Mechanistic information and structure-based design methods have been used to design a series of nonpeptide cyclic ureas that are potent inhibitors of human immunodeficiency virus (HIV) protease and HIV replication. A fundamental feature of these inhibitors is the cyclic urea carbonyl oxygen that mimics the hydrogen-bonding features of a key structural water molecule. The success of the design in both displacing and mimicking the structural water molecule was confirmed by x-ray crystallographic studies. Highly selective, preorganized inhibitors with relatively low molecular weight and high oral bioavailability were synthesized.

High-resolution X-ray structures of the complexes of HIV-1 protease (HIV-1PR) with peptidomimetic inhibitors reveal the presence of a structural water molecule which is hydrogen bonded to both the mobile flaps of the enzyme and the two carbonyls flanking the transition-state mimic of the inhibitors. Using the structure−activity relationships of C 2-symmetric diol inhibitors, computed-aided drug design tools, and first principles, we designed and synthesized a novel class of cyclic ureas that incorporates this structural water and preorganizes the side chain residues into optimum binding conformations. Conformational analysis suggested a preference for pseudodiaxial benzylic and pseudodiequatorial hydroxyl substituents and an enantiomeric preference for the RSSR stereochemistry. The X-ray and solution NMR structure of the complex of HIV-1PR and one such cyclic urea, DMP323, confirmed the displacement of the structural water. Additionally, the bound and “unbound” (small-molecule X-ray) ligands have similar conformations. The high degree of preorganization, the complementarity, and the entropic gain of water displacement are proposed to explain the high affinity of these small molecules for the enzyme. The small size probably contributes to the observed good oral bioavailability in animals. Extensive structure-based optimization of the side chains that fill the S2 and S2‘ pockets of the enzyme resulted in DMP323, which was studied in phase I clinical trials but found to suffer from variable pharmacokinetics in man. This report details the synthesis, conformational analysis, structure−activity relationships, and molecular recognition of this series of C 2-symmetry HIV-1PR inhibitors. An initial series of cyclic ureas containing nonsymmetric P2/P2‘ is also discussed.

No abstract

These results provide a first look into the structure and catalytic residues of a non-metal dehydratase, a large class of hitherto structurally uncharacterized enzymes. It is envisaged that a detailed structural description of scytalone dehydratase will assist in the design of new inhibitors for controlling rice blast disease.

We present a computational method for prediction of the conformation of a ligand when bound to a macromolecular receptor. The method is intended for use in systems in which the approximate location of the binding site is known and no large-scale rearrangements of the receptor are expected upon formation of the complex. The ligand is initially placed in the vicinity of the binding site and the atomic motions of the ligand and binding site are explicitly simulated, with solvent represented by an implicit solvation model and using a grid representation for the bulk of the receptor protein. These two approximations make the method computationally efficient and yet maintain accuracy close to that of an all-atom calculation. For the benzamidine/trypsin system, we ran 100 independent simulations, in many of which the ligand settled into the low-energy conformation observed in the crystal structure of the complex. The energy of these conformations was lower than and well-separated from that of others sampled. Extensions of this method are also discussed.

In cell cultures, the key residues associated with HIV-1 resistance to cyclic urea-based HIV-1 protease (PR) inhibitors are Val82 and Ile84 of HIV-1 PR. To gain an understanding of how these two residues modulate inhibitor binding, we have measured the Ki values of three recombinant mutant proteases, I84V, V82F, and V82F/I84V, for DMP323 and DMP450, and determined the three-dimensional structures of their complexes to 2.1-1.9 A resolution with R factors of 18.7-19.6%. The Ki values of these mutants increased by 25-, 0.5-, and 1000-fold compared to the wild-type values of 0.8 and 0.4 nM for DMP323 and DMP450, respectively. The wild-type and mutant complexes overall are very similar (rms deviations of 0.2-0.3 A) except for differences in the patterns of their van der Waals (vdw) interactions, which appear to modulate the Ki values of the mutants. The loss of the CD1 atom of Ile84, in the I84V mutant complexes, creates a hole in the S1 subsite, reducing the number of vdw contacts and increasing the Ki values. The V82F mutant binds DMP323 more tightly than wild type because the side chain of Phe82 forms additional vdw and edge-to-face interactions with the P1 group of DMP323. The Ki values of the single mutants are not additive because the side chain of Phe82 rotates out of the S1 subsite in the double mutant (the chi 1 angles of Phe82 and -182 in the V82F and V82F/I84V mutants differ by 90 and 185 degrees, respectively), further reducing the vdw interactions. Finally, compensatory shifts in the I84V and V82F/ I84V complexes pick up a small number of new contacts, but too few to offset the initial loss of interactions caused by the mutations. Therefore, our data suggest that variants persist in the presence of DMP323 and DMP450 because of a decrease in vdw interactions between the mutant proteases and inhibitors.

Protease inhibitors are another class of compounds for treatment of human immunodeficiency virus (E1IV)-caused diease. The emergence of resitance to the current antl-HIV drugs makes the determination of potential resistance to protease inhibitors imperative. Here we describe the isolation of an HIV type 1 (HIV-1) resistant to an IHV-protease inhibitor.Serial passage of HIV-1 (strain RF) in the presence of the inhibitor, [2-pyridylacetylisoleucylphenylalanyl-4(CHOH)J2 (P9941), failed to yield a stock of virus with a r t phenotype. However, variants of the virus with 6-to 8-fold reduced sensitivity to P9941 were selcted by using a combination ofplaque assay and endpoint titration. Genetic analysis and computer modeing of the variant proteases revealed a single change in the codon for amino acid 82 (Val -) Ala), which resulted in a protease with lower affinity and reduced sensitivity to this inhibitor and certain, but not all, related inhibitors.

The aspartyl dyad of free HIV-1 protease has apparent pK(a)s of approximately 3 and approximately 6, but recent NMR studies indicate that the aspartyl dyad is fixed in the doubly protonated form over a wide pH range when cyclic urea inhibitors are bound, and in the monoprotonated form when the inhibitor KNI-272 is bound. We present computations and measurements related to these changes in protonation and to the thermodynamic linkage between protonation and inhibition. The Poisson-Boltzmann model of electrostatics is used to compute the apparent pK(a)s of the aspartyl dyad in the free enzyme and in complexes with four different inhibitors. The calculations are done with two parameter sets. One assigns epsilon = 4 to the solute interior and uses a detailed model of ionization; the other uses epsilon = 20 for the solute interior and a simplified representation of ionization. For the free enzyme, both parameter sets agree well with previously measured apparent pK(a)s of approximately 3 and approximately 6. However, the calculations with an internal dielectric constant of 4 reproduce the large pKa shifts upon binding of inhibitors, but the calculations with an internal dielectric constant of 20 do not. This observation has implications for the accurate calculation of pK(a)s in complex protein environments. Because binding of a cyclic urea inhibitor shifts the pK(a)s of the aspartyl dyad, changing the pH is expected to change its apparent binding affinity. However, we find experimentally that the affinity is independent of pH from 5.5 to 7.0. Possible explanations for this discrepancy are discussed.