Exercise & Sport Sciences Reviews:
Circadian Rhythms, Skeletal Muscle Molecular Clocks, and Exercise
Schroder, Elizabeth A.1,2; Esser, Karyn A.1,2
1Center for Muscle Biology and 2Department of Physiology, College of Medicine, University of Kentucky, Lexington, KY
Address for correspondence: Karyn A. Esser, University of Kentucky, Department of Physiology, 800 Rose St., MS508, Lexington, KY 40536 (E-mail: firstname.lastname@example.org).
Associate Editor: LaDora V. Thompson, Ph.D.
Accepted for publication: July 9, 2013.
Skeletal muscle comprises approximately 40% of total body mass and, as such, contributes to maintenance of human health. In this review, we discuss the current state of knowledge regarding the role of molecular clocks in skeletal muscle. In addition, we discuss a new function for exercise as a time-setting cue for muscle and other peripheral tissues.
Circadian rhythms are the approximate 24-h biological cycles that function to prepare the organism for daily environmental changes. Almost all organisms ranging from single-cell bacteria to plants and animals exhibit behavioral, physiological, and biochemical rhythms termed “circadian rhythms” (7,9). The underlying circadian rhythm is a molecular clock mechanism found in most, if not all, cell types, including skeletal muscle. At the cellular level, the presence of a molecular clock is argued to be a necessary timekeeping mechanism to prepare the cell for daily changes in environmental conditions. The ability to synchronize the molecular clock and intracellular physiology with external day-night cycles denotes an evolutionarily conserved adaptation to environmental conditions (7,9).
Environmental stimuli capable of shifting the timing of molecular rhythms are critically important to understanding the synchronization of oscillators to the environment and to each other within a multicellular system. The most well-understood circadian time cue (termed “zeitgeber”) is the photic or light time cue. Whereas lighting conditions have been studied extensively in many species more recently, nonphotic zeitgebers such as time of feeding and time of physical activity/exercise have been shown to influence molecular rhythms and behavior (25,31).
The goal of this review is to present the latest findings demonstrating that the molecular clock exists in skeletal muscle and support for the concept that exercise plays a role as a time cue for clocks in peripheral tissues. The existing data, although limited, suggest the importance of exercise and, more importantly, the timing of exercise in adding stability to the daily rhythms of the circadian system. These observations add a new function for the role of exercise in human health and suggest new concepts for the benefits of exercise in potentially postponing or preventing the development of chronic diseases (25,31).
THE MOLECULAR CLOCK IN MAMMALS
The molecular mechanism underlying circadian rhythms is a gene-regulatory network composed of transcriptional-translational feedback loops referred to as the “core clock” (11). The molecular clock components comprising the positive arm of the core clock are two members of the PAS-bHLH family of transcription factors, Clock (Circadian locomotor output control kaput) and Bmal1 (Brain muscle arnt-like1). The BMAL1 protein is expressed in a circadian pattern in both the suprachiasmatic nucleus (SCN) and peripheral tissues (4,8). CLOCK protein levels do not oscillate in the SCN and only in some peripheral tissues (15). However, the nuclear to cytosolic distribution of CLOCK is circadian in its pattern, with highest levels of CLOCK in the nucleus occurring in the light phase in mice (18,28). The CLOCK:BMAL1 heterodimer activates transcription of additional core clock genes Period (Per1, Per2, and Per3) and Cryptochrome (Cry1 and Cry2) by binding to E-box (CACGTG) sequences in the regulatory region of these genes. The CRY and PER proteins constitute the negative arm of the core molecular clock by forming multimers that inhibit CLOCK:BMAL1 transcriptional activity on translocation to the nucleus. A schematic diagram of the molecular clock mechanism is shown in Figure 1.
Additional components to the core molecular clock family include the orphan nuclear receptors Rora (RAR-related orphan receptor-α) and Rev-erb α/β. These gene products function to link the feedback loops by activating (Rora) or repressing (Rev-erb) Bmal1 transcription (21,24). Most recently, studies have added new elements, known as E3 ligases (e.g., Fbxl3), to the core molecular clock, and these elements function by changing the stability of the PER and CRY proteins (33). In addition to their role in timekeeping, components of the core clock (Bmal1 and Clock) also have been shown to regulate the expression of genes that do not function in timekeeping transcriptionally, and these genes are designated as clock-controlled genes (CCGs). Although the identity of all the direct clock-controlled genes in a specific tissue, like skeletal muscle, has not been defined, they often encode transcription factors (e.g., MyoD1) or proteins that control rate-limiting steps in cell physiology (e.g., PBEF, the rate-limiting enzyme in the NAD+ salvage pathway) (17,18). For more detailed reviews of the molecular clock mechanism, there are several recent reviews by other groups (16).
CENTRAL/PERIPHERAL CLOCKS IN MAMMALS AND ZEITGEBERS
The first suggestion that the central clock (circadian pacemaker) was located in the SCN came when it was discovered that surgical ablation of the SCN resulted in arrhythmic behavior patterns (27). SCN ablation in hamsters creates an actively arrhythmic animal, and transplantation of a healthy SCN (from either a hamster or a mouse) back into the lesioned hamsters restores activity rhythms, with rhythms matching that of the donor animal (26), demonstrating the role of the SCN as a system-wide circadian synchronizer.
Cell autonomy of the molecular clock in peripheral tissue was established first almost 15 yr ago (3). The development of a mouse model in which the luciferase cDNA was knocked into the Per2 coding region to generate a chimeric protein provided the powerful resource (PER2:LUC mice) to test the cell autonomy of the clock mechanism directly (32). For these studies, tissue explants from the SCN, liver, and lung from these PER2:LUC mice were placed in cell culture in the presence of luciferin, and real-time light emission was monitored for up to 2 wk. Although oscillations in luminescence were expected from the central clock (SCN slice), these findings demonstrated that the clock mechanism from peripheral tissue explants maintained normal 24-h circadian periodicity in the absence of systemic neural or humoral factors (32). These studies also validated the use of the PER2:LUC mouse as a valuable resource to study molecular clock function in multiple tissues, including skeletal muscle (31,35).
Although studies have established that the molecular clock mechanism is intrinsic to each cell and that it can run in a cell autonomous manner, a critical feature is that the phase of the molecular clock can be set or reset by cues from the environment. The ability to reset the circadian clock is a critical function to be able to adapt to environmental changes. The timing or phase of the central clock is entrained primarily by cues from light (29). Recent studies have demonstrated that the molecular clock in many peripheral tissues, including skeletal muscle, can be dissociated from the rhythm of the SCN by restricting time of feeding (4). In addition, Zambon et al. (34) reported that there is an interaction between time of day and contraction on expression of clock genes in human muscle, suggesting that contractile activity might be a zeitgeber for the molecular clock mechanism in skeletal muscle. In this study, microarray analysis was used to determine the effects of resistance exercise on gene expression in the quadriceps at 6 and 18 h after an acute bout of exercise. Gene expression also was examined in the nonexercised leg at the same time points for comparison. These results indicate that feeding and, potentially, contractile activity can act as dominant zeitgebers for setting clocks in peripheral tissues (Fig. 2). They also demonstrate that circadian rhythms in peripheral tissues can be dissociated from the SCN through nonphotic environmental signals.
The timing of the molecular clock also can be modulated by posttranslational mechanisms using phosphorylation, acetylation, and/or ubiquitination pathways. These modifications impact the stability and/or translocation of core molecular clock components, with the greatest impact on period length (time for completion of one cycle (24 h)). The most well-known regulator of clock function through changing phosphorylation status is casein kinase 1ε (CK1ε). This serine/threonine kinase has been the subject of studies in both hamster and humans in which the mutations of either CK1ε or PER2 affect period length. The tau mutant hamster has a point mutation (Arg178Cys) in the catalytic region of CK1ε, termed the tau mutation, which results in a hamster that exhibits short circadian periods (23). In these hamsters, CK1ε is unable to phosphorylate PER proteins. With diminished levels of phosphorylation, PER translocates to the nucleus, leading to a more rapid repression of the CLOCK:BMAL1-mediated transcription, effectively shortening the circadian period to 20 h (10). In humans, a mutation in PER2 that affects a phosphorylation site (target of CK1ε) results in the clinical condition familial advanced sleep-phase syndrome (FASPS). Studies of these individuals determined that they have a shortened period length of approximately 20 h compared with the normal 24 h observed in most humans and is associated with early awake and early to sleep (30).
Acetylation of histones unfolds chromatin to expose promoter and regulatory regions of genes and is associated with activation of gene expression, whereas acetylation results in silencing of gene expression. An example for a circadian role in acetylation comes from studies of the histone acetyltransferase p300. The p300 protein associates with CLOCK in a circadian manner, implicating p300 as a component of the core clock transactivation complex. In addition, H3 histone acetylation at the promoter region of Per1 and Per2 shows robust circadian rhythms in phase with the Per1/Per2 mRNA and p300:CLOCK complex formation. p300 is inhibited by the CRY protein, leading to a decrease in CLOCK:BMAL1-mediated transcription at the Per1 promoter. These data suggest that the repressor action of CRY is mediated in part by its actions on chromatin structure (6). The most frequently studied deacetylase involved in circadian regulation is SirtT1. SirT1 deacetylates histone H3. This activity shows a strong circadian rhythm antiphase to the rhythm of histone H3 acetylation (17). This antiphase oscillation of acetylation/deacetylation suggests an important direct or permissive mechanism for controlling clock gene expression and proposes roles for acetylation/deacetylation in the initiation, duration, and termination of both the activating and repressing phases of the circadian cycle.
THE MOLECULAR CLOCK IN SKELETAL MUSCLE
The skeletal muscle circadian transcriptome was first identified by Miller and colleagues (14). This work was followed by a publication that identified approximately 215 mRNA that were expressed in a circadian pattern in the gastrocnemius muscle of wild-type C57BL/6 mice. This list included known core molecular clock components Bmal1, Per2, and Cry1 (13). Gene ontology analysis identified enrichment in multiple cellular processes, including but not limited to metabolic, transcriptional, and degradative processes. The tissue specificity of the circadian transcriptome in skeletal muscle was underscored by the inclusion of known muscle-specific genes such as Myod1, Ucp3, Fbxo32/atrogin, and Myh1(MyHC IIX). The identification of Myod1 as a gene expressed in a circadian manner was exciting as it is well characterized as a transcription factor involved in the skeletal muscle lineage. The fact that it is expressed in a circadian manner in adult tissue and is under direct control of CLOCK:BMAL1 (1) suggests that it may be critical for daily muscle maintenance (Fig. 1). To further understand the implications of molecular clock disruption in skeletal muscle, McCarthy et al. (13) compared gene expression changes between skeletal muscle from wild-type C57BL/6 mice with those from Clock-mutant mice. Clock-mutant mice have a mutation that results in the deletion of exon 19 of CLOCK, and this is associated with longer behavioral rhythms of 27 h (2). We found that several core clock genes, including Bmal1 and Per2, no longer oscillated in the muscle of Clock-mutant mice. Oscillation also was lost in several known clock-controlled genes in the skeletal muscle, such as Tef, Dbp, Myod1, and Pgc1β. In addition to analysis of the circadian genes, McCarthy et al. (13) found that approximately 35% of all the expressed genes were expressed differentially in the muscle of Clock-mutant mice, demonstrating the substantial impact disruption of the core clock machinery has on gene expression, with implications for normal cellular function and health. With continued development of analysis tools for time-course gene expression studies, the list of genes expressed in a circadian manner in skeletal muscle has grown to more than 800. Pizarro et al. (20) have established a valuable Web site to allow for queries into circadian expression from time-course expression experiments in tissues and cell lines (http://circadb.org). These new data will open many exciting avenues of research in the study of skeletal muscle circadian physiology. However, one area that is still understudied is whether the circadian transcriptome differs among individual skeletal muscles of different developmental origin (e.g., limb vs facial muscle) or whether they differ among muscles with markedly different fiber type, metabolic function, and mechanical function.
EXERCISE AND CIRCADIAN RHYTHMS
Wheel running or cage activity has long been used as a readout of circadian behavior, but the idea that physical activity/exercise and, more importantly, the timing of exercise could serve as an entrainment cue is relatively new (25,31). Studies in the late 1980s and early 1990s were the first to show that novel wheel access at different times of day was sufficient to shift the phase of circadian activity rhythms in mice and hamsters (5). Further studies demonstrated that exercise is a sufficient environmental cue to effect clock gene expression in the SCN (central clock) located in the hypothalamus of the brain (12). These studies established that activity, in the form of access to a novel running wheel, during light conditions decreased peak expression of the clock genes Per1 and Per2 in the SCN. Schroeder et al. (25) took this idea farther, examining both timing of exercise as well as multiple bouts of exercise. They explored effects on the central clock using a scheduled exercise paradigm in control and mutant mice in which the central molecular clock mechanism was weakened (25). In attempts to better mimic the timing of exercise in the human population, mice were allowed free access to a wheel, no access to a wheel, or access limited to 6-h time frames at the beginning or end of the dark or active phase for a minimum of 16 d. Similar to previous studies examining single bouts of exercise, they observed changes in the properties of the molecular clock in the SCN after 16 d in the control mice, suggesting that the phase shifts observed in previous studies were not solely an acute-phase response. Moreover, PER2:LUC amplitude was damped in mice with wheel access scheduled early in the dark phase but unaffected with scheduled activity late in the dark phase or with free access to wheel running, suggesting that the timing of exercise may be critical for the maintenance of molecular rhythms in the SCN. Using the vasoactive intestinal polypeptide knockout mouse, shown to have an unstable clock mechanism, they found that scheduled exercise functioned to enhance the stability of both activity and heart rate rhythms.
The core molecular clock gene Clock also has been demonstrated to be critical for healthy skeletal muscle because Clock-mutant mice exhibit approximately 30% reduction in normalized maximal force at both the muscle and single-fiber level (1). In addition, myofiber architecture is disrupted, and mitochondrial volume is diminished. Recent work from Pastore et al. (19) supports these results. They demonstrated that CLOCK protein is critical for mitochondrial maintenance in skeletal muscle. Taking this a step farther, they examined the ability of Clock-mutant mice to adapt to chronic exercise and found that, despite the pathology as a result of the mutant Clock gene, the ability of these mice to adapt to chronic exercise was not changed. Using endurance training, they were able to rescue the metabolic defects resulting from loss of functional CLOCK protein partially.
Most studies of exercise and shifting of circadian rhythms have relied on endurance exercise paradigms. Less is known about the potential for resistance exercise, but Zambon et al. (34) reported that one bout of 60 contractions was associated with changes in molecular clock gene expression in skeletal muscle of humans. Experiments in neonatal cardiomyocytes have shown that contractile activity may modulate the molecular clock through the actions of the CLOCK protein. Histological and biochemical analysis demonstrated that CLOCK localizes to the z-disk in neonatal cardiomyocytes and translocates to the nucleus to influence gene expression in response to contractile activity. CLOCK localization at the z-disk puts CLOCK in an appropriate location to sense mechanical function associated with contractile activity (22). Although these studies are suggestive that resistance exercise also can modulate molecular clock function in muscle, there is still much to be determined.
With more than 600 different muscles in the human body, comprising approximately 40% of total body mass, understanding the effects of exercise on the molecular rhythms in individual skeletal muscles may provide critical insight into systemic mechanisms contributing to daily rhythms. At this stage, there is only one study that has examined more than one muscle in mice exposed to scheduled bouts of either voluntary or involuntary endurance exercise for 2 h d-1 in the light phase 4 h after lights on. In this study, the authors found a significant shift in clock gene expression (PER2:LUC bioluminescence) in three different skeletal muscles and the lung from exercised mice, whereas the molecular clock in the SCN remained unshifted, demonstrating that scheduled exercise can alter the molecular clock in peripheral tissues. In addition, one of the muscles examined, the flexor digitorum brevis was phase advanced more than the other two muscles (soleus and extensor digitorum longus), suggesting the potential for differential regulation of the molecular clock in individual muscles (31). Lumicycle data, in combination with data demonstrating rescue of phenotype resulting from exercise (19,25,31), implicate exercise as a nonphotic time cue in peripheral tissues and suggest that the molecular clock in all muscle tissues may not respond in a similar manner to nonphotic cues. Because muscle is such a large contributor to systems physiology, these data have broad implications for human health and disease, suggesting the power of exercise and, more specifically, the interaction of exercise and muscle as a therapeutic strategy to help stabilize/realign molecular clocks throughout the body.
Circadian rhythms and molecular clocks in skeletal muscle are a new and rapidly emerging area of research. Studies in the last 15 yr have demonstrated that molecular clocks exist in skeletal muscle and, more importantly, studies have demonstrated that the phase of the clocks in skeletal muscle can be reset by altering time of exercise or time of feeding independent of the central clock in the brain. In parallel studies, it has been shown that skeletal muscles from mice in which the core clock genes have been disrupted are weaker and exhibit decreased mitochondrial content and function. These findings suggest a potential new role for exercise in human health through its role in providing timing information to skeletal muscle and other peripheral tissue clocks. Although difficult to extrapolate to humans, exercise studies in mice do suggest an optimal time of day for exercise to enhance the robust nature of circadian rhythms at the molecular level. Until an optimal time is determined, consistently timed daily exercise may be a viable alternative. These studies highlight the recognition that time of day matters for maintenance of proper molecular clock function and is important for issues of muscle strength and endurance.
The authors thank Jonathan England for his help with the figures in the preparation of this article. This work was supported by the following National Institutes of Health grants RC1ES018636 and R01AR55246 (to K.A.E.).
1. Andrews JL, Zhang X, McCarthy JJ, et al. CLOCK and BMAL1 regulate MyoD and are necessary for maintenance of skeletal muscle phenotype and function. Proc. Natl. Acad. Sci. 2010; 107 (44): 19090–5.
2. Antoch MP, Song EJ, Chang AM, et al. Functional identification of the mouse circadian clock gene by transgenic BAC rescue. Cell. 1997; 89 (4): 655–67.
3. Balsalobre A, Damiola F, Schibler U. A serum shock induces circadian gene expression in mammalian tissue culture cells. Cell. 1998; 93 (6): 929–37.
4. Damiola F, Le Minh N, Preitner N, Kornmann B, Fleury-Olela F, Schibler U. Restricted feeding uncouples circadian oscillators in peripheral tissues from the central pacemaker in the suprachiasmatic nucleus. Genes Dev. 2000; 14: 2950–61.
5. Edgar D, Dement W. Regularly scheduled voluntary exercise synchronizes the mouse circadian clock. Am. J. Physiol. 1991; 261: R928–33.
6. Etchegaray JP, Lee C, Wade PA, Reppert SM. Rhythmic histone acetylation underlies transcription in the mammalian circadian clock. Nature. 2003; 421 (6919): 177–82.
7. Idda ML, Bertolucci C, Vallone D, Gothilf Y, Sánchez-Vázquez FJ, Foulkes NS, Andries Kalsbeek, Martha Merrow, Till Roenneberg, Russell G. Foster. Circadian clocks: lessons from fish. In: editors. Progress in Brain Research, Elsevier; 2012. pp. 41–57.
8. Lee Y, Chen R, Lee H, Lee C. Stoichiometric relationship among clock proteins determines robustness of circadian rhythms. J. Biol. Chem. 2011; 286 (9): 7033–42.
9. Loudon Andrew SI. Circadian biology: a 2.5–billion year–old clock. Curr. Biol. 2012; 22 (14): R570–R1.
10. Lowrey P, Shimomura K, Antoch M, et al. Positional syntenic cloning and functional characterization of the mammalian circadian mutation tau. Science. 2000; 288 (5465): 483–92.
11. Lowrey PL, Takahashi JS. Mammalian circadian biology: elucidating genome-wide levels of temporal organization. Annu. Rev. Genomics Human Genet. 2004; 5 (1): 407–41.
12. Maywood E, Mrosovsky N, Field M, Hastings M. Rapid downregulation of mammalian period genes during behavioral resetting of the circadian clock. Proc. Natl. Acad. Sci. USA. 1999; 96: 15211–6.
13. McCarthy JJ, Andrews JL, McDearmon EL, et al. Identification of the circadian transcriptome in adult mouse skeletal muscle. Physiol. Genomics. 2007; 31 (1): 86–95.
14. Miller BH, McDearmon EL, Panda S, et al. Circadian and CLOCK-controlled regulation of the mouse transcriptome and cell proliferation. Proc. Natl. Acad. Sci. 2007; 104 (9): 3342–7.
15. Miyazaki M, Schroder E, Edelmann SE, et al. Age-associated disruption of molecular clock expression in skeletal muscle of the spontaneously hypertensive rat. PLoS ONE. 2011; 6 (11): e27168.
16. Mohawk JA, Green CB, Takahashi JS. Central and peripheral circadian clocks in mammals. Annu. Rev. Neurosci. 2012; 35 (1): 445–62.
17. Nakahata Y, Kaluzova M, Grimaldi B, et al. The NAD+-dependent deacetylase SIRT1 modulates CLOCK-mediated chromatin remodeling and circadian control. Cell. 2008; 134 (2): 329–40.
18. Panda S, Antoch MP, Miller BH, et al. Coordinated transcription of key pathways in the mouse by the circadian clock. Cell. 2002; 109 (3): 307–20.
19. Pastore S, Hood DA. Endurance training ameliorates the metabolic and performance characteristics of circadian Clock
-mutant mice. J. Appl. Physiol. 2013; 114 (8): 1076–84.
20. Pizarro A, Hayer K, Lahens NF, Hogenesch JB. CircaDB: a database of mammalian circadian gene expression profiles. Nucleic Acids Res. 2013; 41 (D1): D1009–13.
21. Preitner N, Damiola F, Lopez-Molina L, et al. The orphan nuclear receptor REV-ERBα controls circadian transcription within the positive limb of the mammalian circadian oscillator. Cell. 2002; 110: 251–60.
22. Qi L, Boateng SY. The circadian protein CLOCK localizes to the sarcomeric z-disk and is a sensor of myofilament cross-bridge activity in cardiac myocytes. Biochem. Biophys. Res. Commun. 2006; 351 (4): 1054–9.
23. Ralph MR, Menaker M. A mutation of the circadian system in golden hamsters. Science. 1988; 241 (4870): 1225–7.
24. Sato TK, Panda S, Miraglia LJ, et al. A functional genomics strategy reveals Rora
as a component of the mammalian circadian clock. Neuron. 2004; 43 (4): 527–37.
25. Schroeder AM, Truong D, Loh DH, Jordan MC, Roos KP, Colwell CS. Voluntary scheduled exercise alters diurnal rhythms of behaviour, physiology and gene expression in wild-type and vasoactive intestinal peptide-deficient mice. J. Physiol. 2012; 590 (23): 6213–26.
26. Sollars P, Kimble D, Pickard G. Restoration of circadian behavior by anterior hypothalamic heterografts. J. Neurosci. 1995; 15 (3): 2109–22.
27. Stephan F, Zucker I. Circadian rhythms in drinking behavior and locomotor activity of rats are eliminated by hypothalamic lesions. Proc. Natl. Acad. Sci. USA. 1972; 69 (6): 1583–6.
28. Storch KF, Paz C, Signorovitch J, et al. Intrinsic circadian clock of the mammalian retina: importance for retinal processing of visual information. Cell. 2007; 130 (4): 730–41.
29. Takahashi J, DeCoursey P, Bauman L, Menaker M. Spectral sensitivity of a novel photoreceptive system mediating entrainment of mammalian circadian rhythms. Nature. 1984; 308 (5955): 186–8.
30. Vanselow K, Vanselow JT, Westermark PO, et al. Differential effects of PER2 phosphorylation: molecular basis for the human familial advanced sleep phase syndrome (FASPS). Genes Develop. 2006; 20 (19): 2660–72.
31. Wolff G, Esser KA. Scheduled exercise phase shifts the circadian clock in skeletal muscle. Med. Sci. Sports Exerc. 2012; 44 (9): 1663–70.
32. Yoo SH, Ko CH, Lowrey PL, et al. A noncanonical E-box enhancer drives mouse Period2
circadian oscillations in vivo
. Proc. Natl. Acad. Sci. USA. 2005; 102 (7): 2608–13.
33. Yoo S-H, Mohawk Jennifer A, Siepka Sandra M, et al. Competing E3 ubiquitin ligases govern circadian periodicity by degradation of CRY in nucleus and cytoplasm. Cell. 2013; 152 (5): 1091–105.
34. Zambon A, McDearmon E, Salomonis N, et al. Time- and exercise-dependent gene regulation in human skeletal muscle. Genome Biol. 2003; 4 (10): R61.
35. Zhang X, Patel SP, McCarthy JJ, Rabchevsky AG, Goldhamer DJ, Esser KA. A non-canonical E-box within the MyoD core enhancer is necessary for circadian expression in skeletal muscle. Nucleic Acids Res. 2012; 40 (8): 3419–30.
circadian; skeletal muscle; exercise; peripheral tissues; zeitgeber
© 2013 American College of Sports Medicine
Highlight selected keywords in the article text.