Protein farnesyltransferase and protein geranylgeranyltransferase type I: Metal requirements and substrate specificities.

Abstract

Prenylation is essential for the function of many cellular signaling proteins, including Ras, which is mutated in 30% of all human cancers. Farnesyltransferase (FTase) and geranylgeranyltransferase type I (GGTase I) are enzymes that catalyze thioether bond formation between a cysteine sulfur near the C-terminus of a protein substrate and the carbon-1 of the isoprenoid substrate using an essential Zn(II) cofactor. FTase also requires Mg(II) for maximal turnover. We have demonstrated that the prenylation rate constant (<italic>k</italic>_chem) catalyzed by WT GGTase I is Mg(II)-independent. In GGTase I, Lys-beta311 is homologous to the FTase Mg(II) ligand, Asp-beta352. Mutation of Lys-beta311 to alanine (Kbeta311A) or aspartate (Kbeta311D) decreases <italic>k</italic>_chem in the absence of Mg(II) 9- to 41-fold. Moreover, <italic>k</italic>_chem for these mutants is enhanced by addition of Mg(II). These results demonstrate that Lys-beta311 of GGTase I partially replaces the catalytic function of Mg(II) in FTase. In FTase, two putative Mg(II) ligands are the diphosphate of farnesyl diphosphate (FPP) and a water molecule located near Lys-beta353 and Lys-beta356. The contributions of Lys-beta353, Lys-beta356, and the diphosphate of FPP to FTase catalysis were examined using mutagenesis and FPP analogs; these results demonstrate significant effects on product dissociation (>20-fold) and suggest a unique mechanism of product release for FTase. The C-terminal residue of the protein substrate was thought to confer specificity for FTase or GGTase I. However, the dual specificity previously observed for some substrates suggests otherwise. The mechanism of action of prenyltransferase inhibitors is complicated by these overlapping substrate specificities, underscoring the need to identify prenylated substrates. We systematically examined the specificity of GGTase I for peptides differing in C-terminal amino acids. In general, turnover of GGTase I increased as the hydrophobicity of the C-terminal amino acid increased, whereas, binding affinities were only slightly altered, distinguishing kinetic and thermodynamic contributions to cross-specificity. Additionally, we scanned 214 peptide sequences corresponding to the C-terminal sequences of potentially prenylatable human proteins and identified 31 novel substrate sequences of GGTase I. Several of these sequences reflect proteins involved in G-protein-dependent signal transduction pathways. However, others are involved in distinct pathways or their cellular functions are unknown. These results add to our knowledge of GGTase I specificity in terms of sequence requirements and substrate function. Insight from this work defines differences in the kinetic mechanisms and substrate specificities of FTase and GGTase I, providing valuable information about these emerging therapeutic targets.