Activation of the p70 S6 Kinase and Phosphorylation of the 4E-BP1 Repressor of mRNA Translation by Type I Interferons*

The Type I IFN receptor-generated signals required for initiation of mRNA translation and, ultimately, induction of protein products that mediate IFN responses, remain unknown. We have previously shown that IFN (cid:1) and IFN (cid:2) induce phosphorylation of insulin receptor substrate proteins and downstream engagement of the phosphatidylinositol (PI) 3 (cid:1) -kinase pathway. In the present study we provide evidence for the existence of a Type I IFN-dependent signaling cascade activated downstream of PI 3 (cid:1) -kinase, involving p70 S6 kinase. Our data demonstrate

The Type I IFN receptor-generated signals required for initiation of mRNA translation and, ultimately, induction of protein products that mediate IFN responses, remain unknown. We have previously shown that IFN␣ and IFN␤ induce phosphorylation of insulin receptor substrate proteins and downstream engagement of the phosphatidylinositol (PI) 3-kinase pathway. In the present study we provide evidence for the existence of a Type I IFN-dependent signaling cascade activated downstream of PI 3-kinase, involving p70 S6 kinase. Our data demonstrate that p70 S6K is rapidly phosphorylated on threonine 421 and serine 424 and is activated during treatment of cells with IFN␣ or IFN␤. Such activation of p70 S6K is blocked by pharmacological inhibitors of the PI 3-kinase or the FKBP 12-rapamycin-associated protein/mammalian target of rapamycin (FRAP/ mTOR). Consistent with this, the Type I IFN-dependent phosphorylation/activation of p70 S6K is defective in embryonic fibroblasts from mice with targeted disruption of the p85␣ and p85␤ subunits of the PI 3-kinase (p85␣؊/؊␤؊/؊). Treatment of sensitive cell lines with IFN␣ or IFN␤ also results in phosphorylation/inactivation of the 4E-BP-1 repressor of mRNA translation. Such 4E-BP1 phosphorylation is also PI3-kinase-dependent and rapamycin-sensitive, indicating that the Type I IFNinducible activation of PI3-kinase and FRAP/mTOR results in dissociation of 4E-BP1 from the eukaryotic initiation factor-4E (eIF4E) complex. Altogether, our data establish that the Type I IFN receptor-activated PI 3kinase pathway mediates activation of the p70 S6 kinase and inactivation of 4E-BP1, to regulate mRNA translation and induction of Type I IFN responses.
Type I interferons (IFNs) 1 are pleiotropic cytokines that exhibit multiple biological effects on cells and tissues, including inhibition of cell proliferation of normal and malignant cells, induction of antiviral responses, as well as immunomodulatory activities (1)(2)(3)(4)(5)(6)(7). Several signaling pathways are activated following binding of Type I IFNs to the multichain Type I interferon receptor complex, whose IFNaR1 and IFNaR2 subunits are constitutively associated with protein members of the Jak family of kinases (reviewed in Refs. [1][2][3][4][5][6][7]. A major Type I IFN-activated cellular pathway is the Jak-STAT signaling cascade (1)(2)(3)(4)(5)(6)(7). Engagement of the Type I IFN receptor results in activation of the Tyk-2 and Jak-1 kinases, which in turn regulate downstream phosphorylation/activation of the STAT1 and STAT2 transcriptional activators. The phosphorylated forms of STAT1 and STAT2 associate with IRF-9 (p48) to form the mature ISGF3 DNA-binding complex that translocates to the nucleus and regulates gene transcription via binding to ISRE elements in the promoters of IFN-stimulated genes (ISGs) (1)(2)(3)(4)(5). Several other STAT complexes are also induced during engagement of the Type I interferon receptor. STAT 1:1 homodimers, STAT 3:3 homodimers, STAT 1:3 heterodimers, STAT 5:5 homodimers, and CrkL:STAT5 heterodimers are formed in a Type I IFN-dependent manner and translocate to the nucleus where they bind to GAS regulatory elements in the promoters of IFN-activated genes to regulate gene transcription (2-5, 8, 9).
In addition to tyrosine phosphorylation of STAT proteins by interferon-activated Jak kinases, phosphorylation on serine residues is required for their full transcriptional activation (10 -14). It appears that, at least in the case of STAT1, phosphorylation on serine 727 is regulated by a member of the protein kinase C family of proteins, protein kinase C ␦ (15). There is also accumulating evidence that the p38 MAPK pathway is activated in a Type I IFN-dependent manner (16,17) and that its function is essential for gene transcription via ISRE (16,17) or GAS elements (18). Such regulatory effects of this pathway play critical roles in IFN signaling, because p38 activation is essential for generation of Type I IFN-dependent antiproliferative responses (19 -21).
The signals generated by the Type I interferon receptor to ultimately regulate mRNA translation are not known. We have previously demonstrated that Type I IFNs activate the insulin receptor substrate (IRS)-PI 3Ј-kinase pathway in human and mouse cells (42)(43)(44)(45) and that both the lipid (42) and serine (45) kinase activities of the p110 catalytic subunit of the PI 3Јkinase are activated during engagement of the Type I interferon receptor. In the present study we sought to determine whether the p70 S6 kinase is activated downstream of the PI 3Ј-kinase to mediate induction of Type I IFN responses. Our data demonstrate that the p70 S6 kinase is rapidly phosphorylated and activated during treatment of sensitive cell lines with IFN␣ or IFN␤. They also show that the IFN␣-dependent phosphorylation/activation of the p70 S6 kinase is defective in mouse embryonic fibroblasts (MEFs) from p85␣Ϫ/Ϫ p85␤Ϫ/Ϫ double knock-out mice. In other studies we establish that the translational mRNA repressor 4E-BP1 is phosphorylated in a Type I IFN-dependent manner, and such phosphorylation is also PI 3Ј-kinase-dependent, further demonstrating that activation of PI 3-kinase/mTOR by Type I IFNs ultimately induces signals important for mRNA translation.

MATERIALS AND METHODS
Cells Lines and Reagents-Human recombinant IFN␣2 was provided by Hoffmann-La Roche. Human recombinant consensus IFN␣ was provided by Amgen Inc. Human recombinant IFN␤ was provided by Biogen Inc. Antibodies against the phosphorylated forms of p70 S6 kinase, mTOR, and 4EBP-1 were obtained from Cell Signaling Technology Inc. An antibody against 4EB-P1 has been previously described (46). The FRAP/mTOR inhibitor, rapamycin, and the PI 3Ј-kinase inhibitors LY294002 and wortmannin were obtained from Calbiochem Inc. (La Jolla, CA). U266 cells were grown in RPMI 1640 supplemented with fetal bovine serum and antibiotics. U2OS and T98G cells were grown in McCoy's and Dulbecco's modified Eagle's media, respectively, supplemented with fetal bovine serum and antibiotics. The generation of p85␣Ϫ/Ϫ␤Ϫ/Ϫ mice, by crossing p85␤ϩ/Ϫ (48) mice with p85␣ϩ/Ϫ mice (47) will be described elsewhere. 2 The p85␣Ϫ/Ϫ␤Ϫ/Ϫ mouse embryonic fibroblasts were obtained from p85␣Ϫ/Ϫ␤Ϫ/Ϫ double knock-out mice. Briefly, mouse embryos were harvested at day 14; the limbs, head, and liver removed, and the resultant torso was finely minced. Following trypsinization, the single cell suspension was transferred onto gelatinized tissue culture dishes, and the mouse embryonic fibroblasts were immortalized after a few passages using SV40 large T antigen, expressed by a retroviral vector. The genotypes of the cells were determined by polymerase chain reaction. All transfections were performed using FuGENE 6, according to the manufacturer's instructions (Roche Applied Science).
Cell Lysis and Immunoblotting-Cells were stimulated with 1 ϫ 10 4 units/ml of the indicated IFNs for the indicated times, then lysed in phosphorylation lysis buffer as previously described (49). Immunoprecipitations and immunoblotting, using an enhanced chemiluminescence (ECL) method, were performed as previously described (49). In the experiments in which pharmacological inhibitors of FRAP/mTOR or the PI 3Ј-kinase were used, the cells were pretreated for 60 min with the indicated concentrations of the inhibitors and subsequently treated for the indicated times with IFNs, prior to lysis in phosphorylation lysis buffer. In some of the experiments to determine the phosphorylation of 4E-BP1, cell extracts were obtained by three freeze-thaw cycles, as previously described (46).
p70 S6 Kinase Assays-Assays to detect the Type I IFN-dependent activation of the p70 S6 kinase were performed as previously described (52). Briefly, cells were lysed in phosphorylation lysis buffer, and cell lysates were immunoprecipitated with an antibody against p70 S6 kinase or control non-immune rabbit immunoglobulin (RIgG). In vitro kinase assays were performed using a synthetic peptide substrate (AKRRRLSSLRA), and p70 S6 kinase activity was measured using an S6 kinase assay kit (Upstate Biotechnology Inc.) according to the manufacturer's instructions (52). Values were calculated by subtracting nonspecific activity, detected in RIgG immunoprecipitates, from kinase activity detected in anti-p70 S6K immunoprecipitates.
Isolation of Normal Peripheral Blood Granulocytes-Informed consent was obtained from healthy volunteers, according to the guidelines established by the Institutional Review Board of Northwestern University Medical School. Polymorphonuclear leukocytes were separated from peripheral venous blood using the Mono-Poly resolving medium (M-PRM, ICN Biomedicals, Aurora, OH), as previously described (20). Briefly, after centrifugation at 300 ϫ g for 30 min at room temperature, the plasma and the mononuclear leukocyte band were discarded, and the polymorphonuclear band was transferred into an individual tube. Cells were washed with culture medium and were subsequently resuspended in culture medium, prior to interferon treatment.
Mobility Shift Assays-Actively growing cells were treated with IFN␣ for the indicated times, in the presence or absence of rapamycin, as indicated. 10 g of nuclear extracts, from untreated or IFN␣-treated cells, was analyzed using electrophoretic mobility shift assays, as described previously (19,20). A double-stranded oligodeoxynucleotide (ATTTCCCGTAAATCCC), representing a sis-inducing element (SIE) of the c-fos promoter, was synthesized and used in the gel shift assays. A double-stranded oligodeoxynucleotide (CTGTTGGTTTCGTTTCCTCA-GA), representing an ISRE element from the ISG-15 gene, was also synthesized and used to detect ISGF3 complexes.
Luciferase Reporter Assays-Cells were transfected with a ␤-galactosidase expression vector and either an ISRE luciferase construct or a luciferase reporter gene containing eight GAS elements linked to a minimal prolactin promoter (8X-GAS), using the SuperFect transfection reagent as per the manufacturer's recommended procedure (Qiagen). The ISRE-luciferase construct (16) included an ISG15 ISRE (TCGGGAAAGGGAAACCGAAACTGAAGCC) cloned via cohesive ends into the BamHI site of the pZtkLuc vector and was provided by Dr. Richard Pine (Public Health Research Institute, New York, NY). The 8X-GAS construct (53) was kindly provided by Dr. Christofer Glass (University of California San Diego, San Diego, CA). Forty-eight hours after transfection, triplicate cultures were either left untreated or treated with 5 ϫ 10 3 units/ml IFN␣, and luciferase activity was subsequently measured using the manufacturer's protocol (Promega). The measured luciferase activities were normalized for ␤-galactosidase activity for each sample.

RESULTS
The p70 S6 Kinase Is Activated by Type I Interferons Downstream of the Phosphatidylinositol 3Ј-Kinase-We initially sought to determine whether the p70 S6 kinase is phosphorylated/activated during treatment of Type I IFN-sensitive cell lines with IFN␣ or IFN␤. Molt-4 cells were incubated for 30 min in the presence or absence of IFN␣. Cells were lysed in phosphorylation lysis buffer, and total cell lysates were analyzed by SDS-PAGE and immunoblotted with an antibody against the phosphorylated form of p70 S6 kinase on threonine 421 and serine 424. IFN␣ treatment resulted in strong phosphorylation of the p70 S6 kinase, whereas there was no change in the amounts of p70 S6 kinase protein detected after IFN␣ stimulation ( Fig. 1, A and B). Similarly, phosphorylation of p70 S6K was detectable when another Type I IFN, IFN␤, was used ( Fig. 1, C and D). In experiments, in which the phosphorylation of the p70 S6 kinase by Type I IFNs was determined, we found that both IFN␣ (Fig. 2, A and B), as well as IFN␤ (Fig. 2, C and D) are capable of inducing phosphorylation of the protein when used at doses as low as 100 IU/ml, further emphasizing the specificity of the process. Furthermore, the Type I interferondependent phosphorylation of the p70 S6 kinase was inducible in various cell types, including U2OS osteosarcoma cells ( Phosphorylation of the S6 kinase was rapid, occurring within 15 min of IFN treatment and was still detectable after 120 min of incubation with IFN (Fig. 3, A and B). As expected, phosphorylation of p70 S6K was also detectable in lysates from cells treated with insulin, used as positive controls for these assays (Fig. 3, A-D). To examine whether the phosphorylation of p70 S6K also occurs in primary human cells, we determined whether IFN␣ is capable of inducing phosphorylation of the protein in human granulocytes isolated from the peripheral blood of healthy donors. Consistent with the data observed in cell lines, IFN␣ treatment induced p70 S6K phosphorylation in normal granulocytes (Fig. 3, E and F), suggesting that this kinase is also activated in primary human cells, and may participate in the induction of IFN responses under physiological conditions.
As our data established that p38 is phosphorylated during engagement of the Type I IFN receptor, we sought to identify the upstream signaling events that ultimately lead to p70 S6K activation. In previous studies we have shown that the serine and lipid kinase activities of the phosphatidylinositol 3Ј-kinase are activated in a Type I IFN-dependent manner, during the interaction of the p85 regulatory subunit of this kinase with IRS proteins (IRS-1 and IRS-2) (42-45). Because the PI 3Ј-kinase pathway is a known regulator of activation of the p70 S6K, we examined whether its activation is required for engagement of the p70 S6 kinase in Type I interferon signaling. Cells were pretreated with the PI 3Ј-kinase inhibitors LY294002 or wortmannin and were subsequently incubated in the presence or absence of IFN␣. Treatment of cells with insulin was included as a positive control. As shown in Fig. 4, inhibition of PI 3Ј-kinase activity, by either LY294002 or wortmannin, abrogated the IFN␣-dependent phosphorylation of p70 S6K (Fig. 4, A and B). Similarly, and as expected, the PI 3Ј-kinase inhibitors also blocked the insulin-dependent phosphorylation of the protein (Fig. 4, A and B).
To definitively establish the requirement of PI3Ј-kinase in the Type I IFN-dependent activation of p70 S6K, we undertook studies using embryonic fibroblasts from mice with targeted disruption of the genes for both the ␣ and ␤ isoforms of the p85 regulatory subunit of the PI 3Ј-kinase. p85␣Ϫ/Ϫ p85␤Ϫ/Ϫ or control p85␣ϩ/ϩ p85␤ϩ/ϩ MEFs were treated with mouse IFN␣, and the phosphorylation of p70 S6K was examined by anti-phospho-p70 S6K immunoblotting. As shown in Fig. 5, IFN␣ treatment resulted in phosphorylation of p70 S6K on threonine 421 and serine 424 in p85␣ϩ/ϩ p85␤ϩ/ϩ cells but not in MEFs lacking expression of the p85 isoforms (p85␣Ϫ/Ϫ p85␤Ϫ/Ϫ) (Fig. 5, A and B). In a similar manner, when p85␣ ϩ/Ϫ p85␤Ϫ/Ϫ cells were compared with p85␣Ϫ/Ϫ p85␤Ϫ/Ϫ cells, phosphorylation of p70 S6K was detectable in the cells expressing p85␣ but not in the cells lacking both isoforms of the p85 regulatory subunit of the PI 3Ј-kinase (Fig.  5, C and D). Thus, activation of the PI 3Ј-kinase pathway is essential for downstream engagement of the p70 S6K by Type I interferons, and the presence of either the p85␣ or p85␤ isoform of the PI 3Ј-kinase may be sufficient to mediate such a response.
FRAP/mTOR Is Activated in a Type I IFN-dependent Manner to Regulate Downstream p70 S6 Kinase Activation-The Equal amounts of total cell lysates were analyzed by SDS-PAGE and immunoblotted with an antibody against the phosphorylated/activated form of the p70 S6-kinase on threonine 421 and serine 424. B, the blot shown in A was stripped and re-probed with an anti-p70 S6K antibody, to control for protein loading. C, Molt-4 cells were treated with IFN␤ for 30 min, as indicated. Equal amounts of total cell lysates were analyzed by SDS-PAGE and immunoblotted with an antibody against the phosphorylated/activated form of the p70 S6-kinase. D, the blot shown in C was stripped and re-probed with an anti-p70 S6K antibody, to control for protein loading.

FIG. 2.
Interferon-dependent phosphorylation of the p70 S6 kinase. A, U266 cells were treated with the indicated doses of IFN␣ for 30 min, as indicated. Equal amounts of total cell lysates were analyzed by SDS-PAGE and immunoblotted with an antibody against the phosphorylated/activated form of the p70 S6-kinase. B, the blot shown in A was stripped and re-probed with an anti-p70 S6K antibody, to control for protein loading. C, Molt-4 cells were treated with the indicated doses of IFN␤ for 60 min, as indicated. Equal amounts of total cell lysates were analyzed by SDS-PAGE and immunoblotted with an antibody against the phosphorylated/activated form of the p70 S6-kinase. D, the blot shown in C was stripped and re-probed with an anti-p70 S6K antibody, to control for protein loading.
FKB12-rapamycin associated protein (FRAP), also called mammalian target of rapamycin (mTOR), has previously been shown to regulate activation of the p70 S6 kinase, downstream of the PI 3Ј-kinase and the PDK-1 kinase (36 -41, 55). It is also well established that activation of mTOR requires its phosphorylation on serine 2448. We investigated whether mTOR is engaged in Type I IFN signaling, to regulate downstream activation of the p70 S6K. U-266 cells were treated with IFN␤, and, after cell lysis, total lysates were resolved by SDS-PAGE and immunoblotted with an anti-phospho mTOR antibody. Some baseline phosphorylation of mTOR was detectable prior to IFN treatment (Fig. 6A). However, type I IFN treatment of the cells resulted in phosphorylation/activation of mTOR, demonstrating that this protein is indeed engaged in IFN-signaling (Fig. 6A).
The demonstration that the p70 S6 kinase is phosphorylated by Type I IFNs strongly suggested that such phosphorylation activates the kinase domain of p70 S6K to regulate generation of IFN responses. We therefore sought to directly determine whether Type I IFN-dependent treatment of cells results in induction of p70 S6 kinase activity. U-266 cells were incubated in the presence or absence of IFN␣, and, after immunoprecipitation of cell lysates with an anti-p70 S6K antibody, immunoprecipitates were subjected to an in vitro kinase assay (52). As shown in Fig. 6B, Type I IFN treatment resulted in an activation of the catalytic domain of the p70 S6 kinase (Fig. 6B). Such activation was blocked by pretreatment of cells with the PI 3Ј-kinase inhibitor LY294002, as well as by the FRAP/mTOR inhibitor rapamycin, demonstrating that the IFN-dependent induction of p70 S6 kinase activity is FRAP/mTOR-dependent (Fig. 6B). Consistent with this, the IFN-dependent phosphorylation of p70 S6 kinase was also blocked when cells were pretreated with rapamycin prior to IFN␣ stimulation (Fig. 6, C and D), confirming that p70 S6K is a downstream effector for mTOR. The effects of rapamycin (Fig. 6C) on p70 S6K phosphorylation were apparently due to suppression of the IFN-dependent phosphorylation and not the minimal baseline phosphorylation of the protein, because they were accompanied by suppression of the IFN-dependent kinase activity of the protein (Fig. 6B).
Activation of the p70 S6 Kinase Downstream of mTOR Does Not Regulate Type I IFN-dependent Gene Transcription-Our data demonstrating activation of p70 S6K downstream of FRAP/mTOR, indicated that this pathway plays a role in mRNA translation, because it is well established that the kinase activity of p70 S6K regulates phosphorylation of the 40 S ribosomal S6 protein. We also examined whether activation of FRAP/mTOR downstream of the PI 3Ј-kinase plays any role in the Type I IFN-inducible activation of the STAT-pathway, which regulates transcriptional regulation of IFN-sensitive genes. We initially performed gel mobility shift assays, using either ISRE or a GAS (SIE) recognition element, to evaluate the effects of FRAP/mTOR inhibition on the formation of STATcontaining DNA-binding complexes. Cells were incubated with IFN␣ in the presence or absence of rapamycin, and the formation of the active ISGF3 or SIF complexes in response to IFN␣ was determined. As expected, treatment of various cell lines with IFN␣ induced formation of ISGF3 complexes (STAT2⅐STAT1⅐IRF-9) (Fig. 7A) or SIF complexes (STAT3:3, STAT1:3, STAT1:1) (Fig. 7, B-D). Rapamycin did not block the induction of ISGF3 or SIF complexes (Fig. 7, A-D), indicating that the activation of the p70 S6 kinase does not exhibit regulatory effects upon STAT activation. In a similar manner, treatment of cells with rapamycin had no effects on the phosphorylation of STAT1 on serine 727 (Fig. 7, E and F), which is required for full transcriptional activation of STAT1 (10 -15).
It is now established that, in addition to the STAT pathway, activation of the p38 MAPK pathway is essential for full transcriptional activation of interferon-stimulated genes (16,18). Furthermore, it has been demonstrated that the p38 MAPK pathway regulates Type I IFN-dependent gene transcription without modifying phosphorylation or activation of STAT proteins or their DNA binding activities in vitro (16,18). Our data demonstrated that the function of the p70 S6 kinase is not FIG. 3. Type I interferon-dependent phosphorylation of p70 S6 kinase. A, U2OS cells were serum-starved overnight and were subsequently incubated with either insulin or IFN␤, for the indicated times. Equal amounts of total cell lysates were analyzed by SDS-PAGE and immunoblotted with an antibody against the phosphorylated/activated form (Thr-421 and Ser-424) of the p70 S6-kinase. B, the blot shown in A was stripped and re-probed with an anti-p70 S6K antibody, to control for protein loading. C, serum-starved U266 cells were treated with either IFN␤ or insulin for 30 min, as indicated. Equal amounts of total cell lysates were analyzed by SDS-PAGE and immunoblotted with an antibody against the phosphorylated form (Thr-421 and Ser-424) of the p70 S6 kinase. D, the blot shown in C was stripped and re-probed with an anti-p70 S6K antibody to assure equal loading. E, normal peripheral blood granulocytes were incubated with IFN␣ for the indicated times. The cells were lysed and total cell lysates were analyzed by SDS-PAGE and immunoblotted with an antibody against the phosphorylated form (Thr-421 and Ser-424) of the p70 S6 kinase. F, the same blot shown in E was stripped and re-probed with an anti-p70 S6K antibody to control for protein loading.

FIG. 4. The Type I IFN-dependent activation of the p70 S6 kinase is PI 3-kinase-dependent.
A, U266 cells were serum-starved for 2 h and were subsequently preincubated for 30 min in the presence or absence of wortmannin (100 nM) or LY294002 (50 M), as indicated. The cells were subsequently treated with IFN␣ or insulin for 30 min, as indicated. Total cell lysates were analyzed by SDS-PAGE and immunoblotted with an antibody against the phosphorylated form of p70 S6 kinase. B, the blot shown in A was stripped and re-probed with an anti-p70 S6 kinase antibody, to control for protein loading. required for STAT activation and their DNA-binding activities but did not exclude the possibility that this pathway may still be facilitating transcriptional regulation of IFN-sensitive genes via effects on auxiliary pathways, including the p38 pathway. To address this issue, experiments were performed in which the effects of rapamycin on gene transcription via ISRE or GAS elements were examined. Cells were transfected with ISRE or 8X-GAS luciferase constructs and treated with IFN␣, in the presence or absence of the FRAP/mTOR inhibitor rapamycin. Luciferase activity was subsequently measured. As expected, IFN␣ induced strong luciferase activity via either ISRE (Fig.  8A) or GAS elements (Fig. 8B), but preincubation with rapamycin had no effect on the induction of such luciferase activities. Thus, based on these findings, it is unlikely that activation of p70 S6K downstream of PI 3Ј-K/FRAP/mTOR exerts any regulatory effects on IFN-dependent gene transcription, and its primary role in IFN signaling appears to be mediation of signals that regulate mRNA translation.

Type I IFN-dependent Phosphorylation of the 4E-BP1 Repressor of mRNA Translation and Its Dissociation from the Eukaryotic Initiation Factor-4E (eIF4E)
Complex-It has previously been shown that, in response to insulin, the 4E-BP1 repressor of mRNA translation is phosphorylated downstream of the PI 3Ј-kinase pathway and that this phosphorylation inhibits its interaction with the initiation factor eIF4E (46). We sought to determine whether IFN␣ treatment also results in phosphorylation of 4E-BP1. U-266 cells were treated with IFN␣, and after cell lysis, total lysates were analyzed by SDS-PAGE and immunoblotted with an anti-4E-BP1 antibody (46,54). IFN␣ treatment resulted in a mobility shift of the 4E-BP1 protein and detection of a slowly migrating form (Fig. 9A), which corresponds to the hyperphosphorylated form of the protein (46,54). To determine whether such phosphorylation of 4E-BP1 requires upstream activation of the PI 3Ј-kinase and FRAP/ mTOR activity, we examined the effects of LY294002 and rapamycin on such phosphorylation. Cells were preincubated with the different pharmacological inhibitors and subsequently treated with IFN␣. The phosphorylation of 4E-BP1 was subsequently analyzed by determination of the mobility shift of the protein. As shown in Fig. 9B, treatment with LY294002 or rapamycin abrogated the IFN␣-inducible hyperphosphorylated form of 4E-BP1 (Fig. 8B), indicating that this event occurs downstream of PI 3Ј-kinase/FRAP/mTOR activation. On the other hand, the SB203580 inhibitor, which selectively blocks activation of the p38 MAPK (16 -18), had no effects on 4E-BP1 phosphorylation (Fig. 9B). To further confirm the phosphorylation of 4E-BP1 in response to Type I IFNs, experiments were performed in which the phosphorylation of the 4E-BP1 protein was detected by using an anti-phospho 4E-BP1 antibody that recognizes the protein only when it is phosphorylated on threonine 70. Using this approach, strong IFN␣-dependent phosphorylation of 4E-BP1 was detectable in U-266 cells (Fig. 10, A and B) as well as COS cells (Fig. 10, C and D), whereas this phosphorylation was inhibited by pretreatment of cells with rapamycin or LY294002 (Fig. 10, A and B).
Recent studies have shown that, in addition to phosphorylation on Thr-70, phosphorylation of additional sites, most likely the priming sites Thr-37 and Thr-46, is essential for the release of 4E-BP1 from eIF4E (60). We examined whether Type I IFN treatment induces phosphorylation of these sites, whose function is critical for optimal phosphorylation of 4E-BP1. Cells were incubated in the presence or absence of IFN␤, and total cell lysates were analyzed by SDS-PAGE and immunoblotted with an anti-phospho 4E-BP-1 antibody, which recognizes the protein only when is phosphorylated on both Thr-37 and Thr-46. As shown in Fig. 11, treatment of cells with IFN␤ resulted in strong phosphorylation of 4E-BP1 on Thr-37 and Thr-46 (Fig. 11, A and B). To directly examine whether the IFN-dependent phosphorylation of 4E-BP1 regulates its activity, we examined the effects of Type I IFN treatment on the interaction of 4E-BP1 with eIF4E, using cap-affinity chromatography. Cells were incubated in the presence or absence of IFN␤, and cell extracts were subjected to cap-affinity isolation of eIF4E. After Type I IFN treatment, the binding of 4E-BP1 to eIF4E was dramatically decreased (Fig. 11D), whereas there was no change in the total amount of 4E-BP1 (Fig. 11C), indicating that 4E-BP1 dissociates from eIF4E, in an interferon-dependent manner. Taken altogether, these studies indicate that, in FIG. 5. Requirement of the p85-regulatory subunit of the PI 3-kinase for the Type I IFN-dependent activation of the p70 S6 kinase. A, p85␣Ϫ/Ϫ␤Ϫ/Ϫ and p85␣ϩ/ϩ␤ϩ/ϩ immortalized mouse embryonic fibroblasts (MEFs) were treated with mouse IFN␣ for the indicated times. Total cell lysates were analyzed by SDS-PAGE and immunoblotted with an antibody against the phosphorylated form of p70 S6 kinase. B, the blot shown in A was stripped and re-probed with an anti-p70 S6 kinase antibody, to control for protein loading. C, p85␣ϩ/Ϫ␤Ϫ/Ϫ and p85␣Ϫ/Ϫ␤Ϫ/Ϫ immortalized mouse fibroblasts (MEFs) were treated with mouse IFN␣ for the indicated times. Total cell lysates were analyzed by SDS-PAGE and immunoblotted with an antibody against the phosphorylated form of p70 S6 kinase. D, the blot shown in A was stripped and re-probed with an anti-p70 S6 kinase antibody, to control for protein loading.
addition to the activation of the p70 S6 kinase, Type I IFN-dependent activation of PI 3-kinase and FRAP/mTOR regulates phosphorylation of 4E-BP1 and its dissociation from eIF4E, providing an additional mechanism by which this pathway regulates mRNA translation in response to Type I IFNs.

DISCUSSION
Type I IFNs exhibit multiple biological activities, including antiproliferative, antiviral, and immunomodulatory effects in vitro and in vivo (1-7). The importance of the biological properties of Type I IFNs has led to extensive studies to understand the mechanisms by which these cytokines exert their biological effects on target cells. Over the last few years, dramatic advances have occurred on our understanding of how signals generated at the Type I IFN receptor level ultimately result in the induction of the effects of interferons. It is now well documented that the signals generated upon binding of IFN␣ or IFN␤ to the Type I IFN receptor result in the production of proteins that exhibit antiviral and/or antiproliferative properties (1-7). The transcription of genes encoding for such proteins is regulated by activation of Type I IFN-dependent Jak-STAT pathways (1)(2)(3)(4). Such transcription appears to be facilitated by other auxiliary, non-STAT pathways, such as the Type I IFNactivated p38 MAPK pathway (16 -21).
In addition to the Jak-STAT and p38 MAPK signaling cascades, Type I interferons engage insulin receptor substrate (IRS) proteins, to regulate activation of the PI 3Ј-kinase (42)(43)(44)(45)58). This pathway is activated by all Type I IFNs (42) downstream of Jak kinases (43,58), and there is evidence that it mediates classic IFN responses, including protection from FIG. 6. Activation of the kinase domain of the p70 S6 kinase by Type I IFNs is FRAP/mTOR-dependent. A, U266 cells were serumstarved overnight and were subsequently treated with IFN␤ for the indicated times. Total cell lysates were resolved by SDS-PAGE and immunoblotted with an antibody against the phosphorylated form of mTOR on serine 2448 (left panel). The same blot was subsequently stripped and re-probed with an anti-mTOR antibody to control for loading (right panel). B, U266 cells were preincubated for 30 min with either the PI 3Ј-kinase inhibitor, LY294002 (20 M), or the FRAP/mTOR inhibitor, rapamycin (20 ng/ml). The cells were subsequently treated for 30 min with IFN␣, in the continuous presence or absence of the inhibitors. The cells were subsequently lysed and immunoprecipitated with an anti-p70 S6 kinase antibody or non-immune rabbit immunoglobulin (RIgG). In vitro kinase assays to detect p70 S6K activity were subsequently carried out on the immunoprecipitates. The data are expressed as -fold increase over control and represent the means Ϯ S.E. of two independent experiments. C, Molt-4 cells were preincubated in the presence or absence of rapamycin (20 nM) for 60 min as indicated. The cells were subsequently treated with IFN␣, in the continuous presence or absence of rapamycin, as indicated. Total cell lysates were analyzed by SDS-PAGE and immunoblotted with an anti-phospho p70 S6 kinase antibody. D, the blot shown in C was stripped and re-probed with an anti-p70 S6 kinase to control for loading.

FIG. 7. The IFN␣-dependent formation of ISGF3 and SIF DNA binding complexes is FRAP/mTOR-independent.
A, actively growing T98G cells were preincubated for 30 min with rapamycin (20 ng/ml), as indicated. The cells were subsequently treated with IFN␣, in the continuous absence or presence of rapamycin as indicated. Nuclear extracts were reacted with 40,000 cpm of a 32 P-labeled ISRE, and complexes were resolved by native gel electrophoresis and visualized by autoradiography. B, T98G cells were preincubated for 30 min with rapamycin (20 ng/ml), as indicated. The cells were subsequently treated with IFN␣, in the continuous absence or presence of rapamycin, as indicated. Nuclear extracts were reacted with 40,000 cpm of a 32 Plabeled SIE, and complexes were resolved by native gel electrophoresis and visualized by autoradiography. C, U-266 cells were preincubated for 30 min with rapamycin, as indicated, and were subsequently treated with IFN␣, as indicated. Nuclear extracts were reacted with 40,000 cpm of a 32 P-labeled SIE, and complexes were resolved by native gel electrophoresis and visualized by autoradiography. D, Molt-4 cells were incubated for 30 min in the absence or presence of rapamycin, as indicated. The cells were subsequently treated with IFN␣, in the continuous absence or presence of rapamycin, as indicated. Nuclear extracts were reacted with 40,000 cpm of a 32 P-labeled SIE, and complexes were resolved by native gel electrophoresis and visualized by autoradiography. E, U20S cells were incubated for 60 min in the presence or absence of rapamycin, as indicated. The cells were subsequently treated with IFN␣ for the indicated times, in the continuous presence or absence of rapamycin. Total cell lysates were analyzed by SDS-PAGE and immunoblotted with an antibody against the phosphorylated form of STAT1 on serine 727. F, the blot shown in E was stripped and re-probed with an anti-STAT1 antibody to control for loading. viral infection (59). However, the precise mechanisms that ultimately lead to the generation of IFN biological effects through this signaling cascade are not known. It is well established that interferons regulate gene transcription and mRNA translation for target genes in a variety of malignant and non-malignant cells, resulting in the production of various protein products, among which proteins that suppress neoplastic cell proliferation, such as the pml gene product (60 -63). On the other hand, it is well established that one mechanism by which interferons mediate antiviral responses is suppression of viral replication for different viruses, via regulation of viral RNA translation (64 -69). In the present study we provide evidence that activation of the PI 3Ј-kinase by Type I IFNs ultimately leads to activation of the p70 S6 kinase, phosphorylation of the 4E-B1 repressor of mRNA translation, and its dissociation from the eIF4E. These data provide the first direct evidence linking an interferon-signaling pathway to events that positively regulate protein translation and address a long outstanding question in the interferon signaling field: the identification of Type I IFN signaling cascades ultimately responsible for generation of protein products that mediate the biological effects of Type I IFNs.
Several different cascades have been previously shown to be activated downstream of the phosphatidylinositol 3Ј-kinase in eukaryotic cells. It is now well established that signaling proteins with pleckstrin homology domains bind directly to phosphatidylinositol 3,4,5-triphosphate, which results from conversion of the plasma membrane phosphatidylinositol 4,5biphosphate by the activated PI 3Ј-kinase (reviewed in Ref. 69).
These include the kinase PDK1 and the Akt kinase (69), which activate diverse downstream signaling cascades. PDK1 phosphorylates and activates the Akt kinase (reviewed in Ref. 75), which in turn activates several downstream effectors, including the mammalian target of rapamycin FRAP/mTOR (70,76,77), which subsequently mediates phosphorylation of p70 S6K FIG. 8. Type I IFN-dependent gene transcription via ISRE or GAS elements is FRAP/mTOR-independent. U2OS cells were transfected with either an ISRE-luciferase construct (A) or an 8X-GAS luciferase construct (B), as indicated. Forty-eight hours after transfection the cells were pretreated for 60 min with the FRAP/mTOR inhibitor rapamycin and were subsequently incubated for 6 h in the presence or absence of IFN␣. The cells were then lysed, and luciferase activity was measured. Data are expressed as -fold increase in response to IFN␣ treatment over control untreated samples for each condition. Mean Ϯ S.E. values of three independent experiments for the ISRE panel, and two independent experiments for the GAS panel are shown. A, serum-starved U-266 cells were preincubated for 60 min with the indicated inhibitors and were subsequently treated with IFN␣ for the indicated times. Total cell lysates were analyzed by SDS-PAGE and immunoblotted with an antibody that recognizes the phosphorylated form of 4E-BP1 on threonine 70. B, the blot shown in A was stripped and re-probed with an anti-4E-BP1 antibody. C, COS cells were serum-starved overnight and were subsequently treated with IFN␣ for the indicated times. Total cell lysates were analyzed by SDS-PAGE and immunoblotted with an antiphospho-Thr-70 -4E-BP1 antibody. D, the blot shown in C was stripped and re-probed with an anti-4E-BP1 antibody, to control for loading. (36 -41) and 4E-BP1 (71)(72)(73)(74). Thus, FRAP/mTOR is required for the phosphorylation/activation of the p70 S6 kinase and phosphorylation/inactivation of 4E-BP1, events essential for the initiation of protein translation. Other pathways activated downstream of Akt include forkhead-related transcription factor 1, the inducer of apoptosis Bad, and the glycogen synthase kinase 3 (70,76,77).
The PI 3Ј-kinase-dependent pathways are traditionally believed to be pathways that mediate events essential for cell growth and anti-apoptotic effects. In fact, most studies have focused on the roles of these proteins in growth factor signaling or in the context of malignant transformation by oncogenes. Our finding, that Type I IFNs also activate the S6 kinase and phosphorylate 4E-BP1, in a FRAP/mTOR-dependent manner, implicates these proteins in the generation of antiproliferative and/or antiviral responses. Thus, it is likely that, as in the case of other pathways (Jak-STAT and MAPK cascades), the PI 3Ј-kinase/FRAP/mTOR pathway is capable of mediating either cell-proliferative or growth inhibitory responses, depending on the stimulus and the cellular context. This is not surprising, because it is well established that Type IFNs regulate transcription of genes, whose protein products suppress cell growth and mediate antiviral responses. Interestingly, a previous study (78) had demonstrated that PI 3Ј-kinase and mTOR mediate lipopolysaccharide-stimulated nitric oxide (NO) production in macrophages via secretion of IFN␤, which functions as an autocrine cofactor for NO production (78). It is of interest that, in that study, addition of exogenous of IFN␤, in the presence of PI 3-kinase and mTOR inhibitors, did not restore NO production (78). Taken together with our data, these findings suggest that this pathway is part of an autocrine loop that regulates production of Type I IFNs and, most importantly, mediates Type I IFN signals. Such downstream signals appear to be activation of the p70 S6K and de-activation of 4E-BP1, which in the context of interferon-induced transcription, facilitate protein translation of mRNA for IFN-sensitive genes and, therefore, induction of IFN-dependent biological effects.
It remains to be determined whether other signaling elements and pathways, beyond the IRS-PI 3Ј-kinase pathway, facilitate activation of p70 S6K and phosphorylation of 4E-BP1. It is well established that several sites of phosphorylation exist in 4E-BP1, and that the phosphorylation of the protein occurs via a two-step mechanism (79). Recently, it was shown that Erk kinases can phosphorylate 4E-BP1 in response to stress (80), whereas the downstream effector of the p38 MAPK, Msk1, phosphorylates 4E-BP1 in response to ultraviolet irradiation (81). Previous studies have also demonstrated that Type I IFNs activate Erk kinases (82) and that such activation is PI 3Јkinase-dependent (45). Similarly, the p38 MAPK pathway is activated by Type I IFNs (16 -18), and in recent studies we have observed that Msk1 is phosphorylated/activated in a Type I IFN-dependent manner. 3 Most importantly, activation of the p38 MAPK pathway is essential for the generation of Type I IFN-dependent antiproliferative and/or antiviral responses in a variety of different cell types (19,20). Thus, it is possible that these kinases are also required for phosphorylation/inactivation of 4E-BP1 and initiation of Type I IFN-dependent mRNA translation, but this remains to be examined in future studies.
Independently of the precise mechanisms involved, our data for the first time establish that the PI 3Ј-kinase pathway and cascades downstream of FRAP/mTOR participate in Type I IFN signaling, to generate IFN-dependent translational responses. These findings may have important clinical implications, because there are ongoing efforts toward the clinical development of rapamycin and the related analog CCI-779 for the treatment of certain malignancies (56,57). Such efforts are conceptually based on the documented ability of these compounds to inhibit growth factor-dependent malignant cell growth. IFN␣ is an antitumor agent widely used in clinical oncology and clinical virology. Therefore, our data indicate that caution should be taken in designing clinical trials combining these agents with IFN␣, because it is possible that they may ameliorate its antitumor properties in vivo.