Oncogene studies have involved ras encoded proteins, which have been analyzed by examining proteins that influence the activity of Ras protein. The work emphasises the complex mechanisms that regulate Ras activity, with the regulators of Ras activity being themselves subject to multiple types of regulation. Schwannoma cell lines from patients with neurofibromatosis had low levels of the NF1 product neurofibromin, which correlated with their containing high levels of GTP*Ras. These results were consistent with NF1 being a tumor suppressor gene whose encoded GTPase stimulation negatively regulates Ras. We have also identified neuroblastoma and melanoma cell lines with genetic abnormalities of NF1 and reduced to absent levels of neurofibromin, suggesting that NF1 is acting as a tumor suppressor gene in these cell lines. In contrast to the schwannoma lines, the level of GTP*Ras was low in all lines and did not correlate with that of neurofibromin. These results suggested that NF1 might inhibit cell growth by a mechanism independent of its GTPase stimulatory activity. To confirm this hypothesis, a full length NF1 cDNA was itnroduced into NIH 3T3 cells. The cells that overexpressed neurofibromin grow more slowly and had normal levels of GTP*Ras. The inhibition was at the level of Ras. Introduction of the NF1 cDNA into melanoma lines slowed their growth and induced a differentiated phenotype, including an increase in cell size, dendrite formation, and an increase in tyrosinase. We have also determined that the regulation of Sos2 stability represents a previously unrecognized mechanism for altering the sensitivity of the Ras-dependent signal transduction pathway. Although sos2 is reported to be as widely expressed as sos1 at the RNA level, the Sos2 protein is unstable in most tissues, in contrast to the Sos1. In vitro and in vivo studies have shown that this instability results from Sos2 being rapidly degraded in a ubiquitin- dependent manner, unlike Sos1. Analysis of chimeras formed between Sos1 and Sos2 indicate that Sos2 contains at least two ubiquitination signals that are lacking in Sos1.
