The inducible form of nitric oxide synthase (NOS2) catalyzes the synthesis of nitric oxide (NO) from arginine in response to injury and infection. NOS2 is expressed predominantly by macrophages and lymphocytes. However, skeletal muscle also expresses NOS2 in response to inflammatory stimuli. The present study sought to determine whether lipopolysaccharide (LPS) stimulates NOS2 in skeletal muscle via Toll-like receptor-4 (TLR4). Intraperitoneal injection of LPS in wild-type mice (C3H/HeSnJ) increased NOS2 mRNA fourfold in skeletal muscle, while no change in NOS2 mRNA was observed in C3H/HeJ mice that harbored a mutation in the LPS receptor. NOS2 coimmunoprecipitated with the muscle-specific caveolin-3 protein, suggesting that myofibers per se respond to LPS in vivo. LPS stimulated NOS2 mRNA expression in C2C12 myocytes, and the regulation of NOS2 mRNA was comparable in myoblasts and differentiated myotubes. LPS transiently stimulated the phosphorylation of the interleukin-1 receptor-associated kinase (IRAK-1) in C2C12 cells and decreased the total amount of IRAK-1 both in vitro and in vivo over time. LPS stimulated the expression of an NF-κβ reporter plasmid, and this was inhibited by the proteasomal inhibitor MG-132. Both myoblasts and myotubes expressed TLR2 and TLR4 mRNA. Expression of a dominant negative form of TLR4 in C2C12 cells blocked LPS-induced NF-κβ reporter activity. SP-600125 [a c-Jun NH2-terminal kinase (JNK) inhibitor] also prevented LPS stimulation of NOS2 expression. Moreover, the JNK inhibitor prevented the LPS-induced increase in NO synthesis. These data indicate that LPS increases NOS2 mRNA expression in muscle via a TLR4-dependent mechanism.
- interleukin-1 receptor-associated kinase
- dominant negative
nitric oxide (no) is synthesized from the amino acid arginine in a reaction catalyzed by NO synthase (NOS). NO has numerous physiological functions. NO is a vasodilator for blood vessels (13) and an antiproliferative agent for smooth muscle cells (2). In skeletal muscle, NO increases both blood flow (13) and glucose uptake (8, 16, 20, 23). NO has disparate roles during infection. NO is toxic to infectious organisms but can also react with superoxide to form NO intermediates, such as peroxynitrite (NO3−), that activate the host stress response and may cause tissue injury (34, 35).
Skeletal muscle contains three forms of NOS. Two of these isoforms, NOS1 and NOS3, are constitutively expressed. Because NOS2 is stimulated by bacterial cell wall components, it is also referred to as inducible NOS (iNOS). NOS2 is expressed in skeletal muscle in response to injury and repair. Dystrophin-deficient muscle has a reduction in the constitutive expression of NOS, and it undergoes a repetitive cycle of damage and repair because of the presence of tissue macrophages. This damage can be prevented by expression of a NOS transgene in muscle (45). NO may therefore be cytoprotective in this model of muscle damage. Yet, in mdx mice, there is overexpression of NOS2 that is reversed by somatic gene transfer of dystrophin (32). These data suggest that the NOS enzymes may play distinct roles in muscle damage and repair and that NO may influence both congenital (22) and acquired muscle-wasting diseases (10, 38).
Multiple investigators have examined the expression of NOS2 in the diaphragm and skeletal muscle after administration of lipopolysaccharide (LPS) (6, 25, 43). Boczkowski et al. (6) and Hussain et al. (25) found NOS2 protein to be abundant in skeletal muscle fibers between 6 and 48 h after LPS. In addition, Thompson et al. (43) found that muscle explants incubated with LPS exhibited NOS2 immunostaining in myocytes. Sambe et al. (40) demonstrated that LPS induces contractile dysfunction in respiratory muscles and that diaphragmatic contraction could be restored by either dexamethasone or a NOS inhibitor.
We (17) and others (21) have shown that human and murine myoblasts produce IL-1β, IL-6, and tumor necrosis factor-α (TNFα) in response to various inflammatory stimuli. Furthermore, investigators at our laboratory recently demonstrated that LPS stimulates cytokine expression in murine skeletal muscle via Toll-like receptor-4 (TLR-4) signaling. Mice that harbor a mutation in TLR-4 (C3H/HeJ mice) have greatly reduced expression of IL-6 mRNA in skeletal muscle in response to LPS (17). Because LPS stimulates NOS2 expression in macrophages, we were interested to know whether NOS2 is also a downstream target of TLR4 in skeletal muscle and muscle cells. To address this question, we first challenged wild-type and C3H/HeJ mice with LPS to examine whether skeletal muscle NOS2 expression was TLR4 dependent. Second, LPS was added directly to C2C12 muscle cells to determine whether this stimulus was capable of increasing NOS2 protein and mRNA. A dominant negative form of TLR4 was used to ascertain whether LPS signals through the TLR4 receptor to increase NF-κβ activity in C2C12 cells. We also examined whether NOS2 is regulated at the transcriptional level and whether MAP kinase and stress-activated protein kinases regulate NOS2 expression. Finally, exogenous NO was added to myocytes to demonstrate the presence of a putative negative feedback loop for NOS2 expression.
MATERIALS AND METHODS
Experimental protocol for C3H/HeSnJ mice.
C3H/HeSnJ and C3H/HeJ mice were obtained from Jackson Laboratories (Bar Harbor, ME). All mice were housed in a controlled environment and provided water and rodent chow ad libitum for 3 wk before use. Mice were 8–9 wk old and weighed 21.4 ± 0.3 g at the time of the study. In the experiment depicted in Fig. 1, mice were injected intraperitoneally with LPS derived from Escherichia coli 026:B6 (25 μg/mouse; DIFCO Laboratories, Detroit, MI) or an equal volume of saline (250 μl/mouse). This dose was based on a preliminary dose-response study and is similar to that used by other investigators (3). After 2, 6, and 18 h, mice were anesthetized with a mixture of ketamine (90 mg/kg; Fort Dodge Animal Health, Fort Dodge, IA) and xylazine (9 mg/kg; Bayer, Shawnee Mission, KS). Blood from the inferior vena cava was collected in heparinized syringes. Hindlimb skeletal muscle (gastrocnemius and plantaris) from both legs was dissected from each animal, wrapped in aluminum foil, and frozen in liquid nitrogen. Mice were killed by cardiac excision and subsequent exsanguination. Tissues were later powdered under liquid nitrogen using a mortar and pestle and stored at −70°C.
In a second series of experiments, male Sprague-Dawley rats weighing 175–200 g were injected intraperitoneally with either LPS (1 mg/kg) or saline (0.5 ml/kg). Six hours after LPS administration, rats were anesthetized with pentobarbital. The gastrocnemius muscle was rapidly excised and weighed, and the tissue was frozen between aluminum blocks precooled to the temperature of liquid nitrogen. All experiments were approved by the Institutional Animal Care and Use Committee at the Pennsylvania State University College of Medicine and adhered to National Institutes of Health guidelines for the use of experimental animals.
The C2C12 murine myoblast cell line used for all studies was purchased from the American Type Culture Collection (Manassas, VA). Cells were grown in 100-mm petri dishes (Becton Dickinson, Franklin Lakes, NJ) and cultured in Eagle's minimum essential medium containing 10% bovine calf serum (BCS), penicillin (100 U/ml), streptomycin (100 μg/ml), and amphotericin B (25 μg/ml) (all purchased from Sigma, St. Louis, MO). Cells were grown to confluence and switched to fresh serum-containing medium before LPS, cytokines, or other agents were added. C2C12 cells were used at the myoblast stage. In some experiments, the cells were switched to medium containing 2% serum and allowed to differentiate into myotubes. Experiments were performed with E. coli LPS (026:B6; DIFCO Laboratories). A variety of compounds purchased from Calbiochem (La Jolla, CA), including SP-600125, PD-98059, SB-202190, MG-132, cycloheximide, and 5,6-dichloro-1-β-d-ribofuranosylbenzimidazole (DRB), were used to characterize the response to LPS.
Transient transfection assays.
C2C12 cells were plated at 50% confluence in 24-well plates. Cells were switched to serum-free medium and transfected with a pNFκβ-Luc reporter vector (BD Biosciences, Palo Alto, CA) and pSV-β-galactosidase control vector (Promega, Madison, WI). Both plasmids were added as a preformed complex with Lipofectamine 2000 at a 5:1 ratio of lipid to DNA. After 2 h, cells were allowed to recover in serum-containing medium for 16 h. Cell extracts were isolated at various times after the addition of LPS in reporter lysis buffer (Promega) and frozen until assay. In some transfection studies, the cells also received a third plasmid expressing a dominant negative form of the TLR4 receptor (pZERO-mTLR4; InvivoGen, San Diego, CA) in which the Toll interleukin-1 receptor (TIR) domain was deleted from the murine TLR4 gene. Luciferase reporter activity was measured with firefly luciferase assay reagents (Promega) on a Turner Biosystems luminometer (Sunnyvale, CA). β-Galactosidase activity was measured with a commercially available kit (Promega) and used to normalize for transfection efficiency.
RNA isolation and ribonuclease protection assay.
Total RNA, DNA, and protein were extracted from C2C12 cells or tissues in a mixture of phenol and guanidine thiocyanate (TriReagent; Molecular Research Center, Cincinnati, OH) according to the manufacturer's protocol. RNA was separated from protein and DNA by the addition of bromochloropropane and precipitation in isopropanol. After being washed in 75% ethanol and resuspended in formamide, RNA samples were quantified by spectrophotometry. RNA (10 μg) was used for each assay. Riboprobes were synthesized from a custom multiprobe mouse template set containing a probe for NOS2, IL-6, and suppressor of cytokine signaling-3 (SOCS-3) mRNA detection (Pharmingen, San Diego, CA). The labeled riboprobe was hybridized with RNA overnight using a ribonuclease protection assay (RPA) kit according to the manufacturer's protocol (Pharmingen). Protected RNA were separated using a 5% acrylamide gel (19:1 dilution of acrylamide to bisacrylamide). Gels were transferred to blotting paper and dried under vacuum on a gel dryer. Dried gels were exposed to a phospho-imager screen (Molecular Dynamics, Sunnyvale, CA), and the resulting data were quantified using ImageQuant software and normalized to the murine ribosomal protein L32 mRNA signal in each lane.
Immunohistochemistry and nitrate measurements.
NOS2 and myocyte enhancer factor (MEF)-2 were detected by performing immunohistochemistry. Briefly, cells were either grown in 10% BCS alone (control) or stimulated with LPS for 6 h and then fixed in 100% methanol for 10 min at −20°C. Fixed cells were washed and blocked with normal goat serum before primary antibodies were added to either NOS2 or MEF-2 (Santa Cruz Biotechnology, Santa Cruz, CA). The specificity of each antibody was determined by exclusion of the primary antibody. Antibodies were detected by adding a complex of biotin-labeled anti-rabbit IgG and horseradish peroxidase (HRP)-labeled avidin (Vectastain Elite; Vector Laboratories, Burlingame, CA). Slides were incubated for equal amounts of time with the HRP substrate 3-amino-9-ethylcarbazole (AEC). AEC formed a red-colored reaction product to demark the amount and position of each antigen.
For the detection of NO, conditioned medium from the cells was incubated with nitrate reductase to convert nitrate to nitrite, and total nitrite was measured colorimetrically as a colored azo dye product of the Griess reaction in a mircrotiter plate at 540 nm. The nitrite concentration of triplicate tissue culture wells was determined by comparing the absorbance of samples against a nitrite standard curve as supplied in a kit purchased from Stressgen Biotechnology (Victoria, BC, Canada).
Western blot analysis.
Cell extracts were electrophoresed onto denaturing polyacrylamide gels and electrophoretically transferred to nitrocellulose with a semidry blotter (Bio-Rad Laboratories, Melville, NY). The resulting blots were blocked with 5% nonfat dry milk for 1.5 h and incubated with antibodies against NOS2, interleukin-1 receptor-associated kinase (IRAK-1), myosin heavy chain (MHC), or Erk kinase as previously described (18, 19). Unbound primary antibody was removed by washing with Tris-buffered saline containing 0.05% Tween 20, and blots were incubated with anti-rabbit or anti-mouse immunoglobulin conjugated with HRP. Blots were briefly incubated with the components of an enhanced chemiluminescence detection system (Amersham, Little Chalfont, UK). Dried blots were used to expose x-ray film for 1–3 min.
Values are means ± SE. Unless otherwise noted, each experimental condition was tested in triplicate and each experiment was repeated twice. Data were analyzed using analysis of variance followed by Student-Newman-Keuls test. Statistical significance was set at P < 0.05. For animal studies, we used four mice in the control group and six in the LPS group. NOS2 mRNA half-life was calculated from the slope of the regression line using the formula t1/2 = 0.5/m, where m is the slope of the line in arbitrary units per hour. Half-lives were compared using a t-test where t = (m1 − m2)/[)]. Statistical significance was set at P < 0.05.
LPS-induced NOS2 expression in murine skeletal muscle was TLR4 dependent.
Mice underwent injection with a nonlethal dose of LPS, and skeletal muscle was assayed for NOS2 mRNA by performing RPA after 2, 6, and 18 h. LPS increased NOS2 mRNA in skeletal muscle fourfold compared with control mice (Fig. 1A, WT/LPS and WT/Sal). The basal expression of NOS2 mRNA did not differ between C3H/HeSnJ (WT/Sal) and C3H/HeJ mice (HeJ/Sal) with a mutation in the TLR4 receptor. LPS failed to increase NOS2 mRNA expression in C3H/HeJ mice, suggesting that the TLR4 receptor is required for the LPS response (Fig. 1A). LPS did not simply increase the steady-state level of muscle mRNA in general. The expression of L32 mRNA was unaltered by LPS. Stimulation of NOS2 mRNA expression by LPS was transient, with maximal expression occurring at 6 h (Fig. 1B). The inhibitor of NF-κβ (IκBα) mRNA was also elevated fourfold in skeletal muscle by LPS, and the response to LPS was TLR4 dependent (Fig. 1C).
LPS also stimulated NOS2 protein expression in rat skeletal muscle 6 h after LPS but not saline administration (Fig. 2A). NOS2 coimmunoprecipitated with the muscle-specific caveolin, caveolin-3, in muscle homogenates when rats underwent injection with LPS. LPS did not alter the total amount of caveolin-3 in muscle homogenates (Fig. 2B). Differentiated C2C12 myotubes also expressed NOS2 protein as detected using immunohistochemistry. Myotubes exhibited diffuse staining throughout the cytoplasm (Fig. 2C, middle). The NOS2 pattern of expression was different from that observed for MEF2, a transcription factor that exhibits predominantly nuclear localization (Fig. 2C, right). Immunostaining for both antigens was dependent on the presence of the primary detection antibodies.
LPS stimulated NOS2 mRNA dose and time dependently in C2C12 cells.
Skeletal muscle contains many cell types that can potentially express NOS2. Therefore, we examined whether LPS directly increases NOS2 mRNA in C2C12 myoblasts, a clonal cell line derived from murine skeletal muscle. LPS stimulated NOS2 mRNA dose dependently (Fig. 2, A and B). Maximal NOS2 mRNA expression in C2C12 cells occurred with 1 μg/ml of LPS and 4 h after exposure of myocytes to LPS (Fig. 3 and Fig. 4, A and B). NOS2 protein was also increased dose dependently by LPS as determined by performing Western blotting (Fig. 3C). LPS stimulated NOS2 mRNA expression in both myoblasts and differentiated myotubes (Fig. 4, A and B). NOS2 expression was also increased by peptidoglycan from Staphylococcus aureus, and NOS2 expression in response to this agent followed a time course comparable to that observed after LPS (data not shown). NOS2 protein expression demonstrated temporal progression similar to NOS mRNA expression, albeit slightly delayed. There was comparable expression of the NOS2 protein in both myoblasts and myotubes (Fig. 4C). Myotube formation was confirmed by greater expression of MHC in the myotubes (Fig. 4C).
LPS stimulated activation of IRAK in C2C12 cells and skeletal muscle.
Although C2C12 cells have been shown to respond to LPS, little is known about the early signaling events that increase the expression of NOS2 and cytokines in muscle cells. In addition, because most of the responses to LPS are transient, it is also not known how the LPS signal is terminated. We investigated whether LPS activates IRAK in C2C12 cells. LPS stimulated transient phosphorylation of IRAK-1 on serine 376 (Fig. 5A). IRAK phosphorylation occurred in as little as 1 min after LPS addition and persisted for 20 min. By 30 min, IRAK phosphorylation had returned to control levels. In the short term, these changes were independent of changes in the total amount of IRAK. In contrast, exposure of C2C12 cells to LPS for a longer period of time (2 h) decreased the total amount of IRAK in cell extracts (Fig. 5B). These changes were independent of changes in the total amount of protein in the cells, as evidenced by equal loading of MHC. To examine whether LPS also decreased the amount of IRAK in skeletal muscle, rats were treated in vivo with LPS for 6 h and skeletal muscle homogenates were probed for IRAK-1. LPS decreased the amount of IRAK-1 in the skeletal muscle homogenates, and this effect was independent of the total amount of protein in the muscle homogenates, as evidenced by equal loading of MHC (Fig. 5C).
LPS-induced NOS2 mRNA expression required ongoing transcription and translation.
In nonmuscle tissues, NOS2 expression is regulated at multiple levels, including transcription (12), mRNA stability (44), translation (28), and protein turnover (27). Yet, how LPS stimulates NOS2 in muscle is not known. Pretreatment of C2C12 cells with the transcriptional inhibitor DRB completely prevented LPS-induced NOS2 expression (Fig. 6A). Cycloheximide also blocked LPS-induced NOS2 mRNA accumulation (Fig. 6B). These data suggest that ongoing transcription and translation must occur for maximal LPS-induced NOS2 expression. Furthermore, LPS-induced NOS2 mRNA half-life was not altered by cycloheximide over a 150-min time frame (Fig. 6C). This observation is in contrast to that regarding IL-6 mRNA half-life, which was markedly enhanced by cycloheximide in LPS-treated cells (Fig. 6D).
LPS stimulated NF-κβ activation in C2C12 cells.
LPS stimulated NOS2 protein expression, and this effect was completely inhibited by pretreatment of the cells (−30 min) with the proteasomal inhibitor MG-132 (Fig. 7A). When MG-132 was added to the cells at a time point after NF-κβ activation (+120 min) but before NOS2 protein synthesis, it failed to inhibit NOS2 expression. LPS stimulated the expression of an NF-κβ reporter plasmid time dependently (Fig. 7B), and LPS-induced NF-κβ reporter activity was inhibited by pretreatment of myocytes with MG-132 (Fig. 7C).
Dominant negative TLR4 receptor inhibited LPS signaling.
Investigators at our laboratory (29) previously showed that skeletal muscle expresses both TLR2 and TLR4, but whether C2C12 cells also express these receptors was not yet known. C2C12 myoblasts and myotubes were treated with LPS, and the level of TLR2 mRNA was quantified using Northern blotting. LPS induced a time-dependent, 50-fold increase in TLR2 but did not change TLR4 mRNA content (Fig. 8, A–D). Maximal expression of TLR2 mRNA occurred 3 h after LPS, and a similar response was observed in both myoblasts and myotubes. Although LPS did not change TLR4 mRNA content, the expression of TLR4 mRNA was almost threefold higher in myotubes than in myoblasts (Fig. 8, C and D).
C2C12 myocytes responded to LPS in the presence and absence of serum, and serum enhanced the response for NOS2 mRNA (Fig. 9, A and B). LPS increased the activity of an NF-κβ reporter plasmid transiently transfected into C2C12 cells, and this effect was completely inhibited by coexpression of a dominant negative form of the TLR4 receptor (Fig. 9C). IL-1α also increased NF-κβ reporter activity in C2C12 cells, and this effect was not altered by coexpression of the dominant negative form of TLR4.
c-Jun NH2-terminal kinase inhibitor SP-600125 blocked LPS-stimulated NOS2 mRNA expression and nitrate accumulation.
Because DRB blocks NOS2 expression, it is likely that LPS stimulates transcription of the NOS2 gene. One possible mechanism is the activation of transcription factors such as Erg, Elk, ATF-2, and c-Jun (5). These transcription factors are activated via MAP kinase phosphorylation cascades emanating from ERK-1 and -2, p38 MAP kinase, and c-Jun NH2-terminal kinase (JNK), respectively. We examined whether PD-98059 (a MEK inhibitor), SB-202190 (a p38 inhibitor), and SP-600125 (a JNK inhibitor) block the LPS-induced increase in NOS2 mRNA expression in C2C12 cells. LPS increased NOS2 mRNA, and the NOS2 mRNA response to LPS was not attenuated by PD-98059 (Fig. 10A). In contrast, SB-202190 partially blocked the LPS-induced increase in NOS2 mRNA expression, whereas the JNK inhibitor completely prevented the increase in NOS2 mRNA. These compounds showed a similar ability to block LPS-induced NOS2 protein expression (Fig. 10B). These changes were independent of changes in the level of the Erk kinase, and the efficacy of the dose of PD-98059 used was confirmed by its ability to selectively decrease LPS-stimulated Erk phosphorylation (Fig. 10B). Stimulation of NOS2 mRNA by LPS was inhibited dose dependently by SP-600125 (Fig. 10C). Complete inhibition was achieved with 50–100 μM SP-600125, and the compound had a half-maximal effective dose of 15–20 μM. Treating C2C12 cells with LPS and assaying the conditioned medium for total nitrite confirmed NO synthesis in a time frame coincident with NOS2 mRNA and protein expression (Fig. 10D). Control myocytes exhibited a low level of constitutive NO production. LPS increased total nitrite sixfold, and this response was completely blocked by pretreating the cells with SP-600125.
Excess NO inhibited LPS-stimulated NOS2 mRNA expression.
Because excess NO might feed back to inhibit NOS2 expression, we examined whether the NO donor sodium nitroprusside (SNP) could block LPS-induced NOS2 mRNA expression. Basal expression of NOS2 mRNA was unaltered by SNP, but pretreatment of C2C12 cells with SNP for 45 min almost completely blocked LPS-stimulated NOS2 mRNA expression (Fig. 11A). Furthermore, excess NO antagonized the increase in IL-6 mRNA observed in response to LPS (Fig. 11B). This inhibitory effect of SNP was not a generalized response, because SNP did not prevent the LPS-induced increase in SOCS-3 mRNA (Fig. 11C).
Bacterial infection stimulates the expression of inflammatory cytokines and the generation of free radicals that, if unchecked, result in tissue damage and organ failure (14). Although immune cells such as lymphocytes and macrophages confront the bulk of infectious agents, skeletal muscle cells also possess both afferent and efferent limbs of the innate immune system (17, 19, 24, 25). LPS, peptidoglycan, and bacterial DNA all have cognate receptors on eukaryotic cells that specifically recognize pathogen-associated molecular patterns (PAMP) (15). These receptors, named Toll-like receptors because of their homology to the Drosophila Toll protein, have been conserved evolutionarily (42).
Under in vivo conditions, NO has beneficial effects in skeletal muscle because it increases blood flow and enhances glucose uptake. However, if NO synthesis is sustained or excessive, it may also have damaging effects. It is not known how NOS2 is regulated in muscle cells. We have demonstrated for the first time that LPS stimulates the expression of NOS2 mRNA in skeletal muscle in vivo and in C2C12 myoblasts in vitro and that TLR4 is necessary for LPS to signal under both conditions. Mice with a mutation in TLR4 (C3H/HeJ) failed to express NOS2 mRNA in skeletal muscle after an intraperitoneal injection of LPS. By comparison, wild-type C3H/HeSnJ mice responded to LPS with a marked increase in the skeletal muscle expression of NOS2. The increase in NOS2 mRNA was not the result of a generalized increase in muscle mRNA, because the skeletal muscle mRNA levels of two housekeeping genes (GAPDH and L32) were unchanged. The LPS-induced expression of IκB mRNA in murine skeletal muscle was also TLR4 dependent. Wild-type C3H/HeSnJ mice responded to LPS with a marked increase in the skeletal muscle expression of IκB, whereas C3H/HeJ mice failed to respond to LPS. Although LPS stimulated NOS2 expression fourfold, the response was transient and therefore most likely was not damaging to muscle. Experimental models that more closely mimic infection, such as cecal ligation and puncture or the infusion of whole bacteria, may produce a greater and/or more sustained rise in NOS2 mRNA (9).
Previous studies using immunohistochemistry have demonstrated that LPS stimulates the expression of NOS2 specifically in muscle fibers (6, 25, 43). In this study, we showed that NOS2 coimmunoprecipitates with caveolin-3. Caveolin-3 is a muscle-specific caveolin that may localize NOS2 to caveolae within the cell membrane. C2C12 myoblasts were also used to determine whether skeletal muscle cells per se respond to LPS or whether the increase in NOS2 expression in vivo resulted from lymphocytes or tissue macrophages that were sequestered in skeletal muscle at the time muscle was removed from the animal. LPS increased NOS2 immunostaining in C2C12 myotubes throughout the cytoplasm, and this staining pattern contrasted with the staining for a muscle transcription factor, MEF2, that exhibited predominantly nuclear localization.
LPS increased NOS2 mRNA expression dose and time dependently in C2C12 cells. NOS2 was also increased by peptidoglycan from S. aureus, suggesting that TLR2 receptors on the myocytes can also signal. Maximal NOS2 expression in response to both PAMP occurred 3–4 h after exposure and within a time frame consistent with that found for NOS2 activation in vivo. NOS2 protein was detected by performing Western blotting of C2C12 cell extracts from both myoblasts and myotubes. LPS also increased total nitrate in the conditioned medium, suggesting that NOS2 mRNA is transcribed and translated and that the enzyme is biologically active in myocytes.
The transient nature of NOS2 mRNA expression in vitro and in vivo suggests that NOS2 transcription is tightly regulated. We (17, 18) previously reported that LPS stimulates the degradation of Iκβα and Iκβε in C2C12 cells but that these proteins are rapidly resynthesized to moderate the NF-κβ response. Although it is likely that continual activation of NF-κβ is necessary for sustained expression of NOS2, NO may also downregulate NOS2 expression by nitrosylating and inactivating NF-κβ (33). We speculate that in other, more severe models of sepsis, NOS2 expression may be more prolonged. However, it remains possible that even a transient exposure of muscle to NO and/or reactive nitrogen intermediates might have a prolonged effect on muscle function if critical transcription factors, enzymes, and/or proteins are irreversibly modified.
LPS from gram-negative bacteria binds to serum proteins and proteins on the cell surface. These proteins present LPS to TLR4, which initiates multiple signal transduction pathways (14). We found that LPS increased NOS2 mRNA in both C2C12 myoblasts and differentiated myotubes, suggesting that TLR4 is present on both cell types. Indeed, we detected TLR4 mRNA in both myoblasts and differentiated myotubes, and this finding is consistent with the detection of the TLR mRNA in rat skeletal muscle (29). TLR2 also was expressed in C2C12 myocytes and was markedly increased by LPS. These data suggest that LPS may prime muscle for a subsequent response to gram-positive bacteria, because these bacteria display the TLR2 ligand peptidoglycan as a component of the cell wall.
TLR4 receptor binding activates tyrosine and serine/threonine kinase cascades that ultimately impinge on transcription factors that activate the expression of inflammatory cytokines and the NOS2 gene. The most extensively studied signaling cascade for LPS involves the activation of NF-κβ. This transcription factor is bound to an inhibitory protein (Iκβ) that, upon phosphorylation and ubiquitination, is targeted for degradation by the 20S proteasome. Release of NF-κβ allows for nuclear translocation of the transcription factor and subsequent gene activation. We report that LPS-stimulated NOS2 protein expression in C2C12 cells is dependent on a functional proteasome because, when administered prophylactically, the proteasomal inhibitor MG-132 completely blocked LPS-induced NOS2 protein expression. In contrast, MG-132 was ineffective when added at a time point after which NF-κβ would have been presumed to be activated. In addition, LPS also stimulated the expression of an NF-κβ reporter plasmid in C2C12 cells, and this effect was prevented by pretreatment of the cells with MG-132. These results suggest that NF-κβ activation is necessary for LPS-induced NOS2 expression in C2C12 cells.
We transfected C2C12 cells with a dominant negative form of TLR4 to determine whether TLR4 mediates the affects of LPS on myocytes. The TLR4 dominant negative construct lacks its TIR domain and therefore cannot form an active complex with other TIR domain-containing proteins such as the adapter protein MyD88 (41). MyD88 recruits IRAK-1 to TLR4 and initiates at least two signaling cascades that lead to the activation of NF-κβ and the JNK pathway. LPS stimulated the expression of an NF-κβ reporter plasmid in C2C12 cells, and this effect was completely abolished by cotransfection with dominant negative TLR4. Hence, TLR4 appears to be necessary for the initiation of a signaling cascade that activates NF-κβ-responsive genes in C2C12 cells and skeletal muscle. In contrast, IL-1α-stimulated NF-κβ reporter activity was not inhibited by coexpression of the dominant negative form of TLR4 in C2C12 cells, thus demonstrating the specificity of the response.
LPS stimulated the activation of IRAK both in vivo and in vitro in our study. In C2C12 cells, LPS increased the phosphorylation of IRAK-1 rapidly and transiently. The significance of this response is not known, because the IRAK-1 kinase domain has been found to be dispensable. IRAK-1 with a point mutation in the ATP binding pocket or even complete deletion of the kinase domain continues to activate NF-κβ and JNK (26, 31). This suggests that other domains within the protein facilitate its activity. In addition, other IRAK family members, such as IRAK-4, are thought to be at least partially functionally redundant with IRAK-1 and can transphosphorylate the kinase. IRAK is both phosphorylated and degraded in response to LPS. Current studies suggest that the degradation of IRAK restrains IL-1 and TLR signaling and may be involved in the development of LPS tolerance. Noubir et al. (36) found that CD14 and TLR4 mediate a rapid degradation of IRAK-1, whereas complement receptor type III mediates a more sustained degradation of IRAK-1. The mechanism by which IRAK-1 is decreased in C2C12 cells, and the significance of this response to the metabolic and mechanical activity of skeletal muscle remains to be determined.
LPS-stimulated NOS2 mRNA expression in C2C12 cells required ongoing transcription and translation and was completely blocked by the transcriptional inhibitor DRB as well as by cycloheximide (an inhibitor of protein synthesis). These data suggest that an intermediary protein, such as a transcription factor that binds to the NOS2 promoter, must be synthesized at the outset for NOS2 mRNA to be expressed. It is noteworthy that NOS2 expression is regulated differently from IL-6 and TNFα expression in this respect, because cycloheximide enhances the LPS-induced expression of these cytokines (17). In the current study, we show that cycloheximide potentiates IL-6 mRNA expression by stabilizing the message and extending its half-life approximately threefold. This response is similar to the ability of cycloheximide to stabilize IL-6 mRNA in lung epithelial cells and human fibroblasts (1, 39). Such a response is not observed for NOS2 mRNA.
MAP kinase pathways that include ERK, p38, and JNK can also be activated during infection or after exposure of immune cells to LPS (11). The JNK pathway is especially noteworthy because it is also activated by eccentric exercise (7) and oxidative stress that can damage skeletal muscle. JNK phosphorylates c-Jun on serine 63, which promotes the heterodimerization of c-Jun with other activator protein-1 (AP-1) transcription factors. The JNK pathway may be important for NOS2 mRNA expression in C2C12 cells, because the JNK inhibitor SP-600125 dose dependently inhibited NOS2 mRNA expression. SP-600125 also inhibited IL-6 synthesis, and therefore it is likely that SP-600125 is a broad inhibitor of the innate immune response and that JNK and the AP-1 transcription factors regulate a diverse set of inflammatory genes.
SP-600125 is a small molecule that inhibits JNK-1, -2, and -3 with similar potency. SP-600125 exhibits 300-fold selectivity against related MAP kinases such as Erk1 and p38 (4). We cannot exclude the possibility that SP-600125 inhibits kinases other than JNK that may be responsible for NOS2 expression. Yet, other kinase inhibitors including PD-98059 (a MEK inhibitor) and SB-202190 (a p38 inhibitor) were less effective or completely ineffective at blocking LPS-induced NOS2 expression. Importantly, we also previously showed that the three inhibitors used in the present study are selective for their respective kinase pathways in C2C12 cells (18).
NO negatively regulates JNK activity by nitrosylating cysteine 116, a critical cysteine in the redox regulation of the kinase (37). Nitrosylation of JNK may be a component of a negative feedback loop in which generation of excess NO by NOS2 inhibits JNK and AP-1 transcription factors. Because JNK activity is important for NOS2 gene expression, excess NO hypothetically could downregulate NOS2 similarly to the end product inhibition observed with some enzymes. It is not known whether endogenous NO downregulates NOS2 in C2C12 cells, but we report that the NO donor SNP blunted LPS-induced NOS2 expression. SNP also blunted LPS-stimulated IL-6 mRNA but not SOCS-3 expression, suggesting that there are both NO-sensitive and -insensitive genes. Further experiments are needed to examine whether SNP alters the binding of c-Jun and NF-κβ to their respective consensus DNA binding sites in the NOS2 promoter.
Lanone et al. (30) reported that overexpression of NOS2 during sepsis may have a negative impact on skeletal muscle function. They also found that NOS2 protein levels in skeletal muscle correlated with the severity of sepsis in humans. In addition, NO3− stably modified muscle proteins with nitrotyrosine and was associated with reduced contractile activity.
In summary, the results of the present study indicate that C2C12 myoblasts are a good model for examining the effects of LPS on NOS2 expression in skeletal muscle. LPS stimulates NOS2 mRNA expression in both murine skeletal muscle and muscle cells with a similar temporal pattern. LPS-induced NOS2 expression in skeletal muscle is TLR4 dependent. LPS-induced NF-κβ activity in C2C12 myoblasts also requires TLR4. NOS2 expression requires ongoing transcription and translation, an active proteasome, and the p38 and JNK pathways. Sustained activation of NOS2 expression in skeletal muscle may participate in the erosion of lean body mass that occurs in inflammatory diseases such as sepsis and acquired immunodeficiency syndrome.
This work was supported in part by National Institute of General Medical Sciences Grant GM-38032.
We thank Danuta Huber for excellent assistance in characterizing TLR2 and TLR4 mRNA in the C2C12 cells.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society