Polymyxin B sensitivity (PBS) 2, a yeast homolog of mitogen-activated protein kinase kinase 3 (Mek3), was originally cloned (in 1987) as a gene that conferred polymyxin B resistance to yeast cells (Boguslawski and Polazzi 1987). The amino acid sequence of the PBS2 gene product showed strong homology to the serine/threonine protein kinase family (Boguslawski and Polazzi 1987). In 1993, it was shown that PBS2 and its downstream HOG1 genes, which code for a Mek3 homologue and a p38 mitogen-activated protein kinase (MAPK) homologue, respectively, are necessary for yeast cells to grow at high osmolarity (Brewster et al. 1993). Two years after that study, Mek3 was first amplified by degenerative PCR as a human homolog of yeast PBS2, and overexpression of its gene product was revealed to activate p38 MAPK in response to osmotic stress, UV irradiation, and inflammatory cytokines (interleukin (IL)-1 and tumor necrosis factor (TNF)) in COS-1 cells (Derijard et al. 1995).
Protein and Gene Structure
The human Mek3 gene has been mapped to 17q11.2 (Derijard et al. 1995). The Mek3 gene has 12 exons and 11 introns, and its coding region contains 957 base pairs organized in nine exons (exons 2–10).
The human Mek protein family contains seven kinases, Mek 1 to 7. Each member of the Mek family contains the Mek-kinase subdomains I–XI (Hanks et al. 1988) (Fig. 1b). Mek3, Mek4, and Mek6 phosphorylate p38 MAPK, whereas Mek1 and Mek2 phosphorylate extracellular signal-regulated kinase 1 (Erk1) and 1 (Erk2). Mek4 and Mek7 phosphorylate c-Jun N-terminal kinase (Jnk (1–3)). Mek5 is the only identified kinase upstream of Erk5.
Mek3 is approximately 50% and 80% homologous to Mek4 and Mek6, respectively (Derijard et al. 1995; Stein et al. 1996). In contrast to Mek1 and Mek2, all p38 MAPK stimulators include a large deletion that is located between subdomains IX and X (Fig. 1b). Both Mek1 and Mek2 contain a proline-rich region between these subdomains, which is required for the efficient activation of downstream Erk1 and Erk2 (Dang et al. 1998). Thus, the deletion could define the kinase substrate specificity of Mek3. A Mek3-specific inhibitor has not yet been developed. UO126 is shown to weakly attenuate Mek3 activity. However, it preferentially suppresses Mek1 and Mek2.
Mek3b is an alternatively spliced form of Mek3. Mek3b is distinct from Mek3 at the 5′-end, encoding 29 extra amino acids at the N-terminus. Reminiscent of Mek3 and Mek6, Mek3b is activated by osmotic shock and phosphorylates p38 when expressed in cultured cells. Both Mek3 and Mek3b activate p38α, whereas only Mek3b phosphorylates p38β2, suggesting a role for the N-terminal region of Mek3 in the determination of substrate specificity (Enslen et al. 2000).
Interaction and Regulation
Mek3 activity is downregulated by phosphatases such as type 2 protein phosphatase (PP2A) (Prickett and Brautigan 2007). Phosphorylated Mek3 binds to alpha-4/immunoglobulin-binding protein 1 (Igbp1) and PP2A, which then enhances the site-specific dephosphorylation of Thr193 (but not Ser189) in the activation loop of Mek3 (Fig. 3). Moreover, Ser189 and Thr193 may be modified by an acetyltransferase, Yersinia outer protein (Yop) J (Fig. 3) (Mukherjee et al. 2006). This acetylation could consequently block phosphorylation of these residues, leading to the attenuation of Mek3 activity (Mukherjee et al. 2006). These results indicate that the level of Mek3 kinase activity is dependent on the phosphorylation, dephosphorylation, or acetylation state of its activation loop.
Mek3 also binds to dynactin1 and microtubules. Dynactin1 binds to and cooperates with a microtubule-dependent motor protein, dynein, and this protein complex induces the movement of various cellular cargos, especially membranous vesicles. The disruption of dynactin1 or treatment with microtubule-disrupting drugs suppresses Mek3 and Mek6 phosphorylation in response to hyperosmolar stresses, suggesting that the function of the dynein-dynactin complex may be to partly regulate the Mek3-p38 MAPK signaling cascade (Fig. 4b).
Apart from the interaction mentioned above, Mek3 is known to associate with several proteins including the osmosensing scaffold for Mekk (Osm), JNK-interacting protein 2 (Jip2), Smad7, c-Src, c-Met, phospholipase C (PLC)-β2, and Rho effector protein kinase N1 (Pkn1) (Yasuda et al. 2009). Such Mek3-interacting proteins are involved in various cellular functions. Details of Osm, Smad7, c-Src, and c-Met are described in the Protein Function section.
Whereas Mek6 activates all p38 isoforms, Mek3 is somewhat selective in that it preferentially phosphorylates the α, γ, and δ subtypes of p38 MAPK at its regulatory Thr and Tyr residues (Cuadrado and Nebreda 2010). The activation of Mek3 and p38 MAPK is associated with several cellular functions, such as cytoskeletal protein remodeling, cytokine expression, apoptosis, and clathrin-mediated endocytosis.
Regulation of Cytoskeletal Proteins
Mek3 also regulates actin polymerization by binding to c-Met (a hepatocyte growth factor/scatter factor (HGF/SF) receptor) and its associated cytoplasmic binding protein kinase c-Src (a member of the cytoplasmic tyrosine kinase family). In this pathway, c-Met stimulates c-Src and its downstream molecule Rac1 that activates the Mek3-p38 MAPK signaling, leading to actin polymerization (Fig. 5a). The active forms of Rac, Mek3, and p38 MAPK are detected in signet-ring carcinoma cell lines (Kobayashi et al. 1999).
How does the activation of the Rac/Mek3/p38 MAPK pathway affect cytoskeletal reorganization? Hitherto, the p38 MAPK substrate, Mapkap kinase 2 (MK2), has been revealed to be involved in this mechanism. MK2 phosphorylates its downstream heat shock protein HSP27, which acts as an actin-capping protein. Phospho-HSP27 causes the release of free actin filament barbed ends, which may result in actin polymerization (Fig. 5a) (Cuadrado and Nebreda 2010). Thus, Mek3 might control cell migration, cell adhesion, or cell shape, by influencing cytoskeletal proteins under pathological conditions.
The stress-mediated activation of Mek3 positively regulates the expression of several cytokines such as interleukin (IL)-1α, IL-1β, IL-6, and IL-12 (Cuadrado and Nebreda 2010). Mek3 knockdown shows selective defects in the response of fibroblasts to tumor necrosis factor (TNF)-α, including reduced p38 MAPK activation and cytokine expression (Cuadrado and Nebreda 2010). In addition, Mek3 mediates lipopolysaccharide (LPS)-triggered post-transcriptional control of cytokine and chemokine expression. LPS stimulation induces the formation of a complex consisting of Irak2 (a member of the IL-1 receptor-associated kinase (Irak) protein family), Traf6, MK2, and Mek3, and this interaction appears to regulate Toll-like receptor-mediated signaling (Fig. 5b). Elevated Mek3, p38, and MK2 activities cause the phosphorylation of its downstream mRNA destabilizing protein, tristetraprolin (TTP), which binds to the AU-rich element (ARE) of cytokine mRNA in the 3′-noncoding region. The phosphorylation of TTP leads to a reduction in the affinity to ARE and consequently stabilizes cytokine and chemokine mRNAs, such as IL-6 and chemokine (C-X-C Motif) ligand 1 (CXCL1) (Fig. 5b) (Hitti et al. 2006). These findings indicate that Mek3 might contribute to the stabilization of cytokine mRNAs after exposure to various forms of stress.
Activation of Mek3 also regulates the induction of apoptosis. That is, the Tak1-Mek3-p38 MAPK signaling complex mediates transforming growth factor (TGF) β1-induced apoptosis. In addition, Smad7 might facilitate this TGF-β-induced apoptosis as a scaffold protein by binding to Tak1, Mek3, and p38 and enhancing each interaction. As a downstream target of p38 MAPK, glycogen synthase kinase (GSK)-3β is phosphorylated and inactivated. The inactive form of GSK-3β causes dephosphorylation and degradation of β-catenin, which binds to a transcriptional regulator, lymphoid enhancer binding factor 1 (LEF1)/T-cell-specific factor (TSF), and may increase the expression of some apoptotic inducers, such as c-myc (Fig. 5c) (Edlund et al. 2005). These results suggest that apoptosis may be regulated by Mek3 kinase activity.
Autophagy and Cell Proliferation
Mek3 may regulate autophagy and cell proliferation. A PB1-domain-containing substrate p62, also known as sequestosome 1 (SQSTM1), is frequently overexpressed in human carcinomas and may be involved in tumorigenesis. In response to amino acids, p62 is phosphorylated by a cascade including PB1-containing kinase Mekk3, Mek3, and p38δ (Fig. 5d) (Linares et al. 2015). Phosphorylated p62 recruits tumor necrosis factor receptor-associated factor (TRAF)6 and promotes translocation of mammalian target of rapamycin complex (mTORC1) to lysosomes. mTORC1 is subsequently activated by K63 polyubiquitination and regulates cell growth and proliferation (Fig. 5d). Thus, a signaling cascade including Mek3 is critical for mTORC1 activation.
Activation of Mek3 is related to the mechanism of clathrin-mediated endocytosis. For instance, the endocytosis of N-cadherin is regulated by the Mek3-p38 MAPK activity in neurons. The cytoplasmic region of a neuronal activity-inducible protein, arcadlin, binds to the TAO2β/Mek3/p38 signalosome and initiates N-cadherin endocytosis, causing neuronal activity-dependent dendritic spine retraction in hippocampal neurons (Fig. 5e) (Yasuda et al. 2007). This is the only situation in which a specific external stimulation regulates the activity of Mek3. In addition, the activation of a signaling cascade containing Mek3 and p38 may also cause α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor endocytosis in the dendritic spines of hippocampal neurons, suggesting that Mek3 may play a pivotal role in endocytosis in CNS neurons.
Mek3-deficient mice appear to be viable and fertile (Cuadrado and Nebreda 2010). By contrast, the Mek3/Mek6-double knockout (KO) is lethal during midgestation at embryonic day 11.0–11.5 because of major defects during the development of the embryonic vasculature. The double KO mice also show developmental delay (Cuadrado and Nebreda 2010). This phenotype exhibited by Mek3/Mek6 double KO mice is similar to that previously described for p38α MAPK KO embryos (Cuadrado and Nebreda 2010). Such a phenotypic difference between single and double KO mice suggests that some compensatory mechanisms can maintain p38 MAPK signaling. In fact, the amount of Mek6 protein is substantially increased in the unilateral ureteric obstruction kidney of Mek3-deficient mice (Ma et al. 2007).
Mek3-deficient mice show several phenotypes, including a reduction of cytokine production and impairment of immune responses (Cuadrado and Nebreda 2010). Additionally, Mek3 KO mice exhibit a reduced progression of arthritis and type 1 diabetes (Fukuda et al. 2008; Inoue et al. 2006). Furthermore, mice with Mek3 deletions show a reduction of bone mass secondary to defective osteoblast differentiation (Greenblatt et al. 2010). Irrespective of the functional significance of Mek3, such minor phenotypes of its deficiency suggest the existence of possible redundant or compensatory mechanisms by Mek6 in vivo.
Mek3 is phosphorylated by various MAPKKKs, which become activated in response to various physical and chemical stresses, such as oxidative stress, UV irradiation, hypoxia, ischemia, and the presence of various cytokines. Activated Mek3 then specifically phosphorylates p38α, δ, and γ MAPK. Conversely, Mek3 kinase activity is downregulated by dephosphorylation or acetylation of the Ser and Thr residues in its activation loop. Although the crystal structure of Mek3 has not yet been determined, the primary structure indicates that it consists of Mek-kinase subdomains I-XI. An interesting structural characteristic of Mek3 is that it has a deletion between subdomain IX and X that could define the kinase substrate specificity for p38 MAPK. Mek3 converts diverse cellular stimuli into various stress responses, such as cytokine expression, cytoskeletal protein remodeling, apoptosis, autophagy, and clathrin-mediated endocytosis. Further studies may be required to clarify the regulatory mechanism of Mek3 in response to various external stresses and the p38 MAPK activation machinery. The development of a Mek3-specific inhibitor is also needed to evaluate its pathophysiological significance of Mek3.
This work was partly supported by KAKENHIs (20591426 to SY, 21500332 to HS, and 20300135 to KY) and the Naito Foundation (to KY).
- Hitti E, Iakovleva T, Brook M, Deppenmeier S, Gruber AD, Radzioch D, Clark AR, Blackshear PJ, Kotlyarov A, Gaestel M. Mitogen-activated protein kinase-activated protein kinase 2 regulates tumor necrosis factor mRNA stability and translation mainly by altering tristetraprolin expression, stability, and binding to adenine/uridine-rich element. Mol Cell Biol. 2006;26:2399–407.PubMedPubMedCentralCrossRefGoogle Scholar
- Yasuda S, Tanaka H, Sugiura H, Okamura K, Sakaguchi T, Tran U, Takemiya T, Mizoguchi A, Yagita Y, Sakurai T, et al. Activity-induced protocadherin arcadlin regulates dendritic spine number by triggering N-cadherin endocytosis via TAO2beta and p38 MAP kinases. Neuron. 2007;56:456–71.PubMedPubMedCentralCrossRefGoogle Scholar
- Yasuda S, Sugiura H, Yamagata K. Mek3. UCSD-Nature Molecule Pages. 2009. doi: 10.1038/mp.a001507.01