| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TRIENNIAL GROWTH SYMPOSIUM |
Department of Cellular and Molecular Physiology, Pennsylvania State University College of Medicine, Hershey 17033
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONTOLL-LIKE RECEPTORS RECOGNIZE...TOLL-LIKE RECEPTORS LEAD DOWN...MAKING AND BREAKING MUSCLE...CONCLUSIONSLITERATURE CITED |
|---|
B (NFB), in the cytoplasm. Phosphorylation and degradation of the inhibitor of NFB allows for translocation of NFB to the nucleus and activation of inflammatory genes. Overexpression of a constitutively active inhibitor of NFB kinase in skeletal muscle causes severe wasting, and we found that inhibitors of either the phosphorylation of IB or its proteolytic degradation prevent TLR ligand-induced expression of cytokines and NOS2. The combination of LPS and interferon 
dramatically enhances the magnitude and duration of LPS-stimulated NOS2 expression and reduces protein translation. Lipopolysaccharide and interferon also downregulates signaling from the mammalian target of rapamycin, a kinase that directs changes in cell size. Inhibitors of NOS block the fall in muscle cell protein synthesis and restore translational signaling, indicating that activation of the NOS2-NO pathway is responsible for the observed decrease in muscle protein synthesis. Our work provides a molecular explanation for reduced muscle growth during infection. Muscle is largely self-sufficient because it expresses receptors, signaling pathways, and effectors to regulate its own size. Prolonged activation of NFB and NOS2 have emerged as detrimental facets of the immune response in muscle. The interplay between inflammatory components and growth factor signaling clearly places muscle at the interface between growth and immunity.
Key Words: cytokine • endotoxin • growth • mammalian target of rapamycin • muscle • pathogen
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONTOLL-LIKE RECEPTORS RECOGNIZE...TOLL-LIKE RECEPTORS LEAD DOWN...MAKING AND BREAKING MUSCLE...CONCLUSIONSLITERATURE CITED |
|---|
We have studied the loss of muscle mass from the clinical perspective of blood-borne infection or sepsis. The spread of bacteria from a focus is the 13th leading cause of death in the United States overall and the foremost cause of death in noncoronary intensive care units. Because sepsis affects muscle tissue, which is necessary for both locomotion and respiration, sepsis has a negative effect on both morbidity and mortality (Ulevitch et al., 2004
). The incidence of sepsis is expected to rise during the next decade due to an aging population, the increased use of invasive catheters, and an escalation in antibiotic-resistant organisms in hospital settings. All of the above portend a potential crisis in the care of septic patients but are also reflective of a growing problem in our management of infectious diseases in general.
Infection and Animal Growth
Infections have a major negative effect on the growth of animals and their ability to accumulate lean body mass. Like other stresses, infections are associated with diminished food intake. But unlike malnutrition per se, infected animals appear to be resistant to the stimulatory effects of nutrients, and, therefore, additional treatment modalities are necessary to mount an anabolic response during infection. Although infection elicits disparate changes in GH secretion in different species, it is universally found that sepsis produces GH resistance (Lang et al., 2005). As a consequence, plasma and tissue levels of IGF-I and IGFBP are modified during infection, and these changes are thought to mediate, at least in part, the negative N balance and muscle wasting associated with sepsis. A systematic discussion of the sepsis-induced changes in the GH-IGF-I axis is beyond the scope of the present review but has been the topic of several recent reviews (Frost and Lang, 2004; Lang et al., 2005, 2007).
The current review will focus on the latest advances in our understanding of how muscle recognizes and responds to pathogens and pathogen-associated molecular patterns (PAMPS). This review will highlight the ability of receptors in skeletal muscle to recognize pathogens and transduce their presence into a signal that activates genes involved in combating infection. We postulate that when the immune response is overzealous and takes place in the absence of adequate feedback, a depletion of muscle mass results. The importance of Toll-like receptors (TLR), the transcription factor nuclear factor B (NFB), the nutrient sensor mammalian target of rapamycin (mTOR), and the enzyme NO synthase (NOS) in this process will be expounded upon.
|
TOLL-LIKE RECEPTORS RECOGNIZE THE ENEMY WITHIN |
|---|
|
TopAbstractINTRODUCTIONTOLL-LIKE RECEPTORS RECOGNIZE...TOLL-LIKE RECEPTORS LEAD DOWN...MAKING AND BREAKING MUSCLE...CONCLUSIONSLITERATURE CITED |
|---|
). The Toll protein binds an endogenous ligand (i.e., Spatzle) that is cleaved by a Ser protease (i.e., Spatzle-processing enzyme) in the presence of fungi or Gram-positive bacteria. The cleaved ligand activates Toll and establishes a positive feedback loop for induction of the antimicrobial peptide Drosomycin (Mulinari et al., 2006). Mammals, being more complex, express almost a dozen proteins with homology to Toll, and these receptors are referred to as TLR (Carpenter and O’Neill, 2007). Toll-like receptors, in a relatively specific manner, recognize a multitude of environmental pathogens. The pathogens are identified by their distinct molecular patterns, commonly referred to as PAMPS, and thus the TLR represent the first line of defense against bacterial cell wall components, such as lipopolysaccharide (LPS) and lipopeptides as well as RNA and DNA from viruses and proteins from protozoa (Leaver et al., 2007).
Ample evidence indicates that the systemic infection generated in experimental models of sepsis such as cecal ligation and puncture, the introduction of a bacteria-laden agar pellet into the abdominal cavity, and even an intraperitoneal injection of LPS can decrease muscle protein synthesis (Vary and Kimball, 1992; Lang et al., 2000; Lang and Frost, 2004). The decrease in protein synthesis appears to be mediated in part by proinflammatory cytokines, such as tumor necrosis factor 
(TNF) and IL-1β, because cytokine antagonists and neutralizing antibodies ameliorate the decrease in protein synthesis (Lang et al., 1996; Cooney et al., 1999a). It is generally thought that these agents inhibit protein synthesis by blocking blood-borne cytokines from affecting muscle, because the infusion of both TNF and IL-1β can also decrease muscle protein synthesis (Cooney et al., 1999b; Lang et al., 2002).
Endotoxemia also decreases muscle protein synthesis in neonatal pigs (Orellana et al., 2002), and sarcocystic infections are notorious for inducing myopathies in cats, dogs, mice, sheep, and cows (Ruiz and Frenkel, 1976; Jeffrey et al., 1989; Elsasser et al., 1998; Chapman et al., 2005). Interestingly, many parasites preferentially infect muscle, and, in the case of mice infected with malaria, this results in both functional and biochemical changes in the fibers themselves (Bagheri et al., 1986; Montes de Oca et al., 2004; Brotto et al., 2005).
TLR in Skeletal Muscle
When LPS is injected i.v. in rabbits, approximately 50% of the initial dose is cleared rapidly with the remaining LPS exhibiting a plasma half-life on the order of 12 h. The majority of i.v.-injected LPS is taken up by the liver (40%), but considerable amounts of LPS are also distributed to the spleen, lung, kidney, and adrenal glands. Because skeletal muscle accounts for a considerable portion of total body mass, muscle also represents a significant depot for LPS accumulation (Mathison et al., 1980).
Skeletal muscle responds rapidly to the i.p. injection of LPS. In the dog, LPS injection increases muscle levels of glucose-6-phosphate, phosphocreatine, and ATP within 5 min (Myrvold et al., 1975). Muscle also quickly reacts to an i.p. injection of LPS by upregulating the expression of TNF and IL-1β mRNA within 30 min (Lang et al., 2003). The time frame of these responses indicate that muscle directly responds to pathogen-associated molecules, although sensory nerve fibers in the liver and the spleen may transmit information to the brain that then initiates responses in skeletal muscle. For the most part, however, the neuronal regulation of cytokine expression has largely been shown to be antiinflammatory and not proinflammatory in nature (Tracey, 2007).
To examine whether muscle can directly respond to LPS independent of immune organ activation, we injected LPS directly into muscle. Injection of LPS into the gastrocnemius strongly induced IL-6 mRNA transcription in the injected muscle but not in the contralateral muscle, indicating a direct effect of LPS on the muscle (Frost et al., 2006). Mice also responded to the direct injection of a tripalmitoylated Cys-, Ser-, and Lys-containing peptide that specifically activates TLR2, indicating that muscle has the capability to respond to multiple PAMPS. Although muscle injury, due to injection of saline alone, did not significantly alter the expression of the cytokines we measured, localized muscle damage did stimulate the expression of the suppressor of cytokine signaling (SOCS)-3. The expression of SOCS-3 after muscle injury is consistent with the upregulation of SOCS-3 mRNA that was observed after strenuous exercise (Spangenburg et al., 2006).
Because muscle responds to LPS, we undertook a 2-pronged approach to confirm the expression and functionality of TLR in muscle. First, we examined which specific TLR are expressed in muscle. Secondly, we performed experiments on skeletal myocytes in culture to eliminate the confounding influence of tissue macrophages, endothelial cells, and inputs from the nervous system. We reverse-transcribed RNA from the mouse C2C12 myoblast cell line and performed 35 cycles of PCR with primers for TLR1 to 9. We found that the muscle cell line expressed TLR1 through TLR7, but we were unable to detect TLR9 and -10 using this technology. A similar procedure carried out on mouse skeletal muscle also failed to detect TLR9 and -10, whereas these TLR were easily detected in the spleen. If the above TLR mRNA are translated into protein, one would expect that muscle has the capacity to respond to bacterial lipopeptides and LPS, single- and double-stranded RNA from viruses, and flagellin, but not bacterial DNA.
We confirmed that skeletal myocytes could respond to TLR2 ligands, such as peptidoglycan, LPS from Porphyromonas gingivalis, and a tripalmitoylated Cys-, Ser-, and Lys-containing peptide that specifically activates TLR2. All of these ligands strongly induced the expression of IL-6 and other inflammatory mediators. Likewise, myocytes responded to LPS from multiple strains of Escherichia coli, indicating they express an active TLR4 receptor. Although myocytes reacted only weakly to a double-stranded RNA mimetic (i.e., TLR3 ligand) and imiquimod (i.e., TLR7 ligand), these compounds were able to synergize with other TLR ligands consistent with their use as immune response modifiers (Frost et al., 2006; Gaspari, 2007; Sel et al., 2007). In contrast, a bacterial DNA mimetic containing cytosine-phosphoguanine di-nucleotides and a strong TLR9 ligand failed to alter IL-6 expression in C2C12 cells, indicating that muscle does not recognize the molecular pattern of bacterial DNA, and this is consistent with the absence of TLR9 mRNA in both whole muscle and myocytes per se. Boyd et al. (2006) have confirmed this finding but suggested that priming C2C12 cells with interferon (IFN) may enhance the response to TLR9 ligands. These authors also demonstrated that C2C12 cells respond to the TLR5 ligand flagellin.
The response of muscle cells to different PAMPS is specified by the TLR to which they bind. In C2C12 cells, both TLR2 and -4 ligands stimulate IL-6 promoter activity and the activity of an NFB reporter plasmid (Frost et al., 2006). However, tripalmitoylated Cys-Ser-Lys, a TLR2 ligand, can be inhibited by cotransfection of myocytes with a dominant negative form of TLR2. In contrast, dominant negative TLR2 does not block the ability of LPS (a TLR4 ligand) to stimulate an inflammatory response in C2C12 cells. The converse also holds true in vivo where direct i.m. injection of LPS stimulates IL-6 mRNA expression in wild-type mice but not in mice that harbor an inactivating mutation of TLR4 (Frost et al., 2006). Then again, i.m. injection of the TLR2 ligand tripalmitoylated Cys-Ser-Lys stimulated IL-6 mRNA transcription equally well in both wild-type and TLR4 mutant mice, demonstrating TLR specificity both in vivo and in vitro.
The ability of muscle cells to respond to a wide variety of PAMPS is consistent with muscle wasting and stunted growth being concurrent with a wide variety of infectious insults. In addition, the epidemiology of sepsis in the United States has changed over the last 10 yr, with Gram-positive bacteria and fungal organisms becoming an increasingly common cause of sepsis (Martin et al., 2003; Hoebe et al., 2006). Therefore, an understanding of the recognition of PAMPS by the innate immune system in general and muscle in particular becomes ever more important when examining potential points of therapeutic intervention.
Giving the NOD to Unconventional Recognition Systems
Although TLR recognize extracellular pathogens or pathogens that have been internalized into specialized cellular compartments, only a recently appreciated set of pathogen recognition molecules, termed the nucleotide oligomerization domain (NOD) proteins, are capable of mounting an immune response to intracellular microbes. These proteins contain a Leu-rich repeat that recognizes Gram-negative type peptidoglycan (i.e., NOD1) and a muramyl dipeptide from both Gram-positive and Gram–negative organisms (i.e., NOD2). The NOD proteins are essential in maintaining intestinal integrity. Mutations in NOD2 are associated with Crohn’s disease, in which a failure of the gut mucosa to respond locally may initiate a systemic response resulting in uncontrolled inflammation. A murine model carrying a NOD2 mutation demonstrates that NOD2 mutations potentiate IL-1β processing and trigger an increased and sustained inflammatory response to NOD2 ligands (Maeda et al., 2005).
Patients with Crohn’s disease exhibit both skeletal muscle weakness and growth retardation, and this may be related to their elevated levels of IL-1β due to NOD2 mutations. Some patients with inflammatory bowel disease also harbor mutations in endogenous IL-1R antagonist and are, therefore, hypersensitive to IL-1β (Witkin et al., 2002; Cormier et al., 2005). These findings are consistent with an IL-1 receptor antagonist restoring muscle protein synthesis in septic rats and the idea that cytokines, in general, may mediate many of the negative effects of pathogens on skeletal muscle (Lang et al., 1996). Although it is not known whether skeletal muscle expresses the NOD proteins, both NOD1 and -2 are expressed in cardiac muscle, and NOD1 agonists can induce multiple organ failure in vivo (Rodriguez-Martinez et al., 2005; Cartwright et al., 2007).
The platelet-activating factor (PAF) receptor also acts as a pattern recognition receptor for phosphorylcholine present in bacterial cell walls. The PAF receptor shepherds these components across endothelial cells where they then exit the vasculature and enter tissues. Trans-endothelial transport is dependent on the PAF receptor, because mice with a mutation in the receptor are protected from the negative effects of bacterial cell wall components rich in phosphorylcholine (Fillon et al., 2006). Phosphorylcholine has a dramatic negative effect on cardiomyocyte contractile proteins, and PAF itself has previously been shown to mediate the negative effects of TNF on skeletal muscle contractility (Alloatti et al., 2000; Fillon et al., 2006). Interestingly, during ischemia reperfusion, there is dramatic muscle injury that can be rescued by infusing a PAF receptor antagonist (i.e., WEB2170). These results indicate that both endogenous and exogenous receptor ligands may influence muscle function and muscle protein synthesis during periods of tissue injury (Lepore et al., 1995; Karlstad et al., 2000; Fillon et al., 2006).
Indigenous TLR Ligands: PAMPS in Sheep’s Clothing
Naturally occurring molecules and proteins may commandeer TLR and mimic the effects of pathogens on muscle and other tissues (Miyake, 2007). Indeed, autoimmune diseases, such as rheumatoid arthritis, are associated with muscle weakness and cachexia (Walsmith and Roubenoff, 2002). Much emphasis has been placed on the release of nuclear components, such as DNA and DNA binding proteins, during tissue damage. In addition, the presence of autoantibodies against DNA has been recognized for many years as a hallmark of autoimmune diseases. Recently, it was suggested that the dual engagement of immunoglobulin M and TLR9 on immune B cells by a complex of chromatin and immunoglobulin G might exacerbate autoimmunity and tissue damage (Leadbetter et al., 2002). The role of TLR9 in this process is intriguing, because TLR9 functions inside the endosome and requires endosomal acidification. Toll-like receptor-9 function can be inhibited by chloroquine, which impairs endosomal acidification. Thus, it is not surprising that chloroquine has been used to treat autoimmune diseases, such as rheumatoid arthritis and lupus (Wallace, 1994; Furst et al., 1999), and the antiinflammatory effects of chloroquine may be related to its ability to disengage the sensing, signaling, or both, of autoimmune DNA by TLR9.
Patients with rheumatoid arthritis and lupus also exhibit autoimmunity against the DNA binding protein high mobility group (HMG)B1 (Uesugi et al., 1998). This is enticing, because HMGB1 is considered to be a late-phase cytokine, because, unlike TNF and IL-1β which are released very early during infection, HMGB1 is released 7 to 30 h after an i.p. injection of LPS (Wang et al., 1999). Serum levels of HMGB1 are also elevated in septic patients compared with healthy control subjects and even further elevated in nonsurvivors. Finally, anti-HMGB1 antibodies protect mice against LPS-induced lethality.
We found that LPS induces HMGB1 after 12 to 24 h and that a previous exposure to alcohol enhances HMGB1 mRNA transcription (Lang et al., 2003; Frost et al., 2005). The HMGB1 binds to TLR4 on primary cells and TLR2 on cell lines. Mice harboring a mutation in TLR4 exhibit a blunted response to HMGB1 (Tsung et al., 2005; Yu et al., 2006). Interestingly, HMGB1 may hamper muscle growth and repair after injury. Mice expressing a muscle-specific form of IGF-I make less HMGB1 after an injection of a muscle toxin and ultimately exhibit more efficient muscle regeneration. These results indicate that this form of IGF-I may stimulate growth by inhibiting HMGB1 expression and inflammation (Pelosi et al., 2007).
Type II diabetes is associated with elevated levels of FFA, and TLR2 is necessary for palmitate-induced insulin resistance (Senn, 2006). In another report, TLR4 was identified as an additional gateway by which fatty acids affect inflammation and metabolism (Kim, 2006; Shi et al., 2006). Likewise, palmitate influences inflammation in adipocytes (Ajuwon and Spurlock, 2005). At the whole-body level, infusion of FFA into rats decreased the basal rate of protein synthesis in skeletal muscle and impaired the anabolic effects of IGF-I on various measures of protein translation (Lang, 2006). These results indicate that sustained levels of fatty acids, as would be present during diabetes, obesity, and sepsis, may negatively affect the ability of insulin and IGF-I to regulate carbohydrate and protein metabolism in skeletal muscle and that they do so by impersonating the signals induced by pathogens (Mayer et al., 2003).
A variety of other putative endogenous TLR ligands have been described in the literature, including heat shock proteins that bind TLR4 (Ohashi et al., 2000), the small ribonuclear RNA U1 that binds TLR3 (Hoffman et al., 2004), and various extracellular matrix and membrane components (Marshak-Rothstein, 2006). In general, these studies were well performed and indicated that endogenous ligands generated by tissue damage may negatively affect both the immune system and the overall health and growth of animals. Yet, all proposed endogenous TLR ligands should be rigorously tested to eliminate the potential confounding effects of contamination by known PAMPS (Tsan and Baochong, 2007).
|
TOLL-LIKE RECEPTORS LEAD DOWN MANY ROADS |
|---|
|
TopAbstractINTRODUCTIONTOLL-LIKE RECEPTORS RECOGNIZE...TOLL-LIKE RECEPTORS LEAD DOWN...MAKING AND BREAKING MUSCLE...CONCLUSIONSLITERATURE CITED |
|---|
). Myeloid differentiation factor-88 (MyD88) was the first TLR adapter protein to be discovered, and it binds to all of the TLR except TLR3. Myeloid differentiation factor-88 and TLR network by homotypic binding of their TIR domains. Myeloid differentiation factor-88 acts as a scaffolding protein that recruits the IL receptor-associated kinase (IRAK)-4 and sets in motion a series of phosphorylation and translocation events, including phosphorylation of IRAK-1, activation of TNF receptor-associated factor-6, phosphorylation of transforming growth factor-β-activated kinase, and finally activation and phosphorylation of the kinase that phosphorylates the inhibitor of NFB or IB kinase (Loiarro et al., 2007). Phosphorylation of IB makes the inhibitor protein susceptible to ubiquitinylation and subsequent degradation by the proteasome. This series of events allows for the translocation of NFkB into the nucleus where NFkB activates the promoters of cytokines and other components of the inflammasome (Campbell and Perkins, 2006).
Myeloid differentiation factor-88 knockout mice tend to be more susceptible to infections, and loss of MyD88 makes mice profoundly unresponsive to TLR2, -4, -5, -7, and -9 ligands (Weighardt et al., 2002). Myeloid differentiation factor-88-null mice exhibit a shift in the bacterial burden to peripheral tissues such that a 3-wk infection with Borrelia burgdorferi, the organism that causes Lyme disease, left blood and splenic levels of the organism unaltered compared with wild-type mice but increased bacterial load in skeletal muscle 100-fold (Behera et al., 2006). As a result, the MyD88-deficient mice exhibit increased inflammation in skeletal muscle and elevated abundance of IFN mRNA. These results indicate that other adapter proteins are present in skeletal muscle and that they mediate the response to the outer surface lipoproteins that are present on B. burgdorferi. In addition, they indicate that animals harboring polymorphisms in TLR, adapter proteins, or both, may be more susceptible to the effect of infection on muscle mass (Carpenter and O’Neill, 2007; O’Neill and Bowie, 2007).
We found that treatment of myocytes with TLR ligands can activate various signaling components in the immune pathway described above. One of the first events detected is the phosphorylation of IRAK-1, and this occurs within 1 min of treatment with LPS (Frost et al., 2004). Lipopolysaccharide also stimulates the phosphorylation and subsequent degradation of IB and the activation of an NFB reporter plasmid in C2C12 cells (Frost et al., 2002). These results indicate that LPS signals down the classical NFB pathway in muscle. Activation of NFB in myocytes is also consistent with the ability of LPS and other TLR ligands to induce the expression of cytokines in muscle both in vivo and in vitro (Frost et al., 2002; Lang et al., 2003).
The IB kinase and the proteasome are necessary for TLR-induced cytokine expression in skeletal myocytes. Pretreatment of C2C12 cells with either an IB kinase (IKK)2 inhibitor or proteasome inhibitor completely blocked LPS-induced IL-6 synthesis. The 2 inhibitors also blocked the ability of a TLR2 ligand to induce IL-6 mRNA transcription, indicating that both TLR2 and -4 utilize the NFB pathway in muscle. This line of reasoning is strengthened by our results showing that LPS also stimulates the expression of luciferase when it was driven by an IL-6 promoter construct, and this activity is blunted if an NFB binding site in the promoter was mutated (Frost et al., 2006). It is likely that we have only begun to identify the genes that are regulated by NFB in muscle. Nuclear factor B is a generic term for transcription factors that contain a Rel homology domain that allows the proteins to form hetero- and homodimers, and as a result, at least 15 different combinations are possible. At present, hundreds of NFkB-regulated genes have been identified (Carmody and Chen, 2007).
The most direct proof of a role for NFB in the etiology of muscle wasting comes from studies of transgenic mice engineered to express a constitutively active IKK2 solely in skeletal muscle. Mice overexpressing IKK2 exhibited greater activation of NFB than wild-type mice, and severe muscle wasting was evident by a 50% decrease in fiber diameter (Cai et al., 2004). Because the wasting phenotype was blocked by crossbreeding IKK transgenic mice with mice carrying the IB superrepressor, it is highly likely the wasting phenotype is due to NFB activation and not activation of other pathways.
Muscle from the IKK-overexpressing mice released 2.5-fold more Tyr than wild-type muscle when incubated ex vivo, indicating that the muscle wasting is in part due to muscle protein breakdown. At least 50% of the change in muscle mass could be attributed to NFB activation of the ubiquitin ligase muscle ring finger-1, because mating the mice overexpressing IKK2 with muscle ring finger-1 knockout mice generated offspring in which the IKK-induced decrease in muscle mass was ameliorated. The above findings are consistent with the ability of salicylate (an IKK inhibitor) to block atrophy in IKK-overexpressing mice and an IKK2 inhibitor to ameliorate muscle wasting in a murine model of human acquired immune deficiency syndrome (Heckmann et al., 2004).
The TLR4 interacts with additional scaffolding proteins that lack well-defined TIR domains. One prominent example is the Jun-N-terminal kinase (JNK) interacting protein (JIP)-3. In general, the JIP proteins allow for JNK to interact with other proteins and thus facilitate its activation by upstream kinases (mitogen-activated protein kinase kinase kinase-1 (MAP3K-1), as well as its ability to phosphorylate substrates. It has been shown that JIP-3 interacts with the last 13 carboxy-terminal AA of TLR4, and deletion of these AA abolished LPS-induced but not anisomycin-induced JNK activation. Therefore, JIP-3 juxtapositions TLR4, MAP3K-1, and JNK, which allows for the phosphorylation of substrates, such as c-Jun (Matsuguchi et al., 2003).
In this context, we found that a JNK inhibitor (i.e., SP600125) completely blocked LPS-induced IL-6 and NOS2 expression in skeletal myocytes (Frost et al., 2003a, 2004). We also reported that JNK inhibition prevents TNF-induced changes in IGF-I mRNA transcription (Frost et al., 2003b). These results indicate that JNK activation by PAMPS and cytokines may negatively regulate both the expression and signaling of growth factors in muscle cells. There is also evidence that JNK mediates the inhibitory effect of TNF on the differentiation of myoblasts to myotubes and the negative effects of mitochondrial dysfunction on insulin receptor substrate-1 expression and glucose uptake in C2C12 myotubes (Lim et al., 2006; Strle et al., 2006). Thus, both NFB and JNK signaling are likely to be important pathways in the negative regulation of muscle mass.
Applying the Brakes to TLR Signaling
Mammals have evolved numerous mechanisms to keep TLR signaling in check to limit the scope and duration of inflammation. A naturally occurring soluble form of TLR4 that inhibits LPS signaling and similar forms of TLR2 were detected in human plasma (Iwami et al., 2000; LeBouder et al., 2003). This indicates that cells may be able to regulate their response to pathogens by actively creating soluble inhibitors to pathogen-associated molecules.
The adapter protein MyD88 exists in a long and short form due to alternative splicing of its mRNA. When the short form of the protein interacts with the long form to create a heterodimer, MyD88 no longer recruits IRAK-1 to TLR4, and the complex fails to phosphorylate IRAK-4 and activate NFB (Janssens et al., 2003). Interestingly, the MyD88 short form does not affect JNK phosphorylation, thereby providing a tool to differentiate between the importance of the NFB and JNK pathways in models of muscle wasting.
Finally, TLR can also be ubiquitinylated and degraded by the proteasome to limit their activity. Triad3a is a ubiquitin ligase that, when overexpressed, degrades TLR-4 and -9 but not TLR-2 or -3 (Chuang and Ulevitch, 2004). Heat shock protein (HSP)90 regulates Triad3 turnover by a direct protein-protein interaction. However, when binding is disrupted, the increase in free Triad3 triggers the degradation of TLR as well as the TNF receptor associated protein receptor-interacting protein (RIP; Fearns et al., 2006). The binding of Triad3 to HSP90 may explain the ability of agents that disrupt HSP90 to alter LPS signaling and prolong survival in murine models of sepsis (Chatterjee et al., 2007; Hsu et al., 2007).
|
MAKING AND BREAKING MUSCLE PROTEIN: IT’S LOST IN THE TRANSLATION |
|---|
|
TopAbstractINTRODUCTIONTOLL-LIKE RECEPTORS RECOGNIZE...TOLL-LIKE RECEPTORS LEAD DOWN...MAKING AND BREAKING MUSCLE...CONCLUSIONSLITERATURE CITED |
|---|
B activation provides a strong stimulus for the degradation of muscle protein, accumulating evidence from rodent studies suggests LPS and sepsis also alter protein synthesis in skeletal muscle. The negative input of infection is largely mediated by disruption of signaling from the mTOR to downstream effectors of translation initiation (Lang et al., 2007). The mTOR is a scaffold protein and kinase that integrates the positive effects of growth factors and nutrients as well as the negative effects of energy deprivation (Dann et al., 2007). Septic rats exhibit altered phosphorylation of multiple mTOR substrates including ribosomal protein S6 and S6-kinase-1 (S6K1), the translational inhibitor eukaryotic initiation factor-4E binding protein-1, as well as mTOR itself. Septic rats are also resistant to the anabolic affects of the AA Leu on mTOR (Lang and Frost, 2004).
Recently, we examined the effect of LPS on protein synthesis in C2C12 myocytes and found that LPS alone was not sufficient to replicate the decrease in protein synthesis that is observed in vivo (R. A. Frost and C. H. Lang, unpublished results). In contrast, the combination of LPS and IFN decreased protein synthesis by almost 80% in myotubes, and this result was consistent with IFN priming cells to respond more robustly to LPS, TNF, or IL-1β (Kapur et al., 1999). The autophosphorylation of mTOR and its substrates S6K1 and 4E binding protein-1 were also decreased in LPS/IFN-treated myocytes. A comparable reduction in the phosphorylation of ribosomal protein S6 was also observed, indicating that S6K1 activity was decreased in the presence of LPS and IFNβ.
Although LPS alone could not diminish mTOR signaling or protein synthesis in myocytes, it induced the expression of IL-6 and NOS2 (Frost et al., 2004). We speculated that simultaneous treatment with LPS and IFN enhanced the magnitude and duration of NOS2 expression in C2C12 cells by stabilizing NOS2 mRNA (Di Marco et al., 2005). Thus, elevated levels of NO expressed for a protracted period of time could cause enhanced damage to regulatory proteins in growth-signaling pathways via nitrosylation. Indeed, we observed that agents that either blunted NOS2 expression or inhibited NOS2 activity prevented both the LPS/IFN-induced decrease in protein synthesis and changes in mTOR signaling (R. A. Frost and C. H. Lang, unpublished results).
It has not been completely established how mTOR activity is altered during sepsis, but the kinase activity is chiefly controlled by an upstream guanosine triphosphatase, designated the Ras homolog enriched in brain. The tuberous sclerosis complex (TSC)2 protein acts as a guanosine triphosphatase-activating protein for Ras homolog enriched in brain and stimulates mTOR activity. Formation of a heterodimer between TSC2 and its homolog, TSC1, antagonizes the mTOR signaling pathway (Li et al., 2004). The TSC2 protein can be phosphorylated by protein kinase B (Akt), and this disrupts the TSC1/2 complex to activate mTOR signaling (Cai et al., 2006). Conversely, TSC2 can be phosphorylated by glycogen synthase kinase 3B and adenosine monophosphate kinase to stabilize the complex and inhibit mTOR (Inoki et al., 2006). After an i.p. injection of LPS in rats, there are no changes in either the amount or phosphorylation of TSC1/2, indicating that other mechanisms are responsible for the observed inhibition of mTOR activity (Lang and Frost, 2005).
One prospective inhibitor of mTOR activity is the proline-rich Akt substrate (PRAS)-40. This Akt substrate associates with mTOR and its coupling protein raptor to prevent the phosphorylation of selective mTOR substrates (Vander Haar et al., 2007; Wang et al., 2007). A TOR-signaling motif in PRAS40 regulates its binding to raptor and, therefore, allows it to be a competitive inhibitor of other mTOR substrates (Fonseca et al., 2007; Oshiro et al., 2007). Phosphorylation of PRAS40 by Akt releases PRAS40 from the regulatory-associated protein of mTOR (i.e., raptor) and may assist the formation of a complex of PRAS40 and 14-3-3 proteins, thus sequestering the protein away from mTOR and favoring the phosphorylation of substrates involved in enhancing translation initiation.
We found that LPS and IFN profoundly inhibit the phosphorylation Akt substrates in C2C12 cells, including PRAS40. These results indicate that local inflammation in skeletal muscle inhibits Akt activity and that this has a direct negative effect on protein translation and mTOR via PRAS40. Although the mechanism by which sepsis alters Akt activity is not known, Akt is a confirmed target for NO action. Donors of NO rapidly inactivate Akt by S-nitrosylation, and mutation of Cys 224 to Ser restores Akt activity (Yasukawa et al., 2005). Although originally described as a mechanism of insulin resistance in diabetes, it is likely that nitrosylation of Akt decreases mTOR signaling in myotubes by enhancing the binding of PRAS40 to raptor, thereby preventing the phosphorylation of mTOR substrates implicated in translation initiation.
|
CONCLUSIONS |
|---|
|
TopAbstractINTRODUCTIONTOLL-LIKE RECEPTORS RECOGNIZE...TOLL-LIKE RECEPTORS LEAD DOWN...MAKING AND BREAKING MUSCLE...CONCLUSIONSLITERATURE CITED |
|---|
). Although this review has primarily focused on TLR4 ligands, such as LPS, many of our findings apply to the ability of a wide variety of pathogens to adversely affect growth via TLR and other recognition systems. The transcription factor NFB is key player in muscle wasting by stimulating the expression of atrogenes involved in muscle protein degradation but also the expression of cytokines and NO that negatively affect anabolic pathways. In the presence of an exaggerated immune response, NO may posttranslationally modify key growth-promoting enzymes such as Akt and limit the phosphorylation of mTOR substrates to curtail growth. Therefore, pharmacological approaches that target components of TLR and or mTOR signaling in skeletal muscle hold promise for alleviating various muscle-wasting diseases.
|
|
Footnotes |
|---|
2 Presented at the Triennial Growth Symposium – Exploring the interface between growth biology and immunology at the annual meeting of the American Society of Animal Science, San Antonio, Texas, July 8 to 12, 2007.
3 Corresponding author: rfrost{at}psu.edu
Received for publication July 31, 2007. Accepted for publication December 31, 2007.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONTOLL-LIKE RECEPTORS RECOGNIZE...TOLL-LIKE RECEPTORS LEAD DOWN...MAKING AND BREAKING MUSCLE...CONCLUSIONSLITERATURE CITED |
|---|
B transcription factor and induces IL-6 and TNF expression in 3T3-L1 adipocytes. J. Nutr. 135:1841–1846. Am. J. Physiol. Regul. Integr. Comp. Physiol. 279:R2156–R2163.B and calcineurin. Infect. Immun. 74:6829–6838.B activation causes severe muscle wasting in mice. Cell 119:285–298.[CrossRef][Medline]B function. Biochem. Soc. Symp. 73:165–180.[Medline]B: Activation and regulation during Toll-like receptor signaling. Cell. Mol. Immunol. 4:31–41.[Medline]B-mediated MyoD decay during muscle wasting requires nitric oxide synthase mrna stabilization, HUR protein, and nitric oxide release. Mol. Cell. Biol. 25:6533–6545. decreases insulin-like growth factor-I messenger ribo-nucleic acid expression in C2C12 myoblasts via a Jun N-terminal kinase pathway. Endocrinology 144:1770–1779.B- and AP-1-dependent gene expression. FEBS Lett. 548:103–107.[CrossRef][Medline] impairs heart and skeletal muscle protein synthesis by altering translation initiation. Am. J. Physiol. Endocrinol. Metab. 282:E336–E347.B activity and IL-1β processing. Science 307:734–738.-3 vs. -6 lipid emulsions exert differential influence on neutrophils in septic shock patients: Impact on plasma fatty acids and lipid mediator generation. Intensive Care Med. 29:1472–1481.[CrossRef][Medline] suppression of differentiation in myoblasts. Endocrinology 147:4363–4373.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
