Ras signaling: PP2A puts Ksr and Raf in the right place

Cells are exposed to a vast array of extracellular stimuli that they have to interpret in a cell-type specific manner. In metazoans, at least five distinct mitogenactivated protein kinase (MAPK) signaling cascades have evolved that couple extracellular signals to a variety of cellular outputs. The core structure of each MAPK signaling pathway is a module of three kinases (MAPK kinase kinase (MAPKKK), MAPK kinase (MAPKK) and MAPK), which become sequentially activated [1]. Pathway specificity is at least in part ensured by scaffold proteins, which help to assemble the active signaling complex and localize it to the appropriate subcellular site. A prominent example is the Raf (MAPKKK) – MEK (MAPKK) – Erk (MAPK) pathway, which relays signals from the activated small G-protein Ras to downstream effector proteins. Previous studies have identified kinase suppressor of Ras (Ksr) as a scaffold protein specific for the Raf–MEK–Erk pathway [2]. Work by Ory et al. [3], published in this issue of Current Biology, provides evidence that protein phosphatase 2A (PP2A) coordinates the assembly of an active signaling complex at the membrane through dephosphorylation of Ksr and Raf. Ksr was originally identified in genetic screens in Drosophila and Caenorhabditis elegans for proteins acting downstream of activated Ras. A comparison of the Ksr proteins found in worm, fly and vertebrates revealed five highly conserved protein domains: a 40 amino acid sequence (CA1) unique to Ksr proteins; a proline-rich region (CA2); a cysteine-rich C1 domain (CA3); a serine-threonine-rich region (CA4), and a predicted carboxy-terminal kinase domain [2]. However, the replacement of a highly conserved lysine residue in the kinase subdomain II of the mammalian Ksr proteins with arginine suggests that mammalian Ksr proteins do not possess kinase activity. Instead, all available data indicate that Ksr acts as a molecular scaffold [2]. In quiescent cells, Ksr is found in the cytoplasm where it interacts, like the inactive Raf protein, with 14-3-3 proteins. The 14-3-3 proteins form dimers that regulate protein function through binding to phosphoserine-containing motifs [4]. In addition, Ksr interacts constitutively with MEK. Upon Ras stimulation, Ksr localises MEK together with Erk to membrane-bound, activated Raf. Raf promotes phosphorylation of MEK, which in turn is required to activate Erk by phosphorylation (Figure 1). But what are the molecular mechanisms that determine the subcellular localisation of Ksr? Two papers from the Morrison lab, one published in Molecular Cell [5] and now one in this issue of Current Biology [3], provide novel and exciting insights. In quiescent cells, Ksr becomes phosphorylated at two serine residues (Ser297 and Ser392) through the action of the Cdc25Cassociated kinase C-TAK1 [5]. Both sites had previously been described as binding sites for 14-3-3 [6]. Treatment of cells with growth factors did not alter the binding or kinase activity of C-TAK1, but resulted in dephosphorylation of Ksr specifically at Ser392, thereby releasing 14-3-3 from this site. Ser297 and Ser392 flank the cysteine-rich C1 domain (CA3), which is essential for translocation and accumulation of Ksr at the membrane [7]. The model that emerged from these studies proposes that the release of 14-3-3 causes a conformational change, allowing the C1 domain of Ksr to contact with as yet unknown membrane-bound ligands. Ory et al. [3] go one step further to unravel the mechanism that controls Ksr dephosphorylation. Using mass spectrometry, they identified the serine/threonine phosphatase PP2A as a component of the Ksr signaling complex in cell culture and in brain lysates from mice. The PP2A core enzyme is a dimer consisting of a catalytic subunit and a structural A subunit. Several distinct B subunits can associate with this core structure to regulate the enzymatic activity [8]. Strikingly, the comparative analysis between growth-factor-stimulated and unstimulated cells revealed that the catalytic subunit and the structural A subunit are bound to Ksr under both conditions whereas the association of the regulatory B subunit with Ksr was dependent on growth factor stimulation. Although this story is exciting enough on its own, one of the surprises is that PP2A is also required to regulate Raf activity [9–11]. A complex and still incompletely understood series of protein modifications, protein–protein and protein–lipid interactions is involved in translocation and activation of Raf at the membrane [12–14]. One critical parameter for Raf activation and membrane recruitment is the dephosphorylation of Ser259, which, like Ser392 in Ksr, serves as a binding site for 14-3-3. Mass spectrometry analysis of purified Raf complexes now revealed the presence of the structural A subunit of PP2A, the catalytic subunit and, in a growth-factor-dependent manner, the regulatory B subunit [3]. Thus, PP2A mediates dephosphorylation of 14-3-3 binding sites in Ksr and Raf. What are the biological effects of PP2A-induced dephosphorylation of Ksr and Raf? Previously it was shown that mutation of Ser392 in Ksr resulted in constitutive localisation of Ksr at the membrane in a Dispatch Current Biology, Vol. 13, R635–R637, August 19, 2003, ©2003 Elsevier Science Ltd. All rights reserved. DOI 10.1016/S0960-9822(03)00568-2