ADP-ribosylation factor 1 (Arf1) is an essential N-myristoylated 21-kDa GTP-binding protein with activities that include the regulation of membrane traffic and phospholipase D activity. Both the N terminus of the protein and the N-myristate bound to glycine 2 have previously been shown to be essential to the function of Arf in cells. We show that the bound nucleotide affects the conformation of either the N terminus or residues of Arf1 that are in direct contact with the N terminus. This was demonstrated by examining the effects of mutations in this N-terminal domain on guanosine 5'-O-(3-thio)triphosphate (GTP gamma S) and GDP binding and dissociation kinetics. Arf1 mutants, lacking 13 or 17 residues from the N terminus or mutated at residues 3-7, had a greater affinity for GTP gamma S and a lower affinity for GDP than did the wild-type protein. As the N terminus is required for interactions with target proteins, we conclude that the N terminus of Arf1 is a GTP-sensitive effector domain. When Arf1 was acylated, the GTP-dependent conformational changes were codependent on added phospholipids. In the absence of phospholipids, myristoylated Arf1 has a lower affinity for GTP gamma S than for GDP, and in the presence of phospholipids, the myristoylated protein has a greater affinity for GTP gamma S than for GDP. Thus, N-myristoylation is a critical component in the construction of this phospholipid- and GTP-dependent switch.

No abstract

The human immunodeficiency virus type 1 internal structural protein precursor, p55, and its corresponding matrix proteolytic fragment, p17, are phosphorylated at Ser111 by protein kinase C. COS-7 cells transfected with plasmids encoding either the wild-type or Ser111-->Ala mutated human immunodeficiency virus type 1 gag gene matrix domain proteins were treated with phorbol 12-myristate 13-acetate (PMA), and the phosphorylation of the expressed p17 proteins was examined by radioimmunoprecipitation, SDS-polyacrylamide gel electrophoresis, and autoradiography. PMA treatment of transfected cells resulted in a 4-5-fold increase in wild-type p17 (but not mutated p17) phosphorylation; however, mutated p17 exhibited a low basal level of phosphorylation that was not affected by PMA, suggesting that additional sites were phosphorylated. PMA treatment of cells expressing wild-type p17 produced a dramatic shift in the localization of p17 from the cytosol to the membrane fraction within 8-15 min, followed by a slow quantitative dissociation of p17 back into the cytosol by 90 min. The cytosol-to-membrane translocation was dependent on N-myristoylated p17 since cells expressing p17 with a Gly2-->Ala mutation did not localize to the membrane. PMA also failed to induce the translocation of fully N-myristoylated Ser111-->Ala p17, suggesting that p17 phosphorylation at Ser111 was responsible for membrane association. This conclusion was confirmed by the finding of phosphorylated wild-type p17 in the membrane fraction only after PMA treatment. These results suggest that a "myristoyl-protein switch" regulates the reversible membrane targeting of p17 by protein kinase C-mediated phosphorylation. This signal may provide a mechanism for the cellular regulation of virus development through modulation of gag protein-related developmental steps such as capsid targeting, assembly, encapsidation, budding, and maturation.

No abstract