Báo cáo Y học: Regulation of mammalian translation factors by nutrients pot

MINIREVIEW Regulation of mammalian translation factors by nutrients Christopher G. Proud Division of Molecular Physiology, School of Life Sciences, University of Dundee, MSI/WTB Complex, Dow Street, UK Protein synthesis requires both amino acids, as precursors, and a substantial amount of metabolic energy. It is well established that starvation or lack of nutrients impairs pro- tein synthesis in mammalian cells and tissues. Branched chain amino acids are particularly effective in promoting protein synthesis. Recent work has revealed important new information about the mechanisms involved in these effects. A number of components of the translational machinery are regulated through signalling events that require the mam- malian target of rapamycin, mTOR. These include transla- tional repressor proteins (eukaryotic initiation factor 4E-binding proteins, 4E-BPs) and protein kinases that act upon the small ribosomal subunit (S6 kinases). Amino acids, especially leucine, positively regulate mTOR signalling thereby relieving inhibition of translation by 4E-BPs and activating the S6 kinases, which can also regulate translation elongation. However, the molecular mechanisms by which amino acids modulate mTOR signalling remain unclear. Protein synthesis requires a high proportion of the cell’s metabolic energy, and recent work has revealed that meta- bolic energy, or fuels such as glucose, also regulate targets of the mTOR pathway. Amino acids and glucose modulate a further important regulatory step in translation initiation, the activity of the guanine nucleotide-exchange factor eIF2B. eIF2B controls the recruitment of the initiator methionyl-tRNA to the ribosome and is activated by insulin. However, in the absence of glucose or amino acids, insulin no longer activates eIF2B. Since control of eIF2B is inde- pendent of mTOR, these data indicate the operation of additional, and so far unknown, regulatory mechanisms that control eIF2B activity. Keywords: translation; elongation factor; mTOR; amino acid; glucose; initiation factor. INTRODUCTION It has long been known that starvation or lack of nutrients influence protein synthesis rates in mammalian tissues and cells. This is not unexpected given that protein synthesis requires both amino acids, as precursors, and metabolic energy. Indeed, protein synthesis is one of the major energy consuming processes of the cell. Recent advances in understanding of the mechanism of translation and its control have facilitated studies at the molecular level into the regulation of protein synthesis by nutrients, and the interplay between nutrients and hormonal signals. An important finding in the last few years is that a number of components of the translational machinery in mammalian cells are subject to acute regulation by the nutrient status of the cell. Regulation of most of these components is linked to the rapamycin-sensitive mTOR (mammalian target of rapamycin) signalling pathway. These targets for mTOR signalling include regulators of translation initiation and elongation, and protein kinases acting on the small ribosomal subunit. This knowledge has allowed investiga- tors to return to the key issue of placing this improved knowledge in a physiological context, and studying the regulation of protein synthesis by nutrients in physiologi- cally important tissues such as skeletal muscle. This article reviews our current understanding of the regulation of translation factors by nutrients and recent studies applying this information to tissues such as pancreatic b-cells, skeletal muscle and heart. Early data suggested that, in muscle in vivo, the rate of elongation may limit protein synthesis under fed conditions [1] while initiation may be limiting in starved animals [2]. Recent work has improved our understanding of the molecular mechanisms involved in regulating both translation initiation and elongation. Early studies focused on the control of protein synthesis in skeletal muscle, as it is a tissue of particular importance for whole body protein metabolism. Overnight fasting led to the disaggregation of polyribosomes in rat skeletal muscle [1] indicating an impairment of translation initiation. Fasting of animals for longer periods involved an additional reduction in the levels of ribosomes in the tissue, manifested as a fall in its RNA content (the bulk of cellular RNA is ribosomal RNA) [3]. In this article I shall discuss mecha- nisms by which nutrients regulate both translation initiation and ribosome biogenesis. REGULATION OF eIF4E BY eIF4E-BINDING PROTEINS The eukaryotic initiation factor (eIF) 4E binds to the 5¢-cap structure of eukaryotic mRNAs and likely provides the first contact between the translational machinery and the mRNA in de novo translation initiation. eIF4E also interacts with several types of protein binding partners. One class comprises the scaffold proteins of the eIF4G group (eIF4G I Correspondence to C. G. Proud, Division of Molecular Physiology, School of Life Sciences, University of Dundee, MSI/WTB Complex, Dow Street, Dundee, DD1 5EH, UK. Fax: + 44 1382 322424, Tel.: + 44 1382 344919, E-mail: c.g.proud@dundee.ac.uk (Received 2 August 2002, revised 23 September 2002, accepted 3 October 2002) Eur. J. Biochem. 269, 5338–5349 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.03292.x and eIF4G II ). eIF4G interacts with a number of other proteins. These include eIF4A, an RNA helicase; the poly(A)-binding protein, PABP; the multisubunit initiation factor, eIF3, which provides a link to the 40S subunit; and the eIF4E kinases, Mnk1 and Mnk2 [4,5] (Fig. 1A). The eIF4E/4G/4A complex is often referred to as the eIF4F complex, although the other components are also likely to be associated with such complexes under physiological conditions. Such complexes are thought to be of key importance in mediating normal, cap-dependent, transla- tion initiation. The second group of eIF4E-interacting proteins comprises low molecular mass proteins that bind to the same (or an overlapping) site on eIF4E and block its interaction with eIF4G (Fig. 1A). In mammals, three eIF4E-binding proteins are known (4E-BP1/2/3) and they will be discussed below. A third type of partner for eIF4E is the nucleocytoplasmic shuttling protein, 4E-T [6], which is considered to be important in conveying eIF4E from the cytoplasm into the nucleus, where a significant proportion of the cellular eIF4E is found [7]. These proteins (eIF4Gs, 4E-BPs, 4E-T) share a common motif through which they interact with the same, or overlapping, sites in eIF4E [6,8]. Binding is therefore mutually exclusive and, for example, eIF4E bound to 4E-BP1 cannot interact with eIF4G to form initiation complexes. 4E-BP1 thus acts as a repressor of cap-dependent translation [9,10]. Of the three 4E-BPs, 4E-BP1 is easily the most intensively studied and best understood. It undergoes phosphorylation at multiple sites in vivo.AsindicatedinFig.1B,thesesites are located almost throughout its short sequence of around 118 amino acids, only the N-terminus being devoid of sites of phosphorylation. Phosphorylation of 4E-BP1 shows a marked hierarchy in vivo [11,12](X.Wang,W.Li,J.L.Parra, A. Beugent & C.G. Proud, unpublished observations). Phosphorylation of the threonines near the N-terminus is required for modification of Thr70, while phosphorylation at Thr70 is required for phosphorylation of Ser65. Earlier data suggested that Ser65 and Thr70 were the most important sites for modulating the binding of 4E-BP1 to eIF4E – phosphorylation at Thr70 promotes its release and phosphorylation at Ser65 may prevent rebinding (Fig. 1B). Phosphorylation of Ser112, at the extreme C-terminus, also appears to be required for release of 4E-BP1 from eIF4E ([14]; C. G. Proud, unpublished data). Phosphorylation of several sites in 4E-BP1 is increased by agents that activate protein synthesis, such as insulin. Phosphorylation of Ser65 and Thr70, and to a lesser extent, Thr37/46, is blocked by rapamycin, indicating an essential role for mTOR in signalling from, e.g., the insulin-receptor to 4E-BP1 (reviewed in [15]). The complex nature of the hierarchy of phosphorylation of the other sites in 4E-BP1 suggests that they are targets for a range of proline-directed kinases that also await identification. Several kinases have been found to phosphorylate 4E-BP1 in vitro, including Erk16, but it remains to be established which kinases actually act on 4E-BP1 in vivo. Several of these, especially those acting at Ser65 and Thr70, the sites most profoundly affected by rapamycin, may be regulated by nutrients through the mTOR pathway. REGULATION OF 4E-BP1 BY AMINO ACIDS In a number of kinds of mammalian cells, amino acids exert marked effects on the phosphorylation and regulation of 4E- BP1 (reviewed in [15, 1 17]). For example, when Chinese Fig. 1. Regulation of eIF4E by 4E-BPs (A) and schematic structure of 4E-BP1 (B). eIF4E (which binds the 5¢-cap of the mRNA, not shown but see text) can interact either with 4E-BPs (such as 4E-BP1, shown) or eIF4G, but not both simultaneously. Phosphorylation of 4E-BP1 at multiple sites induces its release from eIF4E, allowing it to bind to eIF4G and form functional initiation factor complexes that also contain the poly(A)-binding protein PABP, the RNA helicase eIF4A and the eIF4E kinases (such as Mnk1). eIF4G also interacts with eIF3 and thereby recruits the 40S subunit to the 5¢-end of the mRNA. eIF4E can also interact with the shuttling protein 4E-T which is involved in transferring eIF4E to the nucleus: this again involves a binding site on eIF4E that is occluded by 4E-BP1. A number of agents increase the phosphorylation of 4E-BP1 including amino acids and insulin (see text). Regulation by these stimuli is blocked by rapamycin, indicating an essential role in this for signalling via mTOR. (B) 4E-BP1 contains a binding motif for interaction with eIF4E, and two regulatory domains have also been identified – the RAIP motif towards the N-terminus [110] and the TOS motif at the extreme C-terminus [111]. Six sites of phosphorylation have been identified. All except S112 (numbering based on human sequence) are Ser-Pro or Thr-Pro sites (Ser112 is fol- lowed by Gln). Inhibition of mTOR or amino acid withdrawal results in dephosphorylation of a number of sites in 4E-BP1, especially Ser65 and Thr70 (underlined), although Thr37/46 are also affected. Insulin stimulates phosphorylation of Ser65, Thr70 and Ser112, while Ser83 appear to be basally phosphorylated. Thin arrows indicate interplay between sites of phosphorylation that underlies the complex hierarchy of phosphorylation events, while thick arrows indicate the roles of specific sites in regulating the function of 4E-BP1. For example, phos- phorylation at Thr37/46 is required for phosphorylation at Thr70, and phosphorylation at Thr70 is required for phosphorylation at Ser65. Ó FEBS 2002 Control of translation factors by nutrients (Eur. J. Biochem. 269) 5339 hamster ovary cells are transferred to a medium lacking amino acids, 4E-BP1 undergoes dephosphorylation [18–20], which occurs within 15–30 min after amino acid withdrawal. In control cells, in medium containing amino acids, little or no 4E-BP1 is bound to eIF4E and high levels of eIF4F complexes are present. Removal of amino acids quickly causes a marked increase in the amount of 4E-BP1 associated with eIF4E and loss of eIF4F complexes [18,19,21]. These effects are similar to those of adding rapamycin suggesting that the effects of amino acids are mediated via the mTOR pathway. They are reversed, within minutes, by the readdi- tion of amino acids. The most effective single amino acid is leucine, others having very little or no effect in most cells. This role for leucine is a feature that 2 will be discussed later. However, even at concentrations well above those present in the cells’ normal medium, leucine cannot induce the level of 4E-BP1 phosphorylation seen in amino acid replete cells, suggesting that the other amino acids present in growth medium are also important in this effect. These data show that amino acids themselves, in the absence of hormones such as insulin, have marked effects on the phosphorylation of 4E- BP1 and on the formation of translation initiation factor complexes. They imply that mammalian cells have the ability to sense the prevailing availability of amino acids and to relay this information to the translational machinery. The situation is subtly different in certain other types of cells: in human embryonic kidney cells, maintained in medium without serum, basal 4E-BP1 phosphorylation is much lower, so that much of the eIF4E is bound to 4E-BP1 and levels of eIF4F complexes are accordingly low [22,23]. Upon addition of insulin or the phorbol ester, TPA, 4E-BP1 undergoes phosphorylation leading to its release from eIF4E and formation of eIF4F complexes. This does not happen in cells maintained in medium lacking amino acids (E. Hajduch & C. G. Proud, unpublished). In contrast, in adult cardio- myocytes, there is no such requirement for external amino acids, as insulin can induce phosphorylation of 4E-BP1 in cells kept in amino acid-free medium (L. Wang 4 &C.G. Proud, unpublished data). Similarly, in primary adult rat adipocytes insulin can bring about the phosphorylation of 4E-BP1 in the absence of any external amino acids [24,25]. Insulin elicits an increase in 4E-BP1 phosphorylation in amino acid-replete cells, and it can still do so to some extent in CHO cells deprived of amino acids, provided that a metabolizable glucose analogue (or other metabolizable hexose such as D -mannose) is also present [19]. Glucose increases the basal level of phosphorylation of Thr70, but has little effect on basal phosphorylation at Ser65 or Thr37/ 46. However, the presence of glucose does allow insulin to elicit phosphorylation at these sites [19]. Glucose thus exerts a permissive effect with respect to the action of insulin and promotes the release of 4E-BP1 from eIF4E to allow formation of eIF4F complexes. This may reflect an input from metabolic energy to the control of 4E-BP1, perhaps via modulation of the activity of mTOR [26]. This will be discussed in more detail below. Overall, these data indicate a requirement both for amino acids (especially leucine) and an energy source for activation of this key step in translation initiation. This clearly makes excellent sense – amino acids are the precursor for protein synthesis, leucine being an essential amino acid, and protein synthesis consumes a large proportion (perhaps 20–25% [27]) of total cellular energy. REGULATION BY AMINO ACIDS IN PRIMARY TISSUES AND CELL TYPES While many groups have studied the control of translation factors in established cell lines, Kimball, Jefferson and colleagues have extensively investigated the effects of amino acids on 4E-BP1 in primary cells and in tissues. These include isolated adipocytes, perfused liver and skeletal muscle, studied both in vitro and in vivo (reviewed in [28– 30]). Their data again indicate that amino acids exert a positive effect on the phosphorylation of 4E-BP1, with leucine being effective when given alone [31–34]. In muscle, amino acids were required for insulin to enhance the formation of complexes between eIF4E and eIF4G and the rate of protein synthesis. These authors have argued that insulin may serve a permissive function here: for example, giving leucine alone, orally, to rats activates protein synthesis and translation initiation as effectively as a complete meal, but without a rise in plasma insulin concentration [35]. This again underlines the primary role of leucine in regulating protein synthesis. In mice with defective insulin signalling (with similarities to type II diabetes) feeding still stimulates protein synthesis in a similar way to the effects observed in control animals [36]. However, marked reduction in circulating insulin levels (achieved using anti-insulin Ig) does impair the response to feeding [37–39]. Taken together these data indicate that insulin is required for the feeding-induced activation of translation, but that increases in insulin levels may not be, thus suggesting insulin plays a permissive role here. REGULATION OF THE S6 KINASES The protein kinases that phosphorylate ribosomal protein S6 are a second set of proteins that are regulated via mTOR and implicated in the control of mRNA translation [40,41] (Fig. 2). Through alternative splicing, the S6 kinase 1 (also termed a)and2(b) genes each give rise to two distinct proteins, yielding a total of four S6 kinases. The activation of S6K1 and S6K2 by all stimuli so far tested (e.g. insulin, growth factors and phorbol esters) is blocked by rapamycin [40,42,43]. Activation of the S6Ks involves their phosphorylation at multiple sites, some of which lie in the catalytic domain or its Fig. 2. The S6 kinases. The structures of S6K1 and S6K2 are depicted schematically, including their splice variants (forms I and II is each case). The domains within each sequence are indicated, as are major sites of phosphorylation that are associated with activation of these enzymes, and the nuclear localization signals (NLS) in S6K1 I and the S6K2 isoforms. See text for further information. 5340 C. G. Proud (Eur. J. Biochem. 269) Ó FEBS 2002 so-called ÔextensionÕ or ÔlinkerÕ while the majority are located in the C-terminal regulatory domain [40,41] (Fig. 2). Thr229 in the T-loop of the catalytic domain has been shown to be phosphorylated by phosphoinositide-dependent kinase 1 (PDK1) in vitro [44,45]. The protein kinases responsible for phosphorylating the other sites await conclusive identifica- tion. While mTOR can phosphorylate T389 in vitro, it is not clear that it is the physiological T389 kinase (discussed in [40]). The sensitivity of S6K regulation to rapamycin nevertheless shows that mTOR makes an essential input to the control of the S6Ks. The interplay between the phosphorylation sites in S6K1 is complex and for a more detailed discussion the reader is directed to a recent comprehensive review [40]. For the present purposes, this can be summarized to say (a) that phosphorylation of the sites in the C-terminus of S6K1 is believed to facilitate access (by the relevant kinases) to Thr229 and Thr389, phos- phorylation of which is critical for activity; (b) phosphory- lation is again ordered, with modification of Thr229 and especially Thr389 occurring later or even last in the hierarchy; and (c) phosphorylation of multiple sites is sensitive to rapamycin and therefore to signalling via mTOR (reviewed in [40,46]). AMINO ACIDS ARE POSITIVE REGULATORS OF S6K1 In common with the 4E-BPs, the activity and regulation of S6K1 is also sensitive to the nutrient status of the cell [40], although differences are again seen between cell types. Amino-acid-replete CHO cells show a substantial basal activity of S6K1 that is further enhanced by addition of insulin [18,21]. When cells are transferred to amino-acid-free medium, basal activity falls sharply and S6K1 is refractory to stimulation by insulin. Removal of even a single amino acid (especially of arginine or leucine) leads to a marked fall in S6K1 activity [20]. Addition of amino acids partially restores both basal activity and insulin-responsiveness, although the addition of both amino acids and glucose is required for substantial recovery of both effects [18,20,21]. This situation shows similarities and differences with respect to that described above for 4E-BP1/eIF4F in these cells. In both cases, amino acids exert very marked effects in CHO cells: however, whereas amino acids/glucose suffice for complete formation of eIF4F complexes and to maintain a high level of phosphorylation of phosphorylation of 4E-BP1, full activation of S6K1 requires inputs from both amino acids/glucose and insulin. It is currently unclear whether the requirement, in CHO cells, for both amino acids and an additional input (e.g. from insulin) reflects effects of these agents on different (subsets of) phosphory- lation sites in S6K1. In this context it is notable that Hara et al. [20] reported that addition of high levels of amino acids to CHO cells overexpressing the insulin receptor resulted in as high a degree of activation of S6K1 as was observed with normal levels of amino acids plus insulin. This suggests that amino acids can elicit the full response if present at sufficiently high levels. It seems likely that in CHO cells, and probably in other cell types too, amino-acid- replete cells contain only enough amino acids to give partial activation of S6K1 and insulin provides a further input to its activation. Some cell types appear to contain enough amino acids for regulation of S6K1 even when starved for external amino acids (e.g. hepatoma cells [47]). The fact that such cells become dependent upon external amino acids when treated with a compound that inhibits autophagy suggests that this intracellular supply of amino acids is derived from this form of protein breakdown. Autophagy is especially active in hepatocytes and related cell-types, and this is perhaps why some other cell types are more dependent on external amino acids. The effects on S6K1 of nutrient stimuli and agents such as insulin are blocked by rapamycin [30]. In fact, removal of amino acids from CHO cells leads to effects on 4E-BP1 and S6K1, which are qualitatively similar to those of rapamycin treatment. Similar data have been reported for a number of other cell types including adipocytes and HEK 293 cells [20,48], giving rise to the notion that the effects of amino acids are transmitted via mTOR, although there is no formal evidence for this. Evidence in favour of this idea was provided by Hara et al. [20] who showed that amino acid deprivation led to complete dephosphorylation of S6K1 at T389 (T412 in the numbering system used by these authors), this being a major rapamycin-sensitive and thus mTOR- controlled phosphorylation site in S6K1 [49]. Furthermore, and more importantly, they found that a mutant of S6K1 that is resistant to inhibition by rapamycin was also resistant to the effects of amino acid withdrawal. These data suggest that amino acids may signal to S6K1 via mTOR. There is as yet no data on the regulation of S6K2 by amino acids. However, since, like S6K1, S6K2 is regulated via mTOR, it is likely that amino acids also modulate the activity of this enzyme. Protein synthesis itself consumes amino acids. It has long been known that cycloheximide (an inhibitor of protein synthesis) activates S6K1 [50]. Our recent data show that it facilitates the activation of both S6K1 and 4E-BP1 by insulin in amino-acid-deprived CHO cells (A. Beugnet et al. 5 , unpublished data). Consistent with its effect on 4E-BP1 phosphorylation, treatment of cells with cycloheximide also allows insulin to bring about the release of 4E-BP1 from eIF4E and the formation of complexes between eIF4E and eIF4G. Furthermore, three other inhibitors of protein synthesis (anisomycin, emetine or puromycin) each exert very similar effects. Their effects are not due to release of amino acids into the medium. Taken together the data are consistent with the idea (Fig. 3) that regulation of S6K1 and other targets for mTOR signalling is influenced by the size of an intracellular pool of amino acids whose size is determined by the rates of protein degradation and synthesis, and by the availability of extracellular amino acids (which presumably enter this pool following their transport into the cell [51]). Studies using ÔrealÕ cells, adipocytes and skeletal muscle, generally reflect the data obtained in other, transformed, cell lines. For example, amino acids have been shown to stimulate S6K1 in rat adipocytes, and this effect is blocked by rapamycin [48]. However, insulin can activate S6K1 in isolated rat adipocytes in the absence of added amino acids [52]. Orally administered leucine elicits the phosphorylation of S6K1 in skeletal muscle, and this requires insulin, but not an increase in insulin concentration [53]. In human forearm muscle, branched-chain amino acids elicit phosphorylation of S6K1 [54]. These data are largely similar to those discussed above for the regulation of 4E-BP1 by amino acids. Ó FEBS 2002 Control of translation factors by nutrients (Eur. J. Biochem. 269) 5341 A ROLE FOR S6 PHOSPHORYLATION IN RIBOSOME BIOGENESIS? Work, in particular from the laboratory of George Thomas, has suggested that the S6 kinases may play a role in regulating the translation of a set of mRNAs termed the 5¢-TOP (tract of oligopyrimidine) mRNAs (Fig. 3). This group of mRNAs includes those for each of the ribosomal proteins in mammals, and those for certain other proteins involved in mRNA translation such as the elongation factors eEF1A and eEF2 [55] and the poly(A)-binding protein PABP [56]. These mRNAs are characterized by the presence at their extreme 5¢-ends of a short sequence of pyrimidines (the 5¢-TOP), which represses their translation in serum-deprived cells. Following stimulation of the cells by serum, 5¢-TOP mRNAs such as that for eEF1A shift into polyribosomes (i.e. initiation onto them is presumably enhanced [57,58]). Amino acids themselves also promote increased synthesis of proteins encoded by 5¢-TOP mRNAs such as eEF1A [59] and ribosomal proteins [60]. This amino-acid- and hormone-regulated translational control mechanism provides a way in which synthesis of compo- nents of the translational machinery can be quickly switched on following treatment of mammalian cells by an anabolic/ proliferative stimulus, to increase the cellular capacity for protein synthesis. Defects in this may therefore underlie the small size phenotype of cells or animals in which S6K genes have been knocked out [61,62]. The studies of Jefferies and Thomas [57,58,63] indicated that the translational activation of the 5¢-TOP mRNAs was inhibited by rapamycin – the drug prevented or reversed their increased incorporation into polysomes [58,63]. The same group subsequently reported that expression of a mutant of S6K1 that is relatively insensitive to rapamycin resulted in decreased sensitivity of 5¢-TOP mRNA transla- tion to this drug [63]. This seemed to indicate that S6K1, and perhaps phosphorylation of S6, was involved in activating 5¢-TOP mRNA translation. However, more recent work has challenged this view. Tang et al.[64] demonstrated that 5¢-TOP mRNA translation is enhanced by amino acids (which would be consistent with a role for S6K1) but concluded that, by various criteria, S6 phos- phorylation did not appear to be sufficient for increased 5¢-TOP mRNA translation, at least in response to amino acids. For example, amino acid regulation of 5¢-TOP mRNA translation is still observed in cells in which both alleles of the S6K1 gene are knocked out and in which no phosphorylation of S6 is observed in response to amino acids. This also casts doubt on the role of S6Ks and thus S6 phosphorylation in the control of 5¢-TOP mRNA transla- tion at least in response to amino acids. The above findings underline the need for further work to elucidate the mechanisms by which 5¢-TOP mRNA trans- lation is controlled and to define the cellular functions of the S6Ks, which clearly do include roles in events linked to the control of cell and organism size. One such function that has recently been reported is in the control of the elongation factor eEF2. Elongation factor 2 (eEF2) is regulated through the mTOR pathway and by cellular energy status eEF2 mediates the translocation step of elongation. Phosphory- lation of eEF2 at Thr56 inhibits its activity by preventing it from binding to the ribosome. Phosphorylation of eEF2 is catalysed by eEF2 kinase, an unusual and highly specific enzyme. A more detailed discussion of eEF2 and eEF2 kinase can found in the accompanying article by Browne and Proud [65]. In CHO cells overexpressing the insulin receptor, insulin brings about the rapid dephosphorylation of eEF2, con- comitantly with accelerating the rate of elongation. Both effects were blocked by rapamycin [66]. Insulin also elicits the dephosphorylation of eEF2 in other types of cells [67– 70], and decreases the activity of eEF2 kinase, an effect blocked by rapamycin [66,70]. These data suggested a link between mTOR and the control of eEF2 kinase although the nature of the links between mTOR and eEF2 kinase remained unclear for several years. It is now clear that one link between mTOR and the control of eEF2 kinase involves the phosphorylation of eEF2 kinase at Ser366 by Fig. 3. mTOR positively regulates the phosphorylation and function of the ribosomal protein (rp) S6 kinases and of the eIF4E-binding protein, 4E-BP1. Phosphorylation of S6Ks leads to their activation. S6Ks phosphorylate rpS6, which is considered to play a role in the regulation of the translation of the subset of mRNAs containing a 5¢-terminal tractofoligopyrimidines(5¢-TOP mRNAs). S6Ks also phosphorylate elongation factor 2 (eEF2) kinase leading to a decrease in its activity at basal calcium concentrations. Inactivation of eEF2 kinase facilitates the dephosphorylation of eEF2 and the activation of elongation. There appear to be additional mTOR-dependent inputs into the phos- phorylation and control of eEF2 kinase (see accompanying article by Browne and Proud [65]). Phosphorylation of 4E-BP1 at certain sites (see Fig. 1B) leads to its release from eIF4E, which can then interact with eIF4G to form initiation complexes than can recruit the 40S ribosomal subunit to the 5¢-end of the mRNA. Amino acids and cel- lular energy act as positive modulators of mTOR. The mechanisms by which amino acids exert this effect are unclear, but recent evidence suggests this may involve sensing of intracellular amino acid levels. Insulin and a range of other stimuli increase the phosphorylation of S6Ks and 4E-BP1. However, it is not clear (??) whether they do so by modulating the activity/function of mTOR, or whether they provide separate inputs that nonetheless require the mTOR-dependent input. The question mark by the role of S6 phosphorylation in the translation of 5¢-TOP mRNAs denotes the fact that Tang et al. [64] have recently challenged the prevailing concept that these mRNAs are regulated via S6 kinases/phosphorylation of rpS6, at least in response to amino acids. The question mark by the role of ATP in regulating mTOR activity [26] is to indicate that recent data also suggest a role for the AMP-activated kinase in regulating mTOR signalling in skeletal muscle (see text [88]). 5342 C. G. Proud (Eur. J. Biochem. 269) Ó FEBS 2002 S6K1, which results in decreased activity of eEF2 kinase [69]. Given these inputs from mTOR into the control of eEF2 kinase, one might anticipate that the phosphorylation of eEF2 would be modulated by nutrients. Indeed, insulin cannot fully elicit the dephosphorylation of eEF2 in CHO cells lacking amino acids or glucose [21]. A further input from the nutritional status of the cells to the control of eEF2 phosphorylation appears to be related to cellular ATP levels, and may underlie the requirement for glucose referred to above. It clearly makes physiological sense for protein synthesis to be matched to energy availability: protein synthesis is a major energy-consuming process, using up around 25–30% of total cellular energy [27]. This is discussed further in the article by Browne and Proud [65] that accompanies this one. mTOR Frequent mention of mTOR in the regulation of translation has been made in this article without any discussion, so far, of this protein itself. mTOR is large (around 290 kDa) and its primary sequence indicates the presence of a number of potential functional domains [71] (Fig. 4). These include a series of so-called HEAT domains towards its N-terminus, which are likely to be involved in protein–protein interac- tions and a domain with similarity to lipid kinases towards its C-terminus (Fig. 4). This region shows similarity to the phosphoinositide (PI) kinases although mTOR has not been shown to phosphorylate any lipid substrates. Similar domains are also found in several protein kinases such as the ataxia talengiectasia (mutated) kinase ATM and the related kinase ATR. As noted above, mTOR can also phosphorylate certain proteins (e.g. 4E-BP1 and S6K1, see, e.g [72–74]), at least in vitro and in the presence of unphysiologically high concentrations of Mn 2+ -ions. It remains to be established that its in vivo function is as a protein kinase and that it does indeed act on proteins such as 4E-BP1 and S6K1 in vivo. Nonetheless, an intact kinase domain is essential for the function of yeast TOR (see [71]) or of mTOR [73,75] and rapamycin inhibits the kinase activity of mTOR. In addition to possessing in vitro kinase activity itself, mTOR has been shown to be associated with other protein kinase activities, which can be separated from mTOR, e.g. by treatment of mTOR immunoprecipitates with detergent [76,77]. This illustrates that these kinase activities are not intrinsic to mTOR but, presumably, arise from proteins noncovalently bound to mTOR, perhaps via its HEAT domains. The role of these extrinsic kinases in the functions of mTOR and in downstream signalling to proteins such as 4E-BP1, the S6Ks and eEF2 kinase is a potentially important area for future study. One further profitable area for investigation is likely to be the identification of proteins that interact with mTOR; this may shed important light on the regulation of mTOR and on downstream signalling from mTOR to the translational machinery. Indeed, two recent studies identified a novel 150 kDa protein termed Raptor (regulatory associated protein of mTOR) that interacts with mTOR [78,79]. Although raptor has a positive role in nutrient stimulated signalling, its association with mTOR negatively regulates mTOR kinase activity. Nutrient withdrawal results in an increased association of raptor with mTOR [79]. Biochemical studies show that raptor is required for phosphorylation of 4E-BP1 by mTOR and that raptor also enhances phosphorylation of S6k1 by mTOR [78]. Raptor seems to have a positive role in the regulation of cell size. It was shown to interact with 4E-BP1, especially hypophosphorylated forms of the latter and thus appears to act as a scaffold protein by forming ternary complexes with mTOR and 4E-BP1 (and perhaps also S6k1). Further work will clearly be required to determine how nutrients modulate the mTOR–raptor interaction, and whether raptor promotes phosphorylation of all sites in S6k1 and 4E-BP1, or only specific residues. Other recent work has provided insights into the upstream regulation of mTOR. The proteins hamartin and tuberin (also termed TSC1/2) form a complex that suppresses signalling via mTOR (reviewed in [80]). The genes for TSC1/2 are mutated in people suffering from certain types of benign tumours. TSC2 is phosphorylated by protein kinase B, this providing a potential link between phosphatidylinositide 3-kinase signalling and regulation of mTOR, and there is some evidence that this may result in dissociation of complexes between TSC1 and TSC2, thus relieving inhibition of mTOR. However, it is not clear whether TSC1/2 are also involved in the regulation of mTOR by nutrients. Rapamycin binds as a complex with FKBP12 to a region close to the kinase domain in the primary sequence of mTOR (Fig. 4). Recent work suggests that this region may also bind phosphatidic acid and that such binding may be important for the physiological function of mTOR [81]. The authors suggest that the rapamycin-FKBP12 complex may displace PA from mTOR thereby inhibiting its activity [82]. Although proteins that are regulated by mTOR are controlled by a range of stimuli (amino acids, glucose, insulin, growth factors, G-protein-coupled receptor agon- ists), the control of mTOR itself remains very poorly understood. It has proved hard to observe robust changes in mTOR activity (measured in vitro against 4E-BP1, for example) in response to the various cellular treatments tested. This could be for several reasons – for example, the nature of the antibody used to immunoprecipitated mTOR seems to have marked effects upon the (changes in) activity seen in vitro [83]. One way to assess whether amino acids regulate mTOR itself would be to measure the activity of mTOR extracted (immunopurified) from amino acid-replete or -starved cells. Awidelyusedin vitro assay for mTOR relies on its ability to phosphorylate 4E-BP1 or S6K1. However, using this assay (with S6K1 as substrate), Dennis et al. [26] were unable to observe any effect of amino acid withdrawal on the kinase activity of mTOR. This and several considerations lead to doubts whether mTOR is indeed the physiological T389 Fig. 4. mTOR. The principal features of mTOR are indicated: these include domains termed ÔtoxicÕ or ÔrepressorÕ basedonexperiments performed with the yeast homologue TOR. FRB, FKBP12/rapamycin binding domain. Further information is given in the text. Ó FEBS 2002 Control of translation factors by nutrients (Eur. J. Biochem. 269) 5343 kinase [40]. Furthermore, while Thr389 appears to be the final site in the ordered phosphorylation of S6K1, the phosphorylation of other, ÔearlierÕ, sites is sensitive to rapamycin (see [40] and above) strongly implying that mTOR makes further inputs to the regulation of S6K1 additional to any effect it has on Thr389. Such inputs may involve regulation of the (unknown) kinases acting at the C-terminal proline-directed sites or effects on the protein phosphatases acting on S6K1. A potential regulator of the phosphatases acting on S6K1 is a4, which interacts with the catalytic subunit of protein phosphatase (PP)2A [84] and is the mammalian homologue of the yeast phosphatase partner Tap42p [85,86], which in turn is implicated in signalling from yeast TOR to translation. The observations that PP2A interacts with S6K1 and is activated by rapamycin treatment of cells might provide a mechanism by which rapamycin causes dephosphorylation of S6K1 [40,87]. mTOR has a high K m for ATP for its phosphorylation of 4E-BP1 or S6K1 in vitro (around 1 m M [26]). Thus mTOR may act as an energy sensor, its activity decreasing as ATP levels fall. However, this K m value is still considerably lower than normal cellular ATP concentrations and thus ATP levels would have to fall drastically to have a substantial effect on the activity of mTOR (especially given that the relationship between [ATP] and the reaction rate is a hyperbolic rather than a linear one). Indeed, these authors used 2-deoxyglucose at very high concentrations (200 m M ) to deplete ATP. As noted above, effects on another energy sensing system, AMP-activated protein kinase (AMPK), are observed under much milder conditions (1–5 m M 2-deoxy- glucose), where increases in eEF2 phosphorylation have also been seen [67]. Severe energy depletion (such as that obtained with very high 2-deoxyglucose) will interfere with many cellular processes, and it is likely that the Ôenergy- sensingÕ function of mTOR described by Dennis et al.[26] would only come into play under dire cellular conditions! A different mechanism may serve to inhibit translation under conditions of milder energy depletion: this involves activa- tion of 6 the AMPK 7 and inhibition of eEF2, and is discussed in the article by Browne and Proud [65]. Bolster et al.[88] have recently reported that injection of a drug that activates AMPK causes inhibition of mTOR signalling in skeletal muscle. However, activation of AMPK may also interfere with the synthesis and/or release of insulin [89,90], making it hard to interpret these data. The effects on muscle mTOR signalling may, for example, reflect changes in circulating insulin levels. CONTRIBUTION OF mTOR SIGNALLING TO THE REGULATION OF PROTEIN SYNTHESIS IN VIVO An important question is, to what extent do the above regulatory events, linked to mTOR signalling, contribute to the activation of protein synthesis in cells and tissues? This question can be addressed by exploring the effect of rapamycin on the control of overall rates of protein synthesis. In cell lines, rapamycin generally exerts only a small inhibitory effect on the rate of protein synthesis [10]. In skeletal muscle, the activation of protein synthesis elicited by leucine is inhibited by rapamycin, but only partially [91]. This suggests that leucine may operate to stimulate protein synthesis both via mTOR-dependent and -independent pathways. Indeed, leucine is still able to activate muscle protein synthesis in alloxan-diabetic rats where there is no S6K1 phosphorylation or eIF4E/eIF4G binding in response to oral leucine [53]. This points to the operation of perhaps two amino acid regulated responses: firstly, regulation of proteins, such as S6K1 and 4E-BP1, which are linked to mTOR and requires insulin, and, secondly, an insulin-independent pathway which does not involve these two proteins and is therefore distinct from events linked to mTOR (and therefore insensitive to rapamycin). Additional, leucine-sensitive, regulatory inputs must therefore also operate to control protein synthesis in skeletal muscle. Indeed, in L6 myoblasts, regulation of eIF2B (a translation factor not linked to mTOR signalling, see below) appears to be of more importance than the control of eIF4E (by 4E-BP1) for the activation of protein synthesis by amino acids [92]. In isolated rat heart cells (ventricular myocytes), protein synthesis is acutely activated by insulin [70] and this is blocked by around 50% by rapamycin, indicating that mTOR signalling does play an important role in the activation of protein synthesis here. mTOR homologues exist in other eukaryotes, indeed probably in all of them. In yeast there are two homologues, TOR1 and TOR2. Recent studies have revealed roles for these proteins in controlling the cell cycle and cell growth, in transcription and translation, and in the stability of both mRNA and proteins [71,93–97]. It is likely that mTOR also regulates such processes (in addition to the effects on translation described here). OTHER TARGETS FOR THE CONTROL OF TRANSLATION BY NUTRIENTS Cellular nutrition also affects the control of translation factors that are not linked to mTOR signalling, in particular eIF2B. eIF2B is a multisubunit protein that mediates nucleotide-exchange on eIF2, the translation initiation factor that recruits the initiator methionyl-tRNA to the 40S subunit to recognize the start codon during translation initiation [98] (Fig. 5). Binding of eIF2ÆGTPÆMet-tRNAi complexes to the 40S subunit is therefore required for every initiation event. The activity of eIF2B plays a role in regulating overall and transcript specific translational con- trol in eukaryotes from yeast to mammals, and is regulated by a variety of inputs [98]. It can be regulated by amino acids, apparently via several distinct mechanisms, although these do not appear to involve signalling via mTOR. For example, rapamycin does not affect the ability of insulin to activate eIF2B in CHO.T cells [99]. Activation of eIF2B in these cells requires the presence of amino acids and glucose in the medium [21]. These effects do not appear to be connected with defects in the ability of insulin to promote the dephosphorylation of a regulatory (inhibitory) phos- phorylation site at Ser535 in the e-subunit of eIF2B (which still occurs in the absence of glucose or amino acids [21]). In yeast, amino acids regulate the phosphorylation of the a-subunit of eIF2, via the eIF2a kinase GCN2 (Fig. 5). During amino acid starvation, uncharged tRNA accumu- lates and activates Gcn2, leading to phosphorylation of eIF2, to yield eIF2(aP), a potent inhibitor of eIF2B and hence of translation initiation [98]. Inhibition of eIF2B leads to increased translation of the mRNA for GCN4, an 5344 C. G. Proud (Eur. J. Biochem. 269) Ó FEBS 2002 activator of genes required for amino acid biosynthesis, allowing yeast cells to make the necessary amino acids. Although orthologues of Gcn2 exist in mammals, such a mechanism does not seem to be involved in the effects of amino acid withdrawal in CHO cells [21], as this manipu- lation did not affect levels of eIF2a phosphorylation [19,21]. The effects of nutrients on the control of eIF2B therefore appear to be mediated via alternative regulatory mecha- nisms, which may include changes in the phosphorylation of eIF2B. eIF2B undergoes phosphorylation at multiple sites in vivo, and it is quite possible that one or more sites are modulated by amino acids/glucose contributing to the regulation of eIF2B activity. Indeed, our recent data show that dephosphorylation of Ser535 alone does not result in activation of eIF2B, implying the operation of additional regulatory mechanisms [100]. Changes in the activity of eIF2B are implicated in the overall control of protein synthesis in the liver in response to amino acid imbalance (elevated levels of leucine, glutamine and tyrosine), on the basis of correlations between eIF2B activity (but not levels of eIF4F, for example) and overall protein synthetic rates [31]. In L6 myoblasts, histidine and leucine were each able to activate eIF2B and total protein synthesis, while only leucine was able to modulate 4E-BP1, via the mTOR pathway [59,92]. These data also imply that the regulation of the activity of eIF2B, rather than the control of 4E-BP1 (or by implication, S6K1), is important for the overall regulation of protein synthesis in these cells. Control of 4E-BP1 and S6K1, via mTOR, in response to leucine, appears rather to control the translation of specific mRNAs [59]. The molecular mechanisms involved here are unclear, but may involve changes in the activity of a protein kinase that phosphorylates the catalytic e-subunit of eIF2B [31]. This kinase showed decreased activity in response to amino acid imbalance and was probably not one of the kinases previously shown to phosphorylate eIF2Be.This finding further underlines the possibility that nutrient regulation of eIF2B involves changes in its state of phosphorylation. The activity of this unidentified kinase was not affected by rapamycin pretreatment of the cells, again indicating that mTOR is not involved here. Why should mammalian cells have an additional mech- anism to regulate eIF2B in response to amino acids given that they also possess orthologues of GCN2? As discussed above, in yeast, a rise in uncharged tRNA ultimately switches on amino acid biosynthetic pathways. In contrast, mammalian cells are unable to make many of these amino acids, to provide substrates for tRNA charging. An uncontrolled accumulation of uncharged tRNA could have serious consequences for the cell by leading to misincorpo- ration or premature termination during elongation. It may therefore be important for mammalian cells to react to amino acid deficiency before significant accumulation of uncharged tRNAs occurs. HOW DO MAMMALIAN CELLS SENSE AMINO ACIDS? Leucine appears to be the only amino acid capable of eliciting an effect on 4E-BP1 and S6K1 in skeletal muscle [35,91,101], while other branched-chain amino acids (iso- leucine, valine) are also effective in liver [28,60]. In CHO cells, leucine was the only one of the amino acids tested which, when added alone, stimulated 4E-BP1 phosphory- lation [18]. Similarly, it was omission of leucine that had the most profound effect on the activity of S6K1 [20]. Omission of arginine also resulted in a marked fall in S6K1 activity. It seems likely that cell types differ in their sensitivities to the omission or addition of specific amino acids. Whether this reflects the operation of different sensing mechanisms in different cell types awaits further information on the sensing process itself. This raises the key issue of the mechanism by which amino acids are sensed by mammalian cells. Little infor- mation is currently available on this. Iiboshi et al. [102] published information suggesting that levels of tRNA charging may underlie the control of the mTOR pathway. Such effects could conceivably be mediated via mGcn2, which is probably activated by uncharged tRNA. However, the importance of this mechanism on the short term effects of leucine and other amino acids on signalling through the mTOR pathway in questionable. We have consistently been unable to see any change in the state of phosphorylation of eIF2a in response to amino acid addition or withdrawal in CHO cells [19,21]. Only very small changes in eIF2a phosphorylation were seen upon leucine deprivation in L6 myoblasts [92] and Dennis et al. [26] saw no effect of amino acid withdrawal on the level of charging of tRNA. Amino acid alcohols can inhibit amino acyl-tRNA synthetases and thus block tRNA charging. Iiboshi et al. [102] reported that treatment of T-lymphoblastoid (Jurkat) cells with amino Fig. 5. Nutrient inputs into the control of eIF2B in mammalian cells. The role of eIF2B as the guanine nucleotide exchange factor for eIF2 is depicted. In yeast, amino acid starvation leads, via the accumulation of uncharged tRNA, to activation of the eIF2a kinase GCN2 and phosphorylation of eIF2. eIF2(aP) acts a potent inhibitor of eIF2B and thus also of overall translation. The importance of this mechanism in the control of translation by amino acids in mammalian cells is so far less clear. Recent work suggests that also that eIF2(aP) may also positively regulate autophagy [112]. Amino acids do, however, modulate the activity of eIF2B in mammalian cells. Insulin can activate eIF2B by inducing dephosphorylation of an inhibitory site in eIF2B (see text), via a pathway involving PKB and the inactivation of GSK3. Ó FEBS 2002 Control of translation factors by nutrients (Eur. J. Biochem. 269) 5345 acid alcohols led to decreased S6K1 activity. In contrast, neither we ([19]; A. Beugnet & C.G. Proud, unpublished data) 8 nor others [34] have observed effects of amino acid alcohols on targets of mTOR signalling. Inhibition of hepatic protein synthesis by low amino acid levels seems independent of uncharged tRNA [103]. How then are amino acids sensed? Amino acids also regulate (repress) autophagy, e.g. in the liver, and this prompted a number of studies into the mechanisms involved in this effect. Byproducts of leucine metabolism seem unlikely to be involved here (see discussion of [31]) suggesting a role for leucine itself. Mortimore et al. [104,105] were able to show that a non-cell-permeant leucine ÔanalogueÕ could still inhibit autophagy, implying that the effect was mediated by extracellular leucine; they went on to use photoaffinity labelling to show that this reagent could label a membrane-associated protein, suggesting the exist- ence of a plasma membrane leucine ÔsensorÕ [105]. This idea is attractive in view of the discovery of a plasma membrane amino acid sensor (Ssy1p) in yeast [106], but the molecular identity of this protein in mammalian cells has not been established. As mentioned above, other evidence suggests that mTOR may be regulated by intracellular amino acid levels (A. Beugnet, A.R. Tee, P.M. Taylor & C.G. Proud 9 , unpublished data). Various structural homologues of leucine can also mediate the effect [31,107], but the evidence suggests that leucine, rather than its immediate transamination product a-ketoisocaproate, is the mediator of the effect [108]. The specificity for leucine may vary between cell types: while it is the primary regulator in adipocytes, this is not universally the case [109]. Deprivation of amino acids caused a marked depletion of branched chain amino acids [26], consistent with the alternative possibility that vertebrate cells respond to intracellular amino acid levels. Other data consistent with this idea are the observations (a) that injection of leucine into Xenopus oocytes activated TOR signalling (specifically, the phosphorylation and activity of S6K [51] and (b) the finding that manipulation designed to alter intracellular amino acid levels affect mTOR signalling (see above). As mentioned above, blocking autophagy renders hepatoma cells dependent upon added, external amino acids, perhaps by decreasing the pool of intracellular amino acids [47]. A further example of the regulation of translation factors by nutrients is the ability of glucose to modulate the phosphorylation of eIF2a in the pancreas. This appears to involve the unfolded protein response and modulation of the activity of the eIF2a kinase PERK, and may be an important role in regulating insulin synthesis in this tissue (see [113]). FUTURE DIRECTIONS The last four years or so have seen several important advances in our understanding of the control of translation factors by nutrients. However, as is so often the case, the recent data raise even more questions. Particularly import- ant issues concerning the role of mTOR signalling in the control of translation factors include: the nature of the machinery by which amino acids are sensed in mammalian cells, and how this information is relayed to mTOR and the links between mTOR and the control of the S6Ks and 4E-BPs. For example, it is important to identify the protein kinases that act on 4E-BP1 and the S6Ks. The complex hierarchy of phosphorylation of 4E-BP1, in particular, suggests that multiple kinases are involved, some of which may be basally active (due to an input from mTOR) while others may be turned on by insulin. The role of protein phosphatases in the control of 4E-BPs and S6Ks also needs to be explored. In a wider context, it is crucial to identify the other regulatory mechanisms, independent of mTOR, that function to activate protein synthesis in muscle, for example, in response to feeding. ACKNOWLEDGEMENTS Work in the author’s laboratory is supported by the Biotechnology and Biological Sciences Research Council, the British Heart Foundation, The European Union, The Medical Research Council and the Wellcome Trust. REFERENCES 1. Morgan, H.E., Jefferson, L.S., Wolpert, E.B. & Rannels, D.E. (1971) Regulation of protein synthesis in heart muscle. II. Effect of amino acid levels and insulin on ribosomal aggregation. J. Biol. Chem. 246, 2163–2170. 2. Rannels,D.E.,Pegg,A.E.,Rannels,S.R.&Jefferson,L.S.(1978) Effect of starvation on initiation of protein synthesis in skeletal muscle and heart. Am. J. Physiol. Endocrinol. Metab. Cas- trointest. Physiol. 235, E126–E133. 3. Li, J.B., Higgins, J.E. & Jefferson, L.S. (1979) Changes in protein turnover in skeletal muscle in response to fasting. Am.J.Physiol. Endocrinol. Metab. Gastrointest. Physiol. 236, E222–E228. 4. Pyrronet, S., Imataka, H., Gingras, A.C., Fukunaga, R., Hunter, T. & Sonenberg, N. (1999) Human eukaryotic translation initiation factor 4G (eIF4G) recruits Mnk1 to phosphorylate eIF4E. EMBO J. 18, 270–279. 5. Mahalingam, M. & Cooper, J.A. (2001) Phosphorylation of mammalian eIF4E by Mnk1 and Mnk2: tantalizing prospects for a role in translation. Prog. Mol. Subcell. Biol. 27, 131–142. 6. Dostie,J.,Ferraiuolo,M.,Pause,A.,Adam,S.A.&Sonenberg, N. (2000) A novel shuttling protein, 4E-T, mediates the nuclear import of the mRNA 5¢-cap-binding protein, eIF4E. EMBO J. 19, 3142–3156. 7. Strudwick, S. & Borden, K.L. (2002) The emerging roles for translation factor eIF4E in the nucleus. Differentiation 70, 10–22. 8. Haghighat, A., Mader, S., Pause, A. & Sonenberg, N. (1995) Repression of cap-dependent translation by 4E-binding protein 1: competition with p220 for binding to eukaryotic initiation factor-4E. EMBO J. 14, 5701–5709. 9. Pause, A., Belsham, G.J., Gingras, A C., Donze ´ ,O.,Lin,T.A., Lawrence, J.C. & Sonenberg, N. (1994) Insulin-dependent sti- mulation of protein synthesis by phosphorylation of a regulator of 5¢-cap function. Nature 371, 762–767. 10. Beretta, L., Gingras, A C., Svitkin, Y.V., Hall, M.N. & Sonen- berg, N. (1996) Rapamycin blocks the phosphorylation of 4E-BP1 and inhibits cap-dependent translation. EMBO J. 15, 658–664. 11. Gingras, A C., Gygi, S.P., Raught, B., Polakiewicz, R.D., Abraham, R.T., Hoekstra, M.F., Aebersold, R. & Sonenberg, N. (1999) Regulation of 4E-BP1 phosphorylation: a novel two-step mechanism. Genes Dev. 13, 1422–1437. 12. Gingras, A C., Raught, B., Gygi, S.P., Niedzwieka, A., Miron, M., Burley, S.K., Polakiewicz, R.D., Wyslouch-Cieczyska, A., Aebersold, R. & Sonenberg, N. (2001) Hierarchical phosphor- ylation of the translation inhibitor 4E-BP1. Genes Dev. 15, 2852– 2864. 13. Reference withdrawn. 5346 C. G. Proud (Eur. J. Biochem. 269) Ó FEBS 2002 14. Yang, D Q. & Kastan, M.B. (2000) Participation of ATM in insulin signalling through phosphorylation of eIF-4E-binding protein 1. Nat. Cell Biol. 2, 893–898. 15. Gingras, A C., Raught, B. & Sonenberg, N. (2001) Control of translation by the target of rapamycin proteins. Prog. Mol. Subcell. Biol. 27, 143–174. 16. Reference withdrawn. 17. Kimball, S.R. & Jefferson, L.S. (2000) Regulation of translation initiation in mammalian cells by amino acids. In Translational Control of Gene Expression (Sonenberg, N., Hershey, J.W.B. & Mathews, M.B., eds), pp. 561–579. Cold Spring Harbor Labor- atory Press, Cold Spring Harbor, NY. 18. Wang, X., Campbell, L.E., Miller, C.M. & Proud, C.G. (1998) Amino acid availability regulates p70, S6 kinase and multiple translation factors. Biochem. J. 334, 261–267. 19. Patel, J., Wang, X. & Proud, C.G. (2001) Glucose exerts a per- missive effect on the regulation of the initiation factor 4E binding protein 4E-BP1. Biochem. J. 358, 497–503. 20. Hara, K., Yonezawa, K., Weng, Q P., Kozlowski, M.T., Bel- ham, C. & Avruch, J. (1998) Amino acid sufficiency and mTOR regulate p70, S6 kinase and eIF4E BP1 through a common effector mechanism. J. Biol. Chem. 273, 14484–14494. 21. Campbell, L.E., Wang, X. & Proud, C.G. (1999) Nutrients differentially modulate multiple translation factors and their control by insulin. Biochem. J. 344, 433–441. 22. Herbert, T.P., Kilhams, G.R., Batty, I.H. & Proud, C.G. (2000) Distinct signalling pathways mediate insulin and phorbol ester- stimulated eIF4F assembly and protein synthesis in HEK 293 cells. J. Biol. Chem. 275, 11249–11256. 23. Herbert, T.P., Tee, A.R. & Proud, C.G. (2002) The extracellular signal-regulated kinase pathway regulates the phosphorylation of 4E-BP1 at multiple sites. J. Biol. Chem. 277, 11591–11596. 24. Belsham, G.J., Brownsey, R.W. & Denton, R.M. (1980) The effect of insulin and adrenaline on the phosphorylation of a 22 000-molecular weight protein within isolated fat cells; possible identification as the inhibitor-1 of the Ôgeneral phosphataseÕ. Diabetologia 18, 307–312. 25. Diggle, T.A., Moule, S.K., Avison, M.B., Flynn, A., Foulstone, E.J., Proud, C.G. & Denton, R.M. (1996) Both rapamycin-sen- sitive and – insensitive pathways are involved in the phosphory- lation of the initiation factor 4E binding protein (4E-BP1) in response to insulin in rat epididymal fat cells. Biochem. J. 316, 447–453. 26. Dennis,P.B.,Jaeschke,A.,Saitoh,M.,Fowler,B.,Kozma,S.C. & Thomas, G. (2001) Mammalian TOR: a homeostatic ATP sensor. Science 294, 1102–1105. 27. Schmidt, E.V. (1999) The role of c-myc in cellular growth control. Oncogene 18, 2988–2996. 28. Kimball, S.R. & Jefferson, L.S. (2002) Control of protein synthesis by amino acid availability. Curr. Opin. Clin. Nutr. Metab. Care 5, 63–67. 29. Anthony, J.C., Anthony, T.G., Kimball, S.R. & Jefferson, L.S. (2001) Signaling pathways involved in the translational control of protein synthesis in skeletal muscle by leucine. J. Nutr. 131,856S– 860S. 30. Shah, O.J., Anthony, J.C., Kimball, S.R. & Jefferson, L.S. (2000) 4E-BP1 and S6k1: translational integration sites for nutritional and hormonal information. Am.J.Physiol.279, E715–E729. 31. Shah, O.J., Antonetti, D.A., Kimball, S.R. & Jefferson, L.S. (1999) Leucine, glutamine and tyrosine reciprocally modulate the translation initiation factors eIF4F and eIF2B in perfused rat liver. J. Biol. Chem. 274, 36168–36175. 32. Kimball, S.R., Jefferson, L.S., Nguyen, H.V., Suryawan, A., Bush, J.A. & Davis, T.A. (2000) Feeding stimulates protein synthesis in muscle and liver of neonatal pigs through an mTOR- dependent process. Am. J. Physiol. Endocrinol. Metab. 279, E1080–E1087. 33. Balage, M., Sinaud, S., Prod’homme, M., Dardevet, D., Vary, T.C., Kimball, S.R., Jefferson, L.S. & Grizard, J. (2001) Amino acids and insulin are both required to regulate assembly of the eIF4EÆeIF4G complex in rat skeletal muscle. Am. J. Physiol. Endocrinol. Metab. 281, E565–E574. 34. Pham, P.T., Heydrick, S.J., Fox, H.L., Kimball, S.R., Jefferson, L.S. & Lynch, C.J. (2000) Assessment of cell-signalling pathways in the regulation of mammalian target of rapamycin (mTOR) by amino acids in rat adipocytes. J. Cell. Biochem. 79, 427–441. 35. Anthony, J.C., Anthony, T.G., Kimball, S.R., Vary, T.C. & Jefferson, L.S. (2000) Orally administered leucine stimulates protein synthesis in skeletal muscle of postabsorptive rats in association with increased eIF4F formation. J. Nutr. 130, 139–145. 36. Svanberg, E., Jefferson, L.S., Lundholm, K. & Kimball, S.R. (1997) Post-prandial stimulation of muscle protein synthesis is independent of changes in insulin. Am.J.Physiol.35, E841–E847. 37. Preedy, V.R. & Garlick, P.J. (1986) The response of muscle protein synthesis to nutrient intake in postabsorptive rats: the role of insulin and amino acids. Biosci. Report 6, 177–183. 38. Svanberg, E., Zachrisson, H., Ohlsson, C., Iresjo, B.M. & Lundholm, K.G. (1996) Role of insulin and IGF-1 in activation of muscle protein synthesis after oral feeding. Am. J. Physiol. Endocrinol. Metab. 270, E614–E620. 39. Yoshizawa, F., Endo, M., Ide, H., Yagasaki, K. & Funabiki, R. (1995) Translational regulation of protein synthesis in the liver andskeletalmuscleofmiceinresponsetofeeding.Nutr. Biochem. 6, 130–136. 40. Avruch,J.,Belham,C.,Weng,Q.,Hara,K.&Yonezawa,K. (2001) The p70, S6 kinase integrates nutrient and growth signals to control translational capacity. Prog. Mol. Subcell. Biol. 26, 115–154. 41. Fumagalli, S. & Thomas, G. (2000) S6 phosphorylation and signal transduction. In Translational Control of Gene Expression (Sonenberg, N., Hershey, J.W.B. & Mathews, M.B., eds.), pp. 695–717. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 42. Wang, L., Gout, I. & Proud, C.G. (2001) Cross-talk between the ERK and p70, S6 kinase (S6K) signaling pathways. MEK- dependent activation of S6K2 in cardiomyocytes. J. Biol. Chem. 276, 32670–32677. 43. Lee-Fruman, K.K., Kuo, C.J., Lippincott, J., Terada, N. & Blenis, J. (1999) Characterisation of S6K2, a novel kinase homologous to S6K1. Oncogene 18, 5108–5114. 44. Alessi, D.R., Kozlowski, M.T., Weng, Q P., Morrice, N. & Avruch, J. (1998) 3-Phosphoinositide-dependent protein kinase 1 (PDK1) phosphorylates and activates the p70, S6 kinase in vivo and in vitro. Curr. Biol. 8, 69–81. 45. Pullen, N., Dennis, P.B., Andjelkovic, M., Dufner, A., Kozma, S.C., Hemmings, B.A. & Thomas, G. (1998) Phosphorylation and activation of p70 S6k by PDK1. Science 279, 707–710. 46. Pullen, N. & Thomas, G. (1997) The modular phosphorylation and activation of p70s6k. FEBS Lett. 410, 78–82. 47. Shigemitsu, K., Tsujishita, Y., Hara, K., Nanahoshi, M., Avruch, J. & Yonezawa, K. (1999) Regulation of translational effectors by amino acid and mammalian target of rapamycin signaling pathways: possible involvement of autophagy in cultured hepa- toma cells. J. Biol. Chem. 274, 1058–1065. 48. Fox, H.L., Kimball, S.R., Jefferson, L.S. & Lynch, C.J. (1998) Amino acids stimulate phosphorylation of p70 (S6k) and organisation of rat adipocytes into multicellular clusters. Am. J. Physiol. 43, C206–C213. 49. Pearson, R.B., Dennis, P.B., Han, J.W., Williamson, N.A., Kozma, S.C., Wettenhall, R.E.H. & Thomas, G. (1995) The principal target of rapamycin-induced p70 S6k inactivation is novel phosphorylation site within a conserved hydrophobic domain. EMBO J. 14, 5279–5287. Ó FEBS 2002 Control of translation factors by nutrients (Eur. J. Biochem. 269) 5347 [...]... myocytes: implications for cardiac hypertrophy Mol Cell Biol 18, 6729–6736 83 Mothe-Satney, I., Brunn, G.J., Mcmahon, L.P., Capaldo, C.T., Abraham, R.T & Lawrence, J.C (2000) Mammalian target of Ó FEBS 2002 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 Control of translation factors by nutrients (Eur J Biochem 269) 5349 rapamycin-dependent phosphorylation of PHAS-1 in four (S/T) P sites detected by phospho-specific... & Lynch, C.J (1998) Amino acid effects on translational repressor 4E-BP1 are mediated primarily by 1-leucine in isolated adipocytes Am J Physiol 44, C1232–C1238 109 Lynch, C.J (2001) Role of leucine in the regulation of mTOR by amino acids: revelations from structure-activity studies J Nutr 131, 861S–865S 110 Tee, A.R & Proud, C.G (2002) Caspase cleavage of initiation factor 4E-binding protein 1 yields... Horetsky, R.L & Jefferson, L.S (1999) Leucine regulates translation of specific mRNAs in L6 myoblasts through mTOR-changes in availability of eIF4E and phosphorylation of ribosomal protein S6 J Biol Chem 274, 11647–11652 60 Anthony, T.G., Anthony, J.C., Yoshizawa, F., Kimball, S.R & Jefferson, L.S (2001) Oral administration of leucine stimulates ribosomal protein mRNA translation but not global rates of protein... (1997) Rapamycin suppresses 5¢TOP mRNA translation through inhibition of p70S6k EMBO J 16, 3693–3704 64 Tang, H., Hornstein, E., Stolovich, M., Levy, G., Livingstone, M., Templeton, D., Avruch, J & Meyuhas, O (2001) Amino acid-induced translation of TOP mRNAs is fully dependent on phosphatidylinositol 3-kinase-mediated signalling, is partially inhibited by rapamycin, and is independent of S6K1 and... Regulation of protein kinase B and glycogen synthase kinase 3 by insulin and b-adrenergic agonists in rat epididymal fat cells Activation of protein kinase B by wortmannin-sensitive and-insensitive mechanisms J Biol Chem 272, 7713–7719 53 Anthony, J.C., Reiter, A.K., Anthony, T.G., Crozier, S.J., Lang, C.H., Maclean, D.A., Kimball, S.R & Jefferson, L.S (2002) Orally administered leucine enhances protein synthesis... R.K., Cohen, N.A., Snyder, S.H & Sabatini, D.M (1998) RAFT1 phosphorylation of the translational regulators p70, S6 kinase and 4E-BP1 Proc Natl Acad Sci USA 95, 1432–1437 73 Brunn, G.J., Hudson, C.C., Sekulic, A., Williams, J.M., Hosoi, H., Houghton, P.J., Lawrence, J.C & Abraham, R.T (1997) Phosphorylation of the translational repressor PHAS-I by the mammalian target of rapamycin Science 277, 99–101... Avruch, J & Yonezawa, K (1999) Immunopurified mammalian target of rapamycin phosphorylates and activates p70, S6 kinase alpha in vitro J Biol Chem 274, 34493–34498 75 Brown, E.J., Beal, P.A., Keith, C.T., Chen, J., Shin, T.B & Schreiber, S.L (1995) Control of p70, S6 kinase by kinase activity of FRAP in vivo Nature 377, 441–446 76 Nishiuma, T., Hara, K., Tsujishita, Y. , Kaneko, K., Shii, K & Yonezawa,... Characterization of the phosphoproteins and protein kinase activity in mTOR immunoprecipitates Biochem Biophys Res Commun 252, 440–444 77 Heesom, K.J & Denton, R.M (1999) Dissociation of the eukaryotic initiation factor-4E/4E-BP1 complex involves phosphorylation of 4E-BP1 by an mTOR-associated kinase FEBS Lett 457, 489–493 78 Hara, K., Maruki, Y. , Long, X., Yoshino, K., Oshiro, N., Hidayat, S., Tokunaga,... Meyuhas, O & Hornstein, E (2000) Translational control of TOP mRNAs In Translational Control of Gene Expression (Sonenberg, N., Hershey, J.W.B & Mathews, M.B., eds.), pp 671–693 Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY 56 Hornstein, E., Git, A., Braunstein, I., Avni, D & Meyuhas, O (1999) The expression of poly (A)-binding protein gene is translationally regulated in a growth-dependent... Proud, C.G (2002) Evidence that the dephosphorylation of Ser535 in the e-subunit of eukaryotic initiation factor 2B is insufficient for the activation of eIF2B by insulin Biochem J in press 101 Vary, T.C., Jefferson, L.S & Kimball, S.R (1999) Amino acidstimulation of translation in rat skeletal muscle Am J Physiol Endocrinol Metab 277, E1077–E1086 102 Iiboshi, Y. , Papst, P.J., Kawasome, H., Hosoi, H., Abraham, . profoundly affected by rapamycin, may be regulated by nutrients through the mTOR pathway. REGULATION OF 4E-BP1 BY AMINO ACIDS In a number of kinds of mammalian. MINIREVIEW Regulation of mammalian translation factors by nutrients Christopher G. Proud Division of Molecular Physiology, School of Life Sciences, University of