Dual-specificity MAP kinase phosphatases (MKPs): shaping the outcome of MAP kinase signalling

Abstract

Dual-specificity MAP kinase phosphatases (MKPs) provide a complex negative regulatory network that acts to shape the duration, magnitude and spatiotemporal profile of MAP kinase activities in response to both physiological and pathological stimuli. Individual MKPs may exhibit either exquisite specificity towards a single mitogen-activated protein kinase (MAPK) isoform or be able to regulate multiple MAPK pathways in a single cell or tissue. They can act as negative feedback regulators of MAPK activity, but can also provide mechanisms of crosstalk between distinct MAPK pathways and between MAPK signalling and other intracellular signalling modules. In this review, we explore the current state of knowledge with respect to the regulation of MKP expression levels and activities, the mechanisms by which individual MKPs recognize and interact with different MAPK isoforms and their role in the spatiotemporal regulation of MAPK signalling.

Fig. 1

Classification, localization and domain structure of the MKPs. The three subclasses of MKPs are grouped according to localization, substrate specificity and sequence similarity. The JNK/p38 phosphatases are evenly distributed between the nucleus and cytoplasm and so are not represented on either side of the nuclear envelope here. Positions of the major features of MKPs within the primary amino acid sequences of MKPs are represented as defined in the key.

Fig. 2

Regulation of ERK distribution by MEK, DUSP5 and DUSP6/MKP-3. The figure illustrates how MEK, DUSP5 and DUSP6/MKP-3 cooperate to regulate ERK responses in the nucleus and cytoplasm, respectively. The cytosolic kinase, MEK is the only known MKK for ERK and anchors ERK in the cytoplasm under basal conditions. Activation of MEK by the RAF MAPK kinase kinase causes phosphorylation of ERK and dissociation of MEK. This commonly causes nuclear accumulation of ERK unless cytoplasmic anchors or scaffolds of ERK are present at sufficient concentration. The NES of DUSP6/MKP-3 and high affinity for ERK binding, irrespective of phosphorylation state enables competition with MEK and other ERK partners and substrates in the cytoplasm, causing sequestration of dephosphorylated ERK in this compartment. DUSP5 is biochemically similar to DUSP6 in its high selectivity for ERK substrates, but is nuclear targeted because of its NLS, and performs an analogous role by sequestering dephosphorylated ERK in the nucleus.

Fig. 3

MKPs as temporal regulators of MAPK signalling. The signalling diagrams on the left show how MKPs mediate dynamic effects on MAPK substrates, and the graphs on the right illustrate hypothetical scenarios of how temporal changes in MAPK activity may result from such regulation. MAPK1 and MAPK2 represent notional kinases in these pictures responding to prolonged, slowly desensitizing upstream signals. (A) MKPs can act in autoregulatory loops in response to sustained signals and act to desensitize the MAPK that caused their expression. This serves to attenuate MAPK activity and make it more transient in the face of prolonged upstream activation. (B) MKPs are often induced by transient MAPK signals that are desensitized by mechanisms too rapid to involve MKPs. Thus, the MKP expression may alter the response to subsequent pulses of signal or to other stimuli acting through target MAPKs, enabling the cell to retain a transient ‘memory’ that influences the dynamics of subsequent signalling events. (C) MKPs induced by MAPK activity may have substrate preference for other MAPK isoforms, causing decreases in signal flux through parallel pathways in crosstalk mechanisms. Several examples of each of these scenarios have been defined experimentally, and some of the clearest are highlighted in the main text to illustrate these differential regulatory modes.