in Human Monocytes

Inhibitors of the p38 Mitogen-Activated Kinase Modulate IL-4 Induction of Low Affinity IgE Receptor (CD23) in Human Monocytes Lisa A. Marshall,1* Michael J. Hansbury,* Brian J. Bolognese,* Rebecca J. Gum,† Peter R. Young,† and Ruth J. Mayer* CD23, the low affinity IgE receptor, is up-regulated on the surface of IL-4-treated B cells and monocytes and is immediately proteolytically processed, releasing soluble fragments of CD23. Here, we report that inhibitors of the p38 mitogen-activated kinase (p38 MAPK), SK&F 86002 or the more selective inhibitor, SB 203580, reduce the levels of soluble CD23 formed by IL-4-stimulated human monocytes or the human monocytic cell line, U937. In contrast to compounds such as the metalloprotease inhibitor batimastat ([4-(N-hydroxyamino)-2-(R)-isobutyl-3-(S)-(2-thiophenethiomethyl)succinyl]-(S)-phenylalanine-N-methylamide, sodium salt), p38 MAPK inhibitors do not directly inhibit proteolytic processing of CD23. Further, evaluation of surface intact CD23 (iCD23) by flow cytometry demonstrated that SK&F 86002 and SB 203580 reduced the surface expression of iCD23 in a concentration-dependent fashion, while batimastat increased the surface expression of iCD23. The decrease in surface iCD23 was accompanied by a decrease in total cell-associated CD23 protein levels but not CD23 mRNA. IL-4 induced a late (>4-h) increase in p38 MAPK activity and corresponding activation of its substrate MAPKAPK-2. This activation was blocked by addition of SB 203580 before IL-4 induction, in parallel with the inhibition of CD23 expression. Modulation of CD23 by antibodies has been shown to alleviate the symptoms of murine collagen-induced arthritis, implicating CD23 as an important proinflammatory agent. These data show that in addition to the known cytokine inhibitory actions of SK&F 86002 and SB 203580, they also confer an additional potential anti-inflammatory activity through modulation of CD23 expression. The Journal of Immunology, 1998, 161: 6005– 6013. D23, or FceRII, the low affinity IgE receptor, is involved in the regulation of IgE production and has other functions in the modulation of the immune response (1). The a form of CD23 is constitutively expressed at low levels only on B lymphocytes, while a splice variant b form is induced on B lymphocytes as well as other immune and nonimmune cells (2, 3). B lymphocyte surface intact CD23 (iCD23)2 regulates IgE production through negative feedback inhibition, activated by the presence and the binding of excess IgE (4, 5). The binding of IgE to intact CD23 initiates the feedback down-regulation of IgE and apparently inhibits proteolytic cleavage (6), attenuating the formation of the soluble fragments. Intact 45-kDa CD23 is also proteolytically cleaved from the cell surface to give soluble fragments, sCD23, which have immunostimulatory functions such as support of growth and differentiation of germinal center B cells (7, 8), enhancement of IgE expression (9), and activation of monocytes (10 –12).

C

Departments of *Immunopharmacology and †Molecular Biology, SmithKline Beecham Pharmaceuticals, King of Prussia, PA 19406 Received for publication January 16, 1998. Accepted for publication August 6, 1998. 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 with 18 U.S.C. Section 1734 solely to indicate this fact. 1 Address correspondence and reprint requests to Dr. Lisa Marshall, SmithKline Beecham Pharmaceuticals, 709 Swedeland Rd., King of Prussia, PA 19406. 2 Abbreviations used in this paper: iCD23, surface intact CD23; sCD23, soluble CD23; MAPK, mitogen-activated protein kinase; TCM, RPMI 1640 medium with 2 mM glutamine and 10% fetal bovine serum; MBP, myelin basic protein; hsp27, heat shock protein-27; PLA2, phospholipase A2; batimastat, [4-(N-hydroxyamino)-2(R)-isobutyl-3-(S)-(2-thiophenethiomethyl)succinyl]-(S)-phenylalanine-N-methylamide, sodium salt; CSBP, CSAIDS binding protein; MAPKAPK-2, mitogen-activated protein kinase activated protein kinase-2.

Copyright © 1998 by The American Association of Immunologists

The role of CD23 in monocyte/macrophage function is just beginning to be elucidated. sCD23 is a potent activator of monocyte TNF-a production (11, 12) and has been directly implicated in the stimulation of monocyte NO2 2 (13), IL-1, and IL-6 (14). In vitro studies suggest that sCD23 may act as a proinflammatory mediator through binding to monocyte CD11b-CD18 and CD11c-CD18 adhesion molecules (15). Finally, a key role for CD23 in inflammatory processes was demonstrated in vivo using an anti-murine CD23 mAb. This tool was effective in ameliorating the clinical signs of mouse collagen-induced arthritis (16). Thus, CD23 has an important role in inflammatory processes and could be viewed as an anti-inflammatory target in its own right. Currently, the few known modulators of CD23 include IFN-g and IFN-a or IL-10, which down-regulate CD23 (17, 18); hydrocortisone, which augments sCD23 formation (19); cyclosporin A, which is reported to amplify sCD23 release by preactivated B cells (20); rapamycin, which inhibits sCD23 levels in medium of activated B cells (21); and batimastat ([4-(N-hydroxyamino)-2-(R)-isobutyl-3-(S)-(2thiophenethiomethyl)succinyl]-(S)-phenylalanine-N-methylamide, sodium salt), which inhibits the release of sCD23 from the cell surface (22). The identification and development of anti-inflammatory agents directed toward inflammatory cytokines have received considerable attention (23). A novel class of cytokine-suppressive antiinflammatory compounds has been recently described (24, 25) that inhibit the production of cytokines such as TNF and IL-1 by decreasing their biosynthesis. Compounds such as SK&F 86002, SK&F 105809, and SB 203580 inhibit endotoxin-induced IL-1 and TNF formation in human monocytes and are effective in vivo in promoting the survival of mice undergoing septic shock as well as in alleviating symptoms in animal models of acute and chronic 0022-1767/98/$02.00

6006

p38 INHIBITORS MODULATE IL-4 INDUCTION OF CD23 IN MONOCYTES

inflammation, including collagen-induced mouse arthritis (25, 26). These compounds selectively inhibit two p38 mitogen-activated protein kinases (MAPKs), p38/CSBP and p38b (also known as stress-activated protein kinases 2a and 2b) (27–29). Herein, we demonstrate that SK&F 86002 and SB203580 (28) prevent IL-4-induced monocyte or U937 cell CD23 surface expression and protein formation with no effect on CD23 mRNA levels. This identifies CD23 as another protein that is regulated or activated through the p38 MAPK pathway and describes an additional mechanism by which p38 inhibitors may exert their antiinflammatory pharmacologic effect.

Materials and Methods Compounds Batimastat was synthesized at SmithKline Beecham (King of Prussia, PA) as previously described (22). The p38 inhibitors, SK&F 86002 and SB 203580, were synthesized at SmithKline Beecham by the Department of Medicinal Chemistry as previously described (28, 30).

Cell culture Cells were cultured in RPMI 1640 medium with 2 mM glutamine and 10% FBS, referred to as TCM, with additional components as detailed below. All cells were cultured at 37°C in 5% CO2 atmosphere. RPMI 8866 cells, an EBV-transformed B lymphocyte line that constitutively expresses a and b forms of CD23, were obtained from the European Collection of Animal Cell Cultures (Porton Down, U.K.), and U937 and Ramos cells were obtained from American Type Culture Collection (Manassas, VA).

Soluble CD23 release from IL-4-induced human monocytes or U937 cells Monocytes were isolated from human blood using two gradient procedures, resulting in 85–90% purified populations as previously described (31), and were adhered in BD 24-well culture plates at a concentration of 1 3 106 cells/ml in TCM for 2 h at 37°C. After this time, the supernatant was removed, and fresh medium was added to the wells. Monocytes or U937 cells were exposed to 200 ng/ml of IL-4 (added as 10 ml of a 20 ng/ml stock solution made up in TCM) for up to 72 h to induce CD23 expression, accumulation onto the cell surface, and subsequent shedding into the medium (17, 32). Test compounds were added in triplicate (the vehicle DMSO concentration did not exceed 0.1%) together with IL-4 at the beginning of the incubation. Studies were terminated by removal of medium, and sCD23 levels were determined by ELISA using the commercially available kit from The Binding Site (Birmingham, U.K.) modified by use of the antiCD23 mAb, MHM6, as the coating mAb.

RPMI 8866 membrane and cell assays for CD23 RPMI 8866 cells were grown in TCM with 2 mM glutamine and 50 mg/ml gentamicin. Extracted plasma membranes were prepared as previously described (33, 34), and inhibition of sCD23 release from the purified RPMI 8866 cell membranes was measured (22). In brief, extracted membranes were incubated for 1 h in the presence or the absence of compound with a final DMSO concentration of 2%. Assays were quenched with 5 mM batimastat, membranes were removed by filtration, and sCD23 in the supernatant was determined by ELISA. For inhibition of sCD23 release from intact cells, test compounds were dissolved in DMSO at 10 mM, then diluted sequentially in TCM to reach the desired concentration. RPMI 8866 cells were incubated with vehicle or compound for 1 h at 37°C. Assays were quenched by addition of batimastat (30 mM) as previously described (22), and sCD23 was measured in cell-free medium by ELISA. Compound toxicity was monitored by trypan blue exclusion.

Flow cytometric analysis of iCD23 Membrane-bound intact iCD23 was determined by immunofluorescence analysis conducted by flow cytometry using a FACScan (Becton Dickinson, Mountain View, CA). Cells (106/ml) were fixed with 1% paraformaldehyde, followed by incubation with either anti-CD23-FITC Ab or an IgG1 isotype control (Southern Biotechnology Associates, Birmingham, AL) at the recommended dilution on ice for 30 min, followed by standard azide (0.1%) and paraformaldehyde (2%) washes (35).

Western analysis U937 cells (5 3 107) treated with IL-4 and with or without drugs were resuspended in homogenization buffer and disrupted by N2 cavitation as

previously described (36). Total homogenates (100 mg protein/lane) were added to loading buffer, applied to 10% SDS-PAGE (Bio-Rad, Hercules, CA) gels, and blotted to polyvinylidene difluoride. CD23 was detected using the anti-CD23 polyclonal Ab (1/1000 dilution) from The Binding Site CD23 ELISA kit (Birmingham, U.K.) and the conjugate provided by the Binding Site using standard methods. Similarly, total homogenates were analyzed for the 85-kDa phospholipase A2 as previously described (36). Immunoreactive bands were detected using the ECL Western blotting system (Amersham, Aylesbury, U.K.) and quantitated using a Molecular Dynamics densitometry system (Sunnyvale, CA).

Northern analysis Total RNA was isolated from treated or untreated U937 cells using Trizol reagent (Life Technologies, Gaithersburg, MD) according to the manufacturer’s protocol. Equal amounts of total RNA (30 mg/lane) were subjected to electrophoresis in a 1% agarose gel containing formaldehyde, RNA was transferred to Hybond N1 (Amersham) by vacuum blotting (Bio-Rad, Hercules, CA) in 103 SSC according to the manufacturer’s protocol, and fixed by UV cross-linking (0.12 J/cm2). RNA was visualized by staining with 0.02% methylene blue (5 min) and destained in distilled water (15 min), and equal loading per lane of RNA was verified by quantitation of 18S and 28S ribosomal RNA bands. Hybridizations, following standard prehybridization treatment, were performed in prehybridization solution (10 ml/blot) containing 20 ng of denatured specific DNA probe labeled to 1–2 3 109 dpm/mg with 32P (see below) at 68°C for 18 h. Following hybridization, blots were washed twice with 23 SSC/0.1% SDS at 68°C, once with 13 SSC/0.1% SDS at 68°C, and once with 0.23 SSC/0.1% SDS at 68°C. Filters were analyzed and quantitated on a PhosphorImager (Molecular Dynamics). The CD23 probe, a 966-nucleotide full-length cDNA, was labeled using a Rediprime kit and 50 mCi of [32P]dCTP (Amersham, Arlington Heights, IL). Unincorporated nucleotides were removed using Quick Spin columns (Boehringer Mannheim, Indianapolis, IN).

p38/CSBP and MAPKAPK-2 activity determination U937 cells were harvested and lysed, and CSBP2 or its downstream substrate was immunoprecipitated as described previously (37) using antiCSBP2 Ab (24) or MAPKAPK-2 Ab (Upstate Biotechnology, Lake Placid, NY). Half of the immunoprecipitated extract was used for kinase assays. Assays consisted of 30 ml of kinase buffer containing 50 mM [g-32P]ATP (4500 Ci/mmol; 2 mCi/assay; ICN, Costa Mesa, CA) and 10 mg of myelin basic protein (MBP; Life Technologies) or 3.6 mg of hsp27 (StressGen Biotechnologies, Victoria, Canada) as substrate and were incubated for 30 min at 30°C. The reactions were stopped by the addition of SDS sample buffer, and the phosphorylated products were analyzed by SDS-PAGE and autoradiography. The phosphorylated products were quantitated on a Betagen Betascope (Waltham, MA). For Western blots, the other half of the immunoprecipitated extracts was subjected to SDS-PAGE, transferred to nitrocellulose, incubated with either anti-phospho-specific p38 MAPK (Thr180/Tyr182) Ab (New England Biolabs, Beverley, MA) or anti-CSBP2 Ab, and detected by ECL (Amersham) as described by the manufacturer.

Data analysis Data are represented as the mean of triplicate or quadruplicate samples as indicated and represent at least two experiments conducted with two donors where applicable. Data were evaluated using analysis of variation and Duncan’s multiple range test. IC50 values were calculated using nonlinear least squares fits to the meaned data using the curve-fitting utility in the program Kaleidagraph.

Results Assessment of inhibitor effects on IL-4-induced sCD23 release in monocyte or U937 cells Human monocytes are reported to express CD23 upon stimulation by IL-4. Fig. 1A shows a time course of sCD23 accumulation in the medium of human monocytes from one donor treated with IL-4 (200 ng/ml) over 72 h. sCD23 levels, resulting from induced expression of iCD23 and its subsequent cleavage, increased over 24 – 48 h and then begin to plateau over 72 h. Fig. 1B shows the time course of sCD23 accumulation in the medium of U937 treated with IL-4 over 72 h. The sCD23 levels, corresponding to the expression of surface iCD23 and its subsequent cleavage, increased over 48 –72 h. In some cases, at 48 h, medium was removed, and

The Journal of Immunology

6007

FIGURE 2. Effect of inhibitors on sCD23 accumulation over 48 h by IL-4-treated human monocytes. Monocytes were incubated with IL-4 for 36 h in the presence of varying concentrations of SK&F 86002 or SB 203580. Cell-free medium was assessed for sCD23 by ELISA as described in Materials and Methods. Data represent the mean 6 SD nanograms of CD23 per milliliter (n 5 3).

FIGURE 1. Time course of IL-4-induced human monocyte or U937 cell sCD23 accumulation in culture medium. Human monocytes (5 3 106/ml) or U937 cells (1 3 106/ml) were cultured alone or exposed to IL-4 (200 ng/ml) over 72 h. Cell-free medium collected at each time point was tested for sCD23 by ELISA as described in Materials and Methods. Data represent the mean 6 SD nanograms of CD23 per milliliter (n 5 3).

cells were washed and then re-exposed to IL-4-containing medium (200 ng/ml) for an additional 24 h. The initial period following IL-4 activation, is termed the induction phase, while the second period is denoted the cleavage phase, during which the primary events resulting from stimulation with IL-4 have already occurred. The effects of p38 inhibitors on CD23 expression and release were first evaluated in the monocyte IL-4-induced CD23 system. Human monocytes were treated with IL-4 (200 ng/ml) and varying concentrations (0.1, 1, and 10 mM) of SK&F 86002 or SB 203850, for 72 h. Fig. 2 shows one representation of two studies in which both compounds reduced sCD23 in the medium during the induction phase in a concentration-dependent manner. The IC50 concentrations for SK&F 86002 and SB 203580 (IC50, 4.5 and 0.067 mM, respectively) were consistent with their respective LPS-induced cytokine inhibitory actions (IC50, 1.3 mM (30) and 50 –100 nM (28), respectively). More detailed evaluation of the mechanism of action of SK&F 86002 and SB 203580 was conducted in the human monocyte line U937. Table I shows that the p38 inhibitors effectively reduced the levels of sCD23 in the medium of IL-4-stimulated U937 cells incubated with drug and IL-4 for 72 h. To determine the time course of inhibition of CD23 release, SB 203580 or batimastat was added to U937 cells at varying times following stimulation with IL-4. Fig. 3A shows that SB 203580 was effective when added at times up to 4 h following IL-4, but not when added at 20 h or longer after stimulation. In contrast, batimastat (Fig. 3B) inhibited the cleavage of iCD23 at all time points of addition post-IL-4 stimulation. These results suggested that p38 inhibitors have little or no effect on the processing of the iCD23 into soluble fragments, unlike batimastat.

To verify that the p38 inhibitors were not affecting cleavage, SK&F 86002 and SB203580 were compared with batimastat for their ability to alter sCD23 levels generated in 1 h by either RPMI 8866 cell membranes or intact cells (Tables II and III). These EBV-transformed cells constitutively express CD23, so that any observed effect on sCD23 expression would suggest a role of p38 inhibitors outside of the IL-4 signaling pathway. In both RPMI 8866 cell assays there is a lack of direct effect on iCD23 cleavage by either SK&F 86002 or SB 203580, while batimastat potently inhibited the release of sCD23 (34). The effect of SB 203580 on sCD23 accumulation was confirmed in Ramos cells, a B lymphoma line that up-regulates CD23 in response to IL-4 stimulation, using methods identical with those described above for U937 cells. SB 203580 was effective in reducing IL-4-stimulated sCD23 accumulation at 72 h (n 5 3; CD23 (mean 6 SD): IL-4 alone, 22.7 6 1.3 ng/ml; IL-4 and 10 mM SB 203580, 7.8 6 0.4 ng/ml). Evaluation of iCD23 by flow cytometry Human monocytes or U937 cells were exposed to IL-4 (200 ng/ml) for 72 h with vehicle alone, SK&F 86002 (0.1–10 mM), or SB 203580 (10 mM). The incubation was terminated by pelleting the cells and then processing them for flow cytometry and detection of Table I. Effect of SK&F 86002 and SB 203580 on sCD23 levels generated by IL-4-treated U937 cells over 72 h

Treatment

ng sCD23/ml Mean 6 SD (n 5 3)

% Change Compared to IL-4Treated Control

Control SK&F 86002 (10 mM) SK&F 86002 (1 mM) SK&F 86002 (0.1mM) SB 203580 (10 mM) SB 203580 (1 mM) SB 203580 (0.1mM)

19.5 6 0.6 5.2 6 0.8 12.1 6 0.2 14.2 6 0.5 3.7 6 0.6 8.7 6 0.2 13.5 6 0.42

273a 238a 227a 281a 255a 230a

a

Significantly different from control (P , 0.5).

6008

p38 INHIBITORS MODULATE IL-4 INDUCTION OF CD23 IN MONOCYTES Table III. Effect of batimastat, SK&F 86002, and SB 203580 on sCD23 release from RPMI 8866 cells

Treatment

ng sCD23/ml, Mean 6 SD (n 5 3)

% Change Compared to IL-4Treated Control

Control SK&F 86002 (10 mM) SK&F 86002 (1 mM) SK&F 86002 (0.1 mM) SB 203580 (10 mM) SB 203580 (1 mM) SB 203580 (0.1 mM) Batimastat (0.175 mM)

4.4 3.4 6 0.09 3.6 6 0.15 3.7 6 0.13 3.2 6 0.15 4.0 6 0.24 3.8 6 0.12 0.98 6 0.18

222 220 216 228a 210 215 278a

a

FIGURE 3. Dependence on time of addition of SB 203580 on inhibition of sCD23 release. U937 cells (106/ml) were incubated with IL-4 (200 ng/ ml) and SB 203580 (upper; 10 mM) or batimastat (lower; 0.15 mM) added at the times indicated; no measurement was taken at 20 h for batimastat. The level of sCD23 in the medium was determined at 72 h following IL-4 stimulation. Data represent the mean 6 SD (n 5 3). p, Significant difference from IL-4 only control (p , 0.05).

surface iCD23. Fig. 4, A and B, shows one representative histogram of two separate studies showing that exposure of monocytes to IL-4 induced a significant shift of the peak to the right, indicating the induction of CD23-expressing cells. Fig. 4, D–F, demonstrates that treatment with increasing concentrations of SK&F 86002 caused a concentration-dependent shift of the iCD23-expressing cell peak to the left, indicating a reduction in the surface expression of iCD23. Similarly, SB 203580 at 10 mM (Fig. 4C) significantly reduced the number of iCD23-expressing cells by 84% compared with that of the nontreated IL-4-stimulated cells. Fig. 5 illustrates the identical response observed in the U937 monocytic cell line when exposed to SB 203580 or SK&F 86002. IL-4 induced the expression of iCD23 on the surface of the U937 cells (Fig. 5, A and B, one representation of two or three studies). Treatment with 10 mM SK&F 86002 or SB 203580 prevented the IL-4-induced shift to the right and the maintenance of non-CD23expressing cells. In contrast, in a separate study, batimastat (Fig. 5, E, no IL-4; F, with IL-4; and G, with 10 mM batimastat) enhanced the iCD23-expressing cell population by preventing the cleavage and sustaining the iCD23 levels on the surface as has been previously reported (22). These data show that SK&F 86002 and SB 203580 reduce the release of sCD23 by reducing the level of CD23 expressed on the surface of IL-4-stimulated cells. Western and Northern analyses of CD23 expression in inhibitortreated U937 cells The effect of p38 inhibitors on total cellular CD23 and CD23 mRNA was also evaluated in U937 cells. U937 cells (1 3 106/ml) Table II. Effect of SB 203580 and SK&F 86002 on release of sCD23 from purified RPMI 8866 cell membranes

Compound

ng sCD23/ml Mean 6 SD (n 5 3)

% Change

No addition SB 203580 (10 mM) SK&F 86002 (10 mM) Batimastat (0.1 mM)

25.8 6 5.3 22.2 6 0.9 23.1 6 0.4 2.3 6 2.8

214 210 291a

a

Significantly different from control (p , 0.5).

Significantly different from control (p , 0.5).

were treated with IL-4 and vehicle, batimastat (10 mM), SK&F 86002 (10 mM), or SB203580 (10 mM) for 72 h. Cells were pelleted by centrifugation, and total homogenates were generated as described in Materials and Methods. Western analysis of samples from the varying treatments was performed at an equal protein concentration (100 mg) to assess CD23 protein levels as described above. Fig. 6A is one representation of two separate studies showing that cells not stimulated with IL-4 (lane 1; pixel value, 81) had low levels of CD23-immunoreactive protein migrating at approximately 45 kDa, which was significantly enhanced with IL-4 treatment (lane 2; pixel value, 227). Batimastat did not reduce CD23 formation and appeared to enhance protein levels (lane 3; pixel value, 335), while both SK&F 86002 (lane 4; pixel value, 189) and SB 203580 (lane 5; pixel value, 121) significantly reduced CD23 levels by 25 and 72% (compared with the IL-4-induced control minus the basal level), respectively. To show that this was not a nonspecific effect on protein formation, 85-kDa PLA2, which is known not to be affected at the level of expression by p38 inhibitors (J. Lee, SmithKline Beecham, unpublished observations) (38) was monitored in these same samples by Western analysis. Fig. 6B shows that 85-kDa PLA2 is found in U937 as a protein migrating at approximately 100 kDa as previously described (39) (lane 1). IL-4 reduced 85-kDa PLA2 consistent with previous reports (lane 2) (40), but neither the p38 inhibitors (lane 3 and 4) nor batimastat (lane 5) further altered protein levels. U937 cells treated identically to those described above were used to purify total RNA for Northern analysis of CD23 mRNA. Fig. 7A shows that CD23 mRNA expression is apparent in cells not stimulated with IL-4 (lane 1) and enhanced with IL-4 stimulation (lane 2). IL-4-induced CD23 mRNA levels were not altered by either p38 inhibitors or batimastat in two separate experiments, as shown by the quantitative results given in Fig. 7B. p38/CSBP activity To verify that p38/CSBP2 activity was inhibited under the conditions used to evaluate CD23 expression, p38/CSBP2 or its physiologic substrate MAPKAPK-2 was immunoprecipitated from U937 cells treated with IL-4 (200 ng/ml) for varying amounts of time (Fig. 8). The viability of the U937 cells under all conditions and times of treatment, as assessed by trypan blue exclusion, was 88 –93%, so that any changes noted could not be attributed to differences in cell viability. The activities of p38/CSBP2 and MAPKAPK-2 from the immunoprecipitates were determined by phosphorylation of MBP and hsp27, respectively (37). Fig. 7A shows that there is a constitutive level of activity of CSBP2 in untreated cells that increased about twofold within 24 h after IL-4 stimulation. The change in activity was consistent with the increased level of phosphorylated, or activated, CSBP2 (Fig. 7B) in


6009

FIGURE 4. Flow cytometric analysis of sCD23 on human monocytes treated with IL-4 and p38 inhibitors. Human monocytes were exposed to IL-4 (200 ng/ml) for 72 h with vehicle alone, SK&F 86002 (0.1, 1.0, and 10 mM), or SB 203580 (10 mM) as indicated. Mean fluorescence intensities were (A–F, respectively) 16, 436, 119, 497, 309, and 175.

a constant amount of total CSBP2 (Fig. 7C). The constitutive activation in the absence of IL-4 of CSBP2 was also consistent with the activity of MAPKAPK-2 from unstimulated cells (Fig. 7D, lane 1) grown for 24 h. Phosphorylation of hsp27 by MAPKAPK-2 increased twofold after 24 h of IL-4 stimulation (Fig. 7D, lane 3), consistent with the activation of p38/CSBP shown in Fig. 7A. Both the constitutive and stimulated activities of MAPKAPK-2 were completely inhibited by 10 mM SB 203580 (Fig. 7D, lanes 2 and 4), suggesting that phosphorylation/dephosphorylation of MAPKAPK-2 was occurring during the 24-h period. Together these data demonstrate that p38/CSPB2 is activated in 24 h by IL-4 stimulation and that further downstream activation mediated by p38 is inhibited by addition of SB 203580.

Discussion The two p38 inhibitors used in these studies are SK&F 86002, a first generation compound, and SB 203580, a p38 inhibitor that has been shown to be remarkably selective with respect to at least 30 other kinases evaluated (28). It is possible that SB 203580 could also inhibit other, as yet unidentified, kinases. These compounds and other p38 inhibitors are potent inhibitors of proinflammatory cytokine formation and cyclo-oxygenase induction by stress, inflammatory cytokines, and endotoxin (24) and are effective agents in animal models of inflammation and arthritis (26). Elevated

sCD23 levels have been associated with a number of conditions mediated by immunoregulating or proinflammatory processes (16, 41, 42), and sCD23 has been reported to induce monocyte TNF-a formation, cell adhesion, and migration (15, 43). We show here that the p38 inhibitors, SK&F 86002 and SB 203580, in addition to their cytokine modulatory properties also negatively regulate IL-4-induced expression of monocyte CD23. IL-4 stimulation of human monocytes or the human monocyte cell line U937 induces expression of CD23 (32), which is ultimately brought to the cell surface and proteolytically cleaved to form soluble fragments, sCD23. The time course of release of sCD23 suggests that induction of CD23 occurs by 24 h, but that the majority of sCD23 accumulates in the culture medium from 36 –72 h. The sCD23 measured in the medium of IL-4-stimulated monocytes or U937 cells exposed to IL-4 for 72 h was reduced by the p38 inhibitors at concentrations consistent with p38 inhibition (28, 30), with no effect on cell viability. Inhibition of sCD23 accumulation was observed when compound was added within the first 24 h of incubation, but not when it was added at later times. In contrast, batimastat, which directly inhibits the processing of CD23 from the cell surface, inhibits the release of CD23 when added at any time during the incubation, indicating a different mode of action for the p38 inhibitors. This difference was confirmed by the lack of effect of the p38 inhibitors on CD23 cleavage

6010

p38 INHIBITORS MODULATE IL-4 INDUCTION OF CD23 IN MONOCYTES

FIGURE 5. Flow cytometric analysis of sCD23 on U937 cells treated with IL-4 and p38 inhibitors or batimastat. U937 cells were exposed to IL-4 (200 ng/ml) for 72 h with vehicle alone, SK&F 86002 (10 mM), or SB 203580 (10 mM). In a separate study cells were treated with IL-4 and exposed to vehicle or batimastat (10 mM). Mean fluorescence intensities (A–G, respectively) were 12, 76, 28, 29, 32, 53, and 183.

from RPMI 8866 membranes or whole cells in a 1-h assay. SB 203580 did, however, reduce the sCD23 levels in medium from Ramos B lymphoma cells, where CD23 is up-regulated in response to IL-4, suggesting that the effects of p38 inhibition on IL-4 signaling are similar in monocytes and B cells. Flow cytometry demonstrated that SK&F 86002 and SB 203580 reduced iCD23 surface expression induced by IL-4, in direct contrast to the previous observation that batimastat enhanced the number of CD23-expressing cells through inhibition of iCD23 processing (22). Western analysis of p38 inhibitor-treated IL-4-stimulated

U937 cells confirmed that the inhibitors reduced the total cellular CD23 protein, indicating that the reduction in surface expression was not a result of altered protein processing or translocation to the cell surface. In addition, there was no evidence for an increased rate of internalization, as there was no increase in the amount of 16-kDa form of CD23, which results from internalization (data not shown) (44). The p38 inhibitors appear not to directly affect the proteolytic processing of CD23, but instead modulate an event occurring within the first 24 h of incubation, involving either transcriptional or translational regulation. As both SK&F 86002 and


FIGURE 6. Effect of p38 inhibitors or batimastat on IL-4-treated U937 cells CD23 or 85-kDa PLA2 total cellular protein levels. U937 cells (1 3 106/ml) were treated with vehicle, batimastat (10 mM), SK&F 86002 (10 mM), or SB203580 (10 mM) in the presence of vehicle or IL-4 for 72 h. Western analysis of total cell homogenates from the varying treatments was performed at equal protein concentration (100 mg) as described in Materials and Methods. A shows levels of anti-CD23 Ab-immunoreactive protein levels (lane 1, no IL-4 treatment; lane 2, IL-4-treated cells only; lane 3, IL-4-treated cells exposed to batimastat; lane 4, IL-4-treated cells exposed to SK&F 86002; lane 5, IL-4-treated cells exposed to SB 203580) from one representative of two studies. B shows levels of anti-85-kDa PLA2 Ab-immunoreactive protein levels (lane 1, no IL-4 treatment; lane 2, IL-4-treated cells only; lane 3, IL-4-treated cells exposed to batimastat; lane 4, IL-4-treated cells exposed to SK&F 86002; lane 5, IL-4-treated cells exposed to SB 203580).

the much more specific analogue, SB 203580, prevented IL-4-induced CD23 protein formation at concentrations similar to those needed to reduce IL-1 or TNF formation, these data support the identification of CD23 as another protein that is regulated in part through the p38 kinase pathway. The time course of effectiveness of SB 203580 in inhibition of sCD23 accumulation offers some insights into the role of p38 in CD23 expression compared with other roles of p38. Stimulation by TNF-a, IL-1, and other cytokines activates p38 and the downstream kinases within 30 min (45). In contrast, IL-4 signaling has been previously reported to not rapidly activate p38, in contrast to other cytokines (46), but activation at times longer than 1 h after stimulation was not investigated. We observed IL-4 activation of p38 in U937 cells between 4 and 24 h, which was consistent with the lack of effectiveness of SB 203580 in reducing sCD23 when added more than 4 h following IL-4 stimulation. Inhibition of the p38 pathway was confirmed by demonstrating that activation of MAPKAPK-2 by p38 in U937 cells was blocked by SB 203580. Similar extended time courses for effectiveness of SB 203580 have been previously observed in activation of the HIV-1 long terminal repeat by UV stress (47) and in the induction of nitric oxide synthase in bovine chondrocytes following IL-1 stimulation (48). In these cases p38 itself is rapidly activated, and activity is sustained over several hours. These results together suggest that a downstream substrate may not become available until a later time, which once activated retains its phosphorylation state and activity. The delayed time of p38 activation observed in U937 cells stimulated with IL-4 could also suggest that the p38 pathway may not be directly activated by IL-4R engagement. Instead, an autocrine factor could be generated by IL-4 stimulation, which, in turn, leads to activation of the p38 pathway. Once p38 and its downstream effectors have been activated, SB 203580 is no longer effective, suggesting that the effect on CD23 translation is long-lived. In addi-

6011

FIGURE 7. A, Effects of inhibitors on IL-4-treated U937 cell CD23 mRNA expression. U937 cells (1 3 106/ml) were treated with vehicle, batimastat (10 mM), SK&F 86002 (10 mM), or SB203580 (10 mM) in the presence of vehicle or IL-4 for 72 h. Northern analysis of RNA from the varying treatments was performed as described in Materials and Methods. The data show one representative Northern blot from two different experiments (lane 1, no IL-4 treatment; lane 2, IL-4-treated cells only; lane 3, IL-4-treated cells exposed to batimastat; lane 4, IL-4-treated cells exposed to SK&F 86002; lane 5, IL-4-treated cells exposed to SB 203580). B, Northern blots were quantitated on a Molecular Dynamics PhosphorImager, and results were expressed relative to the intensity of the IL-4stimulated band (100%).

tion, it was observed in some experiments that there was a constitutive level of p38 activity in unstimulated cells, which increased further following IL-4 stimulation. Interestingly, U937 cells also showed a constitutive level of CD23 production as has been previously reported (49), which was not inhibited by addition of SB 203580 at the time of IL-4 stimulation. This again suggests that the downstream effect of p38 activation is effective for a long period and further supports a role of p38 in CD23 regulation. The mechanism of action of p38 inhibitors in CD23 regulation was further evaluated by Northern analysis. The increase in CD23 expression upon IL-4 stimulation was relatively small compared with the basal level, consistent with a similar increase previously reported after phorbol ester stimulation (49). Neither p38 inhibitor had any effect on mRNA levels, demonstrating that the signaling pathways directly induced by IL-4, leading to transcriptional upregulation of CD23 through STAT transcription factors (50), are not affected by the p38 inhibitors. Instead, our data suggest that p38 inhibitors modulate events induced by IL-4 that occur downstream of transcriptional activation, leading to inhibition of CD23 translation, in keeping with p38 inhibitor action on other target proteins in monocytes, including IL-1 or TNF-a. Translational regulation of TNF-a by p38 has been shown to be mediated through the AU-rich element in the 39 untranslated region of mRNA that is implicated in mRNA stability and association with proteins 37– 40K (51). Although the exact mode of action is still not clear, it is thought that association of the AUUUA-binding proteins prevent translation. It has been hypothesized that the p38

6012

p38 INHIBITORS MODULATE IL-4 INDUCTION OF CD23 IN MONOCYTES observation adds another proinflammatory protein to the list of cytokines regulated through the p38 kinase pathway and provides additional rationale for the potent anti-inflammatory action of the p38 inhibitors. Further studies are needed to elucidate the exact mechanism of p38 kinase involvement in CD23 induction.

References

FIGURE 8. Effects of IL-4 on p38/CSBP2 activation and MAPKAPK-2 activation in U937 cells. A–C, U937 cells were treated with IL-4 (200 ng/ml) for increasing times up to 24 h. Cells were harvested, lysed, and immunoprecipitated with anti-CSBP2 Ab. Half the immunoprecipitated extract was used for kinase assays (A), which were performed as described in Materials and Methods using MBP as a substrate. The other half was used for two Western blots. B shows a Western blot using anti-phospho-p38 Ab to detect Thr180/Tyr182 phosphorylated (activated) forms of CSBP2. C shows a Western blot using anti-CSBP2 Ab to demonstrate that the CSBP2 protein levels remained constant. In D, U937 cells were treated with or without IL-4 and with either vehicle or SB203580 (10 mM). After 24 h cells were harvested, lysed, and immunoprecipitated with antiMAPKAPK-2 Ab. Kinase assays were performed using hsp27 as a substrate.

MAP kinase pathway may modulate the activity of AUUUA-binding proteins by causing their release and allowing translation. The p38 inhibitors would then interfere with this regulatory process and prevent translation. Analysis of the 39 untranslated region of the CD23 cDNA (GenBank assession no. M14766) indicates that no obvious AU-rich element exists, so that regulation of CD23 translation by p38 must occur through another mechanism or site of action. Numerous downstream substrates of p38 have been identified. A recent report shows that Mnk1, a kinase similar to the p38 substrate MAPKAPK-2, is activated by p38 and that activation is inhibited by SB 203580 (52). Mnk1, in turn, appears to phosphorylate in vitro a translation initiation factor, eIF4E, providing one possible direct link of p38 to translational regulation. Mnk1 phosphorylation would then be one candidate for a downstream event in the p38 MAP kinase pathway involved in IL-4-induced CD23 up-regulation in monocytic cell lines. Recent reports indicate that CD23 may play a significant role in proinflammatory processes (14, 16); its modulation would therefore be an appropriate target for the development of novel antiinflammatory therapeutics. We have demonstrated here that the p38 inhibitors down-regulate CD23 formation in monocytes. This

1. Bonnefoy, J. Y., S. Lecoanet-Henchoz, J.-P. Aubry, J.-F. Gauchet, and P. Graber. 1995. CD23 and B-cell activation. Curr. Opin. Immunol. 7:355. 2. Yokota, A., H. Kikutani, R. Sato, E. L. Barsumian, M. Suemura, and T. Kishimoto. 1988. Two species of human Fce receptor II (FceRII/CD23): tissue specific and IL-4-specific regulation of gene expression. Cell 55:611. 3. Sutton, B. J., and H. J. Gould. 1993. The human IgE network. Nature 366:421. 4. Luo, H., H. Hofstetter, J. Banchereau, and G. Delespesse. 1991. Cross-linking of CD23 antigen by its natural ligand (IgE) or by anti-CD23 antibody prevent B lymphocyte proliferation and differentiation. J. Immunol. 146:2122. 5. Ludin, C., H. Hofstetter, M. Sarfati, C. Levy, U. Suter, D. Alaimo, E. Kilchherr, H. Frost, and G. Delespesse. 1987. Cloning and expression of the cDNA coding for a human lymphocyte IgE receptor. EMBO J. 6:109. 6. Conrad, D. H., K. A. Campbell, W. C. Bartlett, C. M. Squire, and S. E. Dierks. 1994. Structure and function of the low affinity IgE receptor. Adv. Exp. Med. Biol. 347:17. 7. Barnes, P. J. 1991. Biochemistry of asthma. Trends Biochem. Sci. 16:365. 8. Liu, Y., J. A. Cairns, M. J. Holder, S. D. Abbot, K. U. Jansen, J. Bonnefoy, J. Gordon, and I. C. M. MacLennan. 1991. Recombinant 25-kDa CD23 and interleukin-1a promote the survival of germinal center B cells: evidence for bifurcation in the development of centrocytes rescued from apoptosis. Eur. J. Immunol. 21:1107. 9. Pene, J., I. Chretien, F. Rousset, J. Y. Bonnefoy, and J. E. de Vries. 1989. Modulation of IL-4-induced human IgE production in vitro by IFN-g and IL-5: the role of soluble CD23 (s-CD23). J. Cell. Biochem. 39:253. 10. Sarfati, M., B. Bettler, M. Letellier, S. Fournier, M. Rubio-Trujillo, H. Hofstetter, and G. Delespesse. 1992. Native and recombinant soluble CD23 fragments with IgE suppressive activity. Immunology 76:662. 11. Armant, M., H. Ishihara, M. Rubio, G. Delespesse, and M. Sarfati. 1994. Regulation of cytokine production by soluble CD23: costimulation of interferon g secretion and triggering of tumor necrosis factor a release. J. Exp. Med. 180: 1005. 12. Armant, M., M. Rubio, G. Delespesse, and M. Sarfati. 1995. Soluble CD23 directly activates monocytes to contribute to the antigen-independent stimulation of resting T cells. J. Immunol. 155:4868. 13. Fujiwara, H., H. Kikutani, S. Suematsu, T. Naka, K. Yoshida, T. Tanaka, M. Suemura, N. Matsumoto, S. Kojima, T. Kishimoto, et al. 1996. The absence of IgE antibody-mediated augmentation of immune responses in CD23-deficient mice. Proc. Natl. Acad. Sci. USA 91:6835. 14. Bonnefoy, J. Y., C. Plater-Zyberk, S. Lecoanet-Henchoz, J. F. Gauchet, J. P. Aubry, and P. Graber. 1996. A new role for CD23 in inflammation. Immunol. Today 17:418. 15. Lecoanet-Henchoz, S., J. Aubry, P. Graber, P. Life, N. Paul-Eugene, B. Ferrua, C. Plater-Zuberk, and J. Bonnefoy. 1995. CD23 regulates monocyte activation through a novel interaction with the adhesion molecules CD11b-CD18 and CD11c-CD18. Immunity 3:119. 16. Plater-Zyberk, C., and J. Y. Bonnefoy. 1995. Marked amelioration of established collagen-induced arthritis by treatment with antibodies to CD23 in vivo. Nat. Med. 1:781. 17. Kawabe, T., M. Takami, M. Hosoda, Y. Maeda, S. Sato, M. Mayumi, H. Mikawa, K. I. Arai, and J. Yodoi. 1988. Regulation of FceRII/CD23 gene expression by cytokines and specific ligands (IgE and anti-FceRII monoclonal antibody). J. Immunol. 141:1376. 18. Spittler, A., C. Schiller, M. Willheim, C. Tempfer, S. Winkler, and G. Boltz-Nitulescu. 1995. IL-10 augments CD23 expression on U937 cells and down-regulates IL-4 driven CD23 expression on cultured human blood monocytes: effects of IL-10 and other cytokines on cell phenotype and phagocytosis. Immunology 85:311. 19. Wu, C. Y., M. Sarfati, C. Heusser, S. Fournier, M. Rubio-Trujillo, R. Peleman, and G. Delespesse. 1991. Glucocorticoids increase the synthesis of IgE by interleukin-4-stimulated human leukocytes. J. Clin. Invest. 87:870. 20. Hornung, N., and D. Degiannis. 1993. Up-regulation by cyclosporine (CsA) of the in vitro release of soluble CD23 (sCD23) and of the in vitro production of IL-6 and IgM. Scand. J. Immunol. 38:287. 21. Degiannis, D., and N. Hornung. 1995. Rapamycin inhibits the in vitro release of soluble CD23 by SAC/IL-2/IL-4 activated human peripheral blood lymphocytes. Transplantation 59:1076. 22. Christie, G., A. Barton, B. Bolognese, D. R. Buckle, R. M. Cook, M. Hansbury, G. P. Harper, L. A. Marshall, M. E. McCord, K. Moulder, et al. 1997. IgE secretion is attenuated by an inhibitor of proteolytic processing of CD23 (FceRII). Eur. J. Immunol. 27:3228. 23. Lee, J., U. Prabhakar, D. Griswold, D. Dunnington, P. Young, and A. Badger. 1995. Low-molecular weight TNF biosynthesis inhibitors: strategies and perspectives. Circ. Shock 44:97. 24. Lee, J., J. T. Laydon, P. C. McDonnell, T. F. Gallagher, S. Kumar, D. Green, D. McNulty, M. J. Blumenthal, J. R. Heys, S. W. Landvatter, et al. 1994. A protein kinase involved in the regulation of inflammatory cytokine biosynthesis. Nature 372:739.

The Journal of Immunology 25. Lee, J. C., A. M. Badger, D. E. Griswold, D. Dunnington, A. Truneh, B. Votta, J. H. White, P. R. Young, and P. E. Bender. 1993. Bicyclic imidazoles as a novel class of cytokine biosynthesis inhibitors. Ann. NY Acad. Sci. 696:149. 26. Badger, A. M., J. N. Bradbeer, B. Votta, J. C. Lee, J. L. Adams, and D. E. Griswold. 1996. Pharmacological profile of SB 203580, a selective inhibitor of cytokine suppressive binding protein/p38 kinase, in animal models of arthritis, bone resorption, endotoxin shock and immune function. J. Pharmacol. Exp. Ther. 279:1453. 27. Cohen, P. 1997. The search for physiological substrates of MAP and SAP kinases in mammalian cells. Trends Cell Biol. 7:353. 28. Cuenda, A., J. Rouse, R. Doza, R. Meier, P. Cohen, T. Gallagher, P. Young, and J. Lee. 1995. SB 203580 is a specific inhibitor of a MAP kinase homologue which is stimulated by cellular stresses and interleukin-1. FEBS Lett. 364:229. 29. Kumar, S., P. C. McDonnell, R. J. Gum, A. T. Hand, J. C. Lee, and P. R. Young. 1997. Novel homologues of CSBP/p38 MAP kinase: activation, substrate specificity and sensitivity to inhibition by pyridinyl imidazoles. Biochem. Biophys. Res. Commun. 235:533. 30. Lee, J., D. E. Griswold, B. Votta, and N. Hanna. 1988. Inhibition of monocyte IL-1 production by the anti-inflammatory compound, SK&F 86002. Int. J. Immunopharmacol. 10:836. 31. Marshall, L. A., and A. K. Roshak. 1993. Coexistence of two biochemically distinct phospholipase A2 activities in human platelet, monocyte and neutrophil. Biochem. Cell Biol. 71:331. 32. Vercelli, D., H. H. Jabara, B. W. Lee, N. Woodland, R. S. Geha, and D. Y. M. Leung. 1988. Human recombinant interleukin-4 induces FceRII/CD23 on normal human monocytes. J. Exp. Med. 167:1406. 33. Morre, J. D., and D. M. Morre. 1989. Preparation of mammalian plasma membranes by aqueous two-phase partition. BioTechniques 7:946. 34. Marolewski, A. E., D. R. Buckle, G. Christie, D. L. Earnshaw, P. L. Flamberg, L. A. Marshall, D. G. Smith, and R. J. Mayer. 1998. CD23 (FceRII) release from cell membranes is mediated by a membrane-bound metalloprotease. Biochem. J. 333:573. 35. Bonnefoy, J. Y., J. Shields, and J. J. Mermod. 1990. Inhibition of human interleukin 4-induced IgE synthesis by a subset of anti-CD23/FceRII monoclonal antibodies. Eur. J. Immunol. 20:139. 36. Roshak, A., G. Sathe, and L. A. Marshall. 1994. Suppression of monocyte 85 kDa phospholipase A2 by antisense and effects on endotoxin-induced prostaglandin biosynthesis. J. Biol. Chem. 269:25999. 37. McLaughlin, M. M., S. Kumar, P. C. McDonnell, S. Van Horn, J. C. Lee, G. P. Livi, and P. R. Young. 1996. Identification of mitogen-activated protein (MAP) kinase-activated protein kinase-3, a novel substrate of CSBP p38 MAP kinase. J. Biol. Chem. 271:8488. 38. Waterman, W. H., T. F. P. Molski, C. K. Huang, J. L. Adams, and R. I. Sha’afi. 1996. Tumour necrosis factor-a-induced phosphorylation and activation of cytosolic phospholipase A2 are abrogated by an inhibitor of the p38 mitogen-activated protein kinase cascade in human neutrophils. Biochem. J. 319:17. 39. Kramer, R. M., E. F. Roberts, J. Manetta, and J. Putnam. 1991. The Ca21 sensitive cytosolic phospholipase A2 is a 100 kDa protein in human monoblast U937 cells. J. Biol. Chem. 266:5268.

6013 40. Kawaguchi, H., K. Nemoto, L. G. Raisz, J. R. Harrison, O. S. Voznesensky, C. B. Alander, and C. C. Pilbeam. 1996. Interleukin-4 inhibits prostaglandin G H synthase-2 and cytosolic phospholipase A2 induction in neonatal parietal bone cultures. J. Bone Miner. Res. 11:358. 41. Bansal, A., T. Roberts, E. M. Hay, R. S. H. Pumphrey, and P. B. Wilson. 1992. Soluble CD23 levels are elevated in the serum of patients with primary Sjogren’s syndrome and systemic lupus erythematosus. Clin. Exp. Immunol. 89:452. 42. Hellen, E. A., D. C. Rowlands, T. T. Hansel, G. D. Kitas, and J. Crocker. 1991. Immunohistochemical demonstration of CD23 expression on lymphocytes in rheumatoid synovitis. J. Clin. Pathol. 44:293. 43. Paul-Eugene, N., D. Mossalayi, M. Sarfati, K. Yamaoka, J. Aubry, J. Bonnefoy, B. Dugas, and J. Kolb. 1995. Evidence for a role of FceRII/CD23 in the IL-4induced nitric oxide production by normal human mononuclear phagocytes. Cell. Immunol. 163:314. 44. Grenier-Brossette, N., I. Bourget, C. Akoundi, J. Bonnefoy, and J. Cousin. 1992. Spontaneous and ligand-induced endocytosis of CD23 (Fce receptor II) from the surface of B lymphocytes generates a 16-kDa intracellular fragment. Immunology 22:1573. 45. Raingeaud, J., S. Gupta, J. S. Rogers, M. Dickens, J. Han, R. J. Ulevitch, and R. J. Davis. 1995. Pro-inflammatory cytokines and environmental stress cause p38 mitogen-activated protein kinase activation by dual phosphorylation on tyrosine and threonine. J. Biol. Chem. 270:7420. 46. Foltz, I. N., J. C. Lee, P. R. Young, and J. W. Schrader. 1997. Hemopoietic growth factors with the exception of interleukin-4 activate the p38 mitogenactivated protein kinase pathway. J. Biol. Chem. 272:3296. 47. Lee, J. C., and P. R. Young. 1996. Role of CSBP/p38/RK stress response kinase in LPS and cytokine signaling mechanisms. J. Leukocyte Biol. 59:152. 48. Badger, A. M., M. N. Cook, M. W. Lark, T. M. Newman-Tarr, B. A. Swift, A. H. Nelson, F. C. Barone, and S. Kumar. 1998. SB 203580 inhibits p38 MAP kinase, nitric oxide production and inducible nitric oxide synthase in bovine cartilage-derived chondrocytes. J. Immunol. 161:467. 49. Pucillo, C., S. Salzano, S. Pepe, M. Vitale, S. Formisano, and G. Rossi. 1990. Regulation of the expression of the low-affinity IgE receptor (FceRII) in the human monocyte-like cell line U937 by phorbol esters and IgE. Int. Arch. Allergy Appl. Immunol. 93:330. 50. Reichel, M., B. H. Nelson, P. D. Greenberg, and P. B. Rothman. 1997. The IL-4 receptor a-chain cytoplasmic domain is sufficient for activation of JAK-1 and STAT6 and the induction of IL-4 specific gene transcription. J. Immunol. 158: 5860. 51. Han, H., T. Brown, and B. Beutler. 1990. Endotoxin-responsive sequences control cachectin/tumor necrosis factor biosynthesis at the translational level. J. Exp. Med. 71:465. 52. Waskiewicz, A. J., A. Flynn, C. G. Proud, and J. A. Cooper. 1997. Mitogenactivated protein kinases activate the serine/threonine kinases Mnk1 and Mnk2. EMBO J. 16:1909.