Response of the ubiquitin-proteasome pathway to changes in muscle activity

Michael B. Reid

Abstract

The ubiquitin-proteasome pathway plays a critical role in the adaptation of skeletal muscle to persistent decreases or increases in muscle activity. This article outlines the basics of pathway function and reviews what we know about pathway responses to altered muscle use. The ubiquitin-proteasome pathway regulates proteolysis in mammalian cells by attaching ubiquitin polymers to damaged proteins; this targets the protein for degradation via the 26S proteasome. The pathway is constitutively active in muscle and continually regulates protein turnover. Conditions of decreased muscle use, e.g., unloading, denervation, or immobilization, stimulate general pathway activity. This activity increase is caused by upregulation of regulatory components in the pathway and leads to accelerated proteolysis, resulting in net loss of muscle protein. Pathway activity is also increased in response to exercise, a two-phase response. An immediate increase in selective ubiquitin conjugation by constitutive pathway components contributes to exercise-stimulated signal transduction. Over hours-to-days, exercise also stimulates a delayed increase in general ubiquitin conjugating activity by inducing expression of key components in the pathway. This increase mediates a late-phase rise in protein degradation that is required for muscle adaptation to exercise. Thus the ubiquitin-proteasome pathway functions as an essential mediator of muscle remodeling, both in atrophic states and exercise training.

  • disuse atrophy
  • muscle adaptation
  • exercise training
  • cachexia
  • muscle growth

skeletal muscle fibers are subjected to mechanical and metabolic stresses during the course of physical activity. This contributes to damage of cellular proteins, requiring that proteins be continually degraded and resynthesized to maintain normal function. Protein turnover is further stimulated by changes in the pattern of muscle use. Sustained decreases or increases in physical activity cause muscle fibers to adapt, altering protein expression and fiber size. The outcome of these adaptive responses, atrophy vs. hypertrophy, reflects the net balance between protein degradation and protein synthesis. Atrophy requires that protein degradation exceed resynthesis. In muscle hypertrophy, the opposite is true.

The ubiquitin-proteasome pathway is a primary regulator of these dynamic processes, providing a mechanism for selective degradation of regulatory and structural proteins. This pathway is constitutively active in muscle fibers and mediates both intracellular signaling events and normal protein turnover. A substantial body of research indicates that changes in muscle use can stimulate pathway activity, altering activity-related signals and contributing to the remodeling of muscle fibers. The literature describing responses of the ubiquitin-proteasome pathway to altered muscle use has not been reviewed heretofore and is the topic of the present article. The review is presented in three major sections. The first provides a concise overview of the ubiquitin-proteasome pathway for readers who may be unfamiliar with this biology. The second section reviews muscle-specific mechanisms by which pathway activity is known to be regulated, both in intracellular signaling and in muscle adaptation. The third section summarizes our current understanding of pathway responses to decreased or increased muscle use, e.g., during unloading, denervation, or exercise. The goal of the latter section is to consolidate the available data on this emerging topic, integrating observations by a number of investigators to identify general patterns of pathway regulation.

OVERVIEW OF THE UBIQUITIN-PROTEASOME PATHWAY

This first section briefly outlines principles that are generally accepted about function of the ubiquitin-proteasome pathway, including components of the pathway and mechanisms of pathway regulation. This material has been condensed and simplified. No attempt has been made to catalog the primary references that established these concepts. Readers who seek more detailed information may wish to consult the many authoritative reviews on this topic (e.g., 8, 15, 17, 18, 43, 74).

Basic function of the pathway.

The ubiquitin-proteasome pathway is the major pathway for selective protein degradation in eukaryotic cells and is thought to degrade the bulk of all intracellular proteins during muscle remodeling. In brief, the pathway recognizes misfolded or damaged proteins and labels the target proteins by conjugation with the polypeptide ubiquitin. Ubiquitin-conjugated proteins are subsequently recognized and degraded by the 26S proteasome, a multicatalytic enzyme complex. Pathway components are localized to the cytosol but appear to degrade soluble, nuclear, and membrane-bound proteins.

Pathway activity depends upon coordinated interactions among several enzyme families, a process illustrated in Fig. 1. First, ubiquitin monomers are activated by the ubiquitin-activating enzyme (also known as “E1 protein” or “E1”). This is an ATP-dependent step in which ubiquitin is linked to E1 via a high-energy thiolester bond. Only one E1 protein has been identified in mammalian cells, an abundant 110-kDa protein essential for cell survival. Second, activated ubiquitin is transferred to a ubiquitin carrier (E2) protein. Dozens of E2 proteins have been identified, suggesting divergent roles of individual E2s in targeting specific types of protein substrates. Third, the E2 interacts with one of several hundred ubiquitin ligases (E3s) to transfer activated ubiquitin to the substrate, where it is attached via a peptide or isopeptide bond. This begins an iterative process by which E2/E3 partners attach multiple ubiquitin monomers to form polyubiquitin chains, marking the protein substrate for degradation. Finally, the marked protein is degraded by the 26S-proteasome complex. This complex is a major component of the cell, comprising up to 1% of total cellular protein. It consists of two 19S “caps” that identify and unfold targeted proteins. Unfolded proteins are fed into the 20S core, composed of four subunits, which degrades the substrate to small peptides (3–25 amino acid residues) via an ATP-dependent process.

Fig. 1.

The ubiqutin-proteasome pathway. Diagram illustrates protein families that compose the pathway and their interactions; shading identifies proteins that exhibit transcriptional regulation in response to exercise or decreased muscle use. Ub, ubiquitin; E1-UBA, E1-protein or ubiquitin-activating enzyme; E2-UBC, E2-protein or ubiquitin conjugating enzyme; E3-UBL, E3 protein or ubiquitin ligase.

Principles of pathway regulation.

The E1 protein and the 26S proteasome are abundant in most mammalian cells and are constitutively active. These components do not generally function as primary regulators of pathway function. Instead, the interaction between E2/E3 pairs and their corresponding substrates are thought to be the primary regulatory step. Identity of the E2/E3 partners confers substrate specificity, targeting specific proteins for ubiquitin conjugation. E2/E3 activity appears to be rate limiting in many conditions such that upregulation of individual E2 or E3 proteins can increase overall pathway activity.

Conjugation of ubiquitin to specific proteins is regulated moment-to-moment by at least five known mechanisms: 1) Phosphorylation of substrate proteins is a well-recognized example that can either favor or inhibit ubiquitin conjugation, depending on the substrate and E2/E3 proteins involved. 2) Some E2/E3 complexes are not constitutively active and require association with companion proteins to effectively transfer ubiquitin. 3) Viral proteins are known to mediate specific substrate-E2/E3 interactions, thereby promoting ubiquitination. 4) Degradation signals on substrate proteins may be masked by binding to DNA or other proteins, protecting the bound protein from ubiquitin conjugation. 5) Finally, muscle fibers and other mammalian cell types contain deubiquinating enzymes that cleave ubiquitin molecules from conjugated substrates, reversing the degradation signal and protecting the protein from proteasomal targeting.

Over longer time periods, general activity of the ubiquitin-proteasome pathway can be regulated via changes in the amount and composition of pathway components. One example is regulation of histocompatibility complex (MHC) class I antigen. As reviewed by Rock and associates (50), inflammatory mediators stimulate expression of inducible subunits for the 26S proteasome. The inducible subunits substitute for constitutive subunits, thereby altering peptide cleavage products generated by the proteasome and increasing the presentation of MHC class I antigens for T-cell recognition. A more pertinent example is the response of skeletal muscle to catabolic states, including starvation, cancer, renal failure, sepsis, or cytokine exposure. As reviewed elsewhere (13, 25, 29, 41), upregulation of inducible, muscle-specific components causes general activity of the ubiquitin-proteasome pathway to increase, stimulating bulk degradation of muscle proteins (see below).

PATHWAY REGULATION IN MUSCLE

Targeting of specific proteins.

In skeletal muscle, several examples of selective ubiquitin conjugation to a specific protein have been described. Abu Hatoum et al. (1) originally showed that activity of MyoD, a myogenic differentiation factor, is regulated by ubiquitin conjugation that targets MyoD for proteasomal degradation. DNA binding obscures the ubiquitination sequence and stabilizes MyoD, whereas MyoD binding to a second myogenic regulator, Id1, inhibits MyoD/DNA interaction and promotes MyoD degradation. MyoD ubiquitination appears to occur in the G1 phase and to be triggered by phosphorylation at either of two alternative sites, serine 200 as mediated by cyclin E-Cdk2 (68) or lysine 1,333 (4) as mediated by atrogin1/MAFbx (67). Similarly, selective ubiquitin conjugation regulates activity of the transcription factor NF-κB in muscle; phosphorylation of serine residues on an inhibitor protein, either I-κBα (38) or p105 (19), stimulates ubiquitin conjugation to the inhibitor via the E3 protein complex SCFβTrCP (82). Proteasomal degradation of the ubiquitin-tagged inhibitor leads to activation and nuclear translocation of NF-κB, a feed-forward signal that upregulates expression of genes for ubiquitin, E2/E3 proteins, and proteasome subunits (7, 37, 56, 72, 73, 81), favoring an increase in general pathway activity. Finally, Solomon and associates (57) demonstrated that muscle proteins with an acidic or large hydrophobic residue at the N-terminus undergo rapid ubiquitin conjugation via the “N-end rule” pathway. This process is regulated by a specific E2/E3 pair, E2–14k and E3α, with consequences for general pathway activity (see below).

Upregulation of general pathway activity.

General activity of the ubiquitin-proteasome pathway can be increased in skeletal muscle by upregulation of pathway components. This is best understood in the context of overt catabolism where ubiquitin conjugating activity increases markedly. Among many excellent reviews published recently, those by Glass (14), Jackman and Kandarian (24), Jagoe and Goldberg (25), and Lecker (32) provide especially good overviews of this complex topic for readers who seek greater detail. For our purposes, the current section provides a succinct overview of specific elements in the ubiqutin-proteasome pathway that have been shown to affect pathway function. These elements and their effects on general ubiqutin conjugating activity in muscle are described below.

Ubiquitin mRNA levels commonly increase in conditions that stimulate pathway activity. Presumably, increased gene expression expands the pool of ubiquitin monomers available for protein marking. Subunits of the 26S-proteasome complex are also upregulated in a subset of these conditions; this may increase the net ability of the cell to degrade upiquitin-tagged proteins. In contrast, E1 gene expression appears to be unaffected by stimuli that increase pathway activity. Upregulation may be unnecessary since E1 protein is constitutively expressed at high levels and is not thought to regulate pathway specificity.

Upregulation of E2 and E3 gene products appears to be essential for a general increase in pathway activity. Among these, E2–14k and E3α were the first to be clearly identified as inducible pathway elements in muscle. Both proteins are upregulated in experimental models of catabolism (21, 34, 36, 64) and function as an E2/E3 pair to target substrates for degradation via the N-end rule pathway (33). Up to 60% of muscle proteins can be degraded via this mechanism. However, the remainder of muscle proteins, including the myofibrillar proteins myosin and actin, are not N-end rule substrates and are insensitive to E2–14k/E3α action (58).

UbcH2/E2–20k is a second E2 protein that increases general pathway activity and is inducible in striated muscle (37). UbcH2 was originally isolated from human placenta (28). The human gene is a homolog of murine E2–20k (70) and has a strong sequence homology with yeast UBC8. The UbcH2 gene is located on human chromosome 7 and has a complex expression pattern, with at least five different RNA transcripts. Among human tissues, UbcH2 is preferentially expressed by skeletal and cardiac muscles (37), suggesting a role in physiological protein turnover. Constitutive expression is also evident in striated muscles of rodents and in cultured myotubes (37, 47). The UbcH2 gene product is upregulated by unloading of cardiac muscle (47) and by exposing muscle to TNF-α either in vivo or in cell culture (37). In the latter case, UbcH2 upregulation is stimulated by NF-κB and is essential for a general increase in ubiquitin conjugating activity (37).

Atrogin1/MAFbx and MuRF-1 are muscle-specific E3s that are commonly upregulated when general activity increases. The atrogin1/MAFbx gene is expressed constitutively in adult mouse skeletal muscle and in differentiated myotubes. Gene expression is increased in response to fasting, diabetes, cancer, renal failure, and experimental sepsis (6, 16, 80). Atrogin1/MAFbx mRNA is also elevated in cultured myotubes exposed to reactive oxygen species (36), glucocorticoids (55, 61), or TNF-α (35). Transcription of the atrogin1/MAFbx gene is regulated by the Foxo family of transcription factors, a pathway inhibited by IGF-I/Akt signaling (52, 55, 61). Atrogin1/MAFbx expression is also regulated by mitogen-activated protein kinase (MAPK) activity, in particular, p38 MAPK (35). Experimentally induced atrophy is ameliorated in atrogin1/MAFbx-deficient mice (6), an observation that demonstrates the physiological importance of this protein. Similarly, the MuRF-1 gene is constitutively expressed, is often upregulated when general activity increases, and appears to mediate net loss of muscle protein (6, 55, 61).

EFFECTS OF MUSCLE USE ON PATHWAY ACTIVITY

Activity of the ubiquitin-proteasome pathway is transiently increased in response to altered muscle use. This is true for conditions of decreased use that include denervation (6, 39, 40, 44, 78), immobilization (6, 27), and gravitational unloading (6, 20, 21, 44, 48, 49). Markers of pathway activity are also altered by the increase in muscle use that occurs during exercise (27, 46, 53, 54, 59, 62, 66, 69, 77). Response of the pathway to a change in muscle use appears to vary according to intensity and duration of the intervention. Increases in pathway activity are expected to stimulate both signaling events and general proteolysis, as illustrated in Fig. 2. The growing body of research that underlies this model is reviewed in the following sections:

Fig. 2.

Effects of altered muscle use on pathway regulation. Pathway depicts reported response of ubiquitin-proteasome pathway to decreases or increases in muscle use. Alterations in activity act via second messengers (ROS, kinases, phosphatases) to stimulate targeting of specific proteins by constitutive pathway elements. Altered activity also acts via pathway-independent signals (Foxo, p38, other?) and pathway-dependent signals (MyoD, NF-κB) to upregulate inducible pathway elements (ubiquitin, E2/E3 proteins, proteasome subunits), increasing general activity of the pathway and promoting degradation of muscle proteins. Net protein balance reflects the magnitude of two opposing rates, degradation vs. synthesis. Depending on concurrent synthesis rates, increases in pathway activity may occur with net protein loss (atrophic states), net protein increase (hypertrophic remodeling), or no change (adaptation).

Muscle use and pathway-dependent signaling.

The ubiquitin-proteasome pathway regulates signaling events by targeted degradation of individual proteins, a process sensitive to muscle use. The myogenic transcription factor MyoD is a muscle-specific example of such regulation. MyoD signaling is opposed by the ubiquitin-proteasome pathway which selectively degrades the MyoD protein (1). Substantial increases or decreases in muscle use stimulate general activity of the pathway (see below), which predisposes muscle to MyoD depletion. Net loss of MyoD appears to be limited by compensatory upregulation of MyoD gene expression in the affected muscle; elevated mRNA levels have been observed after denervation (12, 23), spinal cord transection (10), tetrodotoxin-induced paralysis (79), spaceflight (22), hindlimb unloading (42, 71), exercise (3, 5, 45), and overloading (2).

A second example is regulation of NF-κB signaling by the ubiquitin-proteasome pathway. As described earlier, ubiquitin conjugation targets NF-κB inhibitor proteins for degradation via the 26S-proteasome, thereby activating NF-κB. Recent studies indicate this ubiquitin-regulated signaling event is sensitive to muscle use. Hunter et al. (19) found that 7 days of mechanical unloading increases NF-κB activity in mouse soleus, a postural muscle that opposes gravity. Protein analyses suggest the response was mediated by an alternative NF-κB pathway that preferentially involved the p50 monomer. Durham and associates (11) confirmed the effect of unloading on mouse soleus, demonstrating increased NF-κB activity after 12 days of hindlimb unloading. Elevated NF-κB signaling is consistent with the increase in general activity of the ubiquitin-proteasome pathway that is observed in muscle after prolonged unloading (see below). In contrast, early studies of exercise effects on NF-κB signaling have yielded conflicting results. Durham et al. (11) found that fatiguing exercise increased NF-κB activity in vastus lateralis of healthy volunteers and in mouse soleus studied in vitro. Several months later, Ji and associates (26) reported the opposite response, an increase of NF-κB activity in limb muscles of rats run to exhaustion on a treadmill. The explanation for this apparent contradiction is not obvious. Both groups studied limb muscle samples obtained immediately after fatiguing exercise; both assessed DNA binding activity of NF-κB in nuclear extracts using electrophoretic mobility shift assay; and both provide credible data that support their (opposite) conclusions. Additional research will be required to resolve this emerging issue.

Pathway upregulation with decreased muscle use.

Markers of ubiquitin-proteasome pathway activity are increased by conditions that diminish muscle use, including gravitational unloading, limb immobilization, and muscle denervation. Table 1 provides a list of these markers and the conditions under which they were found to increase. The most commonly studied markers include ubiquitin-conjugated proteins, ubiquitin protein, and ubiquitin mRNA. Over a decade ago, Riley and associates (49) demonstrated that muscle fiber breakdown in antigravity muscles of space-flown rats was correlated with increased ubiquitin conjugation to myofibrillar proteins. This observation was corroborated by the same group (48) and by Ikemoto and colleagues (21); both of these later studies demonstrated that unloading by hindlimb suspension produces similar changes. Lalani et al. (31) failed to detect changes in ubiquitin mRNA levels in rat antigravity muscles conditioned by 17 days spaceflight. However, Taillandier et al. (64) showed that ubiquitin mRNA content is increased in muscles conditioned by hindlimb unloading. Subsequent studies by Ikemoto and associates (21) and Stevenson et al. (60) defined the time course of ubiquitin upregulation; mRNA levels are increased after 24 h unloading and peak at 4 days with expression remaining elevated for at least 14 days. Medina and colleagues (39, 40) detected similar increases in ubiquitin-conjugated proteins and ubiquitin mRNA in denervated limb muscles. Wing and associates (78) showed that the increase in ubiquitin-conjugated protein after denervation is concomitant with increased proteolysis. However, it was Tawa et al. (65) who established causality by showing that pharmacologic blockade of proteasome activity could inhibit denervation-induced atrophy. Razeghi and colleagues (47) studied atrophic remodeling of the rat heart during mechanical unloading induced by heterotopic transplantation. They found simultaneous activation of pathways that regulate both protein synthesis and protein degradation. Evidence for the latter included an increase in polyubiquitin content that was detectable at 2 days and persisted for 28 days. A similar increase in ubiquitin-protein conjugates was observed by Deruisseau and associates (9) in studies of diaphragm unloading. Atrophy was induced in the diaphragm of rats by continuous mechanical ventilation. After 12 hr, ubiquitin-protein conjugates were increased in both the myofibrillar and cytosolic fractions of the diaphragm.

View this table:
Table 1.

Markers of pathway activity that increase with diminished muscle use

Research to identify the mechanism of increased activity has focused on inducible gene products that code for proteasome subunits and E-proteins (Table 1). Ikemoto and colleagues (21) have established that hindlimb unloading increases enzymatic activity of the proteasome in skeletal muscle. Various models of diminished muscle use suggest that increased proteasome function is regulated at the level of subunit transcription. Medina and colleagues (39) found that denervation upregulated proteasome subunits C1, C2, C5, C8, and C9. Taillandier et al. (64) showed that hindlimb unloading increases expression of two 20S proteasome subunits, C2 and C9; similarly, Ikemoto and colleagues (21) reported upregulation of RC2 and RC9 subunits. In a microarray study, Stevenson and associates (60) detected a broader panel of unloading-responsive genes for seven distinct proteasome subunits. Gene products that code for E2 and E3 proteins are also induced by diminished muscle use. Both Taillandier and colleagues (64) and Ikemoto et al. (21) measured increased levels of E2–14k mRNA after gravitational unloading. Nikawa and associates (44) used DNA microarray analysis to compare rat muscles conditioned by spaceflight, hindlimb unloading, and denervation. Among the many genes altered, these investigators found that atrophy caused by space flight is uniquely linked to upregulation of three E3 proteins: MuRF1, Cbl-b, and Siah-1A. Bodine and associates (6) found differences in the patterns of gene expression among muscles conditioned by gravitational unloading, limb immobilization, and muscle denervation. However, they noted that two E3 proteins, atrogin/MAFbx and MuRF1, were upregulated in all three forms of diminished muscle use. These two genes appear to represent common denominators in the molecular program that regulates muscle atrophy (14). However, recent data by Rourke et al. (51) suggest muscle atrophy is not an obligatory outcome of E3 upregulation. These investigators observed that atrogin1/MAFbx mRNA levels were elevated in soleus and plantaris muscles of ground squirrels after 3–4 mo hibernation. Nevertheless, neither muscle was overtly atrophic compared with values measured before hibernation.

The report by Stevenson et al. (60) provides detailed microarray data on the timing of gene expression changes in unloaded hindlimb muscles of mice. These investigators detected increased mRNA levels for three E3-proteins, confirming the response of atrogin1/MAFbx and MuRF1 (6, 44) and discovering a parallel rise in message for Nedd4. Several E2 proteins were also upregulated but did not achieve the two-fold increase required for formal analysis. Genes for eight proteasome subunits were upregulated in the same muscles. These included C9, as reported previously (39, 64), and six subunits not previously identified as responsive to unloading: HC10-II, RN3, MSS1, Tat-binding 7, p112, and S5A. Expression of all eight proteasome subunits was increased after 4 days of unloading; all but p112 returned to baseline after 14 days.

Changes in E-protein gene expression also appear to regulate the atrophy caused by unloading of respiratory and cardiac muscles. After 12 h of mechanical ventilation, Deruisseau and associates (9) observed selective upregulation of mRNAs for atrogin1/MAFbx and MuRF1. There were no changes in mRNA for E2–14k, the polyubiquitin gene, or components of the proteasome complex (proteasome activating complex PA28, C8 subunit). In studies of unloaded rat hearts, Razeghi et al. (47) found that UbcH2 mRNA levels were transiently increased 2 days after unloading, a response that reversed within 4 days. The response of UbcH2 protein was more prolonged; protein levels increased after 2 days and remained elevated through day 7, returning to baseline at day 28.

Pathway responses to exercise.

Strenuous exercise stimulates adaptation of skeletal muscle. Cellular adaptation requires selective degradation of existing proteins, a process primarily regulated by the ubiquitin-proteasome pathway (15). To meet this challenge, general activity of the pathway is increased. This likely reflects the selective upregulation of inducible, exercise-responsive pathway components. The current list of genes known to respond to exercise is shown in Table 2.

View this table:
Table 2.

Markers of pathway activity that increase with exercise

The response to exercise is best documented for ubiquitin, which appears to be upregulated by several types of exercise conditioning. Sandri et al. (53) reported that ubiquitin protein levels increase in the muscles of mice after a night of spontaneous wheel running. The same group later confirmed this observation (54) and reported that increases in ubiquitin levels correlate with postexercise increases in the number of apoptotic myonuclei (46). Eccentric exercise also increases ubiquitin content. Thompson and Scordalis (66) evaluated muscle biopsies obtained from human biceps 2 days after a bout of damaging eccentric-isokinetic exercise. They found a 55% increase in free ubiquitin after exercise, relative to nonexercised control muscle, and similar increases in ubiquitin-conjugated proteins. Stupka and associates (62) subsequently assessed muscle biopsies from untrained subjects who performed eccentric leg press and knee extension exercise on two occasions, separated by 5.5 wk. Biopsies collected from vastus lateralis 24 h after each bout of exercise showed increases in ubiquitin-conjugated protein content, the response being greater after the second bout. Willoughby and colleagues (77) measured muscle ubiquitin content after eccentric knee extension exercise performed by healthy volunteers on two successive trials. They observed increases in ubiquitin mRNA and protein content that were larger after the first trial than the second. Similar increases were detected in mRNA and protein levels of E2 protein and the 20S proteasome; both proteins were upregulated 6–24 h after each of the repeated exercise trials. Wakshlag and colleagues (69) observed an increase in ubiquitin-conjugated proteins and the proteasomal p31 capping subunit within muscles of working field dogs during the peak hunting season compared with values measured in the preseason. Finally, Sonna et al. (59) studied military trainees with exertional heat injury caused by intense field exercises. These investigators detected increased ubiquitin mRNA levels in peripheral blood mononuclear cells; ubiquitin regulation in muscle was not assessed.

In contrast, Jones and associates (27) reported that mRNA levels for atrogin1/MAFbx and MuRF1 were decreased in human quadriceps after an acute bout of knee extension exercise. This observation appears to conflict with the increases in pathway activity described by other investigators (above) but is not necessarily in opposition. It may be that decrements in atrogin1/MAFbx and MuRF1 mRNAs did not translate into decreased protein levels. It may be that general pathway activity is increased after exercise via other E3 proteins that are under parallel control (32). Or the data of Jones et al. (27) may reflect a specific response of preconditioned muscles, as their subjects underwent 2 wk immobilization that ended only 24 h before the exercise bout.

Taken together, these data suggest that exercise effects on the ubiquitin-proteasome pathway are task specific. Eccentric exercise produces the most robust changes, increasing ubiquitin conjugation to muscle proteins and altering the expression of pathway components including ubiquitin, E2 proteins, and proteasome subunits. Concentric exercise tasks stimulate ubiquitin conjugation but appear to have less effect on expression of pathway genes. Isometric exercise has been shown to stimulate pathway-mediated NF-κB signaling, but transcriptional responses have not been assessed. Interestingly, the effect of exercise intensity on either pathway signaling or transcriptional regulation has not been tested; absence of such dose-response information is an obvious gap in the literature.

The primary data described in preceding sections also suggest that the pathway response to a single bout of exercise is multiphasic and time dependent. This is illustrated in Fig. 3, which depicts a hypothetical model of pathway activity during and after an exercise bout. The onset of exercise appears to stimulate ubiquitin conjugation within seconds-to-minutes. This initial response reflects a transient change in the function of constitutively expressed E-proteins. It contributes to exercise-related signaling via NF-κB, likely influences muscle gene expression, and spontaneously reverses after exercise is terminated. A delayed increase in general pathway activity appears to occur within hours-to-days and to be caused by upregulation of exercise-inducible elements: ubiquitin and selected E-proteins. This secondary response is likely to mediate muscle remodeling, regulating the degradation of damaged or modified proteins for replacement during muscle adaptation. Finally, a late-phase decline in ubiquitin conjugation appears to occur days-to-weeks after a single exercise event. This final decrement reflects recovery from the exercise stimulus and the return to basal state. Note that this model is consistent with the published literature but is not definitive; it is proposed as a testable hypothesis to be refined or refuted by future experiments.

Fig. 3.

Hypothetical model of changes in pathway activity after a bout of exercise. Model integrates the changes in pathway activity that are reported to occur after a single bout of exercise (see text): 1) specific signaling via constitutive pathway components is altered during exercise, a rapid response of short duration (minutes-to-hours); 2) the increase in specific signaling is transient, reversing when exercise ends; 3) upregulation of exercise-inducible components causes a delayed increase in general pathway activity (6–24 h); and 4) in the absence of additional exercise bouts, general activity spontaneously declines to basal levels (24–72 h).

Pathway responses to reloading and training.

Limb muscle use can be abruptly increased by returning animals to normal ambulation after a prolonged period of gravitational unloading. Riley and associates (49) suggested that the return to gravitational loading (reloading) was likely to stimulate muscle remodeling and increase activity of the ubiquitin-proteasome pathway activity. Taillandier and colleagues (63) recently addressed this issue by analyzing rat antigravity muscles conditioned by reloading. After 18 h, they found that ubiquitin-conjugated protein levels were increased. mRNA levels for ubiquitin and two proteasome subunits, C8 and C9, were elevated and were being actively transcribed, whereas E2–14k mRNA levels were paradoxically decreased. The mRNA changes had reversed after 7 days of unloading, with all three gene products returning to basal levels, but ubiquitin conjugates remained elevated. Increases in these markers of pathway activity were accompanied by a sustained increase in protein synthesis rate during reloading. The authors concluded that elevated rates of degradation and synthesis are both required for remodeling of reloaded muscle and that noncoordinate regulation of inducible pathway elements enables more effective targeting of ubiquitin to substrates that require degradation.

Several studies suggest that repetitive exercise has a training effect that tends to inhibit the ubiquitin-proteasome pathway. Willoughby et al. (75) studied muscles of persons with spinal cord injury before and after a 12-wk program of passive leg cycling exercise. The program decreased mRNA levels for E2 protein and the 20S proteasome in exercised muscles; in contrast, expression of myosin heavy chain types IIa and IIx was increased. This same group used a similar protocol and subject population to confirm that training depressed both message and protein content for E2, whereas heat shock protein 72 (HSP-72) and myofibrillar protein contents were both increased (76). In a subsequent study of healthy subjects, Willoughby and associates (77) demonstrated inhibitory effects using only two bouts of eccentric exercise. Ubiquitin, E2 protein, and 20S proteasome mRNAs were upregulated in muscle biopsies taken 6 and 24 h after both exercise bouts. These changes were blunted in the second exercise bout relative to the first. Kee et al. (30) analyzed muscles of rats after five consecutive days of treadmill exercise. Relative to muscles of unexercised rats, they found decrements in protein breakdown rate, chymotrypsin-like activity of the proteasome, and the rate of ubiquitin-dependent casein hydrolysis. These functional changes were not associated with decreases in mRNA for ubiquitin, E2–14k, or the β-1 proteasome subunit. Jones and associates (27) measured changes in gastrocnemius mRNA levels after 2 wk of knee immobilization. During 6 wk of exercise rehabilitation, they detected decrements in mRNA levels for atrogin1/MAFbx, MuRF1, and the 20S proteasome.

CONCLUSIONS

Recent research indicates that changes in muscle use, either increases or decreases, can alter activity of the ubiquitin-proteasome pathway. Pathway activity may change abruptly, by adjusting the function of constitutive components, or may be altered over longer periods via changes in the composition of pathway components. The ubiquitin-proteasome pathway appears to be important both for muscle adaptation and muscle atrophy. Accordingly, pathway regulation is likely to influence processes that range from marathon training to postoperative rehabilitation, from body building to the sarcopenia of aging. This highlights the importance of unanswered questions about underlying the mechanism: What are the signals generated during muscle contraction or muscle inactivity that modulate pathway activity? What regulates transcription of pathway-related genes? What molecular modifications stimulate ubiquitin conjugation to muscle proteins? Which protein substrates are recognized by specific E2/E3 pairs? And what is the rate-limiting step in targeting and degradation of substrate proteins? Clearly, these and related questions can only be addressed by continued research.

GRANTS

Our research in this area is funded by National Institutes of Health Grant HL-59878 and the National Space Biomedical Research Institute.

Acknowledgments

The author is grateful to Drs. Yi-Ping Li, William Durham, Stewart Lecker and Alfred Goldberg for collaboration and generous advice on this topic and to Dr. Jennifer Moylan for assistance with graphics.

REFERENCES

View Abstract