Phosphoinositide 3-Kinase Regulation of T Cell Receptor-mediated Interleukin-2 Gene Expression in Normal T Cells*

Phosphoinositide (PI) 3-kinase has been implicated in T cell receptor (TCR) signaling, either as a positive or a negative regulatory molecule. Here, we show that for normal mouse lymph node T cells, PI 3-kinase activity is required for interleukin-2 (IL-2) production following TCR-mediated activation. Furthermore, in normal T cells, inhibition of PI 3-kinase prevented activation of enzymes in the extracellular signal-regulated protein kinase (ERK) signaling pathway (MEK-1 and ERK-2). Overexpression of a dominant-negative mutant of PI 3-kinase and pharmacological inhibitors of PI 3-kinase prevented transcriptional activation of AP-1 and NF-AT, transcription factors regulated by ERK-2 and pivotal for IL-2 gene expression. Although a constitutively active form of Akt kinase, a downstream mediator of PI 3-kinase function, enhanced TCR-induced IL-2 gene transcription, it could not bypass the requirement for PI 3-kinase activity. Therefore, PI 3-kinase is likely to be involved in signaling for IL-2 production in at least two steps in the TCR-initiated signaling pathway.

The T cell receptor for antigen (TCR) 1 is a multisubunit complex whose ligation initiates a series of signaling events that lead to gene activation, lymphokine production, and cell division. The earliest biochemical event following TCR activation is activation of intracellular protein-tyrosine kinases (1,2). The tyrosine phosphorylation events are regulated by two classes of protein-tyrosine kinases, the Src family (Lck and Fyn) and the Syk/ZAP-70 family. Lck and Fyn phosphorylate tyrosine residues present within a signaling motif in the cytoplasmic regions of the CD3 ␥, ␦, and ⑀ and TCR chains, called the immunoreceptor tyrosine-based activation motif. Phosphorylation of immunoreceptor tyrosine-based activation motifs is followed by the recruitment, tyrosine phosphorylation, and activation of ZAP-70 and Syk. This TCR-associated protein-tyrosine kinase activity is coupled to the phosphorylation and activation of downstream signaling molecules such as phospholipase C␥1. Hydrolysis of inositol phospholipids by phospholipase C␥1 results in the generation of inositol polyphosphates and diacylglycerols, second messengers that lead to activation of protein kinase C family members and to an increased level of intracellular Ca 2ϩ . Increased intracellular Ca 2ϩ in turn activates the phosphatase calcineurin, resulting in the nuclear translocation of the transcription factor NF-AT. Another signaling pathway emanating from the TCR involves the small GTP-binding protein Ras and a kinase cascade that it activates (the Ras/mitogen-activated protein kinase pathway). This pathway results in the activation of the transcription factors Elk1 and subsequently AP-1. Interleukin-2 (IL-2) gene expression, a common measure of productive T cell activation, cannot be induced by a single signaling pathway: it requires the coordinate action of these and other pathways that integrate at the level of multiple transcription factors, including NF-AT, AP-1, NF-B, and Oct-1 (3).
Another enzyme that has recently been implicated in TCR signaling is phosphoinositide (PI) 3-kinase (4,5). Class I A PI 3-kinase is a heterodimer composed of an 85-kDa regulatory subunit (p85) and a 110-kDa catalytic subunit (p110). The regulatory subunit p85 contains an N-terminal SH3 domain, two proline-rich sequences, and two SH2 domains. The SH2 domains bind to phosphorylated tyrosines on receptor and nonreceptor tyrosine kinases and kinase substrates and thus recruit the p85/p110 dimer to activated receptors. The region that separates the two SH2 domains (inter-SH2 domain) mediates the interaction of p85 with the catalytic subunit p110 and is required for PI 3-kinase enzymatic activity. Thus, in activated cells, p110 is targeted to the membrane where its substrates, phosphatidylinositols, reside. In vitro, class I A PI 3-kinase phosphorylates phosphatidylinositol (PtdIns), PtdIns(4)P, and PtdIns(4,5)P 2 at the D3 position of the inositol ring, yielding PtdIns(3)P, PtdIns(3,4)P 2 , and PtdIns(3,4,5)P 3 , respectively. In eukaryotic cells, PtdIns(3)P is constitutively present, and its levels are largely unaltered upon cellular stimulation. In contrast, PtdIns(3,4)P 2 and PtdIns(3,4,5)P 3 are nominally absent from resting cells. A wide variety of stimuli induce a rapid appearance of PtdIns(3,4,5)P 3 in intact cells, followed shortly thereafter by PtdIns(3,4)P 2 (6). Thus, the preferred substrate in vivo is likely to be PtdIns(4,5)P 2 . The rapid appearance of the PI 3-kinase lipid products and their resistance to cleavage by any known phospholipase suggest that they themselves may act as second messengers (7,8).
Several protein kinases have been placed downstream of PI 3-kinase in receptor-stimulated signaling. Protein kinases C, C⑀, C␦, and C are calcium-independent members of the protein kinase family and are activated in vitro by PtdIns(3,4)P 2 and PtdIns(3,4,5)P 3 (9,10). In intact cells, platelet-derived growth factor activates protein kinases C⑀ and C by a mechanism that requires PI 3-kinase (11,12). Activation of the serine/threonine kinase p70 S6K , which plays an important role in the progression of cells from G 1 to S phase of the cell cycle, also involves PI 3-kinase (13). At this time, the most convincing evidence for a downstream mediator of PI 3-kinase signaling is the serine/threonine protein kinase Akt (14). PtdIns(3,4)P 2 binds and activates Akt in vitro and in vivo. This requires that Akt have a functional pleckstrin homology domain, leading to the hypothesis that binding of the lipid to the pleckstrin homology domain targets Akt to the membrane where activating phosphorylation events, which are also PI 3-kinase-dependent, take place. Recently, phosphoinositide-dependent kinase-1, a protein kinase that phosphorylates Akt on one of the two sites required for Akt activity, has been purified and cloned (15)(16)(17). Phosphoinositide-dependent kinase-1 has a pleckstrin homology domain and is activated by PtdIns(3,4)P 2 and PtdIns(3,4,5)P 3 . One substrate of Akt in vivo is glycogen-synthase kinase-3 (18). Glycogen-synthase kinase-3 phosphorylates NF-AT, resulting in nuclear exit of NF-AT and thus opposing the effect of calcineurin (19). Phosphorylation of glycogen-synthase kinase-3 by Akt inactivates glycogen-synthase kinase-3, allowing nuclear localization of NF-AT and transcription of NF-AT-dependent genes such as IL-2. The Bcl-2 family member Bad is another in vivo substrate of Akt; phosphorylation of Bad blocks its death-promoting activity and results in cell survival (20,21).
Recent studies have implicated PI 3-kinase in regulating IL-2 production. In T cells, PI 3-kinase associates with the TCR upon activation due to either a direct interaction between p85 and the phosphorylated immunoreceptor tyrosine-based activation motif of the (22,23) or ⑀ (24) chain or an association between p85 and Src family protein-tyrosine kinases (25). Once bound, PI 3-kinase becomes activated, resulting in the accumulation of PI 3-kinase products (26). One study with the Jurkat T cell line has suggested that inhibition of PI 3-kinase actually induces IL-2 production in the absence of any other stimulus (27), whereas another study found that blocking PI 3-kinase inhibits IL-2 production (28). To address the physiologic role of PI 3-kinase in coupling TCR signaling to transcription of the IL-2 gene, we have used a combination of pharmacological and molecular approaches to characterize its role in primary lymph node T cells. We find that activation of PI 3-kinase is in fact essential for ERK activation and the initiation of IL-2 gene transcription and that this requirement cannot be bypassed by independent activation of Akt.

MATERIALS AND METHODS
Mice and Reagents-(C57BL/6 ϫ B10.D2)F 1 mice were intercrossed for 17-19 generations. Animals of either sex were killed at 6 -10 weeks of age, and single cell suspensions of lymph nodes were prepared. Cells were assayed in RPMI 1640 medium (Biofluids, Inc., Rockville, MD) supplemented with 10% heat-inactivated fetal calf serum, 4 mM glutamine, 100 units/ml penicillin, 150 g/ml gentamicin, and 5 ϫ 10 Ϫ5 M 2-mercaptoethanol. 145-2C11 (2C11) is a hamster monoclonal antibody against mouse CD3-⑀ and was purified from culture supernatant by affinity chromatography with protein A-Sepharose (Zymed Laboratories, Inc., San Francisco, CA). The anti-p85 antibody was purchased from Upstate Biotechnology, Inc. (Lake Placid, NY). The anti-ZAP-70 antibody 1213 and the ZAP-70 substrate cytoplasmic fragment of band 3 protein (cfb3) were a kind gift of Dr. Ron Wange (NIA, National Institutes of Health). Anti-ERK-2 and anti-MEK-1 antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Wortmannin and LY294002 were purchased from BIOMOL Research Laboratories Inc. (Plymouth Meeting, PA). Phorbol 12-myristate 13-acetate and ionomycin were purchased from Sigma. The cDNAs encoding murine p85␣ and p110 were a generous gift of Drs. Lewis Williams and Anke Klippel (Chiron Corp., Emeryville, CA). The 5Ј-IL-2-luciferase reporter plasmid containing the IL-2 minimal promoter and enhancer (nucleotides Ϫ353 to ϩ38 of the murine IL-2 gene) upstream of the luciferase gene was kindly provided by Dr. Jack Ragheb (NIAID, National Institutes of Health). The AP-1-luciferase construct contains six copies of the mouse proximal AP-1 site (at nucleotide Ϫ150), and the NF-AT-luciferase construct contains three copies of the distal NF-AT-binding site (at nucleotide Ϫ280) from the murine IL-2 gene (29). The plasmid encoding histidine-tagged kinase-inactive ERK-2 (K52R) was kindly provided by Dr. Melanie Cobb (University of Texas Southwestern Medical Center).
Construction of Plasmids-For transfection experiments, the coding sequences for p85␣ and p110 were subcloned into the expression vector pCI-neo (Promega, Madison, WI). To generate the deletion mutant ⌬p85, two polymerase chain reaction fragments were amplified from p85␣ cDNA: A1 (nucleotides 774 -1386; flanked by BsmI and BglII restriction sites) and A2 (nucleotides 1693-2232; flanked by BglII and BamHI restriction sites). These two polymerase chain reaction fragments, together with a fragment of p85␣ (from 9 nucleotides upstream of ATG to nucleotide 783; flanked by XbaI and BsmI restriction sites), were cloned into pBluescript SK (II) ϩ (Stratagene, La Jolla, CA) to generate ⌬p85. ⌬p85 was then subcloned into pCI-neo (pCI-⌬p85). Expression of this mutant results in the deletion of 102 amino acids from residues 466 to 567 of p85␣ and the insertion of 2 other amino acids (Arg-Ser) in this deleted position due to the cloning procedure. The correct sequence of ⌬p85 was confirmed by DNA sequence analysis. One additional amino acid (Ser) was present at the deletion site. The cDNA encoding v-Akt (30), the constitutively active viral homologue of c-Akt, was subcloned into pCI-neo (pCI-v-Akt).
IL-2 Assay-10 5 lymph node cells/well were plated in 96-well plates. The cells were stimulated as indicated (with or without pretreatment with inhibitor) and then cultured. Aliquots of the supernatant were removed in triplicate at the indicated time and assayed for IL-2 using a mouse IL-2 immunoassay (Quantikine M, R&D Systems, Minneapolis, MN).
In Vitro Transcription and Translation-In vitro translated proteins were prepared using the TNT coupled reticulocyte lysate system (Promega) using 30 Ci of [ 35 S]methionine (1000 Ci/mmol)/sample. Typically, 0.5 g of circular plasmid DNA was used. After 1.5 h at 30°C, the in vitro translated proteins were either immediately analyzed by SDS-PAGE and autoradiography or first immunoprecipitated and then analyzed by SDS-PAGE and autoradiography to detect protein-protein interactions.
Transient Transfections-Lymph node cells were cultured in phorbol 12-myristate 13-acetate (2 ng/ml) and ionomycin (1 g/ml). After 2 days, the cells were washed, resuspended in fresh medium containing IL-2 (10 units/ml), and cultured for 1 additional day. Prior to transfection, the T cell blasts were washed in medium and resuspended at 50 -75 ϫ 10 6 cells/ml in RPMI 1640 medium containing 20 mM HEPES and 4 mM glutamine. 0.2 ml of the cell suspension was mixed with the indicated amounts of plasmid DNA in electroporation cuvettes (0.4-cm electrode gap; Bio-Rad). Electroporation was performed at 960 microfarads and 250 V using a Bio-Rad Gene Pulser. The cells were then transferred into complete medium with 20 mM HEPES. The transfected cells were aliquoted into 24-well plates and cultured at 37°C for 3-6 h. Then, the cells were transferred to 24-well plates that had been coated with 2C11 (20 g/ml). After 12-14 h, the cells were harvested from the wells; transferred to microcentrifuge tubes; centrifuged; and lysed in 50 l of lysis buffer containing 25 mM Tris-HCl, pH 7.8, 2 mM dithiothreitol, 10% glycerol, 1 mM EDTA, and 1% Triton X-100. Samples (20 l) were then analyzed for luciferase activity.
ZAP-70 Kinase Assay-After treatment with soluble 2C11 (10 g/ml) for the indicated time, cells were lysed in buffer containing 150 mM NaCl, 25 mM Tris, pH 7.5, 1% Triton X-100, 5 mM EDTA, 1 mM Na 3 VO 4 , 10 g/ml leupeptin, 10 g/ml aprotinin, and 10 mM phenylmethylsulfonyl fluoride. Cell lysates were incubated on ice for 30 min and then cleared by centrifugation at 14,000 ϫ g for 10 min. Lysates were then incubated with protein A-Sepharose beads that had been preincubated with polyclonal anti-ZAP-70 antibody. After 2 h at 4°C, the beads were washed twice with lysis buffer and once with kinase buffer. The beads were then subjected to a ZAP-70 kinase assay with 10 Ci of [␥-32 P]ATP in 30 l of 20 mM Tris-HCl, pH 6.8, and 10 mM MnCl 2 (ZAP-70 kinase buffer) using 1 g of Cfb3 as a substrate. After incubating the beads at 37°C for 10 min, the reaction was stopped by the addition of SDS-PAGE sample buffer.
Mitogen-activated Protein Kinase Assay-After treatment with soluble 2C11 (10 g/ml) for the indicated time periods, cells were lysed in buffer containing 150 mM NaCl, 25 mM HEPES, pH 7.4, 1 mM Na 3 VO 4, 1% Triton X-100, 0.5 mM dithiothreitol, 10 g/ml leupeptin, 10 g/ml aprotinin, and 1 mM phenylmethylsulfonyl fluoride. Cell lysates were incubated on ice for 10 min and then cleared by centrifugation at 14,000 ϫ g for 10 min. Lysates were then incubated with protein A-Sepharose beads that had been preincubated with polyclonal anti-ERK-2 antibody. After 2 h at 4°C, the beads were washed twice with lysis buffer and once with kinase buffer. The kinase reaction was carried out in 30 l of 20 mM HEPES, pH 7.4, 10 mM MgCl 2 , 20 M ATP, and 2 mM EGTA (MAPK kinase buffer) containing 2 Ci of [␥-32 P]ATP and 1.5 mg/ml myelin basic protein at 37°C for 15 min. The reaction was stopped by the addition of SDS-PAGE sample buffer.
MEK Assay-After treatment with soluble 2C11 (10 g/ml) for the indicated time periods, cells were lysed in buffer containing 100 mM NaCl, 50 mM Tris-HCl, pH 7.5, 1% Triton X-100, 5 mM EDTA, 50 mM NaF, 40 mM ␤-glycerophosphate, 0.2 mM Na 3 VO 4 , 10 g/ml leupeptin, 10 g/ml aprotinin, 10 g/ml pepstatin A, and 1 mM phenylmethylsulfonyl fluoride. Cell lysates were incubated on ice for 10 min and then cleared by centrifugation at 14,000 ϫ g for 10 min. Lysates were then incubated with protein A-Sepharose beads that had been preincubated with polyclonal anti-MEK-1 antibody. After 2 h at 4°C, the beads were washed twice with lysis buffer and twice with 25 mM HEPES, pH 7.5, and 10 mM MgAc. The kinase reaction was carried out in 25 l of 25 mM HEPES, pH 7.5, 10 mM MgAc, 40 M ATP, and 2 mM dithiothreitol containing 2 Ci of [␥-32 P]ATP and 80 g/ml purified kinase-inactive ERK-2 (K52R) at 30°C for 20 min. The reaction was stopped by the addition of SDS-PAGE sample buffer.
Immunoblotting-Proteins were separated by SDS-PAGE and transferred to nitrocellulose filters (Schleicher & Schuell). Filters were blocked in phosphate-buffered saline containing 0.1% Tween 20 and 10% nonfat dried milk. The respective antibodies were added in phosphate-buffered saline at appropriate dilutions. Bound antibody was detected with anti-rabbit Ig conjugated to horseradish peroxidase (Amersham Pharmacia Biotech, Buckinghamshire, England) in phosphatebuffered saline containing 0.1% Tween 20, washed, and developed by enhanced chemiluminescence (Amersham Pharmacia Biotech).

PI 3-Kinase Antagonists Inhibit IL-2 Production
Induced by TCR Ligation-The effects of wortmannin, a fungal metabolite that irreversibly blocks PI 3-kinase catalytic activity (31,32), on activation-induced IL-2 production were determined. Freshly isolated murine lymph node cells were stimulated with anti-CD3 antibody in the presence of increasing concentrations of wortmannin, and IL-2 was measured (Fig. 1A). Wortmannin inhibited TCR-induced IL-2 production in a dose-dependent fashion, with half-maximal inhibition at a concentration of ϳ20 nM and complete inhibition at 100 nM, concentrations of wortmannin that inhibit PI 3-kinase activity (33,34). Since wortmannin can inhibit at least two other enzymes, PI 4-kinase (35,36) and phospholipase A 2 (37), at nanomolar concentrations, it is possible that the effect of this reagent on T cell activation was independent of its ability to inhibit PI 3-kinase. Therefore, similar experiments were performed with the structurally unrelated PI 3-kinase inhibitor LY294002 (38) (Fig. 1B). As with wortmannin, LY294002 inhibited IL-2 production at concentrations that completely and specifically abolish PI 3-kinase activity (38). Neither wortmannin nor LY294002 was toxic at the concentrations used, as determined by staining with trypan blue (data not shown). These results indicate that PI 3-kinase activity is necessary for IL-2 production in lymph node T cells following activation via the TCR.
A Deletion Mutant of p85 Does Not Associate with p110 and Blocks the Transcriptional Activity of the IL-2 Promoter-To further explore the relationship between PI 3-kinase and activation-induced IL-2 production, a dominant-negative form of p85 was used to inhibit the function of endogenous PI 3-kinase (39,40). We constructed a deletion mutant of p85 (⌬p85) lacking 102 amino acids of the inter-SH2 domain that confers binding to p110 (41,42). To verify the absence of binding of ⌬p85 to p110, ⌬p85 and p110 were transcribed and cotranslated in a reticulocyte translation system in the presence of [ 35 S]methionine. ⌬p85 was immunoprecipitated from the lysate, and the immunoprecipitate was analyzed for the presence of p110 ( Fig. 2A). When wild-type p85 and p110 were cotranslated, anti-p85 antibodies brought down both p85 and p110. In contrast, immunoprecipitation of ⌬p85 from lysates in which ⌬p85 and p110 were cotranslated resulted in the appearance of only ⌬p85. Therefore, the deletion of the inter-SH2 region of p85 did in fact disrupt the interaction between these two molecules.
To determine what effect repression of enzymatic function by dominant-negative ⌬p85 would have on activation-dependent signaling, T cell blasts were cotransfected with a reporter plasmid containing the minimal IL-2 promoter/enhancer upstream of the luciferase gene and either an empty vector or a vector encoding ⌬p85. The transfected cells were stimulated with anti-CD3 antibody, and induction of luciferase activity was assessed (Fig. 2B). As expected, the IL-2-luciferase reporter was induced by 2C11. When ⌬p85 was coexpressed, however, the anti-CD3 antibody-induced IL-2 activity was almost completely inhibited (87% inhibition). Importantly, when the ⌬p85transfected T cell blasts were stimulated with phorbol 12myristate 13-acetate and ionomycin, reagents that bypass the TCR to induce new gene transcription, IL-2-luciferase induction was not inhibited by ⌬p85. In addition, treatment of T cell blasts that had been transfected with the IL-2-luciferase reporter construct with wortmannin before stimulation with anti-CD3 antibody also inhibited IL-2 promoter activity (81% inhibition) (Fig. 2C). As a control, cells were cotransfected with a ␤-galactosidase expression plasmid, and ␤-galactosidase activity was measured in the same cell extracts used to analyze luciferase activity. In all samples analyzed, ␤-galactosidase activity never varied Ͼ2-fold. These results indicate that PI 3-kinase is a necessary mediator in TCR signaling for IL-2 gene transcription.
The IL-2 promoter contains binding sites for members of the NF-AT and AP-1 families of transcription factors (3). NF-AT proteins reside in the cytosol in resting T cells and translocate to the nucleus in TCR-activated T cells. Once in the nucleus, NF-AT and AP-1 form a functional transcription factor complex (43). Besides binding to the NF-AT/AP-1 site, AP-1 also binds to the Oct/AP-1 site and to its own site. It thus participates in IL-2 gene transcription by binding to multiple regulatory regions. To explore whether PI 3-kinase activity was required for transcriptional activation of NF-AT or AP-1, we analyzed the effect of ⌬p85 on the activation of NF-AT-and AP-1-luciferase reporter constructs in transient transfection assays. ⌬p85 inhibited the anti-CD3 antibody-induced NF-AT and AP-1 activities (98 and 89% inhibition, respectively) (Fig. 3, A and C). In addition, pretreatment with wortmannin inhibited the anti-CD3 antibody-induced NF-AT and AP-1 activation by 83 and 77%, respectively (Fig. 3, B and D).
Wortmannin Does Not Affect Proximal TCR Signaling-The earliest biochemical response to TCR ligation is the sequential activation of protein-tyrosine kinases: first Src family kinases and then the Syk/ZAP-70 family. Therefore, to determine if inhibition of PI 3-kinase interfered at a very proximal level of signaling, lymph node cells were treated with 100 nM wortmannin prior to stimulation with anti-CD3 antibody. After stimulation, ZAP-70 was immunoprecipitated, and its kinase activity was assessed by measuring the phosphorylation of a 43-kDa substrate (cfb3). Wortmannin, at a concentration that inhibited IL-2 production and transactivation of IL-2, NF-AT, and AP-1 reporters, had no effect on the anti-CD3 antibody-induced activity of ZAP-70 (Fig. 4).
Anti-CD3 Antibody-induced Activation of ERK-2 and MEK-1 Is Inhibited by Wortmannin and LY294002-A prominent component of the signaling cascade downstream of the activated TCR and leading to IL-2 production is the ERK pathway. ERKs are serine/threonine kinases involved in the activation of AP-1, a transcription factor essential for IL-2 gene transcription. FIG. 2. ⌬p85, a deletion mutant of p85, and wortmannin block the transcriptional activity of the IL-2 promoter. A, ⌬p85 does not associate with p110. p110 was cotranslated in vitro with p85 or ⌬p85 in the presence of [ 35 S]methionine. The reaction mixtures were subjected to immunoprecipitation with anti-p85 antibodies. Bound p110 was detected by SDS-PAGE and autoradiography. B, lymph node T cell blasts were cotransfected with 10 g of IL-2 promoter-luciferase reporter construct and 10 g of pCI-⌬p85 or empty pCI vector. The cells were cultured for 3 h before being transferred into wells coated with 2C11. After 14 h, the cells were harvested, lysed, and assayed for luciferase activity. The mean relative light units of the extracts obtained from cells cultured in medium were 661 (empty vector) and 784 (pCI-⌬p85). The data shown are the means Ϯ S.E. of three independent experiments. C, lymph node T cell blasts were transfected with 10 g of IL-2 promoter-luciferase reporter construct, treated with wortmannin (100 nM) for 30 min, and then cultured on wells coated with 2C11. After 15 h, cell extracts were prepared, and luciferase activity was quantitated. Therefore, the effects of wortmannin and LY294002 on ERK activation following anti-CD3 antibody activation were assessed. Freshly isolated lymph node cells were treated with wortmannin or LY294002 and then activated with anti-CD3 antibody. ERK-2 was immunoprecipitated, and its kinase activity was analyzed using myelin basic protein as a substrate. As expected, ERK-2 was activated by anti-CD3 antibody. In wortmannin-and LY294002-treated cells, however, the anti-CD3 antibody-induced ERK-2 activation was markedly inhibited (Fig. 5A). Treatment of the lymph node cells with wortmannin or LY294002 did not change the levels of ERK-2 protein, as visualized by immunoblotting of the reaction samples and probing with an anti-ERK-2 antibody (Fig. 5B). Furthermore, the addition of wortmannin or LY294002 directly to immunoprecipitated ERK-2 prior to and during the kinase assay had no effect on ERK-2 kinase activity (data not shown). Thus, wortmannin and LY294002 do not directly inhibit ERK-2, but rather inhibit a step necessary for ERK-2 activation.
MEK-1 is a dual-specificity kinase that is activated following TCR stimulation (44) and subsequently phosphorylates and thereby activates ERK-2. Freshly isolated lymph node cells were treated with wortmannin or LY294002 and then activated with anti-CD3 antibody. MEK-1 was immunoprecipitated, and its kinase activity was assessed using kinase-inactive ERK-2 as a substrate. MEK-1 was activated by anti-CD3 antibody, and this activation was markedly inhibited by the addition of wortmannin or LY294002 (Fig. 6A). This was a reflection of an inhibition of specific activity, as the levels of MEK-1 protein were not altered by treatment with wortmannin or LY294002 (Fig. 6B). The addition of wortmannin or LY294002 directly to immunoprecipitated MEK-1 prior to and during the kinase assay had no effect on MEK-1 kinase activity (data not shown). Taken together, these results indicate that PI 3-kinase activity is necessary for a step in the ERK signaling cascade that leads to MEK-1 and ERK-2 activation following TCR activation.
Constitutively Active Akt Enhances TCR-mediated Signaling for IL-2 Transcription, but Does Not Bypass the Requirement of PI 3-Kinase-PI 3-kinase may regulate MEK-1 activation directly or indirectly via one of its known targets. Since Akt has been described as a downstream target of PI 3-kinase (40,45), we asked whether a constitutively active Akt could bypass the need for PI 3-kinase in T cell signaling. Lymph node T cell blasts were transfected with expression plasmids encoding the IL-2-luciferase reporter construct and v-Akt (30). Transient expression of v-Akt in T cell blasts in the absence of an anti-TCR stimulus did not induce IL-2 promoter activity (Fig. 7, see  legend). Expression of v-Akt did, however, lead to augmented IL-2 promoter activation following TCR stimulation, suggesting that Akt participates in signaling for IL-2 gene transcription following TCR activation. Coexpression of v-Akt and ⌬p85 resulted in a marked decrease of IL-2 promoter activation when compared with expression of v-Akt alone. The level of TCRmediated IL-2 promoter activation after coexpression of v-Akt and ⌬p85 was similar to the level observed after transfection of only ⌬p85. Thus, although v-Akt augments signaling via the TCR, it cannot bypass the need for PI 3-kinase in TCR-mediated signaling resulting in IL-2 gene transcription. DISCUSSION The study of PI 3-kinase function has been facilitated by the discovery of two structurally unrelated inhibitors, wortmannin and LY294002. Wortmannin covalently binds Lys-802 of p110 (46) and irreversibly inhibits PI 3-kinase at nanomolar concentrations (33,34). LY294002, a competitive antagonist that prevents the binding of ATP to p110, inhibits PI 3-kinase with an IC 50 of 1.4 M (38). Mutants of p85 in which the inter-SH2 domain has been deleted have proved to act as dominantnegative inhibitors, providing a non-pharmacological method of analyzing PI 3-kinase function in vivo (5). A number of investigators have used these reagents and the Jurkat T cell line to investigate the role of PI 3-kinase in T cell activation and have reported conflicting findings that support either a negative or a neutral role for PI 3-kinase in IL-2 production. In one study, for example, it was concluded that PI 3-kinase has a negative role in TCR-mediated activation because a constitutively active PI Lymph node cells were pretreated with wortmannin (wort.; 100 nM) for 30 min, stimulated with soluble 2C11 (10 g/ml) for 2 min, and then lysed. Anti-ZAP-70 immunoprecipitates were assayed for ZAP-70 kinase activity using cfb3 as a substrate.

FIG. 5. PI 3-kinase inhibitors block ERK-2 activation induced by TCR ligation.
Lymph node cells were pretreated with wortmannin (W) or LY294002 (LY) at the indicated concentrations (nanomolar for wortmannin and micromolar for LY294002) for 30 min and stimulated with soluble 2C11 (10 g/ml) for 15 min. The cell lysates were subjected to immunoprecipitation with anti-ERK-2 antibody. ERK-2 activity was assessed using myelin basic protein as a substrate (A). Shown below the autoradiogram is a quantitation of the bands by PhosphorImager. After the kinase assay, immunoprecipitated ERK-2 was released from the beads and analyzed for protein level by SDS-PAGE and Western blotting (B).
3-kinase mutant ablated the TCR-mediated NF-AT response in transient transfection studies, and a dominant-negative PI 3-kinase mutant very similar to our ⌬p85 led to a basal induction of NF-AT activity and enhanced TCR-induced NF-AT responses in transient transfection experiments (47). It was also reported that wortmannin did not interfere with TCR-mediated NF-AT activation in Jurkat cells in transient transfection studies (48). In this study, the N-terminal SH2 domain of p85 had an inhibitory effect on the NF-AT response, whereas a construct containing both SH2 domains slightly enhanced the basal NF-AT activity without changing TCR-mediated NF-AT activation. This led to the conclusion that the enzymatic activity of PI 3-kinase does not play a role in the NF-AT response, but that p85 plays a complex role as an adaptor molecule in TCR-mediated NF-AT activation.
In contrast to these studies, the data in the present report demonstrate that PI 3-kinase activity is necessary for IL-2 production by normal T cells. The results show that PI 3-kinase regulates IL-2 production at the level of gene transcription since a dominant-negative PI-3 kinase and wortmannin inhibited TCR-mediated IL-2 promoter activation. The role of PI 3-kinase in IL-2 gene transcription was further defined with respect to the transcription factors that participate in IL-2 gene transcription. PI 3-kinase activity was found to be essential for the transcriptional activity of both NF-AT and AP-1. Furthermore, this effect appears to be exerted at the level of signaling molecules within the Ras/ERK pathway since PI 3-kinase activity was required for both MEK and ERK activation following TCR stimulation. Thus, in normal T cells, PI 3-kinase appears to be required for activation-induced up-regulation of MEK activity. We have used three different assays to determine if inhibition of PI 3-kinase alters activation of Ras. Unfortunately, although we can easily detect Ras activation in T cell lines, we have not been able to do so in primary T cells. Efforts are currently underway to develop a system in which the Ras signal can be enhanced to investigate what role, if any, PI 3-kinase plays in Ras activation in normal T cells.
The reason for the discrepancies between our results, indicating a positive role for PI 3-kinase in IL-2 production, and those obtained in the studies mentioned above is not clear, but is likely to reflect differences in TCR-mediated signaling and the regulation of IL-2 production in normal resting T cells compared with Jurkat cells. For example, phorbol ester is needed in addition to TCR ligation to stimulate Jurkat cells, but not normal T cells, to produce IL-2. Another difference is that treatment with wortmannin itself results in a low level of IL-2 production even without TCR activation in Jurkat cells (27), but does not induce IL-2 production or proliferation in normal T cells. In support of this distinction between normal T and Jurkat cells is that IL-2 production by antigen-stimulated TCR transgenic T cells has been shown to be inhibited by wortmannin (28). Surprisingly, inhibition was not observed when activation was achieved with immobilized anti-TCR antibodies. In contrast, we have found that anti-TCR antibodyinduced activation is readily inhibited by both chemical and molecular inhibitors of PI 3-kinase activity. This difference may be explained by different modes of activation. Whereas we used immobilized anti-TCR antibodies only, Shi and co-workers (28) used either syngeneic B7 ϩ B cell blasts that had been lipopolysaccharide-stimulated and mitomycin C-treated, or phorbol 12-myristate 13-acetate together with anti-TCR antibodies. Of interest is the observation that splenic T cells that had been activated by anti-CD3 antibody and B7 ϩ B cells become wortmannin-sensitive when they are propagated in vitro (49). We have identified both MEK-1 and ERK-2 as downstream effectors for PI 3-kinase in normal T cells upon TCR activation. This is in contrast to one study in which it was found that TCR-induced MEK-1 activation in Jurkat cells was not inhibited by wortmannin and ⌬p85, although ERK-2 activation was inhibited by both reagents (50). Again, this discrepancy might be explained by differences between Jurkat and normal T cells.
There are at least two nonexclusive possibilities by which PI 3-kinase might regulate IL-2 gene transcription: directly, by regulating a component of one of the signaling cascades leading to IL-2 gene expression, or indirectly, through one of its known targets. We have tested one of the latter possibilities by asking whether the PI-3 kinase-dependent step involved the PI 3-kinase target Akt. Since a constitutively active Akt kinase did not  7. v-Akt enhances TCR-mediated IL-2 promoter activity, but does not bypass the requirement for PI 3-kinase. Lymph node T cell blasts were cotransfected with 10 g of IL-2 promoter-luciferase reporter construct and pCI-⌬p85 (4 g), pCI-v-Akt (4 g), pCI-v-Akt (4 g) together with pCI-⌬p85 (4 g), or empty vector (8 g). The amount of DNA in each transfection was kept constant with empty vector DNA. After 3 h of culture, the cells were transferred into wells coated with 2C11 and cultured for an additional 14 h. The cells were then harvested, lysed, and assayed for luciferase activity. The mean relative light units of the extracts obtained from cells cultured in medium were 2097 (empty vector), 1970 (pCI-⌬p85), 2595 (pCI-v-Akt), and 1731 (pCI-v-Akt ϩ pCI-⌬p85). The results represent the means Ϯ S.E. of three independent experiments. overcome the block in IL-2 promoter activity mediated by ⌬85, we conclude that there is an additional target of PI 3-kinase that is necessary for IL-2 gene transcription. In addition, the data presented here indicate that the PI 3-kinase target Akt can participate in TCR signaling since constitutively active Akt synergizes with anti-TCR antibodies to induce IL-2 promoter activity in normal T cells. The involvement of Akt in TCRmediated IL-2 gene expression presumably reflects its positive regulation of NF-AT since Akt inactivates glycogen-synthase kinase-3, resulting in decreased phosphorylation of NF-AT and allowing NF-AT to stay in the nucleus. The results of this study suggest that PI 3-kinase participates in TCR-mediated IL-2 gene transcription by at least two distinct pathways, one involving ERK activation and the other Akt/glycogen-synthase kinase-3. The ERK pathway results in the induction of the c-fos gene, which participates in the formation of AP-1, whereas the Akt/glycogen-synthase kinase-3 pathway prevents the nuclear exit of NF-AT. Both AP-1 and NF-AT are necessary for IL-2 gene transcription and thus can account for the essential role that PI 3-kinase has in IL-2 gene expression.