Structure-based design and synthesis of potent, ethylenediamine-based, mammalian farnesyltransferase inhibitors as anticancer agents.

A potent class of anticancer, human farnesyltransferase (hFTase) inhibitors has been identified by "piggy-backing" on potent, antimalarial inhibitors of Plasmodium falciparum farnesyltransferase (PfFTase). On the basis of a 4-fold substituted ethylenediamine scaffold, the inhibitors are structurally simple and readily derivatized, facilitating the extensive structure-activity relationship (SAR) study reported herein. Our most potent inhibitor is compound 1f, which exhibited an in vitro hFTase IC(50) value of 25 nM and a whole cell H-Ras processing IC(50) value of 90 nM. Moreover, it is noteworthy that several of our inhibitors proved highly selective for hFTase (up to 333-fold) over the related prenyltransferase enzyme geranylgeranyltransferase-I (GGTase-I). A crystal structure of inhibitor 1a co-crystallized with farnesyl pyrophosphate (FPP) in the active site of rat FTase illustrates that the para-benzonitrile moiety of 1a is stabilized by a π-π stacking interaction with the Y361ß residue, suggesting a structural explanation for the observed importance of this component of our inhibitors.

Non-peptidic substrate-mimetic inhibitors of Akt as potential anti-cancer agents.

Akt has emerged as a critical target for the development of anti-cancer therapies. It has been found to be amplified, overexpressed, or constitutively activated in numerous human malignancies with oncogenesis derived from the simultaneous promotion of cell survival and suppression of apoptosis. A valuable alternative to the more common ATP-mimetic based chemotherapies is a substrate-mimetic approach, which has the potential advantage of inherent specificity of the substrate-binding pocket. In this paper we present the development of high affinity non-peptidic, substrate-mimetic inhibitors based on the minimum GSK3beta substrate sequence. Optimization of initial peptidic leads resulted in the development of several classes of small molecule inhibitors, which have comparable potency to the initial peptidomimetics, while eliminating the remaining amino acid residues. We have identified the first non-peptidic substrate-mimetic lead inhibitors of Akt 29a-b, which have affinities of 17 and 12 microM, respectively. This strategy has potential to provide a useful set of molecular probes to assist in the validation of Akt as a potential target for anti-cancer drug design.

Modifications of the GSK3beta substrate sequence to produce substrate-mimetic inhibitors of Akt as potential anti-cancer therapeutics.

Amplification, overexpression, and elevated activation of Akt have been detected in many human malignancies making it an important target for cancer therapy. The Akt substrate-binding site offers a large number of potential interactions to an appropriately designed small molecule and can form the basis for the development of selective inhibitors. Here, we report the progression of GSK3beta substrate-mimetic inhibitors towards the development of a potent, small molecule substrate-mimetic inhibitor of Akt.

Design, synthesis, potency, and cytoselectivity of anticancer agents derived by parallel synthesis from alpha-aminosuberic acid.

Chemotherapy in the last century was characterized by cytotoxic drugs that did not discriminate between cancerous and normal cell types and were consequently accompanied by toxic side effects that were often dose limiting. The ability of differentiating agents to selectively kill cancer cells or transform them to a nonproliferating or normal phenotype could lead to cell- and tissue-specific drugs without the side effects of current cancer chemotherapeutics. This may be possible for a new generation of histone deacetylase inhibitors derived from amino acids. Structure-activity relationships are now reported for 43 compounds derived from 2-aminosuberic acid that kill a range of cancer cells, 26 being potent cytotoxins against MM96L melanoma cells (IC50 20 nM-1 microM), while 17 were between 5- and 60-fold more selective in killing MM96L melanoma cells versus normal (neonatal foreskin fibroblasts, NFF) cells. This represents a 10- to 100-fold increase in potency and up to a 10-fold higher selectivity over previously reported compounds derived from cysteine (J. Med. Chem. 2004, 47, 2984). Selectivity is also an underestimate, because the normal cells, NFF, are rarely all killed by the drugs that also induce selective blockade of the cell cycle for normal but not cancer cells. Selected compounds were tested against a panel of human cancer cell lines (melanomas, prostate, breast, ovarian, cervical, lung, and colon) and found to be both selective and potent cytotoxins (IC50 20 nM-1 microM). Compounds in this class typically inhibit human histone deacetylases, as evidenced by hyperacetylation of histones in both normal and cancer cells, induce expression of p21, and differentiate surviving cancer cells to a nonproliferating phenotype. These compounds may be valuable leads for the development of new chemotherapeutic agents.

Structurally simple, potent, Plasmodium selective farnesyltransferase inhibitors that arrest the growth of malaria parasites.

Third world nations require immediate access to inexpensive therapeutics to counter the high mortality inflicted by malaria. Here, we report a new class of antimalarial protein farnesyltransferase (PFT) inhibitors, designed with specific emphasis on simple molecular architecture, to facilitate easy access to therapies based on this recently validated antimalarial target. This novel series of compounds represents the first Plasmodium falciparum selective PFT inhibitors reported (up to 145-fold selectivity), with lead inhibitors displaying excellent in vitro activity (IC(50) < 1 nM) and toxicity to cultured parasites at low concentrations (ED(50) < 100 nM). Initial studies of absorption, metabolism, and oral bioavailability are reported.

Antiproliferative and phenotype-transforming antitumor agents derived from cysteine.

Selective destruction of malignant tumor cells without damaging normal cells is an important goal for cancer chemotherapy in the 21st century. Differentiating agents that transform cancer cells to either a nonproliferating or normal phenotype could potentially be tissue-specific and avoid side effects of current drugs. However, most compounds that are presently known to differentiate cancer cells are histone deacetylase inhibitors that are of low potency or suffer from low bioavailability, rapid metabolism, reversible differentiation, and nonselectivity for cancer cells over normal cells. Here we describe 36 nonpeptidic compounds derived from a simple cysteine scaffold, fused at the C-terminus to benzylamine, at the N-terminus to a small library of carboxylic acids, and at the S-terminus to 4-butanoyl hydroxamate. Six compounds were cytotoxic at nanomolar concentrations against a particularly aggressive human melanoma cell line (MM96L), four compounds showed selectivities of > or =5:1 for human melanoma over normal human cells (NFF), and four of the most potent compounds were further tested and found to be cytotoxic for six other human cancer cell lines (melanomas SK-MEL-28, DO4; prostate DU145; breast MCF-7; ovarian JAM, CI80-13S). The most active compounds typically caused hyperacetylation of histones, induced p21 expression, and reverted phenotype of surviving tumor cells to a normal morphology. Only one compound was given orally at 5 mg/kg to healthy rats to look for bioavailability, and it showed reasonably high levels in plasma (C(max) 6 microg/mL, T(max) 15 min) for at least 4 h. Results are sufficiently promising to support further work on refining this and related classes of compounds to an orally active, more tumor-selective, antitumor drug.

Conformationally homogeneous cyclic tetrapeptides: useful new three-dimensional scaffolds.

The most commonly recognized motifs in protein-protein interactions are gamma and beta turns, which are defined by three to four contiguous amino acids in a peptide sequence. Cyclic tetrapeptides thus represent minimalist turn mimetics, but their usefulness is compromised by strain in their 12-membered rings, making them difficult to cyclize, unstable to hydrolysis/metabolism, and conformationally heterogeneous in polar solvents. Appropriate placement of a beta amino acid in a tetrapeptide creates a 13-membered ring that is shown to be easier to cyclize, hydrolytically more stable, and conformationally homogeneous in polar solvents such as DMSO and water. Three-dimensional structures reveal that these cyclic tetrapeptides are novel rigid scaffolds, their unique side-chain projections matching a structurally diverse range of useful nonpeptidic templates, including sugars and spirocyclic compounds, found as components of natural products. The results provide a potentially useful link between protein architecture and organic natural products. On the basis of protein turn sequences (not protein structures) alone simple cyclic tetrapeptide libraries with a beta amino acid can be rationally designed as conformationally restricted, easily synthesized, and stereochemically controlled screening tools for rapidly identifying pharmacophore space that can then be computer-matched to more complex known natural product templates for drug development.

Mimetics of the peptide beta-strand.

Bioactive structures of peptides represent important clues for drug discovery and development although peptides themselves have substantial limitations as drugs. One promising approach to overcoming the limitations of peptides is to progressively replace amide bonds in peptides with non-peptidic constraints that bring drug-like properties like stability and bioavailability to the molecules. These constraints can also be used to mould molecules into shapes which mimic key elements of protein secondary structure that confer bioactivity to protein surfaces. Preorganizing a molecule into the shape recognized by a receptor results in high affinity binding though a considerable entropy saving and is an effective approach to engineering highly bioactive drug leads. One peptide structure, the extended beta strand, has only recently been identified as a fundamental recognition element in physiological processes. Relatively few molecules have been described as constrained mimics of extended peptide conformations. We now summarize some approaches to mimicking peptide beta strands, and illustrate these with examples of bioactive, stable and bioavailable molecules that are conformationally biased to mimic the extended peptide beta strand.

Regioselective synthesis of antiparallel loops on a macrocyclic scaffold constrained by oxazoles and thiazoles.

The regioselective syntheses and structures are reported for two tris-macrocylic compounds, each possessing two antiparallel loops on a macrocyclic scaffold constrained by two oxazoles and two thiazoles. NMR solution structures show the loops projecting from the same face of the macrocycle. Such molecules are shown to be prototypes for mimicking multiple loops of proteins.[structure: see text]

Beta-strand mimicking macrocyclic amino acids: templates for protease inhibitors with antiviral activity.

New amino acids are reported in which component macrocycles are constrained to mimic tripeptides locked in a beta-strand conformation. The novel amino acids involve macrocycles functionalized with both an N- and a C-terminus enabling addition of appendages at either end to modify receptor affinity, selectivity, or membrane permeability. We show that the cycles herein are effective templates within inhibitors of HIV-1 protease. Eleven compounds originating from such bifunctionalized cyclic templates are potent inhibitors of HIV-1 protease (Ki 0.3-50 nM; pH 6.5, I = 0.1 M). Unlike normal peptides comprising amino acids, five of these macrocycle-containing compounds are potent antiviral agents with sub-micromolar potencies (IC(50) 170-900 nM) against HIV-1 replication in human MT2 cells. The most active antiviral agents are the most lipophilic, with calculated values of LogD(6.5) > or = 4. All molecules have a conformationally constrained 17-membered macrocyclic ring that has been shown to structurally mimic a tripeptide segment (Xaa)-(Val/Ile)-(Phe/Tyr) of a peptide substrate in the extended conformation. The presence of two trans amide bonds and a para-substituted aromatic ring prevents intramolecular hydrogen bonds and fixes the macrocycle in the extended conformation. Similarly constrained macrocycles may be useful templates for the creation of inhibitors for the many other proteins and proteases that recognize peptide beta-strands.