Mitogen-Activated Protein Kinases (MAPKs): ERKs, JNKs, and p38s


Wei Chen, Yu-Chi Juang and Melanie Cobb, UTSouthwestern Medical Center—Pharmacology, Dallas, Texas

doi: 10.1002/9780470048672.wecb348


Mitogen-activated protein kinases (MAPKs) are significant mediators in signal transduction pathways from the membrane to intracellular compartments including the nucleus. They regulate the functions of many gene products and therefore affect cell growth, differentiation, and death. Three main MAPK subfamilies have been identified and are studied widely. All function in cascades that include at least three tiers of protein kinases. Selective kinase inhibitors have been developed;they are powerful tools to study the physiologic functions of MAPKs and, in some cases, are promising and effective drugs. In this article, the regulation of the three main MAPK pathways is sketched, and the status of MAPK inhibitors and their inhibiting mechanisms are highlighted.


Signal transduction pathways play important roles in transducing environmental changes to the regulatory machinery in the cell, which allows cells to alter their behavior rapidly to respond appropriately to those changes. Protein kinases are major and critical components of signaling pathways. More than 500 genes that encode protein kinase domains have been identified in the human genome. They share conserved domains in sequence and in structure, but they have notable differences in regulatory mechanisms that are often similar within subfamilies (1). Mitogen-activated protein kinases (MAPKs) are protein Ser/Thr kinases that are involved in a broad range of cellular events, such as cell proliferation, cell death, homeostasis, acute hormonal responses, and the morphologic changes of embryogenesis and cell differentiation (2-9). Perturbation of these pathways results in diseases such as cancers, diabetes, inflammation disorders, and autoimmune disorders. Therefore, MAPKs are important targets for drug development. The regulation of MAPK pathways has been studied widely. Several model systems have been established, including activating those pathways by various stimuli and identification of in vitro and possible in vivo substrates. However, novel strategies and tools for studying the physiologic functions of MAPKs are still needed. With the identification of the first inhibitors for the ERK1/2 MAPK pathway, PD098059 and U0126 (10, 11), pathway-specific inhibitors have proven indispensable in determining MAPK functions in biologic systems. The lack of absolute specificity is a well-known disadvantage of protein kinase inhibitors and will be discussed in the context of MAPK pathways. The advantages are many and include the possibility of investigating pathways in many cells and in whole animals. Moreover, these inhibitors may have therapeutic potential as anticancer or anti-inflammatory drugs, for example. Thus far, several dozen protein kinase inhibitors are in clinical development currently, and many more are in preclinical studies (12-14). In this article, the main MAPK subfamilies and their regulation are introduced and the status of selected small molecule inhibitors of MAPK pathways will be discussed with a focus on inhibitory mechanisms. For the sake of space, references are made to more extensive reviews that cite the primary literature.



The MAPKs are protein kinases activated by growth factors, hormones, cytokines, and environmental stresses. One or more MAPKs are activated by almost every cell stimulus. The first MAPKs sequenced, Kss1p and Fus3p, are also the kinases most similar to mammalian ERK1/2 and were found in the pheromone response pathway of the budding yeast nearly 20 years ago (8). Subsequently, activities found in mammalian cells that favored Ser/Thr residues followed by Pro in substrates were purified and were shown to be mammalian MAPKs. More than a dozen mammalian MAPKs have now been identified. The hallmark of the MAPK family is the tri-peptide motif (Thr-Xxx-Tyr) located within the activation loop (T-loop) of the kinase domain, which contains the two sites phosphorylated to activate the kinases (2-9). Based on the canonical TXY motif and other features of the primary sequence, MAPKs are classified even more into three major subgroups: the extracellular signal-regulated protein kinase (ERKs), the c-Jun N-terminal kinases or stress-activated protein kinases (JNK/SAPK), and the p38 family of kinases (p38). Although many upstream inputs exist at each step of the cascade, each MAPK pathway can be viewed as a linear kinase cascade that contains at least three layers of protein kinases. Upstream MAPK kinase kinases (MAP3Ks) phosphorylate and activate MAPK kinases (MAP2Ks), which further phosphorylate and activate downstream MAPKs with great selectivity. The activated MAPKs then phosphorylate their target proteins, which include other protein kinases, other enzymes, transcription factors, and cytoskeletal and regulatory factors involved in cell attachment and migration, for example. The cascades that culminate in activation of the three major MAPK groups are summarized briefly below (Fig. 1).



Figure 1


ERK pathways

To date, six MAPKs termed ERKs [ERKs 1, 2, 3, 4, 5, 7 (also called ERK8 depending on species)] have been identified (7-9). All but ERKs 3 and 4 contain the highly conserved TEY (Thr-Glu-Tyr) motif in the activation loop. The atypical ERKs 3 and 4 possess a SEG (Ser-Glu-Gly) activation loop sequence instead. ERK1/2 are the archetype MAPKs and also are the best-studied kinases in this subgroup. To some extent, ERK1/2 respond to most ligands and other cellular stimuli; nevertheless, the most pronounced responses are to growth factors, serum, phorbol esters, and cytokines [frequently reviewed, e.g., (2), and other articles in that volume]. ERK1/2 are 43 and 41 kDa proteins with 83% sequence identity. The MAP3Ks that activate ERK1/2 include Raf-1 (or c-Raf) and B-Raf, most typically, A-Raf where it is expressed, and Mos and Tpl2 under very specific circumstances that have been reviewed in detail (2). Raf-1 is the best studied MAP3K for this pathway and is expressed ubiquitously. Once stimulated, it is phosphorylated and then activates the MAP2Ks MEK1/2, which in turn phosphorylate and activate ERK1/2. B-Raf can associate with and activate Raf-1, and this association can be involved in the transforming events caused by B-Raf in certain tumors (15).

ERK5, which is also well studied, is known as big MAP kinase 1 (BMK1) because it is twice the size of ERK1/2. Although ERK5 shares a TEY motif in its kinase domain, it contains a unique long C terminus that may have important regulatory functions (5). In the ERK5 pathway, the upstream kinase of ERK5 is MEK5 but not MEK1/2. MEK5 may be phosphorylated by the MAP3Ks MEKK2, MEKK3, Tpl2, and mixed-lineage kinases (MLKs). However, scaffolding is thought to distinguish settings in which these MAP3Ks act on ERK5 compared with other MAPKs discussed below (16). In common with ERK1/2, ERK5 is activated in response to serum and growth factors such as nerve growth factor and epidermal growth factor. ERK5 is more sensitive than ERK1/2 to many stress stimuli, such as oxidative stress and hyperosmolarity, although all three may be activated (5).

Compared with ERK1/2 and 5, the other ERKs are studied much less. Their regulatory mechanisms are less understood and may not involve dedicated MAP2Ks (7, 9).


JNK pathways

JNK (c-Jun N-terminal kinase) was first identified as the UV-induced activity responsible for phosphorylating, and thereby activating the proto-oncogene c-Jun (5). At the same time, they were found as SAPKs (stress-activated protein kinases), which are proline-directed kinases activated by growth factors and biosynthetic inhibitors such as anisomycin. Common stimuli that activate JNKs include inflammatory cytokines; fatty acids; and environmental stresses such as UV, osmotic shock, heat shock, ionizing radiation, oxidative stress, and, to a lesser extent, growth factors. In mammals, three genes encode highly related but distinct JNKs/SAPKs: JNK1/SAPKP, JNK2/SAPKa, and JNK3/SAPKy. These proteins, which exist in 10 or more alternatively spliced forms, share more than 85% identity in the core kinase domain. JNK1 and JNK2 are ubiquitous, whereas JNK3 is expressed primarily in neuronal tissues and in the cardiac myocyte. The JNKs are activated by dual phosphorylation on the activation loop TPY (Thr-Pro-Tyr) motif by the MAP2Ks MEK4 and MEK7. Interestingly, MEK4 displays a preference for tyrosine and MEK7 for threonine, which suggests that MEK4 and MEK7 activate JNKs synergistically (5). Many MAP3Ks have been reported to activate MEK4/7, including MEKK 1-4, ASKs, Tpl2, MLKs, and TAK1 (4, 5, 7, 17). Knockout studies have suggested that different MAP3Ks are involved in JNK activation by different stimuli. JNK has important roles in determining cell fate during metazoan development; as well as an involvement in tumorigenesis, inflammation, and obesity (18). JNK inhibits insulin signaling through phosphorylation and desensitization of insulin receptor substrate 1 (25-1). Obesity results in JNK activation, which suggests that JNK inhibitors may hold promise for the treatment of type 2 diabetes, insulin resistance, and obesity (19). JNK activation may lead to apoptosis in appropriate settings. Originally, the phenotype of JNK knockout mice connected JNKs with both the immune response and the apoptosis.


p38 pathways

Originally, p38 (p38α) was identified in three laboratories in three distinct contexts. It was found as the molecular target in a screen for drugs that inhibit tumor necrosis factor α-mediated inflammatory responses, a 38kDa protein rapidly phosphorylated in response to lipopolysaccharide stimulation, and a stress-activated kinase that activated MAPKAP kinase 2 (2, 5-7). Four p38 isoforms, p38α, p38β, p38y/SAPK3/ERK6, and p38S/SAPK4, have been identified in mammals. Although all retain a TGY activation loop motif, they have only slightly more than 60% sequence identity to one another. A result of this relatively low sequence identity is the limited cross-reactivity of antibodies with all the family members and the lack of effects of some inhibitors on all of the isoforms. The isoforms also differ in their tissue distribution, activating stimuli, and downstream substrates. p38α is the best studied. The activity of p38α is stimulated by growth factors; stresses such as UV light, osmotic shock, and ionizing radiation; as well as inflammatory cytokines such as tumor necrosis factor (TNFα) and interleukin 1 (26-1) (2, 5-7). Once stimulated, p38 is activated by dual phosphorylation on the TGY (Thr-Gly-Tyr) motif by the upstream MAP2Ks MEKs 3 and 6. These upstream kinases have preferential effects on different p38 isoforms. p38α, p38γ, and p38S are phosphorylated by both MEK3 and MEK6, and p38β is phosphorylated preferentially by MEK6. MEK3/6 can be activated by the MAP3Ks MEKK1-4, MLKs, ASK1, TAK1, DLK, and TAOs. It is well known that p38 is involved in inflammation, apoptosis, and cell differentiation.


MAPK Inhibitors

The functions of MAPKs have been studied primarily using dominant negative mutants. Constitutively active point mutants of MAPKs are not available, and other constitutive forms involve fusions or mutations of multiple residues. Active MAP3Ks and MAP2Ks have also been expressed, but overexpressed MAP3Ks in particular often activate other pathways in addition to those intended (7). The development of specific inhibitors for each subfamily of MAPKs and inhibitors that act at different levels in these pathways would facilitate our understanding of the complex interactions of these signaling cascades greatly. Inhibitors for some MAPK, MAP2K, and MAP3K family members have been developed. These inhibitors have assisted in identifying physiologic substrates and cellular functions of these enzymes. Selected inhibitors of MAPKs are shown in Fig. 1.



Protein kinases have two substrates: target proteins and ATP-Mg2+. Kinases transfer the γ-phosphoryl group of ATP to hydroxyl acceptor groups of Tyr, Ser, and Thr residues within target proteins. Phosphorylation may control the activation, inactivation, protein interactions, stability, and localization of the substrate. Crystallographic studies have shown that protein kinases consist of two folding domains: a smaller N-terminal domain composed largely of anti-parallel β-strands and a larger C-terminal domain composed primarily of α-helices. The nucleotide is bound in a cleft formed at the interface of the two folding domains (ATP-binding pocket) (20-22). Protein substrates bind largely outside of the active site cleft on the surface of the C-terminal domain. The ATP binding pocket, together with less conserved surrounding pockets, has been the focus of inhibitor design. Most inhibitors target the ATP binding site itself and are competitive with ATP. Thus, the potential for any compound to inhibit multiple kinases is considerable. Inhibitor specificity cannot be deduced from primary sequence similarity among kinases, however, as should become clear in the discussion below. Most JNK and p38 pathways inhibitors fall into this group. A few inhibitors have been developed that inhibit protein kinases by noncompetitive mechanisms. Generally, these inhibitors are allosteric inhibitors that bind outside the ATP pocket to conformations other than the active one or bind in a mode that prevents transition to the active conformation. MEK1/2 inhibitors are good examples of this type of inhibition. Inhibitors that are not competitive with ATP may be more effective than those that are ATP competitive in the cellular milieu with ATP concentrations in the millimolar range. A common binding site for several allosteric inhibitors exists in protein kinases that have a conformational state in which the active site segment containing the conserved DFG (Asp-Phe-Gly) motif is moved out of the active site. This conformation is known as the DFG out state. The aspartate in this motif coordinates Mg bound to ATP and is important for positioning ATP for phosphoryl transfer. Inhibitors of the three major MAPK pathways will be discussed below.


ERK pathway inhibitors

Thus far, no potent ATP competitive inhibitors of ERK1/2 have been reported. Because of the likely cross-reactivity of ATP competitive inhibitors, compounds that inhibit through interactions outside the ATP binding pocket are attractive. Substrates bind to MAPKs on the opposite face from the active site in a region, which are sometimes called the common docking or CD site; this site can also influence kinase conformation (21). Small molecule inhibitors that bind to the CD site of ERK2 have been identified recently but have not received extensive testing (23).

Several inhibitors of upstream kinases Raf and MEK1/2 are used widely in elucidating the physiologic roles of the ERK pathways in a variety of biologic processes in cell culture systems (Figs. 1 and 2) (12, 13). The MEK1/2 inhibitors U0126, PD98059, and PD184352 (IC-1040) are noncompetitive with respect to both MEK substrates, ATP and ERK1/2, which is consistent with an allosteric mechanism of inhibition (24). They are effective largely by preventing the activation of MEK1/2, in addition to inhibiting MEK1/2 activity directly at higher concentrations. The recently solved crystal structure revealed that MEK1/2 have a unique inhibitor binding site located in an interior hydrophobic pocket near but not in the Mg-ATP-binding site. The binding of inhibitors induces several conformational changes in unphosphorylated MEK1/2 that may lock them into nonfunctional species (22). PD0325901 is a recently reported MEK inhibitor, which was modified from PD184352 with improvement of several pharmaceutical limitations. It has a 50-fold greater potency against MEK1 than PD184352, and suppresses ERK1/2 activity longer.

The catalytic domains of MEK5 and MEK1 have just less than 50% amino acid sequence identity and the MEK1/2 inhibitors PD98059, PD184352, and U0126 will inhibit the ERK5 pathway through effects on MEK5. Inhibition of MEK5 occurs at roughly 5-fold greater concentrations (7).

Recently, a benzimidazole derivative ARRY-142886 (AZD-6244) has been reported as a highly potent and selective inhibitor of MEK1/2. The IC50 was determined to be 14 nM against purified MEK1. This inhibitor is also not competitive with ATP, which is consistent with the high specificity of this compound for MEK1/2. It is currently in phase II clinical development (12). A novel compound XL518 was announced by Exelixis (south san Francisco, cA) as a potent and specific inhibitor of MEK1/2, which has highly optimized pharmacokinetic and pharmacodynamic properties.

BAY 43-9006 (Sorafenib; Bayer/Onyx Pharmaceuticals, Emeryville, cA) is a biaryl urea that inhibits Raf-1 kinase activity in vitro with ICs0 value of 6nM, and B-Raf with ICs0 value of 22 nM (25). This compound also inhibits some receptor tyrosine kinases, which include VEGF receptor family members, PDGF receptor, Flt, and c-Kit, at close to the same potency, and other protein Ser/Thr kinases including p38α, p38β, and RIP2 kinase (27, 14). The crystal structure of B-Raf with BAY 43-9006 shows that it binds in an allosteric site near to and partially overlapping the ATP pocket of B-Raf in a DFG out conformation, and it interacts with the residues of the kinase activation loop. The interaction prevents the activation loop and the catalytic residues from adopting a conformation that is competent to bind and phosphorylate substrates. Several other Raf inhibitors exist in different stages of clinical development including Chir-265, PLX4032, GW5074, and ZM336372 (26).




Figure 2


JNK pathway inhibitors

SP600125 and CEP-1347 (KT-7515) are the two most commonly used inhibitors to probe JNK pathways (13). Both are ATP competitive and have demonstrated efficacy for use in vivo, with the successful intervention to decrease brain damage (CEP-1347) or to ameliorate some symptoms of arthritis (SP600125) in animal models (9). Both SP600125 and CEP-1374 inhibit the JNK pathway; however, their targets are different. SP600125 is an anthrapyrazolone that inhibits JNK1, 2, and 3 directly. The IC50 values are 40 nM for JNK1/2, and 90 nM for JNK3. Testing against a broad panel of protein kinases has shown that SP-600125 inhibits at least 13 others with a similar potency as the JNKs, including p38 (27). Therefore, the activity of SP-600125 may not be attributed to the selective inhibition of JNKs. Despite this, SP-600125 has been useful to assess the role of JNK in cell culture and disease models, particularly in combination with p38 inhibitors described below. CEP-1347 inhibits MAP3Ks of the MLK group, which inhibits activation of the JNK pathway in those contexts in which MLKs are the MAP3Ks. The in vitro and cellular potencies are similar. Because of the role of JNK in diseases such as diabetes and obesity, it has been important to find additional novel small molecules that could inhibit the JNK pathway. At least two pan-JNK inhibitors have been reported. The Celgene (Summit, NJ) compound CC-401 has successfully completed a Phase I trial for acute myelogenous leukemia. The Merck compound AS602801 (Merck & Co., Whitehouse Station, NJ) may have therapeutic potential in multiple sclerosis and fibrosis.

JIP-1 is a JNK pathway scaffold protein essential for JNK activation in some systems. Like many substrates, JIP-1 binds to JNK in the region generally called the common docking site in MAPKs (28). With the identification of the key residues of JNK required for interaction with the kinase interaction motif of JIP-1, small peptide inhibitors (TI-JIP: RPKRPTTLNLF) derived from JIP have been described (29). As expected, TI-JIP was found to be competitive with respect to the phosphoacceptor substrate c-Jun and to exhibit noncompetitive inhibition with respect to ATP. It is not yet clear whether peptidomimetics or small molecule inhibitors that might have clinical applicability will be developed using information from JNK-JIP-1 interactions.


p38 pathway inhibitors

Because the activation of p38 plays essential roles in the biosynthesis and release of proinflammatory cytokines such as TNF-α and interleukin-1|8, blocking its activity may offer an effective therapy for treating many inflammatory diseases such as rheumatoid arthritis and inflammatory bowel disease (2, 5-7). Many p38 kinase inhibitors have been developed and have been evaluated extensively in preclinical models of arthritis (12, 13). Among them, SB203580, which is a pyridinyl imidazole compound, has been used as the template for many p38 inhibitors, and it has been extremely useful to delineate the function of p38. SB203580 inhibits the catalytic activity of p38 by binding to the ATP-binding site, but it does not prevent its activation by MAP2Ks, based on ATP competition experiments and crystallographic studies (30). SB203580 inhibits only p38α and β isoforms, but not γ or δ isoforms. VX-745 (Vertex-745; Vertex, Cambridge, MA), another p38α/β/γ inhibitor is a modified pyridinyl imidazole compound with better pharmacologic characteristics. Although it progressed to a Phase II trial for rheumatoid arthritis, additional development has ceased because it crossed into the central nervous system.

A diaryl urea compound BIRB796 is a potent and selective p38 allosteric inhibitor that bears little structural similarity to SB203580. Structural analysis shows that BIRB796 competes indirectly with the binding of ATP; binding requires a large conformational change not observed previously for any of the Ser/Thr protein kinases. BIRB796 interacts with a DFG out conformation of p38 in which the activation loop has been reorganized exposing a critical binding site; this structure incompatible with ATP binding (31, 32). The specific binding mode of BIRB796 suggests it will be a potent reagent in the treatment of chronic autoimmune diseases. In contrast to SB203580, BIRB796 inhibits not only p38α and p38β, but also the y and the 8 isoforms. Other p38 inhibitors in clinical phase II trials include VX-702 (Vertex), Scios 469 (Scios Inc., Mountain View, CA), and PH-797804 (Pfizer Inc., New York, NY). These compounds hold promise for the treatment of rheumatoid arthritis and cardiovascular diseases.

Tpl2 is a MAP3K that can lie upstream of all three MAPK pathways. It activates the translation of TNFα messenger RNA and TNFα production through activation of MAPK pathways. Therefore, Tpl2 inhibitors could be of value for treatment of certain inflammatory diseases. A series of 1,7-naphthyridine-3-carbonitriles and the related quinoline-3-carbonitrile (cyanoquinoline) have been found to inhibit Tpl2 activity and are promising for treatment of rheumatoid arthritis (33).


Key Experiments and Observations

X-ray crystallography is a powerful and direct tool to understand kinase inhibitory mechanisms. Key experiments demonstrate that inhibitors interact with multiple regions of kinases that lead to mechanisms that may or may not be competitive with substrates. As examples, p38 inhibitors fall into both groups.


Noncompetitive mechanism

All MEK1/2 inhibitors described above, U0126, PD098059, PD184352, and ARRY-142886 are noncompetitive inhibitors. From kinetic analysis, it was deduced that these inhibitors bind MEK1 outside of the substrate interaction sites (34). The inhibitor binding site was revealed once the three-dimensional structures of truncated MEK2 and MEK1 as ternary complexes with MgATP and analogs of PD184352 were solved (22). As shown by Ohren et al. (22), MgATP binds in a location comparable with that in other activated protein kinases, whereas the inhibitor binds in a binding pocket separate from but adjacent to the MgATP site. The crystal structure shows that in the presence of MgATP and PD184352-like compounds, the two folding domains of unphosphorylated MEK1 adopt the closed conformation typical of an activated protein kinase. This finding is accompanied by marked changes in the conformation of the kinase activation loop and helix C. The inhibition of the kinase activity of MEK1/2 by PD 184352-like compounds is the result of the stabilization of an inactive conformation of the activation loop and helix C and a deformation of the catalytic site. The presence of inhibitor in this site is thought to lock MEK1/2 into an inactive conformation that enables binding of ATP and substrate but disrupts both the molecular interactions required for catalysis and the proper access to the ERK activation loop.


Competitive mechanism

Most kinase inhibitors compete with ATP. SB203580 is pyridinyl imidazole with structure shown in Fig. 2c. Young et al. (35) first showed that SB203580 could bind to the inactive, unphosphorylated form of p38 with a Kd of about 40 nM. Binding in the active site was consistent with competitive inhibition of p38 by SB203580. They showed that SB203580 could also bind to activated, phosphorylated p38 and inhibit its activity. Inhibition was ATP competitive with a K of 21 nM. The Km ATP was increased from 200 μM to ~1400 mmol/L in the presence of 100 nM SB203580. Frantz et al. (36) compared the affinity of SB 203580 directly for both active and inactive p38 with a radioligand binding assay. They showed that the inhibitor could bind equally well to inactive and active p38, which demonstrates even more that binding of SB 203580 is independent of p38 phosphorylation state. More direct evidence came from the crystal structure of unphosphorylated p38 that showed a pyridinyl-imidazole in the ATP pocket (37).


Ongoing and Future Research Efforts

Selective kinase inhibitors are invaluable tools to dissect the physiologic and pathophysiologic roles of protein kinases, to identify new substrates, to identify model systems that allow evaluation of potential clinical use, and importantly, to develop new therapeutic agents.

Selectivity is critical for the use of protein kinase inhibitors in basic research. Protein kinases are a large class of enzymes related by sequence, all of which share the common capacity to bind ATP (1). Most inhibitors developed to date are competitive with ATP and bind in the active site. Thus, the potential for inhibitor cross-reactivity is great. Hence, establishing whether drug effects are caused by actions on the expected target or to previously unrecognized targets is an important step to validate their use.

We now understand that the entire kinase domain, not just the active site, may provide binding sites for inhibitory interactions. Inhibition may occur through selection of conformations that prevent a functional active state. Allosteric modifiers offer the possibility of high specificity and efficacy with reduced or different cross-reactivities relative to ATP competitive inhibitors. With more and more kinase structures solved, the inactive forms of kinases are becoming attractive targets for drug design, as the inactive kinases may have more distinct active sites compared with their active forms. Targeting the diverse inactive conformations might improve the inhibition specificity. Another approach may come from the development of nonphosphorylatable substrate analogs to interfere with the kinase-protein substrate association. These kinds of substrate-competitive inhibitors may also provide high specificity inhibitors with distinct cross-reactivities because of the varied binding modes of protein substrates.

Some MAPKs have several isoforms that may have different functions. Gene knockout studies have suggested isoform-specific roles. Thus, the development of inhibitors with the ability to interfere with one or a few isoforms as well as those that inhibit all isoforms will be desirable to understand the biology of these enzymes.



1. Manning G, Whyte DB, Martinez R, Hunter T, Sudarsanam S. The protein kinase complement of the human genome. Science 2002; 298:1912-1934.

2. Raman M, Chen W, Cobb MH. Differential regulation of MAPKs. Oncogene 2007; 26:3100-3112.

3. Yoon S, Seger R. The extracellular signal-regulated kinase: multiple substrates regulate diverse cellular functions. Growth Fact. 2006; 24:21-44.

4. Johnson GL, Dohlman HG, Graves LM. MAPK kinase kinases (MKKKs) as a target class for small-molecule inhibition to modulate signaling networks and gene expression. Curr. Opin. Chem. Biol. 2005; 9:325-331.

5. Kyriakis JM, Avruch J. Mammalian mitogen-activated protein kinase signal transduction pathways activated by stress and inflammation. Physiol. Rev. 2001; 81:807-869.

6. Lewis TS, Shapiro PS, Ahn NG. Signal transduction through MAP kinase cascades. Adv.Cancer Res. 1998; 74:49-139.

7. Chen Z, Gibson TB, Robinson F, Silvestro L, Pearson G, Xu B, Wright A, Vanderbilt C, Cobb MH. MAP kinases. Chem. Rev. 2001; 101:2449-2476.

8. Qi M, Elion EA. MAP kinase pathways. J. Cell. Sci. 2005; 118:3569-3572.

9. Bogoyevitch MA, Court N. Counting on mitogen-activated protein kinases-ERKs 3, 4, 5, 6, 7 and 8. Cell Signal. 2004; 16:1345-1354.

10. Dudley DT, Pang L, Decker SJ, Bridges AJ, Saltiel AR. A synthetic inhibitor of the mitogen-activated protein kinase cascade. Proc. Natl. Acad. Sci. U.S.A. 1995; 92:7686-7689.

11. Favata MF, Horiuchi KY, Manos EJ, Daulerio AJ, Stradley DA, Feeser WS, Van Dyk DE, Pitts WJ, Earl RA, Hobbs F, et al. Identification of a novel inhibitor of mitogen-activated protein kinase kinase. J. Biol. Chem. 1998; 273:18623-18632.

12. Sebolt-Leopold JS, English JM. Mechanisms of drug inhibition of signalling molecules. Nature 2006; 441:457-462.

13. English JM, Cobb MH. Pharmacological inhibitors of MAPK pathways. Trends Pharmacol. Sci. 2002; 23:40-45.

14. Fabian MA, Biggs WH III, Treiber DK, Atteridge CE, Azimioara MD, Benedetti MG, Carter TA, Ciceri P, Edeen PT, Floyd M, et al. A small molecule-kinase interaction map for clinical kinase inhibitors. Nat.Biotechnol. 2005; 23:329-336.

15. Wan PT, Garnett MJ, Roe SM, Lee S, Niculescu-Duvaz D, Good VM, Jones CM, Marshall CJ, Springer CJ, Barford D, Marais R. Mechanism of activation of the RAF-ERK signaling pathway by oncogenic mutations of B-RAF. Cell 2004; 116:855-867.

16. Nakamura K, Johnson GL. Noncanonical function of MEKK2 and MEK5 PB1 domains for coordinated extracellular signal-regulated kinase 5 and c-Jun N-terminal kinase signaling. Mol. Cell Biol. 2007; 27:4566-4577.

17. Gallo KA, Johnson GL. Mixed-lineage kinase control of JNK and p38 MAPK pathways. Nat. Rev. Mol. Cell Biol. 2002; 3:663-672.35.

18. Hirosumi J, Tuncman G, Chang L, Gorgun CZ, Uysal KT, Maeda K, Karin M, Hotamisligil GS. A central role for JNK in obesity and insulin resistance. Nature 2002; 420:333-336.

19. Ozcan U, Cao Q, Yilmaz E, Lee AH, Iwakoshi NN, Ozdelen E, Tuncman G, Gorgun C, Glimcher LH, Hotamisligil GS. Endoplasmic reticulum stress links obesity, insulin action, and type 2 diabetes. Science 2004; 306:457-461.

20. Taylor SS, Yang J, Wu J, Haste NM, Radzio-Andzelm E, Anand G. PKA: a portrait of protein kinase dynamics. Biochim. Biophys. Acta 2004; 1697:259-269.

21. Chang CI, Xu B, Akella R, Cobb MH, Goldsmith EJ. Crystal structures of MAP kinase p38 complexed to the docking sites on its nuclear substrate MEF2A and activator MKK3b. Mol. Cell 2002; 9:1241-1249.

22. Ohren JF, Chen H, Pavlovsky A, Whitehead C, Zhang E, Kuffa P, Yan C, McConnell P, Spessard C, Banotai C, Mueller WT, et al. Structures of human MAP kinase kinase 1 (MEK1) and MEK2 describe novel noncompetitive kinase inhibition. Nat. Struct. Mol. Biol 2004; 11:1192-1197.

23. Hancock CN, Macias A, Lee EK, Yu SY, Mackerell AD Jr, Shapiro P. Identification of novel extracellular signal-regulated kinase docking domain inhibitors. J. Med. Chem. 2005; 48:4586-4595.

24. Davies SP, Reddy H, Caivano M, Cohen P. Specificity and mechanism of action of some commonly used protein kinase inhibitors. Biochem. J. 2000; 351:95-105.

25. Adnane L, Trail PA, Taylor I, Wilhelm SM. Sorafenib (BAY 43-9006, Nexavar((R))), a dual-action inhibitor that targets RAF/ MEK/ERK pathway in tumor cells and tyrosine kinases VEGFR/ PDGFR in tumor vasculature. Meth.Enzymol. 2005; 407:597-612.

26. Wang JY, Wilcoxen KM, Nomoto K, Wu S. Recent advances of MEK inhibitors and their clinical progress. Curr.Top.Med.Chem 2007; 7:1364-1378.

27. Bain J, McLauchlan H, Elliott M, Cohen P. The specificities of protein kinase inhibitors: an update. Biochem.J. 2003; 371:199-204.

28. Davis RJ. Signal transduction by the JNK group of MAP kinases. Cell 2000; 103:239-252.

29. Bonny C, Oberson A, Negri S, Sauser C, Schorderet DF. Cell-permeable peptide inhibitors of JNK: novel blockers of beta-cell death. Diabetes 2001; 50:77-82.

30. Wang Z, Canagarajah BJ, Boehm JC, Kassisa S, Cobb MH, Young PR, Abdel-Meguid S, Adams JL, Goldsmith EJ. Structural basis of inhibitor selectivity in MAP kinases. Structure 1998; 6:1117-1128.

31. Pargellis C, Tong L, Churchill L, Cirillo PF, Gilmore T, Graham AG, Grob PM, Hickey ER, Moss N, Pav S, Regan J. Inhibition of p38 MAP kinase by utilizing a novel allosteric binding site. Nat. Struct. Biol. 2002; 9:268-272.

32. Liu Y, Gray NS. Rational design of inhibitors that bind to inactive kinase conformations. Nat.Chem Biol 2006; 2:358-364.

33. Hu Y, Green N, Gavrin LK, Janz K, Kaila N, Li HQ, Thomason JR, Cuozzo JW, Hall JP, Hsu S, et al. Inhibition of Tpl2 kinase and TNFalpha production with quinoline-3-carbonitriles for the treatment of rheumatoid arthritis. Bioorg. Med. Chem Lett. 2006; 16:6067-6072.

34. Alessi DR, Cuenda A, Cohen P, Dudley DT, Saltiel AR. PD098059 is a specific inhibitor of the activation of mitogen-activated protein kinase kinase in vitro and in vivo. J. Biol. Chem. 1995; 270:27489-27494.

35. Young PR, McLaughlin MM, Kumar S, Kassis S, Doyle ML, McNulty D, Gallagher TF, Fisher S, Mcdonnell PC, Carr SA, et al. Pyridinyl imidazole inhibitors of p38 mitogen-activated protein kinase bind in the ATP site. J. Biol. Chem. 1997; 272:12116-12121.

36. Frantz B, Klatt T, Pang M, Parsons J, Rolando A, Williams H, Tocci MJ, O’Keefe SJ, O’Neill EA. The activation state of p38 mitogen-activated protein kinase determines the efficiency of ATP competition for pyridinylimidazole inhibitor binding. Biochemistry 1998; 37:13846-13853.

37. Tong L, Pav S, White DM, Rogers S, Crane KM, Cywin CL, Brown ML, Pargellis CA. A highly specific inhibitor of human p38 MAP kinase binds in the ATP pocket. Nat. Struct. Biol. 1997; 4:311-316.