J Inherit Metab Dis DOI 10.1007/s10545-014-9771-y
GLYCOGENOSES
Exercise in muscle glycogen storage diseases Nicolai Preisler & Ronald G Haller & John Vissing
Received: 7 August 2014 / Accepted: 9 September 2014 # SSIEM 2014
Abstract Glycogen storage diseases (GSD) are inborn errors of glycogen or glucose metabolism. In the GSDs that affect muscle, the consequence of a block in skeletal muscle glycogen breakdown or glucose use, is an impairment of muscular performance and exercise intolerance, owing to 1) an increase in glycogen storage that disrupts contractile function and/or 2) a reduced substrate turnover below the block, which inhibits skeletal muscle ATP production. Immobility is associated with metabolic alterations in muscle leading to an increased dependence on glycogen use and a reduced capacity for fatty acid oxidation. Such changes may be detrimental for persons with GSD from a metabolic perspective. However, exercise may alter skeletal muscle substrate metabolism in ways that are beneficial for patients with GSD, such as improving exercise
Communicated by Georg Hoffmann Presented at the GSD Conference in Heidelberg Germany, November 28– 30, 2013 N. Preisler (*) : J. Vissing Neuromuscular Research Unit, Section 3342, Department of Neurology, Rigshospitalet, University of Copenhagen, Blegdamsvej 9, 2100 Copenhagen, Denmark e-mail: [email protected] J. Vissing e-mail: [email protected] R. G. Haller Department of Neurology, The University of Texas Southwestern Medical Center, Dallas, TX, USA e-mail: [email protected]
tolerance and increasing fatty acid oxidation. In addition, a regular exercise program has the potential to improve general health and fitness and improve quality of life, if executed properly. In this review, we describe skeletal muscle substrate use during exercise in GSDs, and how blocks in metabolic pathways affect exercise tolerance in GSDs. We review the studies that have examined the effect of regular exercise training in different types of GSD. Finally, we consider how oral substrate supplementation can improve exercise tolerance and we discuss the precautions that apply to persons with GSD that engage in exercise. Abbreviations GSD Glycogen storage disease ATP Adenosine triphosphate RPE Rate of perceived exertion PHK Phosphorylase kinase CK Creatine kinase GDE Glycogen debranching enzyme PGM Phosphoglucomutase CDG Congenital disorder of glycosylation PFK Phosphofructokinase NEFA Non esterified fatty acid PGAM Phosphoglycerate mutase GAA α-1,4-glucosidase VO2peak Peak oxygen consumption MET Metabolic equivalents TCA-cycle Tricarboxylic acid cycle ERT Enzyme replacement therapy
R. G. Haller Department of Neurology, The Veterans Affairs Medical Center, Dallas, TX, USA
Introduction
R. G. Haller The Neuromuscular Center, Institute for Exercise and Environmental Medicine of Presbyterian Hospital, Dallas, TX, USA
Glycogen storage diseases (GSD) or glycogenoses are inborn errors of metabolism caused by mutations in the genes coding
J Inherit Metab Dis
for enzymes involved in glycogen or glucose metabolism (DiMauro et al 2004; Laforet et al 2012). GSDs have a broad clinical spectrum and can affect many tissues. Typically though, skeletal muscle and/or liver involvement dominate the clinical presentation, owing to the significance of glycogen– and glucose metabolism in these organs (Laforet et al 2012). In the GSDs that affect muscle, the consequence of a block in skeletal muscle glycogen breakdown (glycogenolysis), or in the anaerobic catabolism of glucose-6-phosphate (glycolysis), is an impairment of muscular performance, owing to 1) an increase in glycogen storage that disrupts contractile function and/or 2) a reduced substrate turnover below the block, which inhibits skeletal muscle ATP production (DiMauro et al 2004; Laforet et al 2012). Exercise, as a treatment for neuromuscular diseases, has received growing attention in the last decades, as new evidence has revoked the older notion that muscle contraction may accelerate the disease process, and because no curative treatment for these conditions are in immediate sight. In the GSDs, exercise as a treatment may seem counter intuitive for many patients and also clinicians, because it is exercise that provoke symptoms in many GSDs, and some types of exercise are not well tolerated by the patients (DiMauro et al 2004). However, it is important to deal with exercise in GSDs for several reasons. First, exercise intolerance is severe in most GSDs, which partly is caused by a sedentary lifestyle in persons with muscle disease. Immobility is associated, not only with skeletal muscle atrophy, but also with metabolic alterations in muscle leading to an increased dependence on glycogen use and a reduced capacity for fatty acid oxidation (Stein and Wade 2005). Such changes may be detrimental for persons with GSD from a metabolic perspective, and these changes can potentially worsen the muscle phenotype leading to more disuse; initiating a “vicious cycle of disuse”. Secondly, exercise is the most powerful natural inducer of metabolic change in skeletal muscle in humans. Theoretically, exercise may alter skeletal muscle substrate metabolism in ways that are specifically beneficial for patients with GSD, such as improving exercise tolerance and increasing the ability to oxidize fat. In addition, like in healthy subjects, a regular exercise program has the potential to improve general health and fitness and improve quality of life, if executed properly (Blair et al 1995; Pedersen and Saltin 2006). However, there are limitations and precautions that apply to persons with GSD that engage in regular exercise training programs and this has to be considered when prescribing exercise to this patient population. In this review we will use the term “exercise” synonymously with dynamic exercise, i.e., involvement of a large muscle mass performing aerobic activities such as walking, running or cycling, unless otherwise stated. Almost all exercise studies and studies of skeletal muscle metabolism in humans have been conducted in adults, and hence our recommendations are
based on what we know about skeletal muscle substrate use and exercise in adults with GSD. In this review, we describe skeletal muscle substrate use during exercise in GSDs, and how blocks in metabolic pathways affect exercise tolerance in GSDs. We review the studies that have examined the effect of regular exercise training in different types of GSD. Finally, we consider how oral substrate supplementation can improve exercise tolerance and we discuss the precautions that apply to persons with GSD that engage in exercise. We limit the review to GSDs that have been examined with exercise interventions, and we begin our review with a brief overview of these disorders.
Glycogen storage diseases affecting muscle In Table 1, we present the GSDs covered by this review. We have not discussed the following GSDs affecting muscle, because they have not yet been studied with exercise intervention; GSD type IV (branching enzyme deficiency), phosphoglycerate kinase deficiency, GSD type XI (muscle lactate dehydrogenase deficiency), GSD type XII (aldolase A deficiency), GSD type XIII (beta enolase deficiency) and GSD type 0 (glycogen synthase deficiency) and GSD type XV (glycogenin deficiency). In the following paragraphs we will briefly present each GSD, in which exercise interventions have been performed, and comment on clinical and metabolic findings that are relevant for prescription of exercise programs. The epidemiology, clinical presentation, disease course and management of the GSDs have been reviewed in detail elsewhere (Baethmann et al 2008; DiMauro et al 2004; Kishnani et al 2010; Laforet et al 2012; Lucia et al 2008; Nakajima et al 2002; Vissing and Orngreen 2014). The enzymatic steps involved in the GSDs reviewed in this article are presented in Fig. 1. General phenotypes, static vs. dynamic symptoms (Fig. 2) GSDs that affect muscle can present with different clinical phenotypes, and the response and tolerance to exercise can also differ, sometimes even within the same type of GSD (Echaniz-Laguna et al 2010; Orngreen et al 2008; Preisler et al 2012b). Despite the variability in phenotype, muscle GSDs can generally be divided into two groups; a group with primarily exercise-related symptoms and a group with fixed symptoms due to muscle atrophy and weakness (Fig. 2) (DiMauro and Lamperti 2001). Dynamic exerciseinduced symptoms signify an “acute energy crisis” within skeletal muscle, caused by a mismatch in the supply and demand for ATP. Even though the classification of muscular symptoms into these two classes is practical from a clinical point of view, it is important to note that phenotypes may overlap (Fig. 2).
Muscle phosphoglycerate mutase PGAM2 (HGNC:8889) No (EC 5.4.2.11) Phosphoglucomutase 1 (EC 5.4.2.2)
GSD type XIV† Phosphoglucomutase 1 deficiency [#614921]
Yes
No
No
No
No
No
No
No
Yes
Yes
No
No
No
No
Yes
No
0–12 %
2–7 %
0–29 %
0–8 %
0%
0–30 %
1–30 %
Yes
Yes
Yes
Yes
Yes
Rare
No
(Tegtmeyer et al 2014)
(Naini et al 2009; Oh et al 2006)
(Orngreen et al 2008; Wuyts et al 2005)
(Nakajima et al 2002)
(Lucia et al 2008)
(Ding et al 1990; Moses et al 1986)
(Bembi et al 2008; Straub 2008)
This table provides an overview of selected clinical and biochemical findings in glycogen storage diseases that have been studied with exercise tests. The data concerns adults and the response to exercise. #OMIM online mendelian inheritance in man® number, GSD glycogen storage disease, EC # enzyme commission number, HGNC ID HUGO gene nomenclature committee ID. * Cardiac hypertrophy is frequent in GSD type IIIa, but frank cardiomyopathy is rare in this GSD. Significant cardiac involvement is rare in GSD type II, but arrhythmias are probably more frequent in this GSD than in the general population (Sacconi et al 2014). §=Skeletal muscle or fibroblast enzyme activity levels (approximate levels, typical ranges). Methods to measure enzyme activity differ among studies. ‡=Whether or not exercise-induced rhabdomyolysis or myoglobinuria is a feature of the GSD. There are a few reports of rhabdomyolysis in GSD type IIIa, but it is rare compared to the other GSDs associated with rhabdomyolysis. †=GSD type XIV is also a congenital disorder of glycosylation of mixed type
PGM1 (HGNC:8905)
PHKA1 (HGNC:8925)
GSD type X (Phosphoglycerate mutase deficiency) [#261670]
GSD type IXd (Muscle phosphorylase kinase deficiency) [#300559]
No No
PYGM (HGNC:9726) PFKM (HGNC:8877)
Muscle Phosphofructokinase (EC 2.7.1.11) Phosphorylase kinase (EC 2.7.11.19)
GSD type VII (Tauri disease) [#232800]
Yes
Yes
Cardiac* Respiratory Liver Residual Rhabdo References§ involvement involvement affection enzyme activity§ myolysis‡
AGL (HGNC:321)
GAA (HGNC:4065)
Official gene symbol (HGNC ID)
Amylo-alpha-1,6-glucosidase & 4-alpha-glucanotransferase (EC 3.2.1.33 & EC 2.4.1.25) GSD type V (McArdle disease) [#232600] Myophosphorylase (EC 2.4.1.1)
Alpha-1,4-glucosidase (EC 3.2.1.20)
GSD type II (Pompe disease) [#232300]
GSD type IIIa (Cori or Forbes disease) [#232400]
Enzyme deficiency (EC #)
Glycogen storage disease (name) [#OMIM]
Table 1 Reviewed glycogen storage diseases affecting the glycogenolytic and glycolytic pathways in skeletal muscle
J Inherit Metab Dis
J Inherit Metab Dis
Fig. 1 Simplified overview of skeletal muscle glycogen and glucose use. The figure provides a simplified overview of the glycogenolytic and glycolytic pathways, the adenylate kinase reaction and the release and uptake of blood borne substrates (within red lines). Enzymes or allosteric effectors are in italics. Enzymes in bold are enzymes deficient in the glycogen storage disease that are covered in this review. Dashed arrows indicate that chemical reactions have been left out, to simplify the figure. PHK phosphorylase kinase, Myophos myophosphorylase, GDE glycogen debranching enzyme, PGM phosphoglucomutase, PFK
phosphofructokinase, PGAM phosphoglycerate mutase, GAA α-1,4-glucosidase, HEX hexokinase (E.C. 2.7.1.1), PGI glucose-6-phosphate isomerase (E.C. 5.3.1.9). LDH lactate dehydrogenase (E.C. 1.1.1.27), ADK adenylate kinase (E.C. 2.7.4.3), AMPD adenosine monophosphate deaminase (E.C. 3.5.4.6), Pi inorganic phosphate, AMP adenosine monophosphate, PKA protein kinase A (E.C. 2.7.11.11), NEFA nonesterified fatty acids, ADP adenosine diphosphate, ATP adenosine triphosphate, IMP inosine monophosphate, P phosphate (Cory 2006; Harris 2006; Smith 2006)
Disorders affecting skeletal muscle glycogen breakdown
oxidation in contracting muscles (Haller and Vissing 2002; Orngreen et al 2009). Glucose is helpful, because it can be metabolized below the metabolic block in this GSD (Fig. 1) (Andersen et al 2008; Haller and Vissing 2002; Vissing and Haller 2003).
Myophosphorylase deficiency, McArdle disease Myophosphorylase deficiency was first described in 1951, and this GSD may be considered the prototype of GSDs with exercise-related symptoms. Myophosphorylase releases glucose-1-phosphate from glycogen (Fig. 1). Almost all mutations in the PYGM gene are null mutations, and the consequence of this is a complete block in skeletal muscle glycogenolysis. This leads to severe exercise intolerance with dynamic, exercise-induced symptoms caused by an acute energy crisis within skeletal muscle. Patients with myophosphorylase deficiency develop a so called second-wind phenomenon during aerobic exercise of moderate intensity (Pearson et al 1961). During the first minutes of exercise there is a gradual and continuous increase in heart rate and rate of perceived exertion (RPE) that is out of proportion to the actual work rate. Heart rate and RPE peak after 6–8 minutes of exercise, at which time heart rate and RPE correspond to values observed at maximal rates of exercise. In the following minutes, heart rate and RPE decline (the second-wind), so that exercise at the same work load becomes much better tolerated (Haller and Vissing 2002). The second-wind is attributable to enhanced uptake of blood-born glucose from the liver and enhanced fat
Muscle phosphorylase kinase deficiency Muscle phosphorylase kinase (PHK) initiates skeletal muscle glycogen breakdown by phosphorylating and activating myophosphorylase (Fig. 1) and therefore it may be anticipated that patients with muscle PHK deficiency present with dynamic, exercise-induced symptoms, similar to patients with myophosphorylase deficiency. This is true in some cases of this rare disease, but generally the dynamic exercise-induced symptoms are milder (Laforet et al 2012). In some cases, persons with muscle PHK deficiency may be almost asymptomatic (Echaniz-Laguna et al 2010; Preisler et al 2012b). Two patients were only diagnosed because of elevated plasma creatine kinase (CK) levels, and three of the less than ten cases reported in the literature never experienced exercise intolerance. Therefore it is likely that some cases go undiagnosed (Echaniz-Laguna et al 2010; Orngreen et al 2008; Preisler et al 2012b).
J Inherit Metab Dis
one third of the patients develop muscle weakness and wasting in their 3rd to 4th decade (Kishnani et al 2010; Moses et al 1986). It is interesting that dynamic exerciseinduced symptoms are not at the forefront of symptoms in this GSD, considering that the GDE works in concert with myophosphorylase to degrade glycogen (Fig. 1) (Harris 2006; Kishnani et al 2010). We have recently documented that young patients with GDE deficiency and without overt muscle weakness, do in fact have significant exercise intolerance to moderate-intensity exercise, suggesting that this metabolic myopathy should also be classified amongst the GSDs that present with dynamic exercise-induced symptoms (Fig. 2) (Kishnani et al 2010; Preisler et al 2013b). In spite of exercise intolerance, patients with GDE deficiency, like patients with muscle PHK deficiency, tolerate exercise much better than patients with myophosphorylase deficiency and myoglobinuria and rhabdomyolysis are very rare in this GSD (Table 1) (DiMauro et al 2004; Kishnani et al 2010). This could relate to the small amount of glucose, which can be released from muscle glycogen at the onset of exercise, until the GDE reaches a branching site on the glycogen molecule. Fig. 2 Static or dynamic skeletal muscle affection in glycogen storage diseases. The figure provides an overview of the typical skeletal muscle presentation (framed boxes) in glycogen storage diseases (GSD). Static muscular symptoms are used to describe muscle affection in GSDs that present with progressive skeletal muscle weakness and wasting, i.e., fixed weakness. Dynamic exercise-induced muscular symptoms are provoked by exercise or other strenuous activities that increase muscular energy requirements, and these symptoms resolve with rest when energy demands diminish. In a proportion of the patients with a GSD that present with dynamic exercise-induced symptoms, a static skeletal muscle phenotype develops later in life (arrows). # Patients with muscle PHK deficiency may be almost asymptomatic. § We suggest that GDE deficiency is classified as a GSD presenting with dynamic exercise-induced symptoms. * In phosphoglucomutase deficiency, approximately half of the cases are exercise intolerant, and a few cases present with weakness, but currently the precise classification of this GSD is unclear. PHK phosphorylase kinase, Myophos myophosphorylase, GDE glycogen debranching enzyme, PGM phosphoglucomutase, PFK phosphofructokinase, PGAM phosphoglycerate mutase, GAA α-1,4-glucosidase (Echaniz-Laguna et al 2010; Kishnani et al 2010; Lucia et al 2008; Nadaj-Pakleza et al 2009; Nakajima et al 2002; Preisler et al 2012b, 2013b; Tegtmeyer et al 2014)
Glycogen debranching enzyme deficiency, Cori or Forbes disease Glycogen debranching enzyme (GDE) deficiency is traditionally considered to be a liver glycogenosis, despite the fact that the GDE is also deficient in skeletal muscle in ~85 % of cases (GDE type a) (Coleman et al 1992; van Hoof and Hers 1967). Skeletal muscle involvement is mild or absent in children and young adults, but approximately
Phosphoglucomutase type 1 deficiency The phosphoglucomutase (PGM) enzyme is not directly involved in the release of glucose residues from the glycogen molecule, however, this enzyme catalyzes the last step of glycogenolysis, converting glucose-1-phosphate to glucose-6-phosphate, which may then enter the glycolytic pathway (Fig. 1) (Harris 2006). The first case of genetically verified PGM type 1 deficiency in skeletal muscle was reported in 2009, and this novel GSDs was added as the 14th (GSDXIV) of the GSDs (Stojkovic et al 2009). In 2014, however, another chapter was added to this disease entity. PGM deficiency has now been documented to not only affect skeletal muscle glycogen breakdown, but also to be the cause of defects in protein glycosylation (Tegtmeyer et al 2014). This is because glucose-1phosphate is an intermediate in protein glycosylation, and the PGM1 isoform is expressed as the predominant isoform in most tissues (McAlpine et al 1970). Protein Nglycosylation is important in cell signaling and multiple organ affection is very common in this novel disorder of mixed-type congenital disorder of glycosylation (CDG) and GSD, with a very broad phenotype (Tegtmeyer et al 2014). Half of the patients report exercise intolerance, and in a few patients, the myopathy dominates the phenotype. The exact mechanism behind the broad disease spectrum and predominant muscle affection in some cases of this GSD/ CDG is not clear at present (Stojkovic et al 2009; Tegtmeyer et al 2014).
J Inherit Metab Dis
Disorders of skeletal muscle glycolysis Muscle phosphofructokinase deficiency, Tauri disease Muscle phosphofructokinase (PFK) deficiency is a disorder of glycolysis that was first described in 1965 (Nakajima et al 2002). The phenotype and response to exercise in patients with muscle PFK deficiency is very similar to myophosphorylase deficiency, albeit these patients are even more intolerant to exercise because of a complete block in glycolysis, limiting skeletal muscle substrate use to almost exclusive dependence on non-esterified fatty acid (NEFA) oxidation (Haller and Lewis 1991; Ono et al 1995; Vissing et al 1996). In contrast to myophosphorylase deficiency, patients with muscle PFK deficiency lack a second-wind. In fact, glucose triggers an out-ofwind phenomenon in this GSD, most likely induced by suppression of plasma fatty acid concentrations by the action of insulin, as a consequence of an increase in blood glucose levels (Haller and Lewis 1991). In addition to skeletal muscle involvement, most patients with muscle PFK deficiency also have a compensated hemolysis, due to co-expression of the muscle isoform of PFK in erythrocytes (Nakajima et al 2002). Phosphoglycerate mutase deficiency The first case of muscle phosphoglycerate mutase (PGAM) deficiency was described in 1981 (DiMauro et al 1981). PGAM converts 3-phosphoglycerate to 2-phosphoglycerate in one of the final steps in the glycolytic pathway (Fig. 2) (Harris 2006). There are two isoforms of the PGAM enzyme; the brain and the muscle form, which are expressed in a tissue specific manner (Sakoda et al 1988; Shanske et al 1987). Patients with PGAM deficiency are much more mildly affected than patients with muscle PFK deficiency and are usually asymptomatic, unless they engage in high-intensity exercise, which may cause contractures, myalgia and myoglobinuria (Naini et al 2009; Vissing et al 2005). The mild phenotype is explained by the expression of the brain isoform of the PGAM enzyme in muscle, which is responsible for the residual enzyme activity measured in muscle in this GSD (Table 1) (Oh et al 2006). A peculiar finding in this GSD is tubular aggregates in muscle biopsies, which are present in around one third of cases (Naini et al 2009). Therefore, PGAM deficiency should be considered in patients with dynamic exercise-induced symptoms and tubular aggregates on biopsy. Especially, if other more common GSDs have been ruled out (Naini et al 2009). Disorders affecting lysosomal glycogenolysis Pompe disease, α-1,4-glucosidase deficiency Pompe disease was described in 1932, and this glycogen storage disease is distinct from other GSDs because the
enzyme defect is within lysosomes where it blocks glycogenolysis (Hers 1963; van der Ploeg and Reuser 2008). The disease is caused by deficiency of α-1,4-glucosidase (GAA) and patients with the late-onset form of GAA deficiency (~70 % of cases), typically present in the 2nd to 3rd decade with progressive muscle weakness and wasting primarily affecting the proximal and axial muscles (Bembi et al 2008). Exercise intolerance is pronounced in many patients with GAA deficiency, but the cause of exercise intolerance in this GSD is probably not related to a supply–demand mismatch of ATP, because lysosomal glycogen breakdown contributes minimally to the ATP production of the myocyte (Preisler et al 2012a). The functional limitation is more likely the consequence of skeletal muscle weakness and wasting and therefore the response to exercise is different from the other GSDs affecting muscle (Preisler et al 2012a).
Skeletal muscle substrate metabolism during exercise in GSD; the metabolic response to single bouts of exercise If the metabolic block causing GSD results in a significant impairment in glycolytic flux, the magnitude of the mismatch in the supply–demand chain of ATP becomes apparent only during exercise. Results from exercise studies show that in most GSDs there is clear association between exercise intolerance, exercise-induced dynamic symptoms, the effect of substrate supplementation therapy and the extent of the derangements in skeletal muscle energy production during exercise (Preisler et al 2012a, b, 2013a, b). Healthy skeletal muscle uses a variety of substrates to generate the ATP that fuels muscular contraction (McArdle et al 2007a). However, the supply and rate of production of ATP from the various sources of fuel differ, and skeletal muscle fuel selection during exercise is intensity- and supply-dependent (van Loon et al 2001). Figure 3 provides an overview of the main substrates metabolized during exercise, the limit of supply for each energy source, and how fast ATP may be produced from the various sources. Table 2 provides an overview of the metabolic and physiological response to selected types of exercises that have been used to study GSDs. Forearm exercise — anaerobic ATP production The forearm exercise test specifically challenges skeletal muscle anaerobic energy production (Haller and Vissing 2004a). During anaerobic conditions, lactate is generated from pyruvate and plasma lactate concentration is a good indicator of the flux through the glycolytic pathway. Plasma ammonia concentrations usually increase slightly after forearm exercise in healthy persons, because the adenylate kinase reaction is used
J Inherit Metab Dis
Fig. 3 Energy yield and supply and time to peak power output from energy sources used by skeletal muscle during exercise. This figure provides an overview of the major substrates used to generate the ATP that fuels muscle contraction during exercise. * Chemical reactions have been simplified for ease of view. § Exercise duration describes the time it takes to exhaust the specific fuel during exercise. Fuels differ in availability and how rapidly ATP can be produced from the different energy sources. † Anaerobic metabolism generates power outputs that are much higher than aerobic pathways and anaerobic metabolism is always engaged to supply ATP during maximal effort. ‡ With respect to oxidative metabolism, carbohydrate oxidation generates a substantially higher power output than fat oxidation, which explains the shift to carbohydrate oxidation as aerobic exercise intensity increases. Therefore, different
types of exercise, performed at different intensities, affect muscle fuel selection, i.e., the proportion of anaerobic or aerobic pathways used to fuel exercise. The figure shows that glycogen metabolism is essential for optimal muscular performance in any activity lasting more than a few seconds, which explains why climbing a flight of stairs may provoke symptoms in persons with glycogen storage disease (GSD). Fat represents an almost unlimited source of fuel; however, exclusive oxidation of non-esterified fatty acids (NEFA) can only maintain lower intensities of exercise ( 10 min of exercise. This type of exercise is dependent on mainly the aerobic combustion of glucose and NEFAs, to drive skeletal muscle contraction (Fig. 3). Measurements of substrate turnover during exercise in GSDs have shown increased oxidation rates of NEFAs to compensate for reduced rates of carbohydrate oxidation (Table 2) (Orngreen et al 2009; Preisler et al 2013a). A high VO2peak is generally associated with an increase in fat use during submaximal exercise, but because NEFAs oxidation rates are already high in some GSDs, it remains to be investigated how much more NEFA oxidation rates may be further increased in these GSDs. A limitation to fat use in GSDs may be a depletion of glycosyl units to feed the TCA-cycle. Thus, there is a shortage of anapleurotic intermediates to spark the TCA-cycle and thus facilitate oxidation of fat. The effect of exercise training in glycogen storage diseases Despite the possible benefits of regular supervised exercise programs in GSDs, there are currently few studies that have assessed the effect of exercise in this group of patients (Quinlivan et al 2011; Voet et al 2013). This is not surprising, considering how rare the GSDs are. Aerobic exercise training in myophosphorylase deficiency Three studies have assessed the effect of exercise training in myophosphorylase deficiency. In all studies, the exercise intervention was aerobic exercise (cycling or walking) of light intensity (60–70 % of maximal heart rate). The subjects exercised 3, 4 or 5 sessions per week; in the study of Haller et al for 14 weeks (eight patients), in the study of Mate-Munoz et al for 8 months (ten patients) and in the study of Ollivier et al for 8 weeks (five patients) (Haller et al 2006; MateMunoz et al 2007; Ollivier et al 2005a). Haller et al reported increases in the subjects average work rate (36 %) and average oxygen uptake (14 %) during exercise. In line with this, there was an increase in cardiac output of 15 %. In addition, increased mitochondrial enzyme activity after training may have played a role in enhanced rates of oxidative phosphorylation (Haller et al 2006). Mate-Munoz et al reported increases in VO2peak, from a mean of 13.0 to 18.8, and increases in peak power output (from 0.8 to 1.1 W/
kg) before vs. after training. Ollivier et al found no increase in VO2max after the training period, probably because the total training volume in this study was low (Ollivier et al 2005a). In all studies, exercise was well tolerated and no adverse events were reported (Haller et al 2006; Mate-Munoz et al 2007; Ollivier et al 2005a). Interestingly, Mate-Munoz et al reported that there was a tendency to a decrease in baseline CK levels with training (Mate-Munoz et al 2007). In addition, two studies have investigated how physically active persons with myophosphorylase deficiency are in their daily lives (Lucia et al 2012; Ollivier et al 2005b). In both of these studies, the authors found that an increase in physical activity level was associated with an amelioration of clinical symptoms in the patients, and in the study by Lucia et al the VO2peak was 23 % higher in physical active patients compared to the inactive patients (21.0 vs. 16.1 ml kg -1 min-1), supporting the conclusion that physical activity improves exercise capacity (Lucia et al 2012). Exercise training in patients with GAA deficiency GAA deficiency is the only GSD for which there is a specific treatment in the form of enzyme replacement therapy (ERT). ERT has improved outcomes for patients with GAA deficiency, but improvements in muscle function on ERT are found to plateau, typically within a year on ERT (van der Ploeg et al 2010, 2012). This may be the reason why health care providers and patients are looking for other additional therapies, such as exercise training, that may improve muscle function. Linda van den Berg et al recently published data from the first randomized controlled study of the effect of exercise in patients with GAA deficiency (van den Berg 2014). In that study, 23 out of 25 patients successfully completed a combined strength (including core exercises) and aerobic exercisetraining program lasting for 12 weeks (three sessions per week) (van den Berg 2014). All patients included had been treated with ERT for at least a year before inclusion. The authors reported improvements in primary outcomes (endurance, muscle strength, and muscle function) and in core stability. Absolute values of oxygen uptake were not reported, but VO2peak increased from 69.4 to 75.9 % (VO2peak % of normal) and 6-minutes walking distance improved 16 meters (from 492 to 508 meters) in the patients. Two patients reported pain and fatigue after the first week of training and CK levels were increased in these patients. However, their CK values normalized and symptoms disappeared within a week, and the participants completed the full exercise program, without additional adverse events (van den Berg 2014). The positive effect of training in GAA deficiency was also documented in a smaller study by Terzis et al, who also studied the effect of a combined strength and aerobic exercise training program, in six patients with GAA deficiency lasting for 6 months (Terzis et al 2011). The authors found that 6-
J Inherit Metab Dis
minute walking distance and strength improved in the patients that completed the exercise program (Terzis et al 2011). Both of the previously mentioned studies are supported by studies in mice deficient of the GAA enzyme (Nilsson et al 2012). In mice, aerobic capacity, grip strength, and motor function improved significantly with endurance training compared to ERT alone, however, glycogen clearance was unaffected by exercise training. Interestingly, the effect of exercise training on strength and motor function was even higher than ERT alone in the mice (Nilsson et al 2012). Exercise has also been used in a few patients during ERT; with the rationale that an increase in blood flow to the exercising muscles may improve delivery of the recombinant enzyme to the working muscles (Strothotte et al 2010; Terzis et al 2012). However, in one study, there were no additional effect of exercise during infusions, and in the other study the patients also exercised in between biweekly infusion, so it could have been the exercise per se that improved walking distance. Slonim et al investigated the effect of exercise in persons with GAA deficiency before ERT became available and the authors reported a plateau in the disease progression in compliant patients (Slonim et al 2007).
Energy supplements to improve exercise tolerance It is popular, even amongst non-professionals, to ingest sports drinks before, during, and after exercise. However, the effect on performance of most of the available ergogenic aids is negligible or absent in persons with healthy skeletal muscle (Rodriguez et al 2009). In GSDs, on the other hand, the effect of substrate supplementation on exercise performance may be striking. The primary purpose of substrate supplementation therapy in GSDs is to bypass the metabolic block in skeletal muscle and deliver substrate, which may be used instead of intramuscular energy sources, to compensate for a reduced ATP production during exercise (Fig. 1). Effect of glucose in glycogen storage diseases affecting glycogenolysis Oral intake of sucrose (glucose) improves exercise tolerance in myophosphorylase deficiency (Andersen et al 2008; Vissing and Haller 2003). Infusions of glucose improve exercise tolerance in GDE deficiency and in PGM deficiency, but the effect of oral ingestion has not yet been investigated (Preisler et al 2013a, b). In muscle PHK deficiency, glucose infusion has had an effect in one patient, but no effect in two other cases (Orngreen et al 2008; Preisler et al 2012b). This discrepancy is partly explained by residual PHK enzyme activity, or activation of myophosphorylase by allosteric
effectors, which serves to preserve substantial glycogenolytic capacity (Preisler et al 2012b). Substrate supplementation in glycogen storage diseases affecting glycolysis Currently, there are no suitable candidates for oral substrate supplementation therapy to give patients with blocks in the glycolytic pathway. Lactate, which is readily use by muscle during exercise, is an obvious candidate, and lactate infusions improve work capacity in muscle PFK deficiency (Haller and Lewis 1991). Unfortunately, orally ingested lactate causes stomach discomfort due to osmotic induced diarrhea, and cannot be used by the patients (unpublished observations). Of note, sucrose ingestion should be avoided before exercise in muscle PFK deficiency, because foods that increase insulin levels reduces exercise tolerance, by suppressing NEFA release from adipocytes, and thus deprives these patients of their primary energy source during exercise (Haller and Lewis 1991). In fact, the best advice to give patients with muscle PFK deficiency is to avoid food before engaging in more strenuous physical activity. Patients with PGAM deficiency were unaffected by infusion of glucose, and neither improvements or worsening of exercise tolerance was observed, probably because residual enzyme activity maintains sufficient flux through the glycolytic pathway in this GSD (Chasiotis et al 1982; Vissing et al 2005). Substrate supplementation to prevent hypoglycemia A fraction of the glucose that is catabolized during exercise is liver-derived glucose (up to 40 % of oxidation rates), and hypoglycemia may develop in GDE deficiency and PGM deficiency due to the combination of reduced liver glucose output and increased skeletal muscle uptake of glucose during exercise (Preisler et al 2013a, b; Richter and Hargreaves 2013; Tegtmeyer et al 2014). Ingestion of carbohydrates serves to prevent hypoglycemia during and after exercise in these conditions. However, it remains to be investigated how and when substrate supplementation should be given to prevent hypoglycemia in these conditions.
Safety Dynamic exercise-induced symptoms — exercise intensity High-intensity exercise, specifically exercises that are dependent mainly on anaerobic energy production, such as resistance exercise should be avoided or performed with caution at a low intensity in the GSDs that present with dynamic exercise-induced symptoms (Table 1 and Fig. 2). Aerobic
J Inherit Metab Dis
exercise at an intensity corresponding to around 60 % of VO2peak (i.e.,
GLYCOGENOSES
Exercise in muscle glycogen storage diseases Nicolai Preisler & Ronald G Haller & John Vissing
Received: 7 August 2014 / Accepted: 9 September 2014 # SSIEM 2014
Abstract Glycogen storage diseases (GSD) are inborn errors of glycogen or glucose metabolism. In the GSDs that affect muscle, the consequence of a block in skeletal muscle glycogen breakdown or glucose use, is an impairment of muscular performance and exercise intolerance, owing to 1) an increase in glycogen storage that disrupts contractile function and/or 2) a reduced substrate turnover below the block, which inhibits skeletal muscle ATP production. Immobility is associated with metabolic alterations in muscle leading to an increased dependence on glycogen use and a reduced capacity for fatty acid oxidation. Such changes may be detrimental for persons with GSD from a metabolic perspective. However, exercise may alter skeletal muscle substrate metabolism in ways that are beneficial for patients with GSD, such as improving exercise
Communicated by Georg Hoffmann Presented at the GSD Conference in Heidelberg Germany, November 28– 30, 2013 N. Preisler (*) : J. Vissing Neuromuscular Research Unit, Section 3342, Department of Neurology, Rigshospitalet, University of Copenhagen, Blegdamsvej 9, 2100 Copenhagen, Denmark e-mail: [email protected] J. Vissing e-mail: [email protected] R. G. Haller Department of Neurology, The University of Texas Southwestern Medical Center, Dallas, TX, USA e-mail: [email protected]
tolerance and increasing fatty acid oxidation. In addition, a regular exercise program has the potential to improve general health and fitness and improve quality of life, if executed properly. In this review, we describe skeletal muscle substrate use during exercise in GSDs, and how blocks in metabolic pathways affect exercise tolerance in GSDs. We review the studies that have examined the effect of regular exercise training in different types of GSD. Finally, we consider how oral substrate supplementation can improve exercise tolerance and we discuss the precautions that apply to persons with GSD that engage in exercise. Abbreviations GSD Glycogen storage disease ATP Adenosine triphosphate RPE Rate of perceived exertion PHK Phosphorylase kinase CK Creatine kinase GDE Glycogen debranching enzyme PGM Phosphoglucomutase CDG Congenital disorder of glycosylation PFK Phosphofructokinase NEFA Non esterified fatty acid PGAM Phosphoglycerate mutase GAA α-1,4-glucosidase VO2peak Peak oxygen consumption MET Metabolic equivalents TCA-cycle Tricarboxylic acid cycle ERT Enzyme replacement therapy
R. G. Haller Department of Neurology, The Veterans Affairs Medical Center, Dallas, TX, USA
Introduction
R. G. Haller The Neuromuscular Center, Institute for Exercise and Environmental Medicine of Presbyterian Hospital, Dallas, TX, USA
Glycogen storage diseases (GSD) or glycogenoses are inborn errors of metabolism caused by mutations in the genes coding
J Inherit Metab Dis
for enzymes involved in glycogen or glucose metabolism (DiMauro et al 2004; Laforet et al 2012). GSDs have a broad clinical spectrum and can affect many tissues. Typically though, skeletal muscle and/or liver involvement dominate the clinical presentation, owing to the significance of glycogen– and glucose metabolism in these organs (Laforet et al 2012). In the GSDs that affect muscle, the consequence of a block in skeletal muscle glycogen breakdown (glycogenolysis), or in the anaerobic catabolism of glucose-6-phosphate (glycolysis), is an impairment of muscular performance, owing to 1) an increase in glycogen storage that disrupts contractile function and/or 2) a reduced substrate turnover below the block, which inhibits skeletal muscle ATP production (DiMauro et al 2004; Laforet et al 2012). Exercise, as a treatment for neuromuscular diseases, has received growing attention in the last decades, as new evidence has revoked the older notion that muscle contraction may accelerate the disease process, and because no curative treatment for these conditions are in immediate sight. In the GSDs, exercise as a treatment may seem counter intuitive for many patients and also clinicians, because it is exercise that provoke symptoms in many GSDs, and some types of exercise are not well tolerated by the patients (DiMauro et al 2004). However, it is important to deal with exercise in GSDs for several reasons. First, exercise intolerance is severe in most GSDs, which partly is caused by a sedentary lifestyle in persons with muscle disease. Immobility is associated, not only with skeletal muscle atrophy, but also with metabolic alterations in muscle leading to an increased dependence on glycogen use and a reduced capacity for fatty acid oxidation (Stein and Wade 2005). Such changes may be detrimental for persons with GSD from a metabolic perspective, and these changes can potentially worsen the muscle phenotype leading to more disuse; initiating a “vicious cycle of disuse”. Secondly, exercise is the most powerful natural inducer of metabolic change in skeletal muscle in humans. Theoretically, exercise may alter skeletal muscle substrate metabolism in ways that are specifically beneficial for patients with GSD, such as improving exercise tolerance and increasing the ability to oxidize fat. In addition, like in healthy subjects, a regular exercise program has the potential to improve general health and fitness and improve quality of life, if executed properly (Blair et al 1995; Pedersen and Saltin 2006). However, there are limitations and precautions that apply to persons with GSD that engage in regular exercise training programs and this has to be considered when prescribing exercise to this patient population. In this review we will use the term “exercise” synonymously with dynamic exercise, i.e., involvement of a large muscle mass performing aerobic activities such as walking, running or cycling, unless otherwise stated. Almost all exercise studies and studies of skeletal muscle metabolism in humans have been conducted in adults, and hence our recommendations are
based on what we know about skeletal muscle substrate use and exercise in adults with GSD. In this review, we describe skeletal muscle substrate use during exercise in GSDs, and how blocks in metabolic pathways affect exercise tolerance in GSDs. We review the studies that have examined the effect of regular exercise training in different types of GSD. Finally, we consider how oral substrate supplementation can improve exercise tolerance and we discuss the precautions that apply to persons with GSD that engage in exercise. We limit the review to GSDs that have been examined with exercise interventions, and we begin our review with a brief overview of these disorders.
Glycogen storage diseases affecting muscle In Table 1, we present the GSDs covered by this review. We have not discussed the following GSDs affecting muscle, because they have not yet been studied with exercise intervention; GSD type IV (branching enzyme deficiency), phosphoglycerate kinase deficiency, GSD type XI (muscle lactate dehydrogenase deficiency), GSD type XII (aldolase A deficiency), GSD type XIII (beta enolase deficiency) and GSD type 0 (glycogen synthase deficiency) and GSD type XV (glycogenin deficiency). In the following paragraphs we will briefly present each GSD, in which exercise interventions have been performed, and comment on clinical and metabolic findings that are relevant for prescription of exercise programs. The epidemiology, clinical presentation, disease course and management of the GSDs have been reviewed in detail elsewhere (Baethmann et al 2008; DiMauro et al 2004; Kishnani et al 2010; Laforet et al 2012; Lucia et al 2008; Nakajima et al 2002; Vissing and Orngreen 2014). The enzymatic steps involved in the GSDs reviewed in this article are presented in Fig. 1. General phenotypes, static vs. dynamic symptoms (Fig. 2) GSDs that affect muscle can present with different clinical phenotypes, and the response and tolerance to exercise can also differ, sometimes even within the same type of GSD (Echaniz-Laguna et al 2010; Orngreen et al 2008; Preisler et al 2012b). Despite the variability in phenotype, muscle GSDs can generally be divided into two groups; a group with primarily exercise-related symptoms and a group with fixed symptoms due to muscle atrophy and weakness (Fig. 2) (DiMauro and Lamperti 2001). Dynamic exerciseinduced symptoms signify an “acute energy crisis” within skeletal muscle, caused by a mismatch in the supply and demand for ATP. Even though the classification of muscular symptoms into these two classes is practical from a clinical point of view, it is important to note that phenotypes may overlap (Fig. 2).
Muscle phosphoglycerate mutase PGAM2 (HGNC:8889) No (EC 5.4.2.11) Phosphoglucomutase 1 (EC 5.4.2.2)
GSD type XIV† Phosphoglucomutase 1 deficiency [#614921]
Yes
No
No
No
No
No
No
No
Yes
Yes
No
No
No
No
Yes
No
0–12 %
2–7 %
0–29 %
0–8 %
0%
0–30 %
1–30 %
Yes
Yes
Yes
Yes
Yes
Rare
No
(Tegtmeyer et al 2014)
(Naini et al 2009; Oh et al 2006)
(Orngreen et al 2008; Wuyts et al 2005)
(Nakajima et al 2002)
(Lucia et al 2008)
(Ding et al 1990; Moses et al 1986)
(Bembi et al 2008; Straub 2008)
This table provides an overview of selected clinical and biochemical findings in glycogen storage diseases that have been studied with exercise tests. The data concerns adults and the response to exercise. #OMIM online mendelian inheritance in man® number, GSD glycogen storage disease, EC # enzyme commission number, HGNC ID HUGO gene nomenclature committee ID. * Cardiac hypertrophy is frequent in GSD type IIIa, but frank cardiomyopathy is rare in this GSD. Significant cardiac involvement is rare in GSD type II, but arrhythmias are probably more frequent in this GSD than in the general population (Sacconi et al 2014). §=Skeletal muscle or fibroblast enzyme activity levels (approximate levels, typical ranges). Methods to measure enzyme activity differ among studies. ‡=Whether or not exercise-induced rhabdomyolysis or myoglobinuria is a feature of the GSD. There are a few reports of rhabdomyolysis in GSD type IIIa, but it is rare compared to the other GSDs associated with rhabdomyolysis. †=GSD type XIV is also a congenital disorder of glycosylation of mixed type
PGM1 (HGNC:8905)
PHKA1 (HGNC:8925)
GSD type X (Phosphoglycerate mutase deficiency) [#261670]
GSD type IXd (Muscle phosphorylase kinase deficiency) [#300559]
No No
PYGM (HGNC:9726) PFKM (HGNC:8877)
Muscle Phosphofructokinase (EC 2.7.1.11) Phosphorylase kinase (EC 2.7.11.19)
GSD type VII (Tauri disease) [#232800]
Yes
Yes
Cardiac* Respiratory Liver Residual Rhabdo References§ involvement involvement affection enzyme activity§ myolysis‡
AGL (HGNC:321)
GAA (HGNC:4065)
Official gene symbol (HGNC ID)
Amylo-alpha-1,6-glucosidase & 4-alpha-glucanotransferase (EC 3.2.1.33 & EC 2.4.1.25) GSD type V (McArdle disease) [#232600] Myophosphorylase (EC 2.4.1.1)
Alpha-1,4-glucosidase (EC 3.2.1.20)
GSD type II (Pompe disease) [#232300]
GSD type IIIa (Cori or Forbes disease) [#232400]
Enzyme deficiency (EC #)
Glycogen storage disease (name) [#OMIM]
Table 1 Reviewed glycogen storage diseases affecting the glycogenolytic and glycolytic pathways in skeletal muscle
J Inherit Metab Dis
J Inherit Metab Dis
Fig. 1 Simplified overview of skeletal muscle glycogen and glucose use. The figure provides a simplified overview of the glycogenolytic and glycolytic pathways, the adenylate kinase reaction and the release and uptake of blood borne substrates (within red lines). Enzymes or allosteric effectors are in italics. Enzymes in bold are enzymes deficient in the glycogen storage disease that are covered in this review. Dashed arrows indicate that chemical reactions have been left out, to simplify the figure. PHK phosphorylase kinase, Myophos myophosphorylase, GDE glycogen debranching enzyme, PGM phosphoglucomutase, PFK
phosphofructokinase, PGAM phosphoglycerate mutase, GAA α-1,4-glucosidase, HEX hexokinase (E.C. 2.7.1.1), PGI glucose-6-phosphate isomerase (E.C. 5.3.1.9). LDH lactate dehydrogenase (E.C. 1.1.1.27), ADK adenylate kinase (E.C. 2.7.4.3), AMPD adenosine monophosphate deaminase (E.C. 3.5.4.6), Pi inorganic phosphate, AMP adenosine monophosphate, PKA protein kinase A (E.C. 2.7.11.11), NEFA nonesterified fatty acids, ADP adenosine diphosphate, ATP adenosine triphosphate, IMP inosine monophosphate, P phosphate (Cory 2006; Harris 2006; Smith 2006)
Disorders affecting skeletal muscle glycogen breakdown
oxidation in contracting muscles (Haller and Vissing 2002; Orngreen et al 2009). Glucose is helpful, because it can be metabolized below the metabolic block in this GSD (Fig. 1) (Andersen et al 2008; Haller and Vissing 2002; Vissing and Haller 2003).
Myophosphorylase deficiency, McArdle disease Myophosphorylase deficiency was first described in 1951, and this GSD may be considered the prototype of GSDs with exercise-related symptoms. Myophosphorylase releases glucose-1-phosphate from glycogen (Fig. 1). Almost all mutations in the PYGM gene are null mutations, and the consequence of this is a complete block in skeletal muscle glycogenolysis. This leads to severe exercise intolerance with dynamic, exercise-induced symptoms caused by an acute energy crisis within skeletal muscle. Patients with myophosphorylase deficiency develop a so called second-wind phenomenon during aerobic exercise of moderate intensity (Pearson et al 1961). During the first minutes of exercise there is a gradual and continuous increase in heart rate and rate of perceived exertion (RPE) that is out of proportion to the actual work rate. Heart rate and RPE peak after 6–8 minutes of exercise, at which time heart rate and RPE correspond to values observed at maximal rates of exercise. In the following minutes, heart rate and RPE decline (the second-wind), so that exercise at the same work load becomes much better tolerated (Haller and Vissing 2002). The second-wind is attributable to enhanced uptake of blood-born glucose from the liver and enhanced fat
Muscle phosphorylase kinase deficiency Muscle phosphorylase kinase (PHK) initiates skeletal muscle glycogen breakdown by phosphorylating and activating myophosphorylase (Fig. 1) and therefore it may be anticipated that patients with muscle PHK deficiency present with dynamic, exercise-induced symptoms, similar to patients with myophosphorylase deficiency. This is true in some cases of this rare disease, but generally the dynamic exercise-induced symptoms are milder (Laforet et al 2012). In some cases, persons with muscle PHK deficiency may be almost asymptomatic (Echaniz-Laguna et al 2010; Preisler et al 2012b). Two patients were only diagnosed because of elevated plasma creatine kinase (CK) levels, and three of the less than ten cases reported in the literature never experienced exercise intolerance. Therefore it is likely that some cases go undiagnosed (Echaniz-Laguna et al 2010; Orngreen et al 2008; Preisler et al 2012b).
J Inherit Metab Dis
one third of the patients develop muscle weakness and wasting in their 3rd to 4th decade (Kishnani et al 2010; Moses et al 1986). It is interesting that dynamic exerciseinduced symptoms are not at the forefront of symptoms in this GSD, considering that the GDE works in concert with myophosphorylase to degrade glycogen (Fig. 1) (Harris 2006; Kishnani et al 2010). We have recently documented that young patients with GDE deficiency and without overt muscle weakness, do in fact have significant exercise intolerance to moderate-intensity exercise, suggesting that this metabolic myopathy should also be classified amongst the GSDs that present with dynamic exercise-induced symptoms (Fig. 2) (Kishnani et al 2010; Preisler et al 2013b). In spite of exercise intolerance, patients with GDE deficiency, like patients with muscle PHK deficiency, tolerate exercise much better than patients with myophosphorylase deficiency and myoglobinuria and rhabdomyolysis are very rare in this GSD (Table 1) (DiMauro et al 2004; Kishnani et al 2010). This could relate to the small amount of glucose, which can be released from muscle glycogen at the onset of exercise, until the GDE reaches a branching site on the glycogen molecule. Fig. 2 Static or dynamic skeletal muscle affection in glycogen storage diseases. The figure provides an overview of the typical skeletal muscle presentation (framed boxes) in glycogen storage diseases (GSD). Static muscular symptoms are used to describe muscle affection in GSDs that present with progressive skeletal muscle weakness and wasting, i.e., fixed weakness. Dynamic exercise-induced muscular symptoms are provoked by exercise or other strenuous activities that increase muscular energy requirements, and these symptoms resolve with rest when energy demands diminish. In a proportion of the patients with a GSD that present with dynamic exercise-induced symptoms, a static skeletal muscle phenotype develops later in life (arrows). # Patients with muscle PHK deficiency may be almost asymptomatic. § We suggest that GDE deficiency is classified as a GSD presenting with dynamic exercise-induced symptoms. * In phosphoglucomutase deficiency, approximately half of the cases are exercise intolerant, and a few cases present with weakness, but currently the precise classification of this GSD is unclear. PHK phosphorylase kinase, Myophos myophosphorylase, GDE glycogen debranching enzyme, PGM phosphoglucomutase, PFK phosphofructokinase, PGAM phosphoglycerate mutase, GAA α-1,4-glucosidase (Echaniz-Laguna et al 2010; Kishnani et al 2010; Lucia et al 2008; Nadaj-Pakleza et al 2009; Nakajima et al 2002; Preisler et al 2012b, 2013b; Tegtmeyer et al 2014)
Glycogen debranching enzyme deficiency, Cori or Forbes disease Glycogen debranching enzyme (GDE) deficiency is traditionally considered to be a liver glycogenosis, despite the fact that the GDE is also deficient in skeletal muscle in ~85 % of cases (GDE type a) (Coleman et al 1992; van Hoof and Hers 1967). Skeletal muscle involvement is mild or absent in children and young adults, but approximately
Phosphoglucomutase type 1 deficiency The phosphoglucomutase (PGM) enzyme is not directly involved in the release of glucose residues from the glycogen molecule, however, this enzyme catalyzes the last step of glycogenolysis, converting glucose-1-phosphate to glucose-6-phosphate, which may then enter the glycolytic pathway (Fig. 1) (Harris 2006). The first case of genetically verified PGM type 1 deficiency in skeletal muscle was reported in 2009, and this novel GSDs was added as the 14th (GSDXIV) of the GSDs (Stojkovic et al 2009). In 2014, however, another chapter was added to this disease entity. PGM deficiency has now been documented to not only affect skeletal muscle glycogen breakdown, but also to be the cause of defects in protein glycosylation (Tegtmeyer et al 2014). This is because glucose-1phosphate is an intermediate in protein glycosylation, and the PGM1 isoform is expressed as the predominant isoform in most tissues (McAlpine et al 1970). Protein Nglycosylation is important in cell signaling and multiple organ affection is very common in this novel disorder of mixed-type congenital disorder of glycosylation (CDG) and GSD, with a very broad phenotype (Tegtmeyer et al 2014). Half of the patients report exercise intolerance, and in a few patients, the myopathy dominates the phenotype. The exact mechanism behind the broad disease spectrum and predominant muscle affection in some cases of this GSD/ CDG is not clear at present (Stojkovic et al 2009; Tegtmeyer et al 2014).
J Inherit Metab Dis
Disorders of skeletal muscle glycolysis Muscle phosphofructokinase deficiency, Tauri disease Muscle phosphofructokinase (PFK) deficiency is a disorder of glycolysis that was first described in 1965 (Nakajima et al 2002). The phenotype and response to exercise in patients with muscle PFK deficiency is very similar to myophosphorylase deficiency, albeit these patients are even more intolerant to exercise because of a complete block in glycolysis, limiting skeletal muscle substrate use to almost exclusive dependence on non-esterified fatty acid (NEFA) oxidation (Haller and Lewis 1991; Ono et al 1995; Vissing et al 1996). In contrast to myophosphorylase deficiency, patients with muscle PFK deficiency lack a second-wind. In fact, glucose triggers an out-ofwind phenomenon in this GSD, most likely induced by suppression of plasma fatty acid concentrations by the action of insulin, as a consequence of an increase in blood glucose levels (Haller and Lewis 1991). In addition to skeletal muscle involvement, most patients with muscle PFK deficiency also have a compensated hemolysis, due to co-expression of the muscle isoform of PFK in erythrocytes (Nakajima et al 2002). Phosphoglycerate mutase deficiency The first case of muscle phosphoglycerate mutase (PGAM) deficiency was described in 1981 (DiMauro et al 1981). PGAM converts 3-phosphoglycerate to 2-phosphoglycerate in one of the final steps in the glycolytic pathway (Fig. 2) (Harris 2006). There are two isoforms of the PGAM enzyme; the brain and the muscle form, which are expressed in a tissue specific manner (Sakoda et al 1988; Shanske et al 1987). Patients with PGAM deficiency are much more mildly affected than patients with muscle PFK deficiency and are usually asymptomatic, unless they engage in high-intensity exercise, which may cause contractures, myalgia and myoglobinuria (Naini et al 2009; Vissing et al 2005). The mild phenotype is explained by the expression of the brain isoform of the PGAM enzyme in muscle, which is responsible for the residual enzyme activity measured in muscle in this GSD (Table 1) (Oh et al 2006). A peculiar finding in this GSD is tubular aggregates in muscle biopsies, which are present in around one third of cases (Naini et al 2009). Therefore, PGAM deficiency should be considered in patients with dynamic exercise-induced symptoms and tubular aggregates on biopsy. Especially, if other more common GSDs have been ruled out (Naini et al 2009). Disorders affecting lysosomal glycogenolysis Pompe disease, α-1,4-glucosidase deficiency Pompe disease was described in 1932, and this glycogen storage disease is distinct from other GSDs because the
enzyme defect is within lysosomes where it blocks glycogenolysis (Hers 1963; van der Ploeg and Reuser 2008). The disease is caused by deficiency of α-1,4-glucosidase (GAA) and patients with the late-onset form of GAA deficiency (~70 % of cases), typically present in the 2nd to 3rd decade with progressive muscle weakness and wasting primarily affecting the proximal and axial muscles (Bembi et al 2008). Exercise intolerance is pronounced in many patients with GAA deficiency, but the cause of exercise intolerance in this GSD is probably not related to a supply–demand mismatch of ATP, because lysosomal glycogen breakdown contributes minimally to the ATP production of the myocyte (Preisler et al 2012a). The functional limitation is more likely the consequence of skeletal muscle weakness and wasting and therefore the response to exercise is different from the other GSDs affecting muscle (Preisler et al 2012a).
Skeletal muscle substrate metabolism during exercise in GSD; the metabolic response to single bouts of exercise If the metabolic block causing GSD results in a significant impairment in glycolytic flux, the magnitude of the mismatch in the supply–demand chain of ATP becomes apparent only during exercise. Results from exercise studies show that in most GSDs there is clear association between exercise intolerance, exercise-induced dynamic symptoms, the effect of substrate supplementation therapy and the extent of the derangements in skeletal muscle energy production during exercise (Preisler et al 2012a, b, 2013a, b). Healthy skeletal muscle uses a variety of substrates to generate the ATP that fuels muscular contraction (McArdle et al 2007a). However, the supply and rate of production of ATP from the various sources of fuel differ, and skeletal muscle fuel selection during exercise is intensity- and supply-dependent (van Loon et al 2001). Figure 3 provides an overview of the main substrates metabolized during exercise, the limit of supply for each energy source, and how fast ATP may be produced from the various sources. Table 2 provides an overview of the metabolic and physiological response to selected types of exercises that have been used to study GSDs. Forearm exercise — anaerobic ATP production The forearm exercise test specifically challenges skeletal muscle anaerobic energy production (Haller and Vissing 2004a). During anaerobic conditions, lactate is generated from pyruvate and plasma lactate concentration is a good indicator of the flux through the glycolytic pathway. Plasma ammonia concentrations usually increase slightly after forearm exercise in healthy persons, because the adenylate kinase reaction is used
J Inherit Metab Dis
Fig. 3 Energy yield and supply and time to peak power output from energy sources used by skeletal muscle during exercise. This figure provides an overview of the major substrates used to generate the ATP that fuels muscle contraction during exercise. * Chemical reactions have been simplified for ease of view. § Exercise duration describes the time it takes to exhaust the specific fuel during exercise. Fuels differ in availability and how rapidly ATP can be produced from the different energy sources. † Anaerobic metabolism generates power outputs that are much higher than aerobic pathways and anaerobic metabolism is always engaged to supply ATP during maximal effort. ‡ With respect to oxidative metabolism, carbohydrate oxidation generates a substantially higher power output than fat oxidation, which explains the shift to carbohydrate oxidation as aerobic exercise intensity increases. Therefore, different
types of exercise, performed at different intensities, affect muscle fuel selection, i.e., the proportion of anaerobic or aerobic pathways used to fuel exercise. The figure shows that glycogen metabolism is essential for optimal muscular performance in any activity lasting more than a few seconds, which explains why climbing a flight of stairs may provoke symptoms in persons with glycogen storage disease (GSD). Fat represents an almost unlimited source of fuel; however, exclusive oxidation of non-esterified fatty acids (NEFA) can only maintain lower intensities of exercise ( 10 min of exercise. This type of exercise is dependent on mainly the aerobic combustion of glucose and NEFAs, to drive skeletal muscle contraction (Fig. 3). Measurements of substrate turnover during exercise in GSDs have shown increased oxidation rates of NEFAs to compensate for reduced rates of carbohydrate oxidation (Table 2) (Orngreen et al 2009; Preisler et al 2013a). A high VO2peak is generally associated with an increase in fat use during submaximal exercise, but because NEFAs oxidation rates are already high in some GSDs, it remains to be investigated how much more NEFA oxidation rates may be further increased in these GSDs. A limitation to fat use in GSDs may be a depletion of glycosyl units to feed the TCA-cycle. Thus, there is a shortage of anapleurotic intermediates to spark the TCA-cycle and thus facilitate oxidation of fat. The effect of exercise training in glycogen storage diseases Despite the possible benefits of regular supervised exercise programs in GSDs, there are currently few studies that have assessed the effect of exercise in this group of patients (Quinlivan et al 2011; Voet et al 2013). This is not surprising, considering how rare the GSDs are. Aerobic exercise training in myophosphorylase deficiency Three studies have assessed the effect of exercise training in myophosphorylase deficiency. In all studies, the exercise intervention was aerobic exercise (cycling or walking) of light intensity (60–70 % of maximal heart rate). The subjects exercised 3, 4 or 5 sessions per week; in the study of Haller et al for 14 weeks (eight patients), in the study of Mate-Munoz et al for 8 months (ten patients) and in the study of Ollivier et al for 8 weeks (five patients) (Haller et al 2006; MateMunoz et al 2007; Ollivier et al 2005a). Haller et al reported increases in the subjects average work rate (36 %) and average oxygen uptake (14 %) during exercise. In line with this, there was an increase in cardiac output of 15 %. In addition, increased mitochondrial enzyme activity after training may have played a role in enhanced rates of oxidative phosphorylation (Haller et al 2006). Mate-Munoz et al reported increases in VO2peak, from a mean of 13.0 to 18.8, and increases in peak power output (from 0.8 to 1.1 W/
kg) before vs. after training. Ollivier et al found no increase in VO2max after the training period, probably because the total training volume in this study was low (Ollivier et al 2005a). In all studies, exercise was well tolerated and no adverse events were reported (Haller et al 2006; Mate-Munoz et al 2007; Ollivier et al 2005a). Interestingly, Mate-Munoz et al reported that there was a tendency to a decrease in baseline CK levels with training (Mate-Munoz et al 2007). In addition, two studies have investigated how physically active persons with myophosphorylase deficiency are in their daily lives (Lucia et al 2012; Ollivier et al 2005b). In both of these studies, the authors found that an increase in physical activity level was associated with an amelioration of clinical symptoms in the patients, and in the study by Lucia et al the VO2peak was 23 % higher in physical active patients compared to the inactive patients (21.0 vs. 16.1 ml kg -1 min-1), supporting the conclusion that physical activity improves exercise capacity (Lucia et al 2012). Exercise training in patients with GAA deficiency GAA deficiency is the only GSD for which there is a specific treatment in the form of enzyme replacement therapy (ERT). ERT has improved outcomes for patients with GAA deficiency, but improvements in muscle function on ERT are found to plateau, typically within a year on ERT (van der Ploeg et al 2010, 2012). This may be the reason why health care providers and patients are looking for other additional therapies, such as exercise training, that may improve muscle function. Linda van den Berg et al recently published data from the first randomized controlled study of the effect of exercise in patients with GAA deficiency (van den Berg 2014). In that study, 23 out of 25 patients successfully completed a combined strength (including core exercises) and aerobic exercisetraining program lasting for 12 weeks (three sessions per week) (van den Berg 2014). All patients included had been treated with ERT for at least a year before inclusion. The authors reported improvements in primary outcomes (endurance, muscle strength, and muscle function) and in core stability. Absolute values of oxygen uptake were not reported, but VO2peak increased from 69.4 to 75.9 % (VO2peak % of normal) and 6-minutes walking distance improved 16 meters (from 492 to 508 meters) in the patients. Two patients reported pain and fatigue after the first week of training and CK levels were increased in these patients. However, their CK values normalized and symptoms disappeared within a week, and the participants completed the full exercise program, without additional adverse events (van den Berg 2014). The positive effect of training in GAA deficiency was also documented in a smaller study by Terzis et al, who also studied the effect of a combined strength and aerobic exercise training program, in six patients with GAA deficiency lasting for 6 months (Terzis et al 2011). The authors found that 6-
J Inherit Metab Dis
minute walking distance and strength improved in the patients that completed the exercise program (Terzis et al 2011). Both of the previously mentioned studies are supported by studies in mice deficient of the GAA enzyme (Nilsson et al 2012). In mice, aerobic capacity, grip strength, and motor function improved significantly with endurance training compared to ERT alone, however, glycogen clearance was unaffected by exercise training. Interestingly, the effect of exercise training on strength and motor function was even higher than ERT alone in the mice (Nilsson et al 2012). Exercise has also been used in a few patients during ERT; with the rationale that an increase in blood flow to the exercising muscles may improve delivery of the recombinant enzyme to the working muscles (Strothotte et al 2010; Terzis et al 2012). However, in one study, there were no additional effect of exercise during infusions, and in the other study the patients also exercised in between biweekly infusion, so it could have been the exercise per se that improved walking distance. Slonim et al investigated the effect of exercise in persons with GAA deficiency before ERT became available and the authors reported a plateau in the disease progression in compliant patients (Slonim et al 2007).
Energy supplements to improve exercise tolerance It is popular, even amongst non-professionals, to ingest sports drinks before, during, and after exercise. However, the effect on performance of most of the available ergogenic aids is negligible or absent in persons with healthy skeletal muscle (Rodriguez et al 2009). In GSDs, on the other hand, the effect of substrate supplementation on exercise performance may be striking. The primary purpose of substrate supplementation therapy in GSDs is to bypass the metabolic block in skeletal muscle and deliver substrate, which may be used instead of intramuscular energy sources, to compensate for a reduced ATP production during exercise (Fig. 1). Effect of glucose in glycogen storage diseases affecting glycogenolysis Oral intake of sucrose (glucose) improves exercise tolerance in myophosphorylase deficiency (Andersen et al 2008; Vissing and Haller 2003). Infusions of glucose improve exercise tolerance in GDE deficiency and in PGM deficiency, but the effect of oral ingestion has not yet been investigated (Preisler et al 2013a, b). In muscle PHK deficiency, glucose infusion has had an effect in one patient, but no effect in two other cases (Orngreen et al 2008; Preisler et al 2012b). This discrepancy is partly explained by residual PHK enzyme activity, or activation of myophosphorylase by allosteric
effectors, which serves to preserve substantial glycogenolytic capacity (Preisler et al 2012b). Substrate supplementation in glycogen storage diseases affecting glycolysis Currently, there are no suitable candidates for oral substrate supplementation therapy to give patients with blocks in the glycolytic pathway. Lactate, which is readily use by muscle during exercise, is an obvious candidate, and lactate infusions improve work capacity in muscle PFK deficiency (Haller and Lewis 1991). Unfortunately, orally ingested lactate causes stomach discomfort due to osmotic induced diarrhea, and cannot be used by the patients (unpublished observations). Of note, sucrose ingestion should be avoided before exercise in muscle PFK deficiency, because foods that increase insulin levels reduces exercise tolerance, by suppressing NEFA release from adipocytes, and thus deprives these patients of their primary energy source during exercise (Haller and Lewis 1991). In fact, the best advice to give patients with muscle PFK deficiency is to avoid food before engaging in more strenuous physical activity. Patients with PGAM deficiency were unaffected by infusion of glucose, and neither improvements or worsening of exercise tolerance was observed, probably because residual enzyme activity maintains sufficient flux through the glycolytic pathway in this GSD (Chasiotis et al 1982; Vissing et al 2005). Substrate supplementation to prevent hypoglycemia A fraction of the glucose that is catabolized during exercise is liver-derived glucose (up to 40 % of oxidation rates), and hypoglycemia may develop in GDE deficiency and PGM deficiency due to the combination of reduced liver glucose output and increased skeletal muscle uptake of glucose during exercise (Preisler et al 2013a, b; Richter and Hargreaves 2013; Tegtmeyer et al 2014). Ingestion of carbohydrates serves to prevent hypoglycemia during and after exercise in these conditions. However, it remains to be investigated how and when substrate supplementation should be given to prevent hypoglycemia in these conditions.
Safety Dynamic exercise-induced symptoms — exercise intensity High-intensity exercise, specifically exercises that are dependent mainly on anaerobic energy production, such as resistance exercise should be avoided or performed with caution at a low intensity in the GSDs that present with dynamic exercise-induced symptoms (Table 1 and Fig. 2). Aerobic
J Inherit Metab Dis
exercise at an intensity corresponding to around 60 % of VO2peak (i.e.,
