Abstract
Abstract Activating mutations of NRAS are found in 20%–40% of acute myeloid leukemia (AML), chronic myelomonocytic leukemia (CMML), and myelodysplastic syndrome (MDS). Like all RAS proteins, oncogenic NRAS must undergo a series of post-translational modifications for differential targeting to distinct cellular microdomains. NRAS is first farnesylated, then subsequently cleaved and methylated at its C-terminal -CAAX region, and is finally reversibly palmitoylated, with each modification catalyzed by a distinct enzyme(s). Once targeted to plasma membrane and/or endomembranes, NRAS engages local activators/effectors to generate distinct signaling outputs that govern cell survival, proliferation and differentiation. The requirement of membrane attachment for transforming activity by RAS oncoproteins makes RAS modification enzymes attractive targets for the development of cancer therapies. To date, however, success of therapies targeting the obligatory first step in the series of RAS post-translational modifications by targeting farnesyl protein transferase have been modest (KRAS and NRAS may alternatively prenylated via geranylgeranylation). Additionally it has been unclear whether RAS is the target affected by blocking farnsylation using inhibitors, since many proteins are modified in this way. Other RAS modifications such as palmitoylation (for HRAS and NRAS) may be more specific (a smaller number of proteins contain -CAAX motifs that are both prenylated and methylated, both of which are thought to be required for palmitoyltransferase activity) and are required for optimal plasma membrane association and targeting of RAS to specific membrane subdomains. However, inconsistent results have been reported as to the relative importance of these latter modifications on RAS function and transforming ability. We have previously shown that expression of oncogenic NRAS in a bone marrow transduction/transplantation model system induces an AML- or CMML-like disease in mice. Using this model in conjunction with mutational analysis of specific post-translational modification sites on the oncogene, we have observed the effect of specifically blocking the palmitoylation modification to oncogenic NRAS in NRAS leukemogenesis. Unlike mutation to the farnesylation site, which causes dislocalization from membranes entirely, the palmitoylation mutation redistributes oncogenic NRAS to internal membranes, indicating that the oncoprotein may retain signaling capacity. However, we find that some signaling pathways upregulated by oncogenic NRAS including the RAS-RAF-MEK-ERK signaling pathway are returned to normal nontransforming signaling levels in cells expressing oncogenic NRAS with the palmitoylation mutation. Furthermore, like farnesylation, we find palmitoylation to be essential for leukemogenesis by oncogenic NRAS in our mouse bone marrow transduction/transplantation model. This study sheds new light on the mechanism of NRAS leukemogenesis and suggests that the palmitoyltransferase enzyme for NRAS may be an effective target for developing therapeutics for hematological malignancies.