nature reviews molecular cell biology https://doi.org/10.

1038/s41580-023-00606-x

Review article Check for updates

Exercise metabolism and

adaptation in skeletal muscle
Jonathon A. B. Smith 1
, Kevin A. Murach 2
, Kenneth A. Dyar 3,4
& Juleen R. Zierath 1,5,6

Abstract Sections

Viewing metabolism through the lens of exercise biology has Introduction

proven an accessible and practical strategy to gain new insights into Skeletal muscle fibre types
local and systemic metabolic regulation. Recent methodological and subcellular characteristics

developments have advanced understanding of the central role of Acute exercise metabolism
in skeletal muscle
skeletal muscle in many exercise-associated health benefits and have
uncovered the molecular underpinnings driving adaptive responses Skeletal muscle responses
to acute exercise
to training regimens. In this Review, we provide a contemporary
Skeletal muscle adaptations
view of the metabolic flexibility and functional plasticity of skeletal to long-term exercise
muscle in response to exercise. First, we provide background on
Future directions
the macrostructure and ultrastructure of skeletal muscle fibres,
highlighting the current understanding of sarcomeric networks and
mitochondrial subpopulations. Next, we discuss acute exercise skeletal
muscle metabolism and the signalling, transcriptional and epigenetic
regulation of adaptations to exercise training. We address knowledge
gaps throughout and propose future directions for the field. This
Review contextualizes recent research of skeletal muscle exercise
metabolism, framing further advances and translation into practice.

1
Department of Physiology and Pharmacology, Karolinska Institutet, Stockholm, Sweden. 2Molecular Mass
Regulation Laboratory, Exercise Science Research Center, Department of Health, Human Performance and
Recreation, University of Arkansas, Fayetteville, AR, USA. 3Metabolic Physiology, Institute for Diabetes and Cancer,
Helmholtz Diabetes Center, Helmholtz Zentrum München, German Research Center for Environmental Health,
Neuherberg, Germany. 4German Center for Diabetes Research (DZD), Neuherberg, Germany. 5Department of
Molecular Medicine and Surgery, Karolinska Institutet, Stockholm, Sweden. 6Novo Nordisk Foundation Center
for Basic Metabolic Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen,
Denmark. e-mail: [email protected]

Nature Reviews Molecular Cell Biology

Review article

Introduction and fast-glycolytic MyHC type IIX (formerly known as type IID; encoded
As a primary site of nutrient storage, energy use and locomotion, skel- by MYH1)21. Muscle fibres have classically been ‘typed’ according to
etal muscle is central to the impact of physical activity on human health. metabolic — oxidative versus glycolytic enzyme — profiles22,23 and the
Periods of inactivity reduce skeletal muscle insulin sensitivity and oxi- predominant abundance22,24 or ATPase activity23 of MyHC isoforms.
dative capacity1, contributing towards impaired systemic metabolic MyHC are the motor proteins of myofibril thick filaments and
flexibility2 and an increased risk for cardiometabolic disease3. Physical determine important aspects of muscle function, such as maximum
activity offers a degree of protection against the deleterious effects shortening velocity25. Yet, the full extent of fibre type characteris-
of sedentary behaviour on whole-body metabolism4: larger volumes4 tics (Supplementary Fig. 1) depends upon matching the excitation–
and more vigorous types5 of physical activity, such as formal exercise, contraction coupling machinery and ATP provision to MyHC activity14,21.
probably convey additional benefit. This coordinated expression is a product of α-motor unit innervation
Bouts of exercise rapidly sensitize skeletal muscle to hormones and the transcriptional synchrony of resident myonuclei26,27.
(Supplementary Table 1) and nutrients. A single (or ‘acute’) exercise During differentiation (myogenesis) (Supplementary Box 1), a com-
session directly increases skeletal muscle transport of amino acids6 bination of intracellular forces ‘squeeze’ centrally located myonuclei to
and glucose7. These effects appear somewhat specific to the contracted the fibre periphery28, where they reside in a generally ordered pattern
musculature, and they enhance postprandial muscle protein synthesis8 of ‘domains’ in mature muscle. Discrete populations of myonuclei serve
and insulin-stimulated glucose disposal9,10 in the recovery period after specialized roles within fibres, including those governing the neuro-
exercise. Consistent exercise training (over weeks, months and years) muscular junction, myotendinous junction26,29,30 (the interface between
further augments skeletal muscle mass 11–13, peripheral insulin- muscle and tendon) and proprioceptive muscle spindles30. Likewise, the
.
sensitivity 12, maximal oxygen consumption (VO 2max ) 12–15 and function of ‘body’26 or ‘canonical’29 myonuclei (constituting ≥90% of
.
strength12,13. VO 2max (ref. 16) and strength17 are well-known predictors total myonuclei)26 is to direct fibre type specificity in muscle, and they
of mortality, and training for their improvement through endurance can be identified in mouse muscle by their Myh isoform signature26,29,30.
and resistance exercise (Box 1) reduces mortality risk in a manner that Myh-positive myonuclei from type I versus type II fibres possess unique
is most effective when both training modalities are performed within chromatin accessibility and transcription factor motif enrichment
the same exercise programme18. profiles that are distinct from one another and from myonuclei in
Although the association between regular physical activity and other cellular compartments, including the myotendinous junction26.
health span has been realized since antiquity, recent methodological This could underlie specific transcription factor-driven myonuclear
advances have allowed the field of exercise physiology to progress programmes26,30 that are partially responsible for regionalized gene
towards comprehensive systems-level profiling of the complex mole­ expression in muscle cells26,30,31 and for the co-expression of specific
cular interplay that occurs with exercise19,20. These advancements have calcium (Ca2+)-handling, sarcomeric and metabolic apparatus14,21,26,31
enabled better mechanistic understanding of the cause-and-effect so that fibre type contractile (also known as ‘twitch’) and metabolic
relationships underlying exercise adaptation and associated health properties generally align (Supplementary Fig. 1b).
benefits. The myonuclei in fibres from endurance-trained younger and older
In this Review, we provide a contemporary summary of the role individuals are more spherical, contain greater lamin A (LMNA) deposi-
of skeletal muscle in response to exercise. First, we address the major tion and are stiffer and less deformable than myonuclei in untrained
cellular makeup of skeletal muscle, highlighting fibre type properties, counterparts32. These structural and mechanical modifications may
and the importance of sarcomeric and mitochondrial networks. Next, facilitate the transduction of cytoskeletal forces towards the nucleus
we discuss skeletal muscle metabolism during acute exercise and the and improve myonuclear resilience against contractile damage32. As
influence of select modifiers such as intensity and timing on this bio­ such, exercise training-induced myonuclear remodelling could have
logy. Finally, we address underlying mechanisms of exercise-induced important implications for muscle adaptation and integrity across
skeletal muscle adaptation and consider differences between training the lifespan.
modalities that may facilitate distinct and complementary responses In general, human muscles often express a greater proportion of
beneficial for human performance and health. slower-twitch fibres than those of other species22,24, and human fibres
are slower-contracting than orthologous fibre types in most mammals,
Skeletal muscle fibre types and subcellular including rats and mice24. Similarly, type IIA fibres are the most oxida-
characteristics tive fibre type in rodent muscle, whereas type I fibres are most oxidative
The human body contains >600 skeletal muscles mainly comprising in humans22 (Supplementary Fig. 1). Such differences might contribute
long contractile cells called muscle fibres. The complex architecture to the lack of conformity between the transcriptomes of human and
of these fibres (Fig. 1a) provides clues into the intricacies of muscle mouse muscle after acute or chronic resistance exercise33 and should
function and homeostatic control. In this section we focus on the con- be considered when inferring human relevance from animal physiol-
tractile and metabolic properties of different muscle fibre types and ogy. Moreover, the properties of muscle can vary markedly among
examine how these attributes are supported by interacting networks individuals23,34, biological sexes23 (Supplementary Box 2), anatomical
of sarcomeres and mitochondria. locations35 and during ageing34,36 (Box 2). For example, discrete spa-
tial metabolomic differences within fibre types have been observed
Contractile, metabolic and myonuclear properties of muscle in mouse muscle37, and age-associated mitochondrial impairments
fibre types may induce a glycolytic shift in human fibres without a correspond-
Human muscles of the torso and limbs express three main fibre ing change in MyHC36. Collectively, this cautions against the use of
types, with slow-to-fast contractile properties in the following order: MyHC as a strict marker of metabolism and vice versa. Future inquiry
slow-oxidative myosin heavy chain (MyHC) type I (encoded by MYH7), fast should better define how covariates — including biological sex, social
oxidative-glycolytic (intermediate) MyHC type IIA (encoded by MYH2) gender, biological versus chronological age, metabolic health and

Nature Reviews Molecular Cell Biology

Review article

Box 1

Different exercise (sub)types and training response heterogeneity

in humans
.
Exercise (sub)types consumption (VO 2max)292–294. Approaches used to standardize
Exercise can be broadly classified as resistance, cardiorespiratory, cardiorespiratory exercise are plentiful (reviewed in ref. 295)
balance and flexibility-based. Flexibility and balance are important and most often include the allocation of intensity against fixed
. .
aspects of physical fitness contributing to fall prevention in older percentages of maximal power (W max), velocity (V max), heart
.
individuals282. However, discussion herein is dedicated to iterations of rate (HRmax) or VO 2max. However, the validity of normalizing
resistance and cardiorespiratory exercise, which are the main focus cardiorespiratory exercise to maximal parameters has been
of this Review. questioned295,296, as this can produce dissimilar physiological and
metabolic perturbations between individuals296. Moving forward,
Resistance exercise studies should strive to establish methods of standardization that
Traditional resistance exercise entails repetitions of dynamic provoke homeostatic disturbances consistent with distinct exercise
concentric (muscle-shortening) and eccentric (muscle-lengthening) intensity domains.
contractions against external load and is an effective intervention to Endurance (also known as aerobic) exercise is typically considered
increase skeletal muscle mass and strength249,283. The total amount to be a continuous bout of formal activity performed at low (<50%),
.
(volume), frequency and intensity of exercise are inherently linked moderate (~50-79%) or high intensities (≥80%) of VO 2max (ref. 292).
training variables that impact adaptation and performance. For High-intensity cardiorespiratory exercise can be further divided
resistance exercise, volume is often reported as the number of into high-intensity interval exercise (HIE) and sprint interval exercise
times (or ‘sets’) a particular muscle group is trained per week and (SIE), both of which constitute several bursts of higher-intensity effort
is the main stimulus for muscle accrual284. Twelve to 20 weekly sets is interspersed with periods of low-intensity active recovery. High-
.
sufficient to maximize hypertrophy285, and the frequency of exercise intensity intervals are conducted at ≥80% of VO 2max, whereas
can be adjusted to disperse training volumes according to personal sprint intervals are supramaximal or ‘all-out’ . Comparison of
292

preference286. cardiorespiratory exercise suggests that training within the high-

Intensity of resistance exercise is generally normalized to intensity range is more effective292,293 or equally as effective294 at
.
a percentage of the maximum load that an individual can lift increasing VO 2max and requires less training volume than moderate
(expressed as the percentage of 1 repetition maximum (RM))287 or intensities to do so294. HIE and SIE training also improve endothelial
how close an exercise set is taken to momentary muscular failure function to greater extents, whereas moderate intensities favour
(in other words, the inability to complete the concentric portion of a long-term glycaemic control (lower glycated haemoglobin A1c
movement)288. Equal gains in muscle mass can be made irrespective (HbA1c))292. Further intricacies are seen between interval training
of load-intensity (<30% 1 RM to ≥80% 1 RM)287 when sets are taken subtypes. Eight weeks of HIE training improved cardiac stroke
.
proximal to failure288. Thus, increasing the number of repetitions volume, VO 2max and endurance (3 km) performance297. By contrast,
against a fixed load or increasing the load lifted for a fixed number of an equivalent period of SIE training potentiated anaerobic capacity
repetitions are both viable progression strategies to promote muscle and sprinting (300 m) performance297.
hypertrophy289. Conversely, absolute strength (1 RM)287, tendon Training variables might also discretely regulate muscle
stiffness290 and running economy (that is, the metabolic cost at a mitochondria (see the section ‘Skeletal muscle adaptations to long-
given velocity of submaximal running)254 are improved more by high- term exercise’). Hence, combinations of endurance exercise, HIE and
load resistance training. Collectively, this implies that implementing SIE are recommended to maximize both health and performance
various loading strategies could represent the best approach for benefits. Consistent with this, just 1–3 h per week of moderate or high-
attaining the breadth of muscle-related adaptations to resistance intensity cardiorespiratory exercise could lower mortality risk, and
training. incorporating resistance exercise confers added protection18.
The stress imposed by a single bout of resistance exercise
upregulates amino acid transporters and sensors in muscle at least Interindividual variation in exercise adaptation
in part through activating transcription factor 4 (ATF4)6. Synergizing The magnitude of adaptation to resistance298 and endurance299,300
with the mammalian target of rapamycin (mTOR) (see the section training differs substantially between individuals. In part, this
‘Skeletal muscle responses to acute exercise’) (Fig. 3), this sensitizes probably stems from the aforementioned challenges regarding
muscle to amino acids for ≥24–48 h after resistance exercise6,8, and exercise standardization295,296, but also from genetic300 and
dietary protein intakes of ≥1.2–1.6 g per kg body mass daily modestly environmental interactions that converge to produce the
complement the hypertrophic response to resistance training291. heterogenous molecular responses that are observed systemically19
and in muscle177 after a common exercise bout.
Endurance and high-intensity cardiorespiratory exercise Interindividual difference in exercise trainability has led
Cardiorespiratory exercise has numerous subtypes that are to the concept of ‘responders’ and ‘non-responders’. Increasing
associated with training-induced improvements in maximal oxygen the volume of fixed-intensity exercise301,302 or the intensity of

Nature Reviews Molecular Cell Biology

Review article

(continued from previous page)

fixed-volume exercise302 can somewhat attenuate exercise As the field of sports genetics continues to grow and be refined, it
non-response, with larger volumes of higher-intensity exercise should provide actionable knowledge for the efficient personalization
perhaps doing so most effectively302. However, lower responders of training programmes. In the meantime, it is important to emphasize
.
still require a greater training stimulus and time commitment that low responders for a given parameter, such as VO 2max, often
to achieve results comparable to those of more-responsive improve in other outcome measures299. Therefore, exercise remains
trainees301,302. beneficial for all who are able to partake.

training status — interact to determine the full spectrum of muscle The myofibrillar matrix
characteristics. The sarcomere is the basic contractile unit of skeletal muscle, and
in-series assembly of sarcomeres forms myofibrils. Although myofi-
Hybrid fibre types brils were initially viewed as single, tube-like structures organized in
Most canonical myonuclei within the same fibre display coordinated parallel within fibres, later evidence suggested that myofibrils instead
transcription of a single Myh isoform in mouse muscle26,27. However, a form branching networks that could act in tandem with the cytoskel-
minority of fibres are hybrid26,27 (reviewed in ref. 38), containing myo- eton to facilitate lateral-force transmission50. Using focused ion beam-
nuclei that express two or more Myh pre-mRNA genes in the same scanning electron microscopy, recent studies have built upon this
nucleus26,29 and/or in different nuclei across the fibre length26. Thus, hypothesis showing that myofibrils indeed form a nonlinear lattice of
the regional distribution of MyHC can vary between muscle biopsy sarcomeres51–53 connected across the width and length of the muscle cell
sites39 and along a single fibre26,40. through three branching subtypes51 (Supplementary Fig. 2a). In mice,
In human vastus lateralis muscle, <10%22,41 to 40%38 of fibres can be the frequency of sarcomere branching decreases during early-to-late
hybrid types. ‘True’ non-transitioning or non-regenerating hybrid fibres postnatal development, but increases again in adult slow-oxidative
have metabolic enzyme22 and single-fibre contractile (force-velocity soleus fibres51. In comparison to soleus muscle, fast-glycolytic gastro­
producing)25,42 properties between those of their co-expressed MyHC cnemius fibres exhibit a different branching morphology (preferring
isoforms, providing further functional nuance to the slow-oxidative myofilament transfer over sarcomere splitting) and utilize fewer total
to fast-glycolytic continuum in the following order of slowest and sarcomeric connections51. Inducing sarcomere connections, such as
most oxidative to fastest and most glycolytic: type I → I/IIA → IIA → IIA/ through gene manipulation in Drosophila, reduces the myofibril cross-
IIX → IIX22,25,42. In adult mice, hybrid fibres are most common in the sectional area52. Thus, myofibrillar matrix assembly appears specific
slow-twitch soleus26,27, and the abundance of hybrid fibres is not altered to functional demand, and its organization in fast-twitch muscle may
by denervation in this muscle group26. By contrast, sciatic nerve tran- support the greater size22 and power24 of these fibres.
section or deletion of Six1 — a gene encoding the transcription factor The relevance of a nonlinear network of sarcomeres is implied
homeobox protein SIX1 driving the fast-glycolytic phenotype in by its structure — providing an elegant mechanism for both longitu-
muscle43 — increases hybrid fibre content in the typically fast-twitch dinal and lateral force transmission from muscles to bones through
extensor digitorum longus26,27. Hence, anatomical position27, innerva- tendons50,51. A linked configuration of myofibrils could also minimize
tion26,27 and transcription factor profiles26 might coalesce to coordinate the impact of localized sarcomere damage in muscle and thereby
Myh expression in muscle. increase the robustness of the contractile machinery across an entire
For a given muscle group, a greater proportion of hybrid and pure fibre. Further interrogation of the function and regulation of the
type IIX fibres seems indicative of sedentary behaviour41. Alterna- myofibrillar matrix in health, disease and exercise adaptation will
tively, exercise tends to reduce hybrid fibre content41,44,45, promoting a be enlightening.
shift away from type IIX fibres towards slower myosin types13,39,41. Con-
sequently, pure type IIX fibres are rare in humans22,24,39,41 and account Mitochondrial complexity in muscle
for <1% of the vastus lateralis fibre pool in healthy individuals40. As Sustained muscle contraction requires a continuous supply of ATP,
discussed in Supplementary Box 1, Myh expression in mouse muscle which is mostly derived from mitochondrial oxidative phosphoryla-
is regulated by competitive promoter–enhancer interactions27,46. tion (OXPHOS) during submaximal54 or longer-duration high-intensity
Exercise epigenetically modifies chromatin accessibility (see the exercise55 (see below; Fig. 2). Depending on fibre type, around 2–10%56
section ‘Skeletal muscle responses to acute exercise’), and resistance of muscle volume is filled by distinct subpopulations of subsarcolem-
exercise increases MyHC-specific protein synthesis47. Yet, it is still mal (also known as peripheral) and intermyofibrillar mitochondria,
unclear how established regulators of fibre type switching, such as differing in structure, function and localization56–59 (Fig. 1).
peroxisome proliferator-activated receptor-γ coactivator 1α (PGC1α, Subsarcolemmal mitochondria often cluster together in the
also known as PPARGC1α)48 and nuclear factor of activated T cells, sarcoplasmic space between myofibrils and the sarcolemma56–59,
cytoplasmic 1 (NFATc1)49, combine with Myh-promoter–enhancer and they dedicate more of their volume to cristae (folds of the inner
dynamics to confer physical activity-dependent fibre type transitions. membrane) and matrix57,58. These globular mitochondria extend deep
Further single-myonuclei RNA and chromatin profiling, alongside into the myofibrillar space and physically join with adjacent intermy-
isolated fibre spatial transcriptomic and proteomic approaches, ofibrillar mitochondria through electron-dense intermitochondrial
should extend the understanding of how gene expression and cel- junctions57,58, forming the mitochondrial reticulum56–59 (Fig. 1b,c).
lular phenotype are regulated among the many nuclei of a syncytial Compared to subsarcolemmal mitochondria, intermyofibrillar mito-
muscle fibre. chondria are more complex in morphology59 and physically interact

Nature Reviews Molecular Cell Biology

Review article

Glossary

Adrenoceptor Maximal oxygen consumption Non-esterified fatty acids Proton-motive force

.
Transmembrane G-protein-coupled (VO2max). The maximum volume of (NEFAs). A metabolic substrate The proton electrochemical gradient
adrenergic receptors (GPCRs) that oxygen (ml kg−1 min−1) that can be inspired utilized by muscle at rest and in an in mitochondria consisting of an
mediate the actions of the endogenous and utilized during exhaustive exercise, intensity-dependent manner during electrical charge gradient (also known
.
catecholamines adrenaline and such that the value (VO2) plateaus exercise. as the ‘membrane potential’) and a
.
noradrenaline. There are nine subtypes despite increasing workloads. VO2max is pH gradient. The proton-motive force
of adrenoceptors: α1A, α1B, α1D, α2A, a measure of aerobic or cardiorespiratory Peak-twitch torque is generated by the proton-pumping
α2B, α2C, β1, β2 and β3. The α2A and α2C fitness and is commonly used to The force produced by muscle (through action of respiratory complexes across
adrenoceptors regulate presynaptic standardize exercise intensity for clinical a moment arm) evoked by a single the inner mitochondrial membrane
.
neurotransmitter release from central trials (for example, x% of VO2max). electrical stimulation from, for example, and couples substrate oxidation to
adrenergic and peripheral sympathetic applied electrodes. ATP generation.
nerves. Mitophagy
A specific form of lysosome-dependent Phosphagen system Transverse tubules
Ergogenic catabolism (autophagy), through A rapid energy-producing (T-tubules). Invaginations in the
A performance-enhancing effect. which damaged mitochondria are pathway comprising the ATP sarcolemmal membrane that insert
selectively removed. Mitophagy of regenerating adenylate kinase between myofibrils. T-tubules
Hypertrophy the mitochondrial reticulum has an (ADP + ADP ⇌ ATP + AMP) and tightly associate with two terminal
Typically refers to an increase in the essential role in maintaining cellular creatine kinase (CrP + ADP ⇌ ATP + Cr) cisternae (calcium-releasing regions)
cross-sectional area (or radial growth) energy homeostasis. reactions. Of these reactions, creatine of the sarcoplasmic reticulum,
of muscle fibres, resulting in gains of kinase has a greater capacity for forming the ‘triads’, which are
skeletal muscle mass in response to Muscle spindles ATP resynthesis in muscle due to essential for excitation–contraction
mechanical loading activities, such as Structures embedded in most the availability of creatine phosphate coupling.
resistance exercise. mammalian skeletal muscles that stores.
continuously relay proprioceptive Voluntary force production
Ketone body information regarding muscle length Proprioceptive The conscious or ‘voluntary’ production
A lipid-derived, water-soluble, organic and movement to the central nervous Able to sense intrinsic information of muscle force (in other words, not
compound produced in the liver that system. Muscle spindles consist of regarding bodily position and triggered by exogenous electrical
can be used as an alternative energy intrafusal muscle fibres enclosed within locomotion. The primary proprioceptive stimulation).
source by extra-hepatic tissues — a capsule layer and are distinct from the sensory organ of the body is the muscle
predominantly the brain, but also heart extrafusal muscle fibres discussed in spindle.
and skeletal muscle. this Review.

with the myofibrillar matrix53,56,57, sarcoplasmic reticulum (SR) and generation for subsequent transfer through the reticulum, into the
intermyofibrillar lipid droplets53,56,58 (Fig. 1b). intermyofibrillar mitochondrial network57. Here, intermyofibrillar
In humans53,59,60 and mice53,56, intermyofibrillar mitochondrial mitochondria, possessing higher ATP synthase (ETC complex V)
(sub)networks are uniform, but their organization varies between expression57 and higher surface area-to-volume ratios59, could utilize
muscle types56,60. In fast-glycolytic extensor digitorum longus fibres this potential energy to support rapid ATP production and diffusion
of mice, intermyofibrillar mitochondria wrap around the I-bands of to myofibrillar ATPases56. Hence, much like an electrical power grid,
sarcomeres, perpendicular to the contraction axis, whereas slow- the mitochondrial reticulum may serve to efficiently disperse energy
oxidative soleus fibres contain larger subnetworks of connected inter- across the fibre56–58.
myofibrillar mitochondria, arranged in grids that surround myosin Within mitochondria, ETC complexes form higher-order struc-
like a cage56. The positioning of mitochondria in the intermyofibrillar tures termed ‘supercomplexes’. The maximal oxygen consumption of
space also directly impacts the structure of adjacent sarcomeres53. permeabilized fibres was positively correlated with the preferential
A greater proportion of mitochondria at the Z-disc reduces the cross- redistribution of complexes III and IV into supercomplexes in the
sectional area of sarcomeres in this region by bending (or ‘curving’) muscle of elderly individuals after 4 months of endurance training61.
peripheral myosin filaments, causing heterogenous myosin–myosin Alternatively, 6 weeks of high-intensity exercise improved muscle
spacing along the sarcomere length53. However, any detrimental impact mitochondrial respiration and ETC enzymatic activity in the absence
on fibre contractility might be offset by the relative preservation of of supercomplex alterations in young adults15. Supercomplexes
cross-sectional area and myofilament spacing towards the sarcomere are clearly important structural features of the ETC, but whether
centre (M-line)53 (the structure of an individual sarcomere is shown in they support exercise adaptation beyond their stoichiometric rela-
Supplementary Fig. 2d). tionship with mitochondrial abundance is yet to be determined.
Subsarcolemmal mitochondria are proximal to nutrient-delivering Furthermore, the assembly and stability of supercomplexes depends
capillaries56–59 and have a greater abundance of proton-motive force on the integrity of mitochondrial cristae62,63, which can be regulated
driving electron transport chain (ETC) complex IV57. As such, this by SR stress-induced signalling based on eukaryotic translation initia-
subpopulation is thought to specialize in membrane potential tion factor 2α (EIF2α) kinase 3 (PERK, also known as EIF1AK3), EIF2α

Nature Reviews Molecular Cell Biology

Review article

a Subsarcolemmal
Capillary lipid droplet Sarcolemma
Subsarcolemmal
glycogen
Intermyofibrillar
mitochondria

Intermyofibrillar
lipid droplet
Intramyofibrillar
glycogen
α-motor
neuron Actin

Myonucleus Myosin
Myofibril

Muscle fibre
Sarcoplasmic
reticulum
Transverse
Subsarcolemmal
tubule
mitochondria
Intermyofibrillar
glycogen

b Fibre type (gene) Mixed muscle fibres Mitochondrial network Coupled mitochondria Organelle proximity
I (MYH7) SR LD LD
IIA (MYH2)
OMM OXPHOS
IIX (MYH1)
IMS H+ H+ H+
Myofibril IMM CoQ C
I III IV V
NADH e– II e– O2
SR ATP
IMF NAD+ 2H2O
FADH2 H+ ADP + Pi
FAD

Matrix

Intermitochondrial
junction

c Shared depolarization between

damaged mitochondria
Propagation of dysfunction
Healthy mitochondria If IMJ remains open throughout coupled
Damaged mitochondria mitochondrial networks
Open IMJ

Electrically coupled mitochondria Local dysfunction

Damage Retraction and

Open IMJ Open IMJ targeting for
Electrical separation Physical separation mitophagy

Closure of IMJ contains

local dysfunction
Closed IMJ

Recovery of healthy
mitochondrial networks

Nature Reviews Molecular Cell Biology

Review article

Fig. 1 | Skeletal muscle fibre ultrastructure. a, Location of mitochondrial three main fibre types14,21,22,24,36, type I (marked by MYH7 expression), type IIA (with
subpopulations and energy stores in muscle fibres. Skeletal muscle is MYH2 expression) and type IIX (expressing MYH1). Differences in mitochondrial
composed of layers of connective tissue and fascicles (also known as muscle protein content14,21 and mitochondrial network configuration56,60 between fibre
bundles). Fascicles contain organized arrangements of individual syncytial types directly impacts muscle metabolism and function. Muscle mitochondria
muscle fibres, each covered by an endomysium, or basal lamina, which is form an interconnected reticulum56–60 that enables swift and efficient
anchored to the fibre membrane (also known as the sarcolemma). Muscle distribution of potential energy from subsarcolemmal (also known as peripheral)
stem cells, termed ‘satellite cells’, reside within this sarcolemma-basal lamina mitochondria to intermyofibrillar mitochondria (IMF), deep within the fibre56–58.
‘niche’ (Supplementary Fig. 3). Specialized components, such as sodium/ The positioning of mitochondria in the intermyofibrillar space influences the
potassium pumps (Na+/K+-ATPase), triads (consisting of transverse tubule and structure of adjacent sarcomeres, resulting in variable cross-sectional areas
sarcoplasmic reticulum (SR)) and proteins of the myofibrils (long arrangements and myofilament spacing at different regions across the sarcomere length53.
of connected sarcomeres) enable fibre contraction through the process of The branching morphology of IMF also accommodates functional interactions
excitation–contraction coupling and sliding filament theory (Supplementary with nearby cellular components, such as the sarcoplasmic reticulum and
Fig. 2). Free ATP in muscle is limited69,70, and fibres possess additional energy intermyofibrillar lipid droplets53,56,58. In oxidative mouse muscle, ~20% of all IMF
depots to maintain contractile activity, including creatine phosphate, glycogen are connected to lipid droplets, which may facilitate efficient ATP production
and intramyocellular lipids (Box 3). Glycogen granules are nonuniformly and distribution56. c, Adjacent mitochondria form networks and share energy
distributed between intramyofibrillar, intermyofibrillar and subsarcolemmal potential through the intermitochondrial junction (IMJ). Analogous to circuit
pools35,267,268. Alternatively, intramyocellular lipids are stored in lipid droplets breakers, intermitochondrial junctions split the reticulum into smaller
(LDs) found predominantly at central (intermyofibrillar) but also peripheral subnetworks, permitting swift separation of defective mitochondria before
(subsarcolemmal) regions within healthy muscle fibres76,78. During submaximal54 their removal through mitophagy58. In this way, intermitochondrial junctions
and longer-duration high-intensity interval55 exercise most ATP in muscle is provide a dynamic layer of quality control, rapidly rewiring the mitochondrial
regenerated by mitochondrial oxidative phosphorylation (OXPHOS) (see the reticulum through healthy network components to sustain muscle function58.
section ‘Acute exercise muscle metabolism’) (Fig. 2). b, Spatial distribution of C, cytochrome c; CoQ, coenzyme Q; IMM, inner mitochondrial membrane;
the mitochondrial reticulum within muscle fibres. Human muscle comprises IMS, intermembrane space; OMM, outer mitochondrial membrane.

and activating transcription factor 4 (ATF4) (PERK–EIF2α–ATF4)63. Muscle substrate utilization is exercise intensity-dependent
Endurance-trained athletes have greater cristae density in mitochon- Muscle fibres are part of a functional ‘motor unit’ comprising an
dria of type I fibres64, and PERK and ATF4 proteins are increased in α-motor neuron and the muscle fibres innervated by its axon. The force
muscle ~48 h after acute resistance exercise65. However, cristae struc- generated by a muscle depends on both the number of activated motor
ture was unchanged after 10 weeks of moderate-intensity endurance units — and thus fibres — and the rate at which motor units discharge
training in sedentary individuals with obesity64. Thus, further study action potentials once recruited (known as rate coding) (reviewed in
is required to distinguish the impact of exercise on mitochondrial ref. 73). Stimulation of motor units conforms to the size-orderly princi-
cristae remodelling. ple of recruitment, such that smaller units are activated first, followed
Moving forward, better understanding of mitochondrial net- sequentially by larger units as contraction intensifies73. Consequently,
works, subpopulations and ETC configurations could benefit strate- the higher power outputs achieved during progressively demanding
gies to combat age-associated decline in muscle function (Box 2) and physical activity relies on the stimulation of a greater number of motor
mitochondria-related diseases. Focus should be directed towards the units, a larger proportion of the muscle fibre pool, and therefore the
heterogenous mitochondrial populations in human muscle, which are recruitment of more type II fibres.
smaller than those in mice59. ATP consumption per unit of time is ~2.5–4-fold higher in type II
fibres than in type I fibres74, and the maximal rate of ATP resynthesis
Acute exercise metabolism in skeletal muscle is fastest through oxygen-independent (anaerobic) versus oxygen-
Relative to weight, the basal thermogenesis of muscle is lower than that dependent pathways, in the following order of descending speed:
of most other organs66 because myosin is maintained in disordered anaerobic phosphagen system (including adenylate kinase and creatine
relaxed and super-relaxed states characterized by slow and extremely kinase (CKM)) and glycolysis reactions provide the fastest supply of
slow ATP kinetics, respectively67 (Supplementary Figs. 2b,c). Upon ATP, followed by carbohydrate oxidation and finally NEFA oxidation.
contraction, mechanosensing rapidly initiates myosin conformational Human type II fibres are hence enriched with creatine phosphate (CrP)
change from relaxed to active68, and muscle ATP consumption increases and glycogen energy depots75 and contain higher levels of adenylate
dramatically during short-duration exhaustive exercise69,70. As free ATP kinase21, glycogenolysis and glycolysis metabolic machinery14,21. Con-
in muscle (~20–25 mmol per kg dry mass) is only sufficient to sustain versely, type I fibres are more abundant in peroxisomes14, mitochon-
maximal exercise for <2 s (refs. 69,70), continued contractile activity dria14,21,76,77 and intramyocellular lipids (IMCLs)76–78, consistent with
requires ATP resynthesis from a combination of intramuscular energy their slower ATP turnover74 (Supplementary Fig. 1b).
stores (Fig. 1a and Box 3) and circulating substrates, such as glucose Together, this indicates that the contribution of specific energy
and non-esterified fatty acids (NEFAs)71,72 (Fig. 2). systems and substrates to working muscle is mainly a function of the
In this section we detail the metabolic responses that enable neuromuscular activation required to match the intensity of the exer-
muscle to match the considerable demands of acute exercise. We also cise being performed. This understanding is envisaged within the
address how this substrate–energy pairing facilitates the integral role ‘crossover concept’ (reviewed in ref. 79), which describes the larger
of muscle in exercise-mediated inter-organ communication (Box 4) relative contribution of NEFA oxidation towards whole-body energy
and touch upon the interaction between exercise metabolism and expenditure at low-to-moderate exercise intensities, with an incremen-
biological rhythms. tal and necessary ‘switch’ towards preferential (oxygen-dependent and

Nature Reviews Molecular Cell Biology

Review article

Box 2

Effects of ageing on skeletal muscle and the benefits of exercise

The loss of total muscle mass303, fibre number34 and type II fibre genome314, and long-term mixed-modality endurance-type exercise
size34,36,250 becomes most evident at ≥50 years of age34,303 and might attenuate methylation events in the promoter regions of
is inherently linked to perturbations in muscle metabolism. important cytoskeletal, sarcomeric, glycolysis, glycogen synthesis
Mitochondrial content is lower in type I and type II fibres from older and tricarboxylic acid cycle-related genes312. These epigenetic
individuals than from younger individuals36, and diminished muscle events could combine to preserve muscle contractile and metabolic
oxidative capacity is linked to mobility decline among adults ≥60 years integrity. Furthermore, exercise training-induced changes in plasma
of age84,304. Glycolytic enzymes and chaperone proteins of myofibrillar apelin (APLN) positively correlated with chair stand, balance and
quality control are also decreased in type II versus type I fibres from walking (Short Physical Performance Battery) test scores in elderly
individuals >65 years of age36. Together with an age-dependent women and men315. This suggests that factors produced from
decline of MYH1 and MYH2 mRNA expression (encoding myosin exercising muscle help to mediate the protective effects of physical
heavy chain IIX isoform (MyHC-IIX) and MyHC-IIA, respectively)47, activity against age-related functional decline (Box 4). APLN released
this disconnect between muscle metabolic and contractile apparatus from muscle during exercise can act in an autocrine manner to
could contribute towards the reduced type II fibre size34,36,250 augment satellite cell-mediated repair, mitochondrial abundance and
that predominantly underlies detriments in muscle force and muscle oxidative capacity315, and/or through paracrine mechanisms
power with advanced age305 (reviewed in ref. 306). Older humans to stimulate endothelial cell expansion316 that could enhance
and aged mice can further accrue subsets of senescent muscle hypertrophy of type II fibres248,250 (Supplementary Fig. 3).
fibres307 and resident mononuclear cells (for example, satellite308, Endurance61, high-intensity interval and concurrent12 training all
.
myeloid307–309 and fibroadipogenic progenitor307,308 cells). The number increase maximal oxygen consumption (VO 2max) and muscle
of senescent cells is typically low in resting muscle but increases after mitochondrial content in older individuals. Reciprocally, resistance
resistance exercise310 and injury308,309 irrespective of age. The defective training improves strength12,240,317 and fat-free mass12 but to a lesser
clearance and potential accumulation of senescent cells during extent than seen in young adults240. The attenuated muscle anabolic
ageing308,309 might impair muscle regeneration308,309, hypertrophy response with age could be due to blunted ribosomal biogenesis240,
(especially of fast-twitch glycolytic fibres)310, strength307,311 and reduced activity of amino acid sensors (such as leucyl tRNA
maximal mitochondrial respiration311, in part through a Cdkn1a-driven synthetase (LARS))318 and lower systemic production and local tissue
transcriptional programme307,311. sensitivity to hormones such as testosterone and insulin-like growth
Exercise can mitigate several aspects of muscle ageing and factor 1 (IGF1)240 (Supplementary Table 1). Progressive resistance
improves systemic insulin sensitivity12. Lifelong endurance exercisers training programmes can further benefit hip and femur bone mineral
have a greater density of muscle mitochondria312,313 and a more density in people ≥65 years of age317, and incorporating power-type
complex and connected mitochondrial reticulum313 relative to resistance exercise (involving explosive concentric movements) may
less-active elderly counterparts. This is coincident with higher be superior for physical performance outcomes, including ‘get-up-
protein levels of inner mitochondrial membrane fusion-factor optic and-go’ and chair stand tests319. Collectively, these findings
atrophy 1 (OPA1)313. Additionally, resistance training was shown emphasize the importance of maintaining a diverse physical activity
to offset age-related CpG-site methylations in the mitochondrial profile across the lifespan.

independent) carbohydrate utilization during exercise at higher levels cycle substrates (pyruvate and glutamate)82 and precursors (acetyl
of mechanical effort71,72. groups)83–85; and sarcoplasmic buffering of the interrelated allosteric
metabolites creatine and ADP86 (Fig. 2a). Indeed, pharmaceutical activa-
Metabolic inertia at the onset of exercise tion of PDH increased muscle acetyl-carnitine availability and reduced
Although the lower metabolic cost of rest72 and light exercise71 is mainly the reliance on phosphagen and glycolytic energy pathways during
fuelled by the oxidation of circulating NEFAs, further upregulation of acute ischaemic exercise85. The ability of previous sprinting intervals to
OXPHOS ATP provision is delayed at the onset of moderate-to-high prime OXPHOS in subsequent high-intensity work bouts could also occur
intensity physical activity, and a larger oxygen-independent contri­ through better coupling of oxidative substrate delivery to mitochondrial
bution fuels muscle in the first ≤30–60 s of exercise54,80. This is despite enzymatic activity, although carryover of residual fatigue probably
sufficient blood flow54,80 and adequate intramuscular oxygen levels for plays an additional role in this scenario by impeding power output69,70.
maximal mitochondrial respiration to occur (≥0.5–2 mmHg) (reviewed
in ref. 81). Oxygen-independent exercise metabolism
The lag in oxygen-dependent metabolism at the beginning of exer- During short-duration maximal efforts, accelerated muscle ATP demand
cise may stem from a combination of linked temporal factors, including is mostly met anaerobically, through rapid stimulation of the phos-
activation of the mitochondrial matrix enzymes pyruvate dehydroge- phagen and glycolytic energy systems69,70. The simplicity and proximity
nase (PDH), 2-oxoglutarate dehydrogenase (OGDC)82 and carnitine of the sarcoplasmic CKM reaction (Box 3) allows CrP to supply equimo-
acetyltransferase (CRAT)83,84; the availability of tricarboxylic acid (TCA) lar amounts of high-energy phosphate to ATPases within milliseconds87,

Nature Reviews Molecular Cell Biology

Review article

which is essential for sprint performance69,70. CrP hydrolysis peaks at the ROS signalling is implicated in adaptations to endurance and
onset of contraction but declines within less than 6 s, and CrP stores can resistance training (reviewed in ref. 105). High-intensity intervals aug-
be >90% depleted after around 30 s of intense exercise70 compared to ment NOX4 mRNA in human vastus lateralis ~3–4 h after exercise, and
.
only ~20% after 10 min of cycling at ~50% of VO 2max (ref. 88). a retrograde NOX4–ROS-axis led to activation of transcriptional
The initial increase in glycogen breakdown is due to posttransla- regulators — such as nuclear factor erythroid 2-related factor 2 (NRF2,
tional modification of glycogen phosphorylase (PYGM) from a less- also known as NFE2L2) and PGC1α — required for mitochondrial bio-
active b form to the constitutively active a form by phosphorylase genesis and the endogenous antioxidant defence response to exercise
kinase (PHK)70. Transients of Ca2+ and AMP are probable regulators training in mouse muscle104 (Fig. 3). Accordingly, blunting oxidative
of PHK and thus the switch to PYGMa (reviewed in ref. 89). AMP can stress through Nox4 deletion in mice104 or high-dose vitamin C and E
simultaneously increase PYGMb activity to support maximal rates of intake in humans106 attenuates the insulin-sensitizing effects of endur-
glycogenolysis70, and the systemic rise of adrenaline during sprint inter- ance training. Still, any potential detriment of antioxidant (such as
.
vals69 or endurance exercise71,90,91 may help to stabilize the PYGMa con- vitamin C and/or vitamin E) supplementation to gains in VO 2max , lean
formation92. Downstream of glycogen and glucose, allosteric regulation body mass or human endurance and strength performance seems
and replenishment of nicotinamide adenine dinucleotide (NAD+) can relatively minor107.
stimulate the rate-limiting enzymes phosphofructokinase (PFK) and
glyceraldehyde-3-phosphate dehydrogenase (GAPDH), respectively, Glucose uptake and carbohydrate oxidation in exercising muscle.
to promote continued flow through glycolysis (Fig. 2c). When workload71,72 and (to a lesser extent) duration71 of moderate-
Average power output drops substantially over the final half of intensity exercise increases, so does the contribution of blood glucose
a 30-s all-out physical effort, corresponding to the pattern of PYGM towards total carbohydrate utilization, although energy provision from
activity (reversion from a to b) and reduced substrate-level phospho- intramuscular glycogen is always higher71,72. Muscle glucose uptake
rylation70. This muscular fatigue might be driven by alterations to the reaches ~15-fold resting levels at the cessation of non-fatiguing exercise
.
intracellular metabolic milieu. Above a critical threshold of ~15 mmol l−1 (~50% of VO 2max) and ~50-fold resting levels at the cessation of exhaus-
.
(ref. 93), inorganic phosphate can enter the SR and precipitate with tive endurance exercise (~100% of VO 2max )88. This increased rate of
Ca2+, thereby impairing Ca2+ release and muscle contraction94. Addi- blood glucose extraction is supported by greater hepatic glucose
tionally, owing to the high activity and near-equilibrium state of lactate output108, a ≥5–10-fold rise in blood flow88, and enhanced perfusion of
dehydrogenase (LDH), the end product of glycolysis is always lactate95. muscle capillaries109. As exercise intensifies, these circulatory and
After ~3.5–4 min of exhaustive cycling, muscle lactate reaches levels microvascular responses collectively maintain relatively constant
more than twofold higher than resting concentrations55, more than plasma–to–interstitial glucose concentrations110 and enlarge the area
fourfold higher following acute resistance exercise96, and ≥8-fold97 available for nutrient exchange to occur109. Such events further serve
to 30-fold69 higher after 10 rounds of short-duration sprinting. The to promote the release of discrete biologically active molecules from
accumulation of lactate (a strong anion) encourages dissociation of muscle20. As discussed in Box 4, many of these muscle-derived
water to HO− + H+, and muscle pH can reach ≤6.5 during strenuous exer- ‘exerkines’ or exercise-induced ‘myokines’ are implicated in aspects of
cise bouts55. Lower intramuscular pH could diminish PYGM activity70, local and systemic exercise adaptation (reviewed in ref. 111).
and H+ may act on group III (mainly mechanosensitive) and IV (mainly The exocytosis and fusion of glucose transporter 4 (GLUT4)-
metabosensitive) muscle afferents in the interstitial space93. Once containing vesicles with the sarcolemmal membrane and T-tubules is
activated, these sensory neurons can feed back, potentially through essential for contraction-stimulated glucose transport into muscle112.
inhibitory γ-aminobutyric acid type B (GABAB) receptors98, to reduce During exercise, translocation of GLUT4 is regulated by a combination
motor cortex excitability and suppress motoneuronal output93. The of Ca2+, metabolic stimuli and mechanosensitive stimuli that converge
severity of peripheral fatigue correlated with the extent of quadriceps on calcium/calmodulin-dependent protein kinase II (CaMKII), 5′-AMP-
activation during self-paced time-trial cycling99. However, mechanisms activated protein kinase (AMPK) and RAS-related C3 botulinum toxin
of exercise-induced peripheral and central fatigue are complex, and the substrate 1 (RAC1) (reviewed in ref. 113), with redundancy between
major contributory factors are probably specific to modality, intensity pathways113,114. The role of RAC1 in contraction-induced glucose uptake
and duration (reviewed in ref. 94). might be mediated in part by downstream NOX2 activation101 but seems
independent of AMPKα2 (ref. 114) during submaximal treadmill running
Oxygen-dependent exercise metabolism in mice. Indeed, the catalytic α subunit of AMPK appears dispensable
Reactive oxygen species production during exercise. In contrast to for exercise-stimulated glucose transport in vivo115,116. Instead AMPKα
maximal sprinting for ≤30 s (refs. 69,70), high-intensity cycling lasting plays a more notable role in post-exercise substrate metabolism117
≥3 min derives >70% of energy from OXPHOS, with a post-exercise and insulin sensitivity116 through the upregulation of pyruvate dehy-
muscle metabolome enriched for pathways using pyruvate, long-chain drogenase kinase 4 (PDK4)117, the promotion of RAS-related protein
NEFAs and amino acids such as alanine, arginine and glutamate55. This RAB8A-perilipin 5 (PLIN5) lipid droplet–mitochondrial tethering118 and
upregulation of OXPHOS reduces mitochondrial superoxide (H2O2) the phosphorylation of TBC1 domain family member 1 (TBC1D1)115
emission100,101 from ETC complexes I–III100 and thus mitochondria and TBC1D4 (also known as AS160)116.
contribute minimally to reactive oxygen species (ROS) generation For glucose to be oxidized in muscle, lactate — simultaneously
in contracting muscle102. Rather, nicotinamide adenine dinucleotide produced from glycolysis and taken up through sarcolemmal
phosphate (NADPH) oxidases (NOX), located in the sarcolemma and monocarboxylate transporters (MCT1 and MCT4)81 — must first
in transverse tubules (T-tubules)102 and the sarcoplasmic reticulum103, be converted to pyruvate by sarcoplasmic or mitochondrial LDH
appear to be the predominant sources of ROS during contraction101,102. (reviewed in ref. 119) (Fig. 2c). Glucose oxidation is higher in men than
Specifically, NOX2 (ref. 101) and NOX4 (ref. 104) are indispensable for in women (Supplementary Box 2) and rises with exercise intensity,
exercise-stimulated sarcoplasmic ROS production in mice. dietary carbohydrate and muscle glycogen levels, and peri-exercise

Nature Reviews Molecular Cell Biology

Review article

VLDL1 Flow
Capillary
NEFA-aIb
LPL
a Action
Glucose Interstitium
potential b
FATP, FABPpm, CD36 T-tubule
c Sarcolemma GLUT4/1 MCT4/1
SERCA Ca2+ HK PHK
Sarcomere b a
HSL PYGM
RYR1 G-1-P
ATGL Glycogenintra
IMFLD G-6-P G-1-P
ATPases ATP Glycolysis ROS O2•–
DHPR Carnitine Glycogeninter
tetrad shuttle
G3PS NAD+
LCaCoA + NAD+ Ca2+ O2
Sarcoplasmic DHAP2– sG3PDH 2
↓Carnfree +
reticulum NADH + Pyr– sLDH La− NOX2
T-tubule sMAS SERCA Ca 2+

Asp2–/
Ca2+ 2OG2–
CPT1B
ACSL1 VDAC VDAC RYR1

4
G3P2– OXPHOS Mal2–/ DHPR
LCaCarn 1 Sarcoplasmic
H+ H+ H+ Glu2– tetrad
mG3PDH Pyr– reticulum
MOE/
CoQ C AGE
MCU CACT I III IV V MPC
mLDH
– II –
CPT2 NADH e e O2
ATP CrP shuttle Na+/K+
Ca2+ PDP NAD+ 2H2O
FADH2 H+ ADP + P Cr pump
3 mCKM +
– + P FAD i
ADP
– + LCaCoA CrP sCKM ATP
PDH ↓CarnFree NAD + mMAS +
– + ANT ADP
– + ATP
PDH NADH aCarn
– + β-Ox CRAT
FADH2 + CoA NADH PDH NAD + Sarcomere
CRAT activity
ANT
aCoA Glycogenintra
ATPases
OGDC TCA CS Energy
cycle ADP + Pi ↑[ADP][Pi]:[ATP]
IDH NADH
FADH2 ↓[NAD+]:[NADH]
CO2 Mitochondrion Matrix IMS Redox
↑ Glucose metabolic flux
Impaired fatty acid transport capacity
Preferential utilization of specific glycogen depots

carbohydrate intake120. Intra-workout consumption121 or ‘mouth- from lipid droplets is almost entirely dependent125 on the redundant
rinsing’122 of exogenous carbohydrates can also benefit exercise per- enzymes125,126 adipose triacylglyceride lipase (ATGL) and hormone-
formance. Whereas the ingestion of carbohydrates feasibly provides sensitive lipase (HSL). The recruitment of ATGL to PLIN5 at lipid drop-
metabolizable substrate120 and spares liver glycogen stores during lets could be enhanced by AMPK-regulated assembly of the
bouts of longer−duration exercise108, the potential ergogenic effect RAB8A–PLIN5 tethering complex118. Alternatively, exercise upregulates
of mouth-rinsing is most probably achieved through oral receptor HSL in an intensity-dependent manner127 through additive contraction
afferents that signal centrally to increase voluntary force production123. (Ca2+–protein kinase C (PKC))128 and adrenaline (cAMP–PKA)91-
mediated pathways that promote its translocation to lipid droplets129.
Muscle lipid metabolism during exercise. The post-exercise serum These signalling events augment rates of muscle lipolysis, and IMCL
metabolome is distinct between exercise modalities. An acute bout contents of type I and type II fibres can be ~45% and ~20% depleted,
.
of endurance or resistance exercise differentially regulates discrete respectively, after ~60–180 min of cycling at ~50–75% of VO 2max
subclusters of metabolites across a 3-h post-exercise period, despite (ref. 130). Muscle HSL activity is transient during acute moderate-
similar trends for metabolome recovery over time124. Of note, amino intensity exercise91 and its downregulation — through negative feed-
acid, nucleotide and carbohydrate (for example, lactate and pyru- back from AMPK91 or allosteric metabolites (for example, long-chain
vate) signatures are prominent after resistance exercise, compared to fatty acyl-CoA) — mirrors the greater relative contribution of circulating
the lipid-derivative enrichment (for example, various acyl-carnitines NEFAs towards whole-body energy expenditure71.
and the ketone body β-hydroxybutyrate) after endurance exercise124. Circulating NEFA availability is increased during longer bouts of
This illustrates unique physiological challenges posed by specific physical activity at low-to-moderate intensity71 due to greater release
interventions. from adipose tissue and the improved affinity of triacylglycerol-rich
.
At intensities eliciting peak fat oxidation (~60–65% of VO 2max), the lipoproteins for hydrolysis by lipoprotein lipases (LPL)131 anchored
contribution of plasma NEFAs and IMCLs is ~1:1 and roughly equal to to the endothelial surface of interstitial capillaries. The utilization
total carbohydrate utilization71,72. In mouse muscle, liberation of NEFAs of plasma NEFAs is enhanced by contraction-induced sarcolemmal

Nature Reviews Molecular Cell Biology

Review article

Fig. 2 | Skeletal muscle metabolism during higher-intensity exercise. pyruvate (Pyr−) and/or lactate (La−) pass through voltage-dependent anion
a, Exercise-onset metabolic inertia (red area). Acetyl-carnitine (aCarn) channels (VDAC), where lactate is converted to pyruvate by mitochondrial
abundance83–85 and the acetyl coenzyme A (aCoA)-producing capacities of lactate dehydrogenase (mLDH) in the intermembrane space119 (step 1). Pyruvate
carnitine acetyltransferase (CRAT)83,84 and pyruvate dehydrogenase (PDH)85 then enters the mitochondrial matrix through the mitochondrial pyruvate
appear rate-limiting for oxidative adenosine triphosphate (ATP) regeneration at carrier (MPC)81,119. The glycerol-3-phosphate (G3P2−) shuttle (G3PS) and malate
the onset of moderate-high intensity exercise. Contraction-induced calcium (Mal2−)/aspartate (Asp2−) shuttle enables mitochondrial oxidation of lactate and
(Ca2+) transients promote mitochondrial Ca2+ uptake into the matrix space269 pyruvate through compartmentalized redox shuttling. G3PS and MAS recycle
through the inner-membrane mitochondrial calcium uniporter (MCU) extra-matrix nicotinamide adenine dinucleotide (NAD+) (step 2) and transport
complex270,271. Increased matrix Ca2+ can upregulate PDH272 through activation of reducing power from glycolysis to the mitochondrial matrix. This occurs
its phosphatase (PDP)273, and can upregulate isocitrate dehydrogenase (IDH)274 through reactions associated with Mal2− and Asp2− delivery into the matrix
and 2-oxoglutarate dehydrogenase (OGDC)272,274,275 directly to fine-tune space81,119 (step 3) and G3P2− donation of electrons directly to coenzyme Q (CoQ)
oxidative metabolism via stimulation of the tricarboxylic acid (TCA) cycle82. of the electron transport chain119 (step 4). As such, saturation of these shuttles
Ca2+ kinetics probably precede an allosteric86 rise in the [ADP][Pi] to [ATP] and increases lactate accumulation and upregulates the lactate-favouring LDHA
[creatine][Pi] to [creatine phosphate] ratios in part because ADP and creatine isoform in vitro278. See Box 3 for discussion of CrP, intermyofibrillar and
(Cr) are buffered by the adenylate kinase (not shown) and creatine phosphate intramyofibrillar glycogen (glycogeninter and glycogenintra, respectively)
(CrP) shuttle reactions. Thus, synchrony between mechanisms of substrate and intermyofibrillar lipid droplet (IMFLD) stores, and section ‘Acute exercise
provision, Ca2+-feedforward and metabolite feedback regulation might underlie muscle metabolism’ for details of their metabolism during acute exercise. β-Ox,
acute metabolic inertia. This could be particularly prominent in type II fibres, β-oxidation; ACSL1, acyl-CoA synthetase long-chain family member 1; AGE,
which have lower CRAT83 and MCU abundance21 and slower mitochondrial aspartate/glutamate exchanger; ANT, adenine nucleotide translocator; ATGL,
Ca2+ import rates276. Furthermore, metabolic inertia is more pronounced in adipose triacylglyceride lipase; b → a denotes posttranslational modification
metabolically compromised and older untrained adults, related to the lower of PYGM from its less-active b form to the constitutively active a form by PHK;
CRAT activity and acetyl-carnitine content of muscle in these individuals84. C, cytochrome c; CACT, carnitine acylcarnitine translocase; mCKM mitochondrial
b, Carbohydrates outcompete non-esterified fatty acids (NEFAs) for oxidation at creatine kinase muscle; sCKM, sarcoplasmic CKM; DHAP2−, dihydroxyacetone
higher intensities (yellow area). Muscle glucose uptake88 and carbohydrate phosphate; DHPR tetrad, four dihydropyridine receptors associated with one
utilization71,72,120 increases with exercise intensity. At workloads above maximum ryanodine receptor 1 (RYR1); FABPpm, fatty acid binding protein plasma
.
fat oxidation (>60–65% of maximal oxygen consumption (VO 2max))71,72, flux of membrane; FAD, flavin AD; FADH2, reduced FAD; FATP, FA transporter protein;
pyruvate to acetyl-CoA progressively exceeds rates of TCA cycle entry at citrate GLUT4/1; glucose transporters 4 and 1; G-1-P, glucose-1-phosphate; G-6-P,
synthase (CS), leading to depletion of the muscle free-carnitine (Carnfree) pool glucose-6-phosphate; HK, hexokinase; HSL, hormone-sensitive lipase; IMS,
through CRAT-dependent acetylation to acetyl-carnitine83. After higher- intermembrane space; LCaCarn, long-chain acyl carnitine; LPL, lipoprotein
intensity submaximal exercise, acetylation of the free-carnitine pool is greatest lipase; sLDH, sarcoplasmic LDH; mMAS, mitochondrial MAS; MCT4/1,
in type I fibres142. Insufficient free-carnitine availability would inhibit NEFA monocarboxylate transporters 4 and 1; mG3PDH, mitochondrial glycerol-3-
mitochondrial import at the first step of the carnitine shuttle — that is, carnitine phosphate dehydrogenase; MOE, malate/2-oxoglutarate exchanger; NADH,
palmitoyl transferase 1B (CPT1B) conjugation of carnitine to long-chain reduced NAD; NEFA-alb, albumin-bound NEFA; NOX2, NAD phosphate oxidase 2;
acyl-CoA (LCaCoA). Reduced fat oxidation is associated with diminished O2•− superoxide; 2OG2−, 2-oxoglutarate; OXPHOS, oxidative phosphorylation
.
free-carnitine levels at ~70% of VO 2max (ref. 72), whereas medium-chain NEFA and associated respiratory complexes; PHK, muscle phosphorylase kinase;
metabolism bypasses carnitine shuttling and is maintained at higher PYGM, glycogen phosphorylase muscle-associated; ROS, reactive oxygen
submaximal workloads277. Therefore, free-carnitine levels appear rate-limiting species; SERCA, sarcoplasmic/endoplasmic reticulum Ca2+-ATPase; sG3PDH,
for long-chain NEFA utilization at increasing exercise intensities. c, Lactate and sarcoplasmic glycerol-3-phosphate dehydrogenase; sMAS, sarcoplasmic MAS;
pyruvate oxidation and NADH shuttles (blue area). Downstream of glycolysis, T-tubule, transverse tubule; VLDL1, very low density lipoprotein 1.

enrichment of long-chain fatty acid transporters (fatty acid translocase and critical ACSL isoform in muscle140. As the inner mitochondrial
(FAT, also known as CD36), plasma membrane-associated fatty acid membrane is impermeable to long-chain NEFAs, long-chain fatty acyl-
binding protein (FABPpm) and fatty acid transport protein 1 (FATP1) CoA molecules are delivered into the mitochondrial matrix through
and FATP4)132, which have varying capacities for increasing NEFA the ‘carnitine shuttle’, where they undergo subsequent β-oxidation
uptake and oxidation in muscle133. Upon exercise, the exocytosis of and OXPHOS.
CD36 (and possibly other transporters) seems independent of AMPK134 Prolonged high-intensity exercise resulting in elevated rates of
but could involve calcium/CaMK kinase (CaMKK)135 and MAPK/ERK lactate production might directly impede IMCL lipolysis over time by
kinase 1 (MEK1) and MEK2 signalling136. Furthermore, although FABPpm downregulating cAMP–PKA signalling141. However, the glycolytic flux
is structurally identical to mitochondrial aspartate aminotransferase of higher-intensity exercise most probably outcompetes NEFAs for
(mAspAT), these proteins serve distinct functions within their respec- oxidation by depleting the muscle free-carnitine pool72,142. This limits
tive subcellular compartments (that is, FABPpm-mediated transport the capacity of the carnitine shuttle and thus impairs the mitochon-
of long-chain NEFAs across the sarcolemma versus mAspAT-based drial import of long-chain fatty acids (Fig. 2b). Longer exercise dura-
delivery of reducing equivalents into mitochondria)137. tions, lower intramuscular glycogen stores, higher dietary fat intakes
.
Compared to the situation at rest, NEFAs entering exercising and greater type I fibre abundance and aerobic fitness (VO 2max) levels
muscle are preferentially directed towards breakdown rather than can reduce carbohydrate reliance and increase fat utilization during
to re-esterification and storage138. Before being used by downstream physical activity120. Nevertheless, although these factors can delay the
metabolic pathways, long-chain NEFAs must be activated by thioes- transition from predominant fat usage to predominant carbohydrate
terification to long-chain fatty acyl-CoA. Several fatty acid transport usage (referred to as the substrate ‘crossover point’), the biphasic
proteins possess intrinsic acyl-CoA synthetase activity139, but acyl-CoA pattern of NEFA oxidation with exercise intensity71,72 is ultimately
synthetase long-chain family member 1 (ACSL1) is the predominant maintained77,143.

Nature Reviews Molecular Cell Biology

Review article

Box 3

Energy depots in skeletal muscle

Creatine phosphate the sarcoplasmic/endoplasmic reticulum Ca2+-ATPase (SERCA)329
The total concentration of muscle creatine is ~120–130 mmol per kg (Fig. 2). Thus, sufficient muscle glycogen is important for many
dry mass in young adults320. Approximately 66-67% of this muscle types of exercise and a critical threshold of ~250 mmol per kg dry
creatine pool is creatine phosphate (CrP), with the rest existing mass has been proposed330, below which self-perceived level of
in unphosphorylated form (free creatine)320. Resting CrP levels effort increases330 and sarcoplasmic reticulum function330,331, power
are ~10–15% higher in type II fibres, which rely on CrP to a greater output331 and repeated-sprint ability330 decline until glycogen
extent than type I fibres75. CrP is the main ATP substrate of the stores are replenished330,331. Nevertheless, exercising with low
phosphagen energy system through a reversible reaction catalysed glycogen availability (~100–300 mmol per kg dry mass) might
by sarcoplasmic creatine kinase (CKM). At physiological pH during augment signal transduction (for example, 5′-AMP-activated protein
exercise (~6.5–7.0)54,55, CKM is bound to sarcomeres through kinase (AMPK), p53) and gene expression (for example, PGC1A and
association with myomesins (MYOM1 and MYOM2) at the M-line321 TFAM)332 associated with oxidative metabolism and mitochondrial
and phosphofructokinase (PFK) at actin filaments of the I-band322— biogenesis (see the main text). This has led to the concept of ‘train
coupling CKM to contraction and glycolysis. CrP hydrolysis peaks low, compete high’ or ‘carbohydrate periodization’ (reviewed
at the onset of maximum contraction but deteriorates within ~6 s in ref. 332). Whether carbohydrate periodization leads to improved
(ref. 70), and stores can be ~75–90% depleted after roughly 30 s of exercise performance over time awaits confirmation, as does
hard exercise70,75. The majority of CrP resynthesis is linked to aerobic delineation of the potential underlying factors (for example,
metabolism through the ‘creatine phosphate shuttle’, whereby glycogen content per se versus hypocaloric diets that cause
mitochondrial CKM re-phosphorylates free creatine to CrP in the weight loss333 and/or proximal glycogen-depleting exercise
intermembrane space, using ATP from oxidative phosphorylation323 bouts that result in higher training volumes245).
(Fig. 2). Intramuscular creatine rarely reaches saturation by diet
and de novo synthesis alone, and supplementing an additional Intramyocellular lipids
~3 g per day (~0.03 g per kg body mass daily) increases stores Intramyocellular lipids (IMCLs) are stored in the hydrophobic core
~15% (to ~140 mmol per kg dry mass) in ~28 days, raising mostly of lipid droplet ellipsoids77 at peripheral (subsarcolemmal, SSLD)
free creatine324. Creatine supplementation appears safe for renal and central (intermyofibrillar, IMFLD) regions within fibres76–78,268.
function325 and consistently improves strength performance326. Women may have ~43% more individual lipid droplets in muscle,
The role of creatine in brain health is an emerging area of interest contributing to a greater (~84%) density of total IMCLs than in
and creatine supplementation may improve the memory function men334. Similarly, type I fibres utilize more IMCLs during moderate-
of older adults327. intensity exercise130 and have ~2–3-fold higher IMCL contents and
lipid droplet numbers than type II fibres76,78. Although IMCLs are
Glycogen mostly deposited (>85%) in IMFLD76–78, the relative distribution and
Muscle glycogen concentrations vary depending on nutritional characteristics of lipid droplet subpopulations (particularly in
state but are ~400–500 mmol per kg dry mass in the vastus lateralis type II fibres) vary depending on training status, body composition
of individuals following mixed macronutrient diets328. Glycogen and metabolic health78.
granules are nonuniformly distributed between three specialized Endurance-trained athletes and adults with T2D have similar
pools in muscle329: (1) the intermyofibrillar pool, found between total IMCL concentrations in muscle78,335 despite markedly
myofibrils, close to mitochondria and the sarcoplasmic reticulum; dissimilar insulin sensitivity profiles78,84,335. This apparent
(2) the subsarcolemmal pool, positioned just beneath the cell contradiction has been termed the ‘athlete paradox’335
membrane; and (3) the intramyofibrillar pool, situated within the but could be partly explained by contrasting lipid droplet
myofibril at the Z-line of the I-band (Fig. 1a). The intermyofibrillar pool properties77,78. Individuals with T2D have a greater number of
is most abundant and accounts for ~77–84% of muscle glycogen, extremely large SSLD in type II fibres77,78, which also possess
whereas intramyofibrillar and subsarcolemmal stores constitute lower subsarcolemmal mitochondrial contents77. This results in
~8–15% and ~6–12%, respectively35,96,97,267,268. The relative concentration higher relative contributions of SSLD to the overall IMCL pool78
of each glycogen reserve appears unaffected by age267, type 2 diabetes and fewer mitochondria-to-SSLD contacts77. Spatially, SSLD could
(T2D)268 or anatomical location (triceps brachii compared to vastus interfere with insulin signalling and the T2D-SSLD phenotype
lateralis)35, but may vary depending on physical activity status35,267,268 negatively correlates with peripheral insulin sensitivity78.
and fibre type35,97. Total glycogen content and utilization is typically Conversely, endurance athletes have approximately twofold more
higher in type II versus type I fibres75,97, and type II fibres have ~23% IMCLs in type I fibres and an increased abundance of adipose
more glycogen in intramyofibrillar stores97. triacylglyceride lipase (ATGL) and perilipin 5 (PLIN5)78 — proteins
Intramyofibrillar glycogen fuels myosin and sodium/potassium associated with lipid droplet turnover118. The higher type I fibre
(Na+/K+)-ATPases329 and is preferentially mobilized during strenuous IMCL content of trained individuals is specifically stored in
endurance35, high-intensity interval97,330 and resistance96 exercise. smaller IMFLD78, which are favourably depleted during prolonged
Myosin-ATPases also utilize intermyofibrillar glycogen, as does endurance exercise76 (Fig. 2b).

Nature Reviews Molecular Cell Biology

Review article

(continued from previous page)

An 8-week programme of high-intensity interval training reduced adults with normal weight, overweight or obesity, or T2D after the
SSLD size, increased the subsarcolemmal mitochondria-to-lipid training intervention, regardless of baseline differences77. Therefore,
droplet ratio and redistributed IMCLs into small IMFLD in type II consistent exercise may alleviate muscle insulin resistance somewhat
fibres77. Accordingly, lipid droplet profiles were similar between through remodelling of lipid droplets77,78.

Biological rhythms and skeletal muscle a normal distribution of early-to-late chronotypes, yet the average
Normal circadian fluctuations in behaviour and physiology — such as chronotype differs according to biological sex (Supplementary Box 2)
sleep–wake cycles, nutritional state, body temperature, cardiovascular and changes during ageing166 (Box 2). Individuals with later chrono-
function, and tissue production and sensitivity to hormones (Supple- types could benefit from exercise-induced phase advances irrespective
mentary Table 1) — coincide to influence acute exercise capacity144 and of exercise timing, whereas earlier chronotypes might suffer circadian
response in a time-specific, tissue-specific manner20. Endogenously gen- misalignment with evening exercise151. Moving forward, chronotype
erated circadian rhythms are predominantly maintained by photic (light- will be an important covariate to consider in exercise biology, especially
dependent) cues relayed through the central pacemaker (hypothalamic for the goal of making personalized therapeutic recommendations.
suprachiasmatic nucleus)145 in concert with cell-autonomous clocks in
peripheral tissues, including skeletal muscle (reviewed in ref. 146). Mus- Skeletal muscle responses to acute exercise
cle clocks regulate transcriptional programmes that prepare the tissue The perturbation of systemic19,124 and local55 homeostasis posed by
for transitions between fasting and feeding147–149 and orchestrate 24-h acute exercise triggers an integrated series of metabolic, hormonal,
rhythms in muscle glucose147,149, lipid148,149 and amino acid148 metabolism. growth factor-related, inflammatory and mechanosensitive events.
Consequently, muscle clock disruption in mice impairs muscle insulin In this section we discuss how these stimuli converge to temporally
sensitivity147,150 and metabolic homeostasis147,148,150. regulate the acute post-exercise adaptive landscape in muscle — from
Hormonal, metabolic and temperature-dependent changes asso- signal transduction90 to gene expression167–169 (Fig. 3).
ciated with exercise shift the central pacemaker151 and impact circadian
clocks in cell types throughout the body152. In human vastus lateralis Post-exercise signalling
muscle, acute endurance153 and resistance153,154 exercise upregulates Sprint and resistance exercise impart stronger whole-body stress
the core clock genes CRY1 and PER2, in part through Ca2+-mediated responses and more robustly impact the muscle phosphoproteome
activation of cAMP response element-binding protein (CREB)153. Recip- than continuous endurance exercise does, although there is similarity
rocally, physical activity patterns directly modulate the expression, in the immediate post-exercise signature90. Over 400 phosphoryla-
phase and/or amplitude of ~15–20% of all rhythmic genes in mouse tion sites on >200 proteins were commonly altered among these exer­
soleus and tibialis anterior muscles independently of the muscle cise modalities90, leading to a shared enrichment of canonical exercise
clock155. This occurs through transcription factors such as NFATc1 that regulatory kinases, including CaMK, AMPK, mammalian target of
are stimulated by motoneuronal firing patterns and downstream Ca2+ rapamycin (mTOR), mitogen-activated protein kinase (MAPK), PKA
dynamics155. Together, these studies provide a mechanistic and bio- and PKC90,170. Much of this post-exercise pathway convergence is con-
chemical rationale for observed time-dependent variations in human served in contracted (exercised) rat and mouse muscle, despite little
athletic performance (reviewed in ref. 156) and the metabolic distur- overlap at specific phosphorylation sites171.
bances157,158 and increased risk of cardiovascular disease159 resulting Modality-driven divergence in the phosphoproteome becomes
from circadian misalignment. more apparent during the 3-h recovery period after exercise90 and may
A high-intensity exercise intervention in humans was shown to facilitate distinct muscle adaptations with endurance, sprint interval
offset the detrimental impact of short-term sleep restriction on whole- or resistance training. Phosphorylation of rapamycin-sensitive sub-
body glucose tolerance and mitochondrial respiration in permeabilized strates of mTOR complex 1 (mTORC1) and p38 MAPK are potentiated
muscle fibres160. Treadmill running exercise in mice also stimulated after resistance exercise90. mTORC1 is a central regulator of muscle
hypoxia-inducible factor-1α (HIF1α) and a broad range of muscle tran- hypertrophy that critically permits mechanical overload-induced
scriptomic161 and metabolic20,161 responses in a time-dependent manner. myofibrillar accretion172. In murine muscle, mTORC1 activity is in part
Such findings have prompted researchers to begin investigating exer- mediated by mechanostimulated diacyl glycerol kinase-ζ (DGKζ) pro-
cise as a therapy to promote circadian alignment and ameliorate meta- duction of phosphatidic acid173 and raptor-dependent translocation of
bolic disease. Preliminary human evidence indicates that training in the the catalytic component of mTORC1 to the late endosomal and lysoso-
afternoon or evening might enhance the beneficial effects of exercise mal system172. By contrast, mTOR in human muscle already associates
on aspects of insulin sensitivity162,163 and blood glucose control164. Fur- with the lysosome, and resistance exercise does not further augment
thermore, moderate-to-vigorous physical activity imparts greater risk this interaction174. Thus, the contraction-induced migration of the
reductions for cardiovascular disease and all-cause mortality when mTOR–lysosomal complex to focal adhesions175 and microvascula-
≥50% of the total activity volume is undertaken after 11:00 h (ref. 165). ture174 at the sarcolemma may be a more important step in humans174–176.
Although rapidly emerging, this branch of ‘chrono-therapeutics’ Here, mTOR associates with RAS homologue enriched in brain (RHEB)
is still in its infancy, and precise protocols to elicit specific metabolic and eukaryotic translation initiation factor 3 (EIF3)174, and activates
outcomes are only beginning to come to fruition20. Additionally, there downstream target ribosomal protein S6 (RPS6)175. The cooperation
is large population-level variation in human chronotypes166, and indi- between mTOR–EIF3 (ref. 174) and Ser235 and Ser236 phosphorylation
vidual chronotype may interact with diurnal exercise timing to modify of RPS6 (ref. 175) at the fibre periphery was greater when a post-exercise
acute responses and longer-term adaptations. Each age group contains protein and carbohydrate beverage was consumed, and this interplay

Nature Reviews Molecular Cell Biology

Review article

Box 4

Select factors produced from skeletal muscle during exercise

Enhanced blood flow during exercise not only improves delivery weight loss in obese mice341. However, post-exercise concentrations
of nutrients and hormones (Supplementary Table 1) to muscle but of Lac-Phe in human plasma were orders of magnitude lower than
also facilitates the release and transport of discrete factors from those administered in the murine experimental model341. As such,
muscle20. These secreted molecules (known as exercise-induced the role of Lac-Phe in the hunger-suppressing effects ~0–3 h after
‘myokines’ or muscle-derived ‘exerkines’) can act in an autocrine, high-intensity exercise342 warrants additional study.
paracrine or endocrine fashion and are thought to promote many Apelin (APLN)315 and succinate343 are other examples of molecules
of the favourable adaptations associated with physical activity111. released from exercising muscle that promote crosstalk between
Interleukin 6 (IL-6) is perhaps the prototypical example of an muscle fibres and resident mononuclear cells. This retrograde
exercise-stimulated, muscle-secreted factor336. Elevations in signalling is particularly important for adaptive extracellular matrix
circulating IL-6 with endurance-type exercise19,336,337 might contribute remodelling and angiogenesis (Supplementary Fig. 3). Longer-duration
towards short-term energy allocation by transiently inhibiting high-intensity intervals55 and prolonged endurance exercise124 may
inflammatory processes (for example, monocyte production increase circulating succinate more than resistance exercise does.
of tumour necrosis factor (TNF)) while preferentially directing Similarly, only endurance exercise lowered the [kynurenine (KYN)]
liberated non-esterified fatty acids (NEFAs) towards working muscle to [kynurenic acid (KYNA)] ratio in plasma124. KYN is a neurotoxic
(reviewed in ref. 338). Accordingly, systemic pharmacological metabolite and the systemic reduction in [KYN] to [KYNA] protects
blockade of the IL-6 receptor promotes re-esterification and storage of against stress-induced depressive-like symptoms in mice344.
NEFAs337. A ventromedial hypothalamic circuit of locally synthesized This occurs through a muscle peroxisome proliferator-activated
IL-6 controls NEFA oxidation after swimming exercise specifically receptor-γ coactivator 1α (PGC1α) isoform 1 (PGC1α1)–peroxisome
in the soleus of mice, through sympathetic α2A/α2C adrenoceptor proliferator-activated receptor-α/δ (PPAα/δ)–KYN aminotransferase
modulation of 5′-AMP-activated protein kinase (AMPK)–acetyl-CoA (KAT) cascade that detoxifies KYN to KYNA344. KYNA subsequently
carboxylase (ACC)339. The effects of centrally produced IL-6 preceded released from muscle could influence energy homeostasis by
a rise in peripheral IL-6 concentrations, and whether IL-6 from muscle activating G protein-coupled receptor 35 (GPR35) on adipocytes
can activate this same neuromuscular axis is unclear339. to stimulate lipid turnover345. Numerous other exercise-responsive
Lactate is an established signalling metabolite and its production secretory factors regulated by PGC1α can reportedly signal from
is enhanced by glycolytic stressors. Elevated rates of muscle muscle (reviewed in ref. 223), including neurturin (NRTN)346. NRTN
glycolysis during higher-intensity (for example, sprint and resistance) operates through both autocrine and paracrine mechanisms to
exercise increases plasma lactate to greater extents than moderate- coordinate slow-oxidative muscle fibre and slow-twitch motor
intensity endurance exercise90. Muscle-derived lactate can initiate neuron property transitions in mice, and its mRNA is upregulated
an adipose tissue-transforming growth factor β2 (TGFβ2) secretion in human vastus lateralis ~72 h after sprint interval exercise346.
axis, which mediates improvements in murine glucose tolerance Collectively, cell-to-cell and inter-organ communication appears
after 11 days of voluntary wheel running340. Lactate exiting muscle to have an important role in the local and global effects of exercise.
can be further converted to N-lactoyl-phenylalanine (Lac-Phe) in cells Although muscle is a mediator of this crosstalk, other metabolically
expressing CNDP2, such as immune cell populations (for example, active tissues (for example, the heart, liver, adipose, nervous,
macrophages and monocytes) or epithelial cell populations341. endocrine and immune systems) also contribute, as discussed in
Intraperitoneal Lac-Phe delivery caused appetite suppression and detail elsewhere (reviewed in ref. 111).

could contribute towards the nutrient-sensitizing effects of resistance Endurance training has been suggested to impair resistance training
exercise on mTORC1 activation6,177. adaptations through AMPK-mediated blunting of mTORC1. Although
Complementary contraction-responsive mechanisms act in tandem this biochemical reaction is known to occur, exercise induction of both
with mTORC1 to promote gains in muscle mass6,178–180. For example, sar- AMPK and mTORC1 signalling is consistent with meta-analyses indicat-
comere shortening recruits the β-isoform of MAP3K20 (also known as ing little, if any, interference effect of concurrent exercise on muscle
ZAKβ) to the Z-disc, resulting in downstream stimulatory phosphoryla- hypertrophy184,185. Bidirectional AMPK–mTORC1 phosphorylation177
tion of p38 MAPK and JUN N-terminal kinase 1 ( JNK1) and JNK2 (ref. 179). could coordinate simultaneous90 or temporally distinct177 post-exercise
Inhibition mediated by JNK1 and JNK2 (ref. 181) and/or Notch182-mediated activity and downstream substrate specificity of these kinases.
inhibition of myostatin–suppressor of mothers against decapentaplegic Physiological proteome and mitochondrial remodelling with
(SMAD) signalling is a purported molecular ‘switch’ in muscle, promoting exercise requires the turnover of select proteins, and autophagic and
resistance-type over endurance-type adaptations by preventing SMAD proteasomal trafficking are integral to these processes. Indeed, genetic
complex nuclear translocation181. This suppression of the myostatin– disruption of proteostasis results in the accumulation of damaged pro-
transforming growth factor-β (TGFβ) pathway might be distinct from teins and impairs force production in fast-twitch extensor digitorum
the regulation of myostatin gene expression182, as both resistance and longus and slow-twitch soleus mouse muscle186. Treadmill running exer-
endurance training downregulate myostatin mRNA levels in human cise increases the activity of an AMPK complex composed of α1, α2, β2 and
muscle183 despite contrasting hypertrophic phenotypes over time13. γ1 isoforms located on the outer mitochondrial membrane (‘mitoAMPK’)

Nature Reviews Molecular Cell Biology

Review article

that could spatially regulate quality control through mitophagy across hypermethylation events are sustained after acute resistance exercise
the mitochondrial reticulum187. This pool of mitoAMPK is conserved and resistance training, providing a plausible mechanism for a ‘muscle
in human muscle187 and may govern mitophagy through phospho- memory’ of previous exercise that facilitates future adaptation199.
rylation of downstream substrates such as UNC-51-like autophagy Modality-dependent DNA methylation signatures might also direct
activating kinase 1 (ULK1)188 and mitochondrial fission factor (MFF)189. specific post-exercise transcriptional programmes in muscle, as acute
In humans, acute high-intensity190,191 and sprint interval90 exercise rap- bouts of resistance199 and sprinting-type202 exercise share enrichment
idly and transiently activates ubiquitin-proteasomal degradation in for only 36 of the top 100 differentially methylated pathways.
a cAMP–PKA-dependent manner190,191 that requires the proteasomal Conceivably, exercise-induced metabolic fluxes, which stimulate
subunit PSD11 (ref. 190). cAMP can then upregulate enzymatic activity glycolysis and OXPHOS, could alter acetyl and methyl group availabil-
of E3 ubiquitin ligases for continued removal of damaged proteins in ity, cellular [NAD+] to [NADH] redox state and the activity of ‘writers’
the post-exercise recovery period191 (Fig. 3). Coordinated proteolytic and ‘erasers’ of epigenetic marks in muscle192. However, the current
signalling is further detected 3 h after resistance exercise, as implied understanding of exercise-regulated DNA methylation and chromatin
by co-regulated phosphorylation of proteins in the N-end rule, calpain remodelling is incomplete. For example, lactate modifies (‘lactylates’)
regulation and ubiquitin receptor RAD23 pathways90. H3K18 in promoters and enhancers of genes related to tissue develop-
Many exercise-responsive phospho-sites have yet to be functionally ment and metabolism in human muscle203, but whether lactylation of
interrogated in muscle90,170,178. Additionally, there are numerous other histone lysine residues contributes to exercise adaptation remains to
protein modifications such as arginine methylation, ubiquitin-like modi- be ascertained.
fication, ribosylation and acylation that remain relatively understudied
in the context of physical activity despite presumably coinciding to Non-coding RNAs. In addition to protein-coding genes, RNA poly-
influence muscle phenotype. Deeper coverage of the posttranslational merase II transcriptional activity produces microRNAs (miRNAs; ≤22
proteome combined with the use of subcellularly targeted biosensors187 nucleotides) and long non-coding RNAs (lncRNAs; ≥200 nucleotides)
should enhance the spatial understanding of how integrated post- that alter mRNA levels and protein abundance without influencing the
exercise signalling governs unique exercise adaptations in muscle, and genetic code. Many miRNAs204 and lncRNAs205 are exercise-responsive
how these networks change with exercise training over time. in muscle, and both classes seem involved in aspects of myogenesis.
During resistance exercise, the downregulation of the muscle-enriched
Epigenetic regulation of exercise adaptation miRNA myomiR-1, through its contraction-mediated release in extra-
Histone modifications and DNA methylation. Upstream signalling, cellular vesicles, may facilitate hypertrophy206. Once in circulation,
together with an altered intramuscular metabolic milieu, can change myomiR-1 could act upon adipocytes to potentiate adrenergic signal-
the chromatin landscape in muscle and thus influence transcription ling and lipolysis206. Still, muscle was not a major source of bloodborne
factor-driven gene expression in response to exercise (reviewed extracellular vesicles upon resistance exercise in humans207 or tread-
in ref. 192). Epigenetic modifications on amino acids in histone tails mill running exercise in mice208. Rather, muscle-derived extracellular
or globular core domains render a permissive or repressive chromatin vesicles — loaded with myomiRs — preferentially accumulated in the
state depending on the type of modification and the residue and/or interstitial space, and treatment with these vesicles promoted dif-
histone protein being targeted192. An acute bout of endurance exercise ferentiation of myoblasts in vitro208. Therefore, extracellular vesicles
can increase H3K36 acetylation193, a mark of euchromatin, and promote originating from muscle might play a more prominent role in local
CaMKII-dependent phosphorylation194,195 and nuclear exclusion of cellular communication (Supplementary Fig. 3). Satellite cells may
the transcriptionally repressive class II histone deacetylases (HDAC) signal to muscle fibres209 and fibroadipogenic progenitor cells210 using
HDAC4 and HDAC5 (ref. 193). High-intensity resistance exercise also extracellular vesicles containing myomiR-206. In mice, this was shown
transiently stimulates myonuclear H3S10 phosphorylation through to regulate the levels of extracellular matrix (ECM)-related genes (for
p38 MAPK and mitogen and stress-activated protein kinase 1 (MSK1) example, Mmp9) in muscle fibres209 and Rrbp1 in fibroadipogenic pro-
in human muscle196. This same axis was necessary for the upregu­ genitor cells210 to coordinate physiological ECM remodelling during
lation of myocardin-related transcription factor B (MRTFB)–serum the early stages of hypertrophy. However, the ability of extracellular
response factor (SRF)-target genes and protein synthesis after in vivo vesicles to deliver biologically active miRNA cargo into recipient cells
contractions in mice196. remains contentious211 and their role in exercise adaptation warrants
Post-exercise histone modifications coalesce with changes in further interrogation.
DNA methylation, typically at the 5′ end of cytosine residues at CpG Though the impact of lncRNAs in human muscle is less studied
sites throughout the genome. DNA methylation in cis-acting regu­ than that of miRNAs, lncRNAs respond to endurance, resistance, con-
latory regions is commonly associated with gene silencing, but some current and high-intensity interval training with distinct differen-
transcription factors, particularly of the homeodomain POU and NFAT tial expression profiles between modalities205. The lncRNA CYTOR is
families, favour methylated CpG sequences197. Acute exercise198–200 and induced by endurance and resistance exercise in human muscle and
exercise training199,201 induce both hypomethylation and hypermethyla- can regulate fast-twitch fibre formation in rodents by sequestering
tion of DNA in muscle, mostly within gene bodies, intergenic regions TEA domain family member 1 (TEAD1) transcriptional activity212. Cytor
and enhancer regions201. The greatest number of CpG-site methylation mRNA is decreased upon ageing and overexpression of this lncRNA
changes and the highest percentage of mRNA transcripts with inversed increased type IIA and type IIB fibre abundance, which was sufficient to
DNA methylation patterns occurred 3 h after a single session of resist- improve grip strength and uphill running performance in aged mice212.
ance exercise, irrespective of workload (80% or 30% of 1 RM)200. These RNAs putatively annotated as lncRNAs may also contain short open
changes began to revert towards baseline methylation levels by 6 h200, reading frames that encode for micropeptides (≤150 amino acids).
suggesting that the muscle methylome is dynamic and tightly regulated The muscle-specific micropeptides dwarf open reading frame213 and
during recovery from exercise. Yet, select DNA hypomethylation and myoregulin214 competitively interact with sarcoplasmic/endoplasmic

Nature Reviews Molecular Cell Biology

Review article

Notch

Resistance exercise Endurance exercise

DAG
RHEB PA NICD
DGKζ Myofibrillar
mTOR protein breakdown Sarcomere E3 Ub ligases
EIF3 and turnover
Calpains
Ub
Lysosome ULK1 AMPK γ1/2/3
P α1/2 β1/2
Ca2+
Autophagy
P
Proteostasis Ca2+
PGC1α1 Sarcoplasmic
Proteasome cAMP reticulum
Protein kinase A

↑ Translation efficiency + capacity Ub

P Ribosomal biogenesis E3 Ub ligases NFATc1 ULK1/MFF

RPS6 Ribosome P
Mitophagy
mRNA PGC1α1
↑RPL3/↓RPL3L
NICD
ERRα NRF2 Transition to Mitochondrial
AMPK γ1
DGKζ slower myosin biogenesis
Myofibrillar IGF1–AKT Promoter Target gene α1/2 β1
protein synthesis signalling type
TFAM P
Nucleolus FOXO p53
Promoter Target gene
SMAD Endurance training > resistance training
MYC ↑ MYC complex Adaptations?
Chromatin Methyl group
remodelling
PGC1α1, TFAM, PPARδ, etc. =
Nucleus Promoter more oxidative phenotype
Target gene

Regulation of mRNA and protein

levels by non-coding RNAs
p38/JNK P

Sarcomere ZAKβ SMAD

M-band
complex
↑PGC1α4 MCU
Ca2+ Mostly RE-associated
Z-disk MSTN/TGFβ
Contraction Ca2+ Mostly EE-associated
↑ Ca 2+ signalling
Shared responses
Sarcoplasmic
Mitochondrion reticulum ACTRIIB

ECM remodelling Capillarization

reticulum Ca2+-ATPase (SERCA) to positively and negatively affect SR transport of mRNA and ribosomes to direct protein translation to spe-
Ca2+ kinetics and muscle contractility, respectively. Other classes of cific sites within the cell217. Microtubule-based trafficking of mRNA
non-coding RNAs, including circular RNAs and tRNA fragments, await facilitates transcript distribution through the crowded sarcoplasmic
future study in the context of human muscle exercise adaptation. space, but dispersion of mRNAs from progenitor myonuclei is still
potentially confined to ≤50 µm (refs. 215,216). By contrast, the move-
mRNA transport and myonuclear propagation ment of proteins translated from transcripts derived from a specific
The dense myofibrillar lattice in adult muscle fibres may impede dif- myonucleus could be ≥5 times more far-reaching218.
fusion of mRNAs expressed from myonuclei. Indeed, recent evidence The use of fluorescent reporters has shown that nuclear proteins
suggests that mRNA distribution away from the perinuclear region is can be detected in surrounding myonuclei and in those distant from
almost entirely dependent on microtubule networks in muscle, whereas the nucleus of origin218–220, a phenomenon termed ‘nuclear propa-
diffusion has a minor role215,216. Transport might be reliant on motor gation’218. Smaller proteins (≤40 kDa) that freely diffuse across the
proteins, reminiscent of cardiomyocytes that use kinesin-1-mediated nuclear pore complex can propagate to many myonuclei throughout

Nature Reviews Molecular Cell Biology

Review article

Fig. 3 | Molecular responses to acute exercise and exercise training. Exercise- myostatin (MSTN)/transforming growth factor-β (TGFβ) signalling. This
induced alterations in circulating molecules19,124 and the intramuscular milieu55,101, represents one of many intracellular changes permitting resistance training-
together with mechanical tension178, initiates a temporal series of biochemical induced hypertrophy over endurance-like adaptations181. The upregulation
and molecular events that lead to muscle adaptation. Activation of signalling of MYC with resistance exercise stimulates ribosomal biogenesis180,198, and the
cascades promote substantial posttranslational modification of the muscle formation of a specialized pool of ribosomes with a high ratio of ribosomal
proteome90,170 and DNA accessibility198,199,230. Collectively, this drives transcription protein large 3 (RPL3) to RPL3-like (RPL3L)238 may favour protein synthesis over
factor-dependent169 changes in gene expression167,183, alongside microRNAs204 and translational fidelity. MYC expression is mTOR-independent but MYC cooperation
long-non-coding RNAs205 that are thought to ‘fine-tune’ the molecular responses with mTOR is necessary to successfully increase ribosomal content180, possibly
to exercise. Endurance exercise (EE) and resistance exercise (RE) are often requiring an mTOR-driven reorganization of nucleoli to aid rRNA transcription281.
considered divergent stimuli, primarily driving oxidative versus hypertrophic Divergence between endurance and resistance exercise at the transcriptomic
muscle adaptations, respectively. However, common processes among exercise level is magnified after a period of training183. Endurance training increases
modalities can result in shared enrichment of signalling cascades90 and electron transport chain complex expression183, mitochon­drial content and
transcriptional networks183 in the post-exercise period. For example, coordinated muscle oxidative capacity12. Conversely, growth-related pathways183, ribosomal
proteolysis is detected following acute exercise, irrespective of exercise type90. abundance198 and muscle mass12 are augmented more by resistance training.
This may require cAMP–protein kinase A (PKA)190,191 and ensures protein quality This could be mediated in part by PGC1α isoforms. Nuclear localization of
control and physiological muscle remodelling. 5′-AMP-activated protein PGC1α1 and Ser15 phosphorylation of p53 are greater in resting muscle after
kinase (AMPK) activity90 and the expression of total PGC1A mRNA183 are also high-intensity interval training239, which might help to preserve mitochondrial
increased after a single bout of either endurance or resistance exercise. content and function257. By contrast, PGC1α1 protein is unchanged after resistance
AMPK phosphorylation activates peroxisome proliferator-activated receptor-γ training, whereas the PGC1α isoform 4 (PGC1α4) protein is preferentially
coactivator 1α isoform 1 (PGC1α1)279, and both AMPK and PGC1α1 potentiate enriched235. PGC1α4 stimulates muscle hypertrophy and is associated with
angiogenic factors260,280 and mitochondrial bioenergetics48,235,260,279 in muscle. enhanced Igf1 expression229, insulin-like growth factor 1 (IGF1)–serine/threonine
After endurance exercise, a distinct pool of mitochondrial AMPK (composed of α1, protein kinase (AKT) and mTORC1 signalling234 and downregulation of Mstn
α2, β2 and γ1 isoforms) regulates mitophagy187–189 and promoter hypomethylation mRNA in mouse muscle229. However, unlike PGC1α1 (ref. 280), PGC1α4 does
facilitates the transcription of peroxisome proliferator-activated receptor-δ not coactivate oestrogen-related receptor-α (ERRα)229 and has no effect on
(PPARδ) and mitochondrial transcription factor A (TFAM)230. Whether resistance oxidative phosphorylation enzymes229,235. Appreciable overlap in the initial
exercise elicits these same effects is unclear. Despite similarities, there are more stages of exercise training probably underlies the degree of shared adaptation
distinct than overlapping post-exercise responses between modalities90,183. between endurance and resistance exercise (see the sections ‘Skeletal muscle
Rapamycin-sensitive substrates of mammalian target of rapamycin (mTOR) responses to acute exercise’ and ‘Skeletal muscle adaptations to long-term
complex 1 (mTORC1) are phosphorylated to a greater extent after resistance exercise’). Depending on individual predisposition (Box 1), dedicated training
exercise90. Mechanical overload initiates translocation of the mTOR–lysosomal of a certain exercise modality could amplify discrete differences in the adaptive
complex174 and diacylglycerol (DAG) kinase-ζ (DGKζ)173 to the sarcolemma. response, resulting in distinct muscle adaptations and the development of
Here, mTOR colocalization with RAS homologue enriched in brain (RHEB) and specific phenotypes over time13. Evidence showing that combined endurance and
eukaryotic initiation factor 3 (EIF3)174 and phosphatidic acid (PA) produced by resistance training can blunt muscle hypertrophy in humans is scarce184,185 but
DGKζ173 might coalesce to fully stimulate mTORC1 (ref. 175) and the translation concurrent training could impede gains in explosive strength184. Still, a combined
of contraction-associated mRNAs. Acute attenuation of UNC-51-like autophagy exercise regime may offer dual benefits for most individuals12. ACTRIIB, activin
activating kinase 1 (ULK1) autophagic signalling after resistance exercise65 receptor type 2B; ECM, extracellular matrix; MFF, mitochondrial fission factor;
and nuclear DGKζ-mediated suppression of forkhead box protein O (FOXO)- MCU, mitochondrial calcium uniporter; NFATc1, nuclear factor of activated T cells,
dependent proteasomal degradation could also support muscle mass by cytoplasmic 1; NICD, Notch intracellular domain; NRF2, nuclear factor erythroid
moderating the breakdown of myofibrillar proteins173. At the sarcomere, 2-related factor 2; p38, p38 mitogen-activated protein kinase; RPS6, ribosomal
contraction recruits ZAKβ to the Z-disc179 where it acts through JUN N-terminal protein S6; SMAD, mothers against decapentaplegic homologue; Ub, ubiquitin;
kinase 1 ( JNK1) and JNK2 (refs. 179,181), potentially alongside Notch182, to inhibit ZAKβ, MAP3K20 isoform-β.

a myotube in vitro218. Conversely, larger proteins (>40–60 kDa) are the total number (from ~0.7–1% to ~3%) of DNA-replicating myonuclei221
limited in their propagation potential, possibly due to increasing in murine muscle. The extent to which these processes occur under
reliance on nuclear transport receptors to traverse the nuclear pore physiological conditions in humans remains unclear.
complex218. Protein properties (such as size) and myonuclear charac-
teristics (such as shape or location within the muscle fibre) probably The post-exercise transcriptome
interact to control nuclear propagation. For example, myonuclei Considerable effort has been dedicated towards characterizing the
in ex vivo fibres differ in capacity for classical versus non-classical transcriptional response of human muscle to exercise, and several key
nuclear import, resulting in a gradient of import rates between studies are highlighted in Supplementary Table 2. High-powered meta-
myonuclei of the same fibre219. analyses have curated much of this scientific endeavour to generate
The potential for myonuclear propagation casts further doubt deeper insight into exercise adaptation167,183. Acute exercise initiates
on strict myonuclear domains in muscle and represents an additional temporal clusters of gene expression167–169, including early induction
layer of spatiotemporal control, combining fibre-wide communication of stress response genes, increased expression of muscle adaptation
with myonuclei-autonomous processes in response to adaptive cues. genes in early and middle stages following exercise167,169, and later167 or
Furthermore, resident myonuclei may be capable of endoreplica- sustained168 increases in immune-related signatures. These transcrip-
tion in mouse muscle (resulting in polyploidy), which could support tional programmes are driven by combinations of different transcrip-
efficient DNA synthesis under heightened transcriptional demand221. tion factors167,169 and alternative transcription start-site usage169. Akin
Myonuclear propagation was detected after wheel-running exercise to the phosphoproteome90, a single bout of resistance exercise alters
training220 and synergistic ablation-induced hypertrophy increased more transcripts than does moderate-intensity endurance exercise183.

Nature Reviews Molecular Cell Biology

Review article

However, there is considerable overlap in pathway enrichments, and select adaptations to specific training interventions and the influence
both training modalities upregulate stress response genes, kinase of muscle fibre type14. We then summarize how muscle-related adapta-
activity and metabolism-related processes, among others183. tions can contribute to improvements in systemic measures of exercise
NR4A3 (also known as NOR1)222 and PGC1A223 are genes commonly performance and discuss whether altered molecular perturbations in
induced after endurance and resistance exercise183 and mainly promote muscle with exercise training33,239–241 reflect an attenuated or refined
oxidative muscle adaptations. NR4A3 and PGC1A have similar tempo- adaptive state.
ral expression patterns after exercise, peaking ~2–3 h into the recov-
ery period167,183, suggesting common upstream regulation through Transcriptome, proteome and contractile property
CREB224,225. The nuclear receptor subfamily 4 group A (NR4A) family adaptations
(including NR4A1 (also known as NUR77) and NR4A2 (also known as Repeated exposure to specific exercise modalities results in shared
NURR1)) share >90% homology in their DNA-binding domains and and divergent alterations in muscle over time12,183 (Fig. 3). At the tran-
can interact as monomers with a single NGFI-B response element226, or scriptomic level, pathways related to ECM remodelling, angiogenesis
as homodimers and heterodimers (alongside other NR4A members) and the TCA cycle are enriched in muscle after exercise training167.
with the Nur response element227. NR4A1 is also exercise-responsive in Whereas resistance exercise robustly effects ECM reorganization183,242
muscle167,202 and may serve partial redundancy with NR4A3, as implied and growth-related pathways242, endurance training favourably impacts
by the slow-oxidative phenotypes of mice expressing either Nr4a1 oxidative metabolic processes and the gene expression of mitochon-
(ref. 228) or Nr4a3 (ref. 222) from skeletal muscle-specific promoters. drial complex subunits183. Crosstalk between the various cell popu­
The PGC1α family of coactivators consists of different iso- lations in muscle appears particularly important for exercise-associated
forms, which are derived from alternative promoter use or splic- adaptations within the interstitial space (Supplementary Fig. 3).
ing events (reviewed in ref. 223). Endurance exercise preferentially Training-induced improvements in muscle respiration often
targets229 and hypomethylates230 the proximal PGC1α promoter in coincide with expansion and modification of the mitochondrial
muscle, increasing PGC1α isoform 1 (PGC1α1) expression. PGC1α1 drives proteome12,15,143 in type I and type II fibres14. Resistance training can
mitochondrial biogenesis and oxidative fibre properties48 at least augment the respiratory and ATP-producing capacity of mitochon-
in part through C-terminal domain RNA binding, which facilitates dria243, but the effects on mitochondrial phenotype are typically less
recruitment in transcriptionally active chromatin condensates231. pronounced than those observed after high-intensity interval12,15,143
A bout of endurance exercise also leads to promoter region hypometh- and endurance14 training. The total volume of endurance exercise
ylation of other key mitochondrial and metabolic genes, including performed during a training intervention might be the strongest stimu-
TFAM and PPARD, with corresponding increases in mRNA expres- lus for mitochondrial biogenesis in muscle (reviewed in ref. 244). This
sion230. The PPARδ-mediated upregulation of lipid metabolism and could be related to the additive effects of proximal exercise bouts on
suppression of glycolytic processes in muscle232 is largely dependent the nuclear enrichment of transcriptional regulators of mitochondrial
on protein–protein interaction with Krüppel-like factor 15 (KLF15)233. abundance (for example, PGC1α1 and transcription factor EB (TFEB))245
Most PGC1α isoforms have overlapping roles in muscle223. The and the upregulation of mito-ribosomes12,14,15 and mitochondrial
truncated variant PGC1α isoform 4 (PGC1α4) differs in that it can be protein synthesis12. During mitochondrial biogenesis, stoichiometry of
induced by intramitochondrial [Ca2+]234 and is transcribed from the the ETC is maintained through proportional translation of ETC complex
alternative PGC1α promoter after resistance exercise229,235 (Fig. 3). subunits from the mitochondrial and nuclear genomes246. Increased
PGC1α4 facilitates muscle hypertrophy229,234,235 and a PPARβ–AMPK- expression of mitochondrial transit peptides and translocase proteins
dependent upregulation of glycolytic metabolism235, which could of the outer and inner mitochondrial membrane following endurance
enhance the anaerobic capacity of muscle after resistance training. Inci- training14 facilitates the transport of nuclear-encoded precursor com-
dentally, the accretion of muscle mass often coincides with a glycolytic ponents into the mitochondrial matrix. However, in the early stages of
switch that may support anabolism by increasing de novo synthesis mitochondrial remodelling, synthesis of fatty acid oxidation and TCA
of nucleotides and non-essential amino acids from intermediates in cycle enzymes may be prioritized over ETC machinery biogenesis,
glucose metabolism (reviewed in ref. 236). to enhance reducing equivalent NADH and FADH2 production and
An acute bout of resistance exercise also stimulates MYC mRNA electron delivery to OXPHOS15.
expression and the hypomethylation of enhancer and MYC-accessible High-intensity interval training at ≥90% of maximum power output
.
intergenic regions of ribosomal DNA198. These events coalesce to aug- (Wmax ) most consistently increases normalized (mass-specific) mito-
.
ment pre-45S rRNA transcription and ribosomal biogenesis180,198. chondrial respiration244, with intensities ≥100% Wmax (as in sprint inter-
Whereas mTORC1 signalling promotes translational efficiency172,180 val training) doing so more efficiently per unit of time spent exercising244.
and directs ribosomal function through Ser235 and Ser236 phospho- The impact of intensity on mitochondrial respiration appears uncou-
rylation of RPS6 (ref. 237), total ribosomal content180 and the altered pled from total training volume, suggesting that these variables drive
composition of ribosomal proteins (for example, an increased ratio complementary but somewhat distinct mitochondrial adaptations244.
of RPL3 to RPL3L)238 determines translational capacity in muscle. High-intensity interval training also induces lysine acetylation of TCA
Collectively, upregulation of the sarcoplasmic and mitochondrial cycle and ETC proteins143, although whether such posttranslational
protein synthetic machinery (such as mito-ribosomes) enables the ben- modifications contribute towards intensity-dependent changes in
eficial proteome remodelling that occurs with high-intensity interval, mitochondrial function requires additional study.
resistance or combined exercise training12. After endurance training, some discrete differences can be seen
in the adaptive response of type I and type II fibres14. Still, most of the
Skeletal muscle adaptations to long-term exercise commonly detected exercise-regulated proteins related to mitochon-
In this section, we contextualize the impact of consistent exercise drial and glucose metabolism behaved similarly between fibre types14.
(exercise training) on muscle phenotype, considering general and For example, training upregulated ACSL1, malate/aspartate shuttle

Nature Reviews Molecular Cell Biology

Review article

.
proteins, the mitochondrial CKM isoform CKMT2, PDH, CRAT and Conversely, any beneficial effect of resistance training on VO 2max appears
LDHB in both type I and type II fibres14. LDHB mRNA is increased after limited to individuals with lower initial aerobic fitness levels (for exam-
.
endurance247 and resistance235 training, potentially through PGC1α1 ple, a VO 2max ≤25 or ≤40 mL kg−1 min−1 in adults >60 and ~20–35 years of
coactivation of distal myocyte enhancer factor 2 (MEF2) and proximal age, respectively) (reviewed in ref. 253) and thus high-intensity interval12
.
oestrogen-related receptor (ERR) binding sites in the LDHB promoter247. and endurance training13,248 typically increase VO 2max to greater extents.
LDHB preferentially catalyses the conversion of lactate to pyruvate A period of aerobic pre-conditioning can potentiate the hyper-
and, together with enhanced mitochondrial density, could improve trophic effects of resistance exercise248, in part by enhancing fibre
lactate clearance in trained muscle through higher rates of conversion capillary density248,250 (Supplementary Fig. 3). Alternatively, high-
to pyruvate and subsequent oxidation247 (Fig. 2c). Further refinement intensity resistance training (≥90% of 1 RM or ≤4 RM) can improve run-
.
of substrate handling with programmed exercise is implied by greater ning economy (that is, reduce the metabolic or VO 2 cost of running)254.
expression of key proteins of metabolic pathways involving glycogen These complementary adaptations further emphasize the utility of
(for example, glycogen synthase)11, NAD+ (nicotinamide phospho- incorporating numerous exercise (sub)modalities into a sustainable
ribosyltransferase) and branched-chain amino acid (branched-chain routine for maximum performance12 and protective health benefits18.
α-ketoacid dehydrogenase kinase) metabolism, as well as ubiqui-
none biosynthesis (5-demethoxyubiquinone hydroxylase)143, after Adaptive responses in muscle are modified with exercise
high-intensity interval training. training
Muscle contractile properties are likewise affected by regular The recurring stimulus of exercise training dampens select signal-
exercise in a somewhat training modality-dependent manner. Pro- ling239,240 and transcriptomic33,241 responses to an acute bout of physi-
tein isoforms regulating SR Ca2+ release143 and myofibrillar Ca2+ sen- cal activity. This is evident after just nine sessions of high-intensity
sitivity14,143 were altered after moderate-intensity endurance14 and interval exercise through a modest reduction in the number (~17%) and
high-intensity interval143 training, coincident with a reduction in fibre amplitude (~30%) of altered mRNA transcripts in human muscle com-
size-adjusted quadriceps peak-twitch torque and prolonged half- pared to the first exercise bout, including glycolysis pathway and HIF1α
relaxation time11 — indicative of a slower muscle phenotype. Conversely, target genes241. The magnitude of change in the muscle transcriptome
stretch-shortening cycle (ballistics or plyometrics) training increased with exercise training becomes even clearer over 30 daily sessions of
contractile velocity, force production and thus peak power of single electrically induced mechanical overload in rat hindlimb muscle33.
type I, type IIA and type IIA/X fibres42. Exercise modalities commonly In particular, the directionality of altered genes contrasted substantially
decrease MYH1 mRNA183 and protein (MyHC-IIX)11,13 content in muscle. between exercise naive and trained muscle. At the 1-h sampling time
However, the upregulation of MYH7 gene expression183 and type I fibre point after exercise, ~70% of the (~2400) differentially expressed tran-
abundance after endurance training44 can be in contrast to the effects scripts were upregulated on day 2 compared with ~83% (of ~3300 genes)
of resistance training, which often downregulates MYH7 (ref. 183) with downregulated on day 30 (ref. 33). This indicates that certain signalling
no corresponding change in the prevalence of type I fibres45. This is mechanisms are sensitive to the modified intracellular environment
consistent with endurance44 and resistance45 exercise differentially after a period of regular exercise. In combination with genetic predis-
shifting hybrid fibres towards pure type I and pure type IIA fibres, position (Box 1), blunted post-exercise molecular responses probably
respectively (reviewed in ref. 38). Indeed, endurance training resulted converge to limit the adaptive potential of muscle, slowing progress
in the modification of more proteins in type I than in type II fibres14, with advanced training experience. Nevertheless, some alterations
and long-term exposure to a particular exercise stimulus41 may largely might reflect a refined (as opposed to impaired) response.
explain the majority of type I or type II fibres detected in biopsies from Pathways related to ribosomal biogenesis and protein synthe-
endurance or strength athletes13. sis are positively enriched in the early adaptive stages in exercised
rat muscle33 but become downregulated in favour of metabolism-
Training-induced adaptations impact exercise performance related processes once hypertrophy plateaus33. Rates of myofibril-
Together, exercise training adaptations in muscle contribute towards lar protein synthesis also correlate better with measures of muscle
.
improvements in gross performance measures such as VO 2max mass accrual after 8 weeks176 to 10 weeks255 of resistance training in
(refs. 12,14,15), peak power output11,143, ballistic power42 and strength12,13. humans, corresponding to the attenuation of exercise-induced mus-
Endurance and resistance training increase muscle capillarization248–250 cle damage255. Furthermore, although 20 days (40 sessions) of high-
and endurance-type training enhances NEFA oxidation at a given exer- frequency, high-intensity interval training reduced the post-exercise
cise intensity77,120,143,251. These glycogen-sparing and metabolite- upregulation of nuclear PGC1α1 and p53 phosphorylation at Ser15,
buffering (for example, lactate clearing)247 effects render muscle more resting levels of these same markers were greater in human muscle
robust to disturbances in metabolic homeostasis, delaying mechanisms after the exercise intervention239. Phosphorylated p53 can translocate
of peripheral and central fatigue94 and raising exercise tolerance and to the mitochondrial genome where it interacts with transcription
.
capacity (that is, higher velocities and percentages of VO 2max pro- factor A, mitochondrial (TFAM) to regulate mitochondrial transcrip-
duced at ventilatory thresholds, and greater maximal running speeds tion256 (Fig. 3). Muscle-specific Trp53-knockout mice have reduced
or distances covered)251. Metabolomics analysis also implies that basal mitochondrial contents and enzymatic activities but retain the
4 weeks (13 bouts) of resistance exercise may suffice to modify the capacity for exercise-stimulated mitochondrial biogenesis257. This sug-
acute muscle metabolome252. Yet, the specific metabolic changes that gests that p53 may be more important for the maintenance of muscle
occur with resistance training are not extensively characterized and mitochondrial integrity than exercise adaptation.
warrant further interrogation. Given these diverse modifications, longer-duration trials are
High-intensity interval11,12 and endurance13 training both induce a needed to fully elucidate the modality-dependent temporal land-
degree of muscle hypertrophy, but gains in muscle mass and strength scape of exercise adaptation in human muscle and to highlight specific
are less pronounced than seen with long-term resistance exercise13. changes in the acute exercise response with consistent training.

Nature Reviews Molecular Cell Biology

Review article

Future directions 10. McConell, G. K. et al. Insulin-induced membrane permeability to glucose in human
muscles at rest and following exercise. J. Physiol. 598, 303–315 (2020).
The complex and intricate macrostructure to ultrastructure of skeletal 11. Hostrup, M., Onslev, J., Jacobson, G. A., Wilson, R. & Bangsbo, J. Chronic β2-adrenoceptor
muscle allows for a high degree of metabolic flexibility during exercise. agonist treatment alters muscle proteome and functional adaptations induced by high
Further study of the regulation and interaction of key cellular com- intensity training in young men. J. Physiol. 596, 231–252 (2018).
12. Robinson, M. M. et al. Enhanced protein translation underlies improved metabolic and
ponents, such as myofibrillar and mitochondrial networks53, should physical adaptations to different exercise training modes in young and old humans.
enable more comprehensive understanding of how muscles mount Cell Metab. 25, 581–592 (2017).
such a malleable response. The use of fluorescent metabolite258 and This study provides a thorough analysis of exercise adaptation at several -omics levels
in human skeletal muscle.
myofilament259 biosensors could aid this effort by elucidating real-time 13. Chapman, M. A. et al. Skeletal muscle transcriptomic comparison between long-term
dynamics of muscle metabolism and contraction. trained and untrained men and women. Cell Rep. 31, 107808 (2020).
Current knowledge of exercise adaptation is heavily biased 14. Deshmukh, A. S. et al. Deep muscle-proteomic analysis of freeze-dried human muscle
biopsies reveals fiber type-specific adaptations to exercise training. Nat. Commun. 12,
towards canonical regulators such as AMPK, PGC1s and mTORC1. 304 (2021).
Although these factors are integral to muscle remodelling with exer- This study highlights notable shared and distinct adaptations in the proteomes of
type I and type II fibres after a period of endurance training.
cise, aspects of adaptation are maintained in the absence of AMPKα1 and
15. Granata, C. et al. High-intensity training induces non-stoichiometric changes in the
AMPKα2 (ref. 260) and of PGC1α and PGC1β (ref. 261), and rapamycin- mitochondrial proteome of human skeletal muscle without reorganisation of respiratory
insensitive pathways also contribute towards muscle hypertrophy178,180. chain content. Nat. Commun. 12, 7056 (2021).
This study interrogates the muscle mitochondrial proteome at multiple time points
Most of the exercise-induced phosphoproteome remains relatively
during a periodized high-intensity training intervention.
unexplored90,170,178, and its functional interrogation will help to identify 16. Laukkanen, J. A. et al. Long-term change in cardiorespiratory fitness and all-cause
redundancies in adaptative signalling and new players in the distinct mortality: a population-based follow-up study. Mayo Clin. Proc. 91, 1183–1188 (2016).
17. García-Hermoso, A. et al. Muscular strength as a predictor of all-cause mortality in an
and complementary effects of resistance, endurance and high-intensity
apparently healthy population: a systematic review and meta-analysis of data from
exercise. approximately 2 million men and women. Arch. Phys. Med. Rehabil. 99, 2100–2113.e5
As exercise biology moves increasingly towards systems-level (2018).
18. Coleman, C. J., McDonough, D. J., Pope, Z. C. & Pope, C. A. Dose-response association
profiling, the multiplexing of new technologies could provide invalu- of aerobic and muscle-strengthening physical activity with mortality: a national
able insight into the spatiotemporal control of gene expression262, cohort study of 416 420 US adults. Br. J. Sports Med. https://doi.org/10.1136/
translation263 and metabolism37 in discrete muscle cell populations. bjsports-2022-105519 (2022).
19. Contrepois, K. et al. Molecular choreography of acute exercise. Cell 181, 1112–1130.e16
If performed in tandem with commensurate functional assays, this (2020).
would enable causal association of specific cellular events with their This study is a multi-omics analysis of human plasma and peripheral blood
phenotypic impact. mononuclear cells at multiple time points across a 1-h recovery period immediately
following a standardized exercise bout.
Exercise is an effective adjunct therapy for certain neurodegen- 20. Sato, S. et al. Atlas of exercise metabolism reveals time-dependent signatures of
erative264 and mental health265 disorders and offers protection against metabolic homeostasis. Cell Metab. 34, 329–345.e8 (2022).
cardiometabolic disease266, sarcopenia (Box 2) and all-cause mortal- In this study, mass spectrometry-based metabolomics is used to characterize
time-of-day differences in metabolic programming in several mouse tissues in
ity18. Multicentre interventions including spectrums of chronological response to an acute bout of exercise, including the net uptake and release of
age, metabolic health, chronotype, ethnicity, biological sex and social metabolites in hindlimb muscles and liver.
gender will move the field closer to uncovering the dynamic changes 21. Murgia, M. et al. Protein profile of fiber types in human skeletal muscle: a single-fiber
proteomics study. Skelet. Muscle 11, 24 (2021).
that coordinate the benefits of exercise training and further define the This study interrogates the proteome of single muscle fibres from young, healthy
role of muscle in the integrative exercise response. Knowledge derived individuals and provides a useful resource for basal proteomic comparisons between
from such approaches should not only inform personalized exercise type I and type II fibres.
22. Bloemberg, D. & Quadrilatero, J. Rapid determination of myosin heavy chain expression
prescription, but also reveal new molecular avenues of drug discovery in rat, mouse, and human skeletal muscle using multicolor immunofluorescence
for the improvement of human health. analysis. PLoS ONE 7, e35273 (2012).
This paper provides a valuable resource for myosin fibre type and enzymatic profile
comparisons, and identifies notable differences between muscles and species.
Published online: xx xx xxxx 23. Simoneau, J. A. & Bouchard, C. Human variation in skeletal muscle fiber-type proportion
and enzyme activities. Am. J. Physiol. 257, E567–E572 (1989).
References 24. Pellegrino, M. A. et al. Orthologous myosin isoforms and scaling of shortening velocity
1. Bilet, L. et al. One-leg inactivity induces a reduction in mitochondrial oxidative with body size in mouse, rat, rabbit and human muscles. J. Physiol. 546, 677–689 (2003).
capacity, intramyocellular lipid accumulation and reduced insulin signalling upon lipid 25. Bottinelli, R., Pellegrino, M. A., Canepari, M., Rossi, R. & Reggiani, C. Specific
infusion: a human study with unilateral limb suspension. Diabetologia 63, 1211–1222 contributions of various muscle fibre types to human muscle performance: an in vitro
(2020). study. J. Electromyogr. Kinesiol. 9, 87–95 (1999).
2. Bergouignan, A. et al. Effect of contrasted levels of habitual physical activity on 26. Dos Santos, M. et al. Single-nucleus RNA-seq and FISH identify coordinated
metabolic flexibility. J. Appl. Physiol. 114, 371–379 (2013). transcriptional activity in mammalian myofibers. Nat. Commun. 11, 5102 (2020).
3. Kivimaki, M. et al. Physical inactivity, cardiometabolic disease, and risk of dementia: This is one of the first studies to perform single myonuclear RNA-sequencing,
an individual-participant meta-analysis. Brit. Med. J. 365, l1495 (2019). revealing heterogeneity between myonuclei within a given muscle fibre.
4. Zhang, D. et al. Leisure-time physical activity and incident metabolic syndrome: 27. Dos Santos, M. et al. A fast myosin super enhancer dictates muscle fiber phenotype
a systematic review and dose-response meta-analysis of cohort studies. Metabolism through competitive interactions with myosin genes. Nat. Commun. 13, 1039 (2022).
75, 36–44 (2017). This study uses a ‘rainbow’ transgenic mouse model of the fast-type myosin locus
5. Millard, L. A. C., Tilling, K., Gaunt, T. R., Carslake, D. & Lawlor, D. A. Association of physical to determine that competitive promoter–super enhancer interactions govern
activity intensity and bout length with mortality: an observational study of 79,503 UK fast-twitch myosin isoform expression in skeletal muscle.
Biobank participants. PLoS Med. 18, e1003757 (2021). 28. Roman, W. et al. Myofibril contraction and crosslinking drive nuclear movement to the
6. D’Hulst, G., Masschelein, E. & De Bock, K. Resistance exercise enhances long-term periphery of skeletal muscle. Nat. Cell Biol. 19, 1189–1201 (2017).
mTORC1 sensitivity to leucine. Mol. Metab. 66, 101615 (2022). 29. Wen, Y. et al. Myonuclear transcriptional dynamics in response to exercise following
7. Goodyear, L. J. et al. Skeletal muscle plasma membrane glucose transport and glucose satellite cell depletion. iScience 24, 102838 (2021).
transporters after exercise. J. Appl. Physiol. 68, 193–198 (1990). 30. Kim, M. et al. Single-nucleus transcriptomics reveals functional compartmentalization
8. Burd, N. A. et al. Enhanced amino acid sensitivity of myofibrillar protein synthesis in syncytial skeletal muscle cells. Nat. Commun. 11, 6375 (2020).
persists for up to 24 h after resistance exercise in young men. J. Nutr. 141, 568–573 (2011). 31. D’Ercole, C. et al. Spatially resolved transcriptomics reveals innervation-responsive
9. Sjoberg, K. A. et al. Exercise increases human skeletal muscle insulin sensitivity via functional clusters in skeletal muscle. Cell Rep. 41, 111861 (2022).
coordinated increases in microvascular perfusion and molecular signaling. Diabetes 66, 32. Battey, E. et al. Myonuclear alterations associated with exercise are independent of age
1501–1510 (2017). in humans. J. Physiol. https://doi.org/10.1113/JP284128 (2023).

Nature Reviews Molecular Cell Biology

Review article

33. Viggars, M. R. et al. Adaptation of the transcriptional response to resistance exercise over 62. Cogliati, S. et al. Mitochondrial cristae shape determines respiratory chain
4 weeks of daily training. FASEB J. 37, e22686 (2023). supercomplexes assembly and respiratory efficiency. Cell 155, 160–171 (2013).
34. Lexell, J., Taylor, C. C. & Sjostrom, M. What is the cause of the ageing atrophy? Total 63. Balsa, E. et al. ER and nutrient stress promote assembly of respiratory chain
number, size and proportion of different fiber types studied in whole vastus lateralis supercomplexes through the PERK-eIF2α axis. Mol. Cell 74, 877–890.e6 (2019).
muscle from 15- to 83-year-old men. J. Neurol. Sci. 84, 275–294 (1988). 64. Nielsen, J. et al. Plasticity in mitochondrial cristae density allows metabolic capacity
This classic study establishes fibre type differences in human skeletal muscle during modulation in human skeletal muscle. J. Physiol. 595, 2839–2847 (2017).
healthy ageing, thereby providing insights on the aetiology of sarcopenia. 65. Hentila, J. et al. Autophagy is induced by resistance exercise in young men, but unfolded
35. Nielsen, J., Holmberg, H. C., Schroder, H. D., Saltin, B. & Ortenblad, N. Human skeletal protein response is induced regardless of age. Acta Physiol. 224, e13069 (2018).
muscle glycogen utilization in exhaustive exercise: role of subcellular localization and 66. Heymsfield, S. B. et al. Human energy expenditure: advances in organ-tissue prediction
fibre type. J. Physiol. 589, 2871–2885 (2011). models. Obes. Rev. 19, 1177–1188 (2018).
36. Murgia, M. et al. Single muscle fiber proteomics reveals fiber-type-specific features of 67. Phung, L. A., Foster, A. D., Miller, M. S., Lowe, D. A. & Thomas, D. D. Super-relaxed state
human muscle aging. Cell Rep. 19, 2396–2409 (2017). of myosin in human skeletal muscle is fiber-type dependent. Am. J. Physiol. Cell Physiol.
37. Luo, L. et al. Spatial metabolomics reveals skeletal myofiber subtypes. Sci. Adv. 9, 319, C1158–C1162 (2020).
eadd0455 (2023). 68. Linari, M. et al. Force generation by skeletal muscle is controlled by mechanosensing in
38. Medler, S. Mixing it up: the biological significance of hybrid skeletal muscle fibers. myosin filaments. Nature 528, 276–279 (2015).
J. Exp. Biol. 222, jeb200832 (2019). 69. Gaitanos, G. C., Williams, C., Boobis, L. H. & Brooks, S. Human muscle metabolism during
39. Horwath, O. et al. Variability in vastus lateralis fiber type distribution, fiber size, and intermittent maximal exercise. J. Appl. Physiol. 75, 712–719 (1993).
myonuclear content along and between the legs. J. Appl. Physiol. 131, 158–173 (2021). 70. Parolin, M. L. et al. Regulation of skeletal muscle glycogen phosphorylase and PDH
40. Murach, K. A. et al. Fiber typing human skeletal muscle with fluorescent during maximal intermittent exercise. Am. J. Physiol. 277, E890–E900 (1999).
immunohistochemistry. J. Appl. Physiol. 127, 1632–1639 (2019). 71. Romijn, J. A. et al. Regulation of endogenous fat and carbohydrate metabolism in relation
41. Bathgate, K. E. et al. Muscle health and performance in monozygotic twins with 30 years to exercise intensity and duration. Am. J. Physiol. 265, E380–E391 (1993).
of discordant exercise habits. Eur. J. Appl. Physiol. 118, 2097–2110 (2018). 72. van Loon, L. J., Greenhaff, P. L., Constantin-Teodosiu, D., Saris, W. H. & Wagenmakers, A. J.
This study of monozygotic twins suggests that fibre type plasticity in response to The effects of increasing exercise intensity on muscle fuel utilisation in humans. J. Physiol.
lifelong endurance training is greater than previously appreciated. 536, 295–304 (2001).
42. Malisoux, L., Francaux, M., Nielens, H. & Theisen, D. Stretch-shortening cycle exercises: 73. Duchateau, J. & Enoka, R. M. Human motor unit recordings: origins and insight into the
an effective training paradigm to enhance power output of human single muscle fibers. integrated motor system. Brain Res. 1409, 42–61 (2011).
J. Appl. Physiol. 100, 771–779 (2006). 74. Stienen, G. J., Kiers, J. L., Bottinelli, R. & Reggiani, C. Myofibrillar ATPase activity in
43. Sakakibara, I., Santolini, M., Ferry, A., Hakim, V. & Maire, P. Six homeoproteins and a skinned human skeletal muscle fibres: fibre type and temperature dependence.
Iinc-RNA at the fast MYH locus lock fast myofiber terminal phenotype. PLoS Genet. 10, J. Physiol. 493, 299–307 (1996).
e1004386 (2014). 75. Greenhaff, P. L. et al. The metabolic responses of human type I and II muscle fibres during
44. Trappe, S. et al. Single muscle fiber adaptations with marathon training. J. Appl. Physiol. maximal treadmill sprinting. J. Physiol. 478, 149–155 (1994).
101, 721–727 (2006). 76. Koh, H. E., Nielsen, J., Saltin, B., Holmberg, H. C. & Ortenblad, N. Pronounced limb and
45. Williamson, D. L., Gallagher, P. M., Carroll, C. C., Raue, U. & Trappe, S. W. Reduction fibre type differences in subcellular lipid droplet content and distribution in elite skiers
in hybrid single muscle fiber proportions with resistance training in humans. J. Appl. before and after exhaustive exercise. J. Physiol. 595, 5781–5795 (2017).
Physiol. 91, 1955–1961 (2001). 77. de Almeida, M. E. et al. Altered intramuscular network of lipid droplets and mitochondria
46. Long, K. et al. Identification of enhancers responsible for the coordinated expression in type 2 diabetes. Am. J. Physiol. Cell Physiol. 324, C39–C57 (2023).
of myosin heavy chain isoforms in skeletal muscle. BMC Genomics 23, 519 (2022). 78. Daemen, S. et al. Distinct lipid droplet characteristics and distribution unmask the
This study, published shortly after Dos Santos et al. (2022), also identifies the apparent contradiction of the athlete’s paradox. Mol. Metab. 17, 71–81 (2018).
fast-myosin super enhancer, as well as enhancers of slow-type myosin. 79. Brooks, G. A. & Mercier, J. Balance of carbohydrate and lipid utilization during exercise:
47. Balagopal, P., Schimke, J. C., Ades, P., Adey, D. & Nair, K. S. Age effect on transcript levels the “crossover” concept. J. Appl. Physiol. 76, 2253–2261 (1994).
and synthesis rate of muscle MHC and response to resistance exercise. Am. J. Physiol. 80. Wust, R. C. et al. Kinetic control of oxygen consumption during contractions in self-perfused
Endocrinol. Metab. 280, E203–E208 (2001). skeletal muscle. J. Physiol. 589, 3995–4009 (2011).
48. Lin, J. et al. Transcriptional co-activator PGC-1α drives the formation of slow-twitch 81. Ferguson, B. S. et al. Lactate metabolism: historical context, prior misinterpretations,
muscle fibres. Nature 418, 797–801 (2002). and current understanding. Eur. J. Appl. Physiol. 118, 691–728 (2018).
49. Kuhnen, G. et al. Genes whose gain or loss of function changes type 1, 2A, 2X, or 2B 82. Wescott, A. P., Kao, J. P. Y., Lederer, W. J. & Boyman, L. Voltage-energized calcium-sensitive
muscle fibre proportions in mice — a systematic review. Int. J. Mol. Sci. 23, 12933 (2022). ATP production by mitochondria. Nat. Metab. 1, 975–984 (2019).
50. Street, S. F. Lateral transmission of tension in frog myofibers: a myofibrillar network 83. Seiler, S. E. et al. Carnitine acetyltransferase mitigates metabolic inertia and muscle
and transverse cytoskeletal connections are possible transmitters. J. Cell. Physiol. 114, fatigue during exercise. Cell Metab. 22, 65–76 (2015).
346–364 (1983). 84. Mancilla, R. F. et al. Skeletal muscle mitochondrial inertia is associated with carnitine
51. Willingham, T. B., Kim, Y., Lindberg, E., Bleck, C. K. E. & Glancy, B. The unified myofibrillar acetyltransferase activity and physical function in humans. JCI Insight 8, e163855
matrix for force generation in muscle. Nat. Commun. 11, 3722 (2020). (2023).
The microscopic analysis of muscle myofibrils in this study provides a new paradigm 85. Timmons, J. A. et al. Substrate availability limits human skeletal muscle oxidative
for the organization of muscle contractile apparatus. ATP regeneration at the onset of ischemic exercise. J. Clin. Invest. 101, 79–85 (1998).
52. Ajayi, P. T. et al. Regulation of the evolutionarily conserved muscle myofibrillar matrix 86. Walsh, B. et al. The role of phosphorylcreatine and creatine in the regulation of
by cell type dependent and independent mechanisms. Nat. Commun. 13, 2661 (2022). mitochondrial respiration in human skeletal muscle. J. Physiol. 537, 971–978 (2001).
53. Katti, P. et al. Mitochondrial network configuration influences sarcomere and myosin 87. Roman, B. B., Meyer, R. A. & Wiseman, R. W. Phosphocreatine kinetics at the onset
filament structure in striated muscles. Nat. Commun. 13, 6058 (2022). of contractions in skeletal muscle of MM creatine kinase knockout mice. Am. J. Physiol.
54. Krustrup, P., Ferguson, R. A., Kjaer, M. & Bangsbo, J. ATP and heat production in human Cell Physiol. 283, C1776–C1783 (2002).
skeletal muscle during dynamic exercise: higher efficiency of anaerobic than aerobic 88. Katz, A., Broberg, S., Sahlin, K. & Wahren, J. Leg glucose uptake during maximal dynamic
ATP resynthesis. J. Physiol. 549, 255–269 (2003). exercise in humans. Am. J. Physiol. 251, E65–E70 (1986).
55. Zagatto, A. M. et al. Impacts of high-intensity exercise on the metabolomics profile of 89. Katz, A. A century of exercise physiology: key concepts in regulation of glycogen
human skeletal muscle tissue. Scand. J. Med. Sci. Sports 32, 402–413 (2022). metabolism in skeletal muscle. Eur. J. Appl. Physiol. 122, 1751–1772 (2022).
This study is an analysis of the immediate post-exercise skeletal muscle metabolome 90. Blazev, R. et al. Phosphoproteomics of three exercise modalities identifies canonical
in response to an exhaustive bout of high-intensity cycling. signaling and C18ORF25 as an AMPK substrate regulating skeletal muscle function.
56. Bleck, C. K. E., Kim, Y., Willingham, T. B. & Glancy, B. Subcellular connectomic analyses of Cell Metab. 34, 1561–1577 e1569 (2022).
energy networks in striated muscle. Nat. Commun. 9, 5111 (2018). Building on work from Hoffman et al. (2015), this study compares the phosphoproteomic
57. Glancy, B. et al. Mitochondrial reticulum for cellular energy distribution in muscle. Nature impact of an acute bout of endurance, resistance or sprinting exercise across a
523, 617–620 (2015). 3-h post-exercise recovery period.
This work advances the understanding of how membrane potential is dispersed across 91. Watt, M. J. et al. Regulation of HSL serine phosphorylation in skeletal muscle and
the mitochondrial reticulum to facilitate efficient ATP distribution in muscle. adipose tissue. Am. J. Physiol. Endocrinol. Metab. 290, E500–E508 (2006).
58. Glancy, B. et al. Power grid protection of the muscle mitochondrial reticulum. Cell Rep. 92. Richter, E. A., Ruderman, N. B., Gavras, H., Belur, E. R. & Galbo, H. Muscle glycogenolysis
19, 487–496 (2017). during exercise: dual control by epinephrine and contractions. Am. J. Physiol. 242,
59. Vincent, A. E. et al. Quantitative 3D mapping of the human skeletal muscle mitochondrial E25–E32 (1982).
network. Cell Rep. 26, 996–1009.e4 (2019). 93. Hureau, T. J. et al. On the role of skeletal muscle acidosis and inorganic phosphates
60. Caffrey, B. J. et al. Semi-automated 3D segmentation of human skeletal muscle as determinants of central and peripheral fatigue: a 31P-MRS study. J. Physiol. 600,
using focused ion beam-scanning electron microscopic images. J. Struct. Biol. 207, 3069–3081 (2022).
1–11 (2019). 94. Cheng, A. J., Place, N. & Westerblad, H. Molecular basis for exercise-induced fatigue:
61. Greggio, C. et al. Enhanced respiratory chain supercomplex formation in response to the importance of strictly controlled cellular Ca(2+) handling. Cold Spring Harb.
exercise in human skeletal muscle. Cell Metab. 25, 301–311 (2017). Perspect. Med. 8, a029710 (2018).

Nature Reviews Molecular Cell Biology

Review article

95. Rogatzki, M. J., Ferguson, B. S., Goodwin, M. L. & Gladden, L. B. Lactate is always the 126. Dube, J. J. et al. Adipose triglyceride lipase deletion from adipocytes, but not skeletal
end product of glycolysis. Front. Neurosci. 9, 22 (2015). myocytes, impairs acute exercise performance in mice. Am. J. Physiol. Endocrinol. Metab.
96. Hokken, R. et al. Subcellular localization- and fibre type-dependent utilization of muscle 308, E879–E890 (2015).
glycogen during heavy resistance exercise in elite power and Olympic weightlifters. 127. Watt, M. J., Heigenhauser, G. J. & Spriet, L. L. Effects of dynamic exercise intensity on the
Acta Physiol. 231, e13561 (2021). activation of hormone-sensitive lipase in human skeletal muscle. J. Physiol. 547, 301–308
97. Vigh-Larsen, J. F. et al. Fibre type- and localisation-specific muscle glycogen utilisation (2003).
during repeated high-intensity intermittent exercise. J. Physiol. 600, 4713–4730 (2022). 128. Donsmark, M., Langfort, J., Holm, C., Ploug, T. & Galbo, H. Contractions induce
98. Sidhu, S. K. et al. Fatigue-related group III/IV muscle afferent feedback facilitates phosphorylation of the AMPK site Ser565 in hormone-sensitive lipase in muscle.
intracortical inhibition during locomotor exercise. J. Physiol. 596, 4789–4801 (2018). Biochem. Biophys. Res. Commun. 316, 867–871 (2004).
99. Ducrocq, G. P. & Blain, G. M. Relationship between neuromuscular fatigue, muscle 129. Prats, C. et al. Decrease in intramuscular lipid droplets and translocation of HSL in
activation and the work done above the critical power during severe-intensity exercise. response to muscle contraction and epinephrine. J. Lipid Res. 47, 2392–2399 (2006).
Exp. Physiol. 107, 312–325 (2022). 130. Stokie, J. R., Abbott, G., Howlett, K. F., Hamilton, D. L. & Shaw, C. S. Intramuscular lipid
100. Goncalves, R. L., Quinlan, C. L., Perevoshchikova, I. V., Hey-Mogensen, M. & Brand, M. D. utilization during exercise: a systematic review, meta-analysis, and meta-regression.
Sites of superoxide and hydrogen peroxide production by muscle mitochondria J. Appl. Physiol. 134, 581–592 (2023).
assessed ex vivo under conditions mimicking rest and exercise. J. Biol. Chem. 290, 131. Ghafouri, K. et al. Moderate exercise increases affinity of large very low-density
209–227 (2015). lipoproteins for hydrolysis by lipoprotein lipase. J. Clin. Endocrinol. Metab. 100,
101. Henriquez-Olguin, C. et al. Cytosolic ROS production by NADPH oxidase 2 regulates 2205–2213 (2015).
muscle glucose uptake during exercise. Nat. Commun. 10, 4623 (2019). 132. Jain, S. S. et al. Additive effects of insulin and muscle contraction on fatty acid transport
102. Sakellariou, G. K. et al. Studies of mitochondrial and nonmitochondrial sources implicate and fatty acid transporters, FAT/CD36, FABPpm, FATP1, 4 and 6. FEBS Lett. 583,
nicotinamide adenine dinucleotide phosphate oxidase(s) in the increased skeletal muscle 2294–2300 (2009).
superoxide generation that occurs during contractile activity. Antioxid. Redox Signal. 18, 133. Nickerson, J. G. et al. Greater transport efficiencies of the membrane fatty acid
603–621 (2013). transporters FAT/CD36 and FATP4 compared with FABPpm and FATP1 and differential
103. Sun, Q. A. et al. Oxygen-coupled redox regulation of the skeletal muscle ryanodine effects on fatty acid esterification and oxidation in rat skeletal muscle. J. Biol. Chem. 284,
receptor-Ca2+ release channel by NADPH oxidase 4. Proc. Natl Acad. Sci. USA 108, 16522–16530 (2009).
16098–16103 (2011). 134. Jeppesen, J. et al. Contraction-induced skeletal muscle FAT/CD36 trafficking and FA
104. Xirouchaki, C. E. et al. Skeletal muscle NOX4 is required for adaptive responses that uptake is AMPK independent. J. Lipid Res. 52, 699–711 (2011).
prevent insulin resistance. Sci. Adv. 7, eabl4988 (2021). 135. Abbott, M. J., Edelman, A. M. & Turcotte, L. P. CaMKK is an upstream signal of AMP-
105. Bouviere, J. et al. Exercise-stimulated ROS sensitive signaling pathways in skeletal activated protein kinase in regulation of substrate metabolism in contracting skeletal
muscle. Antioxidants 10, 537 (2021). muscle. Am. J. Physiol. Regul. Integr. Comp. Physiol. 297, R1724–R1732 (2009).
106. Ristow, M. et al. Antioxidants prevent health-promoting effects of physical exercise in 136. Turcotte, L. P., Raney, M. A. & Todd, M. K. ERK1/2 inhibition prevents contraction-induced
humans. Proc. Natl Acad. Sci. USA 106, 8665–8670 (2009). increase in plasma membrane FAT/CD36 content and FA uptake in rodent muscle.
107. Clifford, T., Jeffries, O., Stevenson, E. J. & Davies, K. A. B. The effects of vitamin C and E Acta Physiol. Scand. 184, 131–139 (2005).
on exercise-induced physiological adaptations: a systematic review and Meta-analysis 137. Holloway, G. P. et al. Fatty acid binding protein facilitates sarcolemmal fatty acid
of randomized controlled trials. Crit. Rev. Food Sci. Nutr. 60, 3669–3679 (2020). transport but not mitochondrial oxidation in rat and human skeletal muscle. J. Physiol.
108. Gonzalez, J. T. et al. Ingestion of glucose or sucrose prevents liver but not muscle 582, 393–405 (2007).
glycogen depletion during prolonged endurance-type exercise in trained cyclists. 138. Sacchetti, M., Saltin, B., Osada, T. & van Hall, G. Intramuscular fatty acid metabolism
Am. J. Physiol. Endocrinol. Metab. 309, E1032–E1039 (2015). in contracting and non-contracting human skeletal muscle. J. Physiol. 540, 387–395
109. Vincent, M. A. et al. Mixed meal and light exercise each recruit muscle capillaries in (2002).
healthy humans. Am. J. Physiol. Endocrinol. Metab. 290, E1191–E1197 (2006). 139. Hall, A. M., Wiczer, B. M., Herrmann, T., Stremmel, W. & Bernlohr, D. A. Enzymatic properties
110. MacLean, D. A., Bangsbo, J. & Saltin, B. Muscle interstitial glucose and lactate levels of purified murine fatty acid transport protein 4 and analysis of acyl-CoA synthetase
during dynamic exercise in humans determined by microdialysis. J. Appl. Physiol. 87, activities in tissues from FATP4 null mice. J. Biol. Chem. 280, 11948–11954 (2005).
1483–1490 (1999). 140. Li, L. O. et al. Compartmentalized acyl-CoA metabolism in skeletal muscle regulates
111. Chow, L. S. et al. Exerkines in health, resilience and disease. Nat. Rev. Endocrinol. 18, systemic glucose homeostasis. Diabetes 64, 23–35 (2015).
273–289 (2022). 141. Chen, S., Zhou, L., Sun, J., Qu, Y. & Chen, M. The role of cAMP-PKA pathway in lactate-
112. Ryder, J. W. et al. Postexercise glucose uptake and glycogen synthesis in skeletal muscle induced intramuscular triglyceride accumulation and mitochondria content increase
from GLUT4-deficient mice. FASEB J. 13, 2246–2256 (1999). in mice. Front. Physiol. 12, 709135 (2021).
113. Sylow, L., Kleinert, M., Richter, E. A. & Jensen, T. E. Exercise-stimulated glucose uptake — 142. Constantin-Teodosiu, D., Howell, S. & Greenhaff, P. L. Carnitine metabolism in human
regulation and implications for glycaemic control. Nat. Rev. Endocrinol. 13, 133–148 muscle fiber types during submaximal dynamic exercise. J. Appl. Physiol. 80, 1061–1064
(2017). (1996).
114. Sylow, L. et al. Rac1 and AMPK account for the majority of muscle glucose uptake 143. Hostrup, M. et al. High-intensity interval training remodels the proteome and acetylome
stimulated by ex vivo contraction but not in vivo exercise. Diabetes 66, 1548–1559 of human skeletal muscle. eLife 11, e69802 (2022).
(2017). 144. Adamovich, Y. et al. Clock proteins and training modify exercise capacity in a daytime-
115. Kjobsted, R. et al. AMPK and TBC1D1 regulate muscle glucose uptake after, but not dependent manner. Proc. Natl Acad. Sci. USA 118, e2101115118 (2021).
during, exercise and contraction. Diabetes 68, 1427–1440 (2019). 145. Bano-Otalora, B. et al. Bright daytime light enhances circadian amplitude in a diurnal
116. Kjobsted, R. et al. Enhanced muscle insulin sensitivity after contraction/exercise is mammal. Proc. Natl Acad. Sci. USA 118, e2100094118 (2021).
mediated by AMPK. Diabetes 66, 598–612 (2017). 146. Martin, R. A., Viggars, M. R. & Esser, K. A. Metabolism and exercise: the skeletal muscle clock
117. Fritzen, A. M. et al. 5′-AMP activated protein kinase α2 controls substrate metabolism takes centre stage. Nat. Rev. Endocrinol. https://doi.org/10.1038/s41574-023-00805-8 (2023).
during post-exercise recovery via regulation of pyruvate dehydrogenase kinase 4. 147. Dyar, K. A. et al. Muscle insulin sensitivity and glucose metabolism are controlled
J. Physiol. 593, 4765–4780 (2015). by the intrinsic muscle clock. Mol. Metab. 3, 29–41 (2014).
118. Ouyang, Q. et al. Rab8a as a mitochondrial receptor for lipid droplets in skeletal muscle. 148. Dyar, K. A. et al. Transcriptional programming of lipid and amino acid metabolism by
Dev. Cell 58, 289–305.e6 (2023). the skeletal muscle circadian clock. PLoS Biol. 16, e2005886 (2018).
119. Glancy, B. et al. Mitochondrial lactate metabolism: history and implications for exercise 149. Hodge, B. A. et al. The endogenous molecular clock orchestrates the temporal
and disease. J. Physiol. 599, 863–888 (2021). separation of substrate metabolism in skeletal muscle. Skelet. Muscle 5, 17 (2015).
120. Rothschild, J. A., Kilding, A. E., Stewart, T. & Plews, D. J. Factors influencing substrate 150. Harfmann, B. D. et al. Muscle-specific loss of Bmal1 leads to disrupted tissue glucose
oxidation during submaximal cycling: a modelling analysis. Sports Med. 52, 2775–2795 metabolism and systemic glucose homeostasis. Skelet. Muscle 6, 12 (2016).
(2022). 151. Thomas, J. M. et al. Circadian rhythm phase shifts caused by timed exercise vary with
121. King, A., Helms, E., Zinn, C. & Jukic, I. The ergogenic effects of acute carbohydrate chronotype. JCI Insight 5, e134270 (2020).
feeding on resistance exercise performance: a systematic review and meta-analysis. 152. Sun, S. et al. A single-cell transcriptomic atlas of exercise-induced anti-inflammatory
Sports Med. 52, 2691–2712 (2022). and geroprotective effects across the body. Innovation 4, 100380 (2023).
122. Hartley, C., Carr, A., Bowe, S. J., Bredie, W. L. P. & Keast, R. S. J. Maltodextrin-based 153. Small, L. et al. Contraction influences Per2 gene expression in skeletal muscle through
carbohydrate oral rinsing and exercise performance: systematic review and meta- a calcium-dependent pathway. J. Physiol. 598, 5739–5752 (2020).
analysis. Sports Med. 52, 1833–1862 (2022). 154. Zambon, A. C. et al. Time- and exercise-dependent gene regulation in human skeletal
123. Gant, N., Stinear, C. M. & Byblow, W. D. Carbohydrate in the mouth immediately facilitates muscle. Genome Biol. 4, R61 (2003).
motor output. Brain Res. 1350, 151–158 (2010). 155. Dyar, K. A. et al. The calcineurin-NFAT pathway controls activity-dependent circadian
124. Morville, T., Sahl, R. E., Moritz, T., Helge, J. W. & Clemmensen, C. Plasma metabolome gene expression in slow skeletal muscle. Mol. Metab. 4, 823–833 (2015).
profiling of resistance exercise and endurance exercise in humans. Cell Rep. 33, 108554 156. Gabriel, B. M. & Zierath, J. R. Circadian rhythms and exercise — re-setting the clock
(2020). in metabolic disease. Nat. Rev. Endocrinol. 15, 197–206 (2019).
125. Alsted, T. J. et al. Contraction-induced lipolysis is not impaired by inhibition of 157. Harmsen, J. F. et al. Circadian misalignment disturbs the skeletal muscle lipidome in
hormone-sensitive lipase in skeletal muscle. J. Physiol. 591, 5141–5155 (2013). healthy young men. FASEB J. 35, e21611 (2021).

Nature Reviews Molecular Cell Biology

Review article

158. Wefers, J. et al. Circadian misalignment induces fatty acid metabolism gene profiles and 187. Drake, J. C. et al. Mitochondria-localized AMPK responds to local energetics and
compromises insulin sensitivity in human skeletal muscle. Proc. Natl Acad. Sci. USA 115, contributes to exercise and energetic stress-induced mitophagy. Proc. Natl Acad. Sci. USA
7789–7794 (2018). 118, e2025932118 (2021).
159. Morris, C. J., Purvis, T. E., Hu, K. & Scheer, F. A. Circadian misalignment increases 188. Laker, R. C. et al. AMPK phosphorylation of Ulk1 is required for targeting of mitochondria
cardiovascular disease risk factors in humans. Proc. Natl Acad. Sci. USA 113, E1402–E1411 to lysosomes in exercise-induced mitophagy. Nat. Commun. 8, 548 (2017).
(2016). 189. Toyama, E. Q. et al. Metabolism. AMP-activated protein kinase mediates mitochondrial
160. Saner, N. J. et al. Exercise mitigates sleep-loss-induced changes in glucose tolerance, fission in response to energy stress. Science 351, 275–281 (2016).
mitochondrial function, sarcoplasmic protein synthesis, and diurnal rhythms. Mol. Metab. 190. VerPlank, J. J. S., Lokireddy, S., Zhao, J. & Goldberg, A. L. 26S Proteasomes are rapidly
43, 101110 (2021). activated by diverse hormones and physiological states that raise cAMP and cause Rpn6
161. Sato, S. et al. Time of exercise specifies the impact on muscle metabolic pathways and phosphorylation. Proc. Natl Acad. Sci. USA 116, 4228–4237 (2019).
systemic energy homeostasis. Cell Metab. 30, 92–110.e4 (2019). 191. Parker, B. L., Kiens, B., Wojtaszewski, J. F. P., Richter, E. A. & James, D. E. Quantification
162. Mancilla, R. et al. Exercise training elicits superior metabolic effects when performed in of exercise-regulated ubiquitin signaling in human skeletal muscle identifies protein
the afternoon compared to morning in metabolically compromised humans. Physiol. Rep. modification cross talk via NEDDylation. FASEB J. 34, 5906–5916 (2020).
8, e14669 (2021). 192. Seaborne, R. A. & Sharples, A. P. The interplay between exercise metabolism, epigenetics,
163. van der Velde, J. et al. Timing of physical activity in relation to liver fat content and insulin and skeletal muscle remodeling. Exerc. Sport Sci. Rev. 48, 188–200 (2020).
resistance. Diabetologia 66, 461–471 (2023). 193. McGee, S. L., Fairlie, E., Garnham, A. P. & Hargreaves, M. Exercise-induced histone
164. Savikj, M. et al. Afternoon exercise is more efficacious than morning exercise at modifications in human skeletal muscle. J. Physiol. 587, 5951–5958 (2009).
improving blood glucose levels in individuals with type 2 diabetes: a randomised 194. Backs, J., Song, K., Bezprozvannaya, S., Chang, S. & Olson, E. N. CaM kinase II selectively
crossover trial. Diabetologia 62, 233–237 (2019). signals to histone deacetylase 4 during cardiomyocyte hypertrophy. J. Clin. Invest. 116,
165. Feng, H. et al. Associations of timing of physical activity with all-cause and cause-specific 1853–1864 (2006).
mortality in a prospective cohort study. Nat. Commun. 14, 930 (2023). 195. Wu, X. et al. Local InsP3-dependent perinuclear Ca2+ signaling in cardiac myocyte
166. Fischer, D., Lombardi, D. A., Marucci-Wellman, H. & Roenneberg, T. Chronotypes in the excitation-transcription coupling. J. Clin. Invest. 116, 675–682 (2006).
US — influence of age and sex. PLoS ONE 12, e0178782 (2017). 196. Solagna, F. et al. Exercise-dependent increases in protein synthesis are accompanied by
167. Amar, D. et al. Time trajectories in the transcriptomic response to exercise — a meta-analysis. chromatin modifications and increased MRTF-SRF signalling. Acta Physiol. 230, e13496
Nat. Commun. 12, 3471 (2021). (2020).
168. Pillon, N. J. et al. Distinctive exercise-induced inflammatory response and exerkine 197. Yin, Y. et al. Impact of cytosine methylation on DNA binding specificities of human
induction in skeletal muscle of people with type 2 diabetes. Sci. Adv. 8, eabo3192 (2022). transcription factors. Science 356, eaaj2239 (2017).
169. Makhnovskii, P. A. et al. Alternative transcription start sites contribute to acute-stress- 198. Figueiredo, V. C. et al. Genetic and epigenetic regulation of skeletal muscle ribosome
induced transcriptome response in human skeletal muscle. Hum. Genomics 16, 24 biogenesis with exercise. J. Physiol. 599, 3363–3384 (2021).
(2022). 199. Seaborne, R. A. et al. Human skeletal muscle possesses an epigenetic memory of
170. Hoffman, N. J. et al. Global phosphoproteomic analysis of human skeletal muscle reveals hypertrophy. Sci. Rep. 8, 1898 (2018).
a network of exercise-regulated kinases and AMPK substrates. Cell Metab. 22, 922–935 This comprehensive interrogation of the resistance exercise-induced methylome
(2015). provides an epigenetic hypothesis for ‘muscle memory’.
This seminal study investigates the exercise-induced human muscle phosphoproteome. 200. Sexton, C. L. et al. Skeletal muscle DNA methylation and mRNA responses to a bout of
171. Nelson, M. E. et al. Phosphoproteomics reveals conserved exercise-stimulated signaling higher versus lower load resistance exercise in previously trained men. Cells 12, 263 (2023).
and AMPK regulation of store-operated calcium entry. EMBO J. 38, e102578 (2019). 201. Lindholm, M. E. et al. An integrative analysis reveals coordinated reprogramming of the
172. You, J. S. et al. The role of raptor in the mechanical load-induced regulation of mTOR epigenome and the transcriptome in human skeletal muscle after training. Epigenetics 9,
signaling, protein synthesis, and skeletal muscle hypertrophy. FASEB J. 33, 4021–4034 1557–1569 (2014).
(2019). 202. Maasar, M. F. et al. The comparative methylome and transcriptome after change of
173. You, J. S. et al. A DGKζ-FoxO-ubiquitin proteolytic axis controls fiber size during skeletal direction compared to straight line running exercise in human skeletal muscle. Front.
muscle remodeling. Sci. Signal. 11, eaao6847 (2018). Physiol. 12, 619447 (2021).
174. Song, Z. et al. Resistance exercise initiates mechanistic target of rapamycin (mTOR) 203. Galle, E. et al. H3K18 lactylation marks tissue-specific active enhancers. Genome Biol. 23,
translocation and protein complex co-localisation in human skeletal muscle. Sci. Rep. 7, 207 (2022).
5028 (2017). 204. Massart, J. et al. Endurance exercise training-responsive miR-19b-3p improves skeletal
175. Hodson, N., Mazzulla, M., Holowaty, M. N. H., Kumbhare, D. & Moore, D. R. RPS6 muscle glucose metabolism. Nat. Commun. 12, 5948 (2021).
phosphorylation occurs to a greater extent in the periphery of human skeletal muscle 205. Bonilauri, B. & Dallagiovanna, B. Long non-coding RNAs are differentially expressed after
fibers, near focal adhesions, after anabolic stimuli. Am. J. Physiol. Cell Physiol. 322, different exercise training programs. Front. Physiol. 11, 567614 (2020).
C94–C110 (2022). 206. Vechetti, I. J. Jr. et al. Mechanical overload-induced muscle-derived extracellular vesicles
176. Abou Sawan, S. et al. Trained integrated postexercise myofibrillar protein synthesis rates promote adipose tissue lipolysis. FASEB J. 35, e21644 (2021).
correlate with hypertrophy in young males and females. Med. Sci. Sports Exerc. 54, 207. Xhuti, D., Nilsson, M. I., Manta, K., Tarnopolsky, M. A. & Nederveen, J. P. Circulating
953–964 (2022). exosome-like vesicle and skeletal muscle microRNAs are altered with age and resistance
177. Needham, E. J. et al. Personalized phosphoproteomics identifies functional signaling. training. J. Physiol. https://doi.org/10.1113/JP282663 (2023).
Nat. Biotechnol. 40, 576–584 (2022). 208. Watanabe, S. et al. Skeletal muscle releases extracellular vesicles with distinct protein
178. Steinert, N. D. et al. Mapping of the contraction-induced phosphoproteome identifies and microRNA signatures that function in the muscle microenvironment. PNAS Nexus 1,
TRIM28 as a significant regulator of skeletal muscle size and function. Cell Rep. 34, pgac173 (2022).
108796 (2021). 209. Murach, K. A. et al. Fusion-independent satellite cell communication to muscle fibers
179. Nordgaard, C. et al. ZAKβ is activated by cellular compression and mediates contraction- during load-induced hypertrophy. Function 1, zqaa009 (2020).
induced MAP kinase signaling in skeletal muscle. EMBO J. 41, e111650 (2022). 210. Fry, C. S., Kirby, T. J., Kosmac, K., McCarthy, J. J. & Peterson, C. A. Myogenic progenitor
180. West, D. W. et al. Acute resistance exercise activates rapamycin-sensitive and -insensitive cells control extracellular matrix production by fibroblasts during skeletal muscle
mechanisms that control translational activity and capacity in skeletal muscle. J. Physiol. hypertrophy. Cell Stem Cell 20, 56–69 (2017).
594, 453–468 (2016). 211. Albanese, M. et al. MicroRNAs are minor constituents of extracellular vesicles that are
181. Lessard, S. J. et al. JNK regulates muscle remodeling via myostatin/SMAD inhibition. rarely delivered to target cells. PLoS Genet. 17, e1009951 (2021).
Nat. Commun. 9, 3030 (2018). 212. Wohlwend, M. et al. The exercise-induced long noncoding RNA CYTOR promotes
182. MacKenzie, M. G., Hamilton, D. L., Pepin, M., Patton, A. & Baar, K. Inhibition of myostatin fast-twitch myogenesis in aging. Sci. Transl Med. 13, eabc7367 (2021).
signaling through Notch activation following acute resistance exercise. PLoS ONE 8, 213. Nelson, B. R. et al. A peptide encoded by a transcript annotated as long noncoding RNA
e68743 (2013). enhances SERCA activity in muscle. Science 351, 271–275 (2016).
183. Pillon, N. J. et al. Transcriptomic profiling of skeletal muscle adaptations to exercise and 214. Anderson, D. M. et al. A micropeptide encoded by a putative long noncoding RNA
inactivity. Nat. Commun. 11, 470 (2020). regulates muscle performance. Cell 160, 595–606 (2015).
This meta-analysis curates an extensive (and frequently updated) library of the 215. Pinheiro, H. et al. mRNA distribution in skeletal muscle is associated with mRNA size.
skeletal muscle transcriptomic response to exercise across human demographics J. Cell Sci. 134, jcs256388 (2021).
and exercise modalities. 216. Denes, L. T., Kelley, C. P. & Wang, E. T. Microtubule-based transport is essential to
184. Schumann, M. et al. Compatibility of concurrent aerobic and strength training for distribute RNA and nascent protein in skeletal muscle. Nat. Commun. 12, 6079 (2021).
skeletal muscle size and function: an updated systematic review and meta-analysis. This study and Pinheiro et al. (2021) suggest that mRNAs are transported along
Sports Med. 52, 601–612 (2022). microtubules in muscle, which has notable implications for the regulation of the
185. Lundberg, T. R., Feuerbacher, J. F., Sunkeler, M. & Schumann, M. The effects of concurrent ‘myonuclear domain’.
aerobic and strength training on muscle fiber hypertrophy: a systematic review and 217. Scarborough, E. A. et al. Microtubules orchestrate local translation to enable cardiac
meta-analysis. Sports Med. 52, 2391–2403 (2022). growth. Nat. Commun. 12, 1547 (2021).
186. Kaiser, M. S. et al. Dual roles of mTORC1-dependent activation of the ubiquitin- 218. Taylor-Weiner, H. et al. Modeling the transport of nuclear proteins along single skeletal
proteasome system in muscle proteostasis. Commun. Biol. 5, 1141 (2022). muscle cells. Proc. Natl Acad. Sci. USA 117, 2978–2986 (2020).

Nature Reviews Molecular Cell Biology

Review article

219. Cutler, A. A., Jackson, J. B., Corbett, A. H. & Pavlath, G. K. Non-equivalence of nuclear 251. Hetlelid, K. J., Plews, D. J., Herold, E., Laursen, P. B. & Seiler, S. Rethinking the role of fat
import among nuclei in multinucleated skeletal muscle cells. J. Cell Sci. 131, jcs207670 oxidation: substrate utilisation during high-intensity interval training in well-trained and
(2018). recreationally trained runners. BMJ Open Sport Exerc. Med. 1, e000047 (2015).
220. Masschelein, E. et al. Exercise promotes satellite cell contribution to myofibers in a 252. Gehlert, S. et al. Effects of acute and chronic resistance exercise on the skeletal muscle
load-dependent manner. Skelet. Muscle 10, 21 (2020). metabolome. Metabolites 12, 445 (2022).
221. Borowik, A. K. et al. Skeletal muscle nuclei in mice are not post-mitotic. Function 4, 253. Ozaki, H., Loenneke, J. P., Thiebaud, R. S. & Abe, T. Resistance training induced increase
zqac059 (2023). in VO2max in young and older subjects. Eur. Rev. Aging Phys. Act. 10, 107–116 (2013).
This study provides the first compelling evidence that myonuclei can synthesize DNA 254. Eihara, Y. et al. Heavy resistance training versus plyometric training for improving running
and that this process can be augmented by muscle mechanical overload. economy and running time trial performance: a systematic review and meta-analysis.
222. Pearen, M. A. et al. The nuclear receptor, Nor-1, markedly increases type II oxidative Sports Med. Open 8, 138 (2022).
muscle fibers and resistance to fatigue. Mol. Endocrinol. 26, 372–384 (2012). 255. Damas, F. et al. Resistance training-induced changes in integrated myofibrillar protein
223. Correia, J. C., Ferreira, D. M. & Ruas, J. L. Intercellular: local and systemic actions of synthesis are related to hypertrophy only after attenuation of muscle damage. J. Physiol.
skeletal muscle PGC-1s. Trends Endocrinol. Metab. 26, 305–314 (2015). 594, 5209–5222 (2016).
224. Goode, J. M. et al. The nuclear receptor, Nor-1, induces the physiological responses 256. Saleem, A. & Hood, D. A. Acute exercise induces tumour suppressor protein p53
associated with exercise. Mol. Endocrinol. 30, 660–676 (2016). translocation to the mitochondria and promotes a p53-Tfam-mitochondrial DNA
225. Wu, Z. et al. Transducer of regulated CREB-binding proteins (TORCs) induce PGC-1α complex in skeletal muscle. J. Physiol. 591, 3625–3636 (2013).
transcription and mitochondrial biogenesis in muscle cells. Proc. Natl Acad. Sci. USA 257. Beyfuss, K., Erlich, A. T., Triolo, M. & Hood, D. A. The role of p53 in determining
103, 14379–14384 (2006). mitochondrial adaptations to endurance training in skeletal muscle. Sci. Rep. 8,
226. Wilson, T. E., Fahrner, T. J., Johnston, M. & Milbrandt, J. Identification of the DNA binding 14710 (2018).
site for NGFI-B by genetic selection in yeast. Science 252, 1296–1300 (1991). 258. Li, X. et al. Ultrasensitive sensors reveal the spatiotemporal landscape of lactate
227. Maira, M., Martens, C., Philips, A. & Drouin, J. Heterodimerization between members of metabolism in physiology and disease. Cell Metab. 35, 200–211.e9 (2023).
the Nur subfamily of orphan nuclear receptors as a novel mechanism for gene activation. 259. Martin, A. A. et al. Sarcomere dynamics revealed by a myofilament integrated FRET-based
Mol. Cell Biol. 19, 7549–7557 (1999). biosensor in live skeletal muscle fibers. Sci. Rep. 12, 18116 (2022).
228. Chao, L. C. et al. Skeletal muscle Nur77 expression enhances oxidative metabolism and 260. Fentz, J. et al. AMPKα is essential for acute exercise-induced gene responses but not
substrate utilization. J. Lipid Res. 53, 2610–2619 (2012). for exercise training-induced adaptations in mouse skeletal muscle. Am. J. Physiol.
229. Ruas, J. L. et al. A PGC-1α isoform induced by resistance training regulates skeletal Endocrinol. Metab. 309, E900–E914 (2015).
muscle hypertrophy. Cell 151, 1319–1331 (2012). 261. Ballmann, C., Tang, Y., Bush, Z. & Rowe, G. C. Adult expression of PGC-1α and -1β in
This is the first report that PGC1A4 is transcribed from the alternative PGC1A promoter skeletal muscle is not required for endurance exercise-induced enhancement of exercise
after resistance exercise and can stimulate muscle hypertrophy. capacity. Am. J. Physiol. Endocrinol. Metab. 311, E928–E938 (2016).
230. Barres, R. et al. Acute exercise remodels promoter methylation in human skeletal muscle. 262. Lu, T., Ang, C. E. & Zhuang, X. Spatially resolved epigenomic profiling of single cells in
Cell Metab. 15, 405–411 (2012). complex tissues. Cell 185, 4448–4464.e17 (2022).
231. Perez-Schindler, J. et al. RNA-bound PGC-1α controls gene expression in liquid-like 263. Gandin, V. et al. Cap-dependent translation initiation monitored in living cells. Nat.
nuclear condensates. Proc. Natl Acad. Sci. USA 118, e2105951118 (2021). Commun. 13, 6558 (2022).
232. Fan, W. et al. PPARδ promotes running endurance by preserving glucose. Cell Metab. 25, 264. Zhen, K. et al. A systematic review and meta-analysis on effects of aerobic exercise in
1186–1193.e4 (2017). people with Parkinson’s disease. NPJ Parkinsons Dis. 8, 146 (2022).
233. Fan, L. et al. Transcription factors KLF15 and PPARδ cooperatively orchestrate genome-wide 265. Yu, Q. et al. Comparative effectiveness of multiple exercise interventions in the treatment
regulation of lipid metabolism in skeletal muscle. J. Biol. Chem. 298, 101926 (2022). of mental health disorders: a systematic review and network meta-analysis. Sports Med.
234. Mammucari, C. et al. The mitochondrial calcium uniporter controls skeletal muscle Open 8, 135 (2022).
trophism in vivo. Cell Rep. 10, 1269–1279 (2015). 266. Battista, F. et al. Effect of exercise on cardiometabolic health of adults with overweight or
235. Koh, J. H. et al. Enhancement of anaerobic glycolysis — a role of PGC-1α4 in resistance obesity: focus on blood pressure, insulin resistance, and intrahepatic fat — a systematic
exercise. Nat. Commun. 13, 2324 (2022). review and meta-analysis. Obes. Rev. 22, e13269 (2021).
236. Wackerhage, H. et al. Does a hypertrophying muscle fibre reprogramme its metabolism 267. Nielsen, J. et al. Subcellular localization-dependent decrements in skeletal muscle
similar to a cancer cell? Sports Med. 52, 2569–2578 (2022). glycogen and mitochondria content following short-term disuse in young and old men.
237. Bohlen, J., Roiuk, M. & Teleman, A. A. Phosphorylation of ribosomal protein S6 Am. J. Physiol. Endocrinol. Metab. 299, E1053–E1060 (2010).
differentially affects mRNA translation based on ORF length. Nucleic Acids Res. 49, 268. Nielsen, J. et al. Increased subsarcolemmal lipids in type 2 diabetes: effect of training on
13062–13074 (2021). localization of lipids, mitochondria, and glycogen in sedentary human skeletal muscle.
238. Chaillou, T., Zhang, X. & McCarthy, J. J. Expression of muscle-specific ribosomal protein Am. J. Physiol. Endocrinol. Metab. 298, E706–E713 (2010).
L3-like impairs myotube growth. J. Cell. Physiol. 231, 1894–1902 (2016). 269. Yi, J. et al. Mitochondrial calcium uptake regulates rapid calcium transients in skeletal
239. Granata, C., Oliveira, R. S. F., Little, J. P. & Bishop, D. J. Forty high-intensity interval muscle during excitation-contraction (E-C) coupling. J. Biol. Chem. 286, 32436–32443
training sessions blunt exercise-induced changes in the nuclear protein content of (2011).
PGC-1α and p53 in human skeletal muscle. Am. J. Physiol. Endocrinol. Metab. 318, 270. De Stefani, D., Raffaello, A., Teardo, E., Szabo, I. & Rizzuto, R. A forty-kilodalton protein of the
E224–E236 (2020). inner membrane is the mitochondrial calcium uniporter. Nature 476, 336–340 (2011).
240. Brook, M. S. et al. Synchronous deficits in cumulative muscle protein synthesis and 271. Baughman, J. M. et al. Integrative genomics identifies MCU as an essential component of
ribosomal biogenesis underlie age-related anabolic resistance to exercise in humans. the mitochondrial calcium uniporter. Nature 476, 341–345 (2011).
J. Physiol. 594, 7399–7417 (2016). This reference and De Stefani et al. (2011) co-identify the core component of the
241. Norrbom, J. M. et al. A HIF-1 signature dominates the attenuation in the human skeletal elusive mitochondrial calcium uniporter.
muscle transcriptional response to high-intensity interval training. J. Appl. Physiol. 132, 272. Denton, R. M., McCormack, J. G. & Edgell, N. J. Role of calcium ions in the regulation
1448–1459 (2022). of intramitochondrial metabolism. Effects of Na+, Mg2+ and ruthenium red on the Ca2+-
242. Stokes, T. et al. Molecular transducers of human skeletal muscle remodeling under stimulated oxidation of oxoglutarate and on pyruvate dehydrogenase activity in intact
different loading states. Cell Rep. 32, 107980 (2020). rat heart mitochondria. Biochem. J. 190, 107–117 (1980).
243. Porter, C., Reidy, P. T., Bhattarai, N., Sidossis, L. S. & Rasmussen, B. B. Resistance exercise 273. Turkan, A., Hiromasa, Y. & Roche, T. E. Formation of a complex of the catalytic subunit of
training alters mitochondrial function in human skeletal muscle. Med. Sci. Sports Exerc. pyruvate dehydrogenase phosphatase isoform 1 (PDP1c) and the L2 domain forms a Ca2+
47, 1922–1931 (2015). binding site and captures PDP1c as a monomer. Biochemistry 43, 15073–15085 (2004).
244. Granata, C., Jamnick, N. A. & Bishop, D. J. Training-induced changes in mitochondrial content 274. Rutter, G. A. & Denton, R. M. The binding of Ca2+ ions to pig heart NAD+-isocitrate
and respiratory function in human skeletal muscle. Sports Med. 48, 1809–1828 (2018). dehydrogenase and the 2-oxoglutarate dehydrogenase complex. Biochem. J. 263,
245. Andrade-Souza, V. A. et al. Exercise twice-a-day potentiates markers of mitochondrial 453–462 (1989).
biogenesis in men. FASEB J. 34, 1602–1619 (2020). 275. Denton, R. M., Pullen, T. J., Armstrong, C. T., Heesom, K. J. & Rutter, G. A. Calcium-insensitive
246. Soto, I. et al. Balanced mitochondrial and cytosolic translatomes underlie the biogenesis splice variants of mammalian E1 subunit of 2-oxoglutarate dehydrogenase complex with
of human respiratory complexes. Genome Biol. 23, 170 (2022). tissue-specific patterns of expression. Biochem. J. 473, 1165–1178 (2016).
247. Liang, X. et al. Exercise inducible lactate dehydrogenase B regulates mitochondrial 276. Sembrowich, W. L., Quintinskie, J. J. & Li, G. Calcium uptake in mitochondria from
function in skeletal muscle. J. Biol. Chem. 291, 25306–25318 (2016). different skeletal muscle types. J. Appl. Physiol. 59, 137–141 (1985).
248. Thomas, A. C. Q. et al. Short-term aerobic conditioning prior to resistance training 277. Sidossis, L. S., Gastaldelli, A., Klein, S. & Wolfe, R. R. Regulation of plasma fatty acid
augments muscle hypertrophy and satellite cell content in healthy young men and oxidation during low- and high-intensity exercise. Am. J. Physiol. 272, E1065–E1070
women. FASEB J. 36, e22500 (2022). (1997).
249. Binet, E. R. et al. Sex-based comparisons of muscle cellular adaptations after 10 weeks of 278. Wang, Y. et al. Saturation of the mitochondrial NADH shuttles drives aerobic glycolysis in
progressive resistance training in middle-aged adults. J. Appl. Physiol. 134, 116–129 (2023). proliferating cells. Mol. Cell 82, 3270–3283.e9 (2022).
250. Verdijk, L. B., Snijders, T., Holloway, T. M., Van Kranenburg, J. & Van Loon, L. J. C. Resistance 279. Jäger, S., Handschin, C., St-Pierre, J. & Spiegelman, B. M. AMP-activated protein kinase
training increases skeletal muscle capillarization in healthy older men. Med. Sci. Sports (AMPK) action in skeletal muscle via direct phosphorylation of PGC-1α. Proc. Natl Acad.
Exerc. 48, 2157–2164 (2016). Sci. USA 104, 12017–12022 (2007).

Nature Reviews Molecular Cell Biology

Review article

280. Arany, Z. et al. HIF-independent regulation of VEGF and angiogenesis by the 309. Dungan, C. M. et al. Deletion of SA β-Gal+ cells using senolytics improves muscle
transcriptional coactivator PGC-1α. Nature 451, 1008–1012 (2008). regeneration in old mice. Aging Cell 21, e13528 (2022).
281. Miyake, T. & McDermott, J. C. Re-organization of nucleolar architecture in myogenic 310. Dungan, C. M. et al. Senolytic treatment rescues blunted muscle hypertrophy in old
differentiation. J. Cell Sci. 136, jcs260496 (2023). mice. Geroscience 44, 1925–1940 (2022).
282. Emilio, E. J., Hita-Contreras, F., Jimenez-Lara, P. M., Latorre-Roman, P. & Martinez-Amat, A. 311. Englund, D. A. et al. p21 induces a senescence program and skeletal muscle dysfunction.
The association of flexibility, balance, and lumbar strength with balance ability: risk of Mol. Metab. 67, 101652 (2023).
falls in older adults. J. Sports Sci. Med. 13, 349–357 (2014). 312. Sailani, M. R. et al. Lifelong physical activity is associated with promoter hypomethylation
283. Roberts, B. M., Nuckols, G. & Krieger, J. W. Sex differences in resistance training: of genes involved in metabolism, myogenesis, contractile properties and oxidative stress
a systematic review and meta-analysis. J. Strength. Cond. Res. 34, 1448–1460 (2020). resistance in aged human skeletal muscle. Sci. Rep. 9, 3272 (2019).
284. Schoenfeld, B. J., Ogborn, D. & Krieger, J. W. Dose-response relationship between weekly 313. Ringholm, S. et al. Impact of aging and lifelong exercise training on mitochondrial
resistance training volume and increases in muscle mass: a systematic review and function and network connectivity in human skeletal muscle. J. Gerontol. A Biol. Sci.
meta-analysis. J. Sports Sci. 35, 1073–1082 (2017). Med. Sci. 78, 373–383 (2023).
285. Baz-Valle, E., Balsalobre-Fernandez, C., Alix-Fages, C. & Santos-Concejero, J. A systematic 314. Ruple, B. A. et al. Resistance training rejuvenates the mitochondrial methylome in aged
review of the effects of different resistance training volumes on muscle hypertrophy. human skeletal muscle. FASEB J. 35, e21864 (2021).
J. Hum. Kinet. 81, 199–210 (2022). 315. Vinel, C. et al. The exerkine apelin reverses age-associated sarcopenia. Nat. Med. 24,
286. Schoenfeld, B. J., Grgic, J. & Krieger, J. How many times per week should a muscle 1360–1371 (2018).
be trained to maximize muscle hypertrophy? A systematic review and meta-analysis 316. Lee, U. et al. A Tead1-Apelin axis directs paracrine communication from myogenic to
of studies examining the effects of resistance training frequency. J. Sports Sci. 37, endothelial cells in skeletal muscle. iScience 25, 104589 (2022).
1286–1295 (2019). 317. O’Bryan, S. J. et al. Progressive resistance training for concomitant increases in muscle
287. Carvalho, L. et al. Muscle hypertrophy and strength gains after resistance training with strength and bone mineral density in older adults: a systematic review and meta-analysis.
different volume-matched loads: a systematic review and meta-analysis. Appl. Physiol. Sports Med. 52, 1939–1960 (2022).
Nutr. Metab. 47, 357–368 (2022). 318. D’Hulst, G. et al. PHD1 controls muscle mTORC1 in a hydroxylation-independent manner
288. Refalo, M. C., Helms, E. R., Trexler, E. T., Hamilton, D. L. & Fyfe, J. J. Influence of resistance by stabilizing leucyl tRNA synthetase. Nat. Commun. 11, 174 (2020).
training proximity-to-failure on skeletal muscle hypertrophy: a systematic review with 319. Balachandran, A. T. et al. Comparison of power training vs traditional strength training on
meta-analysis. Sports Med. 53, 649–665 (2023). physical function in older adults: a systematic review and meta-analysis. JAMA Netw. Open
289. Plotkin, D. et al. Progressive overload without progressing load? The effects of load or 5, e2211623 (2022).
repetition progression on muscular adaptations. PeerJ 10, e14142 (2022). 320. McKenna, M. J., Morton, J., Selig, S. E. & Snow, R. J. Creatine supplementation increases
290. Bohm, S., Mersmann, F. & Arampatzis, A. Human tendon adaptation in response to muscle total creatine but not maximal intermittent exercise performance. J. Appl. Physiol.
mechanical loading: a systematic review and meta-analysis of exercise intervention 87, 2244–2252 (1999).
studies on healthy adults. Sports Med. Open 1, 7 (2015). 321. Hornemann, T. et al. Muscle-type creatine kinase interacts with central domains of the
291. Nunes, E. A. et al. Systematic review and meta-analysis of protein intake to support M-band proteins myomesin and M-protein. J. Mol. Biol. 332, 877–887 (2003).
muscle mass and function in healthy adults. J. Cachexia Sarcopenia Muscle 13, 795–810 322. Kraft, T., Hornemann, T., Stolz, M., Nier, V. & Wallimann, T. Coupling of creatine kinase to
(2022). glycolytic enzymes at the sarcomeric I-band of skeletal muscle: a biochemical study in
292. Mattioni Maturana, F., Martus, P., Zipfel, S. & AM, N. I. Effectiveness of HIIE versus MICT in situ. J. Muscle Res. Cell Motil. 21, 691–703 (2000).
improving cardiometabolic risk factors in health and disease: a meta-analysis. Med. Sci. 323. Perry, C. G. et al. Mitochondrial creatine kinase activity and phosphate shuttling are
Sports Exerc. 53, 559–573 (2021). acutely regulated by exercise in human skeletal muscle. J. Physiol. 590, 5475–5486
293. Milanovic, Z., Sporis, G. & Weston, M. Effectiveness of high-intensity interval training (HIT) (2012).
and continuous endurance training for VO2max improvements: a systematic review and 324. Hultman, E., Soderlund, K., Timmons, J. A., Cederblad, G. & Greenhaff, P. L. Muscle
meta-analysis of controlled trials. Sports Med. 45, 1469–1481 (2015). creatine loading in men. J. Appl. Physiol. 81, 232–237 (1996).
294. Scribbans, T. D., Vecsey, S., Hankinson, P. B., Foster, W. S. & Gurd, B. J. The effect of training 325. de Souza, E. S. A. et al. Effects of creatine supplementation on renal function:
intensity on VO2max in young healthy adults: a meta-regression and meta-analysis. a systematic review and meta-analysis. J. Ren. Nutr. 29, 480–489 (2019).
Int. J. Exerc. Sci. 9, 230–247 (2016). 326. Lanhers, C. et al. Creatine supplementation and lower limb strength performance:
295. Jamnick, N. A., Pettitt, R. W., Granata, C., Pyne, D. B. & Bishop, D. J. An examination and a systematic review and meta-analyses. Sports Med. 45, 1285–1294 (2015).
critique of current methods to determine exercise intensity. Sports Med. 50, 1729–1756 327. Prokopidis, K. et al. Effects of creatine supplementation on memory in healthy
(2020). individuals: a systematic review and meta-analysis of randomized controlled trials.
296. Iannetta, D. et al. A critical evaluation of current methods for exercise prescription in Nutr. Rev. 81, 416–427 (2023).
women and men. Med. Sci. Sports Exerc. 52, 466–473 (2020). 328. Areta, J. L. & Hopkins, W. G. Skeletal muscle glycogen content at rest and during
.
297. Hov, H. et al. Aerobic high-intensity intervals are superior to improve VO 2max compared endurance exercise in humans: a meta-analysis. Sports Med. 48, 2091–2102 (2018).
with sprint intervals in well-trained men. Scand. J. Med. Sci. Sports 33, 146–159 (2023). 329. Nielsen, J., Dubillot, P., Stausholm, M. H. & Ortenblad, N. Specific ATPases drive
298. Hubal, M. J. et al. Variability in muscle size and strength gain after unilateral resistance compartmentalized glycogen utilization in rat skeletal muscle. J. Gen. Physiol. 154,
training. Med. Sci. Sports Exerc. 37, 964–972 (2005). e202113071 (2022).
Similar to Bouchard et al. (1999), this study clearly illustrates interindividual This study finds that specific depots of glycogen fuel different ATPases to varying
heterogeniety in exercise adaptation to a given resistance training programme. extents in skeletal muscle.
299. Vollaard, N. B. et al. Systematic analysis of adaptations in aerobic capacity and 330. Vigh-Larsen, J. F. et al. The role of muscle glycogen content and localization in high-
submaximal energy metabolism provides a unique insight into determinants of human intensity exercise performance: a placebo-controlled trial. Med. Sci. Sports Exerc. 54,
aerobic performance. J. Appl. Physiol. 106, 1479–1486 (2009). 2073–2086 (2022).
300. Bouchard, C. et al. Familial aggregation of VO2max response to exercise training: results 331. Gejl, K. D. et al. Muscle glycogen content modifies SR Ca2+ release rate in elite endurance
from the HERITAGE Family Study. J. Appl. Physiol. 87, 1003–1008 (1999). athletes. Med. Sci. Sports Exerc. 46, 496–505 (2014).
This is a seminal work on the contribution of genetics towards endurance training- 332. Impey, S. G. et al. Fuel for the work required: a theoretical framework for carbohydrate
induced changes in aerobic fitness. periodization and the glycogen threshold hypothesis. Sports Med. 48, 1031–1048
301. Montero, D. & Lundby, C. Refuting the myth of non-response to exercise training: (2018).
‘non-responders’ do respond to higher dose of training. J. Physiol. 595, 3377–3387 (2017). 333. Marquet, L. A. et al. Enhanced endurance performance by periodization of carbohydrate
302. Ross, R., de Lannoy, L. & Stotz, P. J. Separate effects of intensity and amount of exercise intake: “sleep low” strategy. Med. Sci. Sports Exerc. 48, 663–672 (2016).
on interindividual cardiorespiratory fitness response. Mayo Clin. Proc. 90, 1506–1514 334. Tarnopolsky, M. A. et al. Influence of endurance exercise training and sex on
(2015). intramyocellular lipid and mitochondrial ultrastructure, substrate use, and mitochondrial
303. Janssen, I., Heymsfield, S. B., Wang, Z. M. & Ross, R. Skeletal muscle mass and enzyme activity. Am. J. Physiol. Regul. Integr. Comp. Physiol. 292, R1271–R1278 (2007).
distribution in 468 men and women aged 18-88 yr. J. Appl. Physiol. 89, 81–88 (2000). 335. Goodpaster, B. H., He, J., Watkins, S. & Kelley, D. E. Skeletal muscle lipid content
304. Tian, Q. et al. Muscle mitochondrial energetics predicts mobility decline in well- and insulin resistance: evidence for a paradox in endurance-trained athletes. J. Clin.
functioning older adults: the baltimore longitudinal study of aging. Aging Cell 21, e13552 Endocrinol. Metab. 86, 5755–5761 (2001).
(2022). 336. Steensberg, A. et al. Production of interleukin-6 in contracting human skeletal muscles
305. Alcazar, J., Rodriguez-Lopez, C., Delecluse, C., Thomis, M. & Van Roie, E. Ten-year can account for the exercise-induced increase in plasma interleukin-6. J. Physiol. 529,
longitudinal changes in muscle power, force, and velocity in young, middle-aged, and 237–242 (2000).
older adults. J. Cachexia Sarcopenia Muscle 14, 1019–1032 (2023). 337. Trinh, B. et al. Blocking endogenous IL-6 impairs mobilization of free fatty acids during
306. Grosicki, G. J., Zepeda, C. S. & Sundberg, C. W. Single muscle fibre contractile function rest and exercise in lean and obese men. Cell Rep. Med. 2, 100396 (2021).
with ageing. J. Physiol. 600, 5005–5026 (2022). 338. Kistner, T. M., Pedersen, B. K. & Lieberman, D. E. Interleukin 6 as an energy allocator in
307. Zhang, X. et al. Characterization of cellular senescence in aging skeletal muscle. Nat. muscle tissue. Nat. Metab. 4, 170–179 (2022).
Aging 2, 601–615 (2022). 339. Katashima, C. K. et al. Evidence for a neuromuscular circuit involving hypothalamic
308. Moiseeva, V. et al. Senescence atlas reveals an aged-like inflamed niche that blunts interleukin-6 in the control of skeletal muscle metabolism. Sci. Adv. 8, eabm7355
muscle regeneration. Nature 613, 169–178 (2023). (2022).

Nature Reviews Molecular Cell Biology

Review article

340. Takahashi, H. et al. TGF-β2 is an exercise-induced adipokine that regulates glucose and Author contributions
fatty acid metabolism. Nat. Metab. 1, 291–303 (2019). The authors contributed equally to all aspects of the article.
341. Li, V. L. et al. An exercise-inducible metabolite that suppresses feeding and obesity.
Nature 606, 785–790 (2022). Competing interests
342. Hu, M., Nie, J., Lei, O. K., Shi, Q. & Kong, Z. Acute effect of high-intensity interval training The authors declare no competing interests.
versus moderate-intensity continuous training on appetite perception: a systematic
review and meta-analysis. Appetite 182, 106427 (2023). Additional information
343. Reddy, A. et al. pH-gated succinate secretion regulates muscle remodeling in response Supplementary information The online version contains supplementary material available at
to exercise. Cell 183, 62–75.e17 (2020). https://doi.org/10.1038/s41580-023-00606-x.
344. Agudelo, L. Z. et al. Skeletal muscle PGC-1α1 modulates kynurenine metabolism and
mediates resilience to stress-induced depression. Cell 159, 33–45 (2014). Peer review information Nature Reviews Molecular Cell Biology thanks Kamala Sreekumaran
345. Agudelo, L. Z. et al. Kynurenic acid and Gpr35 regulate adipose tissue energy Nair, who co-reviewed with Mark Pataky, Adam P Sharples and Brian Glancy for their
homeostasis and inflammation. Cell Metab. 27, 378–392.e5 (2018). contribution to the peer review of this work.
346. Correia, J. C. et al. Muscle-secreted neurturin couples myofiber oxidative metabolism
and slow motor neuron identity. Cell Metab. 33, 2215–2230.e8 (2021). Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
Acknowledgements
The authors apologize to all colleagues whose work could not be included owing to space Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this
constraints. The authors thank M. Karlén for preparation of the artwork in the supplementary article under a publishing agreement with the author(s) or other rightsholder(s); author self-
figures. K.A.M. was supported by the National Institutes of Health (NIH R00 AG063994). K.A.D. archiving of the accepted manuscript version of this article is solely governed by the terms
was supported by Deutsches Zentrum für Diabetesforschung (2020/21). J.R.Z. was supported of such publishing agreement and applicable law.
by the Swedish Research Council (Vetenskapsrådet) (2015-00165), the Swedish Research
Council for Sport Science (P2022-0013, P2023-0093) and the Novo Nordisk Foundation © Springer Nature Limited 2023
Center for Basic Metabolic Research at the University of Copenhagen (NNF18CC0034900).

Nature Reviews Molecular Cell Biology