Parallel signal processing among mammalian MAPKs.

The intracellular signalling field is dominated by the mitogen-activated protein kinase (MAPK) cascade and its control, which involves the small GTPase Ras and sequential kinase activation. Until recently, ERK1 and ERK2 were the only cloned and well-characterized mammalian MAPKs; diverse ligand-stimulated, proline-directed protein phosphorylation events were attributed to these kinases. The recent discovery of two other MAPK subtypes, the JNK/SAPK subfamily and p38/RK (mammalian equivalents of HOG1 in yeast), reveals extreme complexity within the family and, most intriguingly, the existence in mammalian cells of parallel MAPK cascades that can be activated simultaneously.

p38/RK is essential for stress-induced nuclear responses: JNK/SAPKs and c-Jun/ATF-2 phosphorylation are insufficient.

The ERK, JNK/SAPK and p38/RK MAP kinase subtypes (reviewed in [1]) are differentially activated in mammalian cells by various stimuli, which elicit induction of immediate-early (IE) genes, such as c-fos and c-jun (reviewed in [1-3]), as well as phosphorylation of histone H3 [4] and HMG-14 [5]. Anisomycin and UV radiation have been suggested to induce c-fos and c-jun transcription via JNK/SAPK-mediated phosphorylation of TCF (ternary complex factor), for c-fos induction [6-8], and c-Jun and/or ATF-2 for c-jun induction [9-11] [12,13]. We report here that anisomycin and ultraviolet radiation (UV) activate MAP kinase kinase-6 (MKK6) [14,15], p38/RK [16] [17,18] and MAPKAP kinase-2 (MAPKAP K-2) [17-19]. By using the p38/RK inhibitor SB 203580 [20,21], we show that activation of p38/RK and/or its downstream effectors are essential for anisomycin- and UV-stimulated c-fos/c-jun induction and histone H3/HMG-14 phosphorylation, whereas JNK/SAPK activation and phosphorylation of c-Jun and ATF-2 are insufficient for these responses.

Anisomycin-activated protein kinases p45 and p55 but not mitogen-activated protein kinases ERK-1 and -2 are implicated in the induction of c-fos and c-jun.

Independent of its ability to block translation, anisomycin intrinsically initiates intracellular signals and immediate-early gene induction [L. C. Mahadevan and D. R. Edwards, Nature (London) 349:747-749, 1991]. Here, we characterize further its action as a potent, selective signalling agonist. In-gel kinase assays show that epidermal growth factor (EGF) transiently activates five kinases: the mitogen-activated protein (MAP) kinases ERK-1 and -2, and three others, p45, p55, and p80. Anisomycin, at inhibitory and subinhibitory concentrations, does not activate ERK-1 and -2 but elicits strong sustained activation of p45 and p55, which are unique in being serine kinases whose detection is enhanced with poly-Glu/Tyr or poly-Glu/Phe copolymerized in these gels. Translational arrest using emetine or puromycin does not activate p45 and p55 but does prolong EGF-stimulated ERK-1 and -2 activation. Rapamycin, which blocks anisomycin-stimulated p70/85S6k activation without affecting nuclear responses, has no effect on p45 or p55 kinase. p45 and p55 are activable by okadaic acid or UV irradiation, and both kinases phosphorylate the c-Jun NH2-terminal peptide 1-79, putatively placing them within c-Jun NH2-terminal kinase/stress-activated protein kinase (JNK/SAPK) subfamily of MAP kinases. Thus, the EGF- and anisomycin-activated kinases p45 and p55 are strongly implicated in signalling to c-fos and c-jun, whereas the MAP kinases ERK-1 and -2 are not essential for this process.

Anisomycin and rapamycin define an area upstream of p70/85S6k containing a bifurcation to histone H3-HMG-like protein phosphorylation and c-fos-c-jun induction.

Anisomycin, a translational inhibitor, synergizes with growth factors and phorbol esters to superinduce c-fos and c-jun by a number mechanisms, one of which is its ability to act as a potent signalling agonist, producing strong, prolonged activation of the same nuclear responses as epidermal growth factor or tetradecanoyl phorbol acetate. These responses include the phosphorylation of pp33, which exists in complexed and chromatin-associated forms, and of histone H3 and an HMG-like protein. By peptide mapping and microsequencing, we show here that pp33 is the phosphoprotein S6, present in ribosomes and in preribosomes in the nucleolus. Ablation of epidermal growth factor-, tetradecanoyl phorbol acetate-, or anisomycin-stimulated S6 phosphorylation by using the p70/85S6k inhibitor rapamycin has no effect on histone H3 and HMG-like protein phosphorylation or on the induction and superinduction of c-fos and c-jun. Further, [35S]methionine-labelling and immunoprecipitation studies show that the ablation of S6 phosphorylation has no discernible effect on translation in general or translation of newly induced c-fos transcripts. Finally, we show that anisomycin augments and prolongs S6 phosphorylation not by blocking S6 phosphatases but by sustained activation of p70/85S6k. These results suggest the possible use of anisomycin and rapamycin to define upstream and downstream boundaries of an area of signalling above p70/85S6k which contains a bifurcation that produces histone H3-HMG-like protein phosphorylation and c-fos-c-jun induction in the nucleus.

Anisomycin selectively desensitizes signalling components involved in stress kinase activation and fos and jun induction.

Anisomycin, a translational inhibitor secreted by Streptomyces spp., strongly activates the stress-activated mitogen-activated protein (MAP) kinases JNK/SAPK (c-Jun NH2-terminal kinase/stress-activated protein kinase) and p38/RK in mammalian cells, resulting in rapid induction of immediate-early (IE) genes in the nucleus. Here, we have characterized this response further with respect to homologous and heterologous desensitization of IE gene induction and stress kinase activation. We show that anisomycin acts exactly like a signalling agonist in eliciting highly specific and virtually complete homologous desensitization. Anisomycin desensitization of a panel of IE genes (c-fos, fosB, c-jun, junB, and junD), using epidermal growth factor (EGF), basic fibroblast growth factor, (bFGF), tumor necrosis factor alpha (TNF-alpha), anisomycin, tetradecanoyl phorbol acetate (TPA), and UV radiation as secondary stimuli, was found to be extremely specific both with respect to the secondary stimuli and at the level of individual genes. Further, we show that anisomycin-induced homologous desensitization is caused by the fact that anisomycin no longer activates the JNK/SAPK and p38/RK MAP kinase cascades in desensitized cells. In anisomycin-desensitized cells, activation of JNK/SAPKs by UV radiation and hyperosmolarity is almost completely lost, and that of the p38/RK cascade is reduced to about 50% of the normal response. However, all other stimuli produced normal or augmented activation of these two kinase cascades in anisomycin-desensitized cells. These data show that anisomycin behaves like a true signalling agonist and suggest that the anisomycin-desensitized signalling component(s) is not involved in JNK/SAPK or p38/RK activation by EGF, bFGF, TNF-alpha, or TPA but may play a significant role in UV- and hyperosmolarity-stimulated responses.

2-Aminopurine abolishes epidermal growth factor-stimulated phosphorylation of complexed and chromatin-associated forms of a 33 kDa phosphoprotein.

We have recently reported that several purified polypeptide mitogens such as epidermal growth factor, embryonal carcinoma-derived growth factor, basic fibroblast growth factor and Bombesin induce the rapid appearance of a 33 kDa chromatin-associated phosphoprotein in the nuclei of murine fibroblasts. We show here that in both mouse and human cell lines, a second form of the 33 kDa phosphoprotein exists in a detergent-extractable complexed form which may be pelleted by ultracentrifugation. When quiescent [32P]-labelled cells are treated with EGF, both complexed and chromatin-associated forms of the labelled phosphoprotein are detectable within 10 min, the response peaking at about 1 h and being substantially over 3 h after EGF stimulation. By chymotryptic and cyanogen bromide phosphopeptide mapping studies, the two forms of the 33 kDa phosphoprotein are indistinguishable, as are the mouse and human forms of the protein. The protein kinase inhibitor 2-aminopurine, which has recently been shown to block growth factor-stimulated c-fos and c-myc induction, specifically abolishes the mitogen-stimulated appearance of both forms of the 33 kDa phosphoprotein without affecting the phosphorylation of other cellular proteins. The 33 kDa protein has been purified from Hela cells by a combination of sucrose density gradient centrifugation, preparative electrophoresis and reverse-phase HPLC during which the protein is resolved into two closely-eluting peaks which differ markedly in their specific activity. These results are discussed in relation to the possible role of these events in coupling growth factor-receptor interaction at the cell surface to the early responses of transcriptional activation in the nucleus.

Effects of the inhibition of p38/RK MAP kinase on induction of five fos and jun genes by diverse stimuli.

The ERK, JNK/SAPK and p38/RK MAP kinase subtypes are differentially activated by physiological, pharmacological and stress stimuli; all three subtypes are implicated in immediate-early (IE) gene induction by these agents. Here, we have asked whether inhibition of a single MAP kinase subtype under these conditions would generally alter induction of several IE genes in a similar way or whether this would differentially up- and down-regulate particular IE genes, an issue which bears on the question of whether individual MAP kinases are strictly targeted to specific IE genes, or whether they might catalyse phosphorylation events that affect several IE genes in the same way. SB 203580, an inhibitor of p38/RK, has been used to analyse the role of this kinase in the induction of five IE genes (c-fos, fosB, c-jun, junB and junD) under diverse conditions of stimulation. In C3H 10T1/2 cells, p38/RK and its downstream kinase MAPKAP K-2 are activated by all stimuli used with the exception of TPA. The specificity of SB 203580 as a p38/RK inhibitor in these cells is demonstrated; it does not affect ERKs or JNK/SAPKs but does result in a small increase in the activity of the upstream kinase MKK6, the principal p38/RK activator in these cells. We find that inhibition of p38/RK under these conditions produces general effects on all five IE genes as a group in three ways. First, induction of all five genes in response to okadaic acid or tumour necrosis factor-alpha (TNF-alpha) is not significantly altered by SB 203580. Second, in cells stimulated with anisomycin or U.V. radiation, SB 203580 potently inhibits all of the induced IE genes. Finally, SB 203580 enhances induction of all five IE genes in EGF-treated cells; these enhanced mRNA levels are not due to stabilisation of labile mRNA transcripts. The significance of these results to current thinking on the relationship between distinct MAP kinase subtypes and specific IE genes is discussed.

Identification of anisomycin-activated kinases p45 and p55 in murine cells as MAPKAP kinase-2.

We recently showed that EGF and anisomycin activate two kinases, p45 and p55, whose distinguishing feature is that their detection in in-gel kinase assays is enhanced by copolymerised poly-Glu/Tyr or poly-Glu/Phe (Cano E, Hazzalin CA and Mahadevan LC, Mol. Cell. Biol., 20:117-121). Their activation characteristics and sizes are strikingly similar to those of JNK/SAPKs, which are also strongly activated by anisomycin. However, we show here that p45 and p55 are not JNK/SAPKs but murine forms of MAPKAP kinase-2 because: (i) Detection of immunoprecipitated JNK/SAPKs is completely dependent on the presence of c-Jun as substrate in the in-gel kinase assays, whereas detection of p45 and p55 is not. (ii) Detection of p45 and p55 activity is enhanced by the presence of poly-Glu/Tyr or poly-Glu/Phe, whereas JNK/SAPKs are not detectable under these conditions. (iii) Although the sizes of the murine JNK/SAPKs and MAPKAP K-2 are similar, human JNK/SAPKs migrate at 45 and 55 kDa whereas human MAPKAP K-2 migrates at 50 kDa; the poly-Glu/Tyr-enhanced activity in human cells migrates at 50 KDa. (iv) Purified rabbit muscle MAPKAP K-2 is detectable as two bands of activity on in-gel kinase assays and their detection is enhanced by poly-Glu/Tyr. (v) Finally, the anisomycin-activated poly-Glu/Tyr-enhanced p45 and p55 kinases can be immunoprecipitated from murine cells using an anti-MAPKAP K-2 antibody. Thus, EGF- and anisomycin-activated p45 and p55 are not JNK/SAPKs but MAPKAP K-2, implying that both these agents activate the p38/RK MAP kinase cascade.

2-Aminopurine abolishes EGF- and TPA-stimulated pp33 phosphorylation and c-fos induction without affecting the activation of protein kinase C.

Epidermal Growth Factor (EGF) and Tetradecanoyl Phorbol Acetate (TPA) initiate signalling cascades in C3H 10T1/2 fibroblasts by primarily activating distinct protein kinases, the EGF receptor tyrosine kinase and protein kinase C respectively; there is no signal crossover at the initiation of signalling. Nevertheless, we show here that both agents rapidly elicit common intracellular responses, including the phosphorylation of complexed and chromatin-associated forms of a 33 kDa phosphoprotein (pp33), that of a 15 kDa chromatin-associated phosphoprotein (pp15), as well as the transcriptional activation of a common subset of genes including the c-fos proto-oncogene. 2-aminopurine specifically abolishes complexed and chromatin-associated pp33 phosphorylation in response to EGF and TPA, as well as the induction of c-fos by both agents. The activation of protein kinase C and the levels of transcription factors that bind to the serum response element (SRE), TPA response element (TRE) or NFkB sites in stimulated cells are relatively unaffected by 2-aminopurine. This, to our knowledge, is the first demonstration that it is possible, by using 2-aminopurine which selectively blocks TPA-stimulated pp33 phosphorylation, to block c-fos induction in TPA-treated cells although protein kinase C remains fully active. Further, we show here that although EGF- and TPA-stimulated induction of c-fos is abolished by 2-aminopurine, the appearance of TRE-binding activity in nuclear extracts of stimulated cells is unaffected, suggesting that EGF- and TPA-stimulated induction of TRE-binding activity utilises existing proteins and is not dependent on fresh c-FOS synthesis. These results imply that 2-aminopurine-sensitive complexed and chromatin-associated pp33 phosphorylation may be crucial to c-fos induction in response to EGF and TPA.

Rapid appearance of novel phosphoproteins in the nuclei of mitogen-stimulated fibroblasts.

A sequential extraction procedure has been used to prepare three protein fractions from control and mitogen-stimulated [32P]-PO4-labelled fibroblasts, the first containing soluble and membrane-bound proteins, the second mainly the cytoskeletal proteins vimentin and actin, and the third, a chromatin-derived fraction enriched in histones. We describe here an analysis of the mitogen-stimulated changes in the [32P] labelled phosphoprotein composition of these fractions. The most obvious changes in response to epidermal growth factor, embryonal carcinoma-derived growth factor, basic fibroblast growth factor and bombesin were the rapid appearance of 33 and 15 kDa phosphoproteins in the chromatin-derived fraction. The epidermal growth factor- and basic fibroblast growth factor-stimulated 33 kDa phosphoprotein produced similar chymotryptic peptides and was phosphorylated on serine residues. DNAse/RNAse treatment of the lysates was essential for the extraction of the 33 kDa phosphoprotein. Further, its presence could be demonstrated in preparations of conventionally purified nuclei. An in situ extraction procedure has been used to provide morphological verification of the sequential extraction data. The final structure containing these phosphoproteins is clearly derived from nuclei, enriched in histones, stains for DNA and appears by electron microscopy, to be homogenously composed of chromatin-like material. Thus, we describe here the rapid mitogen-induced appearance of novel phosphoproteins in the nucleus, raising the possibility that they may be involved in orchestrating early nuclear responses to polypeptide growth factors.

Calcium-dependant binding proteins associated with human placental syncytiotrophoblast microvillous cytoskeleton.

Isolated human placental syncytiotrophoblast microvillous plasma membrane vesicles were extracted with Triton X-100 to yield a detergent-insoluble residue. The residue contained approx. 50% of the total membrane protein and was qualitatively different from untreated trophoblast on SDS-polyacrylamide gel electrophoresis, Western blots and dot-immunobinding assay. Three major proteins, with molecular weights of 68, 36 and 34 kDa, dissociated from this non-ionic detergent-insoluble submembranous cytoskeletal fraction in the presence of calcium chelators. They were immunologically related to human lymphocyte cytoskeletal calcium-binding proteins, and the 36 kDa component reacted with antisera to the phospholipase A2 inhibitor, lipocortin II. Anti-lipocortin I sera did not recognise the 34 kDa protein, but did react with a series of trophoblast cytoskeletal proteins in the 34-37 kDa region. Incubation of epidermal growth factor with isolated trophoblast membrane vesicles stimulated the phosphorylation of a 36 kDa protein on tyrosine residues. Immunoprecipitation studies further showed there was no phosphorylation of the 34 kDa protein, but the 68 kDa protein was a major phosphorylated component of isolated syncytiotrophoblast membranes. p68 was principally phosphorylated on serine with slight tyrosine phosphorylation which showed an apparent increase after epidermal growth factor treatment. These results indicate a family of calcium-dependant binding proteins, some of which are phosphorylated, associated with the submembranous cytoskeleton of syncytiotrophoblast microvilli.

Intracellular signalling pathways involved in mesoderm induction by FGF.

We have examined the possible role of two signal transducing mechanisms, tyrosine phosphorylation and activation of protein kinase C (PKC), during fibroblast growth factor (FGF)-induced mesoderm induction in Xenopus. Tyrosine phosphorylation was examined through the use of a monoclonal anti-phosphotyrosine antibody. This antibody was shown to recognize the FGF receptor crosslinked to radioiodinated FGF. We also studied the response of Xenopus ectodermal explants to sodium orthovanadate, a compound that has been shown to elevate intracellular phosphotyrosine levels. Thirty percent of explants cultured in 100 microM vanadate were induced. In addition, vanadate synergized with FGF to give inductions that were more dorsal in nature than either vanadate or FGF alone. The role of PKC was evaluated by measuring PKC activity during mesoderm induction by FGF and by examining the effect of the phorbol ester 12-O-tetradecanoylphorbol 13-acetate (TPA) on explants. TPA did not induce mesoderm, however, activation of PKC was detected in FGF-treated explants. Therefore, activation of the PKC pathway alone is not sufficient for mesoderm induction. Simultaneous treatment with TPA and FGF resulted in a significant inhibition of mesoderm induction by FGF, suggesting that activation of PKC could be part of a negative feedback mechanism. In contrast, TPA had no effect on induction by activin A.

Acetyllysine-binding and function of bromodomain-containing proteins in chromatin.

Acetylated histones are generally associated with active chromatin. The bromodomain has recently been identified as a protein module capable of binding to acetylated lysine residues, and hence is able to mediate the recruitment of factors to acetylated chromatin. Functional studies of bromodomain-containing proteins indicate how this domain contributes to the activity of a number of nuclear factors including histone acetyltransferases and chromatin remodelling complexes. Here, we review the characteristics of acetyllysine-binding by bromodomains, discuss associated domains found in these proteins, and address the function of the bromodomain in the context of chromatin. Finally, the modulation of bromodomain binding by neighbouring post-translational modifications within histone tails might provide a mechanism through which combinations of covalent marks could exert control on chromatin function.

Use of reversed-phase high-performance liquid chromatography on polystyrene-divinylbenzene columns for the rapid separation and purification of acid-soluble nuclear proteins.

The suitability of polystyrene-divinylbenzene reversed-phase HPLC columns for rapid separation and purification of acid-soluble nuclear proteins was evaluated. We used a polystyrene-divinylbenzene reversed-phase HPLC column (PLRP-S) for purification of nuclear proteins extracted with 0.3 M HCl or 5% HClO4. We are able to obtain electrophoretically pure fractions for a number of nuclear proteins including HMG14, HMG17 and variants of histone H3. The identity of proteins in these fractions was confirmed by immunochemical analysis, protein sequencing, mass spectrometry and migration on two-dimensional polyacrylamide gel electrophoresis. These methods do not require special preparation of the sample and are quicker than similar published methods.

Protein synthesis inhibitors differentially superinduce c-fos and c-jun by three distinct mechanisms: lack of evidence for labile repressors.

Protein synthesis inhibitors strongly augment and prolong the usually transient induction of c-fos and c-jun by growth factors, phorbol esters etc., a phenomenon termed superinduction which is conventionally regarded as a secondary consequence of translational arrest. Our recent demonstration that some inhibitors can act positively as nuclear signalling agonists compromises this view and necessitates a re-evaluation of superinduction. First, we show that labile repressors, widely postulated to act negatively on diverse superinducible genes, are not involved in regulating c-fos and c-jun. Secondly, two components of c-fos and c-jun superinduction, namely the delay in shutting off transcription and stabilization of their mRNAs, arise from translational arrest and are common to all protein synthesis inhibitors. Thirdly, the recently described capacity to act positively as nuclear signalling agonists to stimulate pp33/pp15 phosphorylation is restricted to compounds such as anisomycin and cycloheximide; these, but not emetine or puromycin, will induce c-fos/c-jun on their own. Fourthly, the translational arrest-related components of superinduction are dissociable from the signalling agonist effects at sub-inhibitory concentrations of anisomycin, under which conditions a new type of c-fos/c-jun superinduction with 'spike' kinetics is observed. Finally, we show that in response to EGF plus anisomycin, the nuclear signalling responses are themselves augmented and prolonged in a manner that corresponds to c-fos/c-jun superinduction under these conditions.

Rapid histone H3 phosphorylation in response to growth factors, phorbol esters, okadaic acid, and protein synthesis inhibitors.

When quiescent cells are stimulated with growth factors, phorbol esters, okadaic acid, or protein synthesis inhibitors, the early-response genes, which include c-fos and c-jun, are rapidly induced. The earliest growth factor- and phorbol ester-stimulated nuclear signaling events concomitant with proto-oncogene induction are the rapid phosphorylation of two chromatin-associated proteins, pp33 and pp15. We show here that the tumor promoter okadaic acid, which inhibits protein phosphatases 1 and 2A, and the protein synthesis inhibitors anisomycin and cycloheximide also stimulate pp33 and pp15 phosphorylation. Using transcriptional inhibitors, we show that this response is not a consequence of early gene induction. By peptide mapping and microsequencing, chromatin-associated pp15 is identified as histone H3. Upon stimulation, histone H3 is rapidly phosphorylated on serine residues within its highly charged, basic N-terminal domain. Thus, these diverse agents elicit a common early nuclear signal modulating nucleosomal structure or function, potentially contributing to conformational regulation of proto-oncogene induction.

Independent dynamic regulation of histone phosphorylation and acetylation during immediate-early gene induction.

Induction of c-fos and c-jun is associated with phosphoacetylation of histone H3 and acetylation of histone H4. Upon induction, a large population of nucleosomes becomes highly acetylated on histones H3 and H4, whereas a much smaller population of comparable nucleosomes at similar positions along the gene becomes phosphoacetylated. Inhibiting histone H3 phosphorylation with kinase inhibitors does not measurably alter the enhanced acetylation of these nucleosomes. Finally, whereas H3 phosphorylation is a MAP kinase-mediated inducible event, we found acetylation to be continuously turning over by the targeted action of HATs and HDACs in the absence of any stimulation or gene transcription. These studies suggest that phosphorylation and acetylation are independently and dynamically regulated at these genes and reveal the complexity of multiple histone modifications at immediate-early gene chromatin.

MAP kinase-mediated signalling to nucleosomes and immediate-early gene induction.

Extracellular signals are transduced into the nucleus through a variety of signalling mechanisms to elicit changes in patterns of gene expression. This review is focused in the MAP kinase cascades and the part they play in the induction of the immediate-early (IE) genes. We discuss the MAP kinases and their downstream effectors that are known to phosphorylate substrates in the nucleus. In addition to phosphorylating specific transcription factors, MAP kinases and their downstream kinases are implicated in eliciting rapidly targeted alterations in the chromatin environment of specific genes by modulating the phosphorylation and/or acetylation of nucleosomal and chromatin proteins.

Combinations of ERK and p38 MAPK inhibitors ablate tumor necrosis factor-alpha (TNF-alpha ) mRNA induction. Evidence for selective destabilization of TNF-alpha transcripts.

Tumor necrosis factor-alpha (TNF-alpha) is a potent proinflammatory cytokine whose synthesis and secretion are implicated in diverse pathologies. Hence, inhibition of TNF-alpha transcription or translation and neutralization of its protein product represent major pharmaceutical strategies to control inflammation. We have studied the role of ERK and p38 mitogen-activated protein (MAP) kinase in controlling TNF-alpha mRNA levels in differentiated THP-1 cells and in freshly purified human monocytes. We show here that it is possible to produce virtually complete inhibition of lipopolysaccharide-stimulated TNF-alpha mRNA accumulation by using a combination of ERK and p38 MAP kinase inhibitors. Furthermore, substantial inhibition is achievable using combinations of 1 microm of each inhibitor, whereas inhibitors used individually are incapable of producing complete inhibition even at high concentrations. Finally, addressing mechanisms involved, we show that inhibition of p38 MAP kinase selectively destabilizes TNF-alpha transcripts but does not affect degradation of c-jun transcripts. These results impinge on the controversy in the literature surrounding the mode of action of MAP kinase inhibitors on TNF-alpha mRNA and suggest the use of combinations of MAP kinase inhibitors as an effective anti-inflammatory strategy.