnr2, Pgc1a, Pgc1b, Ppara, Pparb, Esrra mRNA levels and Nrf2 and CS tibialis anterior protein levels) that were not normally distributed. Asterisks correspond to p < 0.005 in the statistic tests. Abbreviations: QUAD: quadriceps; TA: tibialis anterior; WT: wild-type; McA: McArdle; Tfam: transcription factor A, mitochondrial; mteNd4: mitochondrially encoded NADH:Ubiquinone oxidoreductase core subunit 4; mt-Rnr2: mitochondrially encoded 16 S rRNA; Pgc1a and b: PPARG Coactivator 1 alpha and beta; peroxisome proliferator activated receptor alpha and beta; Esrra: estrogen related receptor alpha; Nrf1: nuclear respiratory factor 1; Nrf2: nuclear factor erythroid 2-related factor 2; Vdac: voltage-dependent anion channel; CS: citrate synthase. determine whether mitochondrial network disruption affected mito- TA and quadriceps, with the results showing significantly lower levels chondrial OXPHOS complex activities, the enzyme activity of CI, CII and of CI and CIV activities in the gastrocnemius of McArdle mice vs WT CIV was measured in total muscle homogenates from gastrocnemius, controls (Figure 5B). MOLECULAR METABOLISM 66 (2022) 101648 Ó 2022 The Author(s). Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 9 www.molecularmetabolism.com Original Article Figure 5: Analysis of OXPHOS protein levels and activity. (A) Western blot analysis of Ndufa9 (complex I; CI), Sdha (complex II; CII), Uqcrc2 (complex III; CIII) and CoxII (complex IV; CIV) protein relative levels in the gastrocnemius [N ¼ 7 wild-type (57% male) and 7 McArdle mice (29% male)], tibialis anterior [N ¼ 10 wild-type (50% male) and 8 McArdle mice (25% male)] and quadriceps (N ¼ 7 (3 male) in both groups for the different proteins). (B) Citrate synthase and OXPHOS CI, CII and CIV activities in total tissue homogenates from gastrocnemius [N ¼ 10 wild-type (60% male) and 10 McArdle mice (40% male)], tibialis anterior [N ¼ 5 wild-type (60% male) and 4 McArdle mice (50% male)] and quadriceps [N ¼ 8 wild-type (50% male) and 7 McArdle mice (57% male)]. (C, D and E) Mass spectrometry analysis of glucose, glucose-6-phosphate, lactate, pyruvate, NADþ and NADH in total quadriceps homogenates from wild-type and McArdle mice. In these experiments, the same number of mice (N ¼ 5) and sex distribution (40% male) was studied in the two groups. When data followed a normal distribution (Ndufa9, Sdha, Uqcrc2, and Cox II protein levels, as well as CS quadriceps activity) a Student’s t-test was applied, whereas the non-parametric ManneWhitney U-test was used when values were not normally distributed (CS activity in gastrocnemius and tibialis anterior, complex I, II and IV activities, as well as glucose, glucose-6-p, lactate, pyruvate, NADþ and NADH levels in the quadriceps). Asterisks correspond to p < 0.005 in the statistic tests. Abbreviations: GAST: gastrocnemius; QUAD: quadriceps; TA: tibialis anterior; WT: wild-type; McA: McArdle; Ndufa9: NADH: Ubiquinone oxidoreductase subunit A9; Sdha: succinate dehydrogenase complex flavoprotein subunit A; Uqcrc2: ubiquinol-cytochrome c reductase core protein 2; Cox II: cytochrome c oxidase subunit II; CS: citrate synthase; GLUC: glucose (non- phosphorylated); GLUC-6-P: glucose-6-phosphate. 10 MOLECULAR METABOLISM 66 (2022) 101648 Ó 2022 The Author(s). Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). www.molecularmetabolism.com Next, we wanted to assess whether the full blockade of glycogen mitochondrial network in the mouse/patient skeletal muscle fibers, degradation that characterizes McArdle disease reduced glycolytic flux, together with an altered subcellular localization of mitochondrial fission further contributing to OXPHOS impairment. In this effect, mass and fusion proteins, as well as the SR protein calsequestrin, thereby spectrometry results indicated lower levels of glucose, glucose-6- disturbing the proper function of mitochondrial dynamics. We also phosphate, pyruvate and lactate but higher NADH levels in McArdle found an impairment in mitochondrial content and biogenesis, a mouse muscle compared to WT (Figure 5CeE). decrease in OXPHOS complex proteins and activities, together with high muscle NADH but reduced muscle levels of glucose-6-phosphate 3.7. Peak aerobic capacity in patients and of the glycolytic products pyruvate and lactate. In all, these results We finally aimed at providing strong corroborating evidence for a major suggest that the deficiency in oxidative metabolism in McArdle disease impairment of muscle oxidative metabolism in McArdle disease pa- might be caused by not only by reduced levels of OXPHOS substrates tients, as reflected at the whole-body level during intense exertion as previously reportedand corroborated herebut also by a through determination of the universal measure VO2peak. We therefore disruption of the mitochondrial networkwhich was shown here in compared the VO2peak of the largest series of McArdle disease patients detail for the first time. The combined effect of OXPHOS deficiency and ever analyzed (N ¼ 145, 34  14 years) and an age and sex-matched structural mitochondrial network damage and subsequent dysfunction control group (N ¼ 133, 33  13 years). The PYGM genotypes of of these organelles might help to explain why treatment with a com- patients leading to diagnosis of McArdle disease are displayed in pound with an anaplerotic role in the Krebs cycle, triheptanoin, has Supplementary File 2. Both male (N ¼ 82) and female (N ¼ 63) failed to improve muscle oxidative capacity and exercise capacity in patients showed significantly lower VO2peak levels than their sex- patients with McArdle disease [35]. Indeed, although triheptanoin matched controls (by 51% and 53%, respectively, Figure 6A), could replenish Krebs cycle intermediates, this positive effect might be with only 8% of patients (both sexes combined, vs 88% of controls) counterbalanced by the presence of damaged mitochondria in the reaching the minimum threshold for optimal cardiovascular healthd context of a disrupted network of these organelles, thereby suggesting i.e., VO2peak  8 metabolic equivalents of task (MET), with 1 ME T that impairment in glycolytic flux is not the only mechanism under- equivalent to an oxygen uptake level of 3.5 ml O2/kg/min. pinning the impaired aerobic capacity of these patients. In the same line, although the peak aerobic capacity of p.R50*/p.R50* mice in- 3.8. Muscle cytoskeleton and mitochondrial network in two creases significantly with regular endurance training (for eight weeks, patients which would be equivalent to wone-year duration when translated to An important step was to provide preliminary evidence whether the ‘human lifespan’), it still remains remarkably lower (wby 50%) loss of cytoskeleton integrity with subsequent disruption of the mito- compared to untrained WT mice [9], which is in accord with findings in chondrial network and presence of damaged mitochondria found in the patients [4,36]. skeletal muscle of McArdle mice might also occur in affected patients, A major finding of the present study is that most damaged muscle thereby accountingdat least partlydfor their low muscle oxidative fibers in McArdle mice were those with higher glycogen and mito- capacity (as in turn reflected by very low VO2peak values). To this end, chondrial content. In this effect, although a causative association be- vastus lateralis muscle biopsies from two patients were analyzed, with tween the amount of glycogen content, on the one hand, and the the results suggesting apparent disruption of the mitochondrial degree of mitochondrial network disruption, on the other, cannot be network, together with loss of intermediate filaments, actin microfil- inferred, previous studies in McArdle mice have shown that the large aments, and microtubules (Figure 6B). Consecutive muscle sections muscle glycogen depots that characterize this disease cause myofibrils staining for fast MHC and SDH showed that fibersdpresumably type to tear, change direction, and split from adjacent myofibrils, while T- IIAdwith high mitochondrial content were the most affected ones tubules appear misaligned to myofibrils [21]. These previous results (Figure 6B i-ii), with these preliminary data in two patients being in line suggest that glycogen accumulation alters the overall ultrastructure of with the results obtained mice. Additionally, SDH staining showed the muscle fibers. In the present study, we have further observed that several different fibers with high amount of subsarcolemmal mito- most damaged fibers presented with a remarkable disruption of in- chondrial accumulation (Figure 6B iii-iv). Finally, electron microscope termediate and actin microfilaments and microtubules, along with a analysis of vastus lateralis biopsies in two patients showed a large disturbance of the mitochondrial network. In this effect, it is well amount of damaged mitochondria within subsarcolemmal and inter- established that the location of mitochondria within the cell as well as myofibrillar glycogen-rich regions, together with a low amount of the arrangement of the network formed by these organelles is highly cristae and the presence of membrane-like structures, as well as dependent on the cytoskeleton structure [30,31,37]. In addition, the abnormally large mitochondria (Figure 6C). cytoskeleton, more specifically bII-tubulin, plays an important role in regulating the local diffusion of ADP through the VDAC pore in the 4. DISCUSSION mitochondrial outer membrane in order to control the kinetics of ATP synthesis [38,39]. Furthermore, tubulin has been recognized as an In the present work we have analyzed whether mitochondrial network inherent mitochondrial membrane component that acts as a regulator integrity and function is altered in the skeletal muscle of McArdle of the mitochondrial permeability transition pore and thereby plays a mouse model (with preliminary data in two patients), in order to gain role in apoptosis, supporting mitochondrial membrane potential and insight into the potential mechanisms underpinning the poor aerobic regulating OXPHOS [40]. Given all of the above, the large glycogen fitness previously documented in this condition, which was corrobo- deposits inside the skeletal muscle fibers of mice and patients are rated here using a very large cohort. We have found that, in McArdle likely to alter the overall structure of these cells (including the different muscle, most damaged fibers are those with a higher mitochondrial cytoskeleton components) and subsequently the localization and and glycogen content (i.e., type II in patients, and type IIA and IIX in network arrangement of mitochondria within the cell, thereby poten- mice), which show major disruption of the three main cytoskeleton tially affecting, at least partly, the function of these organelles. Addi- componentsactin microfilaments, microtubules and intermediate tionally, and probably as a consequence of mitochondrial network filaments. This, in turn, might contribute to cause a disruption of the disruption, we also observed an altered subcellular localization of the MOLECULAR METABOLISM 66 (2022) 101648 Ó 2022 The Author(s). Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 11 www.molecularmetabolism.com Original Article Figure 6: Peak oxygen uptake (VO2peak) and analysis of the cytoskeleton and mitochondrial network integrity in patients with McArdle disease. (A) VO2peak levels of patients with McArdle disease (N ¼ 145) versus control individuals (N ¼ 133). (B) Immunofluorescence analysis of the cytoskeleton and mitochondrial network integrity in quadriceps biopsies from two different McArdle disease patients; VDAC (red) and plectin (green) immunofluorescence staining (i-iv); VDAC (green) and a-actin (red) immuno- fluorescence staining (v-viii); cell nucleus (blue) are stained with DAPI. Scale bars correspond to 25 mm. (C) Consecutive quadriceps sections staining for fast muscle fibers and SDH activity; i and iii) fast muscle fibers immunohistochemical staining; ii and iv) SDH activity staining; yellow asterisks mark identical muscle fibers in i) and ii), as well as in iii) and iv). (D) Electronic microscopy analysis of mitochondria ultrastructure in quadriceps biopsies from two different McArdle disease patients. Scale bars correspond to 1 or 2 mm as indicated. Abbreviations; QUAD: quadriceps; McA: McArdle; Glcn: glycogen and Mi: mitochondria. 12 MOLECULAR METABOLISM 66 (2022) 101648 Ó 2022 The Author(s). Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). www.molecularmetabolism.com different proteins involved in mitochondrial fission (Drp1 and Fis 1) and profound dysregulation of calcium homeostasis, mitochondrial calcium fusion (Mfn2 and Opa1), as well as lower levels of Mfn2 in the skeletal overload, increase in reactive oxygen species and caspase-independent muscle of McArdle mice, possibly leading to lower mitochondrial apoptosis, and decrease in mitochondrial membrane potential as well as fusion. Furthermore, Drp1 frequently co-localized with isolated (and in the ATP production capacity of mitochondria [55]. Loss of function enlarged) mitochondria, thereby suggesting that mitochondrial fission variants in the chromosome 2 open reading frame 69 (C2orf69) gene, was actively involved in the separation of potentially damaged mito- which controls the levels of the glycogen branching enzyme, affects chondria from the network. Besides their negative effect on mito- mitochondrial membrane potential and oxidative respiration in cultured chondrial dynamics, changes in the cellular distribution of the different neurons, consistent with a glycogen storage-associated mitochondri- fission/fusion proteins might also affect mitochondrial integrity, opathy [56]. Similarly, altered mitochondrial bioenergetics has been membrane potential and oxidative metabolism. In this regard, it has reported in the liver of heterozygous Gbe1þ/ mice [57]. Further studies been observed that Opa1 is involved in maintaining the shape of are needed to corroborate the implication of muscle mitochondrial mitochondrial cristae by stabilizing respiratory chain super-complexes impairment in the pathophysiology of other GSDsespecially those, like [41e43], while loss of Mfn2 function causes metabolic alterations in McArdle disease, associated with exercise intolerance and low aerobic mitochondria (i.e., lower membrane potential and cellular oxygen capacity. On the other hand, the present findings should be kept in mind consumption, as well as depressed substrate oxidation) with the when assessing therapeutic approaches for McArdle disease and other absence of both mitofusins Opa1 and Mfn2 leading to severe mtDNA GSDs affecting the skeletal muscle. Indeed, full restoration of muscle depletion [33]. Consistent with this, we observed an impaired mito- oxidative capacity might not be always possible with drug or exercise chondrial content and biogenesis as well as a decrease in OXPHOS intervention in the context of severe mitochondrial network disruption complex proteins and activities in the skeletal muscle of McArdle mice. inside this tissue. Additionally, we report the presence of c-shaped or “donut” like mitochondria, a phenomenon that has been previously associated with 5. CONCLUSIONS oxidative stress and/or impaired energy production [29,44]; in this regard, the observed increased levels of the Nrf2 protein in the skeletal A hallmark of McArdle disease, severe impairment in muscle oxidative muscle of McArdle mice might represent a compensatory adaptation capacity as reflected by very low VO2peak values in affected individuals, aiming at reducing the oxidative stress caused by mitochondrial could be explained not only by a reduced glycolytic flux due to the metabolism impairment inherited metabolic blockade in muscle glycogen breakdown, but also Mitochondrial network disruption might also affect the interaction of possibly by a disruption of the mitochondrial network in skeletal mitochondria with other organelles such as SR, lysosomes or lipid muscle fibers due to abnormally large glycogen depots, at least in droplets. In this regard, mitochondrial fission 1 protein (commonly those fibers with a higher capacity for accumulating this substrate. abbreviated as ‘FIS1’), which is known to be involved in mitochondria/ Pending corroboration studies in more patients with McArdle disease, SR interaction [45], showed an abnormal subcellular localization in the the present findings might pave the way for future research addressing skeletal muscle of McAdle mice. This, together with the fact that the potential involvement of mitochondrial network alterations in the mitochondria/SR contact sites might be disrupted, would suggest that pathophysiology of other GSDs. mitochondrial calcium metabolism and lipid homeostasis could also be affected. ETHICS APPROVAL Concerning the aerobic capacity of the largest series of patients with McArdle disease ever evaluated with exercise testing, we showed that All experimental procedures were approved by the Vall d’Hebron the VO2peak level of these individuals was much lower compared to Institutional Review Board (protocol number 58/17 CEEA; 35/04/08) healthy controls, which is in line with previous findings indicating and were conducted in accordance with the European Convention for impaired oxidative metabolism and/or low VO2peak (which is also an the Protection of Vertebrate Animals used for Experimental and Other indicator of peak muscle oxidative capacity) in patients with this con- Scientific Purposes (ETS1 2 3) and Spanish laws (32/2007 and R.D. dition [7,13,14,27,46e51]. Additionally, muscle biopsy analyses from 1201/2005). The study in humans adhered the ethics guidelines of the two patients suggested similar figures to those found in miceloss of Declaration of Helsinki, and was approved by the Ethics Committee of mitochondrial and cytoskeleton integrity and disruption of the mito- Hospital 12 de Octubre, Madrid, Spain (approval number 16/081). All chondrial network, especially in type II (glycogen-reliant) fibersthereby participants were informed about the study procedures and signed a providing preliminary support to the notion that mitochondrial network written informed consent. disruption underpins, at least partly, the impairment in muscle oxidative capacity that characterizes this disease. More research in larger patient CONSENT FOR PUBLICATION cohorts is needed to corroborate our pilot results in only two individuals. On the other hand, the role of mitochondria in the pathophysiology of Not applicable. GSDs other than McArdle disease has not been extensively studied. In GSD Ia (deficiency of the catalytic activity of glucose-6-phosphatase, AVAILABILITY OF DATA AND MATERIAL with subsequent intracellular accumulation of glycogen in liver, kidney and small intestines), a decreased mitochondrial content and membrane The datasets generated during and/or analysed during the current potential, deranged mitochondria ultrastructure, impaired oxidative study are available from the corresponding author on reasonable phosphorylation and changes in Krebs cycle metabolites have been request. described in cell cultures as well as in the mouse model for this con- dition [52], with some authors also reporting a decrease in mitochondrial FUNDING number [53] or the presence of swollen and deformed mitochondria [54]. In the case of GSD II (Pompe disease), multiple mitochondrial The present study was funded by grants received from the Fondo de defects have been described in mouse models and patients, including a Investigaciones Sanitarias (FIS, PI17/02052, PI18/00139, PI19/01313, MOLECULAR METABOLISM 66 (2022) 101648 Ó 2022 The Author(s). Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 13 www.molecularmetabolism.com Original Article and PI20/00645) and cofunded by ‘Fondos FEDER’. Gisela Nogales- [8] Villarreal-Salazar M, Brull A, Nogales-Gadea G, Andreu AL, Martin MA, Gadea and Carmen Fiuza-Luces are supported by the Miguel Servet Arenas J, et al. Preclinical research in McArdle disease: a Review of research research contracts (ISCIII CD14/00032 and CP18/00034, respectively models and therapeutic strategies. Genes 2021;13(1). and cofounded by Fondos FEDER0 ). Research by Pedro L. Valenzuela is [9] Fiuza-Luces C, Santos-Lozano A, Llavero F, Campo R, Nogales-Gadea G, Diez- funded by a postdoctoral contract granted by Instituto de Salud Carlos Bermejo J, et al. Muscle molecular adaptations to endurance exercise training III (Sara Borrell, CD21/00138). Monica Villarreal Salazar is supported are conditioned by glycogen availability: a proteomics-based analysis in the by the Mexican National Council for Science and Technology McArdle mouse model. J Physiol 2018;596(6):1035e61. (CONACYT). [10] Grassi B, Porcelli S, Marzorati M, Lanfranconi F, Vago P, Marconi C, et al. Metabolic myopathies: functional evaluation by analysis of oxygen uptake AUTHORS CONTRIBUTIONS kinetics. Med Sci Sports Exerc 2009;41(12):2120e7. [11] Siciliano G, Rossi B, Martini A, Angelini C, Martinuzzi A, Lodi R, et al. Myo- MVS, ARM performed all the mice experiments, AS, AL, and CFL phosphorylase deficiency affects muscle mitochondrial respiration as shown performed patients’ tests, JCRA performed mass spectrometry ana- by 31P-MR spectroscopy in a case with associated multifocal encephalopathy. lyses, GNG, ALA, MAM, JA, JV and AL made major contributions in the J Neurol Sci 1995;128(1):84e91. writing and revision of the manuscript, while TOK and TP designed the [12] Lucia A, Nogales-Gadea G, Perez M, Martin MA, Andreu AL, Arenas J. McArdle experiments and TP wrote the first manuscript draft. disease: what do neurologists need to know? Nat Clin Pract Neurol 2008;4(10):568e77. DATA AVAILABILITY [13] Haller RG, Lewis SF, Cook JD, Blomqvist CG. Myophosphorylase deficiency impairs muscle oxidative metabolism. Ann Neurol 1985;17(2):196e9. Data will be made available on request. [14] De Stefano N, Argov Z, Matthews PM, Karpati G, Arnold DL. Impairment of muscle mitochondrial oxidative metabolism in McArdles’s disease. Muscle ACKNOWLEDGEMENTS Nerve 1996;19(6):764e9. [15] Fiuza-Luces C, Nogales-Gadea G, Garcia-Consuegra I, Pareja-Galeano H, Rufian- Authors would like to thank Javier Torres Torronteras, Yolanda Cámara and Ramon Vazquez L, Perez LM, et al. Muscle signaling in exercise intolerance: insights Martí Seves for their constant scientific inputs. from the McArdle mouse model. Med Sci Sports Exerc 2016;48(8):1448e58. [16] Dong F, Zhu M, Zheng F, Fu C. Mitochondrial fusion and fission are required CONFLICT OF INTERESTS for proper mitochondrial function and cell proliferation in fission yeast. FEBS J 2022;289(1):262e78. The author(s) declare no conflict of interest. [17] Gonzalez-Freire M, Scalzo P, D’Agostino J, Moore ZA, Diaz-Ruiz A, Fabbri E, et al. Skeletal muscle ex vivo mitochondrial respiration parallels decline in vivo APPENDIX A. SUPPLEMENTARY DATA oxidative capacity, cardiorespiratory fitness, and muscle strength: the Balti- more Longitudinal Study of Aging. Aging Cell 2018;17(2). Supplementary data to this article can be found online at https://doi.org/10.1016/j. [18] Weibel ER, Hoppeler H. Exercise-induced maximal metabolic rate scales with molmet.2022.101648. muscle aerobic capacity. J Exp Biol 2005;208(Pt 9):1635e44. [19] Nogales-Gadea G, Pinos T, Lucia A, Arenas J, Camara Y, Brull A, et al. Knock- REFERENCES in mice for the R50X mutation in the PYGM gene present with McArdle dis- ease. Brain 2012;135(Pt 7):2048e57. [1] Nogales-Gadea G, Pinos T, Andreu AL, Martin MA, Arenas J, Lucia A. Next- [20] Real-Martinez A, Brull A, Huerta J, Tarrasó G, Lucia A, Martin M, et al. Low generation sequencing to estimate the prevalence of a great unknown: survival rate and muscle fiber-dependent aging effects in the McArdle disease McArdle disease. Genet Med 2015;17(8):679e80. mouse model. Sci Rep 2019;9(1):5116. [2] De Castro M, Johnston J, Biesecker L. Determining the prevalence of McArdle [21] Krag TO, Pinos T, Nielsen TL, Brull A, Andreu AL, Vissing J. Differential muscle disease from gene frequency by analysis of next-generation sequencing data. involvement in mice and humans affected by McArdle disease. J Neuropathol Genet Med 2015;17(12):1002e6. Exp Neurol 2016;75(5):441e54. [3] Munguia-Izquierdo D, Santalla A, Lucia A. Cardiorespiratory fitness, physical [22] Brull A, de Luna N, Blanco-Grau A, Lucia A, Martin MA, Arenas J, et al. activity, and quality of life in patients with McArdle disease. Med Sci Sports Phenotype consequences of myophosphorylase dysfunction: insights from the Exerc 2015;47(4):799e808. McArdle mouse model. J Physiol 2015;593(12):2693e706. [4] Mate-Munoz JL, Moran M, Perez M, Chamorro-Vina C, Gomez-Gallego F, [23] Dubowitz C, Sewry CA, Oldfors A. Muscle biopsy a practical approach. 5th ed Santiago C, et al. Favorable responses to acute and chronic exercise in 2020. McArdle patients. Clin J Sport Med 2007;17(4):297e303. [24] Andreu AL, Martinez R, Marti R, Garcia-Arumi E. Quantification of mitochondrial [5] Ozemek C, Laddu DR, Lavie CJ, Claeys H, Kaminsky LA, Ross R, et al. An DNA copy number: pre-analytical factors. Mitochondrion 2009;9(4):242e6. update on the role of cardiorespiratory fitness, structured exercise and lifestyle [25] Nogales-Gadea G, Rubio JC, Fernandez-Cadenas I, Garcia-Consuegra I, physical activity in preventing cardiovascular disease and health risk. Prog Lucia A, Cabello A, et al. Expression of the muscle glycogen phosphorylase Cardiovasc Dis 2018;61(5e6):484e90. gene in patients with McArdle disease: the role of nonsense-mediated mRNA [6] Ross R, Blair SN, Arena R, Church TS, Despres JP, Franklin BA, et al. decay. Hum Mutat 2008;29(2):277e83. Importance of assessing cardiorespiratory fitness in clinical practice: a case for [26] Medja F, Allouche S, Frachon P, Jardel C, Malgat M, Mousson de Camaret B, fitness as a clinical vital sign: a scientific statement from the American Heart et al. Development and implementation of standardized respiratory chain association. Circulation 2016;134(24):e653e99. spectrophotometric assays for clinical diagnosis. Mitochondrion 2009;9(5): [7] Scalco RS, Lucia A, Santalla A, Martinuzzi A, Vavla M, Reni G, et al. Data from 331e9. the European registry for patients with McArdle disease and other muscle [27] Santalla A, Nogales-Gadea G, Encinar AB, Vieitez I, Gonzalez-Quintana A, glycogenoses (EUROMAC). Orphanet J Rare Dis 2020;15(1):330. Serrano-Lorenzo P, et al. Genotypic and phenotypic features of all Spanish 14 MOLECULAR METABOLISM 66 (2022) 101648 Ó 2022 The Author(s). Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). www.molecularmetabolism.com patients with McArdle disease: a 2016 update. BMC Genom 2017;18(Suppl 8): [42] Frezza C, Cipolat S, Martins de Brito O, Micaroni M, Beznoussenko GV, 819. Rudka T, et al. OPA1 controls apoptotic cristae remodeling independently from [28] Glancy B, Kim Y, Katti P, Willingham TB. The functional impact of mitochondrial mitochondrial fusion. Cell 2006;126(1):177e89. structure across subcellular scales. Front Physiol 2020;11:541040. [43] Lee H, Smith SB, Yoon Y. The short variant of the mitochondrial dynamin OPA1 [29] Ding WX, Li M, Biazik JM, Morgan DG, Guo F, Ni HM, et al. Electron micro- maintains mitochondrial energetics and cristae structure. J Biol Chem scopic analysis of a spherical mitochondrial structure. J Biol Chem 2017;292(17):7115e30. 2012;287(50):42373e8. [44] Stahon KE, Bastian C, Griffith S, Kidd GJ, Brunet S, Baltan S. Age-related [30] Mado K, Chekulayev V, Shevchuk I, Puurand M, Tepp K, Kaambre T. On changes in axonal and mitochondrial ultrastructure and function in white the role of tubulin, plectin, desmin, and vimentin in the regulation of matter. J Neurosci 2016;36(39):9990e10001. mitochondrial energy fluxes in muscle cells. Am J Physiol Cell Physiol [45] Iwasawa R, Mahul-Mellier AL, Datler C, Pazarentzos E, Grimm S. Fis1 and 2019;316(5):C657e67. Bap31 bridge the mitochondria-ER interface to establish a platform for [31] Moore AS, Holzbaur ELF. Mitochondrial-cytoskeletal interactions: dynamic apoptosis induction. EMBO J 2011;30(3):556e68. associations that facilitate network function and remodeling. Curr Opin Physiol [46] Pizzamiglio C, Mahroo OA, Khan KN, Patasin M, Quinlivan R. Phenotype and 2018;3:94e100. genotype of 197 British patients with McArdle disease: an observational [32] Chen H, Chan DC. Mitochondrial dynamics–fusion, fission, movement, and single-centre study. J Inherit Metab Dis 2021;44(6):1409e18. mitophagy–in neurodegenerative diseases. Hum Mol Genet 2009;18(R2): [47] Lewis SF, Haller RG. 1986. The pathophysiology of McArdle’s disease: clues to R169e76. regulation in exercise and fatigue. J Appl Physiol 1985;61(2):391e401. [33] Chen H, Vermulst M, Wang YE, Chomyn A, Prolla TA, McCaffery JM, et al. [48] Hagberg JM, Coyle EF, Carroll JE, Miller JM, Martin WH, Brooke MH. Exercise Mitochondrial fusion is required for mtDNA stability in skeletal muscle and hyperventilation in patients with McArdle’s disease. J Appl Physiol Respir tolerance of mtDNA mutations. Cell 2010;141(2):280e9. Environ Exerc Physiol 1982;52(4):991e4. [34] Holmstrom KM, Kostov RV, Dinkova-Kostova AT. The multifaceted role of Nrf2 [49] Carroll JE, Hagberg JM, Brooke MH, Shumate JB. Bicycle ergometry and gas in mitochondrial function. Curr Opin Toxicol 2016;1:80e91. exchange measurements in neuromuscular diseases. Arch Neurol 1979;36(8): [35] Madsen KL, Laforet P, Buch AE, Stemmerik MG, Ottolenghi C, Hatem SN, et al. 457e61. No effect of triheptanoin on exercise performance in McArdle disease. Ann Clin [50] Bank W, Chance B. An oxidative defect in metabolic myopathies: diagnosis by Transl Neurol 2019;6(10):1949e60. noninvasive tissue oximetry. Ann Neurol 1994;36(6):830e7. [36] Santalla A, Valenzuela PL, Rodriguez-Lopez C, Rodriguez-Gomez I, Nogales- [51] Andersen KL, Lund-Johansen P, Clausen G. Metabolic and circulatory re- Gadea G, Pinos T, et al. Long-term exercise intervention in patients with sponses to muscular exercise in a subject with glycogen storage disease McArdle disease: clinical and aerobic fitness benefits. Med Sci Sports Exerc (McArdle’s Disease). Scand J Clin Lab Invest 1969;24(2):105e13. 2022;54(8):1231e41. [52] Farah BL, Sinha RA, Wu Y, Singh BK, Lim A, Hirayama M, et al. Hepatic [37] Varikmaa M, Bagur R, Kaambre T, Grichine A, Timohhina N, Tepp K, et al. Role mitochondrial dysfunction is a feature of glycogen storage disease type Ia of mitochondria-cytoskeleton interactions in respiration regulation and mito- (GSDIa). Sci Rep 2017;7:44408. chondrial organization in striated muscles. Biochim Biophys Acta [53] Riede UN, Spycher MA, Gitzelmann R. Glycogenosis type I (glucose 6- 2014;1837(2):232e45. phosphatase deficiency): I. Ultrastructural morphometric analysis of juvenile [38] Tepp K, Mado K, Varikmaa M, Klepinin A, Timohhina N, Shevchuk I, et al. The liver cells. Pathol Res Pract 1980;167(1):136e50. role of tubulin in the mitochondrial metabolism and arrangement in muscle [54] Cho JH, Kim GY, Pan CJ, Anduaga J, Choi EJ, Mansfield BC, et al. Down- cells. J Bioenerg Biomembr 2014;46(5):421e34. regulation of SIRT1 signaling underlies hepatic autophagy impairment in [39] Kuznetsov AV, Javadov S, Guzun R, Grimm M, Saks V. Cytoskeleton and glycogen storage disease type Ia. PLoS Genet 2017;13(5):e1006819. regulation of mitochondrial function: the role of beta-tubulin II. Front Physiol [55] Lim JA, Li L, Kakhlon O, Myerowitz R, Raben N. Defects in calcium homeo- 2013;4:82. stasis and mitochondria can be reversed in Pompe disease. Autophagy [40] Carre M, Andre N, Carles G, Borghi H, Brichese L, Briand C, et al. 2015;11(2):385e402. Tubulin is an inherent component of mitochondrial membranes that in- [56] Wong HH, Seet SH, Maier M, Gurel A, Traspas RM, Lee C, et al. Loss of teracts with the voltage-dependent anion channel. J Biol Chem C2orf69 defines a fatal autoinflammatory syndrome in humans and zebrafish 2002;277(37):33664e9. that evokes a glycogen-storage-associated mitochondriopathy. Am J Hum [41] Quintana-Cabrera R, Quirin C, Glytsou C, Corrado M, Urbani A, Pellattiero A, Genet 2021;108(7):1301e17. et al. The cristae modulator Optic atrophy 1 requires mitochondrial ATP [57] Malinska D, Testoni G, Bejtka M, Duran J, Guinovart JJ, Duszynski J. Alteration synthase oligomers to safeguard mitochondrial function. Nat Commun of mitochondrial function in the livers of mice with glycogen branching enzyme 2018;9(1):3399. deficiency. Biochimie 2021;186:28e32. MOLECULAR METABOLISM 66 (2022) 101648 Ó 2022 The Author(s). Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 15 www.molecularmetabolism.com