The phosphorylation states of three proteins implicated in the action of insulin on translation were investigated, i.e., 70-kDa ribosomal protein S6 kinase (p70S6k), eukaryotic initiation factor (eIF) 4E, and the eIF-4E binding protein 4E-BP1. Addition of insulin caused a stimulation of protein synthesis in L6 myoblasts in culture, an effect that was blocked by inhibitors of phosphatidylinositide-3-OH kinase (wortmannin), p70S6k (rapamycin), and mitogen-activated protein kinase (MAP kinase) kinase (PD-98059). The stimulation of protein synthesis was accompanied by increased phosphorylation of p70S6k, an effect that was blocked by rapamycin and wortmannin but not PD-98059. Insulin caused dephosphorylation of eIF-4E, an effect that appeared to be mediated by the p70S6kpathway. Insulin also stimulated phosphorylation of 4E-BP1 as well as dissociation of the 4E-BP1 ⋅ eIF-4E complex. Both rapamycin and wortmannin completely blocked the insulin-induced changes in 4E-BP1 phosphorylation and association of 4E-BP1 and eIF-4E; PD-98059 had no effect on either parameter. Finally, insulin stimulated formation of the active eIF-4G ⋅ eIF-4E complex, an effect that was not prevented by any of the inhibitors. Overall, the results suggest that insulin stimulates protein synthesis in L6 myoblasts in part through utilization of both the p70S6k and MAP kinase signal transduction pathways.
- translation initiation
- eukaryotic initiation factor 4E
- eukaryotic initiation factor 4G
- eukaryotic initiation factor 4E binding protein
insulin is an anabolic hormone that stimulates the synthesis of protein, glycogen, and lipid in a variety of tissues (reviewed in Ref. 3). In skeletal muscle, insulin deprivation caused by diabetes or starvation results in a reduction in the global rate of protein synthesis that is reversed within 1–2 h of treating insulin-deficient animals with insulin (16). The alterations in protein synthesis that occur in vivo in response to changes in insulin availability are mimicked in preparations of skeletal muscle perfused in the presence or absence of the hormone (15) and in cultures of serum-starved L6 myoblasts treated with insulin (39). The latter results suggest that insulin acts directly on muscle cells to regulate protein synthesis.
The insulin-mediated regulation of protein synthesis observed in skeletal muscle includes changes in initiation of mRNA translation (15). Translation initiation is a complex process beginning with the binding of initiator methionyl-tRNAi to the 40S ribosomal subunit, progressing through binding of mRNA to the 40S subunit, and ending with the joining of the 60S ribosomal subunit to the 40S initiation complex. We (19, 20) and others (1, 41) have shown that insulin regulates translation initiation in part through changes in the activity of proteins involved in binding mRNA to the 40S ribosomal subunit, i.e., the cap-binding protein eukaryotic initiation factor (eIF) 4E and the eIF-4E binding proteins 4E-BP1 and eIF-4G. eIF-4E binds directly to the 7-methylguanosine 5′-triphosphate (m7GTP) cap structure present at the 5′ end of most eukaryotic mRNAs. The eIF-4E ⋅ mRNA complex then associates with eIF-4G, which binds to, or is already bound to, the 40S ribosomal subunit. The binding of mRNA to the 40S ribosomal subunit is regulated in part by the binding of 4E-BP1 to eIF-4E. Binding of 4E-BP1 to eIF-4E does not prevent eIF-4E from binding to mRNA but does prevent its binding to eIF-4G, at least in in vitro reactions. Insulin and other hormones enhance the phosphorylation of 4E-BP1, resulting in dissociation of the 4E-BP1 ⋅ eIF-4E complex and presumably freeing eIF-4E to bind to eIF-4G.
The signaling pathway(s) utilized by insulin to regulate the interaction of eIF-4E with 4E-BP1 has been studied most extensively in 3T3 cells in culture (2, 25, 41). In these cells, as in skeletal muscle, insulin stimulates phosphorylation of 4E-BP1, with a resultant dissociation of the 4E-BP1 ⋅ eIF-4E complex. Recent studies using compounds that inhibit insulin activation of different intracellular signaling pathways suggest that, in 3T3 cells, insulin promotes phosphorylation of 4E-BP1 and dissociation of the 4E-BP1 ⋅ eIF-4E complex by activation of kinases in the phosphatidylinositide-3-OH kinase (PI 3-kinase)/70-kDa ribosomal protein S6 kinase (p70S6k) pathway (2, 25, 41). Two limitations of the previous reports are that protein synthesis was not measured in any of the studies to show that changes in 4E-BP1 and eIF-4E actually correlate with alterations in protein synthesis, and that, furthermore, the amount of eIF-4G bound to eIF-4E was not measured. In addition, neither the phosphorylation state nor the signaling pathway(s) involved in regulating phosphorylation of eIF-4E was examined.
In the present study, the signaling pathway(s) involved in the regulation of protein synthesis by insulin was investigated using L6 myoblasts in culture. The results suggest that activation of both the mitogen-activated protein kinase (MAP kinase) and PI 3-kinase/p70S6k pathways are involved in the stimulation of protein synthesis by insulin. As in 3T3 cells, insulin stimulated both phosphorylation of 4E-BP1 and dissociation of the 4E-BP1 ⋅ eIF-4E complex through activation of the PI 3-kinase/p70S6k pathway. In addition, changes in eIF-4E phosphorylation caused by insulin were shown to occur through the PI 3-kinase/p70S6k pathway. Surprisingly, although insulin increased the amount of eIF-4G bound to eIF-4E, the enhanced association was not prevented by inhibitors of either the MAP kinase or the PI 3-kinase/p70S6k pathway. In contrast, inhibitors of both signal transduction pathways abolished the stimulation of protein synthesis by the hormone. Overall, the results suggest that the stimulation of protein synthesis by insulin in L6 myoblasts involves activation of both the p70S6k and MAP kinase signaling pathways. Furthermore, neither changes in 4E-BP1 association with eIF-4E nor changes in eIF-4G association with eIF-4E are sufficient to account for the changes in protein synthesis caused by insulin in L6 myoblasts.
Enhanced chemiluminescence (ECL) detection reagents and horseradish peroxidase-conjugated sheep anti-mouse immunoglobulin G (IgG) and donkey anti-rabbit IgG were purchased from Amersham Life Sciences. Polyvinylidene difluoride (PVDF) membrane was obtained from Bio-Rad. [35S]Easytag express protein-labeling mix was from NEN Research Products. Antibodies against p70S6k, extracellular signal-regulated kinase (ERK) 1, and ERK2 were purchased from Santa Cruz Biotechnology. Anti-ACTIVE MAPK polyclonal antibody against MAP kinase phosphorylated on Thr-183 and Tyr-185 was obtained from Promega.
L6 myoblast culture.
L6 myoblast cells were grown in culture in 100-mm dishes in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum (HyClone Labs), 100 U/ml benzylpenicillin, and 100 μg/ml streptomycin sulfate. Cells were grown to ∼40% confluence and then transferred to medium without serum or antibiotics for 24 h. The medium was then supplemented with either no additions, insulin (20 nM when present), insulin and rapamycin (25 ng/ml), insulin and wortmannin (1 μM), or insulin and PD-98059 (25 μM), and 10 min later all dishes received 5 μl of [35S]Easytag Express protein-labeling mix (11 mCi/ml). Thirty minutes after the addition of radiolabel, cells were harvested by scraping inbuffer A [1% Triton X-100, 1% deoxycholate, 50 mM NaF, 1 μM microcystin LR, 1 mM EDTA, 0.1 mM phenylmethylsulfonyl fluoride (PMSF), 0.2 mM benzamidine, 0.8 μM leupeptin, and 0.6 μM pepstatin].
Measurement of protein synthesis in L6 myoblasts.
Protein synthesis was monitored by measuring the incorporation of [35S]methionine and [35S]cysteine into protein as described previously (18).
Protein immunoblot analysis.
Blots were developed using an ECL Western blotting kit as described previously (21). Films were scanned using a Microtek ScanMaker III scanner equipped with a transparent media adapter connected to a Macintosh PowerMac 7100 computer. Images were obtained using the ScanWizard plug-in (Microtek) for Adobe Photoshop and quantitated using National Institutes of Health Image software.
An aliquot of the homogenate from L6 cells was combined with an equal volume of sodium dodecyl sulfate (SDS) sample buffer, and the diluted samples were subjected to electrophoresis on a 7.5% polyacrylamide gel (22). The samples were then analyzed by protein immunoblot analysis using rabbit anti-rat p70S6kpolyclonal antibodies as described above. Typically, on activation, p70S6k resolves into multiple electrophoretic forms following separation by electrophoresis on SDS-polyacrylamide gels, with increased phosphorylation being associated with decreased electrophoretic mobility (43).
MAP kinase activation.
Homogenates from L6 cells were subjected to electrophoresis on a 15% polyacrylamide gel (22). The samples were then analyzed by protein immunoblot analysis using Anti-ACTIVE MAPK polyclonal antibodies against MAP kinase phosphorylated on Thr-183 and Tyr-185, as described above. Duplicate samples were analyzed by protein immunoblot analysis using anti-ERK1 and anti-ERK2 antibodies to quantitate the relative amounts of the two forms of MAP kinase in samples from the L6 myoblasts.
Quantitation of phosphorylated and unphosphorylated eIF-4E in extracts of L6 myoblasts.
Homogenates of L6 myoblasts were diluted with 0.25 volumes ofbuffer B [10% (wt/vol) SDS and 150 mM dithiothreitol (DTT)] and were then heated at 90°C for 5 min. The samples were cooled to room temperature and then diluted with a volume of buffer C (0.1 g DTT, 0.4 g 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate, and 5.4 g urea in 6 ml of water) equal to the original sample volume. The phosphorylated and unphosphorylated forms of eIF-4E were separated by isoelectric focusing on a slab gel and quantitated by protein immunoblot analysis as described previously (19) except that the blot was probed with a monoclonal antibody against eIF-4E instead of the polyclonal antiserum that was used previously.
Quantitation of 4E-BP1 ⋅ eIF-4E and eIF-4G ⋅ eIF-4E complexes.
The association of eIF-4E with 4E-BP1 or eIF-4G was quantitated by a modification of the method described previously (19). Briefly, eIF-4E and the 4E-BP1 ⋅ eIF-4E and eIF-4G ⋅ eIF-4E complexes were immunoprecipitated from aliquots of cell homogenate using the anti-eIF-4E monoclonal antibody. The efficiency of eIF-4E immunoprecipitation was >90% under all conditions. The antibody-antigen complex was collected by incubation for 1 h with goat anti-mouse Biomag IgG beads (PerSeptive Diagnostics). Before use, the beads were washed in 1% nonfat dry milk in buffer D [in mM: 50 tris(hydroxymethyl)aminomethane ⋅ HCl, pH 7.4, 150 NaCl, 5 EDTA, 50 NaF, 50 β-glycerophosphate, 0.1 PMSF, 1 benzamidine, and 0.5 sodium vanadate, as well as 0.1% β-mercaptoethanol and 0.5% Triton X-100]. The beads were captured using a magnetic stand and washed twice with buffer D and once withbuffer D containing 500 rather than 150 mM NaCl. Protein bound to the beads was eluted by resuspending the beads in SDS sample buffer and boiling the sample for 5 min. The beads were collected by centrifugation, and the supernatants were subjected to electrophoresis either on a 10% polyacrylamide gel for quantitation of eIF-4G or on a 15% polyacrylamide gel for quantitation of 4E-BP1 and eIF-4E. Proteins were then electrophoretically transferred to a PVDF membrane as described previously (21). The membranes were incubated with either a mouse anti-human eIF-4E antibody, a rabbit anti-rat 4E-BP1 antibody, or a rabbit anti-eIF-4G antibody for 1 h at room temperature. The blots were then developed using an ECL Western blotting kit as described above.
Examination of 4E-BP1 phosphorylation in extracts of L6 myoblasts.
Aliquots of cell homogenate were immunoprecipitated using a monoclonal antibody raised in mice against rat 4E-BP1 using the method described in the previous section for immunoprecipitation of eIF-4E. The immunoprecipitates were solubilized with SDS sample buffer and then subjected to protein immunoblot analysis using a rabbit anti-rat 4E-BP1 antibody.
In the present study, L6 myoblasts were maintained in serum-free medium for 24 h before treatment with insulin. In confirmation of previous reports (32, 36, 39, 40), treatment of serum-deprived L6 myoblasts with insulin caused a rapid (within 40 min) and significant (P < 0.01) increase in protein synthesis (Fig. 1). Inhibitors of the PI 3-kinase/p70S6k signaling pathway (wortmannin and rapamycin) or the MAP kinase pathway (PD-98059) were employed to determine whether insulin was stimulating protein synthesis through activation of a protein kinase(s) in one or the other of these pathways. Each inhibitor reduced protein synthesis in non-insulin-stimulated cells (hereafter referred to as control cells) to a similar extent. In addition, all three inhibitors abolished the stimulation of protein synthesis by insulin and reduced protein synthesis in insulin-treated cells to values not significantly different from those observed in the absence of the hormone. These results suggest that activation of both the PI 3-kinase and MAP kinase pathways are required for insulin to stimulate protein synthesis in L6 myoblasts. They also suggest that prior activation of these pathways contributes to the rate of protein synthesis observed in control cells.
To confirm that the protein kinase inhibitors were acting in an effective and specific manner in the present studies, the insulin-mediated phosphorylation of p70S6k was examined. In a variety of cell types, insulin activates p70S6k through phosphorylation of the kinase on multiple serine and threonine residues, an effect that is blocked by either wortmannin or rapamycin (4). Therefore, the ability of wortmannin and rapamycin to inhibit the insulin-stimulated phosphorylation of p70S6k was investigated in extracts of L6 cells by protein immunoblot analysis. As shown in Fig. 2, insulin caused a decrease in the electrophoretic mobility of p70S6k and the appearance of multiple electrophoretic forms of the kinase in L6 cells, indicative of phosphorylation of the protein. Ribosomal protein S6 kinase p85 (p85S6k), the nuclear form of the kinase, was also detected by the antibodies. However, no change in the electrophoretic mobility of the 85-kDa form of the kinase was observed in cells treated with insulin compared with controls. Both wortmannin and rapamycin blocked completely the insulin-stimulated phosphorylation of p70S6k, as shown by a return in the mobility pattern of the protein to the faster migrating species. In contrast, PD-98059 had no effect on the insulin-stimulated phosphorylation of p70S6k. Insulin also activates MAP kinase through phosphorylation on multiple residues (reviewed in Ref. 5). In particular, phosphorylation of Thr-183 and Tyr-185 in the 42-kDa form of MAP kinase results in its activation. To further define the effectiveness and specificity of the protein kinase inhibitors used in this study, the ability of PD-98059 to inhibit phosphorylation of MAP kinase was examined in L6 cells by protein immunoblot analysis using an antibody that recognizes only the doubly phosphorylated (i.e., active) form of the kinase. As shown in Fig. 2, insulin increased by approximately twofold the amount of both the 42- and 44-kDa forms of MAP kinase in the doubly phosphorylated form. Neither rapamycin nor wortmannin had any effect on the insulin-stimulated phosphorylated of MAP kinase. In contrast, PD-98059 reduced phosphorylation of MAP kinase to almost undetectable levels. As shown in Fig. 2 C, the amount of MAP kinase was unchanged under any of the conditions, showing that the alterations observed in Fig. 2 A were a result of alterations in the amount of phosphate present in the protein rather than of changes in the amount of the protein. These results demonstrate that the three protein kinase inhibitors were acting in an effective and specific manner.
To further define the mechanism through which insulin stimulates protein synthesis in L6 myoblasts, the effects of wortmannin, rapamycin, and PD-98059 on the insulin-induced changes in phosphorylation of eIF-4E were examined by protein immunoblot analysis of isoelectric focusing slab gels. As shown in Fig.3, during isoelectric focusing, eIF-4E was resolved into two bands representing the phosphorylated and unphosphorylated forms of the protein (20). In control cells, ∼58% of eIF-4E was present in the phosphorylated form. Insulin treatment resulted in a net dephosphorylation of eIF-4E such that <30% of the protein was in the phosphorylated form. Both wortmannin and rapamycin blocked completely the insulin-mediated dephosphorylation of eIF-4E, whereas PD-98059 was without effect. These results suggest that insulin utilizes the PI 3-kinase signaling pathway to regulate the phosphorylation state of eIF-4E.
In a variety of cells and tissues, 4E-BP1 can be resolved into multiple electrophoretic bands, termed α, β, and γ, representing differentially phosphorylated forms of the protein (11, 24, 33). The most highly phosphorylated form of 4E-BP1, the γ-form, exhibits the lowest electrophoretic mobility during SDS-polyacrylamide gel electrophoresis, whereas the least phosphorylated form, the α-form, exhibits the greatest mobility. As shown in Fig.4, insulin caused a threefold increase in the amount of 4E-BP1 present in the γ-form relative to total 4E-BP1, indicating that insulin stimulated phosphorylation of the protein. Both wortmannin and rapamycin blocked completely the insulin-stimulated phosphorylation of 4E-BP1. In addition, rapamycin caused a shift in the distribution among the three forms, such that the amount of 4E-BP1 in the least phosphorylated (α-) form, was increased. In contrast to the results observed with rapamycin and wortmannin, PD-98059 had no significant effect on the insulin-stimulated phosphorylation of the protein.
Phosphorylation of 4E-BP1 is thought to regulate translation initiation by causing dissociation of the 4E-BP1 ⋅ eIF-4E complex (reviewed in Ref. 8). To investigate the association of 4E-BP1 and eIF-4E, eIF-4E was immunoprecipitated from extracts of L6 cells, and the amount of the two proteins in the immunoprecipitate was quantitated by protein immunoblot analysis. As shown in Fig.5, and in agreement with previous studies (9, 24, 25), only the α- and β-forms of 4E-BP1 were found associated with eIF-4E. Insulin caused a decrease in the amount of 4E-BP1 that coprecipitated with eIF-4E from L6 myoblasts to ∼50% of the control value. In agreement with the observed changes in 4E-BP1 phosphorylation (Fig. 4), both wortmannin and rapamycin prevented the dissociation of the 4E-BP1 ⋅ eIF-4E complex caused by insulin.
In a similar manner, the association of eIF-4G and eIF-4E was examined by immunoprecipitation of the eIF-4G ⋅ eIF-4E complex using a monoclonal anti-eIF-4E antibody followed by protein immunoblot analysis to quantitate the amounts of the two proteins. As observed in other cell types, eIF-4G was resolved into multiple electrophoretic forms with apparent molecular masses of 200–220 kDa during SDS-polyacrylamide gel electrophoresis (Fig.6 A). Insulin caused a 2.5-fold increase in the amount of eIF-4G that coprecipitated with eIF-4E. Surprisingly, neither wortmannin, rapamycin, nor PD-98059 had any effect on the amount of eIF-4G in the eIF-4E immunoprecipitate. This result was not due to an increase in the amount of eIF-4E in the cells (not shown). Furthermore, it suggests that insulin must act through a signaling pathway distinct from PI 3-kinase/p70S6k and MAP kinase kinase to regulate the binding of eIF-4E to eIF-4G.
A recent report by Servant et al. (38) shows that the simulation of protein synthesis in muscle cells in culture caused by hormones such as insulin or angiotensin II is associated with activation of both MAP kinase and p70S6k. Treatment of the cells with selective inhibitors of either the MAP kinase or p70S6k signaling pathways partially prevents the hormone-induced enhancement of protein synthesis. Complete prevention is only achieved in the presence of both inhibitors, suggesting that activation of both pathways is required for maximal stimulation of protein synthesis. The results of the present study confirm those reported by Servant et al. (38) and show that activation of both the MAP kinase and p70S6k pathways are required for insulin to stimulate protein synthesis in L6 myoblasts in culture. We extend the earlier study to show that activation of the p70S6k pathway is associated with phosphorylation of two proteins that play important regulatory roles in translation, i.e., eIF-4E and 4E-BP1.
Increased phosphorylation of eIF-4E, as assessed by incorporation of [32P]Piinto the protein, is observed under a variety of conditions associated with enhanced protein synthesis (reviewed in Ref. 8). Specifically, insulin stimulates [32P]Piincorporation into eIF-4E in both NIH/3T3 HIR3.5 cells (27) and NIH/3T3 L1 cells (29, 30). The results of the present study, in which the amount of eIF-4E in the phosphorylated form is reduced in L6 myoblasts treated with insulin, seem to contradict the proposal that hyperphosphorylation of eIF-4E is required for stimulation of protein synthesis. However, studies comparing the rate of [32P]Piincorporation into eIF-4E with the actual amount of the protein in the phosphorylated form suggest that increased protein synthesis results in enhanced turnover of phosphate on the protein independent of increases in the amount of phosphorylated eIF-4E (35, 37). Thus, in L6 myoblasts, insulin may stimulate both protein kinase and protein phosphatase activities, with the stimulation of protein phosphatase activity being greater than that of protein kinase activity, resulting in a net decrease in the amount of eIF-4E in the phosphorylated form. An alternative explanation for the observed changes in eIF-4E phosphorylation is that the kinase that phosphorylates eIF-4E preferentially phosphorylates the protein when it is associated with 4E-BP1. Thus rapamycin and wortmannin would promote phosphorylation of eIF-4E by increasing the amount of the protein bound to 4E-BP1 rather than modulating the activity of an eIF-4E kinase.
In contrast to the results of the present study, insulin has been shown to increase the amount of eIF-4E in the phosphorylated form in Chinese hamster ovary (CHO) cell overexpressing the human insulin receptor (CHO.T cells) (7). However, in CHO.T cells insulin stimulates eIF-4E phosphorylation through a MAP kinase-dependent pathway rather than a p70S6k-dependent pathway as observed in L6 myoblasts in the present study. The difference in response to insulin between CHO.T cells and L6 myoblasts may simply reflect a difference in cell type or may be a result of overexpressing the insulin receptor in a cell that normally has few, if any, insulin receptors. Thus overexpressing the insulin receptor in CHO cells may result in a differential activation of signaling pathways compared with cells normally expressing the insulin receptor.
In addition to causing changes in the phosphorylation of eIF-4E, insulin also stimulates phosphorylation of 4E-BP1 and results in dissociation of the 4E-BP1 ⋅ eIF-4E complex. A similar effect of insulin on 4E-BP1 has been reported in skeletal muscle of diabetic rats (19) and incubated diaphragm preparations (1) as well as in insulin-deprived cells in culture (24, 41). Binding of 4E-BP1 does not preclude binding of eIF-4E to m7GTP-Sepharose (24, 33). Instead, it has been proposed (10, 26) that the release of eIF-4E from the 4E-BP1 ⋅ eIF-4E complex stimulates protein synthesis by allowing eIF-4E to bind to eIF-4G and form the translationally active eIF-4F complex. The basis for this proposal is the observation that in in vitro reactions the binding of 4E-BP1 or eIF-4G to eIF-4E is mutually exclusive (10, 26). Furthermore, the region of the eIF-4G protein that is involved in binding to eIF-4E contains a motif that exhibits strong sequence homology to a region in both 4E-BP1 and 4E-BP2 that has been shown to be important in the binding of these proteins to eIF-4E (23, 26). These results suggest that eIF-4G and 4E-BP1 bind to the same site on eIF-4E. However, the finding that rapamycin and wortmannin prevented the dissociation of the 4E-BP1 ⋅ eIF-4E complex without affecting the increase in the amount of eIF-4G bound to eIF-4E caused by insulin suggests that in vivo the mechanism through which insulin stimulates protein synthesis is more complicated than simply causing a dissociation of the 4E-BP1 ⋅ eIF-4E complex. It is noteworthy that a recent study utilizing a cell line lacking insulin receptor substrate-signaling proteins 1 and 2 suggests that phosphorylation of 4E-BP1 may be necessary but is not sufficient to account for the stimulation of protein synthesis by insulin (28). A possible step that insulin may be stimulating is the binding of the mRNA ⋅ eIF-4E complex to the 40S ribosomal subunit. Two mechanisms have been described whereby mRNA becomes bound to 40S ribosomal subunits by cap-dependent mechanisms (reviewed in Ref. 13). In both cases, eIF-4E binds directly to the m7GTP cap at the 5′ end of the message. The mRNA ⋅ eIF-4E complex then binds either to eIF-4G that is already bound to the 40S ribosomal subunit or to free eIF-4G followed by binding of the mRNA ⋅ eIF-4E ⋅ eIF-4G complex to the subunit. The results of the present study could be explained if the latter mechanism were correct and if that step were stimulated by insulin. Thus we speculate that insulin stimulates both dissociation of the 4E-BP1 ⋅ eIF-4E complex as well as the binding of the mRNA ⋅ eIF-4E ⋅ eIF-4G complex to the 40S ribosomal subunit.
A second explanation for the results observed in the present study is that a step distinct from eIF-4E becomes rate limiting in insulin-deprived cells. We (14, 17) and others (12) have shown that insulin stimulates the activity of eIF-2B in skeletal muscle and cells in culture. eIF-2B is a guanine nucleotide exchange factor that mediates the exchange of GDP bound to a second factor, eIF-2, for free GTP (reviewed in Ref. 31). Under a variety of conditions, including amino acid and serum deprivation, the activity of eIF-2B becomes rate limiting for protein synthesis. The apparent discrepancy in the results of the present study, in which eIF-4E ⋅ eIF-4G complex formation is stimulated by insulin in the absence of an increase in protein synthesis, could easily be explained if the activity of eIF-2B were rate limiting in insulin-deprived L6 myoblasts and if PD-98059 prevented the stimulation of eIF-2B activity by insulin. According to this model, rapamycin and wortmannin could either act to prevent the simulation of eIF-2B activity or, as described above, could cause a different step, such as binding of the eIF-4E ⋅ eIF-4G complex to 40S ribosomal subunits, to become rate limiting. It is noteworthy that glycogen synthase kinase-3 (GSK-3) phosphorylates eIF-2B (42) and that the activity of GSK-3 may be regulated by insulin in L6 myoblasts through the MAP kinase pathway (6). Thus it has been proposed that insulin regulates the activity of eIF-2B through modulation of the activity of GSK-3 (34).
The results of the present study clearly show that phosphorylation of 4E-BP1 and dissociation of the 4E-BP1 ⋅ eIF-4E complex occur through a rapamycin- and wortmannin-sensitive pathway. In addition, the changes in eIF-4E phosphorylation that occur in response to insulin involve activation of the same signaling pathway. Finally, the results also show that eIF-4G binding to eIF-4E is not sufficient to account for the stimulation of protein synthesis by insulin, although they do not eliminate the possibility that it is required for the stimulation.
We are grateful to Joan McGwire for technical help.
Address for reprint requests: S. R. Kimball, Jr., Dept. of Cellular and Molecular Physiology, The Pennsylvania State University College of Medicine, P.O. Box 850, Hershey, PA 17033.
This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-15658.
- Copyright © 1998 the American Physiological Society