Activation and function of the MAPKs and their substrates, the MAPK-activated protein kinases

Abstract

The mitogen-activated protein kinases (MAPKs) regulate diverse cellular programs by relaying extracellular signals to intracellular responses. In mammals, there are more than a dozen MAPK enzymes that coordinately regulate cell proliferation, differentiation, motility, and survival. The best known are the conventional MAPKs, which include the extracellular signal-regulated kinases 1 and 2 (ERK1/2), c-Jun amino-terminal kinases 1 to 3 (JNK1 to -3), p38 (α, β, γ, and δ), and ERK5 families. There are additional, atypical MAPK enzymes, including ERK3/4, ERK7/8, and Nemo-like kinase (NLK), which have distinct regulation and functions. Together, the MAPKs regulate a large number of substrates, including members of a family of protein Ser/Thr kinases termed MAPK-activated protein kinases (MAPKAPKs). The MAPKAPKs are related enzymes that respond to extracellular stimulation through direct MAPK-dependent activation loop phosphorylation and kinase activation. There are five MAPKAPK subfamilies: the p90 ribosomal S6 kinase (RSK), the mitogen- and stress-activated kinase (MSK), the MAPK-interacting kinase (MNK), the MAPK-activated protein kinase 2/3 (MK2/3), and MK5 (also known as p38-regulated/activated protein kinase [PRAK]). These enzymes have diverse biological functions, including regulation of nucleosome and gene expression, mRNA stability and translation, and cell proliferation and survival. Here we review the mechanisms of MAPKAPK activation by the different MAPKs and discuss their physiological roles based on established substrates and recent discoveries.

FIG. 1.

Schematic representation of the overall structures of conventional and atypical MAPKs. All MAPKs contain a Ser/Thr kinase domain flanked by N- and C-terminal regions of different lengths. Different additional domains are also present in some MAPKs, including a transactivation domain (TAD), a nuclear localization sequence (NLS), a region conserved in ERK3 and ERK4 (C34), and a domain rich in Ala, His, and Glu (AHQr). The respective MAPKAPKs activated by the MAPKs are show in light gray next to the cognate MAPK. The γ and δ isoforms of p38 are in parentheses to indicate that they have not been shown to promote MAPKAPK activation.

FIG. 2.

MAPK signaling cascades leading to activation of the MAPKAPKs. Mitogens, cytokines, and cellular stresses promote the activation of different MAPK pathways, which in turn phosphorylate and activate the five subgroups of MAPKAPKs, including RSK, MSK, MNK, MK2/3, and MK5. Dotted lines indicate that, although reported, substrate regulation by the respective kinase remains to be thoroughly demonstrated. The γ and δ isoforms of p38 are in parentheses to indicate that they have not been shown to promote MAPKAPK activation.

FIG. 3.

Alignment of MAPK-binding sequences from MAPKAPKs. MAPKs bind regions termed D domains that are found in all MAPKAPKs and required for efficient activation. The D domains, shown in boldface, are characterized by a stretch of positively charged residues surrounded by hydrophobic residues. Some D domains mediate the specific interaction with ERK1/2 or p38, while others are necessary for the interaction with both upstream activators. In some MAPKAPKs, the D domain region overlaps with an NLS (underlined). ERK3/4 binding to MK5 requires a degenerate D domain in the C-terminal region of the protein. MAPK-binding sequences in MAPKAPKs do not fit the optimal consensus sequence for D domains but rather fit a D domain-like sequence termed the kinase-interacting motif (KIM). Similar alignments were previously demonstrated in recent review articles (adapted from references and with permission).

FIG. 4.

Schematic representation of the overall structure of the MAPKAPKs. While RSK1/2/3/4 and MSK1/2 are composed of two nonidentical kinase domains, the MNKs and MK2/3/5 are single-headed kinases that display homology to the CTKD of RSKs and MSKs. The NTKDs of the RSKs and MSKs are members of the AGC family of kinases, which also includes Akt, PKA, and PKC. The CTKDs of RSKs and MSKs and the kinase domains of MK2/3/5 belong to the CAMK family of kinases, which also include AMPK, DAPK, and CAMK1/2. The amino acid composition of each MAPKAPK along with phosphorylation site numbering refers to the human nomenclature. Only the full-length form of each MAPKAPK protein is included in this diagram. NES, nuclear export signal; NLS, nuclear localization signal; PBR, polybasic region; SH3, Src homology 3 domain; X, nonfunctional NES; D, D domain or MAPK docking site; ERK3/4, docking region for ERK3/4; NTKD, N-terminal kinase domain; CTKD, C-terminal kinase domain.

FIG. 5.

Sequence analysis of the different MAPKAPKs. (A) Alignment of sequences around activation loops of MAPKAPKs reveal conservation of the MAPK phosphoacceptor residue followed by a Pro. In the case of RSK and MSK, activation loop sequences are from their respective C-terminal kinase domains. (B) Phylogenetic tree of MAPKAPK family members. All sequences used for the construction of this tree were human. The relative similarities between all MAPKAPKs reflected by this tree suggest that the MAPKAPKs are comprised within five groups, the RSK, MSK, MNK, MK2/3, and MK5 subfamilies. The CLUSTAL X program was used to generate the multiple alignments on which the tree was based.

FIG. 6.

Alignment of the amino acid sequence of the MAPKAPKs containing two kinase domains. Sequences comprising the kinase domains and its subregions are boxed in red and reveal regions of highest homology. The conserved activation loop threonine residue is shown, as well as other conserved phosphorylation sites. The MAPK-binding domain is identified by a line.

FIG. 7.

Signaling cascades leading to activation of the RSKs, MSKs, and MNKs. RSK1/2/3/4 are activated by two inputs originating from ERK1/2 or ERK5, as well as PDK1 enzymes. Similarly, MSK1/2 are activated by ERK1/2 but also by stimuli that activate the p38 module. Depending on the isoforms, MNKs can be stimulated by either ERK1/2 or both ERK1/2 and p38. Different inhibitors of components within these cascades are also shown. Dotted lines indicate that, although reported, substrate regulation by the respective kinase remains to be thoroughly demonstrated.

FIG. 8.

Biological functions and substrates of the MAPKAPKs. Upon activation, the RSKs, MSKs, MNKs, and MK2/3/5 phosphorylate several substrates and regulate many biological responses. The list of substrates indicated in this figure is not exhaustive but emphasizes the many important substrates identified to date for the different MAPKAPKs.

FIG. 9.

Alignment of the amino acid sequence of the MAPKAPKs containing a single kinase domain. Sequences comprising the kinase domains and its subregions are boxed in red and reveal regions of highest homology. The conserved activation loop threonine residue is shown, as well as other conserved phosphorylation sites. The MAPK-binding domain is identified by a line.

FIG. 10.

Signaling cascades leading to activation of MK2/3 and MK5. MK2/3 have been shown to be activated by both ERK1/2 and p38 kinases. Conversely, MK5 was initially shown to be regulated by p38, but recent data suggest a stronger link with ERK3 and ERK4. Different inhibitors of components within these cascades are also shown. Dotted lines indicate that, although reported, substrate regulation by the respective kinase remains to be thoroughly demonstrated.