Interaction of shrimp ras protein with mammalian caveolin-1

An enhanced association of RACK1 with Abl in cells transfected with oncogenic ras

The cellular RACK1 was shown in association with Abl in BALB/3T3 cells transfected with S-ras(Q(61)K) by immunoprecipitation. An identical finding was demonstrated with cells transfected with the embryonic E-ras, but not in cells without transformation. The Abl-RACK1 of transformed cells as resolvable with Triton X-114 was found with little affinity for FAK, PY(397)-FAK and integrin. Of interests, PY(397)-FAK in the membrane skeleton of transformed cells was shown in significant quantities on the Western blot. However the PY(397)-FAK of transformed cells was not functionally able to react with RACK1 and recruit cytokeratin-1, a substrate of Src, indicating that PY(397)-FAK is not operative to transmit integrin signals. In other words, the Abl-RACK1 of transformed cells cannot replace the Src-RACK1 of cells without transformation to bridge PY(397)-FAK and cytokeratin-1 for integrin signals, and the formation of Abl-RACK1 in transformed cells may block the association of PY(397)-FAK-RACK1. We characterized Abl and RACK1 from transformed cells by chromatography on a HiTrap-PEP(Taxol) affinity column, constructed from a beta-tubulin peptide specific for Taxol binding (PEP(Taxol)). However, the Triton X-100 cannot achieve the same resolution of Abl-RACK1 from plasma membrane as is shown with Triton X-114. A significant fraction of Abl was deposited at the membrane skeleton and was therefore not accessible with Triton X-100. Half of Abl resolved with Triton X-100 was demonstrated to have catalytic activity as shown with positive phosphotyrosine staining on the Western blot and competitive elution with a specific phosphate, such as sodium beta-glycerophosphate, from HiTrap-PEP(Taxol), but this was not associated with RACK1. No significant difference of RACK1 was found in Triton X-100 resolvable membrane preparations from cells with and without transformations. Future studies are planned to differentiate the mechanism operative for RACK1 associated and RACK1 freed Abl in cells transformed with oncogenic ras.

Neurofibromin binds to caveolin-1 and regulates ras, FAK, and Akt

Neurofibromin (Nf1) is an approximately 280 kDa protein having tumor suppressor function, presumably by virtue of its GTPase activating domain, but little is known regarding molecular aspects of its effector pathways. Caveolin-1 (Cav-1) regulates diverse signaling molecules and has itself been implicated as a tumor suppressor. Here we demonstrate that Nf1 binds to Cav-1's scaffolding domain and co-immunoprecipitates with Cav-1. Analysis of Nf1's primary structure reveals four potential caveolin binding domains, and interestingly, in individuals with neurofibromatosis I, missense mutations occur with high frequency in 3 of the 4 putative domains. We show that Nf1 modulates ras, Akt, and focal adhesion kinase pathways, thereby affecting cytoskeletal organization; moreover, Nf1's effects on signaling are altered when lipid rafts and caveolae are disrupted by cholesterol depletion. These novel findings provide insight into possible signaling mechanisms of Nf1 and suggest that together Nf1 and Cav-1 may coordinately regulate cell growth and differentiation.

P120-GAP associated with syndecan-2 to function as an active switch signal for Src upon transformation with oncogenic ras

BALB/3T3 cells transfected with plasmids pcDNA3.1-[S-ras(Q(61)K)] of shrimp Penaeus japonicus were applied to reveal a complex of p120-GAP/syndecan-2 being highly expressed upon transformation. Of interest, most of the p120-GAP/syndecan-2 complex was localized at caveolae, a membrane microdomain enriched with caveolin-1. To confirm the molecular interaction between syndecan-2 and p120-GAP, we further purified p120-GAP protein from mouse brains by using an affinity column of HiTrap-RACK1 and expressed mouse RACK1-encoded fusion protein and mouse syndecan-2-encoded fusion protein in bacteria. We report molecular affinities exist between p120-GAP and RACK1, syndecan-2 and RACK1 as well as p120-GAP and syndecan-2. The selective affinity between p120-GAP and syndecan-2 was found to be sufficient to detach RACK1. The p120-GAP/syndecan-2 complex was demonstrated to keep Src tyrosine kinase in an activated form. On the other hand, the syndecan-2/RACK1 complex was found to have Src in an inactivated form. These data indicate that the p120-GAP/syndecan-2 complex at caveolae could provide a docking site for Src to transmit tyrosine signaling, implying that syndecan-2/p120-GAP functions as a tumor promoter upon transformation with oncogenic ras of shrimp P. japonicus.

Novel roles for palmitoylation of Ras in IL-1β-induced nitric oxide release and caspase 3 activation in insulin-secreting β cells

We recently demonstrated that functional inactivation of H-Ras results in significant reduction in interleukin 1 beta (IL-1 beta)-mediated effects on isolated beta cells. Since palmitoylation of Ras has been implicated in its membrane targeting, we examined the contributory roles of palmitoylation of Ras in IL-1 beta-induced nitric oxide (NO) release and subsequent activation of caspases. Preincubation of HIT-T15 or INS-1 cells with cerulenin (CER, 134 microM; 3 hr), an inhibitor of protein palmitoylation, significantly reduced (-95%) IL-1 beta-induced NO release from these cells. 2-Bromopalmitate, a structurally distinct inhibitor of protein palmitoylation, but not 2-hydroxymyristic acid, an inhibitor of protein myristoylation, also reduced (-67%) IL-1 beta-induced NO release from HIT cells. IL-induced inducible nitric oxide synthase gene expression was markedly attenuated by CER. Further, CER markedly reduced incorporation of [3H]palmitate into H-Ras and caused significant accumulation of Ras in the cytosolic fraction. CER-treatment also prevented IL-1 beta-induced activation of caspase 3 in these cells. Moreover, N-monomethyl-L-arginine, a known inhibitor of inducible nitric oxide synthase, markedly inhibited IL-induced activation of caspase 3, thus establishing a link between IL-induced NO release and caspase 3 activation. Depletion of membrane-bound cholesterol using methyl-beta-cyclodextrin, which also disrupts caveolar organization within the plasma membrane, abolished IL-1 beta-induced NO release suggesting that IL-1 beta-mediated Ras-dependent signaling in these cells involves the intermediacy of caveolae and their key constituents (e.g. caveolin-1) in isolated beta cells. Confocal light microscopic evidence indicated significant colocalization of Ras with caveolin-1. Taken together, our data provide the first evidence to indicate that palmitoylation of Ras is essential for IL-1 beta-induced cytotoxic effects on the islet beta cell.

Effects of prenyl pyrophosphates on the binding of S-Ras proteins with KSR

BALB/3T3 cells were transformed by transfection with DNA encoding the mutated ras(Q(61)K) from shrimp Penaeus japonicus (Huang et al., 2001. J. Exp. Zool. 289:441-448). On a Western blot, the kinase suppressor of Ras (KSR) in the membrane fraction was expressed at slightly reduced level as compared to that of the untransformed cells. To understand this in more detail, the interaction of the bacterially expressed shrimp Ras (S-Ras) with KSR was investigated using KSR purified from mice brains. SDS-polyacrylamide gel electrophoresis and Western blot analysis revealed that the monomers of the purified KSR have a relative molecular mass of 60,000. Purified KSR was found to bind with digoxigenylated S-ras-encoding fusion protein (Dig-S-Ras) with high affinity in the absence of ATP, and the binding activity of KSR was sustained upon phosphorylation of Dig-S-Ras with mitogen-activated protein kinase (MAPK). The association of purified KSR with S-Ras was confirmed. Differences between the effects of farnesyl pyrophosphate and geranylgeranyl pyrophosphate on the binding of S-Ras with the purified KSR were assessed. Densitometer analysis revealed that at nanogram concentration, farnesyl pyrophosphate inhibited the binding of S-Ras with KSR competently, but geranylgeranyl pyrophosphate did not. The present study provides the evidence that decrease of the concentration of farnesyl pyrophosphate to sub-microgram levels lower the affinity of Ras proteins with KSR in the signaling pathway.

GTPase stimulation in shrimp Ras(Q(61)K) with geranylgeranyl pyrophosphate but not mammalian GAP

BALB/3T3 cells were transformed by transfection with DNA encoding the mutated ras(Q(61)K) from shrimp Penaeus japonicus (Huang et al., 2000). The GTPase-activating protein (GAP) in the cytosol fraction was significantly expressed and degraded, compared to untransformed cells on the western blot. To understand this in more detail, the interaction of the bacterially expressed shrimp Ras (S-Ras) with GAP was investigated using GAP purified from mouse brains. SDS-polyacrylamide gel electrophoresis revealed the monomers of the purified GAP to have a relative mass of 65,000. Since the purified GAP was bound to the Ras conjugated affinity sepharose column with high affinity and its GTP hydolysis activity upon binding with tubulin was suppressed, the purified enzyme was concluded to be neurofibromin-like. The purified GAP enhanced the intrinsic GTPase activity of the S-Ras, to convert it into the inactive GDP-bound form, in agreement with findings for GTP-bound K(B)-Ras in vitro. To compare the effects between isoprenoids and GAP on the GTP-hydrolysis of Ras, we applied the GTP-locked shrimp mutant S-Ras(Q(61)K) and GTP-locked rat mutant K(B)-ras(Q(61)K). Radioassay studies showed that geranylgeranyl pyrophosphate at microg level catalyzed the GTP hydrolysis of S-Ras(Q(61)K) and K(B)-ras(Q(61)K) competently, but not farnesyl pyrophosphate or the purified GAP. The present study provides the view that the geranylgeranyl pyrophosphate at carboxyl terminal CAAX assists GTP hydrolysis to Ras proteins probably in a manner similar to the substrate assisted catalysis in GTPase mechanism.