MAP Kinase-Activated Protein Kinase 5 (MK5)
Murine and human MK5 cDNAs were initially isolated in 1998 in two independent screens for proteins with sequence homology to MK2 (New et al. 1998; Ni et al. 1998). The novel 54-kDa kinase ubiquitously expressed in all tissues displayed 45% amino acid identity to MK2. Both groups showed that MK5 could be phosphorylated and activated in vitro by the p38 MAP kinase, as detected by 32P incorporation into a substrate peptide (KKRPQRATSNVFS) or Hsp25 (HSPB1). New et al. named this kinase as p38-regulated and activated kinase or PRAK to emphasize its integration into the p38 pathway. More recently, MK5 has also been shown to interact with the atypical MAP kinases, ERK3 (MAPK6) and ERK4 (MAPK4), and these kinases are also involved in the phosphorylation and activation of MK5 (Schumacher et al. 2004; Seternes et al. 2004; Aberg et al. 2006; Kant et al. 2006). An MK5 gene does not appear to be present in either C. elegans or Drosophila, but orthologs are found in most vertebrates.
Structure, Activation, and Expression
T182 located in the activation loop LXTP motif of MK5 is the most well characterized activating phosphorylation site on MK5. While MK2/3 requires phosphorylation at additional regulatory MAPK sites at the C-terminus for full activation, there are no such characterized sites on MK5. A recent large-scale screen identified strong CXCL12-induced phosphorylation of MK5 at T368, but neither the significance of this phosphorylation on MK5 activity nor the upstream kinase involved has been analyzed so far (Yi et al. 2014). Interestingly K364-acetylation of MK5 was shown to enhance its activity (Zheng et al. 2013). p38 can directly phosphorylate T182 and activate MK5 akin to MK2 (New et al. 1998; Seternes et al. 2002) at least in vitro and upon overexpression. Additionally, overexpressed p38 can bind MK5 leading to cytoplasmic relocalization of the resulting complex. However, endogenous MK5 is not significantly activated by classical p38 stimuli such as arsenite and sorbitol (Shi et al. 2003). Moreover, the phenotype of MK5-deficient mice does not resemble one of MK2/3-deficient animals, displaying a normal profile of cytokine production and no increased resistance to LPS challenge (Shi et al. 2003). These observations argue against the physiological relevance of the p38-MK5 signaling axis.
MK5 was shown to interact with atypical MAPKs, ERK3 and ERK4. These kinases form a tight complex with MK5 resulting in mutual stabilization and phosphorylation. Moreover, the coexpression of ERK3/4 not only leads to phosphorylation and activation of MK5 but also cause the translocation of ERK3–MK5 from the nucleus to the cytoplasm (Schumacher et al. 2004; Seternes et al. 2004). However, the physiological conditions, leading to the activation of MK5 by ERK3 and ERK4 remain unknown. Recent studies have identified p21-activated kinases (PAKs) as activators of ERK3/4, suggesting that the cytoskeletal small-GTPases which regulate PAKs could be involved in the activation of ERK3/4 signaling (De la Mota-Peynado et al. 2011; Deleris et al. 2011). In addition to the atypical MAPKs and p38, Protein kinase A (PKA) is another kinase which has been shown to influence the cellular distribution of MK5 (Gerits et al. 2007a). Treatment of cells with forskolin or overexpression of a nuclear-targeted PKAc-α (catalytic subunit) was shown to induce the transient redistribution of MK5 from nucleus to cytoplasm in rat PC12 cells. Interestingly, PKA modulates MK5 functions independent of T182 by phosphorylating it at S115 (Kostenko et al. 2011).
The optimal phosphorylation site motif for MK5 is very similar to MK2/3 and has been defined as (L,M,F,Y,W)-X-R-(Q,M,S)-X-(pS,pT)-X (Ronkina et al. 2015). In vitro, MK5 is able to phosphorylate HSPB1, glycogen synthase, tyrosine hydroxylase (preferentially at S19), and myosin heavy chain (New et al. 1998; Ni et al. 1998; Toska et al. 2002; Gaestel 2006) at the same sites as MK2. Additional in vitro substrates of MK5 include SEPT8 (Shiryaev et al. 2012), BORG (binder of Rho-GTPase) proteins (CDC42EP3 & CDC42EP5), KAL7 (Brand et al. 2012), DNAJB1 (Kostenko et al. 2014), FAK (Dwyer and Gelman 2014), and DJ-1 (Tang et al. 2014).
While cAMP/PKA-induced HSPB1- phosphorylation and F-actin remodeling were shown to be MK5-dependent (Kostenko et al. 2011), stress-dependent phosphorylation of HSPB1 is not impaired in the MK5-KO mice (Shi et al. 2003). A stimulus-specific role for MK5 in HSPB1-phosphorylation is yet to be verified in knockout models. MK5 phosphorylates its interacting partners, atypical MAPKs ERK3 (Schumacher et al. 2004; Seternes et al. 2004) as well as homologous ERK4 (Aberg et al. 2006; Kant et al. 2006). RHEB phosphorylation at S130 was shown to be mediated by MK5 (Zheng et al. 2011) and MK5 was also reported to be the kinase responsible for S37 phosphorylation of p53, downstream to oncogenic Ras-p38 signaling (Sun et al. 2007). In addition, independent studies have shown a role for MK5 in the phosphorylation of the fork-head family transcription factors FOXO1 and FOXO3a at a conserved residue (S215) in the DNA-binding domain (Kress et al. 2011; Chow et al. 2013).
The exact biological function of MK5 is largely unknown. Originally, due to the high structural similarity and substrate consensus, MK5 was expected to be functionally similar to MK2, which is involved in stress response and inflammation. Generation of the MK5-deficient mice challenged this assumption (Shi et al. 2003). Indeed, MK5-deficient mice do not display any of the phenotypic changes characteristic of MK2-deficient animals. Disruption of the MK5 gene in mice on mixed genetic background does not manifest any phenotypical features. MK5 knockout animals backcrossed to C57Bl/6 genetic background resulted in lethality at E11.5 with incomplete penetrance, indicating a role for MK5 in embryogenesis (Schumacher et al. 2004). In studies exploring the Ras-induced senescence, using a second knockout model, MK5 was proposed to be a tumor suppressor with active role in senescence (Sun et al. 2007). In the same study, the authors demonstrated that MK5 phosphorylates p53 at S37, a residue located in the transactivation domain. However, this residue does not lie within the consensus phosphorylation motif for MK5. In addition, a recent report comparing the two different knockout models of MK5 present strong evidence against this proposed role for MK5 as a tumor suppressor in regulating senescence (Ronkina et al. 2015). However, MK5 acts as a tumor suppressor in colon cancer by inhibiting the translation of tumor promoting c-MYC. In this system, MK5 was shown to phosphorylate and activate FOXO3a inducing the expression of microRNA miR-34b/c, which in turn suppresses c-MYC translation (Kress et al. 2011). In addition, there is increasing evidence for the involvement of ERK3 in tumorigenesis and metastasis which makes MK5 a key player in these processes as a modulator of stability, activation, and subcellular localization of ERK3.
Overexpression of MK5 in HeLa cells leads to an increase in both F-actin content and cell migration, an effect which was shown to be counteracted by 14–3-3 epsilon (Tak et al. 2007). MK5 interaction with 14–3-3 epsilon was shown to decrease its kinase activity towards HSPB1. In a contrasting report, MK5-overexpression was shown to suppress cell migration and mutual phosphorylation by Focal Adhesion Kinase (FAK) and MK5 was proposed to regulate FAK-activation and cell migration (Dwyer and Gelman 2014). Another study utilized siRNA-mediated knockdown of MK5 in PC12-cells to show that MK5 is necessary for the forskolin-induced transient increase in F-actin levels (Gerits et al. 2007a). How MK5 mediates cytoskeletal rearrangements and cell migration is still unclear, and further studies are needed to address this.
MK5 is a MAPK-activated kinase integrated into the p38 and ERK3/ERK4 signaling pathways. p38 was initially described as the major MAPK activating MK5, resulting in the name PRAK, which is an acronym for p38 regulated and activated kinase. However, p38 activating stimuli do not usually cause MK5 phosphorylation and activation. Hence, the significance of p38 in MK5 signaling is restricted and needs to be directly addressed in future studies. In contrast, the involvement of ERK3 and ERK4 in MK5 functions is clearly documented in in vivo studies and supported by strong interaction between these proteins, mutual stabilization, and similar pattern of expression in embryogenesis. However, the ERK3/4- MK5 pathway still remains an “orphan” signaling module due to lack of information regarding physiological stimuli as well as regulatory mechanisms. While MK5 seems to have some impact on tumorigenesis, actin remodeling, and cell migration, ERK3-MK5-SEPT7-mediated neuronal morphogenesis seems to be the sole established physiological function of this cascade. Future studies in MK5 knockout animals and dissection of signaling events using genetic approaches involving the p38, ERK3/4, and MK2/3 knockout models and MK5 inhibitors are necessary to further characterize and understand MK5 signaling.
- Brand F, Schumacher S, Kant S, Menon MB, Simon R, Turgeon B, et al. The extracellular signal-regulated kinase 3 (mitogen-activated protein kinase 6 [MAPK6])-MAPK-activated protein kinase 5 signaling complex regulates septin function and dendrite morphology. Mol Cell Biol. 2012;32:2467–78. doi: 10.1128/MCB.06633-11.PubMedPubMedCentralCrossRefGoogle Scholar
- Deleris P, Trost M, Topisirovic I, Tanguay PL, Borden KL, Thibault P, et al. Activation loop phosphorylation of ERK3/ERK4 by group I p21-activated kinases (PAKs) defines a novel PAK-ERK3/4-MAPK-activated protein kinase 5 signaling pathway. J Biol Chem. 2011;286:6470–8. doi: 10.1074/jbc.M110.181529.PubMedCrossRefGoogle Scholar
- Dwyer SF, Gelman IH. Cross-phosphorylation and interaction between Src/FAK and MAPKAP5/PRAK in early focal adhesions controls cell motility. J Cancer Biol Res. 2014;2(1):1045.Google Scholar
- Gerits N, Mikalsen T, Kostenko S, Shiryaev A, Johannessen M, Moens U. Modulation of F-actin rearrangement by the cyclic AMP/cAMP-dependent protein kinase (PKA) pathway is mediated by MAPK-activated protein kinase 5 and requires PKA-induced nuclear export of MK5. J Biol Chem. 2007a;282:37232–43. doi: 10.1074/jbc.M704873200.PubMedCrossRefGoogle Scholar
- Kostenko S, Shiryaev A, Gerits N, Dumitriu G, Klenow H, Johannessen M, et al. Serine residue 115 of MAPK-activated protein kinase MK5 is crucial for its PKA-regulated nuclear export and biological function. Cell Mol Life Sci. 2011;68:847–62. doi: 10.1007/s00018-010-0496-2.PubMedCrossRefGoogle Scholar
- Seternes OM, Johansen B, Hegge B, Johannessen M, Keyse SM, Moens U. Both binding and activation of p38 mitogen-activated protein kinase (MAPK) play essential roles in regulation of the nucleocytoplasmic distribution of MAPK-activated protein kinase 5 by cellular stress. Mol Cell Biol. 2002;22:6931–45.PubMedPubMedCentralCrossRefGoogle Scholar
- Yi T, Zhai B, Yu Y, Kiyotsugu Y, Raschle T, Etzkorn M, et al. Quantitative phosphoproteomic analysis reveals system-wide signaling pathways downstream of SDF-1/CXCR4 in breast cancer stem cells. Proc Natl Acad Sci USA. 2014;111:E2182–90. doi: 10.1073/pnas.1404943111.PubMedPubMedCentralCrossRefGoogle Scholar