IL-3 and IL-4 Activate Cyclic Nucleotide

IL-3 and IL-4 Activate Cyclic Nucleotide Phosphodiesterases 3 (PDE3) and 4 (PDE4) by Different Mechanisms in FDCP2 Myeloid Cells This information is current as of September 28, 2021. Faiyaz Ahmad, Guang Gao, Ling Mei Wang, Tova Rahn Landstrom, Eva Degerman, Jacalyn H. Pierce and Vincent C. Manganiello J Immunol 1999; 162:4864-4875; ; http://www.jimmunol.org/content/162/8/4864 Downloaded from

References This article cites 75 articles, 40 of which you can access for free at: http://www.jimmunol.org/content/162/8/4864.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on September 28, 2021

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 1999 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. IL-3 and IL-4 Activate Cyclic Nucleotide Phosphodiesterases 3 (PDE3) and 4 (PDE4) by Different Mechanisms in FDCP2 Myeloid Cells

Faiyaz Ahmad,* Guang Gao,* Ling Mei Wang,† Tova Rahn Landstrom,‡ Eva Degerman,‡ Jacalyn H. Pierce,† and Vincent C. Manganiello1*

In FDCP2 myeloid cells, IL-4 activated cyclic nucleotide phosphodiesterases PDE3 and PDE4, whereas IL-3, granulocyte-macrophage CSF (GM-CSF), and phorbol ester (PMA) selectively activated PDE4. IL-4 (not IL-3 or GM-CSF) induced tyrosine phosphorylation of insulin-receptor substrate-2 (IRS-2) and its association with phosphatidylinositol 3-kinase (PI3-K). TNF-␣, AG-490 (Janus kinase inhibitor), and wortmannin (PI3-K inhibitor) inhibited activation of PDE3 and PDE4 by IL-4. TNF-␣ also blocked IL-4-induced tyrosine phosphorylation of IRS-2, but not of STAT6. AG-490 and wortmannin, not TNF-␣, inhibited Downloaded from activation of PDE4 by IL-3. These results suggested that IL-4-induced activation of PDE3 and PDE4 was downstream of IRS-2/ PI3-K, not STAT6, and that inhibition of tyrosine phosphorylation of IRS molecules might be one mechnism whereby TNF-␣ could selectively regulate activities of cytokines that utilized IRS proteins as signal transducers. RO31-7549 (protein kinase C (PKC) inhibitor) inhibited activation of PDE4 by PMA. IL-4, IL-3, and GM-CSF activated mitogen-activated protein (MAP) kinase and protein kinase B via PI3-K signals; PMA activated only MAP kinase via PKC signals. The MAP kinase kinase (MEK-1) inhibitor

PD98059 inhibited IL-4-, IL-3-, and PMA-induced activation of MAP kinase and PDE4, but not IL-4-induced activation of PDE3. http://www.jimmunol.org/ In FDCP2 cells transfected with constitutively activated MEK, MAP kinase and PDE4, not PDE3, were activated. Thus, in FDCP2 cells, PDE4 can be activated by overlapping MAP kinase-dependent pathways involving PI3-K (IL-4, IL-3, GM-CSF) or PKC (PMA), but selective activation of PDE3 by IL-4 is MAP kinase independent (but perhaps IRS-2/PI3-K dependent). The Journal of Immunology, 1999, 162: 4864–4875.

yclic nucleotide phosphodiesterases (PDEs)2 are a heter- ity for cAMP (they hydrolyze cGMP poorly), sensitivity to speciﬁc

ogeneous group of structurally related enzymes that hy- inhibitors such as rolipram, and phosphorylation and activation by by guest on September 28, 2021 C drolyze cAMP and cGMP. Nine PDE gene families agents that increase cAMP, and are relatively abundant in inflam- (PDE1 through -9) differ in primary sequences, substrate specific- matory/immune cells (3). ity, responses to effectors and specific inhibitors, and mechanisms In general, agents that increase cAMP, including PDE inhibi- of regulation (1–6). Representatives of several families are usually tors, inhibit activation, cellular functions, and proliferation of im- found in the same cells, i.e., PDE3, -4, and -7 in human T lym- mune/inflammatory cells (3). cAMP inhibited p21ras and Raf ac- phocytes (4), but differ in amounts, proportions, and subcellular tivation and proliferation of NFS-60 myeloid cells (7), and arrested locations in different cells. PDE3 isoforms (PDE3A–B), encoded p210 BRC-ABL-transformed myeloid cells at the G1 phase of the by two genes, are characterized by high affinities for both cAMP cell cycle (8). In peripheral CD4ϩ and CD8ϩ T lymphocytes, ro- and cGMP, sensitivity to specific inhibitors such as cilostamide lipram, a specific PDE4 inhibitor, increased cAMP and inhibited and milrinone, and phosphorylation and activation in response to PHA- and anti-CD3-induced production of IL-2 and cell prolifer- agents that increase cAMP and to insulin (2). PDE4 isoforms, en- ation; specific PDE3 inhibitors enhanced the response to rolipram coded by four genes (PDE4A–D), are characterized by high affin- (9). Conversely, mitogenic stimulation of peripheral lymphocytes with PHA (10) or of T lymphocyte clones sensitized to myelin basic protein (MBP) with Ag (4) resulted in increased total PDE *Pulmonary/Critical Care Medicine Branch, National Heart, Lung, and Blood Insti- tute, and †Laboratory of Cellular and Molecular Biology, National Cancer Institute, activities as well as activities of PDE1 (10) and PDE3 and PDE4 National Institutes of Health, Bethesda, MD 20892; and ‡Section for Molecular Sig- isoforms (4). The cytokines IL-3 and GM-CSF support long-term naling, Department of Cell and Molecular Biology, Lund University, Lund, Sweden proliferation of FDCP2 cells; IL-4 cannot replace IL-3, but en- Received for publication September 15, 1998. Accepted for publication January hances the effects of IL-3 on proliferation of these cells (11–13). 19, 1999. Although IL-3R and IL-4R, which are heterodimeric type 1 cyto- The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance kine receptors (12, 13, 14), lack intrinsic protein tyrosine kinase with 18 U.S.C. Section 1734 solely to indicate this fact. activities, ligand binding induces tyrosine phosphorylation of dis- 1 Address correspondence and reprint requests to Dr. Vincent Manganiello, PCCMB, tinct intracellular substrates (mediated by Janus kinase (JAK) and National Heart, Lung, and Blood Institute, Building 10, Room 5N-307, 10 Center Src family protein tyrosine kinases), including IL-3R and IL-4R Drive, MSC 1434, Bethesda, MD 20892-1434. E-mail address: manganiv@gwgate. nhlbi.nih.gov themselves (11, 15) and transcription factors known as STATs (14, 2 Abbreviations used in this paper: PDE, cyclic nucleotide phosphodiesterase; GM- 16). In FDCP2 cells, IL-4, but not IL-3, induces tyrosine phos- CSF, granulocyte-macrophage CSF; IGF-1, insulin-like growth factor-1; IRS, insulin- phorylation of IRS-2 and association of PI3-K with IRS-2 and with receptor substrate; JAK, Janus kinase; MAP, mitogen-activated protein; MBP, myelin basic protein; MEK, MAP kinase kinase; PI3-K, phosphatidylinositol 3-kinase; PKB, the IL-4R itself (11). Similar to IRS-1 in nonhemopoietic cells, in protein kinase B; PKC, protein kinase C. FDCP2 cells IRS-2 links the IL-4R to PI3-K; activation of PI3-K

Copyright © 1999 by The American Association of Immunologists 0022-1767/99/$02.00 The Journal of Immunology 4865 is important in IL-4-induced mitogenic signals (11, 17–19). Some Cell culture and incubations IL-4R/IRS-2 signals are similar to those generated by insulin and The murine IL-3-dependent hemopoietic cell line, FDCP2, was propagated IGF-1 receptors (12, 17). IL-3 also induces tyrosine phosphoryla- and maintained in RPMI 1640 medium (Life Technologies, Grand Island, tion of cellular proteins, including PI3-K (20, 21) and Shc (22), NY) supplemented with 10% FBS and 5% WEHI-3B (American Type and activates Ras, Raf, MAP kinase, and PI3-K (23) and PKB (24) Culture Collection, Manassas, VA)-conditioned medium and 2 mM glu- in transducing mitogenic signals (25). In FDCP2 cells (11), IL-4 tamine (11). (Conﬂuent WEHI-3B ﬁbroblasts, which produce and secrete IL-3, were maintained in RPMI 1640 medium supplemented with 15% and IL-3 regulate different phosphorylation cascades, and thus, FBS and 2 mM glutamine for 4 days before conditioned medium was presumably, different biologic effects. Little is known of the effects collected and used to supplement growth medium for FDCP2 cells.) For of IL-4 and IL-3 or other cytokines on distinct and overlapping most experiments, exponentially growing FDCP2 cells (2–5 ϫ 105 cells/ signaling pathways that regulate different PDEs in the same cell. ml) were collected, centrifuged (5 min, 1200 ϫ g), washed twice, and cultured overnight in RPMI 1640/10% FBS without conditioned medium. In adipocytes, insulin-induced activation of PDE3B is important Immediately before experiments, cells were washed twice; suspended in in the antilipolytic action of insulin (2). Binding of insulin to its serum-free RPMI 1640 medium containing transferrin (5 ␮g/ml), selenium receptor leads to activation of the intrinsic tyrosine kinase activity (10 ␮M), and BSA (1 mg/ml); and incubated (3 ml cells/well, ϳ7 ϫ 105 of the receptor and tyrosine phosphorylation of IRS proteins (26, cells/ml) in six-well Costar plates (11). After 2–3 h at 37°C, cytokines, 27), resulting in activation of PI3-K (26, 27) and initiation of PMA, etc., were added as indicated. Finally, cells were harvested by centrifugation (1200 ϫ g, 5 min), suspended, homogenized (10–15 strokes in downstream signaling events, including activation of a PDE3B a Dounce homogenizer (Kontes Instruments, Vineland, NJ)) in lysis buffer kinase (2, 28). The PI3-K inhibitor wortmannin blocked insulin- (20 mM Tris, pH 7.5, 150 mM NaCl, 1 mM MgCl2, 1% Nonidet P-40 induced activation of PDE3B kinase and phosphorylation/activa- detergent, 1 mM EDTA, 10 mM NaF, 10 mM sodium pyrophosphate,

␮ Downloaded from tion of PDE3B as well as the antilipolytic action of insulin (28). 1mMNa3V04, 1 mM PMSF (Sigma, St. Louis, MO), and 20 g/ml each of aprotinin and leupeptin (Boehringer Mannheim, Indianapolis, IN)), and Insulin-induced activation of several serine/threonine protein ki- kept for 30 min at 4°C. Protein was measured using the Bradford assay nases, including MAP kinases (29, 30), p70S6 kinase (31, 32), and with BSA as standard. PKB (33–37), known as Rac or Akt, can also be mediated by PI3-K-generated signals. In intact rat adipocytes, insulin-induced cAMP PDE assay activation of PDE3B is independent of MAP kinase and p70S6 Portions of lysates (usually 100 ␮l) were assayed for 10 min at 30°C in a kinase, but may involve PKB (37). total volume of 0.3 ml containing 50 mM HEPES, pH 7.5, 8.3 mM MgCl2, http://www.jimmunol.org/ ␮ 3 In this study, we demonstrate, with the use of TNF-␣ and var- 0.1 mM EDTA, and 0.1 M[H]cAMP (25–35,000 cpm) as substrate (4). After dephosphorylation of 5-AMP to adenosine with Crotalus atrox ious kinase inhibitors, that in FDCP2 cells both PDE3 and PDE4 venom (Sigma), the product was separated from substrate using ion ex- are activated by IL-4 via IRS-2 and PI3-K-dependent signals. change chromatography (QAE-Sephadex; Pharmacia, Piscataway, NJ) and Downstream of PI3-K, however, the signals diverge, resulting in quantified by scintillation counting. Lysates were diluted so that hydrolysis MAP kinase-dependent and -independent pathways for activation of substrate was usually less than 20%. Total PDE, PDE3, and PDE4 activities were measured. PDE3 activity is that activity inhibited by 0.3 ␮M of PDE4 and PDE3, respectively. Furthermore, PDE4 is selec- cilostamide, a specific PDE3 inhibitor; PDE4 activity, that inhibited by tively activated by IL-3 and GM-CSF via PI3-K (not IRS-2) and 0.5 ␮M rolipram, a specific PDE4 inhibitor (38). Inhibitor vehicle MAP kinase signals, and by PMA via PKC and MAP kinase sig- (DMSO), added in equal quantities to samples without inhibitor, did not nals. MAP kinase and PDE4, not PDE3, are activated in FDCP2 influence PDE activities. by guest on September 28, 2021 cells transfected with constitutively active MEK or in FDCP2 cells Immunoprecipitation and immunoblotting transfected with wild-type MEK and treated with IL-3. The MEK inhibitor PD98059 blocks activation of MAP kinase and PDE4 in For most experiments, portions of FDCP2 lysates (2–3 mg protein) were precleared by incubation with 1 ␮g normal or preimmune mouse, rat, or both. Thus, PDE4 can be regulated by MAP kinase-dependent sig- rabbit IgG, as appropriate, for1hatroom temperature before addition of nals involving PI3-K (IL-4, IL-3, GM-CSF) or PKC (PMA). PKB 40 ␮l of trisacryl protein A (Pierce, Rockford, IL), protein G-Sepharose is activated by IL-4 as well as by IL-3 and GM-CSF, not PMA. (Pharmacia Biotech, Piscataway, NJ), or mouse IgG-agarose (Sigma) for TNF-␣, which inhibits tyrosine phosphorylation of IRS-2 and its 30 min before centrifugation (2800 ϫ g, 4°C, 5 min). Precleared cell lysates were incubated with Abs for2hatroom temperature, followed by association with PI3-K, blocks effects of IL-4 on PDE3 and PDE4, incubation with fresh trisacryl protein A, protein G-Sepharose, or mouse but not effects of IL-3 or PMA on PDE4. In FDCP2 cells, IL-4- IgG-agarose for 30 min before centrifugation (2800 ϫ g, 4°C, 5 min). induced activation of IRS-2 and PI3-K may initiate a specific sub- Immunoprecipitates were washed three times with lysis buffer containing set of signaling events that target PDE3 in a manner similar to 0.1% Nonidet P-40, boiled in Laemmli buffer, and subjected to SDS-PAGE activation of PDE3B by insulin in adipocytes. (Novex, San Diego, CA) (39). Proteins were transferred to nitrocellulose membranes in Tris-glycine buffer (25 mM Tris-base and 192 mM glycine at pH 8.3), containing 20% methanol. Membranes were incubated in blot-

ting buffer (150 mM NaCl, 0.05% (v/v) Nonidet P-40, 0.01% NaN3, and 10 Materials and Methods mM Tris, pH 7.4), containing BSA (50 mg/ml) and OVA (10 mg/ml), for Recombinant murine IL-4, IL-3, and GM-CSF were purchased from Pep- 1 h at room temperature with rocking and then for an additional 2 h with the appropriate Abs (40). Membranes were washed three times (10 min rotech (Rocky Hill, NJ); transferrin and selenium from Collaborative Re- 125 3 each) in blotting buffer without Abs, followed by incubation with I- search (Boston, MA); [ H]cAMP from New England Nuclear (Boston, ␮ MA); erk1 mAb from Tranduction Laboratories (Lexington, KY); poly- labeled protein A (2 Ci) (Amersham Life Sciences, Arlington Heights, clonal anti-rat PI3-K, anti-mouse IRS2, polyclonal anti-erk1-CT, and anti- IL)for1hatroom temperature, followed by three additional washes (10 phosphotyrosine Abs from UBI (Lake Placid, NY); goat polyclonal anti- min each) with blotting buffer. Immunoreactive proteins were visualized by PhosphorImager analysis (Molecular Dynamics, Sunnyvale, CA). PKB-CT, goat polyclonal anti-STAT6, anti-JAK 1,2,3, and anti-MEK Abs 125 from Santa Cruz Biotechnology (Santa Cruz, CA); PMA, JAK inhibitor For some immunoblots, instead of I-labeled protein A, HRP-labeled AG-490, and the PI3-K inhibitor wortmannin from Biomol Research Lab- second Ab (mouse, rat, rabbit, or goat) and enhanced chemoluminescence oratories (Plymouth Meeting, PA); MEK-1 inhibitor PD98059 from New (ECL) reagents 1 and 2 (Amersham) were used; immunoreactive proteins England Biolabs (Beverly, MA); the PI3-K inhibitor LY294002 from were visualized by exposing nitrocellulose ﬁlters to Bio-max Kodak ﬁlm Alexis (San Diego, CA); the PKC inhibitor RO31-7549 kindly given by Dr. and developed in a Standard Medical Imaging (Columbia, MD) developer. 32 Michael Beaven (NHLBI, National Institute of Health); and [␥- P]ATP MAP kinase activation and activity assays (3000 Ci/mmol) from ICN Radiochemicals (Costa Mesa, CA). Peptides (KKRNRTLTK) (K9) and Crosstide (GRPRTSSFAEG) were synthesized Activation of MAP kinase was assessed by measuring 1) the change in at the Biomolecular Unit, Lund University (Lund, Sweden). Prestained electrophoretic mobility of MAP kinase on SDS-PAGE/Western immuno- molecular size markers were from Bio-Rad (Richmond, CA). Other mate- blots using anti-erk1 Ab, 2) tyrosine phosphorylation of MAP kinase de- rials were obtained as indicated and were of highest grade available. tected with anti-phosphotyrosine Ab, or 3) phosphorylation of MBP or 4866 ACTIVATION OF PDE3 AND PDE4 BY CYTOKINES IN FDCP2 CELLS

MBP peptide by MAP kinase immunoprecipitates. 1) Portions of cell ly- Table I. PDE activities in FDCP2 cells transfected with MEK sates (ϳ20 ␮g) were boiled in Laemmli buffer (Bio-Rad), subjected to constructsa SDS-PAGE (10% gel, 120 V for 90 min), and transferred to nitrocellulose (NC) filters; MAP kinase was detected with anti-erk1 mAbs using ECL Activity (pmol cAMP hydrolyzed/min/mg) reagents. 2) Samples of precleared FDCP2 cell lysates (ϳ2 mg protein) ␮ ␮ were incubated with anti-erk1-CT Ab (21 l/10 g IgG; UBI) for2hat Transfection Total PDE PDE3 PDE4 room temperature and precipitated with 40 ␮l protein G-Sepharose; proteins were subjected to SDS-PAGE, transferred to nitrocellulose filters, F/V (n ϭ 9) 12.9 Ϯ 1.5 5.9 Ϯ 0.9 6.5 Ϯ 0.8 immunoblotted with polyclonal anti-phosphotyrosine Ab (UBI), followed F/V ϩ IL-3 (n ϭ 8) 21.6 Ϯ 2.8† 6.4 Ϯ 0.9 14.5 Ϯ 2.1† by 125I-labeled protein A, and quantified by PhosphorImager analysis. 3) F/MϪ(n ϭ 3) 14.1 Ϯ 2.3ϩ 6.2 Ϯ 1.7 7.2 Ϯ 1.6ϩ Portions of immunoprecipitates, prepared as in 2, were incubated at 30°C F/MϪ ϩ IL-3 (n ϭ 3) 20.6 Ϯ 2.9† 6.3 Ϯ 1.5 14.7 Ϯ 1.4† ϩ ϩ for 15 min in 50 mM Tris (pH 7.5), 0.4 mM EGTA, 0.4 mM Na3VO4,25 F/M (n ϭ 8) 14.3 Ϯ 1.9 5.9 Ϯ 0.9 7.7 Ϯ 2.1 ␮ ␮ ␥ 32 ␮ † † mM MgCl2, 150 M ATP, 10 Ci [ - P]ATP (10 Ci/mmol), 5 M cAMP F/M ϩ IL-3 (n ϭ 8) 24.8 Ϯ 3.7 6.3 Ϯ 1.6 17.5 Ϯ 2.9 protein kinase inhibitor (Calbiochem, San Diego, CA), and substrate, either F/M* (n ϭ 13) 22.2 Ϯ 2.9† 5.2 Ϯ 1.8 15.5 Ϯ 2.3† 5 ␮M MBP (18 kDa) (UBI) or 5 ␮M MBP peptide (APRTPGGRR (UBI)), F/M* ϩ IL-3 (n ϭ 8) 25.0 Ϯ 3.5† 5.5 Ϯ 1.0 17.6 Ϯ 2.8† in final volume of 50 ␮l. With MBP as substrate, reactions were terminated a FDCP2 cells were permanently transfected with control vector (F/V) or with by addition of 30 ␮lof2ϫ SDS sample buffer. Proteins were separated by Ϫ SDS-PAGE (0.75 mm, 12% gel), according to the method of Laemmli constructs expressing wild-type (F/M), kinase inactive (F/M ), or constitutively active (F/M*) MEK. Cells were incubated without or with IL-3 for 10 min before total (39). Gels were then treated as described by Kameshita and Fujisawa (41) PDE, PDE3, and PDE4 activities in cell lysates were assayed. PDE3 activity repre- with modifications (42). After incubation at room temperature for 1 h, the sents cAMP hydrolysis inhibited by cilostamide (a selective PDE3 inhibitor); PDE4, gels were washed five times with 5% (w/v) TCA and 1% (w/v) sodium that activity inhibited by rolipram (a selective PDE4 inhibitor). PDE activities are pyrophosphate. 32P incorporated into MBP was visualized and quantified reported as pmol cAMP hydroyzed/min/mg (mean Ϯ SEM) from the indicated num- by PhosphorImager analysis of dried gels. ber (n) of experiments (†, p Ͻ 0.001; ϩ, p ϭ NS vs control). Downloaded from When MBP peptide was substrate, reactions were terminated by addition of 10 ␮l 1% BSA, 1 mM ATP, pH 3, and 5 ␮l 30% TCA. For assay reaction blanks, the stopping solution was added to immunoprecipitated samples before assay buffer. Samples were centrifuged, and 25 ␮lofsu- pernatant was applied to phosphocellulose (P81, Whatmann) paper (2 ϫ 2-cm squares), which were washed three times (5 min each) with phos- were isolated. FDCP2 cells expressing similar amounts of wild-type (F/M),

phoric acid (7.5 mg/ml), and once with acetone (5 min) before radioassay Ϫ http://www.jimmunol.org/ of 32P incorporated into substrate. constitutively active (F/M*), and kinase-inactive (F/M ) MEK were selected by analyzing lysates (30–40 ␮g protein) from equivalent numbers of PKB assays with histone 2B, Crosstide, and K9 peptide cells by SDS-PAGE (Novex), immunoblotting with anti-MEK Abs, and substrates assaying MAP kinase activity. Samples of precleared FDCP2 cell lysates (ϳ2 mg protein) were incubated Results ␮ ␮ with anti-PKB-CT Ab (50 l/10 g IgG) for2hatroom temperature and Activation of PDE3 and PDE4 by IL-4, IL-3, GM-CSF, and precipitated with 40 ␮l of protein G-Sepharose beads. Immunoprecipitates were washed three times with lysis buffer (containing 0.1% Nonidet P-40), PMA in FDCP2 cells suspended in 20–30 ␮l of reaction buffer (20 mM HEPES, pH 7.4, 1 mM ␮ ␮ ␥ 32 Both PDE3 and PDE4 enzymes exhibit high affinity for cAMP DTT, 10 mM MnCl2, 10 mM MgCl2,5 M ATP, 10 Ci [ - P]ATP, by guest on September 28, 2021 5 ␮M cAMP protein kinase inhibitor) containing either 13 ␮g of K9 pep- (1–3). PDE3 activity represents cAMP-hydrolyzing activity inhib- tide (KKRNRTLTK), 1 ␮g of Crosstide (GRPRTSSFAEG), or 2.5 ␮gof ited by cilostamide, a specific PDE3 inhibitor; PDE4 activity, ac- histone-2B (H2B) (Boehringer Mannheim) as substrate (35–37), and incu- tivity inhibited by rolipram, a specific PDE4 inhibitor (38). As bated for 15 min at 30°C. For assays with K9 or Crosstide peptides, reac- shown in Table I, PDE3 and PDE4 activities account for most of tions were terminated and phosphocellulose squares were washed and an- the cAMP hydrolytic activity (with 0.1 ␮M[3H]cAMP as sub- alyzed as described for MAP kinase assays. PKB activity was proportional to incubation time and amount of pro- strate) in FDCP2 cell lysates. As seen in Fig. 1A, incubation of tein. K9 peptides, which have also been used to detect p70S6 kinase ac- FDCP2 cells with IL-4 increased both PDE3 and 4 activities (by tivity (43), and the Crosstide peptide, which closely resembles the se- ϳtwofold), whereas IL-3, GM-CSF (GM-CSF and IL-3 receptors quence containing the site phosphorylated by PKB in glycogen synthase share a common ␤ subunit and activate similar signaling pathways kinase-3 (44), have been used to determine PKB activity in both immunoprecipitates and partially purified preparations of PKB from rat adipo- (13)), and PMA selectively increased PDE4 activity. cytes (35–37). In FDCP2 cell lysates, use of either Crosstide or K9 peptide To define pathways utilized by cytokines and PMA to activate gives very similar results. Treatment of FDCP2 cells with 20 nM rapamy- PDE3 and PDE4, we first evaluated their effects on different can- cin (p70S6 kinase inhibitor) had no significant effect on IL-4-induced ac- didate signaling components, especially those apparently involved tivation of the kinase activity detected with K9 as substrate, suggesting that in insulin-induced activation of PDE3B in adipocytes (2), which is p70S6 kinase does not contribute to the IL-4-induced K9-kinase activity in PKB immunoprecipitates. dependent on PI3-K and, perhaps, PKB signaling, but independent With histone 2B as substrate for PKB, reactions were terminated by of MAP kinase (28, 37). To further identify these signaling path- addition of Laemmli buffer and boiling. Proteins were separated by SDS- ways, we used the proinflammatory cytokine TNF-␣ and a panel of 32 PAGE, the gel was dried, and P-labeled histone 2B was detected and protein kinase inhibitors, including AG-490 (which inhibits JAKs, quantified by PhosphorImager (Molecular Dynamics). but not other tyrosine kinases (48)), PD98059 (a selective MEK-1 Expression of MEK (MAP kinase kinase) in FDCP2 cells or MAP kinase kinase-1 inhibitor (49)), wortmannin and Several plasmid vector constructs (provided by J. S. Gutkind, NIDR, Na- LY294002 (PI3-K inhibitors (50, 51)), rapamycin (p70S6 kinase tional Institute of Health) were transformed into Escherichia coli, ampli- inhibitor (52)), and RO31-7549 (PKC inhibitor (53)). fied, and purified. pcDNA3 (F/V), which generates high levels of expression, was used as parent control vector for subsequent constructions and Effects of IL-4, IL-3, GM-CSF, and PMA on tyrosine transfections into FDCP2 cells. Wild-type (WT) MEK (F/M) (45) was phosphorylation of IRS-2, and activation of MAP kinase and cloned into the pcDNA3 vector as a BamHI-NotI fragment. Constitutively PKB active (CA) MEK (F/M*) was generated by replacing Ser218 and Ser222 by glutamic acid (46). Kinase-inactive MEK (F/MϪ) was generated by replac- As reported previously (11), IL-4, not IL-3, increased tyrosine 218 222 ing Ser and Ser by alanine (46). phosphorylation of IRS-2 (Fig. 1B). GM-CSF and PMA also in- Purified DNA (ϳ20 ␮g) was introduced into FDCP2 cells by electro- poration; transformants (F/V, F/M, F/M*, and F/MϪ) were selected by duced little or no tyrosine phosphorylation of IRS-2. On the other growing cells in 24-well culture plates in medium with G418 (750 mg/L), hand, as shown in Fig. 1C, IL-4, IL-3, GM-CSF, and PMA all as previously described (18, 47). Six different colonies of each construct activated MAP kinase. IL-4, which has been reported to activate The Journal of Immunology 4867 Downloaded from

FIGURE 1. Effects of IL-4, IL-3, GM-CSF, and PMA on activation of PDE3, PDE4, MAP kinase, and PKB, and on tyrosine phosphorylation of http://www.jimmunol.org/ IRS-2. FDCP2 cells were incubated with IL-4 (7.4 nM for 10 min), IL-3 (3 nM for 10 min), GM-CSF (3 nM for 10 min), or PMA (50 nM for 30 min). A, Cell lysates were assayed for total PDE activity (Ⅺ) and PDE3 (j) and PDE4 (f) activities. Results represent mean values Ϯ SEM from three or more individual experiments. Where not present, SEs are too small to be presented graphically. IL-4, IL-3, GM-CSF, and PMA activated PDE4, FIGURE 2. Time course of IL-4-induced tyrosine phosphorylation of p Ͻ IRS-2 and activation of PDE, MAP kinase, and PKB. FDCP2 cells were ,ء) while only IL-4 (not IL-3, GM-CSF, and PMA) activated PDE3 p Ͻ 0.01; ϩ, p ϭ NS). B, After incubation of FDCP2 cells with incubated for the indicated times with 7.4 nM IL-4. A, Cell lysates from ,ءء ;0.001 IL-4, IL-3, GM-CSF, and PMA for 10 min, cell lysates were immunopre- control and IL-4-treated cells were assayed for total PDE activity (control, Ⅺ f E cipitated with anti-IRS-2 Ab and immunoblotted with anti-phosphotyrosine ; IL-4 treated, ) or immunoprecipitated with anti-IRS-2 Ab ( ). After by guest on September 28, 2021 and anti-IRS-2 Abs. C, MAP kinase activity was assessed by phosphory- immunoblotting with anti-phosphotyrosine Abs, phosphorylated IRS-2 E lation of MBP protein by MAP kinase immunoprecipitates. D, PKB activ- bands ( ) were quantitated by PhosphorImager analysis and expressed as ity was assessed by phosphorylation of histone 2B by PKB immunopre- arbitrary PhosphorImager units. B, IRS-2 immunoprecipitates were immu- cipitates. Results in B–D are representative of two or more individual noblotted with anti-PI3-K and anti-IRS-2 Abs. C, MAP kinase activity was experiments. Similar results were obtained if MAP kinase and PKB activ- assessed by phosphorylation of MBP peptide by MAP kinase immunopre- ities were assessed by phosphorylation of MBP peptide or Crosstide, re- cipitates. D, PKB activity was assessed by phosphorylation of Crosstide by Ϯ spectively (data not shown). PKB immunoprecipitates. Results represent mean values SEM from three individual experiments in A, C, and D, while B is representative of two individual experiments. Where not present, SEs were too small to be presented graphically. In a time-dependent manner, IL-4 increased PDE, MAP kinase, and PKB activities as well as tyrosine phosphorylation of MAP kinase in some (54, 55), but not all, cells (56), increased .(p Ͻ 0.001; ϩ, p ϭ NS ,ء) IRS-2 MAP kinase activity to a lesser extent than IL-3, GM-CSF, and PMA, assessed by phosphorylation of MBP (Fig. 1C)orMBP peptide (not shown). IL-4 also activated PKB to a lesser extent than did IL-3 or GM-CSF, assessed by phosphorylation of histone (Fig. 1D) or Crosstide (not shown). As reported in adipocytes (36, with anti-p85 PI3-K Ab, an 85-kDa protein was detected in the 37) and lymphocytes (57), PMA did not activate PKB in FDCP2 immunoprecipitates. Transient activation of PDE by IL-4 was also cells (Fig. 1D). associated with transient activation of MAP kinase (Fig. 2C) and PKB (Fig. 2D). Time course of effects of IL-4 on PDE activity, tyrosine phosphorylation of IRS-2, translocation of PI3-K, and activation Effects of TNF-␣, AG-490, and wortmannin on IL-4-induced of MAP kinase and PKB in FDCP2 cells tyrosine phosphorylation of IRS-2 and STAT6 and on As shown in Fig. 2A, incubation with IL-4 resulted in activation of translocation of PI3-K total PDE, which was increased by ϳtwofold within 10 min and The IL-4R lacks intrinsic protein tyrosine kinase activity. IL-4- declined to control values within 25–30 min. This transient acti- induced tyrosine phosphorylation of IRS-2 and other cellular tar- vation of PDE (which was dependent on IL-4 concentration, with gets, including STAT proteins, is thought to be mediated by JAK ϳ ␣ EC50 10 ng/ml IL-4 (data not shown)) was associated with a kinases (1 and 3) (11, 12, 58). TNF- has been shown to reduce transient increase in tyrosine phosphorylation of IRS-2 (Fig. 2A) insulin-induced tyrosine phosphorylation of IRS-1 in Fao hepa- and its association with the p85␣ subunit of PI3-K (Fig. 2B). As toma cells (59, 60). As shown in Fig. 3, TNF-␣, in a concentration seen in Fig. 2B, when lysates from IL-4-stimulated FDCP2 cells (Fig. 3, A and B)- and time-dependent (Fig. 3C) fashion, and the were immunoprecipitated with anti-IRS-2 Ab and immunoblotted JAK inhibitor AG-490 (Fig. 3D) reduced IL-4-induced tyrosine 4868 ACTIVATION OF PDE3 AND PDE4 BY CYTOKINES IN FDCP2 CELLS Downloaded from http://www.jimmunol.org/ FIGURE 3. Effects of TNF-␣ on IL-4-stimulated tyrosine phosphorylation of IRS-2, on IL-4-induced association of the 85-kDa regulatory subunit of PI3-K with IRS-2, and on tyrosine phosphorylation of STAT6. A, After incubation of FDCP2 cells with the indicated concentrations of TNF-␣ for 2 h, and then with IL-4 (7.4 nM) for 10 min, IRS-2 was immunoprecipitated from cell lysates with anti-IRS-2 Ab and immunoblotted with anti-IRS-2 and anti-phosphotyrosine Abs. B, Effects of increasing concentrations of TNF-␣ on IL-4-stimulated tyrosine phosphorylation of IRS-2 were quantiﬁed by p Ͻ 0.001). C, FDCP2 cells were incubated with 5 nM TNF-␣ for the ,ء ,PhosphorImager analysis (mean arbitrary PhosphorImager units Ϯ SEM, n ϭ 3 indicated times before exposure to 7.4 nM IL-4 for 10 min, after which IRS-2 was immunoprecipitated from cell lysates with anti-IRS-2 Ab and immunoblotted with anti-IRS-2 and anti-phosphotyrosine Abs. D–G, After incubation of FDCP2 cells with TNF-␣ (5 nM for 2 h), wortmannin (100 nM for 30 min), or AG-490 (5 ␮M for 30 min), and then with IL-4 (7.4 nM) for 10 min, IRS-2 was immunoprecipitated with anti-IRS-2 Ab, and immunoblotted with anti-phosphotyrosine (D), and anti-p85 and anti-IRS-2 Abs (E). F, Cell lysates were immunoprecipitated with anti-p85 Ab, and immunoblotted with by guest on September 28, 2021 anti-IRS-2 and anti-p85 Abs. G, Cell lysates were immunoprecipitated with anti-STAT6 Ab, immunoblotted with anti-phosphotyrosine and anti-STAT6 Abs, and visualized by PhosphorImager analysis. Results are representative of three or more individual experiments.

phosphorylation of IRS-2, and its association with PI3-K (Fig. 3, did not block tyrosine phosphorylation of STAT6 (Fig. 3G), these D, E, and F) in FDCP2 cells. Wortmannin, an inhibitor of PI3-K, results suggest that activation of PDE by IL-4 is downstream of also blocked association of PI3-K and IRS-2 (Fig. 3, E and F), but IRS-2/PI3-K signals, not STAT signals. The MEK-1 inhibitor did not inhibit IL-4-induced phosphorylation of IRS-2 (Fig. 3D). PD98059 (10 ␮M), which completely blocked IL-4-induced acti- Similar effects of TNF-␣, wortmannin, and AG-490 in blocking vation of MAP kinase (Fig. 4A), only partially blocked activation association of PI3-K and IRS-2 were observed after either immu- of total PDE activity (Fig. 4, A and B). This partial inhibition of noprecipitation of IRS-2 and immunoblotting with anti-p85 (Fig. PD98059 on total PDE reﬂected almost complete inhibition of IL- 3E) or immunoprecipitation of PI3-K and immunoblotting with 4-induced activation of PDE4 with little effect on IL-4-induced anti-IRS-2 (Fig. 3F). The detailed mechanisms whereby TNF-␣ activation of PDE3 (Fig. 4B). Neither rapamycin (20 nM) nor reduces tyrosine phosphorylation of IRS-2, and wortmannin RO31-7549 (2 ␮M) interfered with IL-4-induced activation of blocks the association of IRS-2 with PI3-K, are unknown. Oth- MAP kinase or PDE (Fig. 4A). Thus, downstream of PI3-K, IL- ers have suggested that TNF-␣ stimulation of serine phosphor- 4-induced signals diverge, with PDE4 regulated by MAP kinase- ylation of IRS-1 was associated with inhibition of tyrosine dependent and PDE3 by MAP kinase-independent mechanisms. phosphorylation (60). As shown in Fig. 3G, the JAK kinase inhibitor AG-490 inhibited IL-3 and PMA activate PDE4 by MAP kinase-dependent signals phosphorylation of STAT6 as well as of IRS-2 (Fig. 3D). In con- As seen in Fig. 5, IL-3-induced activation of MAP kinase and trast to its effect on IRS-2 phosphorylation, however, TNF-␣ did PDE4 (Fig. 5A) was transient, with a maximal increase within not block IL-4-induced phosphorylation of STAT6 (Fig. 3G). 10–20 min, and declining thereafter. PD98059 blocked IL-3-induced activation of MAP kinase and PDE4 in a concentration- Downstream of IRS-2 and PI3-K, IL-4 activates PDE3 by MAP dependent manner (Fig. 5B), with maximal effect on both at 10 ␮M kinase-independent and PDE4 by MAP kinase-dependent signals PD98059. As seen in Fig. 6, IL-3-induced activation of MAP ki- Incubation of FDCP2 cells with 5 nM TNF-␣ for 2 h, or with 5 ␮M nase (Fig. 6A) and PDE4 (Fig. 6B) was blocked by AG-490 and AG-490, 100 nM wortmannin, or 50 ␮M LY294002 for 30 min wortmannin and PD98059, but not by TNF-␣, RO31-7549, or before addition of IL-4 completely blocked IL-4-induced activa- rapamycin, suggesting that JAK kinases participate in activating tion of MAP kinase and total PDE activity (Fig. 4A). Since TNF-␣ PI3-K, leading to activation of MAP kinase, PDE4, and PKB. The Journal of Immunology 4869 Downloaded from

FIGURE 5. Time course of IL-3-induced activation of MAP kinase and PDE4 and inhibition by PD98059. A and B, FDCP2 cells were incubated without and with IL-3 (3 nM) for the indicated times (A) or with the in-

dicated concentrations of PD98059 for 30 min and then IL-3 (3 nM) for 10 http://www.jimmunol.org/ min (B). A and B, Cell lysates were assayed for PDE4 activity (Ⅺ) (i.e., activity inhibited by 0.5 ␮M rolipram); MAP kinase activity (F) was assessed by phosphorylation of MBP peptide in MAP kinase immunoprecipitates. Results in A and B represent mean values Ϯ SEM from three or more individual experiments. Where not present, SEs were too small to be presented graphically. IL-4 increased MAP kinase and PDE4 in a time- FIGURE 4. Effects of TNF-␣ and different kinase inhibitors on IL-4- dependent manner; PD98059, in a concentration-dependent manner, inhib- induced activation of MAP kinase and PDE. FDCP2 cells were incubated ,p Ͻ 0.001; ϩ ,ء) ited IL-3-induced activation of PDE4 and MAP kinase with rapamycin (20 nM), wortmannin (100 nM), PD98059 (10 ␮M), p ϭ NS). RO31-7549 (2 ␮M), AG-490 (5 ␮M), and LY-294002 (50 ␮M) for 30 min, by guest on September 28, 2021 and TNF-␣ (5 nM) for 2 h before stimulation with 7.4 nM IL-4 for 10 min. A, MAP kinase activity was assessed by phosphorylation of MBP peptide PKB by IL-4 (Fig. 8A) was blocked by inhibitors of IRS-2 phos- in MAP kinase immunoprecipitates. Total PDE activity was assayed in cell phorylation and PI3-K, i.e., AG-490, TNF-␣, LY 294002, and lysates. Results, expressed as percentage of control (C ϭ 100%), represent p Ͻ wortmannin, findings consistent with the idea that PDE3 and PKB ,ء) mean values Ϯ SEM from three or more individual experiments ,p Ͻ 0.01). Where not present, SEs were too small to be pre- are regulated by PI3-K-dependent mechanisms. As seen in Fig. 8B ,ءء ;0.001 sented graphically. All agents were assessed at least three times, but not IL-3 activation of PKB was blocked by inhibitors acting upstream always in the same experiment. B, After incubation of FDCP2 cells with of, or directly on, PI3-K, including AG-490 and wortmannin. Ac- 7.4 nM IL-4 in the absence or presence of PD98059 for 10 min, cell lysates tivation of PKB (Fig. 8B) or MAP kinase and PDE4 (Fig. 6) by were assayed for total (Ⅺ), PDE3 (j), and PDE4 (f) activities (pmol IL-3, which did not induce tyrosine phosphorylation of IRS-2 (Fig. cAMP/min/mg). Values represent mean Ϯ SEM from five or four (PD ϩ 1), was not blocked by TNF-␣. As seen in Fig. 8, A and B, neither Ͻ ءء Ͻ ء IL-4) individual experiments ( , p 0.001; , p 0.01). PD98059 (which did block effects of IL-3 and IL-4 on PDE4) nor rapamycin inhibited PKB, although higher concentrations of PD98059 (50 ␮M) did partially block activation of PKB by IL-3 As seen in Fig. 7, the time course of PMA-induced activation of (unpublished). As also seen in Fig. 8, A and B, the PKC inhibitor PDE4 and MAP kinase is different from that induced by IL-4 or (RO31-7549) did not block PKB activation, consistent with the IL-3. PMA-induced activation of PDE4 was dependent on PMA finding that PMA did not activate PKB in FDCP2 cells (Fig. 1D). concentration (data not shown), was maximal within 25–30 min, and remained elevated for ϳ60 min (Fig. 7A). The time course of Activation of MAP kinase and PDE4 in FDCP2 cells that PDE4 activation correlated with that for MAP kinase activation overexpress MEK (Fig. 7B). As shown in Fig. 6, PMA-induced activation of MAP Several independently isolated cell lines were generated during kinase (Fig. 6C) and PDE4 (Fig. 6D) was inhibited by RO31-7549 transfection of FDCP2 cells with constructs that expressed vector and PD98059, but not TNF-␣, wortmannin, AG-490, or rapamy- alone (F/V) and wild-type (F/M), constitutively active (F/M*), and cin, suggesting that PMA activates MAP kinase by PKC-depen- kinase inactive (F/MϪ) forms of MEK; immunoreactive MEK was dent signals, not via IRS-2, PI3-K, JAK, or p70S6 kinase-depen- much higher in cells transfected with MEK constructs than in un- dent signals. infected cells (Fig. 9). In the absence of IL-3, MAP kinase activity was similar in F/V and F/MϪ cells, slightly elevated in F/M, and Activation of PKB by IL-4 and IL-3 markedly increased in F/M* cells that overexpress constitutively No specific inhibitors or pharmacological tools are available to activated MEK, as evidenced by the mobility shift of MAP kinase examine the role of PKB in the activation of PDE3 and PDE4 by during SDS-PAGE (Fig. 9B) or phosphorylation of MBP peptide IL-4 and IL-3. As is the case for activation of PDE3, activation of in MAP kinase immunoprecipitates (Fig. 9C). PDE4, not PDE3, 4870 ACTIVATION OF PDE3 AND PDE4 BY CYTOKINES IN FDCP2 CELLS Downloaded from

FIGURE 6. Effects of TNF-␣ and different kinase inhibitors on IL-3- and PMA-induced activation of MAP kinase and PDE4. After incubation without http://www.jimmunol.org/ or with TNF-␣ (5 nM) for 2 h, or PD98059 (10 ␮M), RO31-7549 (2 ␮M), wortmannin (100 nM), AG-490 (5 ␮M), or rapamycin (20 nM) for 30 min, FDCP2 cells were incubated with 3 nM IL-3 for 10 min (left panel), or 50 nM PMA for 30 min (right panel). All agents were tested three or more times, but not always in the same experiment. A and C, MAP kinase activity was assessed by phosphorylation of MBP peptide in MAP kinase immunoprecipitates. B and D, PDE4 activity (B) or total (Ⅺ), PDE3 (j), or PDE4 activities (D) were assayed in cell lysates. Results in A, B, and C represent mean values Ϯ SEM (% control, with control taken as 100%) from three or more individual experiments; in D, mean values (pmol cAMP/mg/min) Ϯ SEM from three or more individual experiments. Where not present, SEs were too small to be presented graphically. PD98059, wortmannin, and AG-490 signiﬁcantly .(p Ͻ 0.001 ,ء) p Ͻ 0.001); RO31-7549 and PD98059, activation by PMA ,ء) inhibited activation of MAP kinase and PDE4 by IL-3

was activated in F/M* cells (Fig. 9, Table I). Incubation with IL-3 of MAP kinase and ϳtwofold activation of PDE4 (not PDE3) in by guest on September 28, 2021 activated MAP kinase in F/V, F/M, and F/MϪ cells (Fig. 9, B and F/V and F/MϪ cells (Table I, Fig. 9), suggesting that kinase-inac- C). Incubation with IL-3 resulted in ϳthree- to fourfold activation tive MEK was not functioning as a dominant negative with respect to activation of either MAP kinase or PDE4. Incubation of F/M cells with IL-3 increased both MAP kinase and PDE4 to levels comparable with those in F/M* cells (Fig. 9, Table I); IL-3 pro- duced a much smaller increase in MAP kinase or PDE4 in F/M* cells than in F/M cells (Fig. 9, Table I). PKB was not increased in cells transfected with MEK constructs; IL-3 activated PKB to the same extent in all FDCP2 cells, indicating that even in cells overexpressing wild-type and constitutively activated MEK, PKB was regulated appropriately by IL-3 (Fig. 9). Although kinase-inactive MEK did not function as a dominant negative with respect to activation of PDE4 (Fig. 9, Table I), in F/M* cells and IL-3-treated F/V and F/M cells, the MEK inhibitor PD98059 inhibited activation of MAP kinase and PDE4. These results are consistent with the idea that PDE4 is activated by MAP kinase-dependent signals (Fig. 10).

Discussion FIGURE 7. Time course of PMA-induced activation of MAP kinase Nine different PDE gene families have been identiﬁed (1–6). Each and PDE4. FDCP2 cells were incubated without and with 50 nM PMA for family contains closely related subfamilies that are products of the indicated times. A, Cell lysates were assayed for total PDE activity (Ⅺ), different genes or arise from alternative transcription initiation छ E and PDE3 ( ) and PDE4 ( ) activities. B, MAP kinase activity was as- sites on the same gene or by alternative splicing of mRNAs. Rep- sessed by phosphorylation of MBP peptide or MBP (data not shown) by resentatives of at least two to three different families are found in MAP kinase immunoprecipitates. Results in A and B represent mean values Ϯ SEM from three or more individual experiments and indicate that most cells. Although they are highly regulated enzymes (1–3), lit- PMA increased total PDE and PDE4 (not PDE3) activities and MAP kinase tle is known of the effects of cytokines on PDEs. Our results dem- ,p Ͻ 0.001; ϩ, p ϭ NS). Where not present, SEs were too small onstrate that PMA and the cytokines, IL-4, IL-3, and GM-CSF ,ء) activity to be presented graphically. activate PDE3 and PDE4 by distinct signaling pathways in FDCP2 The Journal of Immunology 4871 Downloaded from http://www.jimmunol.org/

FIGURE 8. Effect of TNF-␣ and different kinase inhibitors on activation of PKB by IL-3 and IL-4. FDCP2 cells were incubated with rapamycin (20 nM), wortmannin (100 nM), PD98059 (10 ␮M), RO31-7549 (2 ␮M), AG-490 (5 ␮M), or LY-294002 (50 ␮M) for 30 min, or TNF-␣ (5 nM) for 2 h, and then with IL-4 (7.4 nM) (A) or IL-3 (3 nM) (B) for 10 min. PKB activity was assessed by phosphorylation of Crosstide by PKB immunoprecipitates. A, IL-4-stimulated PKB activity was significantly inhibited by by guest on September 28, 2021 wortmannin, TNF-␣, AG-490, and Ly294002 (p Ͻ 0.001), while IL-3- stimulated PKB activity (B) was inhibited significantly by wortmannin and FIGURE 9. Effects of IL-3 on PDE, PKB, and MAP kinase activity in ,(p Ͻ 0.001) and not by TNF-␣. One-fifth of the immunopre- MEK-transfected cells. FDCP2 cells transfected with empty vector (F/V ,ء) AG-490 and wild-type (F/M), constitutively active (F/M*), and kinase-inactive (F/ cipitate was subjected to SDS-PAGE and immunoblotted with anti-PKB Ϫ Ab. Results represent mean values Ϯ SEM (% control, with control values M ) MEK constructs were incubated with or without IL-3 for 10 min. A, ϳ ␮ taken as 100%) from two or three experiments in A and B, respectively. All Proteins ( 30 g) in lysates of control and MEK-transfected cells were agents were tested two or three times, but not always in the same experi- separated by SDS-PAGE and immunoblotted with anti-MEK Ab. MAP ment. Where not present, SEs were too small to be presented graphically. kinase activity was assessed by decreased mobility of Erk-1 in (SDS- PAGE/Western) immunoblots (B) or by phosphorylation of MBP peptide (mean cpm Ϯ SEM, n ϭ 3) in MAP kinase immunoprecipitates from cell lysates (ϳ2.5 mg protein) (C). D, Cell lysates were assayed for total PDE (Ⅺ), PDE3 (j), and PDE4 (f) activities (mean pmol/min/mg Ϯ SEM, n ϭ cells. Activation of PDE3 and PDE4 also exhibits a common char- 3). E, PKB activity was assessed by phosphorylation of Crosstide (mean acteristic of cytokine-mediated events, namely overlapping (acti- cpm Ϯ SEM, n ϭ 3) in PKB immunoprecipitates from lysates (ϳ2.5 mg vation of PDE4 by IL-4, IL-3, GM-CSF) and specific (activation of protein) from control and IL-3-stimulated cells. IL-3 significantly stimu- PDE3 by IL-4) effects. PDEs may thus serve as one locus for lated MAP kinase, total PDE and PDE4 activities in F/V, F/M, F/MϪ cells, cross-talk between cytokine and cyclic nucleotide signaling sys- and PKB in all cells. Total PDE and PDE4 and MAP kinase activities were .(p Ͻ 0.001; ϩ, p ϭ NS ,ء) tems. Whether activation of PDE3 and PDE4 by cytokines medi- increased in F/M* cells in the absence of IL-3 ates specific effects of IL-4 and IL-3 is not known. Where not present, SEs were too small to be presented graphically. Although we utilized pharmacologic agents and inhibitors (with their inherent uncertainties as to absolute specificity of action) to help identify signaling pathways (presented in a schematic fashion in Fig. 10) involved in activation of PDE3 by IL-4 and of PDE4 by phorylation of specific intracellular proteins and activation of sig- IL-4, IL-3, GM-CSF, and PMA, our conclusions are supported by naling pathways, including PI3-K-regulated signals (Fig. 10). their selective effects, i.e., TNF-␣ blocking IL-4, not IL-3, respons- Since IL-4, not IL-3, induces JAK-dependent tyrosine phosphor- es; PD98059 blocking activation of PDE4, not PDE3; AG-490 ylation of IRS-2, its association with PI3-K and activation of PDE3 blocking cytokine, not PMA responses; PKC inhibitor RO31-7549 and PDE4, IRS-2/PI3-K signals may be critical in these effects of blocking PMA, not cytokine, responses, etc. IL-4. This idea is supported by selective effects of TNF-␣ on IL-4 Effects of IL-4 and IL-3 on PDE3 and PDE4 were inhibited by signals. Activation of PDE4 by IL-3, which does not induce ty- JAK and PI3-K inhibitors, consistent with the idea that PI3-K sig- rosine phosphorylation of IRS-2, is blocked by JAK and PI3-K nals are responsible for PDE activation and that receptor-mediated inhibitors, but not TNF-␣. TNF-␣, which does not block IL-4- activation of JAKs is central to cytokine-induced tyrosine phos- induced phosphorylation of STAT6, reduces IL-4-induced tyrosine 4872 ACTIVATION OF PDE3 AND PDE4 BY CYTOKINES IN FDCP2 CELLS

lated p140 protein in forming a Shc/Grb2/SOS complex and initi- ating signals downstream from the IL-3R (22). Although IL-4 and IL-3 activated PDE3 and PDE4 via PI3-K, PMA activated PDE4 via PKC (based on sensitivity to RO31- 7549). Downstream of PI3-K and PKC, IL-4 activated PDE3 by MAP kinase-independent signals, whereas IL-4, IL-3, and PMA activated PDE4 by MAP kinase-dependent signals (Fig. 10). In FDCP2 cells permanently transfected with and overexpressing wild-type MEK, IL-3 activated MAP kinase and PDE4 (not PDE3); in cells transfected with constitutively active MEK, MAP kinase and PDE4 (not PDE3) were activated in the absence of IL-3, and only slightly further increased by IL-3. The MEK inhibitor PD98059 blocked activation of MAP kinase and PDE4 in F/M* cells transfected with constitutively active MEK, or in control (F/V) cells or WT (F/M) cells incubated with IL-3. Whether MAP kinase is the proximate kinase that regulates PDE4 or is part of a kinase cascade is not known. rPDE4 was phosphorylated by MAP kinase in vitro, but with no change in PDE activity (62). The mechanisms whereby PKC and PI3-K activate MAP kinase have Downloaded from not yet been completely defined. MAP kinases are phosphorylated and activated by MEKs that can be phosphorylated and activated by Raf-1 kinase (29). PMA has been reported to activate Raf and FIGURE 10. Effect of PD98059 on IL-3-induced activation of MAP MAP kinases in intact cells (63, 64). Inhibition of PI3-K with kinase and PDE4 in MEK-transfected cells. FDCP2 cells transfected with wortmannin or expression of dominant-negative PI3-K blocked vector alone (F/V) and wild-type (F/M) and constitutively active (F/M*) activation of MAP kinase in some (37, 65–67), but not all (68) http://www.jimmunol.org/ MEK constructs were incubated without or with PD98059 (10 ␮M) for 30 cells. Wortmannin also blocked activation of Raf-1 by platelet- min and then without or with IL-3 for 10 min. A, MAP kinase activity was derived growth factor in Chinese hamster ovary cells (65), by in- assessed by phosphorylation of MBP peptide in MAP kinase immunopre- sulin in 3T3-L1 adipocytes (30), and by IGF-1 in L6 cells (67). cipitates. B, Total PDE (Ⅺ), PDE3 (j), and PDE4 (f) activities were Overexpression of p110␣ PI3-K activated raf-1 and MAP kinase in Ϯ assayed in cell lysates. Results in A and B represent mean values SEM frog oocytes (69), but not other cells (70, 71). Others, however, from three or more individual experiments. PD98059 significantly inhib- have suggested that PI3-K is a direct target of ras (72). ited MAP kinase and total PDE and PDE4 activities in F/M*, F/V, and F/M The initial impetus for the studies in this work came from our p Ͻ 0.001). Where not present, SEs were too ,ء) cells treated with IL-3 small to be presented graphically. interest in regulation of PDE3B in adipocytes (2, 28), from iden- tifying PDE3B mRNAs in human lymphocytes (4), and in recog- by guest on September 28, 2021 nizing analogous signaling mechanisms for IL-4 in FDCP2 cells (11, 12, 17, 18) and insulin in adipocytes (2, 28) (Fig. 10). In some respects, the effects of insulin and IL-4 on PDE3 in adipocytes and phosphorylation of IRS-2 and its association with PI3-K and in- FDCP2 cells are similar. Insulin stimulation of adipocytes induces hibits IL-4-induced activation of MAP-kinase, PKB, PDE3, and association of tyrosine-phosphorylated IRS-1 with the p85 regu- PDE4. Thus, IL-4-induced activation of PDE3 and PDE4 may in- latory subunit of PI3-K, activation of PI3-K, and phosphorylation volve IRS-2/PI3-K signals, not STAT signals, and TNF-␣ may and activation of a microsomal PDE3B (28, 73). Wortmannin, not interfere with IRS-2/PI3-K signals. These observations also sup- PD98059, blocked insulin- and IL-4-induced activation of PKB port the idea that different regions of the IL-4R regulate different and PDE3 in adipocytes (28, 36, 37) and FDCP2 cells, respec- functions, with the region between amino acids 437–557 as critical tively. In these cells, PDE3 is apparently activated by PI3-K-de- for IRS-2 interaction/phosphorylation and IL-4-induced prolifera- pendent/MAP kinase-independent pathways (Fig. 10). Wijkander tion, and the region between amino acids 557–657, for regulation et al. (37) demonstrated that during partial purification of activated of JAK phosphorylation/activation of STAT6 and induction of PKB from insulin-stimulated adipocytes, insulin-sensitive kinase gene expression (19). Some effects of TNF-␣ on IL-4 signaling (as activity that phosphorylated PDE3B in vitro cofractionated with well as on signaling by insulin and other growth factors and cy- PKB. Further characterization of the role of PKB in regulation of tokines) may be mediated via TNF-␣ inhibition of tyrosine phos- PDE3 in intact cells will require pharmacological inhibitors of phorylation of, and signaling via, IRS proteins. While all effects of PKB, and transfection of wild-type and mutant PKB into FDCP2 TNF-␣ are not mediated via inhibition of IRS signaling, disruption cells. Initial studies in FDCP2 cells that overexpress rPKB do in- of these signals may represent one mechanism whereby TNF-␣ dicate that PDE3B (not PDE4) is a downstream target of activated can selectively regulate certain actions of specific cytokines, i.e., PKB.3 those that utilize IRS adapter proteins. Although PI3-K may mediate effects of both IL-4 and IL-3 on Whereas IL-4-induced translocation of PI3-K to IRS-2 repre- PDE3 and PDE4, IL-4 signals differ from those of IL-3. Effects of sents a plausible mechanism for IL-4-induced activation of PI3-K, IL-4 are blocked by TNF-␣ and may depend on PI3-K/IRS-2 sig- other mechanisms are responsible for activation of PI3-K by IL-3 nals. IL-3, on the other hand, apparently activates PI3-K and PKB and GM-CSF (61). The phosphatase SHPTP2 binds to the tyrosine-phosphorylated ␤c subunit of the IL-3R, where it is tyrosine phosphorylated (by JAK2 or src family kinases) and may function 3 F. Ahmad, L.-N. Cong, L. Stenson Holst, L.-M. Wang, T. Rahn Landstrom, J. H. as an adapter molecule between activated IL-3 ␤c and Grb/SOS Pierce, M. J. Quon, E. Degerman, and V. C. Manganiello. 1998. Cyclic nucleotide phosphodiesterase (PDE)3B is a downstream target of Akt/PKB and is involved in and PI3-K, thus regulating both Ras/Raf and PI3-K signaling path- regulation of effects of Akt/PKB on cell proliferation/survival. Submitted for publi- ways (15, 20, 21). Others have implicated a tyrosine-phosphory- cation. The Journal of Immunology 4873

induced STAT1 and STAT4 DNA-binding activity and induction of Fc␥RI mRNA in peripheral human mononuclear cells (80). Whether PDEs play a role in PKB regulation of apoptosis or proliferation (24) is unknown. In recent studies, however, we found that PDE3B is phosphorylated and activated in FDCP2 cells expressing constitutively active PKB. Thymidine incorporation, which was greater in these cells than in control cells, was inhibited to a greater extent by inhibition of PDE3 than by the PDE4 inhibitor rolipram.3 These and other results suggest that PDE3 is a downstream target, if not substrate, for activated PKB, and that activated PDE3B may regulate cAMP pools that are important in effects of PKB on proliferation/survival of FDCP2 cells. In any case, FDCP2 myeloid cells may represent a model system to study regulation by cytokines of distinct and overlapping signaling pathways involved in regulation and integration of the activities of different phosphodiesterases in the same cell. FIGURE 11. Hypothesis regarding signaling pathways involved in activation of PDE3 in adipocytes and of PDE3 and PDE4 in FDCP2 cells. In rat adipocytes, insulin activates PDE3B by MAP kinase- and p70S6 kinase- Acknowledgments Downloaded from independent signals that may involve activation of PKB kinase (PDK-1) and PKB. Whether PKB is the kinase that phosphorylates PDE3B in intact We thank Dr. Martha Vaughan for critical reading of the manuscript and adipocytes is not yet proven. In FDCP2 cells, IL-4 also activates PDE3 by Carol Kosh for excellent secretarial assistance. MAP kinase-independent signals, whereas IL-4, IL-3, GM-CSF, and PMA all activate PDE4 via MAP kinase signals. Whether MAP kinase and PKB directly phosphorylate/activate speciﬁc PDE4 and PDE3 isoforms, respec- References http://www.jimmunol.org/ tively, in FDCP2 cells is also not certain. 1. Conti, M., G. Nemoz, C. Sette, and E. Vicini. 1995. Recent progress in under- standing hormonal regulation of cyclic nucleotide phosphodiesterase. Endocr. Rev. 16:370. 2. Degerman, E., P. Belfrage, and V. C. Manganiello. 1997. Structure, localization, and regulation of cGMP-inhibited phosphodiesterase (PDE3). J. Biol. Chem. 272: 6823. by JAK-dependent signals that are independent of IRS-2 and of 3. Torphy, T. J. 1998. Phosphodiesterase inhibitors: molecular targets for novel ␣ antiasthma agents. Am. J. Crit. Care Med. 157:351. TNF- . If activation of PDE3 in adipocytes by insulin (36, 37), or 4. Ekholm, D., B. Hemmer, G. Gao, M. Vergelli, R. Martin, and V. C. Manganiello. in FDCP2 cells by IL-4, does involve IRS-2/PI3-K-mediated ac- 1997. Differential expression of cyclic nucleotide phosphodiesterase 3 and 4 activation of PKB, mechanisms must exist whereby PKB activated tivities in human T cell clones speciﬁc for myelin basic protein. J. Immunol.

159:1520. by guest on September 28, 2021 by IL-4, not by IL-3, specifically targets PDE3. Perhaps IRS-2 (or 5. Soderling, S., S. Bayuga, and J. Beavo. 1998. Identification of a novel family of the IL-4R) recruits and/or initiates a specific signaling complex (a cyclic nucleotide phosphodiesterases. J. Biol. Chem. 273:15553. specific physical or functional signaling subset utilizing PI3-K and 6. Fisher, D., J. Smith, J. Pillar, S. St. Dennis, and J. Cheng. 1998. Isolation and characterization of PDE8A, a novel human cAMP-specific phosphodiesterase. downstream kinases, including PKB) that targets PDE3, since ac- Biochem. Biophys. Res. Commun. 246:570. tivation of PI3-K and PKB by IL-3 are not in themselves sufficient 7. Csar, X., A. C. Ward, B. W. Hoffman, G. G. Guy, and J. A. Hamilton. 1997. cAMP suppresses p21ras and Raf-1 response, but not the ERK-1 response to for activation of PDE3 (Fig. 10). Whether, for example, IL-3 and G-CSF: possible Raf-1 independent activation of Erk-1. Biochem. J. 322:79. IL-4 induce formation of different signaling complexes (or similar 8. Williamson, E. A., G. S. Burgess, P. Eder, S. Litz-Jackson, and H. S. Boswell. complexes in different subcellular compartments) or utilize differ- 1997. cAMP negatively controls c-myc transcription and G1 cell cycle progres- sion in p210-BCR-ABl-transformed cells: inhibitory activity exerted through cy- ent isoforms of PI3-K, PKB, and/or other signaling molecules is clin D1 and cdk4. Leukemia 11:73. not known. From this perspective, regulation of PKB and PDE3 in 9. Giembycz, M. A., C. J. Corrigan, J. Seybold, R. Newton, and P. J. Barnes. 1996. Identification of cAMP phosphodiesterases 3, 4, and 7 in human CD4ϩ and FDCP2 cells may be analogous to regulation of glucose transport ϩ CD8 T-lymphocytes: role in regulating proliferation and production of IL-2. in adipocytes, where both insulin and platelet-derived growth fac- Br. J. Pharmacol. 118:1945. tor increase PI3-K activity, but only insulin stimulates glucose 10. Hurwitz, R. C., M. Hirsch, D. J. Clarke, V. N. Holcombe, and M. Y. Hurwitz. 1990. Induction of a calmodulin-dependent phosphodiesterase during PHA-stim- transport (74). ulated mitogenesis. J. Biol. Chem. 265:8901. In adipocytes, insulin-induced activation of PDE3B (which de- 11. Wang, L.-M., A. D. Keegan, W. E. Paul, M. A. Heideran, J. S. Gutkind, and creases cAMP and A-kinase and, consequently, hormone-stimu- J. H. Pierce. 1992. IL-4 activates a distinct signal transduction cascade from IL-3 in factor-dependent myeloid cells. EMBO J. 11:4899. lated triglyceride lipase activity) is important in the antilipolytic 12. Wang, L.-M., A. Keegan, M. Frankel, W. E. Paul, and J. H. Pierce. 1995. Signal action of insulin (2, 28). In frog oocytes, insulin- or IGF-1-induced transduction through the IL-4 and insulin receptor families. Nat. Immun. Cell activation of PDE3B, perhaps via PKB, reduces cAMP and stim- Growth Regul. 13:360. 13. Hara, T., and A. Miyajima. 1996. Function and signal transduction mediated by ulates meiotic maturation (75). In 3T3F442A preadipocytes, PDE4 the IL-3 receptor. Nat. Immun. Cell Growth Regul. 14:605. is activated by growth hormone via PI3-K and p70S6 kinase-de- 14. Taniguchi, T. 1995. Cytokine signalling through nonreceptor protein tyrosine kinases. Science 268:251. pendent signals (76); in both 3T3-L1 (77) and 3T3-F442 cells (76), 15. Bone, H., U. Dechert, F. Jurik, J. Schrader, and W. Welham. 1997. SHP1 and PDE4, not PDE3, regulates a cAMP pool important during differ- SHP2 protein tyrosine phosphatases associate with ␤c after IL-3 induced tyrosine entiation of preadipocytes into adipocytes. In hemopoietic/im- phosphorylation. J. Biol. Chem. 272:14470. 16. Ihle, J. 1996. STATs: signal transducers and activators of transcription. Cell mune/inflammatory cells, cyclic nucleotides can modulate multi- 84:331. ple functions, including proliferation, migration, reactive oxygen 17. Wang, L.-M., A. D. Keegan, W. Li, G. E. Lienhard, S. Pacini, J. S. Gutkind, metabolite formation, cytokine and mediator production and se- M. G. Myers, X.-J. Sun, M. F. White, S. A. Aaronson, W. E. Paul, and J. H. Pierce. 1993. Common elements in interleukin-4 and insulin signalling cretion, etc. (3). Cytokine-induced activation of PDE3 and PDE4 pathways in factor-dependent hematopoietic cells. Proc. Natl. Acad. Sci. USA by IL-3 and IL-4 could, for example, indirectly modulate FDCP2 90:4032. 18. Wang, L.-M., M. G. Myers Jr., X. Sun, S. A. Aaronson, M. F. White, and cell proliferation (11), perhaps by preventing cAMP-induced in- J. H. Pierce. 1993. IRS-1: essential for insulin and IL-4-stimulated mitogenesis in hibition of Raf (78) or p70S6K (79). cAMP also inhibited IL-6- hematopoietic cells. Science 261:1591. 4874 ACTIVATION OF PDE3 AND PDE4 BY CYTOKINES IN FDCP2 CELLS

19. Ryan, J. J., L. J. McReynolds, A. Keegan, L.-H. Wang, E. Garfein, 44. Cross, D. A. E., D. R. Alessi, P. Cohen, M. Andjelkovich, and B. A. Hemmings. P. Rothman, K. Nelms, and W. E. Paul. 1996. Growth and gene expression are 1995. Inhibition of glycogen synthase kinase-3 by insulin mediated by protein predominantly controlled by distinct regions of the human IL-4 receptor. Im- kinase B. Nature 378:785. munity 4:123. 45. Crews, C. M., A. Alessandrini, and R. L. Erickson. 1992. The primary structure 20. Welham, M. J., U. Dechert, K. B. Leslie, F. Jirik, and J. W. Schrader. 1994. of MEK, a protein kinase that phosphorylates the ERK gene product. Science Interleukin-3 and granulocyte/macrophage colony stimulating factor, but not 258:478. IL-4, induce tyrosine phosphorylation, activation and association of SHPTP2 46. Cowley, S., H. Paterson, P. Kemp, and C. J. Marshal. 1994. Activation of MAP with Grb2 and phosphatidylinositol-3-kinase. J. Biol. Chem. 269:23764. kinase kinase is necessary and sufficient for PC12 differentiation and for trans- 21. Corey, S., A. Equinoa, K. Puiyana-Theall, J. B. Bolen, L. Cantley, F. Mollinedo, formation of NIH-3T3 cells. Cell 77:841. T. R. Jackson, P. T. Hawkins, and L. R. Stephens. 1993. Granulocyte macrophage 47. Pierce, J., M. Ruggiero, T. Fleming, P. DiFiore, J. Greenberger, L. Varticovski, stimulating factor stimulates both association and activation of phosphatidyl-3- J. Schlessinger, G. Rovera, and S. Aaronson. 1988. Signal transduction through kinase and src-related tyrosine kinases in human myeloid-derived cells. EMBO J. the EGF receptor transfected in IL-3 dependent hematopoietic cells. Science 239: 12:2681. 628. 22. Matsuguchi, T., R. Salgia, M. Hallek, M. Eder, B. Druker, T. J. Ernst, and 48. Meydan, N., T. Grunberger, H. Dadi, M. Shahar, E. Arpaia, Z. Lapidot, J. D. Griffin. 1994. Shc phosphorylation in myeloid cells is regulated by granu- J. S. Leeder, M. Freedman, A. Cohen, A. Gazit, A. Levitzki, and C. M. Roifman. locyte macrophage colony stimulating factor, interleukin-3, and steel factor and 1996. Inhibition of acute lymphoblastic leukemia by a JAK-2 inhibitor. Nature is constitutively increased by p210BCR/ABL. J. Biol. Chem. 269:5016. 379:645. 23. Sato, N., T. Sakamaki, N. Terada, K. Arai, and A. Miyajima. 1993. Signal 49. Alessi, D. R., A. Cuenda, P. Cohen, D. T. Dudley, and A. R. Saltiel. 1995. transduction by the high affinity GM-CSF receptor: two distinct cytoplasmic PD098059 is a specific inhibitor of the activation of mitogen-activated protein regions of the common ␤ subunit responsible for different signalling. EMBO kinase kinase in vitro and in vivo. J. Biol. Chem. 270:27489. J. 12:4181. 50. Povis, G., R. Bonjouklian, M. Berggren, A. Galligos, R. Abraham, C. Ashendel, 24. Songyang, Z., D. Baltimore, L. Cantley, D. Kaplan, and T. F. Franke. 1997. L. Zalkow, L. Mather, J. Dodge, G. Grindley, and C. Vlakos. 1994. Wortmannin, Interleukin-3-dependent survival by the Akt protein kinase. Proc. Natl. Acad. Sci. a potent and selective inhibitor of phosphatidylinositol-3-kinase. Cancer Res. USA 94:11345. 54:2419. 25. Saksmaki, K., I. Miyajima, T. Kitamura, and A. Miyajima. 1992. Critical cyto- Downloaded from plasmic domains of the common B subunit of the human GM-CSF, IL-3 and IL-5 51. Vlakos, C., W. F. Matter, K. Hiu, and R. Brown. 1994. A specific inhibitor of receptors for growth signal transduction and tyrosine phosphorylation. EMBO J. phosphatidylinositol 3-kinase, 2-(4-Morpholinyl)-8-phenyl-4H-1 benzopyran-4- 11:3541. one (LY294002). J. Biol. Chem. 269:5241. 26. Myers, M. Jr., and M. White. 1996. Insulin signal transduction and the IRS 52. Lin, T.-A., and J. C. Lawrence Jr. 1994. Activation of ribosomal protein S6 proteins. Annu. Rev. Pharmacol. 36:615. kinases does not increase glycogen synthesis or glucose transport in rat adipo- 27. Sun, X.-J., S. Pons, L.-M. Wang, Y. Zhang, L. Yenush, D. Burks, M. Myers Jr., cytes. J. Biol. Chem. 269:21255. E. Glastein, N. Copeland, N. Jenkins, J. Pierce, and M. White. 1997. The IRS-2 53. Wilkinson, S. E., P. J. Parker, and J. S. Nixon. 1993. Isoenzyme specificity of gene on murine chromosome 8 encodes a unique signalling adapter for insulin bisindolylmaleimides, selective inhibitors of protein kinase C. Biochem. J. 294: and cytokine action. Mol. Endocrinol. 11:251. 335. http://www.jimmunol.org/ 28. Rahn, T., M. Ridderstrale, H. Tornqvist, V. C. Manganiello, G. Fredrikson, 54. Wery, S., M. Letourneur, J. Bertoglio, and J. Pierre. 1996. Interleukin-4 induces P. Belfrage, and E. Degerman. 1994. Essential role of phosphatidylinositol 3-ki- activation of mitogen activated protein kinase and phosphorylation of Shc in nase in insulin induced activation and phosphorylation of cGMP-inhibited cAMP human keratinocytes. J. Biol. Chem. 271:8529. phosphodiesterase in rat adipocytes: studies using the selective inhibitor wort- 55. Chuang, L. M., T. Y. Tai, R. C. Kahn, H. P. Wu, S. C. Lee, and B. J. Lin. 1996. mannin. FEBS Lett. 350:314. Signal transduction pathways for interleukin-4 and insulin in human hepatoma 29. Denton, R. M., and J. M. Tavare. 1995. Does mitogen-activated protein kinase cells. J. Biochem. 120:111. have a role in insulin action? The cases for and against. Eur. J. Biochem. 56. Welham, M. J., V. Duronio, and J. W. Schrader. 1994. Interleukin-4-dependent 227:597. proliferation dissociates p44crk-1, p42crk2, and p21ras activation from cell growth. 30. Suga, J., Y. Yoshimasa, K. Yamada, Y. Yamamoto, G. Inoue, M. Okamoto, J. Biol. Chem. 269:5865. T. Hayashi, M. Shigemoto, A. Kosaki, H. Kuzuya, and K. Nakao. 1997. Differ- 57. Reif, K., B. Burgering, and P. A. Cantrell. 1997. Phosphatidylinositol 3-kinase ential activation of mitogen-activated protein kinase by insulin and epidermal links the IL-2 receptor to protein kinase B and p70S6 kinase. J. Biol. Chem. by guest on September 28, 2021 growth factor in 3T3–L1 adipocytes. Diabetes 46:735. 272:14426. 31. Proud, C. G. 1996. p70S6 kinase: an enigma with variation. Trends Biochem. Sci. 58. Chin, X. H., B. Patel, L. M. Wang, M. Frankel, N. Ellmore, R. Flavell, 21:181. W. LaRochelle, and J. Pierce. 1997. JAK-1 expression is required for mediating 32. Chung, J., T. Grammar, K. Lemon, A. Kazlanskas, and J. Blenis. 1994. PDGF- IL-4 induced tyrosine phosphorylation of IRS and STAT6 signalling molecules. and insulin-dependent p70S6K activation mediated by phosphatidylinositol- J. Biol. Chem. 272:6556. 3-OH kinase. Nature 370:71. 59. Feinstein, R., H. Kanety, M. Z. Papa, B. Lunenfeld, and A. Karasik. 1993. Tumor 33. Hemmings, B. A. 1997. Akt signalling: linking membrane events to life and death necrosis factor-␣ suppresses insulin induced tyrosine phosphorylation of insulin decisions. Science 275:628. receptor and its substrates. J. Biol. Chem. 268:26055. 34. Franke, T. F., S. I. Yang, T. O. Chan, K. Datta, A. Kaglauskas, D. K. Morrison, 60. Kanety, H., R. Feinstein, M. Z. Papa, R. Hemi, and A. Karasik. 1995. Tumor D. R. Kaplan, and P. N. Tsichlis. 1995. The protein kinase encoded by the Akt necrosis factor ␣ induced phosphorylation of insulin receptor substrate-1 (possi- proto-oncogene is a target of PDGF-activated phosphatidylinositol 3-kinase. Cell ble mechanism for suppression of insulin stimulated tyrosine phosphorylation of 81:727. IRS-1). J. Biol. Chem. 270:23780. 35. Moule, S. K., G. Welsh, N. Edgell, E. Foulstone, C. Proud, and R. Denton. 1997. 61. Gold, M., V. Duronio, S. Saxena, J. Schrader, and R. Aebersold. 1994. Multiple Regulation of protein kinase B and glycogen synthase kinase-3 by insulin and cytokines activate phosphatidylinositol 3-kinase in hematopoietic cells. J. Biol. ␤-adrenergic agents in rat epididymal fat cells. J. Biol. Chem. 272:7713. Chem. 269:5403. 36. Wijkander, J., L. S. Holst, T. Rahn, S. Resjo, I. Castan, V. C. Manganiello, 62. Lenhard, J. M., D. B. Kassel, W. J. Rocque, L. Hamacher, W. D. Holmes, I. Patel, P. Belfrage, and E. Degerman. 1997. Regulation of protein kinase B in rat adi- C. Hoffman, and M. Luther. 1996. Phosphorylation of a cAMP-specific phos- pocytes by insulin, vanadate, and peroxovanadate: membrane translocation in phodiesterase (HSPDE4B2B) by mitogen-activated protein kinase. Biochem. J. response to peroxovanadate. J. Biol. Chem. 272:21520. 316:751. 37. Wijkander, J., T. Rahn Landstrom, V. C. Manganiello, P. Belfrage, and E. Degerman. 1998. Insulin induced phosphorylation and activation of phospho- 63. Troppman, J., J. Bruder, H. Munoz, P. Lloyd, J. Kyriakis, P. Banerjee, J. Avruch, diesterase 3B in rat adipocytes: possible role for protein kinase B but not mito- and U. Rapp. 1994. Mitogen-activated protein kinase/extracellular signal-regu- gen-activated protein kinase or p70S6 kinase. Endocrinology 139:219. lated protein kinase activation by oncogenes, serum, and 12–0-tetradecanoyl- 38. Beavo, J., and D. Reifsnyder. 1990. Primary sequence of cyclic nucleotide phos- phorbol-13-acetate requires Raf and is necessary for transformation. J. Biol. phodiesterase isoenzymes and the design of selective inhibitors Trends Biochem. Chem. 269:7030. Sci. 11:150. 64. Kolch, W., G. Heldecker, G. Kochs, R. Hummell, H. Vahidol, H. Mischak, ␣ 39. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the G. Finkenzeller, D. Marme, and U. Rapp. 1993. Protein kinase C activates Raf-1 head of bacteriophage T4. Nature 277:680. by direct phosphorylation. Nature 364:249. 40. Kamps, M. P. 1991. Generation and use of anti-phosphotyrosine antibodies for 65. Duckworth, B., and L. C. Cantley. 1997. Conditional inhibition of the mito- immunoblotting. Methods Enzymol. 201:101. gen-activated protein kinase cascade by wortmannin. J. Biol. Chem. 272: 41. Kameshita, I., and H. Fujisawa. 1989. A sensitive method for detection of calm- 27665. odulin-dependent protein kinase II activity in sodium dodecyl sulfate-polyacryl- 66. Welsh, G., E. J. Foulstone, S. W. Young, J. Tavare, and C. G. Proud. 1994. amide gel. Anal. Biochem. 183:139. Wortmannin inhibits the effects of insulin and serum on the activities of 42. Cano, E., C. A. Hazzlin, and I. C. Mahadevan. 1994. Anisomycin-activated glycogen-synthase kinase-3 and mitogen-activated protein kinase. Biochem. protein kinases p45 and p55 but not mitogen activated protein kinases Erk-1 J. 303:15. and -2 are implicated in the induction of c-fos and c-jun. Mol. Cell. Biol. 67. Cross, D., D. Alessi, J. Vandenheede, H. McDowell, H. Hundal, and P. Cohen. 14:7352. 1994. Inhibition of glycogen-synthase kinase-3 by insulin or IGF-1 is blocked by 43. Leighton, I. A., K. N. Dalby, F. B. Caudwell, P. T. W. Cohen, and P. Cohen. wortmannin, but not by rapamycin: evidence that wortmannin blocks activation 1995. Comparison of the specificities of p70S6 kinase and MAP KAP ki- of mitogen-activated protein kinase pathway in L6 cells between Ras and Raf. nase-1 identifies a relatively specific substrate for p70S6 kinase: the N-ter- Biochem. J. 303:21. minal kinase domain of MAP KAP kinase-1 is essential for peptide phos- 68. Nakamura, K., C.-J. Zhou, J. Pasente, and C. S. Chew. 1996. Parietal cell MAP phorylation. FEBS Lett. 375:289. kinases: multiple activation pathways. Am. J. Physiol. 271:G640. The Journal of Immunology 4875

69. Hu, Q., A. Klippel, A. Muslin, W. Fantl, and L. Williams. 1995. Ras-dependent 75. Andersen, C., R. Roth, and M. Conti. 1998. Protein kinase B/Akt induces re- induction of cellular responses by constitutively active phosphatidylinositol-3- sumption of meiosis in Xenopus oocytes. J. Biol. Chem. 273:18705. kinase. Science 208:100. 76. Mackenzie, S., S. Yarwood, A. Peden, G. Bolger, R. Vernon, and M. Houslay. 70. Frevert, I., and B. Kahn. 1997. Differential effects of constitutively active phos- 1998. Stimulation of p70S6 kinase via growth-hormone controlled phosphatidyl- phatidylinositol 3-kinase on glucose-transport, glycogen synthase activity and inositol-3-kinase pathway leads to the activation of a PDE4A cyclic AMP-spe- DNA synthesis in 3T3–L1 adipocytes. Mol. Cell. Biol. 17P:190. cific phosphodiesterase in 3T3F442A preadipocytes. Proc. Natl. Acad. Sci. USA 71. Klipel, A., C. Reinhard, W. M. Kavanaugh, G. Aspell, M.-A. Escobedo, and 95:3549. L. Williams. 1996. Membrane localization of phosphatidylinositol-3-kinase is sufficient to activate multiple signal-transducing pathways. Mol. Cell. Biol. 16: 77. Elks, M., and V. C. Manganiello. 1985. A role for soluble cAMP phosphodies- 4117. terases in differentiation of 3T3–L1 adipocytes. J. Cell. Physiol. 124:191. 72. Rodriguez-Viccana, P., P. Warne, R. Dhand, B. Vanhaisebroeck, I. Gout, M. Fry, 78. Stevenson, B. R., X. Kong, and J. C. Lawrence Jr. 1993. Increasing cAMP at- M. Waterfield, and J. Downward. 1994. Phosphatidylinositol-3-kinase is a direct tenuates activation of mitogen activated protein kinase. Proc. Natl. Acad. Sci. target of Ras. Nature 370:527. USA 90:10305. 73. Myers, M. Jr., T. Grammar, L.-M. Wang, X.-J. Sun, J. Pierce, J. Blevis, and 79. Monfar, M., K. Lemon, T. Grammar, L. Cheatham, J. Chung, C. Vlakos, and M. White. 1994. IRS-1 mediates phosphatidylinositol 3-kinase and p70S6k J. Blenis. 1995. Activation of pp 70/85 S6 kinases in interleukin-2-responsive signalling during insulin, IGF-1 and IL-4 stimulation. J. Biol. Chem. 269: 28783. cells is mediated by phosphatidylinositol-3-kinase and inhibited by cAMP. Mol. 74. Isakoff, S. J., G. Taha, E. Rose, J. Marcusohn, A. Klip, and E. Y. Scolnick. 1995. Cell. Biol. 15:326. The inability of phosphatidylinositol 3-kinase activation to stimulate Glut4 trans- 80. Sengupta, T., E. Schmidt, and L. Ivashkiv. 1996. Inhibition of cytokines and location indicates additional signaling pathways are required for insulin-stimu- JAK-STAT activation by distinct signalling pathways. Proc. Natl. Acad. Sci. USA lated glucose uptake. Proc. Natl. Acad. Sci. USA 92:10247. 93:9499. Downloaded from http://www.jimmunol.org/ by guest on September 28, 2021