Prolonged sepsis and exposure to an inflammatory milieu decreases muscle protein synthesis and reduces muscle mass. As a result of its ability to integrate diverse signals, including hormones and nutrients, the mammalian target of rapamycin (mTOR) is a dominant regulator in the translational control of protein synthesis. Under postabsorptive conditions, sepsis decreases mTOR kinase activity in muscle, as evidenced by reduced phosphorylation of both eukaryotic initiation factor (eIF)4E-binding protein (BP)-1 and ribosomal S6 kinase (S6K)1. These sepsis-induced changes, along with the redistribution of eIF4E from the active eIF4E·eIF4G complex to the inactive eIF4E·4E-BP1 complex, are preventable by neutralization of tumor necrosis factor (TNF)-α but not by antagonizing glucocorticoid action. Although the ability of mTOR to respond to insulin-like growth factor (IGF)-I is not disrupted by sepsis, the ability of leucine to increase 4E-BP1 and S6K1 phosphorylation is greatly attenuated. This “leucine resistance” results from a cooperative interaction between both TNF-α and glucocorticoids. Finally, although septic animals are not IGF-I resistant, the anabolic actions of IGF-I are nonetheless reduced because of the development of growth hormone resistance, which decreases both circulating and muscle IGF-I. Herein, we highlight recent advances in the mTOR signaling network and emphasize their connection to the atrophic response observed in skeletal muscle during sepsis. Although many unanswered questions remain, understanding the cellular basis of the sepsis-induced decrease in translational activity will contribute to the rational development of therapeutic interventions and thereby minimize the debilitating affects of the atrophic response that impairs patient recovery.
- translation initiation
- insulin-like growth factor I
the net catabolism of skeletal muscle protein is a hallmark of sepsis (10, 30). In the short term, this reprioritization of muscle protein is not detrimental to the host and indeed provides the amino acid substrates to support hepatic gluconeogenesis and acute-phase protein synthesis. However, when sustained there is an erosion of lean body mass, which is causally linked to increased morbidity and mortality. The clinical implications of protracted muscle protein loss include poor wound healing, loss of muscle strength, decreased ambulation, and an increased risk of thromboembolic complications. Hence, a better understanding of the factors responsible for regulating muscle protein balance under both normal and catabolic conditions offers the greatest potential for developing novel therapeutic interventions.
Catabolism of muscle protein is multifactorial, resulting from an increased rate of protein degradation and/or a decreased rate of protein synthesis. The relative contribution of protein synthesis and proteolysis to the overall net catabolic state may vary with the etiology of the insult, and extrapolation of data from sepsis to other inflammatory conditions must proceed with caution. This review focuses on aspects of translational control pertinent to sepsis and is not intended to provide a detailed description of the latest mechanisms regulating translation initiation in general, for which there are numerous excellent reviews (9, 20, 62). Finally, although we emphasize the role of protein synthesis as a contributor to net catabolism, this is not meant to diminish the potential contribution of protein degradation to the atrophic response. Those interested in mechanisms controlling the proteolytic side of the protein balance equation are directed to one of several recent reviews on this topic (18, 37).
Sepsis Decreases Basal Muscle Protein Synthesis
Sepsis inhibits the synthesis of both myofibrillar and sarcoplasmic proteins preferentially in muscles composed of fast-twitch (e.g., gastrocnemius) fibers (34, 58). However, if the infectious insult is of sufficient severity, as might be seen after the administration of a relatively large dose of endotoxin [lipopolysaccharide (LPS)] or the inflammatory cytokine tumor necrosis factor (TNF)-α (both of which are used to mimic the septic response), a more generalized decrease in muscle protein synthesis is observed regardless of fiber type composition (35, 36). There is no definitive evidence that sepsis reduces the number of ribosomes; therefore, the decreased synthetic rate results primarily from a reduction in translational efficiency (58).
Of the three phases of mRNA translation, initiation, elongation, and termination, the majority of metabolic control in muscle occurs during initiation, and hence, the bulk of the published research (8) pertains to sepsis-induced defects in peptide-chain initiation. Translation initiation is regulated by a number of eukaryotic initiation factors (eIFs), many of which are sensitive to hormonal and nutritional signals and function in a cooperative manner. Regulation of initiation occurs at either 1) the attachment of an initiator methionyl-tRNA onto the small 40S ribosomal subunit (formation of the 43S complex) or 2) the loading of the 43S complex onto “capped” mRNA following the melting of any secondary or tertiary structural elements at the 5′ end. Although defects in both of these processes have been reported in response to sepsis (8), the latter have been the focus of most recent investigations and will be highlighted in this review. Moreover, the ability of anabolic hormones and nutrients to increase muscle protein synthesis appears to be mediated primarily by cap-dependent mechanisms.
Sepsis-Induced Changes in Cap-Dependent Translation Initiation
Alterations in eIF4F complex formation.
Loading of the 43S complex onto the mRNA is regulated by the activity of eIF4F (46). This multifactor complex is composed of several proteins, including eIF4E, that directly bind the m7GpppX cap structure, eIF4A, which together with eIF4B unwinds secondary structure via its ATP-dependent RNA helicase activity, and eIF4G, which serves as an adaptor protein. eIF4G is the nucleus around which the initiation complex forms, as evidenced by its binding sites not only for eIF4E but also for eIF4A and eIF3 (19). As a result, eIF4G recruits the 40S subunit to the 5′ end of mRNA and coordinates the circularization of mRNA via its interactions with eIF4E, poly(A)-binding protein (PABP), eIF3, and mitogen-activated protein kinase (MAPK)-interacting kinase (Mnk)1.
The formation of the eIF4F complex exerts control over the rate of initiation by regulating the availability of “free” eIF4E. Sepsis does not alter the total amount of cellular eIF4E but instead alters its distribution between active and inactive complexes. eIF4E bound to eIF4E-binding protein-1 (4E-BP1) allows association with mRNA but prevents binding to eIF4G and, consequently, the formation of the active eIF4F complex. Because the efficient translation of many mRNAs is facilitated by the active eIF4F complex, the sequestration of eIF4E into an inactive complex with 4E-BP1 impedes initiation and, consequently, protein synthesis. The availability of free eIF4E and the assembly of the competent eIF4F complex is controlled by the ordered phosphorylation of 4E-BP1. This phosphorylation is mediated in part by the Ser/Thr protein kinase mammalian target of rapamycin (mTOR), as discussed in greater detail below. 4E-BP1 undergoes site-specific phosphorylation, resulting in the release of eIF4E from the inactive eIF4E·4EBP1 complex, in response to growth factors and some types of nutrients (Fig. 1). As a consequence, eIF4E is then available to bind to eIF4G and stimulate cap-dependent mRNA translation.
A consensus has now emerged pertaining to the ability of sepsis to impair eIF4F formation. In general, septic insults, such as those imposed by cecal ligation and puncture and LPS, decrease the phosphorylation of 4E-BP1, as evidenced by a reduction in either the hyperphosphorylated γ-isoform or phosphorylation of Thr37/46 residue on 4E-BP1 (26, 31, 32, 34). This change is associated with an increased amount of the inactive eIF4E·4E-BP1 complex and a reciprocal decrease in the active eIF4E·eIF4G complex (26, 31, 32, 34). The sepsis-induced reduction in eIF4E·eIF4G in gastrocnemius is believed to diminish the binding of mRNA with the ribosome and thereby limits protein synthesis. None of these changes can be explained by changes in the total amount of eIF4E, only its availability to form the active eIF4F complex. This is an important caveat because sepsis increases muscle proteolysis (18) and eIF4E has been reported to undergo ubiquitination and proteasome-dependent degradation (39). However, the stimulation of such a mechanism and a reduction in total eIF4E protein in muscle does not appear to be operational in bacterial infection.
Initial inroads have been made to define the mechanism responsible for these sepsis-induced changes in cap-dependent translation initiation. The infusion of TNF-α produces changes in 4E-BP1 phosphorylation and eIF4E distribution that mimic sepsis (36). Furthermore, treatment of septic rats with TNF-binding protein (TNFBP), which neutralizes endogenous TNF-α, prevents the reduction in 4E-BP1 phosphorylation and the redistribution of eIF4E from the active to inactive eIF4F complex (34). However, because these are in vivo studies, it is unclear whether TNF-α acts directly or indirectly on muscle. Contradictory findings have been reported when TNF-α is added to cultured myocytes. For example, TNF-α decreases protein synthesis in human myotubes but fails to alter basal rates of protein synthesis in murine C2C12 cells (51, 60). Furthermore, whereas the circulating glucocorticoid concentration is increased in sepsis and glucocorticoids have been shown to impair protein synthesis and translation initiation in vivo and in vitro (46, 47), inhibition of glucocorticoid action by the type II glucocorticoid receptor antagonist RU-486 fails to prevent the sepsis-induced decrease in 4E-BP1 phosphorylation and altered eIF4E availability (34). Therefore, the decreased formation of the eIF4F complex in sepsis under basal postabsorptive conditions appears to be mediated directly or indirectly via TNF-α, but in a manner independent of excess glucocorticoids.
Alterations in eIF4G.
eIF4G is pivotal in coordinating peptide-chain initiation, as evidenced by its multiple binding domains for various initiation factors, mRNA, and associated cofactors (19). The function of eIF4G can be inhibited via its proteolytic degradation by proteases (5) and specifically by activation of caspase-3 and calpain I in other stress conditions (7). Of the two isoforms, eIF4GI and eIF4GII, the former is most prominent in muscle. eIF4GI represents a family of proteins, which arise through different promoters and alternative splicing, and may differentially affect translational activity (6, 19). Future studies need to assess whether sepsis and inflammatory cytokines selectively modulate the amount of various eIF4GI isoforms. Finally, because eIF4G is a scaffold protein, formation of the eIF4F complex can theoretically be compromised if sepsis alters the subsequent binding of other key regulators, including eIF3, eIF4A, and/or PABP (19). Such decrements might suggest a change in the amount of other proteins known to regulate components of the translational apparatus, such as the PABP-interacting proteins (Paip-1 and Paip-2) (24).
Phosphorylation of eIF4G can enhance the formation of the active eIF4E·eIF4G complex (42). Sepsis, LPS, and TNF-α all decrease the extent of eIF4G Ser1108 phosphorylation (31–34). This reduction occurs within hours and can be sustained for several days if the septic/inflammatory condition persists. During sepsis, reduced phosphorylation of eIF4G appears to be mediated by the overproduction of TNF-α, as evidenced by 1) the reduced eIF4G phosphorylation in muscle of TNF-infused rats and 2) the amelioration of the sepsis-induced decrease in eIF4G phosphorylation in animals pretreated with TNFBP (34). However, the upregulation of other cytokines [such as interleukin (IL)-1] may also be involved because administration of the IL-1 receptor antagonist prevents the decreased eIF4G phosphorylation observed after either sepsis or TNF-α administration (56). It is likely that an overproduction of TNF-α is the primary stimulus, whereas the actions of IL-1 represent a secondary or indirect mechanism because elevated tissue and blood levels of TNF-α precede the rise in IL-1.
Alterations in eIF4E phosphorylation.
eIF4E is an essential component of the functional eIF4F complex and undergoes reversible phosphorylation on Ser209, which is likely mediated physiologically by the MAPKs Mnk1 and Mnk2. However, unlike the other protein factors controlling initiation, the effect of various cellular stresses on the phosphorylation of eIF4E is relatively inconsistent. For example, heat shock, nutrient deprivation, and ischemia lead to a dephosphorylation of eIF4E (12, 45). In contrast, catabolism produced by dexamethasone and TNF-α increases eIF4E phosphorylation in muscle (47, 55). The ability of sepsis per se to alter the phosphorylation state of eIF4E in muscle appears variable (55, 57). Hence, a change in eIF4E phosphorylation seems an unlikely mechanism for the sepsis-induced decrease in translation (49). However, the possibility that eIF4E phosphorylation is necessary for translation of a specific subset of mRNAs, as recently posited (2), is intriguing and deserves attention.
Integration of Metabolic and Nutritional Signals: mTOR, Master Integrator
Adult muscle mass is determined by processes regulating cell growth that are controlled by environmental signals (e.g., hormones such as insulin and IGF-I) and nutrient availability (e.g., amino acids). The metabolic consequences of these mediators are predominantly transduced and integrated intracellularly by mTOR. The seminal features of this mTOR-regulated network have been the subject of a number of recent reviews that summarize the results of an ever-increasing number of eloquent studies performed at the molecular and cellular level (9, 48). However, despite its central importance, there is a paucity of data on the role of mTOR in septic conditions.
Sepsis and LPS coordinately decrease mTOR phosphorylation at Ser2481 and Ser2448 independently of a change in the total mTOR protein (26, 32, 34). Phosphorylation of the first site is indicative of autophosphorylation activity of mTOR and is considered central in the activation of the kinase, whereas phosphorylation of the second site is potentially mediated by several kinases, including the ribosomal protein (rp)S6 kinase (S6K)1 (48). In conjunction with the reduced phosphorylation of rpS6 and 4E-BP1, these data collectively suggest a sepsis- or LPS-induced decrease in mTOR kinase activity, although direct measurements of activity are lacking in these conditions.
The mechanism by which sepsis decreases this constitutive mTOR activity remains speculative. The kinase activity of this protein is in part mediated by the activation of Akt/PKB in response to some stimuli (13). However, the activation of Akt, as assessed by Thr308 phosphorylation, is not altered by LPS or peritonitis (32). The mTOR protein is contained within two large protein complexes having distinct functions. One mTOR complex (mTORC1) consists of mTOR, raptor (for regulatory associated protein of mTOR), and GβL (G protein β-subunit-like protein/mLST8) and integrates the stimulatory actions of nutrients and growth factors in a rapamycin-dependent manner (62). In doing so, mTOR phosphorylates S6K1, S6K2, and 4E-BP1. The importance of raptor in mTOR signaling is evidence of the ability of raptor short interfering RNA to inhibit leucine-dependent phosphorylation of S6K1. In contrast, mTORC2 consists of mTOR, GβL, and rictor (for rapamycin-insensitive companion of mTOR; mAVO3) as well as one of three isoforms of mSin1 (mammalian stress-activated protein kinase-interacting protein), and the activation of this protein complex is largely rapamycin insensitive. Hence, mTORC2 primarily regulates where the cell grows by altering the organization of the actin cytoskeleton, whereas mTORC1 determines when a cell will grow. Because the interactions of raptor and rictor with mTOR appear to be mutually exclusive, the ability of sepsis, LPS, or inflammatory stimuli to influence the distribution of mTOR between these complexes and/or the activity of the complexes themselves may have important ramifications for protein balance. As an example, for effective translation to occur, the mTOR·raptor complex must be recruited to the eIF3 preinitiation complex to phosphorylate and concurrently displace S6K1 from eIF3 (21). S6K1 then binds phosphoinositide-dependent kinase-1, which fully activates S6K1, leading to the phosphorylation of eIF4B and rpS6. The phosphorylation of eIF4B at Ser422 speeds its association with eIF3 (48). Whether sepsis or inflammation modulates such protein-protein interactions, either basally or in response to anabolic stimulation, represents a major gap in our understanding of translation initiation under in vivo conditions.
Sepsis-Induced Changes in rpS6 Phosphorylation
Growth factor and nutrient activation of mTOR leads to the phosphorylation of several downstream effector proteins, including 4E-BP1 (as discussed above) and S6K1. The importance of 4E-BP1 in regulating the active eIF4F complex has been described above and is uncontested. However, the mechanism by which S6K1 activity regulates protein synthesis and/or cell size is still evolving (43). S6K1 was originally proposed to increase protein synthesis by phosphorylating rpS6 and thereby enhance the translation of mRNAs having tracts of oligopyrimidines (TOPs) at their 5′ end. This putative mechanism has been discredited because the translation of these TOP-containing mRNAs is not inhibited in cells lacking both S6K1 and S6K2. On the basis of evidence from the deletion of rpS6 from mice and cells (44), the S6 protein per se does not appear to be essential for the translational control of TOP mRNAs. In fact, contrary to expectations, its deletion actually stimulates protein synthesis. It is possible that rpS6 phosphorylation may alter the affinity of the 40S subunit for a specific subset of mRNAs and thereby effect their translation, but this mechanism remains speculative.
Regardless of the controversy surrounding the exact role of rpS6 in translation, S6K1 per se is necessary for normal protein accretion in skeletal muscle. Its role is exemplified by data from studies using S6K1 null mice, which are smaller than their wild-type littermates and exhibit a disproportionate muscle atrophy (40). Moreover, the reduction in myocyte size was comparable when S6K1 was deleted and when myocytes were exposed to rapamycin, which inhibits mTOR kinase activity. The phosphorylation of sites in the linker region of S6K1 (e.g., Thr389) is necessary for the full and maximal activation of the kinase. Unlike the marked phosphorylation of S6K1 at this specific residue in response to nutrient or growth factor stimulation, there is nominal S6K1 phosphorylation in skeletal muscle under postabsorptive in vivo conditions, which makes detecting decreases problematic. Because sepsis and LPS decrease rpS6 phosphorylation under postabsorptive conditions (26, 31, 32, 34), the suppression of another upstream kinase capable of phosphorylating rpS6 is strongly suggested but unproven. Again, the overproduction of TNF-α directly or indirectly modulates part of this effect because pretreatment of septic rats with TNFBP essentially prevents the decrease in phosphorylated rpS6 (34). Although S6K2 also phosphorylates rpS6, we have routinely failed to detect a reproducible sepsis-induced change in the phosphorylation of this kinase in skeletal muscle (unpublished observations).
S6K1 also phosphorylates at least 10 other substrates [e.g., PDCD4, insulin receptor substrate-1 (IRS-1), mTOR, Bcl-2/Bcl-XL antagonist-causing cell death (BAD), SKAR, eIF4B, eukaryotic elongation factor 2K, cAMP response element modulator-τ, and CEP80], some of which potentially regulate the rate of protein synthesis (43). The influence of sepsis on these proteins for the most part has not been reported. However, a negative feedback mechanism has been described where S6K1 phosphorylates IRS-1 Ser307 phosphorylation, thereby blunting the stimulatory actions of growth factors via inhibition of Akt phosphorylation (53). Such a mechanism has been posited to mediate the insulin resistance observed in type 2 diabetic patients with sustained hyperinsulinemia mediated by overproduction of TNF-α (22). However, the physiological relevance of this mechanism in muscle during catabolic states is questionable. For example, Ser307 phosphorylation of IRS-1 is increased in response to either LPS or TNF-α, but this change is associated with a concomitant decrease in S6K1 and rpS6 phosphorylation (22, 35, 36). Furthermore, sepsis does not alter the ability of IGF-I to increase Akt phosphorylation (32). Hence, although there are compelling data to support such a negative feedback mechanism during conditions of nutrient overload, there is little evidence that this pathway is active during wasting conditions (53).
Anabolic Stimulation of Translation Initiation
Role of growth factors.
The anabolic actions of insulin and IGF-I stimulate muscle protein accretion by increasing the synthesis and decreasing the degradation of protein. Hence, a decrease in the circulating concentration, local synthesis, or biological effectiveness of either hormone may contribute to the sepsis-induced wasting. The blood-borne insulin concentration is normal or slightly elevated in sepsis and trauma, with overt hypoinsulinemia, if present, seen only preterminally (28). However, the development of sepsis-induced insulin resistance is well established with respect to stimulation of glucose uptake and inhibition of gluconeogenesis (27). In contrast, there are conflicting data regarding the stimulatory actions of insulin on muscle protein synthesis, with some studies reporting insulin resistance (57) and others normal insulin action (41, 54).
In contrast to the described changes in insulin, sepsis and other catabolic stimuli have been consistently reported (30) to decrease the circulating IGF-I concentration. This decrease can occur rapidly (e.g., within hours), as seen in response to LPS, and persists for days or weeks during chronic septic conditions. Although plasma IGF-I is determined primarily by the hepatic synthesis and secretion of IGF-I, the hormone is also synthesized locally within muscle and is anabolic by the nature of its autocrine/paracrine effects. Although not the focus of this review, data from cultured hepatocytes and myocytes provide evidence that the overproduction of TNF-α and other inflammatory mediators can directly decrease tissue IGF-I (1, 16). Moreover, skeletal muscle is responsive to various bacterial products and can synthesize and secrete a host of immunomodulatory substances such as cytokines and nitric oxide (15, 17).
Regardless of the etiology, the sepsis-induced decrease in either circulating or muscle-derived IGF-I may be partially responsible for the decreased rate of basal muscle protein synthesis. Such a mechanism is supported by in vivo data (29, 30) showing a strong linear correlation between the content of either IGF-I or the amount of eIF4E·eIF4G and protein synthesis in muscle from septic rats. Moreover, muscle protein synthesis in septic rats can be returned to basal values by administering IGF-I complexed with IGF-binding protein-3, which extends the biological half-life of IGF-I (52). Clinically, this binary complex has been shown (11) to partially restore muscle protein balance in burn patients when infused for 5 days. On the basis of in vitro data where IGF-I stimulates translation, the in vivo action of IGF-I to increase protein synthesis is likely mediated by activation of the phosphatidylinositol (PI) 3-kinase/Akt/mTOR pathway and the subsequent phosphorylation of both 4E-BP1 and S6K1/S6 (13). Sepsis does not appear to alter IGF-I signal transduction via this canonical pathway in skeletal muscle (31, 32). In conclusion, studies from a number of independent laboratories (31, 54) indicate that, although sepsis decreases IGF-I bioavailability, tissue responsiveness to the hormone remains relatively unchanged.
Role of amino acids.
Amino acid availability, especially that of the branched-chain amino acid leucine, represents an important nutritional signal responsible for postprandial stimulation of muscle protein synthesis (25). Although the pathway by which leucine sufficiency is detected and communicated intracellularly remains poorly defined, its anabolic effects clearly involve an increase in translation initiation. Addition of amino acids, especially leucine, increases 4E-BP1 and S6K1 phosphorylation in a rapamycin-sensitive manner. Conversely, these proteins are rapidly dephosphorylated when amino acids are depleted in myocytes (9). Hence, leucine stimulates translational control of protein synthesis chiefly by an mTOR-dependent mechanism.
An important clue to the identity of the sepsis-induced defect in translation initiation was revealed by studies (32, 33) reporting the presence of muscle leucine resistance following injection of LPS or severe peritonitis. In both studies, a maximally stimulating dose of oral leucine failed to increase muscle protein synthesis to the same extent as that seen in nonseptic animals. The sepsis-induced leucine resistance could not be attributed to differences in the gut absorption of leucine or its clearance from the circulation because the prevailing plasma leucine concentration was comparable between control and septic rats. Moreover, although leucine increased the plasma insulin concentration, the degree of hyperinsulinemia was comparable between control and septic animals. These data, in conjunction with the inability of leucine to stimulate Akt phosphorylation under in vivo conditions (32), exclude insulin as a possible mediator of the leucine resistance. However, the insensitivity of muscle to leucine during sepsis was associated with its inability to stimulate the phosphorylation of 4E-BP1 and S6K1/rpS6, data consistent with the observed reduction in mTOR phosphorylation. The inability of leucine to hyperphosphorylate 4E-BP1 was also associated with a decrease in the active eIF4E·eIF4G complex in muscle from septic rats. The presence of sepsis-induced leucine resistance is consistent with the generally disappointing clinical trials in which the muscle wasting observed in trauma patients has not been blunted by nutritional supplements containing branched-chain amino acids (10). The mechanism by which sepsis produces leucine resistance despite a preservation of the anabolic actions of IGF-I remains enigmatic.
Subsequent studies further revealed that pretreatment of septic rats with either TNFBP or RU-486 individually only nominally improved leucine-stimulated phosphorylation of 4E-BP1 and rpS6 (33). In contrast, when the in vivo actions of both TNF-α and glucocorticoids were inhibited in tandem, the leucine responsiveness of septic rats was comparable with that of control animals. Although these findings indicate that TNF-α and glucocorticoids interact in a cooperative manner to suppress nutrient signaling in muscle, the cellular mechanism for this cross-talk has not been elucidated.
As illustrated in Fig. 1, mTOR activity can be controlled by the tuberous sclerosis complex (TSC), consisting of TSC1 (hamartin) and TSC2 (tuberin), which lies upstream (23, 50). In this regard, alterations in the amount of TSC1·TSC2 complex are inversely proportional to the extent of phosphorylation of mTOR, S6K1, and 4E-BP1. In addition, expression of a human TSC1 transgene in mouse skeletal muscle leads to a reduction in muscle mass (59). However, in vivo data (32) do not support such a mechanism in muscle after administration of either LPS or leucine. Neither LPS nor leucine alters TSC2 phosphorylation (Thr1462) under in vivo conditions, a signaling event that inhibits TSC1·TSC2 activity. Furthermore, we have been unable to detect an LPS- or leucine-induced change in the amount of the TSC1·TSC2 heterodimer (Lang CH, unpublished observation). Although it remains possible that sepsis-induced leucine resistance results from impaired TSC1·TSC2 activity, an alternative paradigm would indicate that leucine stimulates mTOR-mediated translation initiation independently of TSC (23). This later possibility is supported by data (50) indicating that amino acid withdrawal still results in dephosphorylation of S6K1, S6, and 4E-BP1 in cells lacking TSC2. The ability of amino acids to stimulate mTOR kinase activity in a TSC2-independent manner has been attributed to activation of hVps34, a Class III PI 3-kinase (4). Hence, sepsis may preferentially impair hVps34 activity; however, such an analysis has yet to be completed.
Future Perspectives and Summary
In addition to the specific questions posed herein, several additional unresolved issues remain. First, future studies are needed to address the mechanism by which the translation of specific mRNAs for various inflammatory cytokines (e.g., TNF, IL-1, and IL-6) is upregulated in skeletal muscle in response to sepsis and LPS (15, 16). This response is expected to further inhibit protein synthesis and the local production of IGF-I via an autocrine/paracrine mechanism. One mechanism for such a dichotomous response may be related to the existence of internal ribosome entry sites (IRES) in the 5′ untranslated regions of many cellular mRNAs (3). Such mRNAs may represent as much as 3–5% of the total cellular mRNAs and are translated by eIF4E-independent mechanisms via the recruitment of ribosomes to the IRES. Of interest, many of the mRNAs containing IRES encode proteins that are upregulated by various types of cellular stress and during apoptosis (20). Hence, whether selected inflammatory cytokines also contain IRES and the potential regulatory sequences that mediate the selective recruitment of their mRNA to the translation machinery deserves attention. There is also a paucity of data pertaining to the topology of protein synthesis (61). No information is available on the localization of translation, which is recognized as a mechanism for synthesizing proteins near their cellular site of function. For example, it is possible that sepsis alters the distribution of the eIF4F complex between the cytosolic and mitochondrial pools, and the endoplasmic reticulum, and cytoskeleton. Although such data can be obtained using classical cell fractionation protocols, newer imaging techniques (such as immunofluorescence microscopy) offer a more precise means to delineate the localization of factors in association with and between specific cell organelles within cultured myocytes and whole muscle.
In summary, the aforementioned signal transduction pathways are undoubtedly not the only pathways altered by infection and the resulting inflammatory state (38). However, alterations in these particular pathways are important in homeostatic control and the development of some metabolic diseases, and they therefore represent a logical starting point to search for the molecular mechanisms mediating the metabolic defects induced by sepsis and injury. Although the existence of multiple regulators of translation initiation provides cells with the plasticity to rapidly respond to changes in the hormonal and nutrient milieu, they also make dissection of their function and interrelations more problematic, especially under in vivo conditions. The reductionist approach has unquestionably provided detailed information related to the molecular mechanisms regulating translation initiation; however, results are nonetheless limited by the artificial conditions used to generate them. There remains the need to apply physiological approaches to elucidate whether these mechanisms derived from eloquent cellular and molecular studies are operational at the more expansive organ, organ system, and organism levels.
This work was supported by National Institute of General Medical Sciences Grants GM-38032 and GM-39277.
We apologize to those authors whose work could not be cited owing to space limitations or our inability to draw connections between the primary literature to the general area of infection and inflammation.
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 © 2007 by American Physiological Society