In the rat, denervation and hindlimb unloading are two commonly employed models used to study skeletal muscle atrophy. In these models, muscle atrophy is generally produced by a decrease in protein synthesis and an increase in protein degradation. The decrease in protein synthesis has been suggested to occur by an inhibition at the level of protein translation. To better characterize the regulation of protein translation, we investigated the changes that occur in various translation initiation and elongation factors. We demonstrated that both hindlimb unloading and denervation produce alterations in the phosphorylation and/or total amount of the 70-kDa ribosomal S6 kinase, eukaryotic initiation factor 2 α-subunit, and eukaryotic elongation factor 2. Our findings indicate that the regulation of these protein translation factors differs between the models of atrophy studied and between the muscles evaluated (e.g., soleus vs. extensor digitorum longus).
- elongation factor
- initiation factor
- protein kinase
- protein synthesis
hypodynamia is defined as reduced load-bearing or locomotor activity by skeletal muscle and generally results in muscle atrophy. Hypodynamia-induced atrophy has been shown to occur following a variety of situations, such as spaceflight (33), prolonged bed rest (18), and cast immobilization (6). One model commonly used to study hypodynamia is hindlimb unloading, in which the hindlimbs are suspended off the ground to remove the normal gravitational load on the muscles. As a result, the soleus muscle, which is composed primarily of slow-twitch fibers in the rat, undergoes substantial atrophy, while the primarily fast-twitch extensor digitorum longus (EDL) muscle remains relatively unaffected (3). The atrophy of the soleus has been partly attributed to a decrease in protein synthesis (34). In fact, rates of protein synthesis in the soleus may decrease as early as 6 h after unloading (34). The inhibition of protein synthesis most likely occurs at the translational level, since the synthesis of proteins such as β-myosin heavy chain decreases, while mRNA levels remain unchanged (34). To date, the mechanisms that regulate translation in this model remain largely undefined.
Denervation is another model of hypodynamia that has been used to study muscle atrophy. In this model, similar to hindlimb unloading, the soleus muscle undergoes greater atrophy than the EDL. Another similarity to the hindlimb unloading model is a rapid decrease in protein synthesis of the soleus muscle (8). Surprisingly, the rate of protein synthesis in the EDL shows a gradual increase and eventually exceeds basal levels by 7 days (8), an effect that has not been observed in the EDL during hindlimb unloading. It has been suggested that the regulation of protein synthesis rates during denervation also occurs at the level of translation (23). However, the mechanisms that regulate the changes in protein translation following denervation have not been determined.
The process of protein translation can be divided into three stages: initiation, elongation, and termination. Each of these stages has been shown to be regulated by numerous protein factors termed initiation, elongation, and release factors (22, 27). It has been well established that initiation and elongation are the steps of translation that are regulated with little regulation occurring during termination. More specifically, most studies to date suggest that initiation is the primary site of regulation for the majority of mRNAs in the cell (31).
There are two major points of control that can regulate the rate of initiation during translation. The first major control point is the binding of the initiator methionyl-tRNA to the 40S ribosomal subunit. This step is regulated by eukaryotic initiation factor 2 (eIF-2; a heterodimer composed of α-, β-, and γ-subunits), which mediates ribosomal binding of the methionyl-tRNA in a GTP-dependent manner (26). As a product of this initiation step, eIF-2 is released in its GDP-bound state. To return eIF-2 to its active GTP-bound state, the GDP must be recycled for another GTP in a reaction catalyzed by the guanine exchange factor eIF-2B (24). The recycling of GTP by eIF-2B can be inhibited by phosphorylation of eIF-2 on the α-subunit (29). Therefore, an increase in eIF-2α subunit phosphorylation can lead to an inhibition of translation initiation.
A second control point during initiation is the binding of the mRNA to the 43S ribosomal subunit. One of the proteins that may regulate this step is the ribosomal S6 kinase (p70S6k). This kinase appears to confer selective translation of mRNAs that contain a 5′ polypyrimidine tract as a common feature. The transcripts from theses mRNAs encode proteins that are generally involved in the translational apparatus, such as the ribosomal proteins (e.g., S6) (12) and elongation factors [e.g., eukaryotic elongation factor 2 (eEF-2)] (13). The translation of these mRNAs is dependent on the activity of p70S6k. The activity of p70S6k is regulated by changes in its phosphorylation state (4).
Rates of protein synthesis can also be regulated at the elongation phase of translation. One factor that regulates elongation is eEF-2. eEF-2 mediates the translocation step of elongation. Similar to what is known for eIF-2α, phosphorylation of eEF-2 results in inhibition of elongation by decreasing its affinity for the ribosome by 10 to 100 times (2).
The purpose of this study was to characterize changes in known translation factors that might regulate protein synthesis during hindlimb unloading and denervation models of muscle atrophy. In this study we demonstrate that initiation and elongation factors are being regulated in both models of atrophy. We show that, in general, the factors investigated (p70S6k, eIF-2α, and eEF-2) change their state of phosphorylation, and/or quantity, in a manner that is consistent with the changes that have previously been reported to occur in protein synthesis rates. However, the results demonstrate that during hypodynamia-induced atrophy, the mechanisms regulating the translation of proteins differ dramatically between the type of muscle and model studied.
MATERIALS AND METHODS
Rabbit polyclonal anti-p70S6k antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Mouse monoclonal anti-eIF-2α antibody was a generous gift provided by Dr. Leonard S. Jefferson (Pennsylvania State University). Rabbit polyclonal anti-phospho-Ser-51 eIF-2α antibody was purchased from Research Genetics (Huntsville, AL). Rabbit polyclonal anti-eEF-2 antibody and rabbit polyclonal anti-phospho-Thr-56 eEF-2 antibodies were generous gifts provided by Dr. Angus C. Narin (The Rockefeller University). Peroxidase-conjugated horse anti-mouse and goat anti-rabbit antibodies were purchased from Vector Laboratories (Burlingame, CA). Polyvinylidene difluoride (PVDF) membranes were purchased from Millipore (Bedford, MA). Enhanced chemiluminescence (ECL) detection reagent was purchased from Amersham Pharmacia Biotech (Amersham, UK). Re-probe buffer was purchased from Geno Tech (St. Louis, MO). DC protein assay kit was purchased from Bio-Rad (Hercules, CA).
Animal models and muscle processing.
All experimental procedures were approved by the University of Illinois at Chicago Animal Care Committee. Animals were housed individually and allowed free access to food and water throughout the experimental period. Female Wistar rats (Charles River Laboratories, Wilmington, MA), 3 mo of age, were randomly assigned to hindlimb unloading control (CNT), sham-operated denervation control (SHM), denervated (DNV), or hindlimb-unloaded (HU) groups. DNV and SHM animals were anesthetized with pentobarbital sodium (40 mg/kg). The muscles of the hindlimb were denervated by isolation and removal of ∼1 cm of the sciatic nerve immediately proximal to the division of the peroneal and tibial branches. SHM animals underwent identical surgeries; however, the sciatic nerve was left intact. Hindlimb unloading was performed by the methods described previously (25). All animals were allowed food and water ad libitum. After 12 h or 7 days of hindlimb unloading or denervation, the soleus and EDL muscles were removed, quickly weighed, and homogenized in a buffer containing 20 mM Tris (pH 7.5), 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM β-glycerophosphate, 1 mM Na3VO4, 1 μg/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride. The homogenate was then centrifuged at 15,000 g for 10 min at 4°C, and the supernatant was used for further analysis. The soleus and EDL muscles from SHM animals were collected 12 h postoperation and processed as described above. Protein concentration was determined by the DC protein assay (Bio-Rad Laboratories).
Quantification of p70S6k phosphorylation.
Tissue supernatant (5 μg protein) was analyzed by SDS-PAGE on a 7.5% acrylamide gel. After electrophoretic separation, proteins were transferred to a PVDF membrane. The membrane was incubated in 5% blotto (5% powdered milk in 1× Tris-buffered saline, 1% Tween 20) overnight at 4°C. After the overnight incubation, the membrane was probed with anti-p70S6k antibody (1:2,500) for 1 h, followed by anti-rabbit antibody (1:5,000) for 45 min. The blots were then developed using an ECL Western blotting kit (Amersham Pharmacia Biotech). p70S6k resolves into multiple bands after electrophoretic separation, with the slower migrating bands representing states of increased phosphorylation. The percent phosphorylation was therefore quantified as previously described (1). All densitometric measurements were carried out on an Alpha Imager 2200 (Alpha Innotech).
Quantification of phosphorylated and total eIF-2α.
Tissue supernatant (25 μg protein) was analyzed by SDS-PAGE on a 12.5% acrylamide gel and Western blotted as described inQuantification of p70S6kphosphorylation. The phosphorylated form of the 36-kDa protein was detected with an anti-phospho-Ser-51 eIF-2α antibody (1:1,250). The membrane was stripped with Re-Probe buffer for 30 min at room temperature, followed by overnight incubation in 5% blotto at 4°C. The total eIF-2α was detected with anti-eIF-2α antibody (1:700). Phosphorylated and total eIF-2α measurements were normalized for any loading differences by quantification of Coomassie blue-stained proteins in the gel that contained proteins in the 50- to 80-kDa range. For the denervated EDL, the relative percent phosphorylation was expressed as the ratio of amount of the phosphorylated form to the total amount of the protein.
Quantification of phosphorylated and total eEF-2.
Tissue supernatant (5 μg protein) was analyzed by SDS-PAGE on a 7.5% acrylamide gel and Western blotted as described in Quantification of p70S6k phosphorylation. The phosphorylated form of the 100-kDa protein was detected with an anti-phospho-Thr-56 eEF-2 antibody (1:500). The membrane was then stripped with Re-probe buffer for 30 min at room temperature, followed by overnight incubation in 5% blotto at 4°C. The total eEF-2 was detected with anti-eEF-2 antibody (1:250). Phosphorylated and total eEF-2 measurements were normalized for any loading differences by quantification of Coomassie blue-stained proteins from the 20- to 65-kDa range.
All data are expressed as means + SE. Unless otherwise noted, all HU samples were compared with CNT samples, and DNV samples were compared with SHM samples. Statistical analysis of the data was performed using analysis of variance (ANOVA). If differences were found, the Student-Newman-Keuls post hoc test was used to locate the source of the difference. Data were tested for normal distribution and homogeneity of variance before analysis was performed. All analyses were performed with SigmaStat statistical software (SPSS, Chicago, IL.).
Hindlimb unloading and denervation induce similar atrophy.
Hindlimb unloading and denervation produced similar effects on muscle weight-to-body weight ratio in the soleus muscles. Muscle weight-to-body weight ratio was not significantly altered after 12 h of hindlimb unloading or denervation. However, by 7 days, this ratio had decreased 29% from control values after hindlimb unloading and 35% after denervation (P < 0.05) (Fig.1 A). The same pattern was observed in the EDL, but the degree of atrophy was smaller. Hindlimb unloading induced a 9% decrease (not statistically significant:P = 0.19), while denervation produced a 12% reduction, in muscle weight-to-body weight ratio by 7 days (P < 0.05) (Fig. 1 B).
Hindlimb unloading and denervation have different effects on p70S6k phosphorylation.
In the soleus muscle, hindlimb unloading and denervation induced similar alterations in p70S6k phosphorylation. Muscles from HU animals showed a significant reduction in p70S6kphosphorylation by 12 h, and this effect was still evident at 7 days compared with muscles from CNT animals (P < 0.05). Muscles from DNV animals also showed a significant reduction in p70S6k phosphorylation at 12 h (P < 0.05) but not at 7 days compared with muscles from SHM animals (Fig.2). It should be noted that there was a trend for a decrease in the phosphorylation state of p70S6kin the soleus of SHM animals compared with CNT animals at 12 h; however, this was not statistically significant (P = 0.06). By 7 days after the sham operation, p70S6kphosphorylation was similar to basal levels (data not shown). This observation suggests that the sham operation may have produced a transient decrease in p70S6k phosphorylation in the soleus muscles. Therefore, it may be more appropriate to compare the p70S6k phosphorylation states from DNV animals to those of CNT animals. In this case, the p70S6k phosphorylation state is markedly reduced at both 12 h and 7 days postdenervation (P < 0.05).
In the EDL muscle, hindlimb unloading and denervation produced very different effects on p70S6k phosphorylation. Hindlimb unloading did not result in any significant alterations in the phosphorylation states at either 12 h or 7 days. However, in the EDL muscles from DNV animals, p70S6k phosphorylation was unchanged at 12 h but increased to 175% of that in SHM animals by 7 days (P < 0.05) (Fig.3). Unlike the effect seen in the soleus, the sham operation did not alter the p70S6k phosphorylation state in the EDL. This may be the result of the low basal phosphorylation state in the EDL muscles from CNT animals (27% phosphorylation) compared with the basal phosphorylation in the soleus muscle from CNT animals (64% phosphorylation).
Alterations in the phosphorylated and total amount of eIF-2α occur with denervation but not hindlimb unloading.
Hindlimb unloading and denervation had no effect on the phosphorylated or total amount of eIF-2α in soleus muscles at either 12 h or 7 days (Fig. 4 A–F). In addition, hindlimb unloading also had no effect on the phosphorylated or total amount of eIF-2α in the EDL muscles at either time points (Fig. 5, A,C, and E). However, the denervated EDL muscles showed a 22% reduction in the amount of phosphorylated eIF-2α (P < 0.05), and this decrease was accompanied by a 200% increase in the total amount of eIF-2α after 7 days of denervation (P < 0.05), an effect that was not observed at 12 h. Together, these changes produced a 78% reduction in the ratio of phosphorylated to total eIF-2α after 7 days of denervation (Fig. 5, B, D, F, and G).
Alterations in the phosphorylated and total amount of eEF-2 occur with hindlimb unloading and denervation.
In the soleus muscles, there was no change in the phosphorylated form or total amount of eEF-2 at 12 h after hindlimb unloading. However, after 7 days, the soleus muscles of HU animals exhibited a 50% decrease in the amount of eEF-2 in the phosphorylated form (P < 0.05), while total eEF-2 was not altered. In contrast, eEF-2 phosphorylation or amount did not change in the soleus muscles of DNV animals at either 12 h or 7 days (Fig.6, A–F).
In the EDL muscles, there was no effect on the phosphorylated or total amount of eEF-2 at either time point in the HU animals or at 12 h in the DNV animals (Fig. 7,A–F). However, at 7 days after denervation, the EDL muscles showed an increased amount of phosphorylated eEF-2 in addition to an increased total amount of eEF-2 (P < 0.05). The changes in both the amount of phosphorylated and total eEF-2 content resulted in no net change in the ratio of phosphorylated to total eEF-2 (Fig. 7, B, D, F, and G).
The effects of muscle unloading and denervation on protein synthesis have been known for many years (7, 34). Muscle atrophy associated with these models of hypodynamia is generally believed to occur through a decrease in protein synthesis with concomitant increases in protein degradation (7, 19). However, this is not always the case, because denervation-induced atrophy of the EDL is characterized by an increase in protein synthesis, implying an even greater contribution from degradation (8). The experiments described in this study focus on the control of protein synthesis through translation factors. Our findings demonstrate that regulation differs between models of hypodynamia as well as between the muscle types that are studied. In general, p70S6k, eIF-2α, and eEF-2 changed their phosphorylation state, and/or quantity (Table 1), in a direction that is consistent with the changes in protein synthesis that have been previously reported (8, 20, 34). These results provide the first insight into the molecular events controlling muscle protein synthesis during hypodynamia-induced atrophy.
Regulation of translation factors in the soleus after hindlimb unloading or denervation.
It has previously been suggested that the rapid decrease in protein synthesis observed in the soleus muscle after 18 h of hindlimb unloading is due to inhibition at the elongation phase of translation (17). This conclusion was based on polysome profiles, which showed a shift in mRNA from the light to heavy polysome pool (17). However, based on the analysis of translation factors, our data suggest that inhibition also occurs during the initiation of translation after hindlimb unloading. This conclusion is based on the decreased phosphorylation of p70S6k.
As mentioned previously, a decrease in p70S6kphosphorylation has been implicated in regulating the translation of mRNAs that contain a polypyrimidine tract in their 5′ untranslated region (11). A decrease in p70S6kphosphorylation was observed after only 12 h of hindlimb unloading, a time point at which the rate of protein synthesis has been shown to be inhibited (34). Additionally, the decrease in p70S6k phosphorylation was still evident after 7 days of hindlimb unloading, again a time point at which the rate of protein synthesis is inhibited (34). The temporal changes in p70S6k phosphorylation are consistent with the temporal changes in protein synthesis rates; therefore, this factor may be an important mediator in the inhibition of protein synthesis. Additionally, many of the mRNAs that are known to be translationally regulated by p70S6k encode proteins such as ribosomal subunits and elongation factors (13). Therefore, it seems possible that a decrease in p70S6k phosphorylation would lead to a decrease in global protein synthesis via a decrease in ribosomal protein components of the translational machinery. This concept of decreased synthetic machinery is also supported by the rapid changes seen in total RNA content after hindlimb unloading and denervation in the soleus muscle (21, 32).
The changes we observed in p70S6k phosphorylation suggest that inhibition is occurring at initiation of translation, but this may not be the case for all mRNAs. A more general inhibition of initiation can occur through a decrease in eIF-2B activity. One of the mechanisms that can decrease eIF-2B activity is an increase in the phosphorylation of eIF-2α, an effect that was not observed with either hindlimb unloading or denervation. However, other mechanisms have been shown to regulate eIF-2B activity; therefore, it remains to be determined whether regulation of eIF-2B activity is involved in the inhibition of translation initiation in these models.
Although our findings suggest that inhibition is occurring at the initiation phase of translation, it does not exclude the possibility that the elongation phase of translation is also being inhibited as suggested by Ku and Thomason (17). However, if elongation is being inhibited, it is not a result of changes in eEF-2 phosphorylation as was determined in this study. In this study we observed a decrease in eEF-2 phosphorylation in the soleus after 7 days of hindlimb unloading. A decrease in eEF-2 phosphorylation has been shown to be associated with increases in the rate of elongation (30). Thus the observation that eEF-2 phosphorylation is decreased in the soleus muscle after 7 days of hindlimb unloading is perplexing. One explanation for this curious change in eEF-2 is that it may be linked to the possible decrease in total protein synthetic machinery mentioned earlier. If the muscle cell senses a loss in total ribosomal capacity, then factors regulating elongation maybe enhanced to facilitate synthesis. However, more studies are clearly needed in this area to elucidate the complex interactions of factors regulating different aspects of translation.
After denervation, the associated muscle atrophy and inhibited protein synthesis has been suggested to result from a decrease in the initiation of translation. This conclusion was based on polysome profiles, which showed a decrease in the polysome-to-monosome ratio, a criterion that was not assessed after hindlimb unloading (16,23). The decreases in p70S6k phosphorylation at the early stage of denervation are consistent with the idea that translation initiation was inhibited. Additionally, like hindlimb unloading, the temporal changes in p70S6k phosphorylation are consistent with the temporal changes in protein synthesis rates (8).
In the soleus muscle, the decreases in p70S6kphosphorylation do not appear to be due to the loss of innervation since both models of hypodynamia produced this change. Therefore, the decreased p70S6k phosphorylation is likely the result of a loss of load and/or stress on the muscle. There are numerous upstream factors that could potentially cause the decrease in p70S6kphosphorylation. One of these upstream factors is the mammalian target of rapamycin (mTOR) (13). Some of the pathways that have been suggested to signal through mTOR include the phosphatidylinositol 3-kinase pathway (14), protein kinase C pathway (9), and amino acids (15). All of these signaling pathways are responsive to the stretch/stress in different cell types (5, 10, 28) and could contribute to the decrease in p70S6k phosphorylation by responding to a decrease in stretch/stress on the muscle.
Regulation of translation factors in the EDL after hindlimb unloading or denervation.
In our study, the EDL muscle does not atrophy significantly after hindlimb unloading. Consistent with this lack of EDL atrophy were unchanged levels of p70S6k phosphorylation. In contrast, there was a significant atrophy of the EDL after denervation, but surprisingly, at 7 days there was an increase in p70S6kphosphorylation levels. The increase in p70S6kphosphorylation after 7 days of denervation is consistent with an increase in the rate of protein synthesis that has been reported to occur at this time point (8). While unexpected, the EDL-specific increase in p70S6k phosphorylation at 7 days could be due to an increased incidence of stretch as a consequence of the rat maintaining its denervated limb in a plantar-flexed position (8). In support of this concept we observed a similar increase in p70S6k phosphorylation in the tibialis anterior muscle after 7 days of denervation, while no change was seen in the denervated plantaris muscle, a fast fiber-type muscle in the posterior compartment of the hindlimb (data not shown).
An increase in p70S6k phosphorylation is not the only factor that could enhance the rate of protein synthesis in the denervated EDL muscle. In addition to the changes in p70S6kphosphorylation, the quantity of eEF-2 and eIF-2α protein increased. An increase in the total amount of these factors in the unphosphorylated state could enhance the rate of protein synthesis if the availability of these factors was previously rate limiting for protein synthesis. The mechanism for an increase in the total amount of these factors is unknown at this time. However, it is interesting to note that the eEF-2 mRNA contains a 5′ polypyrimidine tract, which raises the possibility that it is translationally regulated by p70S6k (11). The increase in p70S6k phosphorylation at 7 days is consistent with the increase in eEF-2 protein, which suggests that the increase in eEF-2 was due to translational regulation. Despite the changes in translation factors that could promote an increase in the rate of protein synthesis, these changes are not sufficient to prevent muscle atrophy. This suggests that protein degradation plays a critical role in the development of atrophy in the EDL.
In summary, this study demonstrates that the regulation of translation factors differs dramatically between the model of atrophy and the type of muscle studied. Understanding how these alterations are mediated by upstream and downstream factors will ultimately lead to a better understanding of how the stress and/or innervation on the muscle signal a change in the rate of protein synthesis. It will be of particular interest to determine why signaling in the denervated EDL differs from that of the denervated soleus. Understanding these differences and their associated mechanisms could ultimately lead to the development of countermeasures of hypodynamia-induced muscle atrophy.
We thank Joshua Lang and Drs. Shann Kim and Mark Fedele for critical review of the manuscript.
This work was supported by National Institute of Arthritis and Musculoskeletal and Skin Diseases Grants AR-45617 (to K. A. Esser) and AR-41705 (to S. C. Kandarian).
Address for reprint requests and other correspondence: K. Esser, School of Kinesiology, 901 W. Roosevelt, Chicago, IL 60608 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2001 the American Physiological Society