INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

LIPOPOLYSACCHARIDE (LPS) derived from the cell wall of gram-negative bacteria mediates many of the inflammatory sequelae of infection. LPS binds to the plasma LPS-binding protein and the complex associates with CD14 (1), heat shock proteins, and a variety of other proteins on the surface of immune cells (43). The interaction of LPS with Toll-like receptor (TLR)-4 is essential for transduction of the LPS signal and the subsequent induction of inflammatory cytokine gene expression (23). Two strains of mice (C3H/HeJ and C57BL/10ScCr) that harbor a mutation in TLR-4 fail to induce cytokine expression and are resistant to the lethal effects of LPS (3, 35).

TLR-4 is part of a larger family of receptors that recognize pathogen-associated molecular patterns (PAMPS). In Drosophilae melanogaster, the Toll protein recognizes the PAMPS of fungi. Orthologues in mammals, such as TLR-2 and TLR-6, recognize PAMPS that are present on yeast and gram-positive bacteria (31). TLR-9 mediates the immune response to bacterial DNA (20). The mammalian TLR family presently consists of 10 members, and this repertoire of receptors provides the immune system with the ability to respond to a wide variety of pathogens.

During infection, resident macrophages in the liver, spleen, and peripheral tissues are activated as part of the innate immune response and synthesize a variety of cytokines. Cytokines play an integrative role in the immune response and may adopt both inflammatory and anti-inflammatory roles. Cytokines can function either locally in a paracrine or autocrine manner or at sites distant from their site of production in a manner comparable to the endocrine hormones. LPS is a strong inducer of many cytokines including tumor necrosis factor (TNF)- and interleukin (IL)-6. Several lines of evidence indicate that these cytokines are important regulators of muscle protein balance. First, TNF- impairs muscle protein metabolism when administered to control animals (7, 15, 18, 27, 42). Second, cytokine antagonists prevent or reverse sepsis- or LPS-induced changes in muscle protein synthesis (7, 24). Finally, proinflammatory cytokines decrease circulating and tissue levels of important anabolic hormones, such as insulin-like growth factor-I, that would be expected to further impair muscle protein balance (12, 13).

LPS stimulates cytokine mRNA and protein expression in classical immune tissues such as the liver, spleen, and lung, as well as nonimmune tissues such as cardiac muscle (8, 16). The cell types that contribute to this signal have not been fully delineated but potentially could include peripheral blood mononuclear cells, endothelial cells, myocytes, and/or satellite cells. The current experiments were designed to test whether LPS plays a direct role in stimulating cytokine expression in skeletal muscle in vivo and C2C12 myoblasts in vitro. In addition, we characterized the in vitro regulation of TNF- and IL-6 mRNA by LPS, dexamethasone, and MG-132.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cell culture. The C2C12 mouse myoblast cell line was purchased from the American Type Culture Collection (Manassas, VA) and used for all studies. Cells were grown in 100-mm Petri dishes (Becton Dickinson, Franklin Lakes, NJ) and cultured in minimal essential media (MEM) containing 5% newborn calf serum, penicillin (100 U/ml), streptomycin (100 µg/ml), and amphotericin B (25 µg/ml) (all from Sigma, St. Louis, MO). Cells were grown to confluence and switched to fresh serum-containing media before addition of LPS, cytokines, or other agents. C2C12 cells were used at the myoblast stage. We and Alvarez et al. (2) observed that differentiated myotubes also secrete IL-6. Experiments were performed with LPS B derived from Escherichia coli 026:B6 (DIFCO Laboratories, Detroit, MI). A variety of compounds was used to characterize the response to LPS including polymyxin B (PMB), MG-132, cycloheximide, and 5,6-dichloro--D-ribofuranosyl-benzimidazole (DRB) (all from Calbiochem, La Jolla, CA). Zymosan A from Saccharomyces cerevisiae, peptidoglycan from Staphylococcus aureus, and dexamethasone were purchased from Sigma Chemical. Additional experiments used the recombinant cytokines IL-1 and TNF- (Peprotech, Rocky Hill, NJ).

TNF- and IL-6 enzyme-linked immunosorbant assay. Conditioned media from C2C12 cells was collected over a 4-h period and frozen at 20°C until assay. Mouse IL-6 was measured with a sandwich ELISA consisting of two anti-mouse IL-6 antibodies and a strepavidin- and horseradish peroxidase (HRP)-linked secondary antibody (Pharminigen, San Diego, CA). Conditioned media was diluted with an equal volume of assay diluent, whereas plasma was diluted 1:12 before assay. Antigen and antibody complexes were detected with tetramethylbenzidine (TMB, an HRP substrate) and the reaction stopped with 2 N H₂SO₄. Ninety-six well plates were read at the absorption maximum for TMB (450 nm). TNF- in mouse plasma was measured as described above with the exception that an anti-mouse TNF- polyclonal antibody and monoclonal antibody were used as the capture and detection antibodies, respectively (Pharminigen).

Experimental protocol for C3H/HeJ and C3H/HeSnJ mice. C3H/HeJ and C3H/HeSnJ 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 their use. At the time of the study, mice were 8-9 wk old and weighed 21.4 ± 0.3 g. In the experiment depicted in Fig. 1, wild-type C3H/HeSnJ mice were injected intraperitoneally with LPS derived from E. coli 026:B6 (25 µg/mouse; DIFCO) or an equal volume of saline (250 µl/mouse). This dose was based on a preliminary dose response in C3H/HeSnJ mice and is similar to that used by other investigators (4). After 2 h, mice were anesthetized with a mixture of ketamine (Fort Dodge Animal Health, Fort Dodge, IA) and xylazine (Bayer, Shawnee Mission, KS), and blood was collected from the inferior vena cava in heparinized syringes. Individual tissues were wrapped in aluminum foil and flash-frozen in liquid nitrogen. Tissues were later powdered under liquid nitrogen using a mortar and pestle and stored at 70°C. For the experiment depicted in Fig. 11, mice were separated into four experimental groups: C3H/HeSnJ mice that received saline (WT/Sal), C3H/HeSnJ mice that were injected with LPS (WT/LPS), C3H/HeJ mice that received saline (HeJ/Sal), and C3H/HeJ mice that were injected with LPS (HeJ/LPS). Mice were injected intraperitoneally with LPS or saline as described above. After 2 h, mice were anesthetized with a mixture of ketamine and xylazine, and blood and tissues were collected as described above. All experiments were approved by the Animal Care and Use Committee at the Pennsylvania State University College of Medicine and adhere to the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Effect of lipopolysaccharide (LPS) on cytokine mRNA abundance in mouse skeletal muscle. Wild-type (C3H/HeSnJ) mice weighing 21.4 ± 0.3 g were injected intraperitoneally with either saline (Sal) or a nonlethal dose of Escherichia coli LPS (LPS, 250 µg/mouse). Tissue samples were flash-frozen in liquid nitrogen, powdered, and analyzed at the peak of cytokine expression (2 h). RNA was isolated and hybridized to a cytokine mRNA ribonuclease protection assay template as described in MATERIALS AND METHODS and run on a 5% acrylamide gel. The dried gel was exposed to a phosphorimage screen and quantified with ImageQuant software. All data are normalized to L32 mRNA as described in MATERIALS AND METHODS and expressed as a fold increase relative to animals injected with Sal alone. A: image of tumor necrosis factor (TNF)- and interleukin (IL)-6 mRNA. B: phosphorimage analysis of TNF- and IL-6 mRNA in the dried gel quantified and plotted. Values are means ± SE. Bars with different lower case letters are significantly different from each other (P < 0.05).

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 (TRI Reagent, Molecular Research Center, Cincinnati, OH) using the manufacturer's protocol. RNA was separated from protein and DNA by the addition of bromochloropropane and precipitation in isopropanol. After a 75% ethanol wash and resuspension in formamide, RNA samples were quantified by spectrophotometry. Ten micrograms of RNA were used for each assay. Riboprobes were synthesized from a multiprobe mouse template set (mCK-2b and mCK-3b, Pharminigen) using an in vitro transcription kit (Pharminigen). The labeled riboprobe was hybridized with RNA overnight using a ribonuclease protection assay (RPA) kit and the manufacturer's protocol (Pharminigen). Protected RNAs were separated using a 5% acrylamide gel (19:1 acrylamide:bisacrylamide). Gels were transferred to filter paper and dried under vacuum on a gel dryer. Dried gels were exposed to a phosphorimage screen (Molecular Dynamics, Sunnyvale, CA), and the resulting data were quantified using ImageQuant software and normalized to the mouse ribosomal protein L32 mRNA signal in each lane.

Western blot analysis. Cell extracts were electrophoresed on 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 either IB-, -, or - (Santa Cruz Biotechnology, Santa Cruz, CA). 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 chemiluminescent detection system (Amersham, Buckinghamshire, UK). Dried blots were used to expose X-ray film for 1-3 min.

Statistics. Values are means ± SE. Unless otherwise noted, each experimental condition was tested in triplicate, and each experiment was repeated two times. Data were analyzed by analysis of variance followed by Student-Newman-Keuls test. Statistical significance was set at P < 0.05. For animal studies, the number of mice per group was four (2 control groups) and six (2 LPS-treated groups).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

LPS increases cytokine mRNA expression in mouse skeletal muscle. LPS is a potent stimulus for cytokine synthesis by a variety of cell types including macrophage, peripheral blood mononuclear cells, and endothelial cells. But less is known about the regulation of cytokine mRNAs in skeletal muscle (39, 40). We examined the ability of LPS to increase TNF- and IL-6 mRNA in the skeletal muscle of mice injected intraperitoneally with LPS by RPA. LPS increased TNF- (9.5-fold) and IL-6 (106-fold) in the gastrocnemius muscle compared with values from time-matched control animals (Fig. 1, A and B).

LPS time and dose dependently increase cytokine mRNA expression in C2C12 myoblasts. Although skeletal muscle is composed primarily of muscle fibers, it also contains blood and blood vessels, connective and nervous tissue, and immune cells that could also respond to LPS. We therefore examined the ability of LPS to increase TNF- and IL-6 mRNA in the clonal C2C12 myoblast cell line (Fig. 2). LPS increased the above cytokine mRNAs time and dose dependently (Figs. 2 and 3). TNF- mRNA was rapidly and transiently increased (Fig. 2, A and B). By comparison, IL-6 mRNA was expressed for a relatively longer period of time with levels still being elevated nearly fivefold above basal 18 h after exposure to LPS (Fig. 2C). LPS also elevated the mRNA content of IL-12, IL-1, IL-1Ra, and TNF- (Table 1). Some cytokines were negatively regulated by LPS exposure. Transforming growth factor-2 and -3 mRNA, two anti-inflammatory cytokines, decreased by 75% in response to LPS (Table 1). Because of the importance of TNF- and IL-6 in the host response to infection and their putative role in muscle wasting, the remainder of our studies focused on changes in these two cytokines. Significant stimulation of the TNF- and IL-6 mRNAs occurred with as little as 0.1 µg/ml of LPS (Fig. 3). The LPS-induced increase in TNF- and IL-6 mRNA expression was completely blocked by the LPS-neutralizing agent PMB (Fig. 4, A and B). LPS also stimulated the synthesis and secretion of IL-6 protein by myocytes as detected by an enzyme-linked immunosorbant assay of the conditioned media (Fig. 4C). Although PMB completely prevented the LPS-induced increase in IL-6 mRNA content, it only partially attenuated the LPS-induced increase in IL-6 protein secretion (compare Fig. 4, B and C). PMB was specific because it blocked the ability of LPS, but not of IL-1, to increase IL-6 synthesis (Fig. 4C).

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Effect of LPS on cytokine mRNA abundance in C2C12 myoblasts. C2C12 myoblasts were grown in 100-mm Petri dishes in minimal essential media containing 5% newborn calf serum (NBCS). After reaching confluence, cells were switched to fresh media and stimulated with 1 µg/ml of E. coli LPS or an equal volume of Sal for 0.3, 0.6, 1, 2, 4, 8, or 18 h. RNA was isolated and hybridized to a cytokine mRNA ribonuclease protection assay template as described in MATERIALS AND METHODS and run on a 5% acrylamide gel. A: image of TNF- and IL-6 mRNA. The dried gel was exposed to a phosphorimage screen and quantified with ImageQuant software. All data are normalized to L32 mRNA as described in MATERIALS AND METHODS and expressed as a fold increase relative to control cells exposed to Sal. Values are single-point determinations and represent each of the time points listed above.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   LPS dose dependently increases cytokine mRNA abundance in C2C12 myoblasts. C2C12 cells were grown as described in Fig. 2 and treated with 0.01, 0.1, 1.0, or 10 µg/ml of E. coli LPS for either 1 h (TNF-) or 4 h (IL-6). RNA was isolated and hybridized to a cytokine mRNA ribonuclease protection assay template as described in MATERIALS AND METHODS and run on a 5% acrylamide gel. The dried gel was exposed to a phosphorimage screen and quantified with ImageQuant software. All data are normalized to L32 mRNA as described in MATERIALS AND METHODS. A and B: fold increases in TNF- and IL-6 quantified. Values are expressed as means ± SE for triplicate dishes. Bars with different lower case letters are significantly different from each other (P < 0.05).

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Polymyxin B (PMB) blocks LPS-induced TNF- and IL-6 mRNA expression. C2C12 myoblasts were grown as described in Fig. 2 and treated with either LPS alone (1 µg/ml) or a mixture of E. coli LPS and PMB (5 µg/ml). RNA was isolated after either 1 h to measure TNF- mRNA (A) or 4 h to measure IL-6 mRNA (B). C: additional cells exposed to IL-1 (40 ng/ml) or IL-1 and PMB. IL-6 protein was detected in the media by ELISA and expressed as a fold increase in secreted IL-6 protein compared with control cells. Control cells secreted 26 ± 8 pg/ml of IL-6 protein per well over a 4-h period. Values are means ± SE of IL-6 in quadruplicate wells. Bars with different lower case letters are significantly different from each other (P < 0.05). All RNA data are normalized to L32 mRNA as described in MATERIALS AND METHODS and are expressed as a fold increase compared with cells not exposed to LPS. Values are means ± SE for triplicate dishes. Bars with different lower case letters are significantly different from each other (P < 0.05).

C2C12 cells respond to multiple bacterial cell wall components. The ability of C2C12 cells to respond to LPS was examined in the presence of serum and under serum-free conditions. Serum components such as the LPS-binding protein and soluble CD14 are often required for cells to respond to LPS (44). C2C12 cells incubated in serum-free media for two consecutive 12-h periods (to remove serum) responded comparably to cells grown in media containing 5% serum when challenged with a maximally stimulating dose of LPS (1 µg/ml). This suggests that C2C12 cells do not require accessory factors, present in serum, to respond to LPS (Fig. 5A). The ability of LPS, a cell wall component of gram-negative bacteria, to induce cytokine expression was cell wall component specific. Zymosan A (a cell wall component from yeast), at a concentration 1,000-fold greater than that of LPS, failed to induce the expression of IL-6 mRNA (Fig. 5B). In contrast, peptidoglycan derived from the cell wall of the gram-positive bacteria Staphylococcus aureus stimulated IL-6 synthesis similarly to LPS (Fig. 5C).

View larger version (13K): [in this window] [in a new window]   Fig. 5.   LPS-induced cytokine expression does not require serum factors. C2C12 myoblasts were grown as described in Fig. 2 and subsequently grown either in 5% serum or serum-free media for 2 successive serum-free periods (12 h). A: fresh media were added to the cells in the presence of E. coli LPS (1 µg/ml) and IL-6 mRNA determined as in Fig. 2. Additional cells were treated with either the yeast cell wall component Zymosan A (ZMA; 1,000 µg/ml; B) or peptidoglycan from Staphylococcus aureus (PGSA; C). RNA was isolated and run on a polyacrylamide gel as described in Fig. 2. A-C: phosphorimages of the dried gels quantified and plotted. All data are normalized for L32 mRNA as described in MATERIALS AND METHODS. Values are means ± SE. Bars with different lower case letters are significantly different from each other (P < 0.05).

Cytokine mRNAs are differentially regulated at the transcriptional and translational level in C2C12 myoblasts. LPS is known to affect cytokine expression at multiple levels including transcription (26), translation (36), processing, and secretion (30). We examined whether ongoing protein synthesis was necessary for LPS to stimulate TNF- and IL-6 mRNA expression. Cycloheximide did not blunt the LPS-induced increase in either TNF- or IL-6 mRNA content. Cycloheximide increased basal IL-6 mRNA content on its own and acted synergistically with LPS to stimulate TNF- mRNA expression (Fig. 6, A and B). DRB, a transcriptional inhibitor, completely prevented the LPS-induced increase in IL-6 mRNA but only partially inhibited TNF- mRNA expression (Fig. 7, A and B). When DRB was added to C2C12 cells at the peak of IL-6 mRNA expression (3 h post-LPS), the message showed identical decay kinetics independent of whether the cells had previously been treated with LPS or saline (Fig. 7C). Because DRB can block LPS-induced IL-6 mRNA expression and LPS does not alter the half-life of IL-6 mRNA (45 min), it is likely that LPS stimulates transcription of the IL-6 gene.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   LPS-induced TNF- and IL-6 mRNA accumulation does not require ongoing protein synthesis. C2C12 myoblasts were grown as described in Fig. 2 and treated with cycloheximide (CHX; 10 µM) for 30 min. Cells were subsequently treated with E. coli LPS (1 µg/ml) and RNA isolated 1 h (TNF-) or 4 h later (IL-6). Phosphorimages of the dried gels were quantified and plotted for TNF- (A) and IL-6 (B). All data are normalized for L32 mRNA as described in MATERIALS AND METHODS. Values are means ± SE of triplicate dishes. Bars with different lower case letters are significantly different from each other (P < 0.05).

View larger version (15K): [in this window] [in a new window]   Fig. 7.   LPS-induced cytokine expression requires ongoing transcription. C2C12 myoblasts were grown as described in Fig. 2 and treated with 5,6-dichloro--D-ribofuranosyl-benzimidazole (DRB; 72 µM; A and B) for 30 min. Cells were subsequently treated with E. coli LPS (1 µg/ml) and RNA isolated 1 h (TNF-) or 4 h later (IL-6). A and B: phosphorimages of the dried gels quantified and plotted. Additional cells were treated with either Sal or LPS for 3 h and then given DRB to examine the half-life of IL-6 mRNA. RNA was isolated 30 and 60 min after the addition of DRB. Some cells were incubated in the absence of DRB to demonstrate that IL-6 mRNA is at a steady state over this time period. All data are normalized for L32 mRNA as described in MATERIALS AND METHODS. Values are means ± SE of triplicate dishes. Bars with different lower case letters are significantly different from each other (P < 0.05).

Cytokine mRNAs are differentially regulated by dexamethasone and MG-132. The ability of LPS to increase IL-6 mRNA in C2C12 myocytes was completely blocked by pretreating the cells with dexamethasone. In contrast, dexamethasone failed to block the LPS-induced increase in TNF- mRNA content (Fig. 8, A and B). Although pretreatment with dexamethasone blocked the expression of IL-6 mRNA, it had no effect on the half-life of the message (Fig. 8C). It is likely that dexamethasone blocks the ability of LPS to stimulate transcription of the IL-6 gene. LPS can also activate cytokine expression via a proteasomal-dependent mechanism (14). We examined whether a proteasomal inhibitor (MG-132) could block LPS-induced cytokine expression in myocytes. LPS-induced TNF- mRNA expression was inhibited by MG-132 (Fig. 9A), but this inhibitor did not affect LPS stimulation of IL-6 mRNA expression (Fig. 9B). C2C12 myoblasts expressed IB-, -, and -. A 30-min exposure to LPS decreased the amount of IB- and - protein but not IB-. Pretreatment with the proteasomal inhibitor MG-132 prevented the LPS-induced decrease in IB- and - (Fig. 9C).

View larger version (14K): [in this window] [in a new window]   Fig. 8.   Dexamethasone (Dex) inhibits LPS-induced IL-6 mRNA expression. C2C12 myoblasts were grown as described in Fig. 2 and treated with Dex (1 µM; A and B) for 30 min. Cells were subsequently treated with E. coli LPS (1 µg/ml) and RNA isolated 1 h (TNF-) or 4 h later (IL-6). A and B: phosphorimage analysis of the dried gel quantified and plotted. Additional cells were treated with either Sal or LPS for 3 h and then given Dex or Sal to examine its effect on IL-6 mRNA half-life in the presence of the transcriptional inhibitor DRB. All data are normalized for L32 mRNA as described in MATERIALS AND METHODS. Values are means ± SE of triplicate dishes. Bars with different lower case letters are significantly different from each other (P < 0.05).

View larger version (20K): [in this window] [in a new window]   Fig. 9.   Proteasomal inhibitor MG-132 inhibits LPS-induced TNF- mRNA expression and IB proteolysis. C2C12 myoblasts were grown as described in Fig. 2 and treated with MG-132 (40 µM; A and B) for 30 min. Cells were subsequently treated with E. coli LPS (1 µg/ml) and RNA isolated 1 h (TNF-) or 4 h later (IL-6). A and B: phosphorimage of the dried gel quantified and plotted. All data are normalized for L32 mRNA as described in MATERIALS AND METHODS. Values are means ± SE of triplicate dishes. Bars with different lower case letters are significantly different from each other (P < 0.05). Additional cells were pretreated with MG-132 for 30 min and treated 30 min with LPS. Total cell extracts from control cells, cells treated with MG-132, LPS alone, or LPS and MG-132, were isolated in sample buffer and run on an SDS-PAGE gel. Triplicate gels were transferred to nitrocellulose and probed with antibodies specific for IB-, -, and -. C: representative Western blot for IB-,-, and -.

TNF- and IL-1 regulate cytokine mRNAs with different kinetics but do not mediate the effect of LPS. The expression of proinflammatory cytokine mRNAs in muscle may not only be regulated as part of the innate immune response to LPS but also by cytokines that are secreted in a paracrine or endocrine fashion in response to injury or infection at a distal site. It was therefore of interest to examine whether the expression of IL-6 mRNA in C2C12 cells could be regulated by TNF- and IL-1 per se. TNF- and IL-1 induced IL-6 mRNA expression in C2C12 cells (Fig. 10A). In general, the response to IL-1 was more rapid than that observed for TNF-. Maximal IL-6 mRNA expression occurred after a 2-h exposure to IL-1, but TNF- required at least 8 h to stimulate IL-6 mRNA to the same magnitude. Because LPS stimulates the synthesis of both IL-1 and TNF-, it is possible they may act as part of an autocrine cytokine network to stimulate other cytokines such as IL-6 in C2C12 cells. We tested this hypothesis by pretreating cells with either IL-1Ra or TNF-binding protein to determine if these inhibitors could sequester endogenous cytokines and block IL-6 mRNA expression. LPS stimulated IL-6 mRNA content up to 18-fold, but this response was unaltered by either antagonist, suggesting that LPS directly stimulates IL-6 synthesis (Fig. 10B).

View larger version (18K): [in this window] [in a new window]   Fig. 10.   TNF- and IL-1 stimulate IL-6 mRNA accumulation with different kinetics. C2C12 myoblasts were grown as described in Fig. 2 and treated with either IL-1 (40 ng/ml) or TNF- (40 ng/ml) and RNA isolated after 0.3, 0.6, 1, 2, 4, or 8 h. A: phosphorimage of the dried gel quantified for IL-6 mRNA. Additional cells were pretreated with either IL-1Ra or TNF-binding protein (TNFBP) for 1 h and then stimulated with LPS for 4 h. B: IL-6 mRNA quantified. All data are normalized for L32 mRNA as described in MATERIALS AND METHODS.

C3H/HeJ mice are hyporesponsive to LPS at the level of cytokine mRNA expression in skeletal muscle. C3H/HeJ mice harbor a mutation in the Toll/interleukin-1R domain of TLR-4 and are hyporesponsive to LPS (46). These mice and a comparable wild-type strain were used to determine whether administration of LPS increased cytokine mRNA content in skeletal muscle in vivo and whether the TLR-4 receptor is necessary for such stimulation. C3H/HeJ and wild-type mice (C3H/HeSnJ) were injected with a nonlethal dose of LPS, and blood and tissue samples were obtained near the peak of cytokine expression (2 h). LPS increased the plasma concentration of TNF- (17-fold) and IL-6 (14-fold) as determined by ELISA (Table 2). This response was severely blunted in C3H/HeJ mice (TNF-, 2% of wild type and IL-6, 11% of wild type). The cytokine expression pattern in skeletal muscle in response to LPS was similar to that seen with C2C12 cells consisting of a large increase in IL-6 (107-fold) and TNF- (9-fold) mRNA (Fig. 11C and Table 2). LPS also stimulated IL-6 mRNA expression in the spleen, liver, and heart of wild-type mice (Fig. 11, A-D). C3H/HeJ mice were hyporesponsive to LPS in multiple tissues including the spleen (30%), liver (0%), gastrocnemius (4%), and heart (3%) when the level of IL-6 mRNA content was compared with wild-type mice. C3H/HeJ mice were also hyporesponsive to LPS at the level of TNF- mRNA in skeletal muscle (Table 2). Cardiac muscle from C3H/HeJ mice was hyporesponsive to LPS at the level of IL-6 (3%) compared with control mice (Table 2). In contrast, C3H/HeJ mice were surprisingly responsive to LPS at the level of TNF- mRNA expression in cardiac muscle with a response that achieved 72% of that seen in wild-type mice. Cardiac IL-6 mRNA was elevated in the basal state of C3H/HeJ mice compared with wild-type mice (Table 2).

View this table: [in this window] [in a new window]   Table 2.   Plasma IL-6 and TNF- concentrations and expression of their mRNAs in skeletal and cardiac muscle from wild-type and TLR-4-mutant mice after a 2-h LPS challenge

View larger version (28K): [in this window] [in a new window]   Fig. 11.   LPS-induced changes in the expression of IL-6 mRNA in vivo are Toll-like receptor (TLR)-4 dependent. Mice with a nonfunctional mutation in TLR-4 are hyporesponsive to LPS. Wild-type (WT) mice (C3H/HeSnJ) and TLR-4-mutant mice (C3H/HeJ) were injected with either Sal or a nonlethal dose of E. coli LPS (250 µg/mouse). Tissue samples were flash-frozen in liquid nitrogen, powdered, and RNA was obtained at the peak of cytokine expression (2 h). A-D: phosphorimage analysis of IL-6 mRNA in the dried gel quantified and plotted. IL-6 RNA in the spleen (A), liver (B), skeletal muscle (C), and heart (D) are shown. All data are normalized for L32 mRNA as described in MATERIALS AND METHODS. Values are means ± SE, n = 4 control and 6 LPS for each group. Bars with different lower case letters are significantly different from each other (P < 0.05).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

This study demonstrates that LPS stimulates the expression of cytokines in both skeletal muscle in vivo and C2C12 myoblasts in vitro. Although LPS is known to increase cytokine expression in immune cells, less is understood about its effect on skeletal muscle or muscle cells. For the first time, we showed that LPS increases the expression of TNF- and IL-6 in gastrocnemius muscle. Thus LPS can alter the expression of both pro- and anti-inflammatory cytokines in skeletal muscle. Given the heterogeneous composition of skeletal muscle, we next examined whether LPS could directly stimulate cytokine expression in a clonal population of C2C12 myocytes. LPS increased the expression of a set of cytokine mRNAs in C2C12 cells that was similar to that observed in mouse skeletal muscle, including TNF-, IL-6, IL-12, IL-1, and IL-1Ra.

TLR-4 appears to be necessary for cytokine (TNF- and IL-6) expression in mouse skeletal muscle, liver, spleen, and heart in response to LPS. C3H/HeJ mice that harbor a mutation in TLR-4 fail to express cytokines in response to LPS, whereas a similar strain (C3H/HeSnJ), which does not carry this mutation, demonstrates a robust response. It is possible that LPS interacts directly with TLR-4 in skeletal muscle and stimulates cytokine mRNA expression. Muscle cells therefore are capable of mounting an innate immune response much like cells of the immune system.

The response of cells and animals to LPS can be characterized in a number of ways. We showed that the activation of C2C12 myoblasts by LPS does not require serum factors. Although factors such as the LPS-binding protein and soluble CD14 may enhance the LPS response, they do not appear to be essential for LPS activity in myoblasts. This result is consistent with the CD14-independent increase in TNF- mRNA, in cardiomyocytes treated with LPS (9). Cytokine mRNA expression was blocked by PMB, an LPS-neutralizing agent, in C2C12 cells. This effect is relatively specific because PMB blocked the ability of LPS to stimulate the synthesis of a representative cytokine (IL-6) but failed to prevent IL-1-stimulated secretion of the cytokine.

Other investigators showed that C2C12 myoblasts respond to TNF- and IL-1. Indeed, there is some controversy whether cytokines such as TNF- stimulate or prevent C2C12 cells from differentiating into myotubes (5, 28). We found that both TNF- and IL-1 increased the expression of IL-6 mRNA and protein in C2C12 cells. The kinetics of the IL-6 mRNA response to IL-1 were similar to those observed for LPS, whereas the increase in response to TNF- was slower and less robust. These results are consistent with IL-1R and TLR-4 being part of the same TLR superfamily and the two receptors sharing many intracellular signaling pathways (31). Although we have not specifically investigated whether LPS alters markers of myoblast differentiation, such as creatine kinase or myogenin, we did not observe a change in either the total cell number or the number of myotubes over a 24-h period. However, it is also likely that any physiological changes that result from LPS exposure would depend on the net balance of both pro- and anti-inflammatory cytokine expressions. Our cultures contained mostly C2C12 myoblasts. Although we (unpublished data) and others (2) found that C2C12 myotubes secrete IL-6, it is also likely that this gene may be regulated slightly differently in myoblasts and fully differentiated myotubes.

We examined whether ongoing transcription and translation were necessary for LPS-induced expression of cytokine mRNAs in C2C12 cells. Paradoxically, cycloheximide increased IL-6 and TNF- mRNA. This response was completely blocked by the transcriptional inhibitor DRB. These data suggest that the stress produced by cycloheximide alone stimulates transcription of the above genes. Cycloheximide also enhanced the ability of LPS to stimulate the expression of TNF- and IL-6 mRNA. Cycloheximide has previously been shown to increase IL-6 mRNA in other cell types (29, 38). The transcriptional inhibitor DRB blocked LPS-stimulated IL-6 mRNA expression, but DRB only partially inhibited TNF- mRNA accumulation. If cells were stimulated with LPS for 3 h, allowing for maximal expression of IL-6, and then treated with DRB, the half-life of IL-6 was found to be identical in the presence and absence of LPS. This finding shows that LPS does not alter IL-6 mRNA half-life and strongly suggests that LPS stimulates transcription of the IL-6 gene. Additional studies examining IL-6 and TNF- promoter activity, in C2C12 cells, are needed to determine if LPS truly regulates these cytokines at the transcriptional level.

The immune response in vivo can be suppressed by anti-inflammatory cytokines as well as by glucocorticoids. We examined whether a synthetic glucocorticoid could suppress LPS stimulation of cytokine mRNA expression in C2C12 cells. Dexamethasone completely blocked the LPS-induced increase in IL-6 mRNA content but was ineffective at suppressing TNF- mRNA accumulation. This inhibitory pattern is similar to that seen with DRB and suggests that dexamethasone may be functioning at the transcriptional level. This conclusion is consistent with our observation that dexamethasone did not alter IL-6 mRNA half-life. Such a transcriptional mechanism would also be consistent with the ability of dexamethasone to bind to the glucocorticoid receptor (GR). The GR often acts as a transcriptional inhibitor when bound to specific glucocorticoid-responsive elements (GREs). Many cytokine genes have GREs in their promoters (37). The inability of dexamethasone to inhibit the accumulation of TNF- mRNA is somewhat surprising. Dexamethasone has previously been shown to inhibit the activity of the TNF- promoter in human monocytic THP-1 cells (41). Yet, the response of various cells and genes to glucocorticoids is often dependent on the context of transacting factors with which the GR associates and the nucleotide sequence of the cis elements close to the GRE. It is possible that muscle cells provide a different transcriptional context than that found in monocytes. A direct comparison of TNF- promoter activity in C2C12 and monocytic cells will be necessary to determine if tissue-specific regulation of the TNF- gene occurs in response to LPS and glucocorticoids.

The LPS and IL-1 receptors are part of a larger family of TLRs that share the ability to activate the transcription factor nuclear factor (NF)-B. NF-B binds to the promoters of many genes involved in the immune response. A key regulatory step in NF-B activation is the proteolytic degradation of its inhibitory protein IB by the proteasome. We examined whether MG-132, a proteasomal inhibitor, could block LPS-induced cytokine mRNA expression in C2C12 cells. MG-132 selectively inhibited TNF- mRNA expression but had no detectable effect on the LPS-induced increase in IL-6 mRNA content. These data suggest that proteasomal activation plays a role in the accumulation of TNF- mRNA in response to LPS. LPS decreased IB, -, and - levels in C2C12 cells, and this was prevented by pretreatment with MG-132, suggesting that NF-B activation may play a role in regulating the TNF- promoter in C2C12 cells. The fact that MG-132 can completely block TNF- mRNA accumulation, whereas DRB only attenuates this response, suggests that the proteasome may regulate TNF- mRNA at both a transcriptional and posttranscriptional level. It is likely, in C2C12 cells, that a labile protein factor keeps the constitutive amount of TNF- mRNA at a relatively low level. Many protein factors have been shown to bind to AUUUA (A = adenine, U = uracil) elements in the 3'-untranslated region of TNF mRNA and to control both its stability and translation. This includes RNA-binding proteins such as TIAR (19) and Hu (45) as well as the 20S proteasome itself (22).

The responsiveness of C2C12 cells to LPS is in many ways similar to other LPS-inducible cell types. LPS activates C2C12 cells in the same concentration range as peripheral blood mononuclear cells and cardiomyocytes (9). C2C12 cells also become LPS resistant like immune cells and whole animals (unpublished observation). C2C12 cells respond to a variety of pathogen-associated patterns including LPS from gram-negative bacteria and peptidoglycan from gram-positive microorganisms. C2C12 cells also respond to proinflammatory cytokines such as IL-1 and TNF-. Finally, C2C12 cells show a complex regulation of cytokine expression that includes transcriptional and posttranscriptional regulation in response to LPS, glucocorticoids, and proteasomal inhibitors.

Critical illness is often associated with a loss of muscle protein due to both increased muscle proteolysis and a decrease in muscle protein synthesis (6). Amino acids are mobilized from skeletal muscle and reused by other tissues, such as the liver, for the synthesis of acute-phase proteins (21). Although the response to infection is often viewed as a systemic event, local synthesis of proinflammatory cytokines in skeletal muscle may also promote muscle wasting. TNF- inhibits protein synthesis in human skeletal muscle cells in vitro (17) and rat skeletal muscle in vivo (25). IL-6 also enhances protein degradation in C2C12 myoblasts (11) and dramatically stunts growth in transgenic mice that overexpress the protein (10). Conversely, local expression of anti-inflammatory cytokines such as IL-1Ra may limit the loss of muscle protein. Systemic infusion of an IL-1Ra prevents the decrease in skeletal muscle protein synthesis that occurs in a rat model of sepsis (24). IL-12 also prevents deterioration of diaphragmatic muscle function in septic rats (32).

In general, the cytokine response of C3H/HeJ mice to LPS was greatly reduced compared with C3H/HeSnJ mice. This was reflected systemically with a suppressed concentration of TNF- and IL-6 in the plasma of LPS-treated C3H/HeJ mice and is consistent with many studies showing that this mouse strain is hyporesponsive to LPS. C3H/HeJ mice were also hyporesponsive to LPS at the level of the gastrocnemius, liver, spleen, and heart. Surprisingly, the response of TNF- mRNA to LPS in cardiac muscle was comparable between C3H/HeJ and wild-type mice. These data agree with a recent study by Baumgarten et al. (4) where the TNF- mRNA responses of C3H/HeJ and C3HeB/Fej mice 2 h post-LPS were equivalent. However, the LPS-induced increase in cardiac TNF- mRNA was diminished in C3H/HeJ mice at an earlier time point (0.5 h).

Although C3H/HeJ mice have a mutation in TLR-4, they do show some residual activation of cytokine expression. This may be due to activation of other TLRs such as TLR-2 and -6 (34) or the B lymphocyte-associated LPS receptor RP-105 (33). The reduced ability of LPS to stimulate cytokine expression in skeletal muscle of C3H/HeJ mice in vivo also does not necessarily exclude the possibility that the lack of a functional TLR-4 receptor on immune cells is responsible for the hyporesponsiveness of skeletal muscle. Immune cells from C3H/HeJ mice may be incapable of transmitting a signal to myocytes and thus fail to activate IL-6 and TNF- mRNA expression in skeletal muscle.

In summary, the results of the present study indicate that C2C12 myoblasts are a good model system for examining the effects of LPS on cytokine expression in skeletal muscle. LPS regulates both pro- and anti-inflammatory cytokines in these cells. LPS also regulates cytokine expression in mouse skeletal muscle. The activation of myocytes in vivo requires a functional TLR-4 either in immune cells or skeletal muscle itself. Residual cytokine activation in C3H/HeJ mice, in response to LPS, suggests that other receptors also exist that can contribute to the recognition of gram-negative bacteria in vivo. Direct activation of cytokine expression in skeletal muscle may promote the muscle wasting that occurs in inflammatory diseases such as sepsis and the acquired immune deficiency syndrome.