Osmolarity – an overview | ScienceDirect Topics

3 The HOG Signaling System

The HOG pathway (Fig. 2.1) is one of the best understood and most intensively studied MAPK systems. First, components (Hog1 and Pbs2) were identified in a genetic screen for osmosensitive mutants deficient in glycerol accumulation (Brewster et al., 1993). In parallel, inactivation of SLN1, encoding the single yeast sensor histidine kinase, was found to be lethal (Ota and Varshavsky, 1993). This lethality, which was later shown to be because of inappropriate overactivation of the Hog1 kinase, was suppressed by knockout of any of the genes SSK1, SSK2, PBS2, and HOG1, thereby defining a linear pathway from Sln1 to Hog1. In addition, overexpression of PTC1, PTP2, or PTP3 suppressed lethality of the sln1 mutant, defining those as negative elements of the pathway (Maeda et al., 1994; Posas and Saito, 1997; Posas et al., 1996). Finally, the observation that ssk1 as well as ssk2 ssk22 mutants were osmotolerant while deletion of PBS2 and HOG1 caused osmosensitivity prompted genetic screens employing synthetic enhancement that identified SHO1, STE20, and STE11 as encoding components of the Sho1 branch (Maeda et al., 1995). Identification of the pathway components and characterization of their order of function represent textbook examples of the power of both targeted and global yeast genetics approaches. In fact, forward genetics, suppressor mutation, multicopy suppression, synthetic enhancement, epistasis analysis, and yeast two hybrid screens were all employed in this context. Particular powerful genetic tools are mutations that activate signaling constitutively. Significant knowledge has emerged on the flow of information through the pathway and hence the mechanisms of signal transduction by combining the genetic tools with in vitro and in vivo protein interaction assays, as well as in vitro protein kinase assays (de Nadal et al., 2002; Hohmann, 2002; O'Rourke et al., 2002; Saito and Tatebayashi, 2004; Tatebayashi et al., 2006).

The HOG signaling system consists of two branches that converge on the MAPKK Pbs2, the Sln1, and the Sho1. Components of the Sho1 branch also take part in pseudohyphal development and mating in S. cerevisiae (O'Rourke and Herskowitz, 1998). In many fungi, it appears that the Sho1 module is not connected to Pbs2 and hence is not involved in osmotic responses (Furukawa et al., 2005; Krantz et al., 2006). This indicates that the Sho1 module might not primarily have a role in osmosensing but rather perceives signals related to cell shape and/or cell surface conditions, in accordance with the role in activation played by the cell polarity machinery. Sho1 is specifically located at sites of cell growth and does not appear to sense turgor changes (Reiser et al., 2000, 2003).

The Sho1 branch consists almost exclusively of proteins shared with the pseudohyphal development pathway and the pheromone response pathway. Signaling specificity seems to be assured by recruitment to scaffold proteins (Sho1, Opy2, Pbs2) and requires the Hog1 kinase. In hog1 mutants, exposure to osmotic stress causes activation of the pseudohyphal and pheromone response pathways and morphological aberrations (Davenport et al., 1999; O'Rourke and Herskowitz, 1998; Rep et al., 2000). The mechanism by which Hog1 prevents such cross talk has not yet been elucidated. Mechanisms involved in activation of the Sho1 branch following osmotic shock have been described in detail using constitutively active Stell and Sho1 mutants as well as protein interaction studies (Tatebayashi et al., 2006). As indicated earlier, the sensing mechanism of osmotic changes in the Sho1 branch is not understood at this point but must be closely related to Sho1 (Tatebayashi et al., 2006). The observation that Sho1 can be replaced by engineered proteins that recruit Pbs2 to the plasma membrane suggests that Sho1 does not function as a sensor itself (Raitt et al., 2000). Sho1 shows much less variation in size than in primary sequence (Krantz and Hohmann, 2006), indicating a structural rather than an enzymatic function.

Sln1 is a sensor histidine kinase related to bacterial twocomponent systems. Such proteins are widespread in fungi and plants (Catlett et al., 2003). Sln1 has a similar domain organization as the bacterial osmosensing histidine kinase EnvZ. Both proteins have two transmembrane domains at their N terminus, which are connected by a large extracellular loop, about 300 amino acids in yeasts. It is believed that the extracellular loop and the transmembrane domains sense turgor changes (Reiser et al., 2003), perhaps by responding to movements of the plasma membrane relative to the cell wall. The homodimer is likely regulated by a structural change, which is propagated from the extracellular sensing domain to the intracellular histidine kinase domain of Sln1 (Posas et al., 1996; Reiser et al., 2003). In S. cerevisiae the Sln1 histidine kinase is a negative regulator of the downstream MAPK cascade; deletion of SLN1 or inactivation of the kinase results in lethal Hog1 overactivation (Maeda et al., 1994). When active (i.e., under ambient conditions), the Sln1 histidine kinase crossphosphorylates within a dimer (Posas et al., 1996), and the phosphate group is transferred via the Sln1 receiver and response regulator domains as well as the Ypd1 phosphotransfer protein to the Ssk1 response regulator protein. Hyperosmotic shock causes inactivation of Sln1 kinase activity and dephosphorylation of Ssk1. This scenario is well supported by mutational analysis of all steps in the phosphorelay system (Posas et al., 1996). Unphosphorylated Ssk1 mediates activation of the redundant MAPKKKs Ssk2 and Ssk22, which in turn activate Pbs2.

The activity and the relative contribution of the two pathway branches to Hog1 kinase activity are usually measured in mutants that are blocked in either branch (Maeda et al., 1995; O'Rourke and Herskowitz, 2004). Whether such experiments reflect activity of the two branches in wildtype cells is presently unknown. It appears that the Sho1 branch has a higher stress threshold for activation (Maeda et al., 1995; O'Rourke and Herskowitz, 2004) and that it is insufficient to mediate maximal pathway activation alone (unpublished data).

Excerpt from:
Osmolarity - an overview | ScienceDirect Topics