alpha(1A) adrenergic receptor induces eukaryotic initiation factor 4E-binding protein 1 phosphorylation via a Ca(2+)-dependent pathway independent of phosphatidylinositol 3-kinase/Akt.

Phosphorylation of the translation repressor eukaryotic initiation factor 4E-binding protein 1 (4E-BP1) is thought to be partly responsible for increased protein synthesis induced by growth factors. This study investigated the effect of a G(q)-coupled receptor on protein synthesis and the phosphorylation state and function of 4E-BP1 in Rat-1 fibroblasts expressing the human alpha(1A) adrenergic receptor. Treatment of cells with phenylephrine (PE), a specific alpha(1) adrenergic receptor agonist, increased protein synthesis and induced the phosphorylation of 4E-BP1 and its release from translation initiation factor 4E. Although the PE-induced phosphorylation of 4E-BP1 was blocked by the phosphatidylinositol 3-kinase inhibitor LY294002, neither phosphatidylinositol 3-kinase nor Akt, its downstream effector, is activated in cells treated with PE (Ballou, L. M., Cross, M. E., Huang, S., McReynolds, E. M., Zhang, B. X., and Lin, R. Z., J. Biol. Chem. 275, 4803-4809). The effect of PE on 4E-BP1 phosphorylation was also abolished in cells depleted of intracellular Ca(2+) and in cells pretreated with calmodulin antagonists. By contrast, phosphorylation of 4E-BP1 still occurred in cells in which the Ca(2+)- and diacylglycerol-dependent isoforms of protein kinase C were down-regulated by prolonged exposure to a phorbol ester. We conclude that activation of the alpha(1A) adrenergic receptor in Rat-1 fibroblasts leads to phosphorylation of 4E-BP1 via a pathway that is Ca(2+)- and calmodulin-dependent. Phosphatidylinositol 3-kinase, Akt, and phorbol ester-sensitive protein kinase C isoforms do not appear to be required in this signaling pathway.

tion complex, the complex binds to the 5Ј end of the mRNA and translocates to the initiation codon, and then the 60 S ribosomal subunit is added to form an active 80 S ribosome. Binding of the 43 S preinitiation complex to mRNA is mediated by eukaryotic initiation factor (eIF) 1 4F. eIF4F in mammals contains three subunits; one of them, eIF4E, binds directly to the m 7 GpppN (where N is any nucleotide) cap on the 5Ј end of the mRNA. Together with eIF4B, eIF4F unwinds the secondary structure in the 5Ј untranslated region of the mRNA to create a binding site for the 43 S preinitiation complex (1,2). Activation of protein synthesis by growth factors is a complex process that involves phosphorylation of a number of translation initiation factors, regulatory proteins and the 40 S ribosomal subunit (1)(2)(3)(4). eIF4E-binding protein 1 (4E-BP1) is a 12-kDa translation repressor that is thought to be a key player in the regulation of protein synthesis (3). In resting cells, hypophosphorylated forms of 4E-BP1 bind tightly to eIF4E on the mRNA cap, thus preventing formation of a functional eIF4F complex (3,5). Treatment of cells with growth factors leads to phosphorylation of 4E-BP1 on multiple sites and its dissociation from eIF4E, thereby relieving the translational block. Translation of mRNAs with extensive secondary structure at the 5Ј end is thought to be particularly sensitive to regulation by 4E-BP1 (3). Phosphorylation of the S6 protein in 40 S ribosomal subunits is another mechanism that mediates growth factor-induced activation of protein synthesis (4). The major kinase that phosphorylates S6 is the M r ϭ 70,000 S6 kinase (p70 S6 kinase) (4,6). p70 S6 kinase is activated by phosphorylation of the enzyme at multiple sites (7,8). Phosphorylation of the 40 S ribosomal subunit by p70 S6 kinase is thought to selectively up-regulate translation of certain mRNAs that contain a polypyrimidine tract adjacent to the mRNA cap (5Ј-TOP mRNAs; Refs. 9 and 10).
Intense study has been aimed at identifying upstream regulators in the signaling pathways that lead to phosphorylation of 4E-BP1 and p70 S6 kinase. These pathways appear to be quite similar. First, it has been demonstrated in many cell systems that growth factor-induced phosphorylation of both proteins is blocked by the immunosuppressant rapamycin (11)(12)(13)(14). Rapamycin, when bound to its intracellular receptor FKBP12, inhibits the function of the mammalian target of rapamycin (mTOR), a kinase of which the catalytic domain resembles that of phosphatidylinositol (PI) 3-kinase (15,16). mTOR has been found to undergo autophosphorylation and to phosphorylate exogenous protein substrates in a rapamycin/FKBP12-sensitive manner. Indeed, it was recently reported that mTOR in immunoprecipitates phosphorylates 4E-BP1 and fragments of p70 S6 kinase in vitro (17,18). Phosphorylation of recombinant 4E-BP1 was reported to occur on five Ser/Thr-Pro sites (19) that also become phosphorylated in vivo in response to insulin treatment (20). However, more recent reports (18,21) suggest that mTOR phosphorylates 4E-BP1 only at two sites, which serves as a priming event for subsequent phosphorylation of other Ser/Thr-Pro sites by unknown kinases that co-immunoprecipitate with mTOR (22).
A second similarity between the pathways leading to phosphorylation of 4E-BP1 and p70 S6 kinase is their apparent dependence on PI 3-kinase and its downstream effector, the protein kinase Akt. Treatment of cells with wortmannin or LY294002, two inhibitors of PI 3-kinase, prevents the phosphorylation of both proteins following growth factor treatment (23)(24)(25). Likewise, overexpression of a dominant-negative mutant of Akt causes a reduction in insulin-induced phosphorylation of 4E-BP1 (26,27) and p70 S6 kinase (28). Conversely, expression of activated forms of PI 3-kinase (26,29) or Akt (30,31) induces 4E-BP1 phosphorylation and activation of p70 S6 kinase in a rapamycin-sensitive manner. Finally, some mutants of the platelet-derived growth factor (PDGF) receptor that cannot bind PI 3-kinase fail to induce 4E-BP1 phosphorylation (32) and p70 S6 kinase activation (33,34) upon PDGF treatment. These results have led to the proposal of a signaling pathway leading from growth factor receptors to PI 3-kinase, Akt, mTOR, and phosphorylation of 4E-BP1 and p70 S6 kinase (26).
In contrast to this proposed signaling pathway, we recently found that stimulation of the ␣ 1A adrenergic receptor (AR) leads to an increase in p70 S6 kinase activity without activation of PI 3-kinase or Akt (35). ␣ 1 ARs have been implicated in the pathogenesis of cardiac hypertrophy, but little is known about the signaling pathways utilized by these receptors to regulate translation (36). Treatment of rat neonatal cardiac myocytes in vitro with the ␣ 1 AR agonist phenylephrine (PE) was reported to activate p70 S6 kinase and stimulate protein synthesis and hypertrophic cell growth (37). However, the study of these events in cardiac myocytes is complicated by the fact that they express all three of the known ␣ 1 AR subtypes (␣ 1A , ␣ 1B , and ␣ 1D ; Refs. 38 and 39). In this and our recent (35) study, we used Rat-1 fibroblasts that stably express the human ␣ 1A AR as a simplified model system to study signaling pathways that regulate translation. We show here that stimulation of the ␣ 1A AR leads to an increase in protein synthesis accompanied by phosphorylation of 4E-BP1 and its dissociation from eIF4E. This response is not mediated by PI 3-kinase/Akt signaling, but rather by a Ca 2ϩ -and calmodulin-dependent pathway.

EXPERIMENTAL PROCEDURES
Materials-PE was purchased from Sigma; phorbol 12-myristate 13acetate (TPA), A23187, ophiobolin A, and calmidazolium chloride were from Calbiochem (San Diego, CA); BAPTA-AM was purchased from Molecular Probes (Eugene, OR); 7-methylguanosine 5Ј-triphosphate (m 7 GTP) Sepharose 4B was from Amersham Pharmacia Biotech; [ 3 H]leucine (Ͼ140 Ci/mmol) was from NEN Life Science Products; rabbit polyclonal antibodies against 4E-BP1 and protein A-agarose were from Santa Cruz Biotechnology (Santa Cruz, CA); and human recombinant PDGF A/B was from Roche Molecular Biochemicals. All remaining reagents were from common commercial sources.
Cell Culture and Extract Preparation-Rat-1 fibroblasts stably transfected with the human ␣ 1A AR were a gift from G. Johnson of Pfizer Laboratories (40). These cells do not express endogenous ␣ 1 ARs (41). They were maintained in Dulbecco's modified Eagle's medium with 10% fetal bovine serum at 37°C in a humidified environment of 5% CO 2 and 95% air. Cells were seeded in 60-mm dishes at a density of 1 ϫ 10 5 cells/plate and used 3-4 days later when they were 80 -90% confluent. The cells were incubated for 20 -24 h in serum/antibiotic-free medium before starting experimental treatments. For experiments involving Ca 2ϩ , the cells were preincubated for 1 h in high salt glucose buffer (10 mM Hepes, pH 7.4, 140 mM NaCl, 4 mM KCl, 2 mM MgSO 4 , 1 mM KH 2 PO 4 and 10 mM glucose) plus either 2 mM EGTA or 1 mM Ca 2ϩ . Drugs were then added directly to the buffer at the indicated doses. To prepare lysates, cells were washed twice with cold phosphate-buffered saline containing 1 mM sodium orthovanadate, and the cell layers were incubated with cell lysis buffer (1% Triton X-100, 25 mM Hepes, pH 7.5, 50 mM NaCl, 5 mM EDTA, 50 mM NaF, 1 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride, and 10 g/ml each of aprotinin and leupeptin) for 15 min on ice. Homogenates were centrifuged for 15 min at 14,000 ϫ g at 4°C and supernatants were retained. Protein concentration was determined by a Bradford microprotein assay (Bio-Rad).
Immunoblotting-Proteins in cell extracts were separated on SDSpolyacrylamide gels and electrophoretically transferred onto nitrocellulose or polyvinylidene difluoride membranes. Membranes were blocked in 5% nonfat milk for 1 h and then incubated in the primary antibody overnight at 4°C. After washing, membranes were incubated with horseradish peroxidase-linked secondary antibody (Amersham Pharmacia Biotech) for 1 h at room temperature. The signal was visualized using an enhanced chemiluminescence kit (NEN Life Science Products).
p70 S6 Kinase Assay-S6 kinase activity in cell lysates was measured using 40 S ribosomal subunits as substrate as described previously (24). The amount of 32 P incorporated into S6 was quantitated by liquid scintillation counting. One unit of enzyme incorporates 1 pmol of P i into S6 per min.
Binding of 4E-BP1 to m 7 GTP Sepharose-4E-BP1 binding assays were performed as described previously (42) with some modifications. Cell lysates were prepared in lysis buffer as described above. Equal amounts of cell lysate protein were diluted with 4E-BP1 binding assay buffer (50 mM MOPS, pH 7.2, 0.5 mM EDTA, 0.5 mM EGTA, 100 mM KCl, 1 mM dithiothreitol, 50 mM NaF, 80 mM 2-glycerophosphate, 100 M GTP, and 0.5 mM phenylmethylsulfonyl fluoride) to bring the volume to 350 l. Twenty-five microliters of m 7 GTP Sepharose beads were added to each sample and incubated overnight at 4°C. Then, the beads were washed three times with the same buffer, and 4E-BP1 protein bound to the beads was eluted by boiling for 5 min in SDS-polyacrylamide gel electrophoresis sample buffer. The samples were subjected to SDS-polyacrylamide gel electrophoresis, and 4E-BP1 was detected by Western blotting as described above.
Protein Synthesis-Cells were seeded in 12-well culture dishes at a density of 7.5 ϫ 10 4 cells/well, and the next day, they were placed in serum-free medium. Twenty hours later, the cells were treated with 10 M PE or 50 ng/ml PDGF. After 2 h, 1 Ci/ml of [ 3 H]leucine was added to each well, and the cells were incubated for 4 h more in the presence of growth factors. Then, the cells were washed twice with cold phosphate-buffered saline and lysed with 100 l of lysis buffer. One-half ml each of 0.1 mg/ml bovine serum albumin and 20% trichloroacetic acid were added to each sample, and after 30 min on ice, they were passed through glass microfiber filters. The filters were washed twice with 5 ml of 10% trichloroacetic acid and once with 2 ml of ethanol and air dried for 30 min. Radioactivity was determined by liquid scintillation counting. Data were analyzed by one-way analysis of variance using the StatView program (Abacus Concept, Berkeley, CA). Pairwise comparisons were obtained using Fisher's post hoc tests. Values were considered significantly different when p Յ 0.01.

Effect of PE on Protein Synthesis and 4E-BP1
Phosphorylation-To evaluate the effect of ␣ 1A AR stimulation on protein synthesis, Rat-1 fibroblasts stably expressing the receptor were treated with PE and incorporation of [ 3 H]leucine into trichloroacetic acid-insoluble cell material was measured. Protein synthesis in cells treated for 6 h with PE was 34% higher than the basal level in control cells (p Ͻ 0.001; Fig. 1). By comparison, PDGF treatment increased protein synthesis by 41% above the control (Fig. 1).
Phosphorylation of 4E-BP1 at sites that cause it to dissociate from eIF4E is thought to contribute to growth factor-induced increases in the rate of protein synthesis (3). Therefore, we assessed the extent of 4E-BP1 phosphorylation by gel mobility shift assays after exposing Rat-1 cells to PE for various periods of time. 4E-BP1 migrates in SDS-polyacrylamide gels as three species designated ␣, ␤ and ␥. The ␥ band is the most highly phosphorylated species, and the ␣ band is the least phosphorylated form. Stimulation of cells with PE led to a moderate increase in 4E-BP1 phosphorylation, as judged by the mobility shift of the protein on Western blots (Fig. 2). The fraction of 4E-BP1 migrating as the ␥ band reached a maximum after 10 min in the presence of PE, and thereafter it appeared that 4E-BP1 underwent some dephosphorylation, as the ␣ band became more intense (Fig. 2). In contrast to PE, PDGF treatment induced a much higher level of 4E-BP1 phosphorylation at every time examined, and the protein remained highly phosphorylated over the entire time course (Fig. 2).
Sensitivity of PE-induced 4E-BP1 Phosphorylation to Rapamycin and LY294002-Prior studies have indicated that rapamycin, an inhibitor of mTOR, interferes with the pathway that leads to phosphorylation of 4E-BP1 in response to a variety of growth factors (13,14). To evaluate the role of this molecule in signaling by the ␣ 1A AR, cells were preincubated in the presence of rapamycin prior to addition of PE and the phosphorylation state of 4E-BP1 was determined by mobility shift assays. Similar to the results in Fig. 2, the predominant isoform of 4E-BP1 in control cells was the hypophosphorylated ␤ band (Fig. 3A, left panel). PE treatment converted about half of the protein to the hyperphosphorylated ␥ species, whereas in cells exposed to PDGF, essentially all of the 4E-BP1 was converted to the ␥ band. In control cells incubated with rapamycin, the basal level of 4E-BP1 phosphorylation was decreased so that equal amounts of the ␣ and ␤ bands were visible (Fig. 3A, middle panel). The PE-induced phosphorylation of 4E-BP1 was completely inhibited in the presence of rapamycin, and the drug also strongly inhibited phosphorylation of the protein in response to PDGF. p70 S6 kinase activity measured in the same cell extracts was similarly inhibited by rapamycin treatment (Fig. 3C). This result suggests that phosphorylation of 4E-BP1 induced by the ␣ 1A AR is mediated by a pathway that requires mTOR.
We have recently shown that PE treatment of these cells activates p70 S6 kinase with no corresponding increase in PI 3-kinase or Akt activity (35). However, the response of p70 S6 kinase to PE is still inhibited by the PI 3-kinase inhibitor LY294002 (Fig. 3C and Ref. 35). Incubation of cells with LY294002 also reduced the basal phosphorylation of 4E-BP1 in control cells and completely inhibited phosphorylation of the protein in response to both PE and PDGF (Fig. 3A, right panel).
These results indicate that 4E-BP1 phosphorylation and p70 S6 kinase activation induced by the ␣ 1A AR are similar in that both responses are independent of PI 3-kinase/Akt signaling but exhibit sensitivity to LY294002 inhibition.
Binding of 4E-BP1 to eIF4E-Cell stimuli such as insulin induce the phosphorylation of 4E-BP1 at multiple sites, and it is not yet clear how these sites contribute to the control of 4E-BP1 function (20). Because of this complexity, gel mobility shift assays reveal little about the functional state of 4E-BP1. To investigate the effect of ␣ 1A AR stimulation on binding of 4E-BP1 to eIF4E, we measured the amount of 4E-BP1 that was coprecipitated with eIF4E using the mRNA cap affinity resin m 7 GTP Sepharose. Extracts of control and treated cells were incubated with m 7 GTP Sepharose and 4E-BP1 associated with the beads was visualized on Western blots. A large amount of 4E-BP1 was recovered from extracts of untreated control cells, due to tight binding between eIF4E and hypophosphorylated 4E-BP1 (Fig. 3B, left panel; Ref. 5). The amount of 4E-BP1 precipitated by m 7 GTP Sepharose was greatly reduced in ex- tracts of cells treated with PE, consistent with the interpretation that PE-induced phosphorylation of 4E-BP1 disrupts the 4E-BP1-eIF4E complex. A similar result was obtained using extracts of cells treated with PDGF (Fig. 3B, left panel). In addition, we determined that the PE-induced release of 4E-BP1 from m 7 GTP Sepharose and activation of p70 S6 kinase are insensitive to pertussis toxin (data not shown), indicating that these events are mediated by ␣ 1A AR signaling through G q and not G i proteins. m 7 GTP Sepharose binding assays were also done to evaluate the effect of rapamycin and LY294002 on 4E-BP1 function. In agreement with the mobility shift assays (Fig. 3A), rapamycin completely inhibited the PE-induced dissociation of 4E-BP1 from eIF4E, whereas some release of 4E-BP1 from eIF4E still occurred in PDGF-treated cells (Fig. 3B, middle panel). By contrast, pretreatment with LY294002 completely blocked dissociation of the 4E-BP1-eIF4E complex induced by both PE and PDGF (Fig. 3B, right panel). Western blots from an experiment similar to the one shown in Fig. 3B were stripped and reprobed with an antibody to eIF4E. Equal amounts of eIF4E were found in each lane, indicating that PE treatment of cells does not affect the ability of eIF4E to bind to the mRNA cap affinity resin (data not shown). These data show that stimulation of the ␣ 1A AR promotes the phosphorylation of 4E-BP1 at sites that cause it to be released from eIF4E. This mechanism may account at least in part for the PE-induced increase in protein synthesis illustrated in Fig. 1.
Effect of PE on 4E-BP1 and p70 S6 Kinase Is Ca 2ϩ -dependent-As for all G q -coupled receptors, stimulation of the ␣ 1A AR leads to an increase in the intracellular Ca 2ϩ concentration ([Ca 2ϩ ] i ) and activation of diacylglycerol-and Ca 2ϩ -dependent isoforms of protein kinase C (PKC) (36,38,41). To examine the role of Ca 2ϩ in ␣ 1A AR signaling to 4E-BP1, cells were incubated in medium containing EGTA under conditions known to completely abolish the PE-induced increase in [Ca 2ϩ ] i (35). Then, the cells were treated with or without PE, and binding of 4E-BP1 to eIF4E was measured in m 7 GTP Sepharose binding assays. In the presence of Ca 2ϩ , PE induced the phosphorylation of 4E-BP1 and its release from eIF4E (Fig. 4A, left panel). This response was inhibited in Ca 2ϩ -depleted cells, as indicated by an increased amount of 4E-BP1 coprecipitating with eIF4E on the affinity resin. A more pronounced effect was observed in cells treated with BAPTA-AM to chelate intracellular Ca 2ϩ (Fig. 4A, right panel). The amount of 4E-BP1 bound to eIF4E was greatly increased in the presence of the chelator, indicating that both the basal and PE-induced phosphorylation of 4E-BP1 was abolished. Finally, the effect of artificially raising the [Ca 2ϩ ] i using the Ca 2ϩ ionophore A23187 was examined. Very little 4E-BP1 from cells treated with A23187 bound to eIF4E, suggesting that a high [Ca 2ϩ ] i alone is an effective inducer of 4E-BP1 phosphorylation (Fig. 4A, right panel). Thus, functional phosphorylation of 4E-BP1 in response to stimulation of the ␣ 1A AR occurs via a Ca 2ϩ -dependent pathway.
It was recently shown that growth factors, including angiotensin II and PDGF, signal to p70 S6 kinase via a Ca 2ϩ -dependent pathway (43,44). We therefore tested whether PEinduced activation of p70 S6 kinase is also Ca 2ϩ -dependent. Cells were treated as described above to manipulate the [Ca 2ϩ ] i , and p70 S6 kinase activity was measured after various cell treatments. In the presence of Ca 2ϩ , PE activated p70 S6 kinase about 3-fold over the basal level (Fig. 4B, left panel). In Ca 2ϩ -depleted cells the basal level of kinase activity was not changed but PE-induced activation of the enzyme was almost totally blocked. Similarly, chelation of intracellular Ca 2ϩ with BAPTA-AM caused a reduction in basal and PE-induced p70 S6 kinase activity (Fig. 4B, right panel). As seen for 4E-BP1 phos-phorylation (Fig. 4A, right panel), treatment of cells with A23187 also activated p70 S6 kinase (Fig. 4B, right panel). Thus, phosphorylation of 4E-BP1 and activation of p70 S6 kinase in response to stimulation of the ␣ 1A AR are Ca 2ϩ -dependent processes.
Effect of PE on 4E-BP1 Is Dependent on Calmodulin but Not PKC-The increase in [Ca 2ϩ ] i following stimulation of G q -coupled receptors leads to activation of the Ca 2ϩ -and diacylglyceroldependent PKCs. This prompted us to investigate the role of these enzymes in ␣ 1A AR-mediated phosphorylation of 4E-BP1. Cells were treated with or without TPA for 24 h to downregulate PKCs, and then binding of 4E-BP1 to m 7 GTP Sepharose was measured after challenge with an agonist. Treatment of control cells with either PE or TPA for 20 min induced the phosphorylation of 4E-BP1 and disruption of the 4E-BP1-eIF4E complex (Fig. 5A). Pretreatment of cells for 24 h with TPA abolished phosphorylation of 4E-BP1 promoted by a subsequent challenge with TPA, but there was little effect on PE-induced phosphorylation of 4E-BP1 (Fig. 5A). Activation of p70 S6 kinase by PE was also unaffected in cells after PKC down-regulation (35). Thus, phosphorylation of 4E-BP1 promoted by the ␣ 1A AR does not require TPA-sensitive PKCs.
We next used m 7 GTP Sepharose binding assays to test whether calmodulin might be required for the Ca 2ϩ -dependent phosphorylation of 4E-BP1. Pretreatment of cells with the calmodulin inhibitors ophiobolin A or calmidazolium prevented the PE-induced release of 4E-BP1 from eIF4E (Fig. 5B). This result suggests that stimulation of the ␣ 1A AR activates a Ca 2ϩ /calmodulin-dependent pathway that leads to phosphorylation of 4E-BP1 and dissociation of the 4E-BP1-eIF4E complex. DISCUSSION The results presented here demonstrate that stimulation of the ␣ 1A AR in Rat-1 cells promotes increased phosphorylation of the translation repressor 4E-BP1. Even though phosphorylation of 4E-BP1 as judged by gel mobility shift assays was modest in cells treated with PE as compared with PDGF ( Fig.  2), PE-induced phosphorylation caused almost all of the protein to be released from eIF4E (Fig. 3B). Furthermore, the results in this and our previous study (35) indicate that functional phosphorylation of 4E-BP1 in response to ␣ 1A AR stimulation is independent of PI 3-kinase/Akt signaling. We base this conclusion on the observations that PI 3-kinase activity in phosphotyrosine immunoprecipitates did not increase, the three known isoforms of Akt were not activated, and the levels of PI 3,4bisphosphate and PI 3,4,5-trisphosphate were not elevated in cells treated with PE (35). One possible explanation for why 4E-BP1 phosphorylation is blocked by LY294002 (Fig. 3) is that the compound inhibits a protein distinct from PI 3-kinase, such as mTOR (17,19,45), that is required for 4E-BP1 phosphorylation.
To our knowledge, this is the first report of growth factor receptor-mediated stimulation of 4E-BP1 phosphorylation in the absence of PI 3-kinase/Akt activation. The functional consequences of growth factor-induced phosphorylation of 4E-BP1 have been most intensively studied in cells treated with insulin, which acts through a tyrosine kinase receptor to activate the PI 3-kinase/Akt pathway. Although other G protein-coupled receptors, including the -opioid (46), gastrin/cholecystokinin type B (47), prostaglandin F 2␣ (48), and angiotensin II type 1 (49) receptors, can also signal to 4E-BP1, stimulation of these four receptors has been reported to activate PI 3-kinase and/or Akt in addition to inducing 4E-BP1 phosphorylation in the same cellular context (46 -51). Use of PI 3-kinase inhibitors and co-expression studies using highly active or dominantnegative mutants of PI 3-kinase and Akt have suggested that these signaling molecules act upstream of 4E-BP1 and p70 S6 kinase (23)(24)(25)(26)(27)(28)(29)(30)(31)(32)(33)(34). However, recent work by Dufner et al. (52) has shown that although expression of membrane-bound and cytosolic active mutants of Akt induces phosphorylation of 4E-BP1, only those Akt mutants that are constitutively targeted to the membrane can activate p70 S6 kinase. In addition, a membrane-targeted kinase-dead mutant of Akt blocked insulin-induced phosphorylation of 4E-BP1 but had no effect on insulininduced p70 S6 kinase activation. These workers concluded that Akt plays a dominant role in signaling to 4E-BP1 but is not necessary for p70 S6 kinase activation (52). Our data indicate that activation of Akt is not necessary for either the functional phosphorylation of 4E-BP1 (this study) or the activation of p70 S6 kinase (35).
Similar to the results shown here, amino acids have also been reported to elicit 4E-BP1 phosphorylation independently of PI 3-kinase/Akt signaling. Cells incubated in medium lacking amino acids exhibit reduced 4E-BP1 phosphorylation and p70 S6 kinase activity, and readdition of amino acids to these cells induces the functional phosphorylation of 4E-BP1 and activation of p70 S6 kinase (53)(54)(55). Even though these events are inhibited by wortmannin treatment in some cell types, amino acids do not promote an increase in PI 3-kinase or Akt activity. It would be of interest to determine whether amino acids and the ␣ 1A AR use a similar PI 3-kinase-independent signal transduction pathway to promote phosphorylation of 4E-BP1 and p70 S6 kinase.
It was shown earlier that incubation of a variety of cell types in medium containing EGTA to deplete intracellular Ca 2ϩ stores leads to a sharp and rapid decrease in the rate of protein synthesis (reviewed in Ref. 56). Polysomes were converted to monosomes in cells treated with EGTA, and analysis of ribosome transit times indicated that average elongation rates were relatively unaffected in Ca 2ϩ -depleted cells. These observations led to the conclusion that one or more steps in translation initiation require Ca 2ϩ . It has been proposed that Ca 2ϩ depletion might inhibit protein synthesis initiation by increasing the phosphorylation of eIF2␣, but not all data support this hypothesis (56). Our results suggest that 4E-BP1 and p70 S6 kinase are two proteins that confer Ca 2ϩ dependence on translation initiation. Treatment of Rat-1 fibroblasts with EGTA leads to the loss of PE-induced 4E-BP1 phosphorylation and p70 S6 kinase activation (Fig. 4). In addition, use of BAPTA-AM to chelate intracellular Ca 2ϩ reduces both the basal and hormone-activated levels of phosphorylation of the two proteins (Fig. 4). Thus, our expectation is that cap-dependent translation (regulated by 4E-BP1) and translation of 5Ј-TOP mRNAs (regulated by p70 S6 kinase) are inhibited in Ca 2ϩdepleted cells. This idea could be tested by analyzing translation of specific mRNAs on polysome gradients.
The Ca 2ϩ dependence of 4E-BP1 phosphorylation and p70 S6 kinase activation does not appear to be mediated by PKCs. Down-regulation of Ca 2ϩ -dependent PKCs by long term TPA treatment had little effect on the PE-induced phosphorylation of 4E-BP1 (Fig. 5A) or the activation of p70 S6 kinase (35).
Instead, it appears that these Ca 2ϩ -dependent events require calmodulin or a closely related protein. We show here that treatment of Rat-1 cells with two structurally unrelated calmodulin antagonists inhibits the functional phosphorylation of 4E-BP1 induced by the ␣ 1A AR (Fig. 5B). It was reported earlier that calmidazolium and other calmodulin antagonists inhibit protein synthesis when added to Ehrlich ascites tumor cells (57). Treatment of cells with calmidazolium induced the disaggregation of polysomes, indicating that a step in translation initiation was inhibited. Interestingly, the concentration of calmidazolium that gave half-maximal inhibition of protein synthesis in intact cells (10 M) was much lower than that required to inhibit cell-free translation (125 M; Ref. 57). One interpretation of this result is that the major target of calmidazolium is a calmodulin-dependent signaling pathway that mediates 4E-BP1 phosphorylation in intact cells.
Functional phosphorylation of 4E-BP1 and activation of p70 S6 kinase in response to stimulation of the ␣ 1A AR were sensitive to rapamycin (Fig. 3). This suggests that mTOR is a positive regulator for both of these events, as mTOR is the only known intracellular target of the rapamycin-FKBP12 complex (16). Thomas and co-workers (58) showed that overexpression of catalytically active or inactive versions of p70 S6 kinase blocked the insulin-induced phosphorylation of 4E-BP1. They suggested that a signaling pathway leading to phosphorylation of 4E-BP1 and p70 S6 kinase bifurcates immediately upstream of the two proteins and that overex- FIG. 5. Effect of TPA and calmodulin antagonists on PE-induced phosphorylation of 4E-BP1. Cells were treated as described below, and then binding of 4E-BP1 to m 7 GTP Sepharose was assayed in cell lysates (see under "Experimental Procedures"). A, cells were preincubated in serum-free medium without (control) or with 100 nM TPA for 24 h, followed by treatment with 100 nM TPA or 10 M PE for 20 min as indicated. B, serum-starved cells were pretreated for 30 min without (control) or with 10 M ophiobolin A or 10 M calmidazolium chloride, followed by stimulation with or without 10 M PE for 20 min. pressed p70 S6 kinase proteins inhibit 4E-BP1 phosphorylation by sequestering a common rapamycin-sensitive upstream activator that might be mTOR. Furthermore, other investigators have proposed that insulin positively regulates 4E-BP1 phosphorylation by increasing the kinase activity of mTOR through a PI 3-kinase/Akt-dependent pathway (59). 4E-BP1 and p70 S6 kinase responded similarly to all the cell treatments examined herein, supporting the idea that phosphorylation of the two proteins is controlled by overlapping signal transduction pathways. However, activation of mTOR by PI 3-kinase/Akt cannot account for the functional phosphorylation of 4E-BP1 induced by the ␣ 1A AR (35). We are currently testing whether mTOR activity might be controlled by a Ca 2ϩ /calmodulin-dependent pathway.
Up-regulation of protein synthesis plays a critical role in physiologic and pathologic cell growth and proliferation. Recent advances using in vitro models to delineate the signal transduction pathways that regulate translation have furthered our understanding of this important cellular process. Additional studies using in vivo models are needed to determine whether observations made in cell culture systems are relevant at the organism level.