Interleukin-10 stimulation of phosphatidylinositol 3-kinase and p70 S6 kinase is required for the proliferative but not the antiinflammatory effects of the cytokine.

Interleukin-10 (IL-10) is a powerful suppressor of the proinflammatory monokine production by lipopolysaccharide-stimulated monocytes as well as a T- and B-cell growth cofactor. The signal transduction cascades initiated by IL-10 ligation to its cognate receptor remain to be elucidated. Here, we demonstrate that in both primary monocytes and the D36 cell line, IL-10 rapidly and transiently stimulated phosphatidylinositol 3-kinase activity associated with the p85 subunit of the enzyme. IL-10 also activated p70 S6 kinase in both cell types. The activation of both of these kinases was sensitive to wortmannin, an inhibitor of phosphatidylinositol 3-kinase. The activation of p70 S6 kinase was also inhibited by the immunosuppressive drug rapamycin. Both rapamycin and wortmannin inhibited the IL-10-induced proliferation of D36 cells but in contrast had no effect on the antiinflammatory effects of the cytokine on lipopolysaccharide-stimulated monocytes. Similar results on D36 proliferation and lipopolysaccharide-stimulated monocyte inhibition by IL-10 were obtained with another phosphatidylinositol 3-kinase inhibitor, LY294002. This suggests that the activation of phosphatidylinositol 3-kinase and p70 S6 kinase is involved in the proliferative functions of IL-10 and that other as yet uncharacterized pathways affect the suppressive effects on monocytes, indicating that multiple and distinct signaling pathways mediate the various pleiotropic activities of IL-10. Furthermore, these findings suggest that it may be possible in the future to modulate the antiinflammatory effects of IL-10 for therapeutic benefit without disrupting other functions of the cytokine.

Interleukin-10 (IL-10) is a powerful suppressor of the proinflammatory monokine production by lipopolysaccharide-stimulated monocytes as well as a T-and B-cell growth cofactor. The signal transduction cascades initiated by IL-10 ligation to its cognate receptor remain to be elucidated. Here, we demonstrate that in both primary monocytes and the D36 cell line, IL-10 rapidly and transiently stimulated phosphatidylinositol 3-kinase activity associated with the p85 subunit of the enzyme. IL-10 also activated p70 S6 kinase in both cell types. The activation of both of these kinases was sensitive to wortmannin, an inhibitor of phosphatidylinositol 3-kinase. The activation of p70 S6 kinase was also inhibited by the immunosuppressive drug rapamycin. Both rapamycin and wortmannin inhibited the IL-10-induced proliferation of D36 cells but in contrast had no effect on the antiinflammatory effects of the cytokine on lipopolysaccharide-stimulated monocytes. Similar results on D36 proliferation and lipopolysaccharide-stimulated monocyte inhibition by IL-10 were obtained with another phosphatidylinositol 3-kinase inhibitor, LY294002. This suggests that the activation of phosphatidylinositol 3-kinase and p70 S6 kinase is involved in the proliferative functions of IL-10 and that other as yet uncharacterized pathways affect the suppressive effects on monocytes, indicating that multiple and distinct signaling pathways mediate the various pleiotropic activities of IL-10. Furthermore, these findings suggest that it may be possible in the future to modulate the antiinflammatory effects of IL-10 for therapeutic benefit without disrupting other functions of the cytokine.
Interleukin-10 (IL-10), 1 originally identified as an inhibitor of cytokine synthesis by T-cells (1), was subsequently recognized to have a wide range of additional properties. It has potent antiinflammatory effects, suppressing the production of IL-1, tumor necrosis factor ␣ (TNF-␣), and ␥-interferon in polymorphonuclear cells (2) and monocytes (3) while increasing the release of soluble TNF receptor, a natural inhibitor of TNF-␣ function (4). IL-10 down-regulates major histocompatibility complex class II expression (5) and inhibits the expression of other cell surface markers on macrophages (6). It also acts as a cofactor for mast cell, T-cell, and B-cell growth (7)(8)(9) and prevents apoptosis in IL-2-starved T-cells (10).
IL-10 mediates these activities via a high affinity cell surface receptor structurally related to the interferon receptors (11). However, little is known of the signaling mechanisms initiated by IL-10. Like many cytokines, IL-10 can induce phosphorylation and activation of members of the Janus kinase family, in this case Tyk2 and Jak1 (12) and their effectors, the signal transducer and activator of transcription proteins, STAT1␣ and STAT3 (12,13). IL-10 can activate NF-B in T-cells (14), although inhibition of this transcription factor by this cytokine has been observed in monocytes (15).
The aim of this study was to investigate IL-10-mediated intracellular signaling mechanisms and their relationship to cellular functions of the cytokine. The study shows that IL-10 activates PI 3-kinase and p70 S6 kinase in both primary monocytes and a murine mast cell line, D36. IL-10-mediated proliferation in these cells was sensitive to wortmannin (16 -18) and LY294002 (19), inhibitors of PI 3-kinase, and rapamycin, which inhibits the activation of p70 S6 kinase (20). However, PI 3-kinase and p70 S6 kinase activation was not required for IL-10 suppression of lipopolysaccharide (LPS)-induced TNF-␣ or soluble TNF receptor production. These data indicate a selective role for PI 3-kinase and p70 S6 kinase in IL-10 function and suggest the existence of multiple pathways with different functional end points induced by IL-10.
Isolation of Peripheral Blood Monocytes-Single donor plateletphoresis residues were purchased from the North London Blood Transfusion Service (Colindale, UK). Mononuclear cells were isolated by Ficoll-Hypaque centrifugation (specific density, 1.077 g/ml) preceding monocyte separation in a Beckman JE6 elutriator. Elutriation buffer consisted of RPMI 1640 medium supplemented with 1% fetal bovine serum. Monocyte purity was assessed by flow cytometry using directly conjugated anti-CD45 and anti-CD14 antibodies (Leucogate, Becton Dickinson, UK) and was routinely greater than 90%. All media used in the separation and culture of monocytes were tested for endotoxin using the Limulus amebocyte lysate test (BioWhittaker Inc.) and were rejected if the endotoxin concentration exceeded 0.1 unit/ml.
Incubation Conditions-D36 cells were washed three times and rested for 3 h in cytokine-free medium at 37°C. Before the addition of stimulus, the cells were washed twice more. Monocytes and D36 cells (without IL-3) were suspended at the stated densities in culture medium and stimulated with various concentrations of Chinese hamster ovary cell-derived human recombinant IL-10 (a gift of Dr. K. Moore, DNAX, CA). The IL-10 contained Ͻ0.2 unit/ml of endotoxin. Wortmannin (Sigma) and LY294002 (a gift of Drs. G. Panayotou and R. Stein, Ludwig Institute, London, UK) were dissolved as stock solutions in dimethyl sulfoxide and rapamycin (Calbiochem) in ethanol, respectively; diluted in culture medium; and preincubated with the cells for 15 min at 37°C before the addition of stimuli.
Analysis of Cell Viability-Cell viability was assessed by the ability of the cells to exclude trypan blue and by the amount of merocyanine 540 (Sigma) incorporated into the cell membrane of live gated cells (22) analyzed on a FACStar Flow cytometer (Becton Dickinson, Palo Alto, CA). 50 l of 1 mg/ml merocyanine 540 solution were added to 100 l of cells removed from culture. Before use, the merocyanine was filtered through a 0.2-m pore size filter and stored in the dark at 4°C.
Proliferation Assays-Proliferation assays were performed at 37°C in round-bottomed 96-well plates (Becton Dickinson) using 100-l aliquots of a 2 ϫ 10 5 cells/ml suspension. Stimulus was added to a final volume of 200 l. After 18 h, 1 Ci of [ 3 H]thymidine (Amersham International, Amersham, UK) was added to each sample, and the cells were harvested with an automated harvester (Skatron, UK) after a further 6 h. All results are expressed as the mean of triplicate cultures Ϯ S.D.
PI 3-Kinase Assay-Lipid kinase assays were essentially as described (23), with the following modifications. The reaction was terminated by the addition of 100 l of 1 M HCl. Lipids were extracted by addition of 200 l of chloroform-methanol (1:1). The lower organic layer was washed with 80 l of methanol, 1 M HCl (1:1), and the resultant lipid layer was evaporated at 60°C. Phosphorylated products were separated by TLC on Silica Gel 60 plates (Merck & Co., Inc., Rahway, NJ) developed in chloroform, methanol, water, 30% ammonium hydroxide (180:140:29.2:10.8). Reaction products (i.e. phosphatidylinositol 3-phosphate, PI-3P) were visualized by autoradiography using hyperfilm (Amersham, UK) and quantified by scintillation counting of the excised spots and scanning of the autoradiogram using an imaging densitometer (model GS-670; Bio-Rad, UK). Local subtraction was utilized to adjust for the intensity of each spot, which corrects for the background absorption, averaging the absorption taken 2 pixels wide around each spot, and subtracting it from the integration of the spot itself, using the Molecular Analyst software program (Bio-Rad, UK).
p70 S6 Kinase Assay-Beads containing immunoprecipitates were washed three times in p70 S6 kinase lysis buffer and once in p70 S6 kinase assay buffer (50 mM MOPS (pH 7.2), 1 mM dithiothreitol, 30 M ATP, 5 mM MgCl 2 , 10 mM p-nitrophenylphosphate). The precipitates were resuspended in 9 l of kinase assay buffer containing either 1 l of 40 S ribosome extract (a kind gift of Dr. G. Thomas, Friedrich Miescher Institute, Switzerland) or 1 l of a 250 M solution of substrate peptide (KKRNRTLTK), 1 l of 5 M protein kinase A inhibitor (Santa Cruz Biotechnology Inc.), and 5 Ci of [␥-32 P]ATP. The reactions were performed at 37°C for 30 min, and the products were separated by 15% (40 S) or 20% (peptide) (w/v) acrylamide, 0.07% bisacrylamide gel electrophoresis in the presence of 0.72 M ␤-mercaptoethanol. Products were visualized by autoradiography.
Effect of Wortmannin on IL-10-mediated Inhibition of LPS-induced TNF-␣, IL-1␤, and IL-6 Production in Primary Monocytes-Monocytes were seeded at 1 ϫ 10 6 /ml in triplicate in flat-bottomed plates. Cells were preincubated with wortmannin (0 -1 M) for 15 min prior to stimulation with LPS (10 ng/ml Escherichia coli-derived, Sigma), LPS, and IL-10 (10 ng/ml), IL-10 (10 ng/ml) alone or media alone. Supernatants were removed at 6 and 18 h and TNF-␣ levels were determined by ELISA as described previously (24). The coat monoclonal anti-TNF-␣ antibody (61E71) and the developing polyclonal anti-TNF-␣ antibody were provided by Dr. W. Buurman, Rijks Universiteit Limburg, Maastricht, The Netherlands. An anti-rabbit horseradish peroxidase conjugate (Amersham, UK) was used to detect the polyclonal antibody followed by a homogenic substrate of tetramethylbenzidine dihydrochloride (Sigma). The detection range of the assay was 1.4 -5000 pg/ml. Reagents for the IL-1␤ ELISA were a gift from Drs. J. Kenney and D. Webb (Syntex, CA) The ELISA was performed using the coat anti-IL-1␤ mAb (IL-B1-H6) and detected with anti-IL-1␤ mAb (IL-B1-H67) as described previously (25). The range of the assay was 10 pg/ml to 7.4 ng/ml. Reagents for the IL-6 ELISA were a gift from Dr. F. Padova (Sandoz, Basel, Switzerland), and the ELISA was performed using the coat anti-IL-6 mAb (LN1 314-14) and developed using the detection anti-IL-6 mAb (LN1 101-14), as described previously (25). The range of the assay was 100 pg/ml to 10 ng/ml. All results are expressed as the mean of triplicate cultures Ϯ S.D.

IL-10 Induces the Proliferation of D36 Cells
Which Is Inhibited by Rapamycin, Wortmannin, and LY294002-IL-10 induced [ 3 H]thymidine incorporation in the murine mast cell line D36 in a dose-dependent manner, maximal at 100 ng/ml (Fig.  1a). However, unlike IL-3 (21), IL-10 is unable to support the long-term propagation of these cells. We investigated whether the proliferation of D36 cells was directly attributable to IL-10 or the result of the induction of an autocrine growth factor induced by the cytokine. Thus, fresh D36 cells were cultured with supernatants from IL-10-treated cells in the presence or absence of neutralizing levels of anti-IL-10 antibodies or an isotype-matched control. As shown in Fig. 1b, neutralization of the IL-10 in the supernatants inhibited proliferation by ϳ90% whereas the control antibody had no effect. This suggests that the proliferation is directly attributable to IL-10.
The immunosuppressive drug rapamycin inhibits the IL-10 response of D36 cells (Fig. 2a). The concentrations of rapamycin required (IC 50 Ϸ 0.1 ng/ml) were similar to those used to inhibit the proliferation induced by other cytokines, e.g. IL-2 (26), IL-3 (27), and IL-4 (28). Wortmannin, an inhibitor of PI 3-kinase, which like rapamycin can prevent the activation of p70 S6 kinase (29), also inhibited IL-10-induced proliferation of D36 cells (Fig. 2b) in a dose-dependent manner. Furthermore, LY294002 (2-(4-morpholino)-8-phenyl-4H-benzopyran-4-one), another inhibitor of PI 3-kinase that is structurally unrelated to wortmannin, also inhibited the proliferation of D36 cells induced by IL-10 ( Fig. 2c). At the concentrations tested, none of the drugs had any effect on cell viability (Ͼ95%), as measured by merocyanine or trypan blue exclusion (results not shown), except at 100 M LY294002 where cell viability was 75%. Wortmannin was less effective than rapamycin or LY294002, which may be as a result of the rapid degradation of the wortmannin in cell culture medium, with a half-life of 30 min. 2 Higher concentrations of wortmannin were not used because they may evoke nonspecific effects of the compound.

Interleukin-10 Induces PI 3-Kinase Activity in D36 Cells and
Primary Monocytes-The effect of wortmannin on IL-10 activity led to an investigation of the effects of this cytokine on PI 3-kinase. IL-10 induced an increase in lipid kinase activity associated with immunoprecipitates of the anti-p85␣ subunit of PI 3-kinase from D36 cells (Fig. 3, a and b), and monocytes isolated from human peripheral blood (Fig. 4). Similar results were obtained in the monocytic cell line U937 (data not shown).
No activity was observed in control immunoprecipitates of isotype-matched monoclonal antibodies ( Fig. 3c and data not shown). The kinetic was rapid, was maximal between 2 and 10 min, and was reduced to background by 30 min. Pretreatment of cells with wortmannin totally inhibited kinase activity, confirming the specificity of this activity to PI 3-kinase. The variation in PI 3-kinase activity was not due to differences in amounts of p85␣ present in immunoprecipitates (Figs. 3b and  4b). Densitometry results are expressed as an absorption value of the PI-3P enzymic product alone, which was calculated by adjusting for the local absorption around each spot, as described under "Experimental Procedures." IL-10 activation of PI 3-kinase activity was consistently observed in monocytes prepared from a further nine donors (results not shown), with some variation (-fold activation was between 1.5 and 10.3). 2 L. Stephens, personal communication. ). After cell lysis, the p85␣ subunit was immunoprecipitated (IP), the associated lipid kinase activity was assayed as described under "Experimental Procedures," and the 32 P-labeled lipid product (PI-3P) was separated by TLC. b, densitometric analysis of the autoradiographs of PI-3P spots on the TLC plate and Western immunoblots with anti-p85␣ antibodies of parallel immunoprecipitates from the same lysates used in a. Lipid kinase activity associated with U5 versus isotypematched control immunoprecipitates (Ig) was assayed as above (c). Similar experiments performed with LY294002 were not possible, because unlike wortmannin, this compound does not covalently bind to the enzyme and thus is removed during the assay procedure.
The increase in kinase activity was specific for the p85 ␣ subunit as a comparison of the PI 3-kinase activity in immunoprecipitates of the ␣, ␤, or ␥ isoforms of the p85 subunit showed that IL-10 stimulates the p85␣ pool specifically (Fig. 5).
IL-10 Induces p70 S6 Kinase Activation in Primary Monocytes and D36 Cells-The effect of rapamycin on IL-10-induced proliferation in D36 cells led to an investigation of p70 S6 kinase activity. IL-10 induced the activation of p70 S6 kinase in both D36 cells (Fig. 6a) and monocytes (Fig. 6c), although the kinetics appeared slightly slower in the former (60 min) than the latter (45 min) cell type. p70 S6 kinase activation was not seen in immunoprecipitates using an isotype-matched antibody control (Fig. 6, b and d) and was inhibited by both wortmannin (Fig. 6, a and c) and rapamycin (Fig. 6, b and d) in both cell types. LY294002 was also able to inhibit p70 S6 kinase activation by IL-10 (results not shown).

The Inhibition of LPS-induced Monocyte TNF-␣ Production in Primary Monocytes Is Not Sensitive to the Effects of Wortmannin, LY294002
, and Rapamycin-One marked effect of IL-10 on peripheral blood mononuclear cells is the inhibition of proinflammatory cytokine production. Because we have shown that IL-10 induces activation of both PI 3-kinase and p70 S6 kinase in these cells, the effect of wortmannin, LY294002, and rapamycin on this was investigated. None of these compounds had any effect on the inhibition by IL-10 of LPS-induced TNF-␣ production at 18 h (Fig. 7) or 6 h (data not shown); in addition, IL-10-induced release of soluble TNF receptor was similarly unaffected (data not shown). An intriguing and consistent finding was that wortmannin (Fig. 7a) synergized with LPS to enhance TNF-␣ production by as much as 300% (Fig. 7a). This was consistently seen in monocytes isolated from four separate donors. In contrast, the other PI 3-kinase inhibitor LY294002 suppressed TNF production by ϳ25% (Fig. 7b), whereas rapamycin had no effect (Fig. 7c). One can only postulate that these various effects on TNF production induced by LPS are possibly due to other effects of these drugs. As with D36 cells, none of the drugs exhibited significant toxicity to monocytes (Ն95% viable) as judged by merocyanine and trypan blue staining. DISCUSSION This study has shown that IL-10 can activate PI 3-kinase and p70 S6 kinase. Furthermore, the activation of these factors is required for the proliferative but not the antiinflammatory effects of IL-10, indicating that multiple signaling pathways are activated by this cytokine.
The accumulation of the 3Ј-phosphorylated phospholipids due to the activation of PI 3-kinase correlates with the proliferative response to many growth factors and cytokines (30 -35). However, unlike the receptors for these factors, the IL-10 receptor is related to the receptors for the interferons (11). This paper demonstrates that PI 3-kinase is activated via a receptor of this family, although the delineation of the precise structure of the 3Ј-phosphorylated lipids induced by IL-10 was beyond the limit of this study.
The PI 3-kinase enzyme is composed of two subunits, the catalytic p110 and regulatory p85. Whether the observed increase in PI 3-kinase activity in anti-p85 immunoprecipitates really does represent a bulk activation of PI 3-kinase is unclear. More likely, the observed modest activation of kinase activity would suggest that some subpopulation of the cellular PI 3-kinase pool is being activated. The mechanisms of activation of PI 3-kinase remain largely undetermined, as is the case here.
Previous studies have generally measured factors activating FIG. 4. IL-10 induces PI 3-kinase activity in primary human monocytes. a, freshly elutriated monocytes were unstimulated (un) or treated with 100 ng/ml IL-10 for the given times in the presence of wortmannin (100 nM) where indicated (Wort.). After cell lysis, the p85␣ subunit was immunoprecipitated (IP), the associated lipid kinase activity was assayed as described under "Experimental Procedures," and the 32 P-labeled lipid product (PI-3P) was analyzed by TLC. b, densitometric analysis of the autoradiographs of PI-3P spots on the TLC plate and Western immunoblots with anti-p85␣ antibodies of parallel immunoprecipitates from the same lysates used in a.

FIG. 5. IL-10-induced PI 3-kinase activity in primary human
monocytes is associated only with immunoprecipitates of the p85␣ subunit. a, freshly elutriated monocytes were unstimulated (un) or treated with 100 ng/ml IL-10 for 10 min in the presence of wortmannin (100 nM) where indicated (Wort.). Cell lysates were immunoprecipitated (IP) with antibodies to the ␣, ␤, and ␥ isoforms of p85 and the associated lipid kinase activity assayed as described under "Experimental Procedures." The 32 P-labeled lipid product (PI-3P) was analyzed by TLC. b, densitometric analysis of the autoradiographs of PI-3P spots on the TLC. PI 3-kinase by precipitating the enzyme with anti-phosphotyrosine antibodies or when associated with other signaling molecules (e.g. insulin receptor substrate 1 (36)). In addition, the p85 subunit is often tyrosine phosphorylated in response to cytokine activation (30,31,37,38). However, thus far we have been unable to demonstrate IL-10-induced activity in antiphosphotyrosine or anti-insulin receptor substrate 1 immunoprecipitates (results not shown), although further studies in this area are still in progress.
Although the downstream effects of the 3Ј-labeled PI second messengers are still elusive, in vitro they can activate protein kinase C (39), and the serine/threonine protein kinase Akt (40,41). Furthermore, in platelet-derived growth factor and insulin-stimulated HepG2 cells (29) or in IL-2-stimulated CTLL T-cells (42) and CD28-stimulated T-cells (43,44), PI 3-kinase activity has been correlated with an increase in p70 S6 kinase activity. Our studies showed that IL-10 can also induce p70 S6 kinase activation, and the observations with wortmannin and LY294002 also demonstrate a possible causal link between the activation of PI 3-kinase and p70 S6 kinase, this time by IL-10. However, recent studies (45) on insulin signaling have dissociated the activation of the two enzymes, suggesting that other wortmannin-sensitive targets may exist. Such a hypothesis is supported by the recent observation that the catalytic subunit of DNA dependent kinase, a molecule with a PI 3-kinase-like domain, is also sensitive to wortmannin (46).
As for IL-2-(26), IL-3- (27), and IL-4-induced 3 p70 S6 kinase activation, rapamycin inhibited IL-10 activation of this enzyme with a concomitant inhibition of IL-10-induced proliferation of D36 cells. The intracellular target of rapamycin appears to be a family of molecules identified variously as a target of rapamycin in yeast (47), rapamycin and FKBP targets in rats (48,49), and FKBP12-rapamycin-associated protein in humans (50). FKBP12-rapamycin-associated protein has recently been shown to lie proximal of the activation of p70 S6 kinase (51). The precise role of these molecules in cell physiology is unknown. Although like DNA-dependent kinase they have some structural homology to the kinase domain of PI 3-kinase, FKBP12-rapamycin-associated protein, at least, is not a target for wortmannin (51). A recent study has identified a rapamycin-sensitive step in the induction of the key cell survival factor Bcl-2 by IL-2 (52). Because IL-10 can induce the Bcl-2 protein (53), one could speculate that rapamycin inhibits this function 3 D. Taylor-Fishwick, unpublished data.
FIG. 6. IL-10 induces p70 S6 kinase activity in D36 cells and primary monocytes. Resting D36 cells (a) or fresh primary monocytes (c) were unstimulated (un) or treated with 100 ng/ml IL-10 for the given times in the presence of wortmannin (100 nM) where indicated (Wort.). After cell lysis, p70 S6 kinase was immunoprecipitated, and the associated kinase activity assayed as described under "Experimental Procedures"; the 32 P-labeled 40 S ribosomal subunit was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and autoradiography. Resting D36 cells (b) or primary monocytes (d) were unstimulated (un) or treated with 100 ng/ml IL-10 for the given times in the presence of rapamycin (10 ng/ml) where indicated. After cell lysis, p70 S6 kinase was immunoprecipitated with specific antibodies or a nonspecific isotype-matched control (Ig control), and the associated kinase activity was assayed as described under "Experimental Procedures"; the 32 P-labeled substrate peptide or 40 S ribosome subunit as indicated was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and autoradiography. of IL-10. This is currently under investigation.
The inhibitory effects of wortmannin and rapamycin on IL-10-induced proliferation demonstrated the functional relevance of PI 3-kinase and p70 S6 kinase activation and agree with many previous studies indicating a role for these enzymes in cytokine and growth factor-induced mitogenesis. The possibility that other drug targets are mediating the inhibitory effects of wortmannin is unlikely, since LY294002 had similar effects on IL-10-induced proliferation. Its spectrum of cellular targets other than PI 3-kinase is likely to different, as the data obtained with monocytes suggest. However, until the effects of wortmannin, LY294002, and rapamycin are fully characterized, such a possibility cannot be totally excluded.
In contrast, none of the drugs had any inhibitory effect on two of the key antiinflammatory functions of IL-10 in monocytes, the inhibition of LPS-induced TNF-␣ production and the release of soluble TNF receptor release, indicating that neither PI 3-kinase nor p70 S6 kinase participates in this function of IL-10. Similar results were also obtained when monocyte culture supernatants were assayed for IL-1␤ and IL-6 (data not shown). Interestingly, wortmannin actually synergized with LPS to increase TNF expression by as much as 3-fold. Furthermore, even for shorter assays of TNF production (6 h), wortmannin still had no effect on IL-10 function. This observation would indicate that the lack of effect of wortmannin on IL-10 function in monocytes is not due to the short half-life of the drug but may be due to other effects of the drug. The fact that LY294002 had the opposite effect on TNF-␣ production would support this conclusion. A recent publication has shown that wortmannin can stimulate the stress-activated protein kinase pathway (54), and this may have a positive effect on TNF-␣ production.
In summary, this study shows that IL-10 can activate two key intracellular signaling enzymes, PI 3-kinase and p70 S6 kinase. The effects of wortmannin, LY294002, and rapamycin implicate these molecules in IL-10-induced proliferation in the D36 mast cell line, but not in the antiinflammatory effects of the cytokine in monocytes. These data would indicate that there are multiple signaling mechanisms activated by IL-10, involved in different functions of the cytokine.