J Biol Chem, Vol. 275, Issue 10, 7395-7402, March 10, 2000

The Pyridinyl Imidazole Inhibitor SB203580 Blocks Phosphoinositide-dependent Protein Kinase Activity, Protein Kinase B Phosphorylation, and Retinoblastoma Hyperphosphorylation in Interleukin-2-stimulated T Cells Independently of p38 Mitogen-activated Protein Kinase^*

Ferdinand V. Lali, Abigail E. Hunt, Sarah J. Turner, and Brian M. J. Foxwell^§

From the Kennedy Institute of Rheumatology, London W6 8LH, United Kingdom

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Pyridinyl imidazole inhibitors, particularly SB203580, have been widely used to elucidate the roles of p38 mitogen-activated protein (MAP) kinase (p38/HOG/SAPKII) in a wide array of biological systems. Studies by this group and others have shown that SB203580 can have antiproliferative activity on cytokine-activated lymphocytes. However, we recently reported that the antiproliferative effects of SB203580 were unrelated to p38 MAP kinase activity. This present study now shows that SB203580 can inhibit the key cell cycle event of retinoblastoma protein phosphorylation in interleukin-2-stimulated T cells. Studies on the proximal regulator of this event, the phosphatidylinositol 3-kinase/protein kinase B (PKB)(Akt/Rac) kinase pathway, showed that SB203580 blocked the phosphorylation and activation of PKB by inhibiting the PKB kinase, phosphoinositide-dependent protein kinase 1. The concentrations of SB203580 required to block PKB phosphorylation (IC₅₀ 3-5 µM) are only approximately 10-fold higher than those required to inhibit p38 MAP kinase (IC₅₀ 0.3-0.5 µM). These data define a new activity for this drug and would suggest that extreme caution should be taken when interpreting data where SB203580 has been used at concentrations above 1-2 µM.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Interleukin-2 (IL-2)¹ is a potent T cell growth factor that mediates its effects via a high affinity heterotrimeric receptor comprising , , and _c subunits. Several intracellular signaling pathways are known to be activated by IL-2, including the p42/44 mitogen-activated protein kinase (MAP kinase, also known as ERK2/1), the p38 and p54 MAP kinases (also called stress kinases, or HOG and JNK, respectively), the phosphatidyl inositol 3' (PI) 3-kinase pathway and the Jak/STAT (signal transducer and activator of transcription) pathways. Our earlier studies using the MEK (mitogen-activated protein kinase/extracellular signal-regulated kinase kinase) inhibitor PD098059 (1) and those of others (2, 3) have indicated that the p42/44 MAP kinase pathway is not required for IL-2-driven proliferation. In contrast, a pyridinyl imidazole inhibitor of p38 MAP kinase, SB203580, inhibited IL-2-driven T cell proliferation with an IC₅₀ of 3-5 µM, suggesting a possible role for p38 MAP kinase in this process (4). Recently, we have further investigated the role of p38 MAP kinase in proliferation by mapping the subdomains of the IL-2 receptor  chain involved in the activation of the kinase. As previously shown for p42/44 MAP kinase, activation of p38 and p54 MAP kinases required the acidic rich A region of the IL-2 receptor  chain (5). However, the A region is not required for proliferation (2, 5), indicating that neither p38 MAP kinase nor p54 MAP kinase is essential for this function. Furthermore, CNI-1493 (6, 7), an inhibitor of p38 and p54 MAP kinase activation by IL-2 was unable to inhibit proliferation (5). Surprisingly, SB203580 was still able to inhibit proliferation in the absence of IL-2 stimulated p38 MAP kinase activation. It has already been reported that SB203580 does inhibit p54 MAP kinase activity (8, 9), but the possibility that the anti-proliferative effects of SB203580 may be mediated by effects on p54 MAP kinase can also be discounted by the studies of Hunt et al. (5).

The above data suggest the existence of a novel target of SB203580 that is critical for mitogenic signaling. To help understand the mechanism of action and possible targets of SB203580, the effect of this inhibitor on cell cycle-regulated proteins was examined. We found that the IL-2-induced hyperphosphorylation of Rb was greatly reduced in SB203580-treated cells, whereas the drug had no effect on cytokine-induced p27^kip degradation or Myc expression. Subsequent studies investigated the effects of SB203580 on PI 3-kinase/protein kinase B (PKB), as these kinases have been implicated in the regulation of Rb phosphorylation (10). We observed that although PI 3-kinase activity was unaffected, the phosphorylation of PKB on Thr³⁰⁸ and Ser⁴⁷³ was inhibited, resulting in an inhibition of IL-2-induced kinase activity. The concentrations of SB203580 required were commensurate with antiproliferative effects of the drug and provide evidence that 3-phosphoinositide-dependent protein kinase 1 (PDK1; the Thr³⁰⁸ kinase) is inhibited by SB203580. It has recently been reported that PDK1 also acts as the Ser⁴⁷³ kinase (11). These results could provide a mechanism for the antiproliferative effect of SB203580. Furthermore these data will have important implications for the interpretation of results from numerous published studies in which SB203580 has been used at concentrations in excess of 2 µM.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials

IL-2 was a generous gift from Dr. P. Lomedico (Roche Molecular Biochemicals). Antiphospho-specific Ser⁴⁷³ and Thr³⁰⁸ PKB were from New England Biolabs (Hitchin, Herts, UK), whereas sheep anti-PKB and other reagents for in vitro PKB kinase assay were sold as a kit by Upstate Biotechnology, Inc. (Lake Placid, NY). Anti-Rb was from Pharmingen (San Diego, CA), and the monoclonal anti-cyclin D3 and anti-p27^kip1 were from Pierce. Rabbit anti-p70S6 kinase was from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA) and was used for both immunoprecipitation and Western blotting. Rabbit anti-c-Myc antibody was from Santa Cruz (Santa Cruz, CA). SB203580 was purchased from Calbiochem-Novabiochem, and U5 anti-p85 monoclonal antibody was a generous gift from Professor D. Cantrell (Imperial Cancer Research Fund, London). Rabbit antisera to p38 MAP kinase was from Professor J. Saklatvala (Kennedy Institute of Rheumatology, London). Second-layer antibodies (horseradish peroxidase-conjugated) were purchased from DAKO (DAKO A/S Denmark). Recombinant PDK1 and recombinant PKB were kindly provided by Dr. L. Stephens (Babraham Institute, Cambridge, UK) and Dr. D. Alessi (Dundee, UK). CN1-1493 was from Dr. K. Tracey (Picower Institute of Medical Research New York).

Methods

Cells and Proliferation Assay The IL-2-dependent murine T cell line, CT6, was grown and maintained as described previously (12). These cells were rested by washing three times in RPMI and culturing overnight in RPMI, 5% fetal calf serum in the absence of growth factor, antibiotics, or -mercaptoethanol supplements. 2-5 × 10⁶ rested CT6 cells were resuspended in 2 ml of RPMI, 5% fetal calf serum and preincubated with inhibitors or vehicle control as indicated in figure legends. Cells were then stimulated with 20 ng/ml recombinant human IL-2 for 5 min at 37 °C and pelleted in a minifuge for 30 s, medium was aspirated, and the pellet was lysed in the appropriate buffer. BA/F3 cells stably expressing deletion mutants of IL-2 receptor  chain (a generous gift from Professor T. Taniguchi, Tokyo, Japan) were maintained in glutamine containing RPMI further supplemented with 5% fetal calf serum and 0.2 µg/ml G418 (Calbiochem-Novabiochem) as described previously (13). Human peripheral blood mononuclear cells were prepared from buffy coat leukophoresis residues (North London Blood Transfusion Service, Colindale, London UK) and activated with 50 ng/ml OKT3 for 48 h. The cells were then washed extensively, rested overnight, and washed again before activating with IL-2; such cell preparations were >90% T cells (14). Cellular proliferation assays were performed by measurement of [³H]thymidine incorporation as described previously (12).

Kinase Assays

PKB Kinase Assay-- Cells were lysed in Buffer A (see below) for Western blotting and PKB kinase assays. Kinase assays were performed according to the manufacturer's instructions. Briefly, 4 µg of sheep anti-PKB was immobilized on 25 µl of protein G-Sepharose overnight (or 1.5 h) and washed in Buffer A (50 mM Tris, pH 7.5, 1 mM EDTA, 1 mM EGTA, 0.5 mM Na₃VO₄, 0.1% -mercaptoethanol, 1% Triton X-100, 50 mM sodium fluoride, 5 mM sodium pyrophosphate, 0.1 mM phenylmethylsulfonyl fluoride, 1 µg/ml aprotinin, pepstatin, leupeptin, and 1 µM microcystin). The immobilized anti-PKB was then incubated with 0.5 ml of lysate (from 5 × 10⁶ cells) for 1.5 h and washed three times in 0.5 ml of Buffer A supplemented with 0.5 M NaCl, two times in 0.5 ml of Buffer B (50 mM Tris-HCl, pH 7.5, 0.03% (w/v) Brij-35, 0.1 mM EGTA, and 0.1% -mercaptoethanol), and twice with 100 µl of assay dilution buffer; 5× assay dilution buffer is 100 mM MOPS, pH 7.2, 125 mM -glycerophosphate, 25 mM EGTA, 5 mM sodium orthovanadate, 5 mM DTT. To the PKB enzyme immune complex was added 10 µl of assay dilution buffer, 40 µM protein kinase A inhibitor peptide, 100 µM PKB-specific substrate peptide, and 10 µCi of [-³²P]ATP, all made up in assay dilution buffer. The reaction was incubated for 20 min at room temperature with shaking, then samples were pulse spun, and 40 µl of reaction volume were removed into another tube to which was added 20 µl of 40% trichloroacetic acid to stop the reaction. This was mixed and incubated for 5 min at room temperature, and 40 µl was transferred onto P81 phosphocellulose paper and allowed to bind for 30 s. The P81 pieces were washed three times in 0.75% phosphoric acid then in acetone at room temperature. -³²P incorporation was then measured by scintillation counting.

PI 3-Kinase Assay-- Cells were lysed in PI 3-kinase lysis buffer (40 mM Tris-HCl, pH 7.5, 200 mM NaCl, 1 mM EGTA supplemented with 1 mM DTT, 1 mM Na₃VO₄, and 10 µg/ml each of aprotinin, pepstatin, leupeptin) at 10 × 10⁶ cells/ml, and the post-nuclear lysate was precleared with 25 µl of protein G-Sepharose for 1 h then preincubated with 5 µg of monoclonal anti-p85 (U5) and further with 25 µl of protein G-Sepharose for the final 1 h. The pellets were washed three times in 0.5 ml of PI 3-kinase assay buffer. The pellet was then resuspended in 25 µl of kinase assay buffer. To this, 10 µl of a 1 mg/ml mixture of phosphatidylinositol and phosphatidylserine (made up in 100 mM HEPES, pH 7.5, and sonicated just before use) was added. The mixture was then preincubated at room temperature for 10 min, and the reaction was started by the addition of 15 µl of ATP mixture (340 µl of water, 4.2 µl of 1 M MgCl₂, 16 µl of 100 mM ATP) supplemented with 5 µCi of [-³²P]ATP. The reaction proceeded for 15 min and was stopped by the addition of 100 µl of 1 M HCl and vortexing, adding a further 200 µl of a 1:1 chloroform:methanol and vortexing again, and microfuge-spinning the tubes for 5 min. The lower layer was removed and dried in vacuo (or at 60 °C on dry block) then redissolved in 10 µl of 4:1 chloroform:methanol before spotting onto silica plates. The plate was developed in a preequilibrated vertical tank with chloroform, methanol, 28% ammonium hydroxide, water (180:140:10.8:27.5) for 3 h (or overnight) followed by phosphorimaging analysis (Fuji FLA-2000).

p70S6 Kinase Assay-- Cells were lysed in 0.5 ml of p70S6 kinase lysis buffer (10 mM potassium phosphate, pH 7.05, 0.5% Triton X-100, 1 mM EDTA, 5 mM EGTA, 1 mM Na₃VO₄, 1 mM phenylmethylsulfonyl fluoride, 10 µg/ml leupeptin, 1 µg/ml pepstatin, 1 µg/ml aprotinin), and the postnuclear lysate was precleared with 20 µl of protein A-agarose for 30 min. The precleared supernatant was then preincubated with 5 µl of rabbit antiserum to p70S6 kinase for 1 h and additionally with 25 µl of protein A-agarose with mixing for a further 1 h, all at 4 °C. The final immune complex was washed twice in 0.5 ml of lysis buffer and twice in 0.5 ml of kinase assay buffer (50 mM MOPS, pH 7.2, 1 mM DTT, 30 mM ATP, 5 mM MgCl₂, 10 mM p-nitrophenylphosphate). To washed immune complex pellet was added 45 µl of assay mixture (made up of 35 µl of kinase assay buffer, 5 µl of 125 mM substrate peptide (KKRNRTLTK), 5 µl of 50 µM protein kinase A inhibitor, 5 µCi of [-³²P]ATP), and the reaction was allowed to continue for 30 min at room temperature. The reaction was stopped by the addition of reducing sample buffer and boiling for 5 min. After separation on a peptide gel as described before (15), radioactivity incorporated into peptide was quantitated by phosphorimaging.

Recombinant PDK1/PKB Assays-- Lipid vesicles were made by drying down a mixture of phosphatidylcholine and phosphatidylserine in vacuo and reconstituting with lipid buffer (0.2 M NaCl, 20 mM HEPES, 2 mM EGTA) to a final 5 times working stock (500 µM phosphatidylcholine, 500 µM phosphatidylserine, and 100 µM phosphatidylinositol 3,4,5-trisphosphate (PIP₃) and sonicated before use. EE-tagged recombinant PDK1 and PKB (both >98% pure) were prediluted in enzyme dilution buffer (1 mM DTT, 0.1 M NaCl, 1 mM EGTA, 20 mM HEPES). PDK1 assays were performed with 1 µM EE-PKB and 50 nM EE-PDK1 in the presence of appropriately diluted lipid vesicles, 0.5 µM ATP, and 1 µCi of [-³²P]ATP in assay buffer (8 mM MgCl₂, 0.12 M NaCl, 1.2 mM DTT, 1.2 mM EGTA, 0.01% azide) supplemented with protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 10 µg/ml aprotinin, 10 µg/ml leupeptin) in a final volume of 5 µl. The reaction was allowed to continue for 5 min at 30 °C and stopped by boiling with 10 µl of 1.5 times SDS sample buffer (with 5 mM EDTA). The PKB autokinase assays were performed as above for PDK1 but in the absence of PDK1 and PIP₃. Samples were then resolved on a 10% SDS-polyacrylamide electrophoresis gel and quantitated by phosphorimaging (Fuji FLA-2000).

Western Blotting-- After separation by SDS-polyacrylamide gel electrophoresis, proteins were transferred onto nitrocellulose or polyvinylidene difluoride membranes that were then blocked with 5% (w/v) milk powder in Tris-buffered saline/Tween 20 (0.1%) for 1 h at room temperature. The blocked membranes were then incubated in 1:1000 dilution of first-stage antibody (made up in Tris-buffered saline Tween, 2% bovine serum albumin supplemented with 0.1% azide) overnight, then washed and incubated with 1:1000 horseradish peroxidase-conjugated secondary antibody (made up in Tris-buffered saline Tween, 5% (w/v) milk powder) for 1 h at room temperature before developing with ECL.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

SB203580 Inhibits IL-2-induced Rb Hyperphosphorylation-- As we have shown previously (4, 5), the IL-2-induced proliferation of primary human T cells, murine CT6 T cells, or BAF F7 B cells is prevented by the p38 MAP kinase inhibitor with an IC₅₀ of 3-5 µM (Fig. 1). However, as our recent studies (5) showed that IL-2-induced proliferation and the inhibitory effects of SB203580 on this event were independent of p38 MAP kinase or even p54 MAP kinase activity in both T cells and B cells, we endeavored to identify other possible targets involved in mediating the anti-proliferative effects. To do this, we investigated events associated with cell cycle progression. SB203580 had no effect on Myc expression, except for a small reduction at 30 µM only (Fig. 2a). Furthermore, nuclear staining of SB203580-treated CT6 cells with propidium iodide showed no evidence of apoptosis after stimulation with IL-2 for 20 h (data not shown). The expression of hyperphosphorylated Rb and degradation of p27^kip1 were also measured as markers of S-phase entry. The addition of IL-2 to resting CT6 cells caused the hyperphosphorylation of Rb as detected by Western blotting (Fig. 2b). The presence of SB203580 in the antiproliferative (0-30 µM) range resulted in a dose-dependent reduction in the hyperphosphorylated form (Fig. 2b). There also appeared to be some reduction in the total levels of Rb protein. Similar inhibitory effects on Rb hyperphosphorylation and protein levels were obtained with wortmannin and LY294002, both inhibitors of PI 3-kinase. The decrease in Rb protein is likely to be due to the IL-1-converting enzyme (ICE)-mediated proteolysis of the hypophosphorylated form, which has been previously reported (16-18). These results would also agree with previous studies on the role of PI 3-kinase in IL-2-induced Rb hyperphosphorylation and protein by Brennan et al. (10) and would support previous indications that PI 3-kinase is a proximal regulator of Rb (10, 19, 20). The effects of SB203580 on Rb hyperphosphorylation were confirmed in similar studies on activated primary human T cells (Fig. 2d).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Effect of SB203580 on IL-2-induced proliferation. Effect of SB203580 on the proliferation of CT6 (diamonds), BA/F3 cell line F7 (squares), and PBMC/T cells (circles). Rested cells were preincubated with 0-30 µM SB203580 or vehicle control for 1 h and cultured for 24 h in the presence of 20 ng/ml IL-2. 0.5 µCi/well of [3H]thymidine was added for the last 6 h. Thymidine incorporation was measured by scintillation counting, and the data points are expressed as the mean of triplicate measurements ± S.E. The data show [3H]thymidine incorporation as a percentage of [3H]thymidine incorporated into cells without inhibitor. The average IL-2-induced thymidine incorporation (cpm ± S.E.) for the different cell lines were 166,977 ± 7,465 for CT6 cells, 93,219 ± 3,537 for PBMCs, and 130,571 ± 661 for the BA/F3 F7 cells. Data are representative of at least three separate experiments.

View larger version (35K): [in this window] [in a new window]   Fig. 2.   Effect of SB203580 on cell cycle proteins. Resting CT6 cells (a, b, c) and PBMC (d and e) were preincubated with the indicated doses of SB203580, wortmannin, LY294002, or vehicle control for 1 h before stimulating with 20 ng/ml IL-2 or vehicle control for 3 h (c-Myc) or 20 h (Rb and p27kip). Cells were then lysed and prepared for total cellular c-Myc (a) Rb (b and d) or p27kip (c and e) measurement by Western blotting as described under "Experimental Procedures." The data are representative of three separate experiments. pRb, hypophosphorylated; ppRb, hyperphosphorylated Rb.

We also investigated a second cell cycle-regulated protein, p27^kip1. The addition of IL-2 to resting CT6 cells induces the degradation of p27^kip1(Fig. 2c). This degradation was unaffected by SB203580, which if anything, further reduced levels of the protein. Wortmannin and LY294002 similarly had no inhibitory effect on p27^kip1 degradation. Again, these studies on p27^kip1 degradation were repeated in activated primary human T cells, with no significant inhibition observed with SB203580 or wortmannin, although LY294002 had some inhibitory effect (Fig. 2e).

SB203580 Inhibits the Phosphorylation and Activation of PKB-- The characteristic, S-phase hyperphosphorylation of Rb induced by IL-2 has been reported to be mediated by the PI 3-kinase pathway via the distal effector PKB (10). Furthermore, both the mitogenic and survival functions of the PI 3-kinase pathway have, in several reports, been attributed to PKB (10, 15, 19-22). We were therefore interested in investigating the possibility that SB203580 mediates its effects on Rb by inhibiting these kinases, especially as wortmannin and LY294002 displayed similar effects. The activation of PKB requires the PI 3-kinase-generated second messenger PIP₃ as well as phosphorylation on Thr³⁰⁸ and Ser⁴⁷³ mediated by the PIP₃-dependent kinases, PDK1 and PDK2, respectively (23, 24). We investigated the effect of SB203580 on PKB activation by looking at IL-2-induced phosphorylation of residue Ser⁴⁷³ of PKB in whole cell lysates using a phospho-specific antibody. In both CT6 and activated human T cells, SB203580 inhibited the phosphorylation of Ser⁴⁷³ in a dose-dependent manner (Fig. 3, a and b). Similar studies on the IL-2-responsive BA/F3 F7 B cells gave the same result (Fig. 3c). The approximate IC₅₀ for the effect of SB203580 on this parameter is ~5 µM, similar to the concentration required for the inhibitory effects on proliferation (Fig. 1). As expected, wortmannin (Fig. 3) and LY294002 (not shown) also inhibited Ser⁴⁷³ phosphorylation, whereas rapamycin (not shown) had no effect. The phosphorylation of PKB on Thr³⁰⁸ was similarly investigated. As the antibody was not so effective, PKB was first immunoprecipitated, and phospho-Thr³⁰⁸ was detected by Western blotting. SB203580 inhibited Thr³⁰⁸ phosphorylation in CT6 cells with similar efficacy to the Ser⁴⁷³ phosphorylation (Fig. 4). Wortmannin, as expected, also inhibited this Thr³⁰⁸ phosphorylation. To confirm that the effects of SB203580 on PKB phosphorylation correlated with kinase activity, assays were performed on immunoprecipitated PKB from IL-2-stimulated CT6 cells (Fig. 5). The drug inhibited PKB activation with an IC₅₀ of 3-10 µM, in agreement with its effects on phosphorylation of the kinase and cell proliferation.

View larger version (47K): [in this window] [in a new window]   Fig. 3.   Effect of SB203580 on serine 473 phosphorylation of PKB. CT6 cells (a), activated human T cells (b), and BA/F3 F7 cells (c) were preincubated with 0-30 µM SB203580 or vehicle control for 1 h or 100 nM wortmannin for 15 min before stimulating with 20 ng/ml IL-2 or vehicle control for 5 min. PKB phosphoserine 473 (upper panels) was detected by Western blotting. Membranes were stripped and reprobed for total PKB to confirm equal loading (lower panels). The data are representative of three separate experiments.

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Effect of SB203580 on phosphorylation of threonine-308 on PKB. Rested CT6 were preincubated with the indicated doses of SB203580 or vehicle control for 1 h or 100 nM wortmannin for 15 min before stimulating with 20 ng/ml IL-2 or vehicle control for 5 min. Cells were then lysed, and PKB was immunoprecipitated (IP) with anti-PKB antiserum or irrelevant control antibody. Threonine 308 phosphorylation was assayed by Western blotting (upper panel). Membranes were stripped and reprobed for total PKB to confirm equal loading (lower panel).

View larger version (29K): [in this window] [in a new window]   Fig. 5.   SB203580 inhibits IL-2-induced PKB activity. Rested CT6 cells per sample were preincubated with indicated concentrations of SB203580, wortmannin, or vehicle control before stimulating with 20 ng/ml IL-2 or vehicle control for 5 min. Immunokinase assays were performed as described under "Experimental Procedures." Panel a is representative of three separate experiments. b, the accumulated data from the three experiments is represented graphically showing inhibition of PKB activity as a percentage of uninhibited, IL-2-activated controls. Error bars equal ± S.D. (n = 3).

SB203580 Inhibits the Activation of p70S6 Kinase but Not PI 3-Kinase-- Although the above results showed that PKB activation is inhibited, it is still possible that PKB is one of several mitogenic effector molecules downstream of the actual SB203580 target. Therefore the effect of SB203580 on IL-2-induced activation of the PI 3-kinase/PKB pathway was examined. Exposure of CT6 cells to IL-2 leads to a reproducible 2-fold increase in anti-p85-precipitable PI 3-kinase activity. This was unaffected by preincubating the cells with SB203580. In contrast, wortmannin totally inhibited this activity (Fig. 6). Furthermore, direct addition of SB203580 to PI 3-kinase assays did not have any effect (results not shown), indicating that PI 3-kinase is not the target of the drug. The effect of SB203580 on PI 3-kinase/PKB pathway was also examined indirectly. Several studies have shown that p70S6 kinase is a distal mediator of PI 3-kinase activity in several systems (15, 25). As expected, wortmannin (Fig. 6b) and LY294002 (not shown) inhibited the activation of p70S6 kinase by IL-2, as measured in immunokinase assays. We observed that SB203580 could also inhibit IL-2-induced p70S6 kinase activation, although the concentration required was slightly higher with an IC₅₀ above 10 µM. These observations place the target of SB203580 downstream of PI 3-kinase but upstream of p70S6 kinase. Furthermore, it suggests that the SB203580 target is a common activator of both PKB and p70S6 kinase. The most likely candidate is PDK1, which has been reported to phosphorylate and activate p70S6 kinase. The higher IC₅₀ for p70S6 kinase activation may reflect the fact that PDK1 contributes to only one of several phosphorylations required for p70S6 kinase activation.

View larger version (41K): [in this window] [in a new window]   Fig. 6.   Effect of SB203580 on the activation of PI 3-kinase and p70S6 kinase. Rested cells were preincubated with indicated concentrations of SB203580, wortmannin, or vehicle control before stimulating with 20 ng/ml IL-2 or vehicle control for 5 min. Representative immunokinase assays for PI 3-kinase showing fold activation over unstimulated cells (a) and p70S6 kinase with % inhibition of IL-2-stimulated activity labeled on top of peptide gel panel (b) are shown. The PI 3-kinase experiment is representative of two, and the p70S6 kinase experiment is representative of three repeat experiments.

SB203580 Can Act as an Inhibitor of PDK1-- The data so far suggest that PDK1 and/or PDK2, the PKB kinases, are possible targets for SB203580. The Thr³⁰⁸ kinase, PDK1, is a constitutively active enzyme (26), but the phosphorylation of PKB on Thr³⁰⁸ is regulated by PIP₃. The kinase for Ser⁴⁷³, a putative PDK2, is so named because it is also dependent on PIP₃, but PDK2 has not been fully characterized, and as a result, no direct assays are available to examine its activity. Because we had evidence that PI 3-kinase activity and, therefore, PIP₃ production are not inhibited (Fig. 6a), we examined whether SB203580 could act as a PDK1 inhibitor. To do this, recombinant kinase was used with recombinant PKB as the substrate. Activity of the enzyme was assessed by measuring the incorporation of [³²P]phosphate into PKB (Fig. 7a). SB203580 was able to inhibit the activity of PDK1 in a dose-dependent manner with an IC₅₀ in the 3-10 µM range, but CNI-1493, an inhibitor of p38MAP kinase activation (6), did not affect PDK1 activity. The recombinant PKB used in this assay, but not the PDK1, had endogenous autokinase activity (results not shown), so we tested the effect of the drug on this (Fig. 7b). SB203580 was unable to inhibit PKB autokinase activity. The inhibition of PDK1 by SB203580 identifies it as the putative target in the PI 3-kinase pathway, in agreement with its reported role in p70S6 kinase activation (27, 28) and supported by our finding above (Fig. 6b).

View larger version (29K): [in this window] [in a new window]   Fig. 7.   SB 203580 inhibits PDK1 activity. Recombinant PDK1 activity was assayed after preincubation with 0-30 µM SB203580 or 5 µM CNI-1493 using recombinant PKB as substrate (a). PKB autokinase activity was measured using recombinant PKB in the absence of PDK-1 or PIP3 (b). Reactions were performed as described under "Experimental Procedures." Data is expressed as the percentage of total activity obtained with a 5-min incubation in the absence of inhibitor. The figures are each representative of three separate experiments.

p38 MAP Kinase Is Not Involved in PKB Phosphorylation-- Although the above data demonstrate that SB203580 is an inhibitor of the PKB kinases (PDK1 and by inference PDK2), it was important to discriminate between the effects of SB203580 on PKB activation and the p38 MAP kinase pathway. The compound, CNI-1493 can inhibit the phosphorylation and activation of p38 MAP kinase induced by IL-2 and lipopolysaccharide (5-7). Therefore the effect of CNI-1493 on the IL-2-induced phosphorylation of p38 MAP kinase was compared with that of PKB. This study was performed in BA/F3 F7 cells because the IL-2-induced phosphorylation of p38 MAP kinase was easier to detect in these cells compared with T cells.² CNI-1493 inhibited IL-2-induced phosphorylation of p38 MAP kinase in the dose range 0.5 to 5 µM while having no effect on PKB phosphorylation at Ser⁴⁷³ (Fig. 8). In contrast, wortmannin inhibited PKB phosphorylation and was consistently found to enhance the IL-2-induced phosphorylation of p38 MAP kinase (data not shown), an observation supported by kinase assays (not shown). These data would discount any role for p38 MAP kinase in the regulation of PKB either distal or proximal of PI 3-kinase. The data would also indicate that IL-2-induced PKB and p38 MAP kinase are likely to be on separate IL-2-mediated signaling pathways.

View larger version (9K): [in this window] [in a new window]   Fig. 8.   p38 MAP kinase is not required for PKB phosphorylation. BA/F3 F7 cells were stimulated with 20 ng/ml IL-2 for 15 min following preincubation with 0-5 µM CNI-1493 or vehicle control. Cell extracts were then separated by SDS-polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride. The membrane was cut in half and immunoblotted for the phosphorylation of p38 MAP kinase (upper panel) and PKB (lower panel). The upper panel was reprobed for p38 MAP kinase (not shown) and showed equal loading on all tracks.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Since the original description of the pyridinyl imidazole SB203580 as a specific inhibitor of p38 MAP kinase, the drug has been widely used as a research tool. The correlation drawn between the effects of SB203580 and a role for p38 MAP kinase has been applied to many systems with several hundred publications to date. Although the IC₅₀ for inhibition of p38 MAP kinase activity has been shown to be approximately 0.3 µM (4, 29), many studies have used SB203580 at higher concentrations, often at 10-30 µM or even up to 100 µM (30, 31). Recently two studies have shown that SB203580 could inhibit certain subtypes of the related p54 MAP kinase (JNK/SAPK1) at concentrations of 5-10 µM (8, 9). This is an added complication in the interpretation of results obtained with SB203580, as p54 MAP kinase is often activated in tandem with p38 MAP kinase. However, our recently published findings (5) indicate that the inhibitory effect of SB203580 on proliferation driven by IL-2 was not related to either p38 or p54 MAP kinase function and that the drug was likely to have some other target.

We found that SB203580 in excess of 3 µM had effects on IL-2-induced hyperphosphorylation of Rb that were similar to those of the PI 3-kinase inhibitors, wortmannin and LY294002. This led us to investigate the effects of the inhibitor on the PI 3-kinase/PDK1/PKB pathway. Our data show that SB203580 inhibits the activation of PKB by blocking the phosphorylation of the kinase on the residues Thr³⁰⁸ and Ser⁴⁷³, that are required for kinase function. Furthermore, we have shown that SB203580 inhibits PDK1, the Thr³⁰⁸ kinase, in vitro. The inhibition of Ser⁴⁷³ phosphorylation by the drug is in agreement with the recent publication by Balendran et al. (11) that PDK1 also acts as PDK2 once bound to protein kinase C-related protein kinase-2 (PRK2). The IC₅₀ of SB203580 for the inhibition of PKB phosphorylation is commensurate with that for the antiproliferative effects of the drug. The putative role of PKB in mediating proliferation induced by IL-2 (10) could provide a mechanism for this effect of SB203580. The compound also inhibited the phosphorylation of PKB induced by IL-4³ or the antibody-induced cross-linking of CD45,⁴ showing that the effect of SB203580 is not confined to IL-2 signaling. However, our data disagree with the recent study of Sweeney et al. (32), which shows no inhibition of insulin-induced PKB activation in adipocytes by SB203580. A recent study by Shaw et al. also reports no inhibition of PKB activation in response to cellular stress in SW 3T3 and 293 cells (33). Although both studies used 10 µM SB203580, a concentration at which we see up to 50% inhibition of IL-2-induced PKB activity, they were performed on different lineages of cells from those used in our studies, and furthermore, the stress-activated PKB responses display much slower kinetics compared with the mitogen-induced activation in our studies. It is possible that the cell type or stimulus may modulate the effects of SB203580 on PKB activation. In our studies, the drug inhibited IL-2 responses in both T and B cells.

Following our observations that, at higher concentrations, SB203580 inhibited PKB activation, it was important to distinguish between the effects of the drug on the p38 MAP kinase and PKB pathways. Alessi et al. (34) show that the p38 MAP kinase substrate, MAPKAP kinase-2, could also act as a Ser⁴⁷³ kinase. However, this effect could only be shown with the purified proteins, and the study found no evidence that this occurred in intact cells. Furthermore, a recent study by the same group reported that insulin-like growth factor-1 activated PKB in the absence of p38 MAP kinase activation (33). These studies support our own finding that the compound CNI-1493, which blocks p38 MAP kinase activation by IL-2 (5), had no effect on PKB phosphorylation. In agreement with these observations, we have previously shown that inhibition of IL-2-induced MAPKAP kinase-2 activation by SB203580 had an IC₅₀ of 0.3 µM (4), which is 10-20-fold lower that that required to inhibit PKB phosphorylation. Also, the kinetics of p38 MAP kinase activation is not consistent with a role in IL-2-induced PKB activation. Phosphorylation of PKB is maximal at 5 min, whereas activation of p38 MAP kinase takes 15 min (4). Finally, the study of Alessi et al. (34) shows that MAPKAP kinase-2 did not act as a Thr³⁰⁸ kinase.

Although our data discount a role for p38 MAP kinase in PKB activation, a recent work by Wang et al. (35) reports that Fas-mediated hyperphosphorylation of Rb is p38 MAP kinase-dependent. Because the concentrations of SB203580 used in our study totally inhibit the p38 MAP kinase signaling, the observations of Wang et al. could provide an explanation for the changes we see in Rb phosphorylation. However the reduction we see in Rb phosphorylation does not occur at the lower concentrations of SB203580, which totally inhibit p38 MAP kinase activity in our cells (IC₅₀ = 0.3-0.5 µM (4)). The need to use higher concentrations of SB203580 dissociates the mitogen-induced hyperphosphorylation of Rb from the p38 MAP kinase pathway in our system. Furthermore, it would be interesting to see if an inhibitor of p38 MAP kinase activation, CNI-1493 has a similar effect on Fas-induced Rb hyperphosphorylation.

The effect of SB203580 on the PI 3-kinase-mediated activation of PKB supports the reported role for these kinases in cellular proliferation and survival. In particular, the identification of PDK1 as a putative target of SB203580 provides a possible mechanism for this effect. However in the context of IL-2 signaling, the effect of SB203580 appears to be solely on cell cycle progression, as we have found no evidence of increased apoptosis in treated cells. It would be interesting to see if the phosphorylation of Bad, which can be mediated by PKB (36, 37), is affected by SB203580. The inhibitory effect of SB203580 on the activation of p70S6 kinase in vivo would agree with recent studies that have identified this kinase as a substrate for PDK1 (26, 28). Additionally, PKB has been reported to play a role in p70S6 kinase activation. Although data from membrane-targeted PKB support this putative role in p70S6 kinase activation (38), Dufner et al. (39) find this activity only in the membrane-targeted form of PKB and argue that it may not reflect the normal mechanism of p70S6 kinase activation. The observations made in our study, however, remain consistent with a model of PI 3-kinase signaling in which PDK1 is upstream of both 70S6 kinase and PKB. Besides the established downstream effectors of PI 3-kinase-signaling discussed above, the pathway has also been implicated in the regulation of cell cycle proteins (10). Our findings on Rb regulation agree with these observations. The lack of effect of SB203580 (or wortmannin and LY294002) on p27^kip1 degradation, however, was surprising, since previous work on Kit-225 T cells suggested a role for PKB in IL-2-induced p27^kip1 degradation (10).

These observations raise questions concerning studies that have used the inhibitory effects of SB203580 as a confirmation of the involvement of p38 MAP kinase in a given cellular activity. Although previous reports (29, 40) and our own work (4) would suggest that at concentrations of <1 µM the effects of SB203580 can still be correlated with inhibition of p38 MAP kinase activity, we now submit that at higher concentrations, extreme caution must be exercised in interpreting data. Our observations do not establish a requirement for PDK1 or PKB activity in Rb phosphorylation but suggest a role for these kinases and their distal effectors in Rb regulation. Furthermore, the data do not preclude a role for novel targets of SB203580 in mediating Rb phosphorylation and T cell proliferation.

In summary, we have putatively identified PDK1/PDK2 as targets for the pyridinyl imidazole p38 MAP kinase inhibitor SB203580, which could, at least in part, explain the antiproliferative effect of the drug. The observation that SB203580 can inhibit PI 3-kinase/PDK1/PKB pathway could have major implications for the interpretation of data obtained with this drug.

	ACKNOWLEDGEMENTS

We thank Professors Feldmann, Saklatvala, and Cantrell and Drs. Davies, Dean, and Finch for reading the manuscript and S. Johns for typing the manuscript. We also thank Professor Taniguchi, Dr. K. Tracey, and Drs. Alessi and Stephens for their gifts of reagents.

	FOOTNOTES

^* The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Supported by grants from the BBSRC (UK), Rhone Poulenc-Rorer, The Arthritis and Rheumatism Campaign (UK), and the Wellcome Trust.

^§ To whom all correspondence should be addressed: Kennedy Instititute of Rheumatology, 1 Aspenlea Rd., London W6 8LH, UK. Tel.: 44-208 383 4444; E-mail: b.foxwell@cxwms.ac.uk.

² F. V. Lali, A. E. Hunt, S. J. Turner, and B. M. J. Foxwell, unpublished observations.

³ A. Hunt and B. M. J. Foxwell, unpublished observations.

⁴ L. Hayes, unpublished observations.

	ABBREVIATIONS

The abbreviations used are: IL, interleukin; MAP, mitogen-activated protein; PKB, protein kinase B; PDK, 3-phosphoinositide-dependent protein kinase; Rb, retinoblastoma; PI, phosphatidylinositol; JNK, c-Jun NH₂-terminal kinase; SAPK, stress-activated protein kinase; MOPS, 4-morpholinepropanesulfonic acid; DTT, dithiothreitol; PIP3, phosphatidylinositol 3,4,5-trisphosphate.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES