The role of the proteasome in the regulation of cellular levels of the transcription factor CCAAT/enhancer-binding protein β (C/EBPβ) is poorly understood. We tested the hypothesis that C/EBPβ levels in cultured myotubes are regulated, at least in part, by proteasome activity. Treatment of cultured L6 myotubes, a rat skeletal muscle cell line, with the specific proteasome inhibitor β-lactone resulted in increased nuclear levels of C/EBPβ as determined by Western blotting and immunofluorescent detection. This effect of β-lactone reflected inhibited degradation of C/EBPβ. Surprisingly, the increased C/EBPβ levels in β-lactone-treated myotubes did not result in increased DNA-binding activity. In additional experiments, treatment of the myotubes with β-lactone resulted in increased nuclear levels of growth arrest DNA damage/C/EBP homologous protein (Gadd153/CHOP), a dominant-negative member of the C/EBP family that can form heterodimers with other members of the C/EBP family and block DNA binding. Coimmunoprecipitation and immunofluorescent detection provided evidence that C/EBPβ and Gadd153/CHOP interacted and colocalized in the nuclei of the β-lactone-treated myotubes. When Gadd153/CHOP expression was downregulated by transfection of myotubes with siRNA targeting Gadd153/CHOP, C/EBPβ DNA-binding activity was restored in β-lactone-treated myotubes. The results suggest that C/EBPβ is degraded by a proteasome-dependent mechanism in skeletal muscle cells and that Gadd153/CHOP can interact with C/EBPβ and block its DNA-binding activity. The observations are important because they increase the understanding of the complex regulation of the expression and activity of C/EBPβ in skeletal muscle.
- CCAAT/enhancer-binding protein β
- skeletal muscle
- gene transcription
the ccaat/enhancer-binding proteins (C/EBPs) are transcription factors that participate in the regulation of multiple important cell functions during normal and pathophysiological conditions (19, 25, 29). The C/EBP family consists of six members, i.e., C/EBPα, -β, -γ, -δ, -ε and growth arrest DNA damage/C/EBP homologous protein (Gadd153/CHOP). The C/EBP family members bind to DNA and regulate gene transcription after forming homodimers or heterodimers with other members in the C/EBP family or unrelated transcription factors (19, 25). The expression and activity of C/EBP transcription factors are tissue and cell specific, and their role in the regulation of gene transcription varies in different physiological and pathological conditions. Previous studies suggest that the expression and activity of C/EBPβ may be particularly important for gene regulation during inflammation (24). In recent experiments, we found evidence that C/EBPβ is involved in sepsis- and glucocorticoid-induced muscle wasting (23, 38) and in the inflammatory response in enterocytes (14).
The transcriptional activity of C/EBPβ may be influenced by posttranslational modifications, including phosphorylation (4) and acetylation (36), as well as the formation of homodimers or heterodimers with other transcription factors (25). An additional important factor that can influence the activity of C/EBPβ is its cellular abundance. Similar to other C/EBP family members (7, 30, 31), C/EBPβ has a rapid turnover, and, in recent experiments, we found that the half-life of C/EBPβ in cultured myotubes was ∼1–2 h (34). The rapid turnover of C/EBPβ is consistent with the tight control typically seen for many regulatory proteins and that is often accomplished by changes in the degradation of the proteins, although other mechanisms may be involved as well.
Different proteolytic mechanisms regulate the degradation and cellular levels of transcription factors. The ubiquitin-proteasome proteolytic pathway plays an important role in the degradation of many cellular regulatory proteins, including transcription factors (12, 21). Recent studies suggest that the degradation of some of the C/EBP family transcription factors, including C/EBPα, -γ, and -δ and Gadd153/CHOP, is proteasome dependent (7, 11, 30). In contrast, the role of the proteasome system in the regulation of C/EBPβ levels is controversial. Thus, whereas C/EBPβ expression and activity were not influenced by proteasome activity in cultured keratinocytes (30) and human melanoma cells (11), results from experiments in other cell types suggested that C/EBPβ expression was regulated by the proteasome. For example, Yamamoto et al. (37) reported that the homolog of C/EBPβ in the sea slug Aplysia californica, ApC/EBP, was degraded by the proteasome. In a recent study from our laboratory, treatment of cultured human enterocytes with the proteasome inhibitors MG-132 and lactacystin resulted in increased nuclear levels of C/EBPβ (15). Interestingly, this response to the proteasome inhibitors was, at least in part, caused by induction of the heat shock response. Thus it is possible that the role of the proteasome in the regulation of C/EBPβ expression and activity is cell specific and may reflect different mechanisms.
In recent experiments, we found that the degradation of C/EBPβ in cultured myotubes was calpain dependent (34). This finding does not rule out that other mechanisms may be involved in the regulation of C/EBPβ expression in muscle cells. Indeed, studies in other cell types suggest that multiple proteolytic pathways may contribute to the degradation of transcription factors (27). The role of the proteasome in the regulation of C/EBPβ levels in muscle cells is not known. Here, we tested the hypothesis that the degradation of C/EBPβ in cultured myotubes is proteasome dependent.
MATERIALS AND METHODS
L6 rat skeletal muscle cells were purchased from American Type Culture Collection (Manassas, VA). DMEM was purchased from Mediatech (Herndon, VA). Clasto-lactacystin β-lactone was from Boston Biochem (Cambridge, MA). Calpeptin was from Calbiochem (San Diego, CA). Cycloheximide and quercetin were from Sigma (St. Louis, MO). Protein A/G PLUS-Agarose, oligonucleotide encoding the sequence for C/EBP binding, and a corresponding mutant oligonucleotide were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). NE-PER nuclear and cytoplasmic extraction reagents were from Pierce Biotechnology (Rockford, IL). In Situ Cell Death Detection Kit, TMR red, and DNase I recombinant, grade I were from Roche (Mannheim, Germany). [γ-32P]ATP and the Western Lightning Kit for enhanced chemiluminescence were purchased from Perkin-Elmer Life Sciences (Boston, MA). Poly(dI-dC) was from Boehringer Mannheim (Indianapolis, IN). Alexa Fluor 594 goat anti-mouse IgG and Alexa Fluor 488 goat anti-mouse IgG (2 mg/ml; Molecular Probes) as well as Oligofectamine Transfection Reagent were purchased from Invitrogen (Grand Island, NY). VECTASHIELD Mounting Medium [with 4′,6-diamidino-2-phenylindole (DAPI)] was from Vector Laboratories (Burlingame, CA). Kodak X-Omat blue, X-Omat AR, and BioMax MR films were from Eastman Kodak (Rochester, NY). Minigels (10 and 12%) were purchased from Bio-Rad Laboratories (Hercules, CA). Coomassie plus protein assay reagent was from Pierce. Tri-Reagent was from Molecular Research Center (Cincinnati, OH). The TaqMan One Step PCR Master Mix Reagents Kit was from Roche Molecular Systems (Foster City, CA).
Plasmids and antibodies.
siGENOME SMARTpool reagent (Rat DDIT3) and siCONTROL Non-Targeting small interfering RNA (siRNA) were purchased from Dharmacon (Chicago, IL). Mouse monoclonal antibodies to C/EBPβ (H-7), Gadd153 (B-3), and 70-kDa heat shock protein (HSP 70; 3A3), rabbit polyclonal antibodies to C/EBPβ (C-19), Gadd153 (F-168) and Oct-1 (C-12), and normal mouse IgG and normal rabbit IgG were from Santa Cruz Biotechnology. Monoclonal antibody to α-tubulin was from Sigma.
L6 rat skeletal muscle cells were thawed and maintained by repeated subculturing at low density in 162-cm2 culture flasks and were used between passages 2 and 8. Cells were grown in DMEM supplemented with 10% FBS, 100 U/ml penicillin, and 100 μg/ml streptomycin in a 10% CO2 atmosphere at 37°C. When cells reached ∼80% confluence, they were removed by trypsinization (0.25% trypsin in PBS) and seeded in 10-cm culture dishes or six-well culture plates. The cells were grown in the presence of 10% FBS until they reached ∼80% confluence at which time the medium was replaced with DMEM containing 2% FBS for induction of myotube differentiation. Cultures were used for experiments ∼5 days later, when myotube formation was observed. Myotubes were treated with different concentrations of β-lactone, as indicated in results. Untreated myotubes served as controls. In separate experiments, 10 μg/ml of cycloheximide or 5 μg/ml of actinomycin D were added to the culture medium, either alone or in combination with 5 μM β-lactone. After the different treatments, myotubes were harvested for determination of C/EBPβ, Gadd153/CHOP, and HSP 70 protein levels or DNA-binding activity, as described below.
Preparation of nuclear and cytoplasmic extracts and whole cell lysates.
Nuclear and cytoplasmic fractions were prepared with Pierce NE-PER nuclear and cytoplasmic reagents according to the manufacturer's protocol (Pierce Biotechnology). NE-PER cytoplasmic extraction reagents (CER I and CER II) and nuclear extraction reagent enable the stepwise separation of cytoplasmic and nuclear extracts from cultured cells. In other experiments, whole cell lysates were prepared by lysing myotubes in three volumes of lysis buffer consisting of 50 mM Tris·HCl, pH 7.4, 300 mM NaCl, 5 mM EDTA, 1% Nonidet P-40, and protease inhibitor cocktail. Protein concentrations in the nuclear extracts and whole cell lysates were determined by using the Pierce Coomassie blue R-250 method (Pierce Biotechnology) with BSA as standard.
Whole cell lysates or nuclear and cytoplasmic extracts were incubated at 100°C for 10 min, and aliquots (50 μg protein) were loaded on 10 or 12% Bio-Rad polyacrylamide ready gels. The separated proteins were transferred electrophoretically using semidry transfer methodology to nitrocellulose membranes (Millipore). The membranes were treated with blocking buffer (5% nonfat dry milk, 20 mM Tris·HCl, pH 7.5, 150 mM NaCl, and 0.05% Tween 20) for 1 h at room temperature. The membranes were then incubated overnight with the appropriate primary antibody, washed with 20 mM Tris·HCl, pH 7.5, 150 mM NaCl, and 0.05% Tween 20 (TTBS) three times, and incubated for 1 h with the appropriate peroxidase-conjugated secondary antibody. α-Tubulin was determined as loading control for the cytoplasmic fraction and whole cell lysates. The transcription factor Oct-1 (28) was determined as loading control for the nuclear fraction. Membranes were then washed extensively in TTBS. Immunoreactive protein bands were detected by using the Western Lightning Kit for enhanced chemiluminescence and exposure on Kodak X-Omat blue or X-Omat AR film.
Coimmunoprecipitation was used to detect protein-protein interaction between C/EBPβ and Gadd153/CHOP. Aliquots (50 μg protein) of whole cell lysates were immunoprecipitated with a rabbit polyclonal anti-C/EBPβ (C-19) antibody or control IgG (nonimmune rabbit IgG) as well as with mouse monoclonal anti-Gadd153 antibody or control IgG (nonimmune mouse IgG) for 1 h at 4°C. After 1 h of mixing at 4°C, 25 μl of packed Protein A/G PLUS-Agarose beads were added, and the samples were mixed at 4°C overnight. The beads were washed five times with 1 ml of buffer containing 50 mM Tris·HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, and 1% Nonidet P-40, and bound proteins were eluted with 2× SDS sample buffer at 95°C for 5 min and were separated by 12% Bio-Rad polyacrylamide ready gels. Western blotting for C/EBPβ or Gadd153/CHOP was performed as described above.
ATP and lactate dehydrogenase measurements.
ATP was measured in control and β-lactone-treated myotubes using the commercially available CellTiter-Glo Luminescent Cell Viability Assay kit from Promega (Madison, WI). Lactate dehydrogenase (LDH) release from myotubes was measured using the CytoTox-ONE Homogeneous Membrane Integrity Assay kit from Promega.
Terminal deoxynucleotidyltransferase-mediated dUTP nick end-labeling assay.
Apoptotic cell death was examined by terminal deoxynucleotidyltransferase-mediated dUTP nick end-labeling (TUNEL) assay and fluorescence microscopy (Nikon Eclipse TE300; Nikon, Avon, MA). Control (untreated) myotubes and myotubes treated with 5 μM β-lactone for 6 h were fixed with 4% paraformaldehyde in PBS, pH 7.4, for 1 h at room temperature, rinsed with PBS, permeabilized with 0.1% Triton X-100 in 0.1% sodium citrate for 2 min on ice, and washed two times with PBS. Staining was performed by incubating myotubes for 1 h at 37°C in a humidified atmosphere in the dark in 100 μl of TUNEL reaction mixture (In Situ Cell Death Detection Kit, TMR Red, No. 12 156 792 001; Roche). Positive controls were obtained by treating permeabilized myotubes with DNase I, grade I (1,000 U/ml in 50 mM Tris·HCl, pH 7.5, and 1 mg/ml BSA), for 10 min at room temperature to induce DNA strand breaks before the staining procedure. Negative controls were performed by incubating fixed and permeabilized myotubes in 100 μl/well of Label Solution (without terminal transferase) instead of TUNEL reaction mixture (data not shown).
To reduce the expression of Gadd153/CHOP, myotubes were transfected for 48 h with Gadd153/CHOP siRNA (siGENOME SMARTpool reagent; rat DDIT3), siCONTROL Nontargeting siRNA, or were mock transfected (treated with transfection reagents only without oligonucleotide). siRNA was diluted to 1.08 μM in serum-free DMEM, and oligofectamine was diluted 1:5 in serum-free DMEM. These mixtures were incubated at room temperature for 5 min. The diluted lipid reagent was then added to the siRNA mixture and incubated a further 15 min at room temperature. The cells were rinsed one time with serum-free DMEM and bathed in serum-free DMEM. Lipid-RNA complexes were applied to the myotubes, and an equivalent amount of lipid reagent was added to mock-transfected myotubes. After transfection, myotubes were treated with β-lactone for 6 h or were untreated, whereafter Gadd153/CHOP levels were determined by Western blotting and C/EBPβ DNA binding was determined by electrophoretic mobility shift assay (EMSA) with supershift analysis.
C/EBP gelshift oligonucleotide (5′-TGC AGA TTG CGC AAT CTG CA-3′; Santa Cruz Biotechnology) was end labeled with [γ-32P]ATP using T4 polynucleotide kinase (GIBCO-BRL). End-labeled probe was purified from unincorporated [γ-32P]ATP using a purification column (Bio-Rad Laboratories) and recovered in Tris-EDTA buffer, pH 7.4. Labeled DNA probe was incubated with 3 μg of nuclear protein extracts (20 min, room temperature) in a buffer containing 12% glycerol (vol/vol), 12 mM HEPES, pH 7.9, 4 mM Tris·HCl, pH 7.9, 1 mM EDTA, 1 mM dithiothreitol, 25 mM KCl, 5 mM MgCl2, and 0.04 μg/μl poly(dI-dC). An excess (15×) of nonradiolabeled wild-type or mutant oligonucleotide (5′-TGC AGA GAC TAG TCT CTG CA; nucleotide substitutions underlined) was added to test the specificity of the EMSA. Supershift analysis was performed by adding 2 μl of mouse monoclonal antibody to C/EBPβ, and the mixture was incubated at room temperature for 30 min. The resulting protein-DNA complexes were separated on a 7% polyacrylamide gel at 200 volts for 5 h. The gel was dried under vacuum for 2 h at 80°C, and protein-DNA complexes were visualized by autoradiography.
Localization and colocalization of C/EBPβ and Gadd153/CHOP.
L6 myotubes were cultured on German glass coverslips in six-well plates and treated with 1 or 5 μM β-lactone for 6 h before fixation in 3.5% paraformaldehyde for 15 min. The fixed myotubes were permeabilized with 0.2% Triton X-100 for 10 min and incubated with mouse monoclonal antibody to C/EBPβ (H-7) or mouse monoclonal antibody to Gadd153 (B-3) for 1 h at room temperature. After three washes in PBS containing 0.1% Tween, the coverslips were incubated with Alexa 594-conjugated goat anti-mouse secondary antibody for 1 h at room temperature. The coverslips were then washed three times in PBS containing 0.1% Tween and mounted on glass slides using DAPI mounting solution. Images were obtained using a Zeiss Laser Scanning Confocal Microscope 510 (Thornwood, NY) with ×40 objective. Background nonspecific fluorescence was determined by the same protocol in the absence of primary antibodies. All images were collected under identical settings.
To examine colocalization of C/EBPβ and Gadd153/CHOP, the fixed L6 myotubes were permeabilized with 0.2% Triton X-100 and then incubated for 1 h with mouse monoclonal antibody to C/EBPβ (H-7) and rabbit polyclonal antibody to Gadd153 (F-168) at room temperature and washed three times in PBS. Fluorescein Alexa 488-conjugated goat anti-mouse and Alexa 594-conjugated goat anti-rabbit secondary antibodies were added for 1 h and then rinsed with three washes of PBS. Cover slips were mounted with DAPI mounting solution. Immunofluorescence was detected using a Zeiss Laser Scanning Confocal Microscope 510 with ×40 objective. All images were collected under identical settings, and their respective backgrounds among the various conditions were tested in each experiment.
Results are presented as means ± SE. Statistical analysis was performed by using Student's t-test or ANOVA as appropriate. Experiments were repeated at least three times to provide evidence of reproducibility.
Several isoforms of C/EBPβ have been described, including a full-length 38-kDa isoform [liver-enriched transcriptional activating protein (LAP)], a 35-kDa isoform of LAP, a 21-kDa liver-enriched transcriptional inhibitory protein (LIP), and a 14-kDa isoform (1, 8, 35). Although the different C/EBPβ isoforms are commonly thought to arise from alternative translational start sites in the coding sequence (8), some studies suggest that the truncated 21- and 14-kDa isoforms of C/EBPβ may be the result of artifactual proteolysis during tissue preparation (2, 3). Other reports suggest that truncated C/EBPβ isoforms may be generated by both alternative translation and proteolytic cleavage of full-length C/EBPβ (35). In our experiments, the 35-kDa LAP was the predominant isoform in the cultured myotubes, whereas the expression of the 38-kDa full-length LAP and the 21-kDa LIP was lower and not detectable in some experiments. The 14-kDa isoform of C/EBPβ was not detectable in any of our experiments. In the present study, when a certain isoform is not shown in Figs. 1–9, it reflects the fact that it was not detected in that experiment.
When cultured L6 myotubes were treated for 6 h with the specific proteasome inhibitor β-lactone (18), cellular levels of C/EBPβ increased in a concentration-dependent manner (Fig. 1A). The maximal effect was noticed at a β-lactone concentration of 5 μM, and, at this concentration, the levels of all three C/EBPβ isoforms expressed in the myotubes were increased. When myotubes were treated with 5 μM β-lactone, C/EBPβ levels increased after 3 h and were maximally increased after 6 h (Fig. 1B). Treatment of the myotubes with 5 μM β-lactone for 6 h did not induce appreciable cell injury. Thus there was no evidence of cell rounding or detachment (data not shown), and myotube ATP levels and LDH release were unaffected (Fig. 1, C and D). In addition, TUNEL assay showed that the treatment did not result in apoptotic cell death, and phase-contrast microscopy revealed no apparent changes in myotube morphology (Fig. 2). In most subsequent experiments, myotubes were treated with 5 μM β-lactone for 6 h.
Increased C/EBPβ levels in myotubes treated with β-lactone are consistent with proteasome-dependent degradation of C/EBPβ, although other mechanisms could be involved as well. To examine the influence of β-lactone on the degradation of C/EBPβ, the decrease in C/EBPβ levels was determined in myotubes cultured in the presence of cycloheximide during 24 h. Because protein synthesis was blocked, the decrease in C/EBPβ levels reflected degradation of the protein. To increase the accuracy of this experiment, myotubes were first treated with 1 μM dexamethasone for 12 h to increase C/EBPβ levels in the myotubes (38). This allowed us to examine the degradation of C/EBPβ in myotubes that contained high levels of the protein. Using this experimental approach, C/EBPβ levels declined relatively rapidly in myotubes incubated in the absence of β-lactone (Fig. 3A). When C/EBPβ levels were quantified by densitometry (including both isoforms expressed in this experiment), results were consistent with a half-life of C/EBPβ of ∼2–4 h (Fig. 3B). Treatment of the myotubes with β-lactone stabilized C/EBPβ and prolonged the half-life to almost 24 h, suggesting that the proteasome inhibitor reduced the degradation of C/EBPβ.
Although the results reported here suggest that the increased C/EBPβ levels noticed in β-lactone-treated myotubes reflected inhibited degradation of the transcription factor, increased expression of C/EBPβ may also reflect increased synthesis of the protein. To test the potential role of protein synthesis in the β-lactone-induced increase in C/EBPβ levels, the effect of the proteasome inhibitor was tested in myotubes incubated in the absence or presence of cycloheximide. In this experiment, basal levels of C/EBPβ were reduced after 6 h treatment with cycloheximide (as expected), but even in these myotubes, an increase in C/EBPβ levels was noticed after treatment with β-lactone (Fig. 4A), suggesting that the β-lactone-induced increase in C/EBPβ levels did not reflect stimulated synthesis. Additional experiments supported this interpretation. Thus treatment of the myotubes with the transcriptional inhibitor actinomycin D reduced basal levels of C/EBPβ but did not prevent the β-lactone-induced increase in C/EBPβ levels (Fig. 4B). Somewhat unexpectedly, treatment of the myotubes with actinomycin D resulted in a more pronounced decrease in basal C/EBPβ levels than treatment with cycloheximide (compare Fig. 4, A and B). To test whether this effect of actinomycin D reflected cell toxicity, we determined the effect of actinomycin D on cell viability. Determined from LDH release, cell survival was not affected by treatment with 5 μg/ml of actinomycin D for 6 h (Fig. 4C).
One consequence of treatment with proteasome inhibitors is induction of the heat shock response (5). In a recent study, we found that treatment of cultured human intestinal epithelial cells with the proteasome inhibitor MG-132 resulted in increased expression of heat shock protein 72 (HSP 72; see Ref. 15). In the same experiments, treatment of the cells with MG-132 increased C/EBPβ levels, and this effect of the proteasome inhibitor was reduced by quercetin, a substance known to block induction of the heat shock response (20). Those results suggested that the increased levels of C/EBPβ in enterocytes treated with proteasome inhibitor were, at least in part, regulated by heat shock proteins. In the present study, treatment of the cultured myotubes with β-lactone for 6 h did not increase HSP 70 levels (Fig. 4D). Because this finding does not rule out the possibility that other heat shock proteins were increased (16) and may have contributed to the increased C/EBPβ levels, we treated myotubes with quercetin. Treatment of the myotubes with 100 μM quercetin, a concentration that effectively inhibited the heat shock response in other cell types treated with proteasome inhibitors (15), did not prevent the β-lactone-induced increase in C/EBPβ levels (Fig. 4E), suggesting that the increased C/EBPβ expression noticed in myotubes treated with β-lactone under the present experimental conditions was not regulated by the heat shock response. In fact, densitometric analysis of multiple experiments in which myotubes were treated with quercetin showed that C/EBPβ expression was actually increased by ∼15% in β-lactone-treated myotyubes when quecetin was added to the treatment (Fig. 4F). This result further supports the interpretation that the β-lactone-induced increase in C/EBPβ levels was not secondary to induction of the heat shock response. These results, however, do not rule out the possibility that the heat shock response may be induced and play a role in C/EBPβ expression after more prolonged (>6 h) treatment of myotubes with β-lactone.
It should be noted that the experiments described above were performed in whole cell lysates. When C/EBPβ expression was examined separately in the nuclear and cytoplasmic fractions, C/EBPβ was detected almost exclusively in the nuclear fraction both in control and β-lactone-treated myotubes (Fig. 5A). The separation of the nuclear and cytoplasmic fractions was confirmed by the expression of the transcription factor Oct-1 (28) in the nuclear fraction and α-tubulin in the cytoplasmic fraction. The nuclear localization and the effect of β-lactone treatment were also confirmed by immunofluorescent detection (Fig. 5B).
Having shown that treatment of myotubes with β-lactone resulted in increased nuclear expression of C/EBPβ, we next examined the functional aspects of the observations and hypothesized that C/EBPβ DNA-binding activity would be increased in β-lactone-treated myotubes. Surprisingly, EMSA with supershift analysis using an anti-C/EBPβ antibody showed that C/EBPβ DNA-binding activity was not increased in the β-lactone-treated myotubes (Fig. 6A, lane 3). To make certain that this lack of C/EBPβ DNA binding did not reflect a methodological error, we determined the effect of the calpain inhibitor calpeptin, a treatment that we found recently to increase C/EBPβ expression and DNA-binding activity in cultured myotubes (34). Treatment of the myotubes with 50 μM calpeptin for 6 h resulted in increased C/EBPβ DNA-binding activity (Fig. 6A, lane 2). Interestingly, the increased C/EBPβ DNA binding noticed in calpeptin-treated myotubes was blunted by β-lactone (Fig. 6A, lanes 2 and 4). Of note was the finding that the lower bands in the EMSA were not substantially weaker when the supershifted band appeared (see Fig. 6A, lanes 2 and 4). Although we do not have a definitive explanation for this finding at present, it is possible that the lack of a substantial decrease of the lower band reflects the fact that the C/EBPβ antibody was added to the reaction in all four lanes in Fig. 6A. Hence, it is possible that there was a weak supershift present under basal conditions and that this supershift was too weak to generate a clear supershifted band but strong enough to reduce the lower band somewhat, making it more difficult to detect a difference in the lower band in Fig. 6A, lanes 2 and 4.
As expected, the cellular levels of C/EBPβ were increased in myotubes treated with β-lactone or calpeptin (Fig. 6B). Interestingly, the effects of β-lactone and calpeptin on C/EBPβ levels were additive, supporting the concept that the regulation of C/EBPβ degradation is complex and may reflect the activity of multiple degradation mechanisms. It was interesting to note that C/EBPβ DNA-binding activity was only increased to a small degree after the combined treatment with β-lactone and calpeptin (Fig. 6A, lane 4) despite a substantial increase in C/EBPβ levels (Fig. 6B, lane 4).
The lack of increased C/EBPβ DNA binding noticed here in β-lactone-treated myotubes (despite the increased levels of the transcription factor) may reflect binding of C/EBPβ to an inhibitory protein also increased by the β-lactone treatment. The C/EBP family members form complexes with other transcription factors, including other members of the C/EBP family. When Gadd153/CHOP interacts with other members of the C/EBP family, DNA binding is inhibited (6, 29). Because experiments in other cell types showed that treatment with proteasome inhibitors results in increased expression of Gadd153/CHOP (11), it is possible that, in the present experiments, treatment of the myotubes increased the expression of the inhibitory Gadd153/CHOP, in addition to C/EBPβ, and that Gadd153/CHOP interacted with C/EBPβ and inhibited its DNA-binding activity.
To examine the potential role of Gadd153/CHOP, we first determined the effect of β-lactone on myotube Gadd153/CHOP levels. This was important because, although previous studies suggest that Gadd153/CHOP is degraded by a proteasome-dependent mechanism in other cell types (11), the effects of proteasome inhibitors on Gadd153/CHOP levels in muscle cells have not been reported. Here we found that treatment of cultured L6 myotubes with β-lactone resulted in increased Gadd153/CHOP levels in the nuclear fraction (Fig. 7A). The nuclear localization of Gadd153/CHOP and the effect of β-lactone treatment were confirmed by immunofluorescent detection (Fig. 7B). Importantly, the Gadd153/CHOP levels were not influenced by calpeptin (Fig. 7, A and B), similar to a recent report from our laboratory (34).
We next tested whether treatment of the myotubes with β-lactone resulted in formation of heterodimers between C/EBPβ and Gadd153/CHOP. Coimmunoprecipitation experiments provided evidence for an increased protein-protein interaction between C/EBPβ and Gadd153/CHOP in the nuclear fraction of β-lactone-treated myotubes, consistent with the formation of heterodimers between the two transcription factors (Fig. 8, A and B). Evidence for increased interaction between C/EBPβ and Gadd153/CHOP was found both when an anti-C/EBPβ antibody was used for pull down and an anti-Gadd153/CHOP antibody was used for immunoblotting (Fig. 8A) and when the reversed order was used for coimmunoprecipitation (Fig. 8B). Immunofluorescent detection showed that C/EBPβ and Gadd153/CHOP colocalized to the same nuclei in β-lactone-treated myotubes, supporting the possibility that they interacted and formed heterodimers (Fig. 8C).
To further test the role of increased Gadd153/CHOP levels in the lack of increased C/EBPβ DNA binding in β-lactone-treated myotubes, we transfected myotubes with siRNA targeting Gadd153/CHOP mRNA. Other myotubes were transfected with nontargeting siRNA. The increase in Gadd153/CHOP levels caused by treatment with β-lactone was almost completely abolished in myotubes transfected with Gadd153/CHOP siRNA but was not influenced by nontargeting siRNA (Fig. 9A). When EMSA with supershift analysis was performed, results showed that C/EBPβ DNA-binding activity was increased in β-lactone-treated myotubes in which the Gadd153/CHOP gene had been silenced (Fig. 9B, lane 6) but not in myotubes treated with non-targeting siRNA (Fig. 9B, lane 5). This result supports the concept that increased Gadd153/CHOP prevents C/EBPβ DNA-binding activity in β-lactone-treated myotubes and may also explain why the increased C/EBPβ DNA-binding activity seen in myotubes treated with calpeptin was blunted by β-lactone (see Fig. 6A, lane 4).
Although inhibited DNA-binding activity of C/EBPβ is probably the most important consequence of the C/EBPβ-Gadd153/CHOP interaction, another possible consequence of the interaction could be that Gadd153/CHOP stabilized C/EBPβ and prevented its degradation (in which case the increase in C/EBPβ levels would be secondary to increased Gadd153/CHOP levels). To test that possibility, we determined C/EBPβ levels in myotubes treated with Gadd153/CHOP siRNA. Downregulation of Gadd153/CHOP did not prevent the β-lactone-induced increase in C/EBPβ levels (Fig. 9C, lanes 3 and 4) despite the fact that the β-lactone-induced increase in Gadd153/CHOP was efficiently inhibited (see Fig. 9A, lanes 3 and 4). This result argues against a model in which increased C/EBPβ levels after treatment with β-lactone were caused by stabilization of the protein by increased Gadd153/CHOP levels.
In the present study, treatment of cultured myotubes with the specific proteasome inhibitor β-lactone resulted in a dose- and time-dependent increase in nuclear C/EBPβ levels. Additional experiments showed that the stability of C/EBPβ was increased in the presence of β-lactone and that the proteasome inhibitor increased C/EBPβ levels independent of protein synthesis. Taken together, the results suggest that C/EBPβ is degraded, at least in part, by a proteasome-dependent mechanism in skeletal muscle cells.
In previous studies, conflicting results were reported with regard to the role of the proteasome in the regulation of C/EBPβ expression, possibly reflecting cell specificity. Thus, whereas evidence of proteasome-dependent degradation of C/EBPβ was found in cultured intestinal epithelial cells (15), hepatocytes (22), lung fibroblasts (17), and in the sea slug A. californica (37), other studies suggested that C/EBPβ degradation was not regulated by the proteasome in human melanoma cells (11), keratinocytes (30), or preadipocytes (32). To our knowledge, the present report is the first study examining the role of the proteasome in the degradation of C/EBPβ in skeletal muscle cells.
In recent experiments, we found evidence that C/EBPβ degradation was calpain dependent in cultured myotubes (34). Taken together with those observations, the present results suggest that the degradation of C/EBPβ is regulated by multiple proteolytic mechanisms in skeletal muscle. Because calpains typically cleave proteins into products that are subsequently further degraded by other mechanisms (10), it is possible that C/EBPβ is first cleaved by calpains and then degraded by the proteasome. Indeed, results in our recent report were consistent with the concept that calpain activity generates C/EBPβ cleavage products, including the 21-kDa LIP, that are further degraded by the proteasome (34). It should be noted, however, that the interaction between the calpain and proteasome systems in skeletal muscle is probably complex and may reflect not only a sequential cleavage and degradation of protein substrates. For example, in recent experiments, we found evidence that the two systems may be regulated in parallel, rather than in tandem, in sepsis-induced muscle wasting (9). It should also be noted that, because the proteasome degrades proteins into small peptides rather than free amino acids, it is likely that additional proteolytic activities participate in the degradation of C/EBPβ “downstream” of the proteasome. Interestingly, there is evidence that multiple proteolytic mechanisms regulate the degradation of other transcription factors as well, including c-Fos, c-Jun, and p53 (27).
In contrast to the apparently conflicting reports on proteasome-dependent degradation of C/EBPβ, there is more uniform evidence that other members of the C/EBP family, including Gadd153/CHOP, are degraded by the proteasome (11, 30, 32). In the present study, treatment of cultured myotubes with β-lactone resulted in increased nuclear expression of Gadd153/CHOP, consistent with proteasome-dependent degradation of Gadd153/CHOP in muscle cells. This observation is similar to results in other cell types in which treatment with various proteasome inhibitors resulted in increased Gadd153/CHOP levels (11, 32).
Although Gadd153/CHOP has been reported to activate the transcription of certain genes, including genes referred to as DOCs (for downstream of CHOP; see Ref. 33), Gadd153/CHOP more commonly acts as a dominant negative member of the C/EBP family by forming non-DNA-binding heterodimers with other C/EBP family members (6, 26, 29). Because, in the present study, the increased nuclear levels of C/EBPβ in β-lactone-treated myotubes were not associated with increased C/EBPβ DNA-binding activity, we hypothesized that treatment of the myotubes with β-lactone resulted in increased levels of Gadd153/CHOP that inhibited C/EBPβ DNA-binding activity. Our results did, indeed, support that hypothesis. Thus treatment of the myotubes with β-lactone resulted in increased levels of Gadd153/CHOP that colocalized and formed heterodimers with C/EBPβ in the nuclei. Importantly, when the β-lactone-induced increase in Gadd153/CHOP levels was prevented by Gadd153/CHOP siRNA, the increased C/EBPβ levels resulted in increased DNA binding. A similar increase in Gadd153/CHOP levels and inhibition of C/EBPβ DNA binding as observed here were reported recently in cultured preadipocytes treated with N-acetyl-Leu-Leu-norleucine, a nonspecific protease inhibitor that inhibits proteasome and calpain activities, further supporting the important role of interaction between C/EBP family members in the regulation of gene transcription (32).
In contrast to the present study, increased C/EBPβ levels caused by treatment of cultured enterocytes with the proteasome inhibitors MG-132 and lactacystin resulted in increased DNA-binding activity in a recent study from our laboratory (15). Although Gadd153/CHOP levels were not measured in that study, the increase in C/EBPβ DNA binding suggests that the Gadd153/CHOP expression was not increased in those experiments. Thus the role of the proteasome in the degradation of Gadd153/CHOP may be cell specific, similar to the regulation of C/EBPβ degradation.
Although most proteins degraded by the proteasome are ubiquitinated (12), some proteins, for example ornithine decarboxylase (13), are degraded by the proteasome without prior ubiquitination. It is not known from the present experiments whether C/EBPβ is ubiquitinated before being degraded by the proteasome. Previous studies suggest that other C/EBP family members, including C/EBPα, C/EBPγ, and Gadd153/CHOP, are ubiquitinated before degradation (11, 30). Interestingly, in a recent study by Kuang and Goldstein (17), treatment of cultured lung fibroblasts with MG-132 resulted in increased C/EBPβ protein levels without generation of high-molecular-weight, presumably ubiquitin-linked, forms of C/EBPβ, indicating that C/EBPβ may be degraded by the proteasome without prior ubiquitination in lung fibroblasts.
The present study is important because it defined some of the potential mechanisms involved in the regulation of C/EBPβ expression and activity in skeletal muscle. In previous studies, we observed that sepsis- and glucocorticoid-induced muscle wasting was associated with increased expression and activity of C/EBPβ (23, 38). Because several genes involved in the development of muscle wasting have putative C/EBPβ-binding sites in their promoter regions (23), an increased understanding of mechanisms regulating C/EBPβ expression and activity in skeletal muscle may have important clinical implications. It should be noted that, although the present and a recent report from our laboratory (34) suggest that C/EBPβ proteins are degraded by both calpain- and proteasome-dependent mechanisms, it is not known at present which degradation mechanism is most important in physiological and pathological conditions. In sepsis- and glucocorticoid-induced muscle wasting, C/EBPβ expression and DNA-binding activity are increased despite upregulated activity of both calpain- and proteasome-dependent protein degradation (23, 38). Those observations suggest that the increased C/EBPβ levels noticed in atrophying skeletal muscle during sepsis and glucocorticoid treatment reflect increased synthesis of C/EBPβ, an interpretation supported by increased C/EBPβ mRNA levels in those conditions. Thus it is possible that, in various muscle-wasting conditions, the turnover of C/EBPβ is stimulated with increased calpain- and proteasome-dependent degradation of C/EBPβ and an even more pronounced increase in the synthesis of the transcription factor, resulting in a net increase in C/EBPβ levels.
Because, in the present study, increased Gadd153/CHOP expression resulted in inhibited C/EBPβ DNA-binding activity, it may be speculated that sepsis- and glucocorticoid-induced muscle wasting is not associated with a substantial increase in the expression of Gadd153/CHOP. It will be important in future studies to determine the influence of sepsis and other muscle-wasting conditions on the expression and activity of Gadd153/CHOP and its role in the regulation of C/EBPβ DNA-binding activity.
The study was supported in part by National Institutes of Health Grants R01 DK-37908 and R01 NR-8545 (both to P.-O. Hasselgren).
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 © 2007 the American Physiological Society