DESIGN, SYNTHESIS AND BIOLOGICAL EVALUATION OF NON-PEPTIDIC INHIBITORS OF FARNEYLTRANSFRERASE

Ras proteins are plasma membrane–bound GTP-binding proteins which play an important role in normal cellular physiology and pathophysilogy. They are involved in the transmission of signals from extracellular stimuli to the nucleus. Mutations in ras genes, that lead to uncontrolled cell growth, have been identified in human cancers especially those of pancreas, colon, lung and bladder [1].
Ras proteins are initially sinthesized in the cytoplasma where they undergo farnesylation as the first step of the post-translational modifications. Farnesylation of the cysteine unit of the so-called CAAX box is followed by the proteolytic cleavege of the three terminal aminoacids (AAX) and by methyl esterification at the new C terminal Cysteine residue by a protein methyltransferase [2]. The last step is the acylation with palmitic acid of Cysteine residue located upstream of the farnesylated Cysteine. This palmitoylation increses the binding of Ras proteins to the cell membrane.
As the farnesylation is catalyzed by farnesyltransferase (Ftase), inhibition of this enzyme represents a potential target for the development of anticancer agents. Many classes of Farnesyltransferase inhibitors (FTIs) have been reported in the literature. They have been designed based on: a) the modification of tetrapeptide CAAX (peptidomimetics); b) the farnesyl moiety of farnesyl pyrophosphate (FPP mimetics); c) the bisubstrate analogues incorporating the structural motif of both farnesyl pyrophosphate and the CAAX tetrapeptide; d) the natural inhibitors.
Our studies on FTIs focused on peptidomimetics. On the basis of the structures of known inhibitors and of the results of docking searches [3], the compounds of formula 1were developed.
The interaction capabilities of the selected compounds were highlighted with detailed molecular docking searches, using BioDock. More in details, MonteCarlo based conformational scans were combined with docking analyses in order to take in account the ligand flexibility. The preferred conformations for each derivative (15/20 structures per ligand) were docked in the FTase structure and the best obtained complexes were minimized, keeping fixed all residues outside a 10 Å radius sphere around the ligand. This study underlines how these novel inhibitors enclose at least three interacting substructures: 1) the carboxyl moiety, which realizes strong electrostatic interactions with Zn++ ion reinforced by similar interactions with Arg-202 and Arg-359; 2) the biphenyl group, which realizes  stacking with the Tyr-361 residue; 3) the benzodioxane moiety, which interacts with Lys-294 and Lys-164, establishing both charge transfer interaction with the phenyl ring and hydrogen bonds with benzodioxane oxygen atoms. These results seem to indicate that the methylthioethyl chain of Methionine is not involved in the interaction with the active site of the enzyme. Therefore, we have planned the replacement of such an aminoacid with Glycine, Isoleucine or Homoserine.
For this new class of FTIs, structure activity relationship will be proposed on the basis of the results of the binding studies.