| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
NONRUMINANT NUTRITION |
Muscle Biology Group, University of Arizona, Tucson 85721
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONSKELETAL MUSCLE STRUCTURE AND...PROTEOLYTIC SYSTEMS IN MUSCLETHE CALPAIN SYSTEM AND...THE PROTEASOME AND SKELETAL...LITERATURE CITED |
|---|
Key Words: calpain • easily releasable myofilament • myofibrillar protein turnover • proteasome
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONSKELETAL MUSCLE STRUCTURE AND...PROTEOLYTIC SYSTEMS IN MUSCLETHE CALPAIN SYSTEM AND...THE PROTEASOME AND SKELETAL...LITERATURE CITED |
|---|
). Emphasis on enhancing muscle growth in domestic animals traditionally has focused on increasing rate of muscle protein synthesis by optimizing nutrition, management practices, etc. This emphasis has at least partly been due to the amount of information available on muscle protein synthesis and ways to modify it. Until recently, there has been much less information available on intracellular protein turnover. Nearly 40 yr ago, Goldberg (1969a,b) first showed that rate of intracellular protein turnover in skeletal muscle was not an unchanging background against varying rates of protein synthesis but that rate of protein degradation varied substantially in response to physiological demand. Therefore, the rate of skeletal muscle growth can be enhanced not only by optimizing the rate of muscle protein synthesis, as has been the emphasis in animal science for the past 100 yr, but also by decreasing the rate of muscle protein degradation. Moreover, because 15 to 25% of the ingested energy in domestic animals is used to replace muscle protein that has been degraded during metabolic turnover (Young et al., 1975; Millward and Garlick, 1976), decreasing the rate of muscle protein turnover results directly in an increase in the efficiency with which ingested nutrients are converted into muscle protein (Goll et al., 1992, 1998). Consequently, understanding how muscle proteins are turned over metabolically and how this turnover is regulated has important implications to muscle growth. |
SKELETAL MUSCLE STRUCTURE AND PROTEIN TURNOVER |
|---|
|
TopAbstractINTRODUCTIONSKELETAL MUSCLE STRUCTURE AND...PROTEOLYTIC SYSTEMS IN MUSCLETHE CALPAIN SYSTEM AND...THE PROTEASOME AND SKELETAL...LITERATURE CITED |
|---|
Classes of Muscle Proteins
Skeletal muscle contains 3 groups of protein when classified by solubility and their location in the muscle tissue.
Sarcoplasmic proteins constitute ~30 to 35% of total protein in muscle tissue by weight. These are the cytoplasmic proteins that are soluble in low (<0.2 M)-salt solutions and comprise all the glycolytic enzymes, enzymes involved in metabolic pathways, etc. This class of muscle proteins contains at least several hundred different polypeptides and is located entirely intracellularly.
Myofibrillar proteins constitute 55 to 60% of total protein in muscle tissue by weight. These are the proteins that constitute the myofibril or contractile structure in skeletal muscle. Although some myofibrillar proteins such as 
-actinin and CapZ are soluble at low ionic strength after they have been extracted from the myofibril and separated from their normal binding partners, high ionic strengths (0.3 M) are required for their initial extraction. Myosin and actin are the 2 main proteins in this group, although more than 15 other proteins are also present in the myofibrillar structure. The myofibrillar proteins are located entirely intracellularly.
Stroma proteins constitute 10 to 15% of total protein in muscle tissue and are defined as those proteins that are insoluble in an aqueous solvent at neutral pH. Many of the proteins in this group are extracellular. Collagen and extracellular matrix proteins are the main proteins in this group, although some membrane proteins may also be included, because they are not soluble in the absence of detergents.
The myofibrillar proteins are not only the largest class of skeletal muscle proteins but also are responsible for the contractile properties of muscle and for most of the functional and culinary properties of muscle and meat. Thus, studies on muscle growth and muscle protein turnover need to focus on the myofibrillar proteins.
Myofibrillar Protein Turnover
Myofibrils are unique to striated muscle and also present a special situation for metabolic turnover. The contractile function of myofibrils requires that the myofibrillar structure extend continuously from one end of the muscle cell to the other (Figure 1). Thus, turnover of myofibrillar proteins must be accomplished without disrupting this continuous structure. It was proposed over 30 yr ago (Dayton et al., 1975) that myofibrillar proteins can be turned over by releasing filaments from the surface of the myofibril, leaving a myofibril with a diameter that is smaller by 1 layer of myofilaments (Figure 2). This mechanism is consistent with the observations that atrophying muscle in different muscular dystrophies, after denervation or during fasting, has smaller diameter myofibrils than unaffected muscle (Badalamente and Stracher, 2000). Dayton et al. (1975) proposed that the calpains, which had just been isolated and purified from skeletal muscle (Dayton et al., 1976a,b), were responsible for release of myofilaments from the surface of myofibrils, because the calpains cleave many of those proteins that are involved in keeping myofilaments attached to the myofibril. The calpains rapidly cleave titin and nebulin at the point where these 2 polypeptides enter the Z disk (Goll et al., 2003). These cleavages, together with cleavage of desmin and filamin, which encircle the myofibril at the Z disk and tether it to the sarcolemma, would release -actinin (Goll et al., 1991), the principal Z disk protein, from the myofibril. Release of -actinin would result in release of thin filaments from the surface of the myofibril. The calpains also degrade M proteins, tropomyosin and troponin, albeit at slower rates than the degradation of titin and nebulin (Goll et al., 1992, 1999). Cleavage of the M proteins, together with the cleavage of titin, severs the attachments of the thick filament to the myofibril, and in the presence of ATP to dissociate myosin crossbridge binding to thin filaments, the thick filaments would be released from the myofibril. Calpain cleavage of tropomyosin and troponin would facilitate disassembly of the thin filament to G-actin monomers, and calpain cleavage of C-protein would facilitate disassembly of the thick filament to myosin monomers (lower right, Figure 2), making both actin and myosin susceptible to degradation by the proteasome. The calpains do not cleave proteins to AA but rather make a few selective cleavages leaving large fragments. Because the calpains cleave those proteins that are involved in maintaining the myofibrillar proteins in the myofibrillar structure, they seem to be ideally suited for catalyzing the first step in myofibrillar protein turnover. Although a role of the calpains in myofibril disassembly as a first step in their metabolic turnover has been widely accepted, there has been little experimental evidence to directly support this role.
|
|
If myofibrillar protein turnover proceeds via the mechanism described in Figure 2
, muscle should contain a small population of myofilaments that have been dissociated from the surface of the myofibril and that represent those myofilament intermediates on the path to turnover. In the late 1970s and early 1980s, Etlinger and coworkers (Etlinger et al., 1975; van der Westhuyzen et al., 1981) discovered that approximately 10 to 15% of total muscle protein in skeletal and cardiac muscle can be dissociated from intact myofibrils in the form of myofilaments by gentle agitation in an ATP-containing solution (Figure 3). The release of these myofilaments did not require hydrolysis of ATP, because it occurred equally as well in the presence of AMP-PNP, a nonhydrolyzable analog of ATP, as in the presence of ATP and is not affected by pH in the range of 6.5 to 7.5 (Belcastro et al., 1991b). These easily releasable myofilaments (ERM) lack -actinin, desmin, titin, and nebulin (van der Westhuyzen et al., 1981; Belcastro et al., 1991a,b), proteins that are degraded or released from myofibrils by the calpains (Dayton et al., 1975; Goll et al., 1991, 1992, 2003), but contain actin and myosin (Figure 4), proteins that are not degraded or are degraded only very slowly (S. W. Mares, V. F. Thompson, and D. E. Goll, unpublished results) by the calpains.
|
|
) and decreases upon addition of calpain inhibitors such as leupeptin (van der Westhuyzen et al., 1981) or E-64 (Dahlmann et al., 1986). We have found that incubation with purified calpain at very low ratios (1:2,000 wt/wt) of calpain to myofibrillar protein followed by trituration in the presence of ATP increases the amount of ERM protein in a crude myofibril preparation by 3- to 4-fold (G. Neti and D. E. Goll, unpublished results). The ERM released by calpain treatment are ultrastructurally identical to the ERM obtained by using trituration of crude myofibrils without calpain treatment in the presence of ATP and have the same polypeptide composition in SDS-PAGE as the ERM obtained by using the normal procedure (G. Neti and D. E. Goll, unpublished results). There are several reasons to believe that the ERM are part of the upstream events in myofibrillar protein turnover and are not simply recently synthesized myofibrillar proteins that have assembled and are ready to be added to the myofibril. First, addition of corticosterone, which increases the rate of muscle protein turnover, to cultures of chick muscle cells increases the amount of ERM that can be obtained from this muscle model, whereas nifedipine, a Ca2+ channel blocker, prevents this increase in ERM (Machida et al., 2003). Second, sepsis (Williams et al., 1999), starvation (Dahlmann et al., 1986), and diabetes (Belcastro et al., 1991a) also all increase the rate of myofibrillar protein turnover and increase the amount of ERM in muscle from the affected animals. Finally, pulse-chase studies initially suggested that the ERM were intermediates in the turnover of myofibrillar proteins (van der Westhuyzen et al., 1981). Kinetic studies (Clark, 1993) have shown that proteins in adult cardiac myocytes are degraded from 2 different pools: 1 that comprises ~10% of total muscle protein and turns over with a half-life of 11.9 h and 1 that turns over with a half-life of 15.6 d. This observation is consistent with the hypothesis that the former rapidly turning-over pool contains ERM, whereas the latter pool consists of intact myofibrils. Although it is not certain that the mechanism or pathway used by muscle cells to remove myofibrillar proteins during periods of rapid muscle loss is the same as that used to turn over myofibrillar proteins during muscle growth, the evidence showing that treatments that induce muscle wasting also increase the amount of ERM in that muscle indicates that the ERM have some role in myofibrillar protein turnover. Although a substantial amount of circumstantial evidence supports the role of ERM and their release by the calpains as the first step in metabolic turnover of myofibrillar proteins, direct evidence supporting this mechanism is lacking. Release of ERM from the surface of the myofibril to turn over myofibrillar proteins would imply that the interior core of myofibrils turns over very slowly, if at all.
Exchange of Proteins
Many reports, most of them appearing in the 1980s and early 1990s, indicated that proteins in myofibrils may exchange with their counterparts in the cell cytoplasm or the surrounding medium and suggested that direct exchange of proteins may be involved in turnover of the myofibrillar proteins in striated muscle. The protein composition of myofibrils changes throughout life as embryonic isoforms of myosin and other myofibrillar proteins are replaced by mature isoforms. Even in mature muscle, the myosin isoform content of myofibrils changes in response to physiological demand. This isoform change must occur in continuously functioning muscle, and the available evidence suggests that the newly incorporated isoforms are present throughout the myofibril from its interior to its surface. Such a distribution would not occur if myofibrillar proteins turned over only via loss of easily releasable myofilaments and addition of new myofilaments on the surface of the myofibril. It has been difficult, however, to obtain conclusive evidence showing that proteins in myofibrils, especially in the interior of the myofibril, can exchange with their counterparts in the surrounding cytoplasm.
An early study showed that if muscle-specific poly(A)+RNA is translated in a reticulocyte in vitro translation system, and skeletal muscle myofibrils were added to the system after translation, a specific subset of the translated proteins bound to the myofibrils (Bouché et al., 1988). These bound proteins were all identified as sarcomeric muscle proteins. If, on the other hand, skeletal muscle myofibrils were added to a reticulocyte in vitro translation system after translation of proteins encoded by brain poly(A)+RNA, only 1 polypeptide, identified as β-actin, bound to the myofibrils. It wasn’t clear at the time whether the newly synthesized myofibrillar proteins had been incorporated into functioning myofibrils or had merely adhered to the surface of the added myofibrils, but the results raised the possibility that the protein composition of myofibrils may change rapidly in response to the type of sarcomeric protein being translated by an adjacent mRNA.
Many studies have suggested that the proteins assembled in myofibrils are in an exchange equilibrium with a pool of protein monomers in the cytoplasm. A critical shortcoming of many of these studies, however, is that they have used actin (thin) filaments or myosin (thick) filaments that were assembled in vitro from purified proteins; these synthetic thick and thin filaments do not contain the accessory proteins that thick and thin filaments contain in situ. Thick myosin filaments in situ contain the following: a) C-protein, which surrounds the thick filaments like staves around a barrel every 43 nm (Bennett et al., 1986); b) H-protein and X-protein, which colocalize with certain C-protein bands (Bennett et al., 1986); c) the M-line proteins, myomesin and creatine kinase, and d) titin. Because these accessory proteins surround the myosin molecules that constitute the shaft of thick filaments, it seems highly probable that they would have important effects on the ability of myosin molecules to exchange with molecules in the surrounding medium. Similarly, thin actin filaments in situ contain the following: a) tropomyosin, which forms 2 strands that lie in the groove of the actin filament; b) troponin, which sets on the surface of the actin filament at 38.7-nm intervals; c) nebulin, which lies on the surface of the actin filament and runs from the Z disk to the pointed end (the M-line end) of the actin filament; d) tropomodulin, which caps the pointed end of the actin filament; and e) many Z-disk proteins, which cap the barbed end of actin filaments. It seems probable that these accessory proteins will also have important effects on the ability of actin monomers to exchange with actin monomers in the surrounding medium. Furthermore, studies using synthetic thick and thin filaments to learn whether myosin and actin in these filaments exchange with myosin and actin in the surrounding medium do not allow for the constraints that having these filaments assembled in a myofibril would impose on the ability of actin and myosin molecules to diffuse in and out of the myofibril lattice. Therefore, it is difficult to interpret studies showing that, in a population of synthetic thick filaments in which 1 set had been made with labeled myosin and the other set contained only unlabeled myosin, 75% of the myosin molecules had been exchanged within 3 h of incubation (Saad et al., 1986, 1991). Is this exchange related to turnover of myosin in vivo? Similarly, it is difficult to determine whether the exchange of 5% of the actin monomers in purified F-actin after 20 min of incubation with labeled actin indicates that actin monomers would also exchange in thin filaments in situ (Pardee et al., 1982).
Several studies have examined the exchange of myofibrillar proteins in myofibrils, which contain native thin and thick filaments that have all the accessory proteins, or the exchange of myosin (Johnson et al., 1988) or actin (Glacy, 1983; McKenna et al., 1985) that had been microinjected into living cells with myosin or actin in the thick and thin filaments in the cells. Myosin heavy chain or myosin light chains (LC), LC1, LC2, and LC3, that had been synthesized in an in vitro reticulocyte cell-free translation system associated with chick skeletal muscle myofibrils (Goldfine et al., 1991). The association differed among the different myosin subunits; LC1 and LC3 were preferentially associated with the myofibrils, and less LC2 was associated. The myosin heavy chain bound by itself, indicating that the myosin subunits did not have to assemble before binding (i.e., being associated with) to myofibrils. No attempts were made to determine whether the myosin that associated with the myofibrils was actually incorporated into the interior of the myofibrils or was simply bound to the myofibrillar surface. Actin that also had been synthesized in a reticulocyte system bound to myofibrils with most of the binding occurring at the Z-disk level, and no binding was observed in the M-line region (Peng and Fischman, 1991). Confocal microscopy using optical analysis at every 0.5 µm of section depth indicated that the fluorescein isothiocyanate-actin incubated with the myofibrils could be observed throughout the myofibril, suggesting incorporation of the labeled actin and not binding to the surface of the myofibril (Figure 5).
|
-actinin bound to the Z disk. Either confocal or image deconvolution is required to determine at the light microscope level of resolution whether microinjected proteins are incorporated into the myofibrils in the living muscle cells or whether they are simply bound to the surface of the myofibril, and such studies have not been done. That microinjected nonmuscle actin bound to the myofibrils just as efficiently as microinjected muscle actin (McKenna et al., 1985) suggests that the binding was nonspecific and may not be a physiological mechanism involved in myofibrillar protein turnover.
A recent study found that ~15% of the -actinin in rabbit soleus myofibrils was labeled after 8 h of incubation of the myofibrils with fluorescently labeled -actinin in the medium; this change occurred with no change in total -actinin content of the myofibrils, indicating that the labeled -actinin was being exchanged with the endogenous -actinin (Swartz, 1999). Pretreatment of the myofibrils with fluorescent -actinin followed by incubation with unlabeled -actinin resulted in a decrease in fluorescence, again suggesting that an exchange was occurring. It was noted, however, that there seemed to be a pool of -actinin in the Z disks that was not exchanged regardless of incubation time.
Only the confocal microscopy observations and the recent -actinin studies have provided any evidence that the proteins being incubated with myofibrils are actually incorporated into the interior of the myofibril rather than simply being bound to the myofibrillar surface. It remains unclear how myosin polypeptides with a mass of 200 kDa could diffuse into a myofibril matrix without disrupting the contractile properties of the myofibril (this assumes that a myosin heavy chain can assemble with other myosin molecules in a thick filament; if the myosin molecule must be assembled before it is incorporated into a thick filament, then the molecular mass would be 550 kDa). In sum, therefore, it is still unclear whether the myofibrillar proteins can, in living cells, exchange with their counterparts in the cytoplasm or whether the studies thus far have simply observed proteins bound to the surface of myofibrils and not actually incorporated into them. As will be discussed subsequently in this review, it seems clear that the spacing of the myofibril lattice will not allow for exchange of proteins or polypeptides larger than ~200 kDa with other proteins in the interior of the myofibril. It is interesting that a great deal of information is available on the genetic mechanisms that regulate expression of genes encoding myofibrillar proteins, but there is very little definitive information available on how, in a living functioning muscle, the expressed proteins are assembled into a functional myofibril.
|
PROTEOLYTIC SYSTEMS IN MUSCLE |
|---|
|
TopAbstractINTRODUCTIONSKELETAL MUSCLE STRUCTURE AND...PROTEOLYTIC SYSTEMS IN MUSCLETHE CALPAIN SYSTEM AND...THE PROTEASOME AND SKELETAL...LITERATURE CITED |
|---|
The Lysosomal System
Proteases in this system (cathepsins) are located inside lysosomal structures and have acidic pH optima ranging from 3.5 up to 6.5. Because of their low pH optima, cathepsins are not active at the pH of cell cytoplasm (cathepsin B with a pH optimum of 6.0 to 6.5 may have some activity), so any role of cathepsins in muscle protein turnover would have to occur inside lysosomes. Moreover, cells including muscle cells contain cystatin, a potent inhibitor of Cys proteases such as cathepsin B and L. Because of their low pH optima and cystatin, cathepsins would not be active in cell cytoplasm. Myofibrils, which are 0.5 to 3.0 µm in diameter, are too large to be engulfed by lysosomes (which would result in severing the myofibril and loss of function). It is unclear whether the ERM discussed in the preceding section and that have dimensions of 8 nm (diameter) to 1,000 nm (length; i.e., thin filaments) or 15 nm (diameter) to 1500 nm (length; i.e., thick filaments) could be engulfed by lysosomal structures. Finally, normal skeletal muscle cells contain very few lysosomes, especially in comparison with organs like the liver or the spleen. Thus, it seems unlikely that lysosomal proteases are involved in metabolic turnover of myofibrillar proteins (Wildenthal et al., 1980; Lowell et al., 1986), although they may be responsible for necrotic degradation, especially during times of macrophage invasion of cells (Furuno and Goldberg, 1986; Lowell et al., 1986; Lecker et al., 1999). The primary role of lysosomal cathepsins is degradation of extracellular proteins that have been taken up via pinocytosis of receptor-mediated endocytosis and then transported by a series of vesicles to the lysosome, where they are degraded at the acidic pH in lysosomes.
The Caspase System
The caspases are responsible for degradation of proteins during apoptosis. The caspases are Cys proteases, as are the calpains and some of the cathepsins, but they do not require Ca2+ for activity, as do the calpains. It is not known at the present time whether the caspases can efficiently degrade myofibrils. Because the caspases are activated by the events that initiate apoptosis, it seems unlikely that they have significant activity in normal-functioning muscle cells, although they may become activated during periods of muscle wasting. Although Du et al. (2004, 2005) have shown that purified, activated caspase-3 can degrade actin to a 14-kDa fragment and have suggested that caspase-3 acts upstream from the proteasome in turnover of myofibrillar proteins, it is not clear how degradation of actin would contribute to disassembly of a myofibril while retaining its functionality. The actin degradation reported by Du and coworkers was limited, detected by Western blotting, and it seems unlikely that normal muscle cells (i.e., not apoptotic) would activate sufficient caspase-3 to contribute to metabolic turnover of myofibrillar proteins. Thus, it is unlikely that the caspases have an important role in metabolic turnover of myofibrillar proteins.
The Calpain System
The calpain system includes 14 different members, plus calpastatin, in those mammals that have been studied carefully (Goll et al., 2003). Skeletal muscle contains significant amounts of the 2 ubiquitous well-characterized calpains, the micromolar Ca2+, requiring Ca2+-dependent protease (µ-calpain), and the millimolar Ca2+, requiring Ca2+-dependent protease (m-calpain), and their specific inhibitor, calpastatin, and as discussed previously, the available information suggests that these 2 calpains and calpastatin are involved in turnover of myofibrillar proteins.
The Proteasome
The proteasome has a major role in intracellular protein degradation in all cells, including muscle cells. The unique properties of the proteasome indicate that it could not degrade intact myofibrils, and several studies have shown that it has no effect on intact myofibrils in in vitro systems. Thus, of the 4 major protease systems in skeletal muscle, the available evidence indicates that only 2, the calpain system and the proteasome, have a major role in metabolic turnover of myofibrillar proteins.
|
THE CALPAIN SYSTEM AND SKELETAL MUSCLE PROTEIN TURNOVER |
|---|
|
TopAbstractINTRODUCTIONSKELETAL MUSCLE STRUCTURE AND...PROTEOLYTIC SYSTEMS IN MUSCLETHE CALPAIN SYSTEM AND...THE PROTEASOME AND SKELETAL...LITERATURE CITED |
|---|
). The structural changes observed in rapidly atrophying muscle are mimicked closely by treatment of myofibrils from normal muscle with µ-calpain (Goll et al., 1991, 1992) or m-calpain (Dayton et al., 1975, 1976b; Goll et al., 1991, 1992). Additionally, SDS-PAGE has shown that the calpains cause degradative changes in skeletal muscle myofibrils similar to those seen in Duchenne muscular dystrophy (Sugita et al., 1980) or other rapidly atrophying muscle.
The calpain system is also involved in skeletal muscle growth. Administering various β-adrenergic agonists to animals results in a 10 to 30% increase in rate of accumulation of muscle mass. Although different studies using different species, different β-adrenergic agonists, and different conditions have produced slightly different results, most studies agree that administration of β-adrenergic agonists increases both the rate and efficiency at which muscle protein is accumulated. Administration of β-adrenergic agonists also affects the calpains system (Forsberg et al., 1989). Muscle calpastatin activity is increased significantly (Forsberg et al., 1989; Kretchmar et al., 1989, 1990; Bardsley et al., 1992; Parr et al., 1992, 2000) with this increase, ranging from 52 (Forsberg et al., 1989) to 430% (Kretchmar et al., 1990). It has been concluded, therefore, that β-adrenergic agonists increase the rate of skeletal muscle growth primarily by decreasing the rate of muscle protein degradation and that this decrease in rate of muscle protein degradation is principally due to an increase in calpastatin activity.
The callipyge sheep provide additional evidence for a role of the calpain system in skeletal muscle growth. Muscle from the hindquarters (e.g., the biceps femoris, semitendinosus, and longissimus dorsi) of sheep having the callipyge gene are 30 to 40% larger than the corresponding muscles from normal sheep, and callipyge sheep have greater feed efficiency than normal sheep of the same breed (Koohmaraie et al., 1995). Calpastatin activities in the hypertrophied muscles are 100 to 125% greater than calpastatin activities in the same muscles from normal sheep. On the other hand, in those muscles that are not affected by the callipyge gene and are not hypertrophied, the calpastatin activities are the same in the callipyge and normal sheep. Again, the evidence suggests that decreased calpain activity, likely mediated by increased calpastatin activity, is associated with an increased rate and efficiency of skeletal muscle growth.
Several studies have provided direct evidence that the calpain system is involved in turnover of skeletal muscle proteins. Overexpression of a dominant-negative form of m-calpain or of an inhibitory domain of calpastatin in a rat myogenic cell line, L8 myoblasts, decreased the rate of muscle protein degradation in these cells by 30 and 63%, respectively (Huang and Forsberg, 1998). Overexpression of a full-length human calpastatin transgene in mice reduced loss of muscle mass in the soleus during a 10-d unweighting period by 30% (Tidball and Spencer, 2002). Because calpastatin is absolutely specific for inhibition of the calpains, these studies demonstrate that the calpains are involved in muscle protein degradation. That calpastatin overexpression and the calpastatin domain did not inhibit 100% of muscle protein degradation is due to inability of the calpains to participate in sarcoplasmic protein turnover and to the fact that overexpression of calpastatin does not completely inhibit all calpain activity in the cell (thus, not even all myofibrillar protein degradation is ablated by overexpression of calpastatin).
In 1989, a mRNA encoding a polypeptide having 51 to 54% sequence homology to the 80-kDa subunit of µ-and m-calpain was identified in skeletal muscle (Sorimachi et al., 1989). The polypeptide encoded by this mRNA was named calpain-3; calpain-3 mRNA is transcribed at very high levels in skeletal muscle, in lower levels in cardiac and smooth muscle, and only at very low levels in other tissues. There has been a great deal of interest in calpain-3, because it was shown that disruption of the calpain-3 gene caused limb girdle muscular dystrophy type 2A. Because loss of calpain-3 results in muscle wasting, it seems unlikely that calpain-3 has a general degradative role in skeletal muscle but rather that it acts as a signaling protease. These properties and that calpain-3 has low proteolytic activity and has not been shown to degrade myofibrillar structures such as the Z disk makes it very unlikely that calpain-3 has a significant role in metabolic turnover of the myofibrillar proteins.
The µ- and m-calpains are unique in the following ways: 1) they do not degrade polypeptides to AA or even large peptides but rather make a few selective cleavages in proteins, leaving large polypeptide fragments that often retain catalytic but unregulated activity (Goll et al., 2003), 2) they do not degrade undenatured actin and degrade undenatured myosin only very slowly (S. W. Mares, V. F. Thompson, and D. E. Goll, unpublished results), and 3) they do not catalyze the bulk degradation of sarcoplasmic proteins in muscle, although they degrade kinases and phosphatases that are present in the sarcoplasm (Tan et al., 1988; Smith and Dodd, 2007). Because of these unique properties, the calpains cannot be involved in bulk turnover of sarcoplasmic proteins, and they do not degrade myofibrillar proteins to AA. Therefore, the role of the calpains in myofibrillar protein turnover must be restricted to the release of actin and myosin filaments from the myofibrils, that is, the release of ERM. A recent report has provided some direct evidence that the calpains act upstream of the proteasome in turnover of myofibrillar proteins and that calpain activation seems to provide protein substrates to the proteosome, thereby increasing its activity (Smith and Dodd, 2007). Also, because the calpains do not degrade polypeptides to AA, using release of AA to estimate muscle protein turnover cannot be used to implicate or discount the role of the calpains in this turnover.
It has been assumed that calpains release ERM from the surface of the myofibril, and the possibility that the calpains could release molecules from the interior of the myofibril has not been discussed. The calpain molecule is a prolate ellipsoid with dimensions of approximately 10 x 6 x 5 nm (Goll et al., 2003). The lattice spacing in myofibrils varies with state of contraction but ranges from 20 to 50 nm between thick filaments, approximately 10 to 20 nm between thick and thin filaments in the overlap area, and approximately 20 to 26 nm between thin filaments as they enter the Z disk (Goldstein et al., 1991). The lattice spacing in the Z disk is less than in the I- or A-band areas, ranging from 17 to 20 nm at the edge of the Z disk to less than that in the interior of the Z disk where -actinin crosslinks exist. Immunolocalization studies have shown that the calpains are concentrated in the I-band and Z-disk areas of the myofibril (Kumamoto et al., 1992), and the lattice spacings in the myofibril suggest that the calpain molecule could diffuse into the interior of the myofibril in the I-band area. Thus, the calpains could release actin filaments from the interior of the myofibril, although the effect that such release would have on function of the myofibril is unclear. It also seems very unlikely that any thick and thin filaments released in the interior of the myofibril by calpain could diffuse out, even if they were released by calpain.
Spacing of the myofibril lattice also is important when considering whether myofibrillar proteins could diffuse in and out of the myofibril as proposed by the exchange theory of myofibrillar protein turnover. Actin monomers (5.5-nm diameter) could easily diffuse in and out of the lattice. The myosin molecule, however, is a rod approximately 160- to 165-nm long, so it would be very difficult for it to diffuse in and out of the myofibril, even in the I-band area, and nearly impossible for it to diffuse into the A-band area, which is where it is located. The individual myosin heavy chains (which have been suggested to associate with myosin molecules in the myofibril; Goldfine et al., 1991) would also be 160 to 165 nm in length, but they may be in a random coil conformation that may enable their diffusion into the myofibril. The -actinin molecule is a prolate ellipsoid with dimensions of approximately 4 x 50 nm (Suzuki et al., 1976), so its diffusion into the Z disk would be very slow, although the results reported by Swartz (1999) suggest that it can exchange with -actinin molecules in the Z disk. Other myofibrillar proteins or protein subunits, such as troponin C, myosin light chains, and troponin I, are also small and could easily diffuse in and out of the myofibril lattice. The large myofibrillar proteins such as titin and nebulin could not diffuse in or out of the myofibril lattice. Calpain cleavage fragments of titin and nebulin could diffuse out of the myofibril, but such intramyofibrillar cleavage and loss of titin or nebulin fragments would seem likely to disrupt normal functioning of the myofibril.
|
THE PROTEASOME AND SKELETAL MUSCLE PROTEIN TURNOVER |
|---|
|
TopAbstractINTRODUCTIONSKELETAL MUSCLE STRUCTURE AND...PROTEOLYTIC SYSTEMS IN MUSCLETHE CALPAIN SYSTEM AND...THE PROTEASOME AND SKELETAL...LITERATURE CITED |
|---|
), but it was not until the early 1990s that the nature of the molecules responsible for this activity began to emerge (Bochtler et al., 1999; Glickman and Cienchanover, 2002; Cienchanover, 2005). It soon became evident that the proteasome is to intracellular protein turnover what the ribosome is to intracellular protein synthesis, and interest in the proteasome has mushroomed, leading to the Nobel Prize being awarded to its discoverers in 2004 (A. Hersko, A. Cienchanover, and I. Rose). Properties of the Proteasome
The proteasome is ubiquitous and essential in eukaryotes; ubiquitous but not essential in archaea, and rare and not essential in bacteria, which have a different ATP-dependent proteolytic system. The eukaryotic proteasome is involved in muscle protein degradation and is the proteasome that will be described here. The proteasome has 2 parts: a 20S core particle (CP) and a 19S complex that binds to the CP and acts to regulate its activity (S = the sedimentation coefficient of each particle); together they form the 26S proteasome). The 20S CP is a barrel-like ~700-kDa complex that is 15 nm in height and 11 nm in diameter and that in eukaryotic cells contains 28 different subunits grouped into 2 classes: and β subunits (14 of each). There are 7 different gene products for each of the and β classes named 1, 2, 3, .....7 and β1, β2, β3, .....β7. The and β subunits have all been cloned and sequenced; they range in size from 20 to 35 kDa. Although the individual subunits have sequence homology with each other and the individual β subunits have sequence homology with each other (but not with the subunits), they have no sequence homology with any other proteolytic enzyme. The 20S particle is the proteolytic particle and is arranged in 4 rings of 7 subunits each, with the subunits forming the 2 outer rings and the β subunits forming the 2 inner rings (Figure 6). Proteolytic activity of the proteasome resides exclusively with the β subunits; only the β1, β2, and β5 subunits have functional catalytic sites, hence, 6 catalytic sites in one 20S proteasome. The catalytic residue for all 3 sites is the hydroxyl group of a N-terminal Thr, making the catalytic mechanism of the proteosome different from the other 4 classes of proteolytic enzymes: Ser proteases, Cys proteases, Asp proteases, and metalloproteinases. The catalytic mechanism has been studied in detail and is similar to the well-characterized mechanisms of trypsin and chymotrypsin.
|
), and the crystallographic structure showed that all 6 catalytic sites are located in the interior of the central cavity (Löwe et al., 1995; Unno et al., 2002). The central cavity is guarded by the proteins constituting the 19S regulatory particle, so access to the central cavity is only 1.2 to 1.5 nm in diameter. Thus, protein molecules cannot enter the catalytic chamber without first being unfolded, and access to the central cavity is the mechanism that ultimately controls proteasome activity.
|
). The combination of the 20S CP and the regulatory subunit forms the 26S proteasome, a ~2,500-kDa particle that is extremely labile and tends to dissociate into the 20S particle and other subunits. The 19S regulatory complex can be divided into a lid and a base (Figure 8); the lid contains 8 different polypeptides that are involved in binding polyubiquitin chains and removing them from the polypeptide that is to be degraded so they can be recycled. The base contains 6 homologous ATPases and 3 polypeptides that are not ATPases. The ATPases use the energy of ATP to unfold the polypeptide entering the proteasome chamber; these enzymes may also be used in substrate binding. As will be discussed briefly in the next paragraph, ATP hydrolysis is also required to attach ubiquitin to the doomed polypeptide, so protein degradation via the proteasome pathway requires a significant amount of energy. It has been estimated that as many as 300 to 400 molecules of ATP may be required for degradation of a protein by the proteasome. If this is accurate, the proteasome qualifies as the SUV (stupid, useless vehicle) of biology.
|
, or with only one end or the other (Figure 7); the 20S particle may also exist as a free particle in cells. Although there is still controversy, it seems that both the 26S and the 20S proteasome can degrade proteins in living cells; degradation by the 26S proteasome requires ubiquitination and seems to be the principal route of intracellular protein degradation (Orlowski and Wilk, 2000, 2003). Targeting a Protein for Degradation: Ubiquitination
Ubiquitination of polypeptides and degradation of the ubiquitinated polypeptides by the 26S proteasome has been studied extensively. In the ubiquitination pathway, polypeptides that have been designated for destruction (how polypeptides are selected for this designation is unknown), are first labeled by attachment through an isopeptide bond to ubiquitin (a 76-AA, 8,565-Da protein found in all eukaryotic cells but not in prokaryotes). Attachment involves the C-terminal end of ubiquitin and usually occurs through an 
-amino group on the selected protein; it may also occur through the N-terminal AA of the selected protein, but the N-terminal amino group of many proteins (including most myofibrillar proteins) is blocked by acetylation, etc., precluding attachment to this group in these proteins. Ubiquitination is done in a series of enzymatic steps.
The first enzyme, called E1, activates the ubiquitin molecule in an ATP-dependent reaction to produce a high-energy E1-S ubiquitin complex (S designates the selected protein). The activated ubiquitin is transferred to 1 of a family of enzymes, the E2 enzymes (Figure 9) called the ubiquitin carrier proteins or ubiquitin-conjugating enzymes. The E2 complex then interacts with 1 of a large family of proteins, the E3 proteins, called ubiquitin ligases; the ubiquitin is transferred to the E3 enzyme, the E3-ubiquitin conjugate selects a doomed protein and transfers the ubiquitin to the selected protein; the E2-E3-ubiquitin-protein is then recognized by the proteasome (Figure 9).
|
). The ubiquitin molecules that are added after the initial ubiquitin all use the same E1, E2, and E3 series of enzymes and attach the second (third, etc.) ubiquitin to the -amino group of Lys48 of the first (second, third, etc.) ubiquitin. A single E1 enzyme can transfer the ubiquitin to any 1 of several E2 enzymes, which, in turn, can transfer their ubiquitin to any 1 of many E3 enzymes (Figure 9). This hierarchical arrangement evidently allows for specificity in designating proteins for degradation, but the mechanism underlying this specificity is not yet understood. Because the E3 ubiquitin ligases are responsible for selecting and recognizing those proteins to be degraded, there over 500 ubiquitin ligases in the human genome. The E3 ligases seem to have a tissue specificity; that is, there seem to be ubiquitin ligases specific for muscle protein degradation, specific for metabolic enzyme degradation, etc.
|
The Proteasome and Muscle Protein Turnover
It is estimated that 80 to 90% of all proteins in a cell are ultimately degraded via the proteasome pathway, and it is not surprising therefore that numerous reports have implicated the proteasome in muscle protein turnover. Many studies have reported that proteasome activity or expression of proteasome subunits or ubiquitinating enzymes is elevated during muscle atrophy (Lecker et al., 1999). Two recent studies found that the mRNA encoding 2 E3 ligases were significantly upregulated specifically during muscle atrophy induced by unweighting, denervation, or immobilization (rat muscle; Bodine et al., 2001) or during fasting, uremia, or streptozotocin-induced diabetes (mouse model; Gomes et al., 2001). The E3 ligases upregulated by unweighting, denervation, or immobilization were identified as MuRF1, for muscle ring finger, and MAFbx, for muscle atrophy F-box. The upregulation was specific for skeletal muscle and occurred in all 3 models of muscle wasting (Bodine et al., 2001). The gene whose expression was upregulated 7- to 9-fold by fasting was named atrogin (Gomes et al., 2001); sequence analysis showed that its sequence was 96% homologous to the rat MAFbx gene, suggesting that atrogin is the mouse homologue of rat MAFbx and that the 2 studies identified the same gene in 2 different species and 2 different models of muscle wasting. The ubiquitin ligases identified by Bodine and Gomes were expressed in skeletal and cardiac muscle but not in liver, brain, spleen, pancreas, placenta, or testis, indicating that they were E3 ligases specific for muscle protein degradation. Since these 2 reports, other studies have found that expression of MuRF1 and MAFbx E3 ligases were upregulated in a rat model of muscle wasting induced by sepsis.
It was found that MuRF1 binds to titin at the Z-disk and M-line regions of the titin molecule. As stated previously in this review, titin is one of those proteins that is rapidly degraded by the calpains (leaving a fragment of ~500 kDa and several smaller fragments) and whose degradation at the Z-disk and M-line levels would be needed for dissociation of the thick and thin filaments from the surface of the myofibril. Location of MuRF1 at these regions of titin would position it for immediate ubiquitination and proteasomal degradation of any titin polypeptide if it were released by the calpains.
It is clear that muscle protein degradation has an important role in enhancing both the rate and efficiency of skeletal muscle growth in domestic animals. The myofibrillar protein fraction constitutes over 50% of the protein in skeletal muscle, and it presents a unique challenge for turnover of its protein constituents. The proteasome undeniably is responsible for much of the intracellular protein degradation that occurs in muscle and other cells. It would be very surprising, therefore, if studies using proteasome inhibitors and estimating intracellular protein turnover by measuring release of free AA did not find that the proteasome had a major role in degradation of muscle proteins. It seems likely that the proteasome is directly responsible for degradation of the sarcoplasmic proteins that can be ubiquitinated and presented to the 26S proteasome. The myofibrillar proteins, however, cannot be directly degraded by the proteasome, because neither the myofibril itself (0.5 to 3.0 µm in diameter) nor thick and thin filaments (14 to 15 nm and 8 nm in diameter, respectively) could enter the central chamber of proteasome where the catalytic residues reside. The opening to this central chamber is ~1.2 to 1.5 nm, and in eukaryotic cells, it is blocked by the N-termini of some of the subunit polypeptides. Indeed, many studies have found that intact myofibrils are not degraded when they are incubated with a proteolytically active proteasome (Koohmaraie, 1992; Solomon and Goldberg, 1996). Solomon and Goldberg (1996) found that although the proteasome degraded myosin and actin, it had no effect on intact myofibrils, and they concluded that the rate-limiting step in degradation of myofibrillar proteins was their release from the myofibril. A subsequent study also found that the myosin heavy chain was degraded by the proteasome, but only after its disassembly from the myofibril, a process that was not associated with proteasome activity. Thus, the current state of knowledge on the mechanism(s) regulating metabolic turnover of myofibrillar proteins can be summarized as follows. Neither myofibrils nor thick and thin filaments can be degraded by the proteasome, because they cannot enter the central catalytic chamber of the proteasome, whether they are ubiquitinated or not. The calpains do not degrade proteins to AA, so measuring release of free AA to estimate rate of protein turnover cannot be used to determine whether the calpains are contributing to this turnover. The calpains do not degrade sarcoplasmic proteins (Tan et al., 1988; Smith and Dodd, 2007), so turnover of sarcoplasmic proteins is not caused by the calpains but is likely due to proteasomal activity. The myofibrillar proteins must be dissociated from the myofibril before they can be degraded downstream to AA by the proteasome and cellular peptidases. Does this dissociation occur by release of myofilaments from the surface of the myofibril (the ERM), by protein exchange, or by both these mechanisms? When are the myofibrillar proteins ubiquitinated? Before they are dissociated from the myofibril or after they are in the monomeric form? Or, are they ubiquitinated at all? Both the proteasome and the calpains are present in cells, including skeletal muscle cells, in large excess (Goll et al., 2003). Thus, it is unclear whether increased calpain or proteasome activity in in vitro assays would have any direct effect on the rate of myofibrillar protein turnover in vivo. Increased ubiquitination or a change in how activity of the calpains is regulated (e.g., calpastatin levels) may have greater significance to in vivo activity. Measuring mRNA levels for proteolytic enzymes does not necessarily reflect the activities of these enzymes (Wang et al., 1998). Less than 50% of the enzymes in cells have their activity regulated at the transcriptional level. Both the proteasome and the calpains are phosphorylated, and phosphorylation affects their activity. Measurements of protein levels and enzyme activities are more directly related to intracellular enzyme activity than message levels. It should be noted that a large majority of the studies relating the calpains and especially the proteasome to muscle protein degradation involved studies of muscle that was atrophying because of denervation, unweighting, or other treatment to induce rapid muscle protein turnover. It is possible that the mechanism used to turnover muscle proteins under these conditions differs from that used to turnover muscle proteins during muscle growth. Only the studies on callipyge sheep, administration of β-adrenergic agonists, and effects of the dominant negative and calpastatin expression in L6 myoblasts can be related to muscle growth. Should this be a concern?
This review has not discussed the kinetic considerations that are important to any analysis of myofibrillar protein turnover (Swartz, 1999). The rates at which myofibrillar proteins are released from myofibrils, whether it be via exchange or by calpain cleavage, will control their availability to the next downstream protease (likely the proteasome or possibly, lysosomal cathepsins in some instances). And, release of proteins from the myofibril will involve an equilibrium that is given by the rate of their association back onto the myofibril divided by the rate of their dissociation.
It may be useful to note that although intracellular protein turnover seems energetically inefficient, it is nevertheless an essential process for life. Intracellular protein turnover provides AA to cells between meals or during periods when no dietary sources of AA are available; it is required to eliminate those polypeptides that contain errors of translation or transcription and that are likely harmful to the cell if not removed (do the Alzheimer plaques fall in this category?); and it is needed to degrade those proteins/isoforms that are being displaced during development or in response to a physiological demand (e.g., changes in myosin isoforms). Thus, intracellular protein turnover has been maintained through evolution and selection, and knockout models that eliminate such turnover are invariably embryonically lethal.
In summary, it is surprising that over 30 yr after it was first proposed, the calpain-dissociation-downstream degradation of dissociated proteins hypothesis for turnover of myofibrillar proteins (Dayton et al., 1975) has never been critically examined, especially since it is clear that neither the calpains nor the proteasome can turn over the myofibrillar proteins alone. The myofibrillar proteins are the major protein fraction in skeletal muscle, and alterations in the rates at which they are turned over would have significant effects on the rate of skeletal muscle growth.
|
Footnotes |
|---|
2 Presented at the Nonruminant Nutrition symposium at the annual meeting of the American Society of Animal Science, San Antonio, TX, July 8 to 12, 2007.
3 Corresponding author: darrel.goll{at}arizona.edu
Received for publication July 3, 2007. Accepted for publication August 15, 2007.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONSKELETAL MUSCLE STRUCTURE AND...PROTEOLYTIC SYSTEMS IN MUSCLETHE CALPAIN SYSTEM AND...THE PROTEASOME AND SKELETAL...LITERATURE CITED |
|---|
-actinin/actin interaction in the Z-disk by using calpain. J. Biol. Chem. 266:8501–8510.-actinin. J. Biol. Chem. 251:6860–6870.-actinin in isolated rigor myofibrils. J. Muscle Res. Cell Motil. 20:457–467.[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
  J. M. Giger, P. W. Bodell, M. Zeng, K. M. Baldwin, and F. Haddad Rapid muscle atrophy response to unloading: pretranslational processes involving MHC and actin J Appl Physiol, October 1, 2009; 107(4): 1204 - 1212. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. McClung, A. R. Judge, E. E. Talbert, and S. K. Powers Calpain-1 is required for hydrogen peroxide-induced myotube atrophy Am J Physiol Cell Physiol, February 1, 2009; 296(2): C363 - C371. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. P. Walker, T.K. Rajendra, L. Saieva, J. L. Fuentes, L. Pellizzoni, and A. G. Matera SMN complex localizes to the sarcomeric Z-disc and is a proteolytic target of calpain Hum. Mol. Genet., November 1, 2008; 17(21): 3399 - 3410. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. E. Chibisa, G. N. Gozho, A. G. Van Kessel, A. A. Olkowski, and T. Mutsvangwa Effects of Peripartum Propylene Glycol Supplementation on Nitrogen Metabolism, Body Composition, and Gene Expression for the Major Protein Degradation Pathways in Skeletal Muscle in Dairy Cows J Dairy Sci, September 1, 2008; 91(9): 3512 - 3527. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. McClung, M. A. Whidden, A. N. Kavazis, D. J. Falk, K. C. DeRuisseau, and S. K. Powers Redox regulation of diaphragm proteolysis during mechanical ventilation Am J Physiol Regulatory Integrative Comp Physiol, May 1, 2008; 294(5): R1608 - R1617. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. G. Bergen Measuring in vivo intracellular protein degradation rates in animal systems J Anim Sci, April 1, 2008; 86(14_suppl): E3 - E12. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Z. Fan, S. W. Kim, T. J. Applegate, and M. Cervantes Nonruminant Nutrition symposium: Understanding protein synthesis and degradation and their pathway regulations J Anim Sci, April 1, 2008; 86(14_suppl): E1 - E2. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
