Effect of hydroxyl group configuration in hydroxyethylamine dipeptide isosteres on HIV protease inhibition. Evidence for multiple binding modes

Identification of HIV inhibitors guided by free energy perturbation calculations

Free energy perturbation (FEP) theory coupled to molecular dynamics (MD) or Monte Carlo (MC) statistical mechanics offers a theoretically precise method for determining the free energy differences of related biological inhibitors. Traditionally requiring extensive computational resources and expertise, it is only recently that its impact is being felt in drug discovery. A review of computer-aided anti-HIV efforts employing FEP calculations is provided here that describes early and recent successes in the design of human immunodeficiency virus type 1 (HIV-1) protease and non-nucleoside reverse transcriptase inhibitors. In addition, our ongoing work developing and optimizing leads for small molecule inhibitors of cyclophilin A (CypA) is highlighted as an update on the current capabilities of the field. CypA has been shown to aid HIV-1 replication by catalyzing the cis/trans isomerization of a conserved Gly-Pro motif in the Nterminal domain of HIV-1 capsid (CA) protein. In the absence of a functional CypA, e.g., by the addition of an inhibitor such as cyclosporine A (CsA), HIV-1 has reduced infectivity. Our simulations of acylurea-based and 1-indanylketone-based CypA inhibitors have determined that their nanomolar and micromolar binding affinities, respectively, are tied to their ability to stabilize Arg55 and Asn102. A structurally novel 1-(2,6-dichlorobenzamido) indole core was proposed to maximize these interactions. FEP-guided optimization, experimental synthesis, and biological testing of lead compounds for toxicity and inhibition of wild-type HIV-1 and CA mutants have demonstrated a dose-dependent inhibition of HIV-1 infection in two cell lines. While the inhibition is modest compared to CsA, the results are encouraging.

Synthesis and biological activity of potent HIV-1 protease inhibitors based on Phe-Pro dihydroxyethylene isosteres

Peptidomimetic inhibitors of HIV-1 PR are still a key resource in the fight against AIDS. Here we describe the synthesis and biological activity of HIV-1 PR inhibitors based on four novel dihydroxyethylene isosteres of the Phe-Pro and Pro-Pro dipeptides. The isosteres, containing four stereogenic centers, were synthesized in high yield and excellent stereoselectivity via the cyclization of epoxy amines derived from α-amino acids. The inhibitors were assembled by coupling the isosteres with suitable flanking groups and were screened against recombinant HIV PR showing activities in the subnanomolar to micromolar range. Two Phe-Pro-based inhibitors active at the nanomolar level were further investigated: both inhibitors combine the ability to suppress HIV-1 replication in infected MT-2 cells with low cytotoxicity against the same cells, thereby displaying a high therapeutic index. These results demonstrate the potential of the new Phe-Pro dihydroxyethylene isostere as a core unit of powerful HIV-1 PR inhibitors.

Minimalist and universal peptidomimetics

Many "new generation" peptidomimetics are designed to present amino acid side chains only; they do not have structural features that resemble peptide main chains. These types of molecules have frequently been presented in the literature as mimics of specific secondary structures. However, many "side-chain only" peptidomimetics do not rest in single conformational states, but exist in a limited number of freely interconverting forms. These different conformations may resemble different secondary structures, so referring to them as, for instance, turn- or helical-mimics understates the ways they could adapt to various binding situations. Sets of scaffolds that can be used to mimic aspects of nearly every secondary structure, i.e. universal peptidomimetics, can be constructed. These may assume a privileged place in library design, particularly in high throughput screening for pharmacological probes for which binding conformations, or even the target itself, is unknown at the time the library is designed (critical review, 101 references).

Universal peptidomimetics

This paper concerns peptidomimetic scaffolds that can present side chains in conformations resembling those of amino acids in secondary structures without incurring excessive entropic or enthalpic penalties. Compounds of this type are referred to here as minimalist mimics. The core hypothesis of this paper is that small sets of such scaffolds can be designed to analogue local pairs of amino acids (including noncontiguous ones) in any secondary structure; i.e., they are universal peptidomimetics. To illustrate this concept, we designed a set of four peptidomimetic scaffolds. Libraries based on them were made bearing side chains corresponding to many of the protein-derived amino acids. Modeling experiments were performed to give an indication of kinetic and thermodynamic accessibilities of conformations that can mimic secondary structures. Together, peptidomimetics based on these four scaffolds can adopt conformations that resemble almost any combination of local amino acid side chains in any secondary structure. Universal peptidomimetics of this kind are likely to be most useful in the design of libraries for high-throughput screening against diverse targets. Consequently, data arising from submission of these molecules to the NIH Molecular Libraries Small Molecule Repository (MLSMR) are outlined.

Design, synthesis and biological evaluation of novel dual inhibitors of acetylcholinesterase and beta-secretase

To explore novel effective drugs for the treatment of Alzheimer's disease (AD), a series of dual inhibitors of acetylcholineterase (AChE) and beta-secretase (BACE-1) were designed based on the multi-target-directed ligands strategy. Among them, inhibitor 28 exhibited good dual potency in enzyme inhibitory potency assay (BACE-1: IC(50)=0.567 microM; AChE: IC(50)=1.83 microM), and also showed excellent inhibitory effects on Abeta production of APP transfected HEK293 cells (IC(50)=98.7 nM) and mild protective effect against hydrogen peroxide (H(2)O(2))-induced PC12 cell injury. Encouragingly, intracerebroventricular injection of 28 into amyloid precursor protein (APP) transgenic mice caused a 29% reduction of Abeta(1-40) production. Therefore, 28 was demonstrated as a good lead compound for the further study and more importantly, the strategy of AChE and BACE-1 dual inhibitors might be a promising direction for developing novel drugs for AD patients.

Design, Synthesis, and Crystal Structure of Hydroxyethyl Secondary Amine-Based Peptidomimetic Inhibitors of Human β-Secretase

The design and synthesis of a novel series of potent and cell permeable peptidomimetic inhibitors of the human beta-secretase (BACE) are described. These inhibitors feature a hydroxyethyl secondary amine isostere and a novel aromatic ring replacement for the C-terminus. The crystal structure of BACE in complex with this hydroxyethyl secondary amine isostere inhibitor is also presented.

An unusual functional group interaction and its potential to reproduce steric and electrostatic features of the transition states of peptidolysis

The donor-acceptor interaction between a tertiary amine and an aldehyde, first observed among a select class of alkaloids, was deliberately established in a peptidomimetic (1a-c) to mimic features of the two principal transition states of peptide hydrolysis. Compounds 1a-c show preferential adoption in methanol and water of a 'folded' conformation displaying the interaction. Proportions of the folded form in MeOH range from 45% to 70% and can reach 84% in buffer. Significantly, three tendencies for the folded/unfolded equilibrium are observed: increasing solubility and polarity of the medium and decreasing temperature results in a higher extent of folding. In the absence of any parameter set available for this weak bond, no modeling studies were conducted to aid in the design of 1a-c. The successful straightforward synthesis of 1 and its folding and inhibition results with HIV-1 peptidase using FRET technology encourage studies to further pre-organize candidate molecules and to screen the structure space by modeling and parallel combinatorial chemistry.

Synthesis, antiviral activity, and pharmacokinetic evaluation of P3 pyridylmethyl analogs of oximinoarylsulfonyl HIV-1 protease inhibitors

As a continuation of the recently communicated discovery of oximinoarylsulfonamides as potent inhibitors of HIV-1 aspartyl protease, compounds bearing pyridylmethyl substituents at P3 were designed and synthesized. Potent analogs in this series provided low single-digit nanomolar EC50 values against both wild-type HIV and resistant mutant virus (A17), attenuated some 3- to 12-fold in the presence of 50% human serum. Pharmacokinetic results for compounds in this series showed good to excellent exposure when co-administered orally with an equal amount of ritonavir (5mg/kg each) in the rat, with average AUC >8 microg h/mL. Similar dosing in dog resulted in significantly lower plasma levels (average AUC <2 microg h/mL). The 3-pyridylmethyl analog 30 gave the best overall exposure (rat AUC=7.1 microg h/mL and dog AUC=4.9 microg h/mL), however, this compound was found to be a potent inhibitor of cytochrome P450 3A (Ki=2.4 nM).

Rational approach to AIDS drug design through structural biology

The discovery and development of more than a dozen drugs in the past 15 years for the treatment of AIDS offer an excellent example of progress in the field of rational drug design. At this time, the principal targets are reverse transcriptase and protease, enzymes encoded by the human immunodeficiency virus. The introduction of protease inhibitors, in particular, has drastically decreased the mortality and morbidity associated with AIDS. This review presents the methods used to develop such drugs and discusses the remaining problems, such as the rapid emergence of drug resistance.

Peptidomimetics of efflux pump inhibitors potentiate the activity of levofloxacin in Pseudomonas aeruginosa

Several classes of peptidomimetics of the efflux pump inhibitor D-ornithine-D-homophenylalanine-3-aminoquinoline (MC-02,595) have been prepared and evaluated for their ability to potentiate the activity of the fluoroquinolone levofloxacin in Pseudomonas aeruginosa. A number of the new analogues were as active or more active than the lead, demonstrating that a peptide backbone is not essential for activity.

Peptide interactions with G-protein coupled receptors

Peptide recognition by G-protein coupled receptors (GPCRs) is reviewed with an emphasis on the indirect approach used to determine the receptor-bound conformation of peptide ligands. This approach was developed in response to the lack of detailed structural information available for these receptors. Recent advances in the structural determination of rhodopsin (the GPCR of the visual system) by crystallography have provided a scaffold for homology modeling of the inactive state of a wide variety of GPCRs that interact with peptide messages. Additionally, the ability to mutate GPCRs and assay compounds of similar chemical structure to test a common binding site on the receptor provides a firm experimental basis for structure-activity studies. Recognition motifs, common in other well-studied systems such as proteolytic enzymes and major histocompatibility class receptors (MHC) are reviewed briefly to provide a basis of comparison. Finally, the development of true peptidomimetics is contrasted with nonpeptide ligands, discovered through combinatorial chemistry. In many systems, the evidence suggests that the peptide ligands bind at the interface between the transmembrane segments and the extracellular loops, while nonpeptide antagonists bind within the transmembrane segments. Plausible models of GPCRs and the mechanism by which they activate G-proteins on binding peptides are beginning to emerge.

Structure-based drug design approaches for predicting binding affinities of HIV1 protease inhibitors

Computational assessment of the binding affinity of enzyme inhibitors prior to synthesis is an important component of computer-assisted drug design (CADD) paradigms. The free energy perturbation (FEP) methodology is the most accurate means of estimating relative binding affinities between two inhibitors. However, due to its complexity and computation-intensive nature, practical applications are restricted to analysis of structurally-related inhibitors. Accordingly, there is a need for methods that enable rapid assessment of large number of structurally-unrelated molecules in a suitably accurate manner. In this review, the FEP method is compared with regression-based methods that employ multivariate models to assess the advantages of each in the estimation of relative binding affinities of inhibitors to an enzyme. Semiquantitative predictions of relative binding free energies of human immunodeficiency virus 1 (HIV1) protease inhibitors are also presented and compared with the corresponding FEP results. The results indicate that the regression-based methods and the FEP method are useful in the semi-quantitative and quantitative assessment of relative binding affinities of enzyme inhibitors, respectively, prior to synthesis.

Predicting relative binding affinities of non-peptide HIV protease inhibitors with free energy perturbation calculations

The relative binding free energies in HIV protease of haloperidol thioketal (THK) and three of its derivatives were examined with free energy calculations. THK is a weak inhibitor (IC50 = 15 microM) for which two cocrystal structures with HIV type 1 proteases have been solved [Rutenber, E. et al., J. Biol. Chem., 268 (1993) 15343]. A THK derivative with a phenyl group on C2 of the piperidine ring was expected to be a poor inhibitor based on experiments with haloperidol ketal and its 2-phenyl derivative (Caldera, P., personal communication). Our calculations predict that a 5-phenyl THK derivative, suggested based on examination of the crystal structure, will bind significantly better than THK. Although there are large error bars as estimated from hysteresis, the calculations predict that the 5-phenyl substituent is clearly favored over the 2-phenyl derivative as well as the parent compound. The unfavorable free energies of solvation of both phenyl THK derivatives relative to the parent compound contributed to their predicted binding free energies. In a third simulation, the change in binding free energy for 5-benzyl THK relative to THK was calculated. Although this derivative has a lower free energy in the protein, its decreased free energy of solvation increases the predicted delta delta G (bind) to the same range as that of the 2-phenyl derivative.

Asymmetric dihydroxylation route to a dipeptide isostere of a protease inhibitor: enantioselective synthesis of the core unit of ritonavir

An enantioselective synthesis of the dipeptide isostere of ritonavir has been accomplished utilizing Sharpless asymmetric hydroxylation as the key step.

Transition-State Mimetics for HIV Protease Inhibitors: Stereocontrolled Synthesis of Hydroxyethylene and Hydroxyethylamine Isosteres by Ester-Derived Titanium Enolate Syn and Anti-Aldol Reactions

Stereocontrolled syntheses of hydroxyethylene dipeptide isostere and aminoalkyl epoxides for hydroxyethylamine isosteres are described. The stereochemistry of both stereogenic centers of the aminoalkyl epoxides 10 and 15 as well as the gamma-lactone 17 was assembled by our recently developed highly selective ester-derived titanium enolate aldol reactions. The Ti-enolate of 6 reacted with (benzyloxy)acetaldehyde and cinnamaldehyde to provide the syn-aldol product 7 and anti-aldol product 12, respectively. Removal of the chiral template followed by Curtius rearrangement of the resulting acid provided the desired amine functionality. The present syntheses represent practical and enantioselective entry to a range of other dipeptide isosteres, which are not limited to amino acid derived substituents.

Inhibitors of HIV-1 protease: a major success of structure-assisted drug design

Retroviral protease (PR) from the human immunodeficiency virus type 1 (HIV-1) was identified over a decade ago as a potential target for structure-based drug design. This effort was very successful. Four drugs are already approved, and others are undergoing clinical trials. The techniques utilized in this remarkable example of structure-assisted drug design included crystallography, NMR, computational studies, and advanced chemical synthesis. The development of these drugs is discussed in detail. Other approaches to designing HIV-1 PR inhibitors, based on the concepts of symmetry and on the replacement of a water molecule that had been found tetrahedrally coordinated between the enzyme and the inhibitors, are also discussed. The emergence of drug-induced mutations of HIV-1 PR leads to rapid loss of potency of the existing drugs and to the need to continue the development process. The structural basis of drug resistance and the ways of overcoming this phenomenon are mentioned.

cis-1-Aminoindan-2-ol in Asymmetric Syntheses

A review on cis-1-aminoindan-2-ol derived asymmetric syntheses is described.

Exploring the energy landscapes of molecular recognition by a genetic algorithm: analysis of the requirements for robust docking of HIV-1 protease and FKBP-12 complexes

Energy landscapes of molecular recognition are explored by performing "semi-rigid" docking of FK-506 and rapamycin with the Fukisawa binding protein (FKBP-12), and flexible docking simulations of the Ro-31-8959 and AG-1284 inhibitors with HIV-1 protease by a genetic algorithm. The requirements of a molecular recognition model to meet thermodynamic and kinetic criteria of ligand-protein docking simultaneously are investigated using a family of simple molecular recognition energy functions. The critical factor that determines the success rate in predicting the structure of ligand-protein complexes is found to be the roughness of the binding energy landscape, in accordance with a minimal frustration principle. The results suggest that further progress in structure prediction of ligand-protein complexes can be achieved by designing molecular recognition energy functions that generate binding landscapes with reduced frustration.

Preparation of Aminoalkyl Chlorohydrin Hydrochlorides: Key Building Blocks for Hydroxyethylamine-Based HIV Protease Inhibitors

Enantiomerically pure N,N-dibenzyl-alpha-amino aldehydes reacted with (chloromethyl)lithium, generated in situ from bromochloromethane and lithium metal, to give predominantly erythro aminoalkyl epoxides. Treatment of the crude epoxides with aqueous hydrochloric acid gave crystalline (2S,3S)-N,N-dibenzylamino chlorohydrin hydrochlorides in 32-56% overall yield and high isomeric purity. These compounds are versatile synthetic intermediates for the preparation of hydroxyethylamine-based HIV protease inhibitors, either directly as such, or via conversion to the corresponding N-Boc-(2S,3S)-aminoalkyl epoxides. The processes described do not make use of hazardous reagents or intermediates, do not require chromatographic purifications, and are thus amenable to the preparation of large quantities of these versatile building blocks.

Design and synthesis of substrate-based peptidomimetic human immunodeficiency virus protease inhibitors containing the hydroxymethylcarbonyl isostere

The human immunodeficiency (HIV) codes for an aspartic protease known to be essential for retroviral maturation and replication. The HIV protease can recognize Phe-Pro and Tyr-Pro sequences as the virus-specific cleavage site. These features provided a basis for the rational design of selective HIV protease-targeted drugs for the treatment of acquired immunodeficiency syndrome (AIDS). HIV protease is formed from two identical 99 amino acid peptides. We replaced the two Cys residues by L-Ala to synthesize [Ala67,95]-HIV-1 protease by the solid phase method and then prepared [Tyr6,42, Nle36,46, (NHCH2COSCH2CO)51-52, Ala67,95] HIV-1 protease (NY-5 isolate) using the thioester chemical ligation method. Based on the substrate transition state, we designed and synthesized a novel class of HIV protease inhibitors containing an unnatural amino acid, (2S, 3S)-3-amino-2-hydroxy-4-phenylbutyric acid, named allophenylnorstatine (Apns) with a hydroxymethylcarbonyl (HMC) isostere. Among them, the conformationally constrained tripeptide kynostatin (KNI)-272 (iQoa-Mta-Apns-Thz-NHBut) was a highly selective and superpotent HIV protease inhibitor (Ki = 0.0055 nM). KNI-272 exhibited potent antiviral activities against both AZT-sensitive and -insensitive clinical HIV-1 isolates as well as HIV-2 with low cytotoxicity. After i.d. administration, bioavailability of KNI-272 was 42.3% in rats. Also, KNI-272 exhibited in vivo anti-HIV activities in human PBMC-SCID mice. The x-ray crystallography and molecular modeling studies showed that the HMC group in KNI-272 interacted excellently with the aspartic acid carboxyl groups of HIV protease active site in the essentially same hydrogen-bonding mode as the transition state. This result implies that the HMC isostere is an ideal transition-state mimic and contributes to the high activity of KNI-272.

A preference-based free-energy parameterization of enzyme-inhibitor binding. Applications to HIV-1-protease inhibitor design

The interface between protein receptor-ligand complexes has been studied with respect to their binary interatomic interactions. Crystal structure data have been used to catalogue surfaces buried by atoms from each member of a bound complex and determine a statistical preference for pairs of amino-acid atoms. A simple free energy model of the receptor-ligand system is constructed from these atom-atom preferences and used to assess the energetic importance of interfacial interactions. The free energy approximation of binding strength in this model has a reliability of about +/- 1.5 kcal/mol, despite limited knowledge of the unbound states. The main utility of such a scheme lies in the identification of important stabilizing atomic interactions across the receptor-ligand interface. Thus, apart from an overall hydrophobic attraction (Young L, Jernigan RL, Covell DG, 1994, Protein Sci 3:717-729), a rich variety of specific interactions is observed. An analysis of 10 HIV-1 protease inhibitor complexes is presented that reveals a common binding motif comprised of energetically important contacts with a rather limited set of atoms. Design improvements to existing HIV-1 protease inhibitors are explored based on a detailed analysis of this binding motif.