The human family of MAP kinase signal-integrating kinases (Mnks) comprises four related proteins derived from two genes by alternative splicing. The Mnk1 gene gives rise to two proteins, Mnk1a and Mnk1b, which possess distinct C-termini and properties.Despite lacking the C-terminal MAP kinase-binding site, Mnk1b shows higher basal activity than Mnk1a. In contrast, the activity of Mnk1a is tightly regulated by signalling through ERK and p38 MAP kinase.We show that the short C-terminus of Mnk1b confers on it a 'default' behaviour of substantial, but unregulated, activity. In contrast, the longer C-terminus of Mnk1a represses the basal activity and T (activation)-loop phosphorylation of this isozyme while allowing both properties to be stimulated by upstream MAP kinase signalling.Two features of the C-terminus of Mnk1a appear to account for this behaviour: the known MAP kinase-binding site and a region (predicted to be -helical) which occludes access to the catalytic domain and the T-loop. The activation of Mnk1a results in a marked conformational change leading to a more 'open' structure. We also identify a conserved phenylalanine in an Mnk-specific insert as playing a key role in governing the ease with which Mnk1a can be phosphorylated.These studies help to identify the features that give rise to the diverse properties of human Mnk isoforms.Keywords: protein kinase; Mnk; ERK; p38 MAP kinase; eIF4E Abbreviations used: eIF, eukaryotic initiation factor; ERK, extracellular ligand-regulated kinase; HEK, human embryonic kidney; hMnk, human Mnk; MAP kinase, mitogen-activated protein kinase; mMnk, mouse Mnk; Mnk, MAP kinase signal-integrating kinase (or MAP kinase-interacting kinase); NES, nuclear export sequence; NLS, nuclear localisation sequence. A c c e p t e d M a n u s c r i p t Licenced copy. Copying is not permitted, except with prior permission and as allowed by law.
© 2009 The Authors Journal compilation © 2009 Portland Press Limited
IntroductionThe MAP kinase signal-integrating kinases (also termed 'MAP kinase-interacting' kinases; Mnks) were first discovered by virtue of their abilities to be phosphorylated by [1] or to bind to [2] ERK and/or p38 MAP kinase. There are two Mnk genes in mice and humans, Mnk1 and Mnk2. The Mnk proteins that were originally reported (here termed Mnk1a and Mnk2a) each contain, within their C-terminal regions, a motif for binding to MAP kinases (Fig. 1A). Mnk1a binds well to both ERK and p38 MAP kinases ( / ) while Mnk2a binds better to ERK and only weakly to p38 MAP kinases / [2;3]. ERK and p38 MAP kinases phosphorylate two threonine residues in Mnk1/2 which are located in the activation or 'T'-loop. In many protein kinases phosphorylation in this region leads to their activation.Subsequently it was shown that the human Mnk1 and Mnk2 genes each give rise to two mRNAs, and thus to two distinct polypeptides, as a consequence of alternative mRNA splicing [4][5][6][7] (Fig. 1A). In each case, the second form of each polypeptide (Mnk1b or Mnk2b) contains a distinct C-termina...