The Ras GDP/GTP cycle is regulated by oxidizing agents at the level of Ras regulators and effectors.

Reactive oxygen species (ROS) have been found to play important roles in regulating cellular functions. Their action in vivo has been related to specific effects on signal transduction pathways, such as Ras pathway. In order to characterize which elements of Ras pathway are affected by ROS, we have analyzed the action of different oxidizing agents on the ability of GTPase activating protein GAP and nucleotide exchange factor GEF to enhance the intrinsic activities of Ras. The action of these agents on the binding between H-Ras and its effector c-Raf-1 was also investigated. No effects were observed on the intrinsic activities of H-Ras or Ras2p. On the other hand, reversible inhibitions of GEF and GAP actions on Ras were found, whose extent was dependent on the agent used. As tested with the scintillation proximity assay, these agents also inhibited the binding of c-Raf-1 to H-Ras. Our data reveal new potential targets for the action of ROS on Ras pathway and suggest that they can influence the Ras activation state indirectly via regulators and effectors.

Sites of phosphorylation by protein kinase A in CDC25Mm/GRF1, a guanine nucleotide exchange factor for Ras.

Activation of the neuronal Ras GDP/GTP exchange factor (GEF) CDC25Mm/GRF1 is known to be associated with phosphorylation of serine/threonine. To increase our knowledge of the mechanism involved, we have analyzed the ability of several serine/threonine kinases to phosphorylate CDC25Mm in vivo and in vitro. We could demonstrate the involvement of cAMP-dependent protein kinase (PKA) in the phosphorylation of CDC25Mm in fibroblasts overexpressing this RasGEF as well as in mouse brain synaptosomal membranes. In vitro, PKA was found to phosphorylate multiple sites on purified CDC25Mm, in contrast to protein kinase C, calmodulin kinase II, and casein kinase II, which were virtually inactive. Eight phosphorylated serines and one threonine were identified by mass spectrometry and Edman degradation. Most of them were clustered around the Ras exchanger motif/PEST motifs situated in the C-terminal moiety (residues 631-978) preceding the catalytic domain. Ser745 and Ser822 were the most heavily phosphorylated residues and the only ones coinciding with PKA consensus sequences. Substitutions S745D and S822D showed that the latter mutation strongly inhibited the exchange activity of CDC25Mm on Ha-Ras. The multiple PKA-dependent phosphorylation sites on CDC25Mm suggest a complex regulatory picture of this RasGEF. The results are discussed in the light of structural and/or functional similarities with other members of this RasGEF family.

In budding yeast, reactive oxygen species induce both RAS-dependent and RAS-independent cell cycle-specific arrest.

The role of mild oxidative stresses elicited by diethylmaleate (DEM)-induced glutathione depletion in the progression of the yeast cell cycle has been investigated. We found that different wild-type strains are sensitive to oxidative stresses induced by similar DEM doses: approximately 1 mM on YPD plates, 5-10 mM in shaken flasks. At lower doses, DEM caused a transient decrease in growth rate, largely because of a decreased G1-to-S transition. Treatment with higher DEM doses leads to complete growth arrest, with most cells found in the unbudded G1 phase of the cell cycle. DEM treatment resulted in transcriptional induction of stress-responsive element (STRE)-controlled genes and was relieved by treatment with the antioxidant N-acetyl cysteine. Reciprocal shift experiments with cdc25 and cdc28 mutants showed that the major cell cycle arrest point was located in the Start area, at or near the CDC25-mediated step, before the step mediated by the CDC28 cyclin-dependent kinase. The DEM-induced G1 arrest requires a properly regulated RAS pathway and can be bypassed by overexpressing the G1-specific cyclin CLN2. However, cells with either a deregulated RAS pathway or overexpressing CLN2 failed to grow and arrested as budded cells, indicating that a second DEM-sensitive cell cycle step exists.