help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2003-1749
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
145/10/4592    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Strle, K.
Right arrow Articles by Kelley, K. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Strle, K.
Right arrow Articles by Kelley, K. W.
Endocrinology Vol. 145, No. 10 4592-4602
Copyright © 2004 by The Endocrine Society

Proinflammatory Cytokine Impairment of Insulin-Like Growth Factor I-Induced Protein Synthesis in Skeletal Muscle Myoblasts Requires Ceramide

Klemen Strle, Suzanne R. Broussard, Robert H. McCusker, Wen-Hong Shen, Rodney W. Johnson, Gregory G. Freund, Robert Dantzer and Keith W. Kelley

Laboratory of Immunophysiology (K.S., S.R.B., R.H.M., W.-H.S., K.W.K.) and Integrative Biology (R.W.J.), Department of Animal Sciences and Department of Pathology (G.G.F.), College of Medicine, University of Illinois, Urbana, Illinois 61801; and Neurobiologie Intégrative (R.D.), Unité Mixte de Recherche, Institut National de la Recherche Agronomique-Université de Bordeaux 2, Centre National de la Recherche Scientifique, Rue Camille Saint-Saëns, 33077 Bordeaux Cedex, France

Address all correspondence and requests for reprints to: Keith W. Kelley, University of Illinois, Laboratory of Immunophysiology, Department of Animal Sciences, 207 Edward R. Madigan Laboratory, 1201 West Gregory Drive, Urbana, Illinois 61801. E-mail: kwkelley{at}uiuc.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
GH and IGF-I control over 80% of postnatal growth. We recently established that TNF{alpha} impairs the ability of IGF-I to increase protein synthesis and promote expression of myogenin in myoblasts. Here we extend these results by showing that ceramide, a second messenger in both TNF{alpha} and IL-1ß receptor signaling pathways, is a key downstream sphingosine-based lipid that leads to IGF-I resistance. A cellpermeable ceramide analog, C2-ceramide, inhibits IGF-I-induced protein synthesis by 65% and blocks the ability of IGF-I to increase expression of two key myogenic factors, myogenin and MyoD. Identical results were obtained with both TNF{alpha} and IL-1ß (1 ng/ml). Consistent with these data, neutral sphingomyelinase (N-SMase), an enzyme that catalyzes formation of ceramide from sphingomyelin, blocks IGF-I-induced protein synthesis and expression of both myogenin and MyoD. The possibility that cytokine-induced ceramide production is required for disruption of IGF-I biologic activity was confirmed by treating C2C12 myoblasts with inhibitors of all three ceramide-generating pathways. A N-SMase inhibitor, glutathione, as well as an acidic sphingomyelinase (A-SMase) inhibitor, D609, reverse the cytokine inhibition of IGF-I-induced protein synthesis by 80% and 45%, respectively. Likewise, an inhibitor of de novo ceramide synthesis, FB1, causes a 50% inhibition. Similarly, all three inhibitors significantly impair the ability of both TNF{alpha} and IL-1ß to suppress IGF-I-driven expression of myogenin. These experiments establish that ceramide, derived both from sphingomyelin and de novo synthesis, is a key intermediate by which proinflammatory cytokines impair the ability of IGF-I to promote protein synthesis and expression of critical muscle-specific transcription factors.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
ELEGANT GENETIC STUDIES have recently established that over 80% of postnatal growth is controlled by GH, IGF-I, and their interaction (1). The growth-promoting properties of these two hormones are being exploited in chronic diseases that can lead to muscle wasting. We recently summarized results from over 300 subjects involved in clinical trials that tested the effects of GH administration in AIDS patients (2). In all these experiments, plasma IGF-I concentrations increased by 2- to 4-fold after GH therapy, but the increase in lean body mass was less than what is normally obtained in GH-deficient children and adults. Even direct injections of recombinant human IGF-I, along with GH, into AIDS patients with wasting were ineffective in increasing a sustained anabolic response (3, 4). These in vivo data confirm the concept that a major problem with GH therapy for wasting AIDS patients is IGF-I resistance.

Muscle wasting in AIDS (5) and cachectic cancer patients (6), as well as in normal aging (7, 8), is closely associated with increased expression of two major proinflammatory cytokines, TNF{alpha}, and IL-1ß (reviewed in Refs.5 and 6). These cytokines can interfere with muscle growth and regeneration, in part by disrupting both IGF-I (Ref.9 ; and reviewed in Refs.7 and 10) and insulin (11, 12) signaling cascades. Skeletal muscle growth and repair are dependent upon activation, proliferation, and differentiation of myoblasts. We recently reported that TNF{alpha}, at as little as 100 pg/ml, significantly impairs the ability of IGF-I to increase protein synthesis in primary myoblasts. Nearly identical results were found in murine C2C12 myoblasts treated with as little as 10 pg/ml TNF{alpha} (13) or IL-1ß (14). We also found that very low concentrations of either TNF{alpha} (13) or IL-1ß (14) significantly reduce the ability of IGF-I to increase expression of myogenin, a critical muscle differentiation factor (reviewed in Refs.15 and 16). This inhibition is not mediated by TNF{alpha} impairing the ability of IGF-I to activate the intrinsic tyrosine kinase activity of the IGF-I receptor. Instead, TNF{alpha} and IL-1ß induce IGF-I resistance by reducing tyrosine phosphorylation of both insulin-receptor substrate (IRS)-1 and IRS-2 docking proteins (13, 14). These findings are identical to results we reported earlier with TNF{alpha} and IL-1ß in breast cancer epithelial cells (17, 18). Although it is clear that TNF{alpha} and IL-1ß attenuate the anabolic actions of IGF-I in vivo and in vitro, the downstream mediators that cause this inhibition are only beginning to be elucidated.

Ceramide is a sphingosine-based lipid second messenger that mediates many of the actions of proinflammatory cytokines. Recent reports indicate that intracellular ceramide increases linearly with age in animals and humans, and is elevated in chronic diseases such as AIDS, type II diabetes, and cancer (Ref.19 ; and reviewed in Refs.20 and 21). All of these conditions are characterized by increased expression of proinflammatory cytokines, particularly TNF{alpha}. Activation of TNF{alpha} receptors is well accepted to increase intracellular ceramide by three major pathways: neutral (N) and acidic (A) sphingomyelinase (SMase) pathways that catalyze the breakdown of membrane sphingomyelin to ceramide (22, 23, 24, 25) and a de novo ceramide synthesis pathway requiring ceramide synthase (25, 26). Similarly, several investigators have determined that ceramide is significantly elevated after IL-1ß receptor activation via both sphingomyelinases (27, 28) and probably the de novo synthesis pathway (29), but the specific IL-1ß receptor-associated subunits that induce intracellular ceramide expression have not yet been defined.

Ceramide has been reported to impair the ability of the insulin receptor, a close relative to the IGF-I receptor, to increase glucose transport in muscle cells (30, 31). However, the potential roles of an increase in ceramide production caused by the two major proinflammatory cytokines in regulating IGF-I-dependent protein synthesis and myogenin expression in myoblasts are unknown. Here we confirm that TNF{alpha} (13) and IL-1ß (14) inhibit the ability of IGF-I to promote protein synthesis and increase expression of myogenin, and we extend these results to the MyoD transcription factor in C2C12 myoblasts. More importantly, we determine that inhibitors of all three ceramide-generating pathways significantly reverse the ability of TNF{alpha} and IL-1ß to suppress IGF-I induction of both protein synthesis and expression of myogenin. Collectively, these results establish that ceramide from both de novo- and SMase-dependent pathways is required for cytokines to inhibit the myogenic actions of IGF-I in skeletal muscle myoblasts.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Reagents
Recombinant human IGF-I, murine TNF{alpha}, and murine IL-1ß were purchased from Intergen (Purchase, NY). DMEM with 4.5 g/liter glucose, 0.584 g/liter glutamine, and penicillin/streptomycin were obtained from Bio Whittaker Cambrex (Walkersville, MD). Fetal bovine serum (FBS; < 0.25 EU/ml of endotoxin) was purchased from HyClone (Logan, UT). Tritiated phenylalanine, L-[2,3,4,5,6-3H]-phenylalanine (3H-phenylalanine) and protein G-Sepharose beads were obtained from Amersham Biosciences (Piscataway, NJ). The myogenin murine monoclonal antibody (F5D) and MyoD rabbit polyclonal antibody (M-318) were from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit anti-IRS-1 antibody was from Upstate Biotechnology (Lake Placid, NY) and the murine PY-20 phosphotyrosine-specific antibody from Transduction Laboratories (Lexington, KY). A mouse monoclonal antibody to {alpha}-tubulin (B-5-1-2), Fumonisin B1 (FB1), O-Tricyclo[5.2.1.0 (2, 6)]dec-9-yl dithiocarbonate potassium salt (D609), L-glutathione (GSH), manumycin A from Streptomyces parvulus, N-SMase from Staphylococcus aureus, and C2-ceramide (N-acetyl-D-sphingosine), a cell-permeable ceramide analog, were purchased from Sigma-Aldrich Chemical (St. Louis, MO).

We evaluated data from at least three published articles to establish a general range of effective concentrations for each inhibitor. We then conducted dose-response experiments to determine the optimal concentration of the inhibitors based on their ability to reverse TNF{alpha}-dependent inhibition of IGF-I-induced protein synthesis. The most effective nontoxic dose was used in our experiments. Dose response curves for C2-ceramide (0.01–100 µM) and N-SMase (5–100 mU) were used to determine the lowest concentration that inhibited IGF-I-induced protein synthesis (data not shown). Similarly, we ensured that the optimal concentration of C2-ceramide, N-SMase and the ceramide pathway inhibitors did not reduce cell survival over a 36-h incubation period, as assessed by a colorimetric methylthiazolyldiphenyl-tetrazolium bromide assay (data not shown), as we previously described (32). The optimal concentrations used in our experiments were consistently lower for both C2-ceramide (1 µM) and N-SMase (25 mU) than has been reported in a variety of other cells [10–40 µM C2-ceramide (33, 34, 35, 36); 100–200 mU N-SMase (37, 38)]. Preliminary experiments also established that the optimal inhibitory concentrations of FB1 (1 µM), D609 (5 µg/ml), and GSH (3 mM) in C2C12 myoblasts were generally lower than reported for other cells. FB1 is a ceramide synthase inhibitor from Fusarium moniliforme that is effective in a variety of cells including hepatocytes, COS-1 cells, MCF-7 human breast cancer epithelial cells and neural-derived PC12 cells at concentrations of 0.2–100 µM (25, 34, 39), with an optimal concentration of 1 µM in hepatocytes (25, 34, 39) to 60 µM in COS-1 cells (34). D609 inhibits A-SMase activity and is effective between 1 and 100 µg/ml, with an optimal concentration of 50 µg/ml in U937 cells (22). GSH is a N-SMase inhibitor that is effective at physiological concentrations of 2 mM to 20 mM, with an optimal concentration between 3 and 5 mM for human leukemia and MCF-7 cells (40, 41).

The diluent for C2-ceramide was dimethylsulfoxide (final concentration of 0.01%), for FB1 it was methanol (final concentration of 0.1%), and for N-SMase it was glycerol (final concentration of 0.1%). These diluents were added to the medium of all treatments for the respective experiments. All other reagents were from Sigma-Aldrich.

Cell culture
Murine C2C12 myoblasts (ATCC, Manassas, VA) were maintained at 37 C, 95% humidity, and 7% CO2 in DMEM supplemented with 10% heat-inactivated FBS, 100 µg/ml streptomycin, and 100 U/ml of penicillin. For all experiments, cells were grown to 70–80% confluence, washed three times in DMEM to remove growth factors, and deprived of serum for 5 h before initiation of treatments. In experiments measuring protein synthesis, C2C12 myoblasts were treated with TNF{alpha}, IL-1ß, or with optimal concentrations of C2-ceramide, N-SMase, FB1, D609, or GSH for 1 h before addition of IGF-I and 1.5 µCi of 3H-phenylalanine for an additional 5 h. Incorporation of 3H-phenylalanine into protein was determined by harvesting cells onto glass fiber filters (Whatman International Ltd., Maidstone, UK) with deionized water. The free 3H-phenylalanine that was not incorporated into protein was removed by washing the filters four times with deionized water. Cells were harvested using a PHD cell harvester model 200A (Cambridge Technology, Watertower, MA). The amount (counts per minute) of 3H-phenylalanine incorporated into protein was determined using an LS 6000IC Beckman Scintillation Counter (Beckman Coulter, Fullerton, CA). In experiments measuring myogenin and MyoD protein expression, myoblasts were pretreated as described above but incubated with IGF-I without 3H-phenylalanine for 12, 24, or 36 h. For immunoprecipitation experiments, C2C12 myoblasts were pretreated with TNF{alpha} or C2-ceramide for 1 h before a 3-min stimulation with IGF-I.

Protein synthesis in myotubes
C2C12 myoblasts were grown to 90% confluence and then induced to differentiate into myotubes by being cultured in differentiation medium [DMEM containing 2% horse serum (HS)] for 5 d. Myotubes were washed three times with DMEM and deprived of serum for 4 h before a 1-h treatment with 1 µM C2-ceramide. Cells were then treated with IGF-I for an additional 40 min and pulsed with 1.5 µCi 3H-phenylalanine for 1 h. Protein synthesis was expressed as incorporation of 3Hphenylalanine per milligram of total trichloroacetic acid (TCA)-precipitable protein, as previously described (42). Briefly, 5% TCA was added to the cells. The plates were scraped, washed twice with 5% TCA, and the samples were pooled and placed on ice for 1 h. To pellet the precipitated protein, the samples were centrifuged at 6000 x g. The pelleted proteins were washed three times in 5% TCA to remove free 3H-phenylalanine and then dissolved in 0.1 N NaOH. Samples were then heated to 70 C for 2 h in 0.1% sodium dodecyl sulfate (SDS) to aid in solubilizing the precipitate. Protein concentration was obtained with a Bio-Rad Dc Protein Assay (Bio-Rad, Hercules, CA). The samples were suspended in water-based scintillation fluid (Ecoscint H, National Diagnostics) and 3H-phenylalanine incorporation measured with a LS 6000IC Beckman Scintillation Counter. The values were expressed as counts per minute per milligram of precipitated protein.

Ceramide analysis by HPLC
C2C12 cells were cultured in control medium or TNF{alpha} (10 ng/ml) for 4 h. Cells were rinsed in ice-cold PBS and maintained at –80 C. As a control treatment, C2-ceramide (10 µg) was added to other cells after the final wash with PBS. Concentration of ceramide in lysates of 2 x 107 cells was measured by Avanti Lipids (Alabaster, AL) using an HPLC method that is validated for quantifying intracellular ceramide.

Western blotting for myogenin or MyoD
Cells were lysed in homogenization buffer [50 mM HEPES, 1% Triton X-100, 150 mM NaCl2, 1 mM EGTA, 10 mM Na pyrophosphate, 10% glycerol, 1.5 mM MgCl2, 1% Na deoxycholate, 0.1% SDS, 0.1 mM sodium orthovanadate, 100 mM NaF, 1 mM phenylmethylsulfonyl fluoride, 2 µg/ml aprotinin, 40 nM leupeptin, and 2 µg/ml pepstatin (pH 7.4)]. The protein concentration in whole cell lysate samples was determined using the Bio-Rad Dc Protein Assay kit (Bio-Rad Laboratories). None of the experimental treatments, including TNF{alpha}, IL-1ß, C2-ceramide, N-SMase, or the ceramide pathway inhibitors, affected the amount of protein in whole cell lysates compared with control C2C12 cells at any of the times tested (data not shown). Cellular proteins (50 µg) were separated in 12.5% polyacrylamide gels containing SDS (SDS-PAGE) and transferred to Trans-Blot polyvinylidene difluoride (PVDF) membranes (Bio-Rad Laboratories). The PVDF membranes were then blocked with 1% BSA for 1 h and incubated overnight with myogenin, MyoD, or {alpha}-tubulin-specific antibodies diluted 1/1000 (1/2000 for {alpha}-tubulin) in 1% BSA in T-TBS [20 mM Tris-HCl (pH 7.4), 150 mM NaCl, 0.05% Tween 20]. Subsequently, the blots were washed with T-TBS, incubated for additional 1 h with antirabbit or antimouse horseradish peroxidase-labeled IgG (1:1000) and then developed with enhanced chemiluminescence (ECL) Western Blotting Detection Reagent (Amersham Biosciences). Finally, the membranes were exposed to autoradiographic film (Classic Film BX, Midwest Scientific, St. Louis, MO). The intensity of protein bands on autoradiograms was quantified by scanning with an Agfa Duosacan T1200 scanner (NucleoTech, San Mateo, CA) followed by the densitometric analysis using GelExpert 3.5 software (NucleoTech). The densitometric summaries (see Figs. 6–8GoGoGo) are presented are ratios of mass of target proteins myogenin or MyoD to {alpha}-tubulin protein loading controls. Individual values were divided by the experimental mean was used to standardize the data. Experiments were repeated independently three times, unless noted differently.



View larger version (37K):
[in this window]
[in a new window]
  		FIG. 6. IGF-I-induced myogenin expression is blocked by C2ceramide and N-SMase. C2C12 myoblasts were treated with TNF{alpha} (1 ng/ml), IL-1ß (1 ng/ml), N-SMase (25 mU/ml), or C2-ceramide (1 µM) before addition of IGF-I. Expression of myogenin and the control protein {alpha}-tubulin were analyzed by Western blotting at 12 (B), 24 (C), and 36 (D) h. The densitometric summaries are presented as a ratio of the mass of myogenin to that of loading control protein, {alpha}-tubulin (B–D). A representative Western blot of myogenin expression after 24-h exposure to the 10 treatments is shown in panel A. IGF-I induced a significant increase in expression of myogenin at 12, 24, and 36 h. TNF{alpha} and IL-1ß, as well as C2-ceramide and N-SMase, completely blocked IGF-I-induced myogenin expression at all three time points. *, P < 0.05; **, P < 0.01, n = 3 at each time point.

 


View larger version (34K):
[in this window]
[in a new window]
  		FIG. 7. C2-ceramide and N-SMase inhibit expression of the MyoD transcription factor that is induced by IGF-I. Expression of the transcription factor MyoD and the control protein {alpha}-tubulin were determined by Western blotting. A representative Western blot of MyoD expression after a 36-h exposure to the 10 treatments is shown in panel A. IGF-I caused a significant increase in MyoD expression at 24 (C) and 36 (D), but not at 12 h (B). The mass of {alpha}-tubulin was not affected at any time point by any treatment. The IGF-I-induced increase in MyoD was fully blocked by both TNF{alpha} and IL-1ß, as well as C2-ceramide and N-SMase at both 24 and 36 h. Summaries of the densitometric values are presented as ratios of MyoD to control protein {alpha}-tubulin (B–D). **, P < 0.01, n = 3 at each time point.

 


View larger version (37K):
[in this window]
[in a new window]
  		FIG. 8. The ability of TNF{alpha} and IL-1ß to impair IGF-I-induced myogenin expression requires both de novo and SMase-dependent ceramide generating pathways. After a 1-h treatment with TNF{alpha} (1 ng/ml) or IL-1ß (1 ng/ml) alone or in the presence of the de novo ceramide synthesis inhibitor FB1 (1 µM, n = 3) (A), the A-SMase inhibitor D609 (5 µg/ml, n = 4) (B) or the N-SMase inhibitor GSH (3 mM, n = 4) (C), myoblasts were incubated with or without IGF-I for an additional 36 h. In control medium, IGF-I consistently increased (**, P < 0.01) expression of myogenin in the presence or absence of any inhibitor. Similarly, both TNF{alpha} and IL-1ß suppressed (**, P < 0.01) the ability of IGF-I to increase expression of myogenin. Expression of {alpha}-tubulin remained unaffected by any of the treatments. The densitometric values are presented as a ratio of myogenin to control protein {alpha}-tubulin (A–C). FB1, the de novo ceramide synthesis inhibitor, partially but significantly impaired the inhibition of IGF-I-induced myogenin expression caused by both TNF{alpha} and IL-1ß. The inhibitor of A-SMase, D609, blocked (*, P < 0.05) the suppressive effects of TNF{alpha}, but it only modestly affected the inhibition caused by IL-1ß. In contrast, GSH, a N-SMase inhibitor, almost completely reversed TNF{alpha} and IL-1ß inhibition of IGF-I-induced myogenin expression.

 
Immunoprecipitation of IRS-1
Procedures for immunoprecipitation of IRS-1 have been previously described by our laboratory (14, 17). In brief, 1 x 106 C2C12 myoblasts were treated as described in Results and lysed in homogenization buffer. IRS-1 was immunoprecipitated by incubating cell lysates overnight at 4 C with 2 µg/ml of a specific anti-IRS-1 antibody and 25 µl of protein G-Sepharose beads. Protein bound to the beads was washed four times with homogenization buffer, heated at 100 C for 10 min in 100 µl Laemmli buffer and separated in 7.5% SDS-PAGE gels. These proteins were then transferred to PVDF membranes and blotted with antibodies specific to phosphotyrosine residues or IRS-1. Autoradiography was performed as described above for Western blotting.

Statistical analysis
Data for both protein synthesis and myogenic transcription factors were analyzed as a completely randomized design as a General Linear Model using the Statistical Analysis System V8 (43). Treatment differences were determined by an F-protected Duncan’s multiple range test.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
C2-ceramide mimics proinflammatory cytokine inhibition of IGF-I-induced protein synthesis and IRS-1 tyrosine phosphosphorylation
TNF{alpha} inhibits IGF-I-induced protein synthesis in human myoblasts (9), a finding we recently extended to both TNF{alpha}- (13) and IL-1ß (14)-induced suppression of IGF-I-induced myogenin expression in murine C2C12 myoblasts. It is well known that ceramide is a common mediator of many biologic actions of both TNF{alpha} and IL-1ß, but the role of ceramide in regulating IGF-I-induced protein synthesis and myogenin expression has not been reported. To test this hypothesis, we first conducted preliminary experiments to confirm that TNF{alpha} causes ceramide accumulation in C2C12 myoblasts. The concentration of ceramide in cells cultured in medium alone was 23 µg/ml, and this increased to 38 µg/ml after addition of TNF{alpha} (10 ng/ml). As an internal control, 10 µg of exogenous C2-ceramide was added at initiation of the experiment to cells cultured in medium alone. The amount of ceramide recovered in these cells was 31 µg/ml, indicating that the assay could reliably detect ceramide. Collectively, these data show over a 50% increase in ceramide concentration in cells treated with TNF{alpha} for 4 h. These results are consistent with another report demonstrating that C2C12 myoblasts generate ceramide (31).

Next, C2C12 myoblasts were pretreated with TNF{alpha}, IL-1ß, or C2-ceramide for 1 h before a 5-h incubation in the presence or absence of IGF-I. As expected, IGF-I doubled the rate of protein synthesis (P < 0.01), and this effect was inhibited (P < 0.01) by 70% by both TNF{alpha} and IL-1ß (Fig. 1AGo). C2-ceramide mimicked this inhibition and reduced (P < 0.01) IGF-Iinduced 3H-phenylalanine incorporation into protein by 65% (Fig. 1AGo). Similar results were obtained in multinucleated differentiated C2C12 myotubes (Fig. 2AGo). As with C2C12 myoblasts, C2-ceramide inhibited (P < 0.01) IGF-I-induced protein synthesis in C2C12 myotubes, as assessed by TCA precipitation (Fig. 2AGo).



View larger version (41K):
[in this window]
[in a new window]
  		FIG. 1. IGF-I-induced protein synthesis in murine myoblasts and tyrosine phosphorylation of IRS-1 is inhibited by proinflammatory cytokines or their downstream activation products. A, C2C12 myoblasts were treated with TNF{alpha} (1 ng/ml), IL-1ß (1 ng/ml), or C2-ceramide (1 µM) for 1 h before a 5 h incubation with 1.5 µCi of 3H-phenylalanine (3H-Phe) in the absence or presence of IGF-I (50 ng/ml). 3H-phenylalanine incorporation into new protein was quantified as described in Materials and Methods. IGF-I significantly increased protein synthesis, an effect that was inhibited (**, P < 0.01, n = 3) by both TNF{alpha} and IL-1ß. C2-ceramide mimicked this inhibition by decreasing IGF-I-induced protein synthesis. B, To examine if ceramide and TNF{alpha} inhibit IGF-I receptor-proximal signaling in myoblasts, myoblasts were pretreated with C2-ceramide (0.1 and 1 µM) or TNF{alpha} (1 ng/ml) for 1 h before addition of IGF-I for 3 min. Immunoprecipitation of IRS-1 followed by Western blotting with an antiphosphotyrosine antibody revealed that both TNF{alpha} and C2-ceramide inhibit the ability of IGF-I to induce tyrosine phosphorylation of IRS-1. C, Myoblasts were treated as described in panel A, but N-SMase (25 mU/ml) was substituted for ceramide. N-SMase acted similar to its downstream product, ceramide, as well as TNF{alpha} and IL-1ß, to inhibit (**, P < 0.01, n = 3) IGF-I-induced protein synthesis.

 


View larger version (30K):
[in this window]
[in a new window]
  		FIG. 2. C2-ceramide inhibits IGF-I-induced protein synthesis in myotubes. C2C12 myoblasts were cultured in DMEM containing 2% HS and allowed to differentiate. Myotubes were then treated with C2-ceramide (1 µM) for 1 h before addition of IGF-I (50 ng/ml) for another 100 min. Cells were pulsed with 1.5 µCi of 3H-phenylalanine for the last 60 min of treatment. IGF-I induced an increase in protein synthesis (**, P < 0.01, n = 2) and this increase was inhibited by C2-ceramide.

 
Because C2-ceramide inhibited the biological actions of IGF-I, we investigated the possibility that ceramide regulates early IGF-I receptor-proximal signaling events in C2C12 myoblasts. We have previously established that TNF{alpha} and IL-1ß inhibit IGF-I-induced tyrosine phosphorylation of IRS-1 (13, 14, 17), a large docking protein that transduces signals from both IGF-I and insulin receptors. Early work showed that ceramide inhibits tyrosine phosphorylation of insulininduced activation of IRS-1 (44, 45). Here we confirm the inhibition caused by TNF{alpha} (13) and extend these results by showing that C2-ceramide also inhibits IGF-I-induced tyrosine phosphorylation of IRS-1 in C2C12 myoblasts (Fig. 1BGo, representative blot). It is therefore possible that the cytokine-dependent inhibition of IGF-I-induced protein synthesis by may be due, at least in part, to suppression of early upstream signaling events associated with activation of the IGF-I receptor.

The inhibition in myoblasts caused by C2-ceramide was extended by incubating myoblasts with N-SMase (25 mU/ml). This enzyme generates intracellular ceramide and was predicted to have similar inhibitory effects as the addition of exogenous C2-ceramide. Indeed, N-SMase inhibited (P < 0.01) IGF-I-induced protein synthesis by 85% (Fig. 1CGo). Notably, TNF{alpha}, IL-1ß, C2-ceramide and N-SMase did not affect protein synthesis in the absence of IGF-I. These results are consistent with the idea that ceramide generated from the N-SMase-dependent pathway may mediate cytokine inhibition of IGF-I-induced protein synthesis.

TNF{alpha} and IL-1ß inhibition of IGF-I-induced protein synthesis is mediated by ceramide
Ceramide can be newly synthesized or generated by degradation of membrane sphingomyelin. Ceramide from either source serves as a second messenger for TNF{alpha} and IL-1ß receptor signaling pathways. Recently, cell-permeable ceramide analogs have been implicated in blocking insulin-induced AKT activity (30, 46) and PDGF-induced thymidine incorporation in muscle cells (47). However, those experiments did not examine the possibility that ceramide mediates cytokine receptor-induced inhibition of IGF-I biologic activity. Here we show that ceramide not only mimics cytokine action but also that ceramide is required for both TNF{alpha} and IL-1ß to inhibit IGF-I-induced protein synthesis. These experiments were accomplished by treating C2C12 myoblasts with inhibitors of each of the three ceramide-generating pathways. As expected, none of the ceramide inhibitors affected the ability of IGF-I to promote protein synthesis (Fig. 3Go, A–C). However, these inhibitors were effective, to varying degrees, in suppressing the biological actions of TNF{alpha} and IL-1ß. C2C12 myoblasts were first treated with an inhibitor of de novo synthesis of ceramide, FB1, in the presence of TNF{alpha} or IL-1ß with and without IGF-I. Whereas TNF{alpha} and IL-1ß inhibited IGF-I-induced 3H-phenylalanine incorporation, treatment with FB1 reversed this inhibition by approximately 50% (P < 0.01; Fig. 3AGo). Importantly, FB1 did not affect protein synthesis in the absence of either TNF{alpha} or IL-1ß. Comparable results were obtained when myoblasts were treated with the A-SMase inhibitor, D609, which reduced by 45% (P < 0.01) the inhibition of IGF-I-induced protein synthesis caused by TNF{alpha} and IL-1ß (Fig. 3BGo). Lastly, C2C12 myoblasts were incubated with GSH, a N-SMase inhibitor. GSH markedly impaired the ability of TNF{alpha} and IL-1ß to inhibit the anabolic actions of IGF-I, causing an 80% (P < 0.01) and a 100% (P < 0.01) reversal, respectively (Fig. 3CGo). Another N-SMase inhibitor, manumycin A, was toxic to C2C12 myoblasts at concentrations (≥0.5 µM) that are well below that required to inhibit N-SMase activity (100 µM) (48) (data not shown). Conversely, neither D609 nor GSH were active alone, and they did not affect IGF-I action in the absence of TNF{alpha} and IL-1ß. These results establish that ceramide generated from both de novo and SMase-dependent pathways is required for cytokines to inhibit IGF-I-induced protein synthesis in myoblasts.



View larger version (37K):
[in this window]
[in a new window]
  		FIG. 3. Ceramide synthase and sphingomyelinase inhibitors reverse the suppression of IGF-I-induced protein synthesis caused by TNF{alpha} and IL-1ß in myoblasts. TNF{alpha} (1 ng/ml) and IL-1ß (1 ng/ml), alone or in the presence of FB1 (1 µM), D609 (5 µg/ml), or GSH (3 mM), were added to C2C12 myoblasts for 1 h before addition of IGF-I (50 ng/ml) and 3H-phenylalanine. A, FB1, a ceramide synthase inhibitor, impaired both TNF{alpha} and IL-1ß inhibition of IGF-I-induced protein synthesis (**, P < 0.01, n = 2). B, Similarly, D609, an A-SMase blocker, reduced cytokine inhibition of IGF-I-induced protein synthesis (**, P < 0.01, n = 3). C, GSH, a N-SMase inhibitor, reversed TNF{alpha} and IL-1ß inhibition of IGF-I-induced protein synthesis (**, P < 0.01; n = 6).

 
Inhibitors of endogenous ceramide-generating pathways do not ameliorate exogenous ceramide- and N-SMase-induced inhibition of IGF-I-stimulated protein synthesis
The usefulness of pharmacological inhibitors is sometimes limited due to nonspecificity and toxicity. Although all of the ceramide-generating inhibitors used in these experiments act in diverse ways to impair ceramide generation, the inhibitors should not reduce the actions of exogenously added, preformed ceramide or N-SMase. To test this possibility, we treated C2C12 myoblasts with inhibitors of de novo ceramide synthesis (FB1) or A-SMase (D609). Immediately after addition of inhibitors, cells were incubated with exogenous C2-ceramide or N-SMase for 1 h before addition of IGF-I. Synthesis of new proteins was then determined. These results established that FB1 and D609 do not reverse C2-ceramide- (Fig. 4AGo) or N-SMase-dependent (Fig. 4BGo) inhibition of IGF-I-induced protein synthesis. Similar results were obtained when myogenin was used as the end-point for IGF-I-induced activity rather than protein synthesis. In these separate experiments, C2C12 myoblasts were treated with either FB1, (Fig. 5AGo), D609 (Fig. 5BGo), or the N-SMase inhibitor GSH (Fig. 5CGo). None of the inhibitors reversed the inhibition of IGF-I-induced myogenin expression after 36 h in culture that was caused by exogenous ceramide. These results are consistent with the notion that all three inhibitors, FB1, D609, and GSH, act to block key components of ceramide-generating pathways without interfering with other signaling intermediates.



View larger version (37K):
[in this window]
[in a new window]
  		FIG. 4. Inhibitors of endogenous ceramide-generating pathways do not reverse exogenous ceramide or N-SMase inhibition of IGF-Iinduced protein synthesis. C2-ceramide (1 µM), N-SMase (25 mU/ml), FB1 (1 µM) and D609 (5 µg/ml) were added to C2C12 myoblasts for 1 h before addition of IGF-I (50 ng/ml) and 3H-phenylalanine. A, The ceramide synthase inhibitor, FB1, and the A-SMase inhibitor, D609, did not reverse exogenous C2-ceramide-dependent inhibition of IGF-I-induced protein synthesis (**, P < 0.01, n = 2). B, Similarly, neither FB1 nor D609 reversed the inhibition of IGF-I-stimulated protein synthesis caused by exogenous N-SMase (**, P < 0.01, n = 2).

 


View larger version (48K):
[in this window]
[in a new window]
  		FIG. 5. Inhibitors of de novo and SMase-dependent ceramide generating pathways do not impair the ability of exogenous C2-ceramide to inhibit IGF-I-induced myogenin expression. Myoblasts were treated for 1 h with FB1 (1 µM), D609 (5 µg/ml), or GSH (3 mM) alone or in the presence of C2-ceramide. Cells were then incubated with or without IGF-I for an additional 36 h. C2-ceramide consistently inhibited IGF-I-induced myogenin expression, whereas the expression of {alpha}tubulin remained unaffected by any of the treatments. A, FB1, the de novo ceramide synthesis inhibitor; B, D609, the A-SMase inhibitor; and C, GSH, the N-SMase inhibitor, did not impair the inhibition of IGF-I-induced myogenin expression caused by exogenous C2ceramide. A representative Western blot of two independent experiments is shown for FB1 (A), D609 (B), and GSH (C).

 
Ceramide and N-SMase inhibit IGF-I-induced expression of myogenin and MyoD
Myoblast differentiation and fusion into myotubes are regulated by sequential expression of key myogenic transcription factors, particularly myogenin and MyoD (15, 16). Previous results established that TNF{alpha} inhibits IGF-I-induced expression of myogenin (13, 49). Here we extend these results to IL-1ß and show that ceramide can also inhibit IGF-Iinduced expression of myogenin and MyoD. C2C12 myoblasts were pretreated with TNF{alpha}, IL-1ß, N-SMase, or C2-ceramide before addition of IGF-I, followed by incubation for an additional 12, 24, or 36 h. Cellular proteins (50 µg) were separated by SDS-PAGE, and protein expression was detected by probing Western blots with antibodies specific for myogenin and MyoD, as described in Materials and Methods. Representative Western blots are presented for myogenin (Fig. 6AGo) and MyoD (Fig. 7AGo). To control for minor variations in protein loading, data were expressed as a ratio of the mass of myogenin or MyoD to the mass of {alpha}-tubulin in the same lane. A densitometric summary of at least three independent experiments revealed that IGF-I induced a modest, but statistically significant (P < 0.05) increase in myogenin at as early as 12 h (Fig. 6BGo). Longer incubations with IGF-I resulted in greater expression of myogenin (Fig. 6Go, C and D). Similar results were obtained for MyoD, but the increase in MyoD caused by IGF-I was delayed until 24 h (Fig. 7Go, B and C). TNF{alpha} and IL-1ß reversed IGF-I-induced protein expression of both myogenin (Fig. 6Go) and MyoD (Fig. 7Go) at all three time points. Consistent with the protein synthesis data (Figs. 1Go and 3Go), C2-ceramide and N-SMase mimicked TNF{alpha} and IL-1ß and also completely blocked IGF-I-induced myogenin (Fig. 6Go) and MyoD (Fig. 7Go) expression.

Ceramide mediates TNF{alpha} and IL-1ß inhibition of IGF-I-induced expression of myogenin
Addition of C2-ceramide or an enzyme that leads to formation of ceramide, N-SMase, blocks the ability of IGF-I to induce myogenic factor expression (Figs. 6Go and 7Go). In this objective, we used well-characterized inhibitors of endogenous de novo (25, 34, 39) and SMase-dependent pathways (22, 40, 41) to determine if intracellular ceramide generating pathways are involved in this phenomenon. To investigate this possibility, C2C12 myoblasts were pretreated with TNF{alpha} or IL-1ß in the presence or absence of de novo and SMase-dependent inhibitors for 1 h. Cells were then incubated with IGF-I for an additional 36 h, at which time whole cell lysates were prepared. Western blotting was used to determine expression of myogenin and {alpha}-tubulin. Consistent with data in Fig. 6DGo, a densitometric summary confirmed that IGF-I induced at least a 3-fold increase (P < 0.01) in mass of myogenin relative to {alpha}-tubulin at 36 h (Fig. 8Go). Similarly, both TNF{alpha} and IL-1ß consistently blocked (P < 0.01) myogenin expression induced by IGF-I. In the absence of IGF-I, neither of the two proinflammatory cytokines or the inhibitors, FB1, D609, and GSH, affected expression of myogenin. However, in the presence of IGF-I, FB1 reversed the inhibition caused by both TNF{alpha} and IL-1ß by approximately 50% (P < 0.01) (Fig. 8AGo). A similar trend was observed with the A-SMase inhibitor, D609. In this case, D609 blocked (P < 0.05) TNF{alpha}-induced inhibition of myogenin expression (Fig. 8BGo). A similar trend was observed for IL-1ß, but the inhibition caused by D609 did not significantly reverse the inhibitory effect of IL-1ß (27% reversal, Fig. 8BGo). The N-SMase inhibitor, GSH (Fig. 8CGo), was consistently the most effective in abrogating the ability of both TNF{alpha} and IL-1ß to suppress IGF-I-driven expression of myogenin (P < 0.01). Indeed, there was no statistical difference in the mass of myogenin between cells treated with IGF-I alone or cells treated with IGF-I and either TNF{alpha} or IL-1ß plus GSH. Collectively, these results establish that ceramide is a key intermediate by which proinflammatory cytokines, TNF{alpha} and IL-1ß, induce resistance to IGF-I in muscle cells.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Elevated concentrations of proinflammatory cytokines are closely and positively correlated with muscle wasting and reduced regenerative ability that occur during prolonged illness and aging. TNF{alpha} and IL-1ß act directly on muscle cells to block the anabolic actions of several growth factors, including IGF-I (9, 13, 14). However, the mechanisms that mediate the ability of TNF{alpha} and IL-1ß to inhibit IGF-Iinduced biological activity in myoblasts are unknown. We first confirmed our recent results (13) showing that TNF{alpha}, at 1 ng/ml, significantly inhibits the ability of IGF-I to promote protein synthesis (Fig. 1Go) and to increase expression of myogenin (Fig. 6Go). We then extended these data to show that at this same low concentration of 1 ng/ml, both TNF{alpha} and IL-1ß block IGF-I-induced expression of another key muscle-specific transcription factor, MyoD (Fig. 7Go). Addition of synthetic C2-ceramide to myoblasts, as well as an endogenous inducer of intracellular ceramide, purified N-SMase, mimics the inhibition of IGF-I caused by both TNF{alpha} and IL-1ß, as determined by three biological events: 1) protein synthesis (Figs. 1Go and 4Go); 2) expression of myogenin (Fig. 6Go); and 3) expression of MyoD (Fig. 7Go). In addition, C2-ceramide inhibits tyrosine phosphorylation of IRS-1 induced by IGF-I in myoblasts (Fig. 1BGo). Finally, C2-ceramide acts similarly in differentiated myotubes to inhibit IGF-I-induced protein synthesis (Fig. 2AGo).

Although it is well accepted that ceramide acts as a second messenger for both TNF{alpha} and IL-1ß receptor signaling pathways, the inhibitory effects of ceramide on IGF-I-induced expression of myogenin and MyoD have not been reported. We therefore examined this possibility by treating C2C12 myoblasts with inhibitors of de novo ceramide generating pathways (FB1), N-SMase (GSH), and A-SMase (D609). These experiments demonstrated that all three inhibitors significantly impair the effectiveness of TNF{alpha} and IL-1ß, but not of exogenous C2-ceramide and N-SMase, in blocking both IGF-I-induced protein synthesis (Figs. 3Go and 4Go) and myogenin expression (Figs. 5Go and 8Go). Collectively, these data establish that ceramide, a sphingosine-based lipid second messenger derived from both de novo and SMase-dependent pathways, is required for TNF{alpha} and IL-1ß to reduce the ability of IGF-I to induce protein synthesis and expression of muscle-specific transcription factors in myoblasts.

The signaling pathways of TNF{alpha} and IL-1ß are involved in regulating many of the same biological processes, such as the activation of stress proteins, proliferation, and differentiation. Both cytokines have been implicated in blocking the anabolic actions of IGF-I and are closely associated with muscle wasting that occurs in AIDS and cancer patients (reviewed in Refs.5 and 6). It is therefore not surprising that signal transduction pathways for both TNF{alpha} and IL-1ß receptors share several signaling components, including ceramide (50). These findings are supported by synthetic ceramide analogs and exogenous SMases that can bypass receptor activation and mimic biological activities of both cytokines (23, 24, 27). Consistent with these reports, we found that C2-ceramide and exogenous N-SMase mimic cytokine inhibition of IGF-I-induced protein synthesis and myogenic factor expression in myoblasts.

Myogenin and MyoD are critical muscle-specific transcription factors required for myoblast differentiation and fusion into myotubes (reviewed in Refs.15 and 16). Findings from in vivo and in vitro experiments indicate that both myogenin and MyoD heterodimerize with E proteins and induce expression of skeletal muscle-specific genes by binding to E-boxes (CANNTG) (reviewed in Ref.16). Consistent with results of others (51, 52), we showed that IGF-I increases expression of both myogenin and MyoD. Furthermore, we confirmed our recent results showing that TNF{alpha} (13) and IL-1ß (14) inhibit expression of myogenin and extended these results to C2-ceramide and MyoD. We then confirmed that IL-1ß acts similarly to TNF{alpha} because both cytokines increase intracellular concentrations of ceramide (22, 23, 24, 25, 27, 28). Indeed, IL-1ß acts at very low concentrations (1 ng/ml) to inhibit the ability of IGF-I to increase protein synthesis and augment expression of both myogenin (Fig. 6Go) and MyoD (Fig. 7Go). Similarly, C2-ceramide and N-SMase completely inhibit expression of myogenin (Fig. 6Go) and MyoD (Fig. 7Go). N-SMase appeared to be critical for both TNF{alpha} and IL-1ß action because addition of the N-SMase inhibitor, GSH, in most cases completely reverses the ability of these two proinflammatory cytokines to inhibit protein synthesis and myogenin expression induced by IGF-I (Figs. 3Go and 8Go). To begin to explore the possibility that ceramide also affects the anabolic activity of IGF-I in terminally differentiated cells, we performed a similar experiment on C2C12 myotubes. Indeed, our results show that C2-ceramide almost completely inhibits IGF-I-induced protein synthesis in differentiated myotubes (Fig. 2AGo). These data indicate that ceramide inhibits both IGF-I-induced protein synthesis and differentiation in myoblasts and decreases the ability of IGF-I to promote the synthesis of protein in myotubes.

Numerous reports have established that both TNF{alpha} (22, 23, 24, 25) and IL-1ß (27, 28, 29) increase intracellular concentrations of ceramide. However, ceramide does not mediate all biological responses of these proinflammatory cytokines (50) and may not mediate the responses to both cytokines equally (53). It is important to note that TNF{alpha} and IL-1ß receptors use several distinct signaling components that induce similar biological responses. Our results are consistent with this concept. For example, the A-SMase inhibitor, D609, reversed TNF{alpha}-mediated inhibition of IGF-I while it only partially inhibited IL-1ß-induced inhibition (Fig. 8BGo). This occurred despite the fact that both cytokines completely blocked IGF-I-induced myogenin expression (Fig. 8BGo). This finding indicates that in myoblasts, IL-1ß-induced inhibition of IGF-I biological activity may be less dependent than TNF{alpha} on A-SMase. Studies involving mutations in components of the IL-1 receptor signaling pathway are also consistent with this conclusion. For example, cells deficient in the IL-1 receptor accessory protein lose the ability to activate A-SMase, yet A-SMase activation remains sensitive to TNF{alpha} stimulation (53).

Another important conclusion can be drawn from in vivo experiments in which mice were injected with optimal concentrations of the two proinflammatory cytokines, either separately or together (54). These data established that although both TNF{alpha} and IL-1ß increase SMase activity in vivo, the extent to which they did so differ. Furthermore, the A-SMase inhibitor D609 reverses the inhibition of IGF-I-induced protein synthesis by both cytokines. This finding demonstrates that, although initial signal transduction from both receptors is distinct, these signals converge on intermediary factors such as SMases and ceramide synthase. The differences in TNF{alpha} and IL-1ß signaling pathways could therefore explain the efficacy of inhibitors to block proinflammatory cytokine-induced inhibition of IGF-I-induced biological activities (Fig. 8Go). Collectively, these in vivo and in vitro results establish that while both TNF{alpha} and IL-1ß generate similar responses, they can do so through distinct receptor-associated factors and to different degrees.

Three pharmacological blockers of ceramide synthesis were used in the experiments reported here: FB1 (de novo ceramide synthesis), D609 (A-SMase), and GSH (N-SMase). Unfortunately, many pharmacological compounds are either toxic or lack specificity. For example, GSH is well known to inhibit N-SMase and also regulate oxidation state of cells. A recent report (55) has now established that oxidative stress, GSH and N-SMase activity are tightly linked. These data clearly showed that amyloid ß, which induces oxidative stress and depletes intracellular GSH, leads to N-SMase activation, ceramide generation and consequently cell death. These events are reversed by addition of N-acetyl cysteine (NAC), a precursor that restores intracellular GSH concentrations and inhibits N-SMase activity, as well as by antisense oligonucleotides to N-SMase (55). Similarly, in C2C12 cells, high concentrations of TNF{alpha} (10 ng/ml) reduces intracellular GSH, suppresses transcription from the troponin promoter and inhibits differentiation into myotubes in the presence of 0.5% FBS (56). Addition of NAC does not affect the TNF{alpha}-induced reduction of troponin synthesis, but partially reverses TNF{alpha} inhibition of creatine kinase activity and the myogenic index. The authors concluded that inhibition of protein synthesis by TNF{alpha} does not depend upon oxidative stress, although this may not be the case for myotube formation. Unfortunately, these investigations with C2C12 cells did not explore the possibility that the reduction in intracellular GSH caused by TNF{alpha} leads to the activation of N-SMase, or whether NAC inhibits TNF{alpha}-induced N-SMase activation in C2C12 cells. Collectively, these new data, along with several other reports (40, 41, 57, 58), confirm that GSH directly inhibits N-SMase but not A-SMase, and that the downstream reduction in N-SMase activity blocks biological processes that are dependent upon GSH depletion (e.g. oxidative stress). Similarly, neither TNF{alpha} nor IL-1ß alone increase the concentrations of free oxygen radicals in C2C12 cells (59). These results indicate that in C2C12 cells treated with either TNF{alpha} or IL-1ß, GSH regulates N-SMase directly without altering the redox state of the cell. In addition, we show that the ceramide-generating inhibitors do not block the actions of exogenous, preformed ceramide (FB1 and D609 in Fig. 4AGo; and FB1, D609, and GSH in Fig. 5Go) or N-SMase (FB1, D609 in Fig. 4BGo). These results support the hypothesis that the pharmacological blockers used in these experiments inhibit ceramide-generating pathways but are unlikely to interfere with other signaling components.

Our results clearly show that FB1, D609, and GSH differ in their ability to block TNF{alpha}- and IL-1ß-induced inhibition of IGF-I actions in myoblasts. Although FB1 and D609 only partially reversed the cytokine-induced inhibition of IGF-I, GSH completely blocked this inhibition in all but one instance (Fig. 3Go). Indeed, the results showing that GSH completely blocked TNF{alpha} and IL-1ß inhibition of IGF-I-induced myogenin suggest that N-SMase is the most important ceramide-generating pathway in C2C12 myoblasts. However, several lines of evidence are not totally consistent with this conclusion. Although GSH blocked IL-1ß inhibition of IGF-I-induced protein synthesis, GSH only partially inhibited TNF{alpha}-induced inhibition (Fig. 3Go). In addition, inhibitors of A-SMase (D609) and de novo ceramide synthesis (FB1) also significantly reversed TNF{alpha}- and IL-1ß-induced inhibition of IGF-I. Therefore, although GSH exhibited the greatest reversal of the inhibitory actions by proinflammatory cytokines, the possibility cannot be dismissed that differences in the inhibitory ability between FB1, D609, and GSH were simply due to GSH being the most potent inhibitor.

IGF-I is a potent anabolic factor that induces muscle hypertrophy by promoting expression of myogenic differentiation factors, including myogenin and MyoD. Elevated expression of IGF-I in vivo is associated with increased strength, mass, and regenerative capacity of skeletal muscles (51, 60). Conversely, in vitro expression of a dominant-negative form of the IGF-I receptor dramatically decreases expression of myogenin and MyoD (61), delays muscle differentiation (62), and causes muscle hypoplasia (61). Proinflammatory cytokines have been shown to inhibit IGF-I actions in myoblasts (9, 13, 14, 49), and our new results establish that ceramide is required for TNF{alpha} and IL-1ß to block IGF-I-induced muscle-specific events. However, the precise mechanism of how ceramide blocks IGF-I-induced protein synthesis and myogenin expression is not yet understood. Previous results from our laboratory demonstrated that TNF{alpha} and IL-1ß inhibit tyrosine phosphorylation of IRS-1 (13, 14, 17). IRS proteins are critical for mediating cellular responses to insulin as well as IGF-I. For example, IRS-1-deficient mice have impaired insulin- and IGF-I-stimulated glucose uptake as well as significantly stunted muscle growth (63, 64, 65). Inhibition of IRS-1 and IRS-2 tyrosine phosphorylation could therefore significantly inhibit muscle growth and development. Our results are consistent with these previous findings indicating that ceramide mimics TNF{alpha} inhibition of IGF-I-induced IRS-1 tyrosine phosphorylation (Fig. 1BGo). Furthermore, ceramide also inhibits IGF-I-induced S473 phosphorylation of Akt (data not shown). However, it remains to be determined whether these inhibitory properties of ceramide are direct or caused by intermediate, more downstream pathways.

Other reports have established that cytokines and ceramide act to inhibit myoblast differentiation in the absence of exogenous IGF-I or -II. For example, in a report by Meadows et al. (66), TNF{alpha} (20 ng/ml) and ceramide (20 µM) alone decreased expression of creatine kinase compared with untreated controls. However, at the lower more physiological concentrations used in our experiments (TNF{alpha} and IL-ß at 1 ng/ml and C2-ceramide at 1 µM), proinflammatory cytokines were inhibitory only in presence of IGF-I. The discrepancies between our results and those of others can be primarily attributed to the differences in experimental design. In fact, we recently confirmed that TNF{alpha} at higher concentrations (10 ng/ml) reduces basal protein synthesis (13). Furthermore, both TNF{alpha} (10 ng/ml) and C2-ceramide (10 µM) reduce cell viability in both the absence and presence of IGF-I (24 h; our unpublished observations). In addition, the majority of reports studying the actions of cytokines in C2C12 cells used cell culture media supplemented with low concentrations of HS (66) or fetal bovine serum (FBS) (67) at the time of treatment, whereas we used DMEM alone. Both HS and FBS contain numerous growth-promoting factors, including IGF-I. We recently demonstrated that as little as 1 ng/ml IGF-I significantly increases protein synthesis in C2C12 myoblasts (14). Therefore, it is probable that low serum differentiation media contain sufficient concentrations of IGF-I or IGF-II to induce differentiation. Differentiation media are also likely to provide a supportive environment for endogenous synthesis of IGF-II, as shown by a significant increase in IGF-II concentrations in myoblasts (66) and MyoD transfected 10T1/2 fibroblasts (68) incubated for 48 h in medium supplemented with 2% HS. Indeed, certain biological events associated with differentiation, such as increased expression of creatine kinase (66) or expression of myosin heavy chain (68) occur concurrently with the increase in IGF-II expression and are inhibited by TNF{alpha}, ceramide, or adenovirus-encoding antisense cDNA to IGF-II. However, because a significant increase in endogenous IGF-II protein expression does not occur until 48 h, it is unlikely to account for our results that were measured at much earlier time points (protein synthesis at 6 h and expression of myogenin and MyoD at 12–36 h). Collectively, data presented here are consistent with the reports discussed above and significantly extend them by showing that even lower more physiological concentrations of proinflammatory cytokines and exogenous ceramide act only in presence of growth factors.

The amount of membrane phospholipids, such as sphingomyelin and its product ceramide (19, 20, 21), as well as TNF{alpha} and IL-ß (5, 6, 7), increase in several chronic diseases and aging. Alterations in sphingomyelin metabolism and the subsequent generation of high concentrations of ceramide are positively correlated with growth factor resistance and reduced glucose uptake into several target tissues, including skeletal muscle (20, 21). Here we directly establish that ceramide inhibits the ability of IGF-I to induce protein synthesis, tyrosine phosphorylate IRS-1 and increase expression of both myogenin and MyoD in myoblasts. Furthermore, we show that ceramide also inhibits IGF-I-induced protein synthesis in differentiated myotubes. More importantly, inhibitors of each ceramide-generating pathway reduce the ability of TNF{alpha} and IL-1ß to block IGF-I-induced protein synthesis as well as expression of specific differentiation-inducing transcription factors. It is particularly interesting that none of the experimental treatments, including C2-ceramide, N-SMase, TNF{alpha}, and IL-1ß, affect protein synthesis or expression of myogenic transcription factors in the absence of growth factor stimulation. These findings are consistent with the concept of reciprocal systems of communication between the endocrine and immune systems because they define, in muscle cells, two biologically important activities that are controlled jointly by hormones and cytokines. Collectively, these data establish that ceramide is required for proinflammatory cytokines to induce a state of IGF-I resistance in myoblasts and could provide a new target for treatment of skeletal muscle wasting disorders.


   Footnotes
 
This work was supported by grants from the National Institutes of Health to K.W.K. (AI50442 and MH 51569) and R.W.J. (AG-16710).

Abbreviations: A-SMase, Acidic sphingomyelinase; C2-ceramide, N-acetyl-D-sphingosine; SMase, sphingomyelinase; FB1, Fumonisin B1; FBS, fetal bovine serum; GSH, glutathione; HS, horse serum; IRS-1, insulin receptor substrate-1; NAC, N-acetyl cysteine; N-SMase, neutral sphingomyelinase; PVDF, polyvinylidene difluoride; SDS, sodium dodecyl sulfate; TCA, trichloroacetic acid.

Received December 24, 2003.

Accepted for publication July 6, 2004.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Lupu F, Terwilliger JD, Lee K, Segre GV, Efstratiadis A 2001 Roles of growth hormone and insulin-like growth factor 1 in mouse postnatal growth. Dev Biol 229:141–162[CrossRef][Medline]
  2. Kelley KW 2004 From hormones to immunity: the physiology of immunology. Brain Behav Immun 18:95–113[CrossRef][Medline]
  3. Ellis KJ, Lee PD, Pivarnik JM, Bukar JG, Gesundheit N 1996 Changes in body composition of human immunodeficiency virus-infected males receiving insulin-like growth factor I and growth hormone. J Clin Endocrinol Metab 81:3033–3038[Abstract]
  4. Lee PD, Pivarnik JM, Bukar JG, Muurahainen N, Berry PS, Skolnik PR, Nerad JL, Kudsk KA, Jackson L, Ellis KJ, Gesundheit N 1996 A randomized, placebo-controlled trial of combined insulin-like growth factor I and low dose growth hormone therapy for wasting associated with human immunodeficiency virus infection. J Clin Endocrinol Metab 81:2968–2975[Abstract]
  5. Chang HR, Dulloo AG, Bistrian BR 1998 Role of cytokines in AIDS wasting. Nutrition 14:853–863[CrossRef][Medline]
  6. Baracos VE 2000 Regulation of skeletal-muscle-protein turnover in cancer-associated cachexia. Nutrition 16:1015–1018[CrossRef][Medline]
  7. Grounds MD 2002 Reasons for the degeneration of ageing skeletal muscle: a central role for IGF-1 signalling. Biogerontology 3:19–24[CrossRef][Medline]
  8. Carter CS, Pahor M, Sonntag WE 2001 Assessment of physical function in aging rodents: a strategy for improving preclinical testing. Exp Aging Res 27:271–282[CrossRef][Medline]
  9. Frost RA, Lang CH, Gelato MC 1997 Transient exposure of human myoblasts to tumor necrosis factor-{alpha} inhibits serum and insulin-like growth factor-I stimulated protein synthesis. Endocrinology 138:4153–4159[Abstract/Free Full Text]
  10. Broussard SR, Zhou JH, Venters HD, Bluthé RM, Freund GG, Johnson RW, Dantzer R, Kelley KW 2001 At the interface of environment-immune interactions: cytokine and growth factor receptors. J Anim Sci 79:E268–E284
  11. Uysal KT, Wiesbrock SM, Marino MW, Hotamisligil GS 1997 Protection from obesity-induced insulin resistance in mice lacking TNF-{alpha} function. Nature 389:610–614[CrossRef][Medline]
  12. Hirosumi J, Tuncman G, Chang L, Gorgun CZ, Uysal KT, Maeda K, Karin M, Hotamisligil GS 2002 A central role for JNK in obesity and insulin resistance. Nature 420:333–336[CrossRef][Medline]
  13. Broussard SR, McCusker RH, Novakofski JE, Strle K, Shen WH, Johnson RW, Freund GG, Dantzer R, Kelley KW 2003 Cytokine-hormone interactions: tumor necrosis factor {alpha} impairs biologic activity and downstream activation signals of the insulin-like growth factor I receptor in myoblasts. Endocrinology 144:2988–2996[Abstract/Free Full Text]
  14. Broussard SR, McCusker RH, Novakofski JE, Strle K, Shen WH, Johnson RW, Dantzer R, Kelley KW 2004 IL-1ß impairs insulin-like growth factor I-induced differentiation and downstream activation signals of the insulin-like growth factor I receptor in myoblasts. J Immunol 172:7713–7720[Abstract/Free Full Text]
  15. Buckingham M, Bajard L, Chang T, Daubas P, Hadchouel J, Meilhac S, Montarras D, Rocancourt D, Relaix F 2003 The formation of skeletal muscle: from somite to limb. J Anat 202:59–68[CrossRef][Medline]
  16. Rudnicki MA, Jaenisch R 1995 The MyoD family of transcription factors and skeletal myogenesis. Bioessays 17:203–209[CrossRef][Medline]
  17. Shen WH, Zhou JH, Broussard SR, Freund GG, Dantzer R, Kelley KW 2002 Proinflammatory cytokines block growth of breast cancer cells by impairing signals from a growth factor receptor. Cancer Res 62:4746–4756[Abstract/Free Full Text]
  18. Shen WH, Yin Y, Broussard SR, McCusker RH, Freund GG, Dantzer R, Kelley KW 2004 Tumor necrosis factor {alpha} inhibits cyclin A expression and retinoblastoma hyperphosphorylation triggered by insulin-like growth factor-I induction of new E2F-1 synthesis. J Biol Chem 279:7438–7446[Abstract/Free Full Text]
  19. Haughey NJ, Cutler RG, Tamara A, McArthur JC, Vargas DL, Pardo CA, Turchan J, Nath A, Mattson MP 2004 Perturbation of sphingolipid metabolism and ceramide production in HIV-dementia. Ann Neurol 55:257–267[CrossRef][Medline]
  20. Cutler RG, Mattson MP 2001 Sphingomyelin and ceramide as regulators of development and lifespan. Mech Ageing Dev 122:895–908[CrossRef][Medline]
  21. Schmitz-Peiffer C 2000 Signalling aspects of insulin resistance in skeletal muscle: mechanisms induced by lipid oversupply. Cell Signal 12:583–594[CrossRef][Medline]
  22. Schutze S, Potthoff K, Machleidt T, Berkovic D, Wiegmann K, Kronke M 1992 TNF activates NF-{kappa}B by phosphatidylcholine-specific phospholipase C-induced "acidic" sphingomyelin breakdown. Cell 71:765–776[CrossRef][Medline]
  23. Wiegmann K, Schutze S, Machleidt T, Witte D, Kronke M 1994 Functional dichotomy of neutral and acidic sphingomyelinases in tumor necrosis factor signaling. Cell 78:1005–1015[CrossRef][Medline]
  24. Adam-Klages S, Adam D, Wiegmann K, Struve S, Kolanus W, Schneider-Mergener J, Kronke M 1996 FAN, a novel WD-repeat protein, couples the p55 TNF-receptor to neutral sphingomyelinase. Cell 86:937–947[CrossRef][Medline]
  25. Dbaibo GS, El-Assaad W, Krikorian A, Liu B, Diab K, Idriss NZ, El-Sabban M, Driscoll TA, Perry DK, Hannun YA 2001 Ceramide generation by two distinct pathways in tumor necrosis factor {alpha}-induced cell death. FEBS Lett 503:7–12[CrossRef][Medline]
  26. Xu J, Yeh CH, Chen S, He L, Sensi SL, Canzoniero LM, Choi DW, Hsu CY 1998 Involvement of de novo ceramide biosynthesis in tumor necrosis factor-{alpha}/cycloheximide-induced cerebral endothelial cell death. J Biol Chem 273:16521–16526[Abstract/Free