Sphingomyelinase stimulates oxidant signaling to weaken skeletal muscle and promote fatigue

Leonardo F. Ferreira, Jennifer S. Moylan, Laura A. A. Gilliam, Jeffrey D. Smith, Mariana Nikolova-Karakashian, Michael B. Reid


Sphingomyelinase (SMase) hydrolyzes membrane sphingomyelin into ceramide, which increases oxidants in nonmuscle cells. Serum SMase activity is elevated in sepsis and heart failure, conditions where muscle oxidants are increased, maximal muscle force is diminished, and fatigue is accelerated. We tested the hypotheses that exogenous SMase and accumulation of ceramide in muscle increases oxidants in muscle cells, depresses specific force of unfatigued muscle, and accelerates the fatigue process. We also anticipated that the antioxidant N-acetylcysteine (NAC) would prevent SMase effects on muscle function. We studied the responses of C2C12 myotubes and mouse diaphragm to SMase treatment in vitro. We observed that SMase caused a 2.8-fold increase in total ceramide levels in myotubes. Exogenous ceramide and SMase elevated oxidant activity in C2C12 myotubes by 15–35% (P < 0.05) and in diaphragm muscle fiber bundles by 58–120% (P < 0.05). The SMase-induced increase in diaphragm oxidant activity was prevented by NAC. Exogenous ceramide depressed diaphragm force by 55% (P < 0.05), while SMase depressed maximal force by 30% (P < 0.05) and accelerated fatigue—effects opposed by treatment with NAC. In conclusion, our findings suggest that SMase stimulates a ceramide-oxidant signaling pathway that results in muscle weakness and fatigue.

  • exercise
  • oxidative stress
  • ceramide
  • diaphragm
  • inflammation

sphingomyelinase (SMase) hydrolyzes sphingomyelins (SM) in cell membranes to generate ceramides, a family of signaling molecules that induce cell death, inflammation, and growth arrest (35). Low levels of SMase activity are detectable in serum of humans (43) and mice (44, 53) and are attributed to a secretory form of the acid SMase (smpd1). Serum SMase is secreted by macrophages and vascular endothelial cells (44) in response to tumor necrosis factor-α or TNF-α (53), a proinflammatory cytokine. Serum SMase can act either on SM in circulating lipoproteins or on SM at the extracellular leaflet of the plasma membrane of distant cells. Serum SMase activity is elevated during systemic inflammation (53), cancer radiotherapy (43), and chronic heart failure (11). Possible consequences of these increases in serum SMase activity for skeletal muscle have not been examined systematically.

Intriguingly, the pathophysiological conditions in which serum SMase activity is elevated are also known to weaken muscle and exacerbate fatigue. During chronic heart failure, for example, serum SMase activity correlates with patient weakness independent of muscle atrophy (11). The intracellular product of serum SMase activity, ceramide, stimulates the production of cellular-derived oxidants—reactive oxygen species (ROS) and reactive nitrogen species (RNS)—in a cell type-specific manner (7, 9, 17, 38, 52, 54). Oxidant production can be a key element of sphingolipid signaling in nonmuscle cells (52). In skeletal muscle, oxidant signaling can depress force and accelerate fatigue (1, 4, 25, 37, 46, 49). These observations suggest that circulating SMase might act on skeletal muscle to depress contractile function.

As a first step to understanding the effects of SMase on muscle, we tested three hypotheses: 1) SMase exposure increases ceramide content and oxidant activity in skeletal muscle cells; 2) SMase/ceramide signaling depresses specific force and promotes fatigue; and 3) increased oxidant activity is essential for the loss of muscle function stimulated by SMase (Fig. 1). The effects of SMase exposure on muscle ceramide content, cytosolic oxidant activity, specific force, and fatigue characteristics were examined in mature C2C12 myotubes and murine diaphragm fiber bundles under in vitro conditions that excluded SMase effects on blood flow, circulating mediators, or neural activation.

Fig. 1.

Schematic representation of hypotheses tested. Steps from sphingomyelin/sphingomyelinase (SMase) to oxidants are proposed on the basis of studies in nonmuscle cells (12, 19, 30, 52), while oxidants cause skeletal muscle weakness and fatigue (1, 4, 25, 37, 46, 49). Oxidants include reactive oxygen and reactive nitrogen species. NAC, N-acetylcysteine.


Cell culture.

C2C12 myoblasts (American Type Culture Collection, Manassas, VA) were grown in Dulbecco's modified Eagle's medium (DMEM) containing 100 μg/ml of streptomycin, 100 U/ml of penicillin (Sigma), and 10% fetal bovine serum. Myoblasts were stimulated to differentiate 48 h after being plated (∼80% confluent) by switching the cells to DMEM supplemented with antibiotics and 2% heat-inactivated horse serum. Cells were maintained in this medium for 5–6 days to enable formation of differentiated, multinucleated myotubes. Early cellular responses to SMase or SMase plus N-acetylcysteine (NAC) were tested by exposing myotubes to experimental solutions for 30 min before collection. We observed no morphologic evidence of cell injury or death (shortening, rounding, altered distribution, bright field intensity, detachment) under these conditions. Similarly, preliminary studies showed that SMase exposure for 72 h had no effect on myotube myosin heavy chain content, actin content, or total protein content (data not shown).

Animal care.

Experiments were performed using 62 male C57BL6/J mice (6–8 wk old; Harlan, Indianapolis, IN). Mice were maintained in a 12:12-h dark-light cycle and received water and food ad libitum. Each animal was deeply anesthetized by inhalation of isoflurane (Aerrane, Baxter Healthcare, Deerfield, IL) and killed by exsanguination followed by cervical dislocation. All procedures conformed to the guiding principles for use and care of laboratory animals of the American Physiological Society and were approved by the Institutional Animal Care and Use Committee of the University of Kentucky.

Diaphragm preparation.

The diaphragm muscle was quickly excised and placed in Krebs-Ringer solution (in mM: 137 NaCl, 5 KCl, 1 MgSO4, 1 NaH2PO4, 24 NaHCO3, and 2 CaCl2) equilibrated with 95% O2-5% CO2 (pH ∼7.4). Muscle fiber bundles were surgically isolated with portions of the rib and central tendon attached for measurements of oxidant activity and contractile function.

Experimental solutions.

Experimental preparations were exposed to 0.1–0.5 U/ml bacterial SMase (Staphylococcus aureus; Sigma-Aldrich, St. Louis, MO) and 10–20 μM C2- or C6-ceramide (Biomol International, Plymouth Meeting, PA). These final concentrations are consistent with previous studies testing the effects of SMase and ceramide in vitro (6, 18). The stock solution of ceramide consisted of a ceramide-bovine serum albumin (BSA) complex (1:1 mol in 2% ethanol vol/vol). Control preparations were treated with equal final concentrations of the appropriate vehicle (SMase, <0.2% glycerol; ceramide, BSA in 0.02% ethanol vol/vol) to control for nonspecific effects. The vehicle solutions did not alter oxidant activity or force production (data not shown). In diaphragm bundles, we used the antioxidant NAC (10 mM; Sigma-Aldrich) to test the role of oxidants in signaling events downstream of SMase. We also used ng-nitro-l-arginine methyl ester (l-NAME, 10 mM; Sigma-Aldrich) to inhibit nitric oxide synthases, thereby testing the role of RNS as mediators of SMase action. NAC and l-NAME were dissolved in Krebs-Ringer buffer. Muscles were treated with NAC or l-NAME for 10 min before addition of SMase to the muscle-bathing solution. All solutions were prepared on the day of the experiment.

Sphingolipid analyses.

C2C12 myotubes (0.5 mg cell wet wt) were harvested in methanol. The lipid extracts were dried under vacuum and shipped on dry ice to the Lipidomics Core at the Medical University of South Carolina for further extraction and analysis of sphingolipid content. Sphingolipids were separated by high-performance liquid chromatography, introduced into the electrospray ionization source, and then analyzed by tandem mass spectrometry using a TSQ 7000 triple quadrupole mass spectrometer (Thermo-Fisher Scientific) as described previously (45).

Oxidant activity assays.

The fluorochrome probe 2′,7′-dichlorofluorescin diacetate (DCFH-DA; Molecular Probes, Eugene, OR) was used to measure overall oxidant activity (33, 41), and 4,5-diaminofluorescein diacetate (DAF-DA) was used to measure RNS activity (27) in the cytoplasm of intact cells. Differentiated C2C12 myotubes were loaded with DCFH-DA (10 μM) or DAF-DA (5 μM) for 30 min at 37°C. For diaphragm, we used 20 μM DCFH-DA. During the loading period, myotubes and diaphragm preparations were exposed to vehicle, ceramides, or SMase. We also tested the effect of NAC treatment on oxidant activity elicited by SMase in diaphragms. In these experiments, muscles were placed in NAC or buffer control (Krebs-Ringer) at 37°C for 10 min before the addition of DCFH-DA and vehicle or SMase to the incubation dish. Accumulation of the oxidized derivative (DCF or DAF; 480-nm excitation, 520-nm emission) was measured in myotubes and fiber bundles (area ∼0.27 mm2). Background emissions were determined from a myotube- or muscle-free area of the dish. Measurements were made using an epifluorescence microscope (Nikon Eclipse TE2000; Nikon Instruments, Melville, NY), a charge-coupled device camera (Cool Snap-ES, Photometrics, Tucson, AZ), and a computer-controlled shutter in the excitation light pathway. Final values for DCF and DAF fluorescence intensity were corrected for background emissions.

Contractile studies.

Diaphragm fiber bundles were mounted for study in vitro. The rib of each fiber bundle was tied to a glass rod, and the central tendon was attached to a force transducer (BG Series 100g, Kulite, Leonia, NJ) using silk suture (4-0). The force transducer was mounted on a micrometer, which was used to adjust muscle length. The muscle was placed between platinum plate electrodes in a water-jacketed organ bath containing buffer continuously gassed with 95% O2-5% CO2. The diaphragm strip was positioned at the length that elicited the highest twitch force (optimal length, Lo).

Contractile characteristics were determined by electrical stimulation using supramaximal voltage (Grass S48, Quincy, MA) and pulse and train durations of 0.3 and 300 ms, respectively. Muscles were then treated with experimental solutions (see above) at 37°C. Experiments to test the effects of C6-ceramide were performed by discontinuing O2-CO2 flow at the time of ceramide administration to avoid foaming of the solution. In C6-ceramide experiments, we assessed changes in maximal tetanic force (Po) by delivering 300-Hz stimuli at 15 and 30 min after exposure to experimental solutions. To test effects of SMase, the solution was gassed with O2-CO2 throughout the experiment. We determined the time course of effects of SMase by delivering 300-Hz stimuli at 5, 10, 15, 30, 45, and 60 min after the onset of treatment.

To define force-frequency characteristics, we measured force in response to stimulus frequencies of 1, 15, 30, 50, 80, 120, and 150 every 2 min—interspersed by maximal, 300-Hz stimulations—starting 60 min after the onset of treatment with experimental solutions. Peak twitch tension (Pt), time to peak tension (TPT), and one-half relaxation time (1/2-RT) were measured in response to 1-Hz stimulation. We also calculated the Pt-to-TPT (dP/dt) and Pt-to-Po (Pt:Po) ratios to study twitch characteristics. All measurements were performed in solutions containing d-tubocurarine (0.025 mM).

One minute after the force-frequency protocol, we started a fatigue protocol consisting of tetanic contractions every 2 s for 600 s (stimulation parameters: frequency 40 Hz, train duration 500 ms, pulse duration 0.3 ms). On the basis of the preliminary findings, we also tested fatigue of muscles exposed to SMase and NAC/SMase using higher stimulus frequencies (see results). Frequency was adjusted to match the initial forces of treated muscles to the initial forces developed by control muscles [“matched-initial force” protocol (25)]. For analysis of matched-initial force, we used data from muscles with initial force within the mean ± 2 SD of controls. Data were excluded when this criterion was not met.

For all measurements of muscle force, the output of the force transducer was monitored using a digital oscilloscope (546601B; Hewlett Packard, Palo Alto, CA). After the end of the fatigue protocol, muscle length was measured using a hand-held electronic caliper (CD-6“ CS, Mitutoyo America, Aurora, IL). The fiber bundle was then removed from the organ bath, trimmed of bone and connective tissue, blotted dry, and weighed. Muscle weight and Lo were used to estimate cross-sectional area (5). Absolute forces are expressed as N/cm2. Forces measured during the fatigue protocol were also normalized for the initial force of the trial (time 0). During the fatigue protocol, unstimulated force was determined by measuring baseline force immediately before a stimulus train.

Statistical analysis.

All comparisons were performed using commercially available software (Prism 5.0b, GraphPad Software, La Jolla, CA). For clarity, statistical tests are described in results following each comparison. Data are presented as means ± SE unless stated otherwise. Significance was accepted at the P < 0.05 level.


Sphingolipid analyses.

Ceramide subspecies measured in mature C2C12 myotubes under control conditions are shown in Table 1. As expected (29), C16-Cer was the most abundant species followed by C24:1-Cer. The content of C20:4-ceramide was below the detection level of our assay. SMase exposure increased the content of C14-, C16-, C16-dihydroceramide-, C18-, C18:1-, C20:1-, C24-, C24:1-, and C26:1-ceramide, yielding a 2.8-fold increase in total ceramide content (Fig. 2). With the exception of C14-ceramide, which was elevated 15-fold, the relative increases in individual species ranged between 3- and 6-fold. The net increase in ceramide content was due mainly to the accumulation of C16-Cer and C24:1-Cer.

View this table:
Table 1.

Content of ceramide subspecies in control myotubes

Fig. 2.

Increase in myotube ceramide content after SMase exposure. Left: total myotube ceramide content normalized for lipid phosphate (Pi) levels (control, n = 5; SMase, n = 3). Right: fold change in ceramide subspecies (C14 to C26 carbon chain length) in response to SMase exposure (0.5 U/ml). dh, Dihydroceramide. Data are means ± SE. Statistical analysis: unpaired t-test (*P < 0.05).

Ceramides can be metabolized to free sphingoid bases by ceramidases; sphingoid bases can be further phosphorylated to (dihydro) sphingosine-1-phosphates by sphingosine kinases. These molecules may function as secondary mediators of SMase signaling (36). The total content of free and phosphorylated sphingoid bases in mature myotubes (119 ± 10 pmol/pmol Pi) was far less than total ceramide content of ∼6%. Sphingosine (58 ± 6) and sphingosine-1-phosphate (41 ± 12) were the most abundant species, while their saturated counterparts dihydrosphingosine (7 ± 1) and dihydrosphingosine-1-phosphate (12 ± 4) were detected at much lower levels. SMase treatment did not change total content of the free sphingoid bases and their phospho-derivatives (142 ± 9) or the composition of individual sphingoid species (data not shown). Accordingly, subsequent evaluations of muscle function were restricted to studies of SMase and ceramide effects.

Cytosolic oxidant activity.

Exposure to SMase for 30 min increased cytosolic oxidant activity in mature myotubes as reflected by DCF fluorescence [control, 136 ± 5 arbitrary units (AU); SMase, 182 ± 10; P < 0.05]. Data in Fig. 3 show that direct exposure to ceramide subspecies had similar effects. DCF fluorescence was increased by exposure to either C2-ceramide (control, 96 ± 4; treated, 112 ± 6; P < 0.05), or C6-ceramide (control, 56 ± 2; treated, 70 ± 3; P < 0.05). In contrast, SMase exposure had no effect on DAF fluorescence (control, 179 ± 4 AU; SMase, 180 ± 4; P > 0.05, n = 6/group), suggesting that cytosolic RNS activity was unaltered.

Fig. 3.

C2C12 myotube oxidant activity is increased by ceramide and SMase. Myotubes were exposed to vehicle (control), C2-ceramide (C2-Cer; 10 μM, n = 11/group), C6-ceramide (C6-Cer; 10 μM, n = 12/group), and SMase (0.5 U/ml, n = 12/group) for 30 min at 37°C. Bars represent percentage change in arbitrary fluorescence units [%ΔAFU; (treated − control) × 100/control]. Images are representative of DCF fluorescence in myotubes collected at ×10 magnification. Images of treated myotubes are paired with respective controls. Statistical analysis: paired t-test for arbitrary fluorescence units from treated vs. respective controls (*P < 0.05).

In diaphragm muscle fibers, SMase exposure increased cytosolic oxidant activity from 28 ± 2.5 AU (control) to 62 ± 5.6 AU (SMase), an effect abolished by NAC pretreatment (24 ± 1.8 AU; Fig. 4). C6-ceramide increased diaphragm cytosolic oxidant activity from 136 ± 11 AU to 216 ± 31 AU. These findings suggest that myotubes and muscle fibers have homologous responses to SMase. They also identify NAC as an experimental tool for inhibiting oxidant-mediated effects of SMase on muscle function (see below).

Fig. 4.

SMase and ceramide increase diaphragm oxidant activity. Left: representative images of DCF fluorescence in diaphragm muscle fibers (×10 magnification). Middle: arbitrary units of DCF fluorescence normalized for mean of control group. SMase effect was prevented by the antioxidant NAC. Muscles were exposed to vehicle (control, n = 4), SMase (0.5 U/ml, n = 9), or NAC/SMase (10 mM, n = 5) for 30 min at 37°C. Statistical analysis: one-way ANOVA and Bonferroni post hoc test. *P < 0.01 vs. control and NAC/SMase. Right: arbitrary units of DCF fluorescence normalized for control in diaphragm muscles treated with vehicle (control, n = 5) and C6-ceramide (20 μM, n = 5) for 10 min; images not shown. *P < 0.05 by paired t-test.

Contractile function of unfatigued muscle.

Direct exposure of diaphragm fiber bundles to C6-ceramide stimulated a progressive decrease in maximal specific force (Po) within 30 min (Fig. 5). Data in Fig. 6 illustrate the effects of SMase exposure on Po. A time-dependent decline was evident within 30 min and persisted for at least 60 min. Decrements in Po were also dose dependent after 60-min exposures to SMase of 0.1 (−18%), 0.25 (−31%), and 0.5 U/ml (−33%).

Fig. 5.

Exogenous ceramide depresses maximal force (Po) of diaphragm strips. Maximal force measured immediately before (0 min) and after 15 and 30 min exposure to vehicle (open bars; n = 3) and C6-ceramide (closed bars; n = 3) at 37°C. Specific forces of control muscles at 0, 15, and 30 min were 18.1 ± 2.4, 22.9 ± 1.3, and 20.4 ± 0.2 N/cm2, respectively. Statistical analysis: two-way repeated-measures ANOVA and Bonferroni post hoc test; *P < 0.05.

Fig. 6.

Time- and concentration-dependent effects of exogenous SMase on maximal force of diaphragm strips. Open circle, control; solid square, SMase. Top: force measured before (room temperature) and during exposure (37°C) to vehicle (n = 4 mice) or SMase 0.5 U/ml (n = 4 mice). Dotted line indicates onset of exposure to vehicle or SMase and change in organ bath temperature to 37°C. Bottom: force after 60 min of exposure to vehicle (control, n = 6) or SMase 0.1 (n = 3), 0.25 (n = 4), and 0.5 U/ml (n = 19). Statistical analysis: top, two-way repeated-measures ANOVA and Bonferroni post hoc test; bottom, one-way ANOVA and Dunnett's post hoc test. *P < 0.05 for control vs. SMase.

Table 2 depicts changes in twitch characteristics. SMase exposure decreased Pt and twitch dP/dt. NAC pretreatment prevented these decrements. Other twitch properties (TPT, 1/2 RT, Pt:Po) were unaffected by SMase.

View this table:
Table 2.

Twitch characteristics of diaphragm muscles

As shown in Fig. 7, the specific force-frequency relationship was depressed by SMase exposure. SMase-treated fiber bundles developed 30% less specific force than controls at stimulus frequencies of 30 to 300 Hz. NAC pretreatment blunted this decrement by one-half, identifying oxidant activity as a mediator of SMase effects on specific force. Normalizing force for Po abolished differences among groups; the relative force-frequency relationships were essentially identical. This protocol involved repetitive measurements of maximal tetanic force. These values did not decline over time (overall changes: control −1.5 ± 1.5%, SMase −2.6 ± 1.8, NAC/SMase −0.1 ± 2.2; P > 0.05, paired t-test), which confirms stability of the fiber bundles.

Fig. 7.

Force-frequency characteristics of diaphragm muscles. Specific forces are expressed normalized for cross-sectional area (N/cm2) or peak tetanic force (%Po). Open circle, control (n = 6); solid square, SMase (0.5 U/ml; n = 19); shaded square, NAC/SMase (10 mM, n = 7). Solid and dotted lines are best regression fit of mean data using Hill equation. Statistical analysis: two-way repeated-measures ANOVA with Bonferroni post hoc test. *P < 0.05 for control vs. SMase; θP < 0.05 for NAC/SMase vs. control or SMase.

To test for RNS involvement, a fourth group of fiber bundles were pretreated with 10 mM l-NAME [nitric oxide (NO) synthase inhibitor (40)] before SMase exposure. l-NAME had no effect on SMase-induced weakness at any stimulus frequency, a finding illustrated by mean Po values: l-NAME/SMase 14.6 ± 1.6 N/cm2 vs. 16.3 ± 0.5 N/cm2 SMase alone. This result is consistent with our earlier observation that NO activity is unaffected by SMase and suggests that RNS do not mediate SMase effects on force.


Our fatigue protocol elicited an exponential decrease in specific force. In control muscles stimulated at 40 Hz, specific force decreased to 36 ± 2% of the initial value after 600 s of repetitive contractions (Fig. 8, top). SMase depressed specific force evoked by 40-Hz stimulation at every time point in the fatigue trial.

Fig. 8.

Diaphragm fatigue characteristics during repetitive contractions. Top: specific force during matched-frequency protocol in control (open circle, n = 6) and SMase (solid square, n = 8). Middle: specific force during matched-force protocol in SMase (71 ± 3 Hz, n = 7; solid square) and NAC/SMase (45 ± 3 Hz, n = 5; shaded square). Mean control data (top) are shown by dotted line for reference. Bottom: unstimulated force during matched-force protocol. Data from 420 to 600 s were similar to 360 s and are omitted for clarity. Statistical analysis: two-way repeated-measures ANOVA and Bonferroni post hoc test. *P < 0.05 for control vs. SMase; ϕP < 0.05 NAC/SMase vs. SMase, #P < 0.05 control vs. NAC/SMase.

In our matched-force protocol, SMase-treated muscles were stimulated at 71 ± 3 Hz to evoke the same specific force as control muscles at time 0 (Fig. 8, middle). After 30 contractions (time 60 s), SMase-treated muscles developed less specific force than controls. NAC/SMase-treated muscles were stimulated at 45 ± 3 Hz. They developed initial specific forces and displayed fatigue characteristics that were indistinguishable from controls.

After 2 min of fatiguing contractions, control muscles exhibited incomplete relaxation between contractions (Fig. 8, bottom). The specific force sustained by unstimulated muscle was decreased by SMase. NAC pretreatment did not alter this response.


Our main findings are that ex vivo exposure of myotubes and mouse diaphragm to SMase increases cellular ceramide content and cytosolic oxidant activity, weakening the diaphragm and accelerating fatigue. Ceramide analogs mimicked the effects of SMase on oxidant activity and specific force. The decrements in contractile function elicited by SMase were either blunted or abolished by treatment with NAC. These findings identify extracellular SMase as a potential regulator of muscle function and muscle-derived oxidants as essential mediators of weakness and fatigue stimulated by SMase.

Sphingolipids in myotubes.

SMase hydrolyzes sphingomyelin into phosphoryl choline and ceramide. Naturally occurring ceramides have fatty acid chain lengths varying from 2 to 28 carbon atoms, with longer-chain fatty acids being more abundant (6). In myotubes, we detected ceramide subspecies with fatty acid residues ranging from C14 to C26, the most prevalent being C16-ceramide (45% of total ceramide). These findings are consistent with previous ceramide analyses in myotubes (23) and skeletal muscle (10, 22).

Our data indicate that exposure to SMase increased 9 of 13 ceramide subspecies in myotubes. C14-ceramide was most affected, as its level increased 15-fold. However, net increases in the content of C16 and C24:1-ceramides accounted for the bulk of ceramide accumulation. To our knowledge, this is the first report of SMase effects on the ceramide profile of a muscle-derived cell type.

Approximately 75% of the total SM in cells is found at the extracellular leaflet of the plasma membrane. The seemingly uniform increase in all subspecies of ceramide indicates that exogenous SMase affects the bulk of SM, rather than a specific pool. In contrast, neutral SMase2 is an intracellular plasma membrane-associated SMase that hydrolyzes SM at the inner leaflet of the plasma membrane and preferentially metabolizes sphingomyelins with longer fatty acid chains, such as C24 (30). The extracellular and intracellular SMase seems to act on different SM pools and consequently may have different cellular functions.

Muscle also contains detectable levels of free and phosphorylated sphingoid bases which are bioactive products of ceramide breakdown. We found that myotubes contain similar levels of sphingosine and sphingosine-1-phosphate, while the steady-state contents of the saturated dihydrosphingosine (sphinganine) and dihydrosphingosine-1-phosphate were much lower. On average, the sum of all free and phosphorylated sphingoid bases is an order of magnitude less than total ceramides, results consistent with prior reports (23, 42). SMase exposure did not increase sphingosine or sphingosine-1-phosphate. These findings suggest that sphingoid bases do not mediate SMase effects in our system, but we cannot exclude a role for local or compartmentalized changes in sphingoid bases in muscle cells.

SMase, ceramide, and oxidant activity.

The effect of SMase/ceramide signaling on cellular redox status appears to be cell type specific. Oxidant activity is increased in oocytes (6) and hepatocytes (16) but decreased in jurkat T cells (38). We found that SMase stimulated cytosolic oxidant activity in both myotubes and intact muscle fibers. This response was reproduced by ceramide exposure and abolished by NAC pretreatment. These novel findings suggest that SMase/ceramide signaling stimulates production of oxidants by muscle.

The pool of ceramide subspecies generated by exogenous SMase is not unique in its capacity to stimulate oxidant production. For example, the de novo pathway generates ceramide in response to palmitic acid (20), and treatment of cultured muscle cells with palmitic acid heightens cytosolic oxidants (13, 28). This suggests that de novo synthesis is a second pathway for ceramide-stimulated oxidants in muscle.

In nonmuscle cells, ceramide enhances ROS production by mitochondria (7, 16) and NADPH oxidase (54) and also stimulates NO synthesis (2, 31). SMase acts on vascular endothelial cells to increase NO production via the endothelial isoform of NO synthase (31). This enzyme is constitutively expressed by skeletal muscle fibers (26). However, fluorescence of the NO-sensitive probe DAF did not change with SMase exposure, arguing that RNS do not contribute to the rise in oxidant activity seen in our study. Rather, the rise in DCF fluorescence likely reflects greater ROS production (33), which is consistent with muscle weakness and fatigue (1, 4, 41, 49).

Contractile depression.

This study is the first to demonstrate that direct SMase or ceramide exposure causes skeletal muscle weakness. C6-ceramide and SMase elicited progressive, monotonic declines in maximal force. Examined in greater detail, we found that SMase depressed specific force across a broad range of twitch and submaximal tetanic stimulus frequencies. These changes were observed in curarized fiber bundles activated via electrical field stimulation. Accordingly, the loss of force is independent of nerve function or neuromuscular coupling. Rather, this weakness must reflect a cellular response of component muscle fibers to SMase/ceramide stimulation.

Muscle-derived oxidants appear to be essential for this response. Both SMase and ceramide elevated cytosolic oxidant activity. More importantly, the loss of specific force elicited by SMase was blunted by NAC, a nonspecific antioxidant (3). SMase does not appear to weaken muscle via RNS signaling since 1) SMase did not increase NO activity in myotubes and 2) blockade of NO synthesis did not alter SMase-induced weakness. The more likely alternative is that SMase effects are mediated by ROS, which are widely reported to depress specific force of skeletal muscle (1, 4, 21, 49, 50).

SMase and fatigue.

Muscles fatigued faster after SMase exposure, an effect abolished by NAC pretreatment. As in SMase-induced weakness, muscle-derived oxidants appear to be downstream signals by which SMase accelerates fatigue. This contention is consistent with studies showing that muscle fatigue is accelerated by oxidants (39) and delayed by antioxidants (8, 14, 25, 32, 41, 48).

The matched-initial force protocol developed by our group (25) is relevant for translational biology. In vivo, the central nervous system increases skeletal muscle activation to achieve the mechanical output required for a given task, e.g., work of breathing for diaphragm muscle. In response to muscle weakness, mechanical output is maintained (in part) by increasing the firing frequency of motor neurons. This physiologic strategy is mimicked in vitro by our matched-initial force protocol. We increased stimulation frequency such that weaker, SMase-exposed muscles developed the same specific force as control muscle at the onset of exercise. The capacity to sustain force during repetitive contractions was clearly compromised by SMase exposure.

Clinical relevance.

Our study was developed as an initial step to elucidate mechanisms of muscle dysfunction in diseases with heightened serum SMase activity. In patients with heart failure, serum SMase activity is correlated with skeletal muscle weakness (11). Sepsis also increases serum SMase activity (53) and causes respiratory muscle weakness (50). In heart failure and sepsis, oxidants are increased in skeletal muscle (24, 34, 51), and diaphragm muscle weakness is blunted by antioxidant treatment (15, 51). These effects are similar to those we found with SMase exposure in vitro. Therefore, our findings identify increased serum SMase activity as a potential mechanism by which heart failure or sepsis might cause skeletal muscle weakness.

Methodological considerations.

This proof-of-concept study tested the effects of direct SMase stimulation on skeletal muscle function. The bacterial SMase used herein has the same enzymatic action and molecular products as the secretory SMase found in patients' serum (30, 44). In vitro experimentation eliminated systemic actions of SMase on neural and cardiovascular targets that might alter muscle function indirectly. SMase activities in our study (6–30 μmol·ml−1·h−1) were chosen on the basis of published reports of previous investigators (6, 18). These values exceed the activities measured in patient serum [∼ 0.4–1 μmol·ml−1·h−1 (11, 43)] to compensate for in vitro conditions, i.e., greater diffusion distances in isolated fiber bundles and shorter exposure times. Importantly, the increase in total ceramide content induced by SMase exposure was within the pathological range measured in humans (43, 47).


Our study demonstrates that SMase exposure increases ceramide levels and cytosolic oxidant activity in skeletal muscle, depressing specific force, and accelerating fatigue. Exogenous ceramide reproduces the effects of SMase on oxidant activity and force, whereas the antioxidant NAC inhibits SMase effects on muscle function. Our data do not support RNS as mediator of SMase action. We conclude that SMase acts via ceramide/oxidant signaling to weaken skeletal muscle and promote fatigue.


This study was supported by National Institutes of Health (NIH) grants to M. B. Reid (R01 AR055974) and M. Nikolova-Karakashian (R01 AG026711). L. F. Ferreira was supported by postdoctoral fellowships from the American Heart Association (GRA 0725334B and 09POST2020082) and L. A. A. Gilliam was supported by an NIH training grant (T32 HL086341).


No conflicts of interest, financial or otherwise, are declared by the authors.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
  50. 50.
  51. 51.
  52. 52.
  53. 53.
  54. 54.
View Abstract