We have found that microinjection of the transforming but not the normal p21 protein into Xenopus laevis oocytes induced the production of 1,2-diacylglycerol (DAG) and inositol triphosphate (IP-3). While the transforming H-ras p21 was an effective mitogen for normal 3T3 cells, its mitogenic function was substantially reduced (about 80%) in protein kinase C (PKC)-depleted cells. The activity was almost completely recovered by co-microinjection of the ras p21 protein and PKC. These results provide evidence for a functional requirement of PKC for the mitogenic activity of the H.ras protein. In contrast with the results obtained in Xenopus oocytes, 3T3 cells transformed by a variety of ras oncogenes do not show any increase of basal levels of IP-3. However, DAG levels are increased about 40-50% over control, normal cells, suggesting a different source for the production of DAG than the hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP-2). We have observed elevated levels of the catabolites resulting from the hydrolysis of other major phospholipids, like phosphatidyl-choline (PC) and phosphatidyl-ethanolamine (PE). Transformation by the sis oncogene, as well as treatment with serum or PDGF, induced production of DAG and IP-3 but not the hydrolysis of PC or PE. These results provide evidence for at least two independent mechanisms for the production of DAG, and that both mechanisms can be activated by individual oncogene products. In a different set of experiments, we have been able to express in E. coli and purify to homogeneity the product of the rho gene from Aplysia californica, a ras-related gene. We have demonstrated that, indeed, the gene product of the rho gene (p21 rho) is a G. protein with a similar GTPase activity to that of the normal ras protein. Finally, we have been able to restore membrane localization and transforming activities of ras p21 mutants devoid of both activities by insertion of a membrane signaling peptide of the p60src product from the amino terminal.
