Search

The nature of the inhibition of thiol proteases by a new class of mechanism-based inhibitors, 1,5-diacylcarbohydrazides, is described. These potent, time-dependent, active-site spanning inhibitors include compounds that are selective for cathepsin K, a cysteine protease unique to osteoclasts. The 1,5-diacylcarbohydrazides are slow substrates for members of the papain superfamily with inhibition resulting from slow enzyme decarbamylation. Enzyme-catalyzed hydrolysis of 2,2'-N, N'-bis(benzyloxycarbonyl)-L- leucinylcarbohydrazide is accompanied by formation of a hydrazide-containing product and a carbamyl-enzyme intermediate that is sufficiently stable to be observed by mass spectrometry and NMR. Stopped-flow studies yield a saturation limited value of 43 s(-)(1) for the rate of cathepsin K acylation by 2,2'N, N'-bis(benzyloxycarbonyl)-L-leucinylcarbohydrazide. Inhibition potency varies among proteases tested as reflected by 2-3 orders of magnitude differences in K(i) and K(obs)/I, but all eventually form the same stable covalent intermediate. Reactivation rates are equivalent for all enzymes tested (1 x 10(-)(4) s(-)(1)), indicating hydrolysis of a common carbamyl-enzyme form. NMR spectroscopic studies with cathepsin K and 2,2'-N,N'-bis(benzyloxycarbonyl)-L-leucinylcarbohydrazide provide evidence of inhibitor cleavage to generate a covalent carbamyl-enzyme intermediate rather than a tetrahedral complex. The product Cbz-leu-hydrazide does not appear enzyme-bound after cleavage in the NMR spectra, suggesting that the stable inhibited form of the enzyme is the thioester complex. 1, 5-diacylcarbohydrazides represent a new class of unreactive cysteine protease inhibitors that share a common mechanism of action across members of the papain superfamily. Both S and S' subsite interactions are exploited in achieving high selectivity and potency.

Potential advantages of C2-symmetric inhibitors designed for the symmetric HIV-1 protease include high selectivity, potency, stability, and bioavailability. Pseudo-C2-symmetric monools and C2-symmetric diols, containing central hydroxymethylene and (R,R)-dihydroxyethylene moieties flanked by a variety of hydrophobic P1/P1' side chains, were studied as HIV-1 protease inhibitors. The monools and diols were synthesized in 8-10 steps from D-(+)-arabitol and D-(+)-mannitol, respectively. Monools with ethyl or isobutyl P1/P1' side chains were weak inhibitors of recombinant HIV-1 protease (Ki > 10 microM), while benzyl P1/P1' side chains afforded a moderately potent inhibitor (apparent Ki = 230 nM). Diols were 100-10,000x more potent than analogous monools, and a wider range of P1/P1' side chains led to potent inhibition. Both classes of compounds exhibited lower apparent Ki values under high-salt conditions. Surprisingly, monool and diol HIV-1 protease inhibitors were potent inhibitors of porcine pepsin, a prototypical asymmetric monomeric aspartic protease. These results were evaluated in the context of the pseudosymmetric structure of monomeric aspartic proteases and their evolutionary kinship with the retroviral proteases. The X-ray crystal structure of HIV-1 protease complexed with a symmetric diol was determined at 2.6 A. Contrary to expectations, the diol binds the protease asymmetrically and exhibits 2-fold disorder in the electron density map. Molecular dynamics simulations were conducted beginning with asymmetric and symmetric HIV-1 protease/inhibitor model complexes. A more stable trajectory resulted from the asymmetric complex, in agreement with the observed asymmetric binding mode. A simple four-point model was used to argue more generally that van der Waals and electrostatic force fields can commonly lead to an asymmetric association between symmetric molecules.

Simian immunodeficiency virus protease (SIV-PR) was produced in Escherichia coli with a recombinant expression system in which the mature enzyme autoprocessed from a precursor form. Recombinant SIV and HIV-1 (human immunodeficiency virus, type 1) proteases were purified from bacterial cell lysates by use of sequential steps of ammonium sulfate precipitation and size-exclusion and ion-exchange chromatography. The amino acid composition, amino-terminal sequence, and molecular weight (monomer) of the recombinant SIV-PR were in accord with that of the 99 amino acid polypeptide predicted from the SIVMac-PR nucleotide sequence. The active form of SIV-PR was shown to be dimeric by gel filtration chromatography. Inhibition by pepstatin A, time-dependent inactivation by 1,2-epoxy-3-(4-nitrophenoxy)propane, and pH rate profiles using oligopeptide substrates demonstrated that SIV-PR behaves as an aspartic protease. Recombinant HIV-1 Pr55gag precursor was processed in vitro by SIV-PR and HIV-1 PR with indistinguishable proteolytic patterns upon NaDodSO4-polyacrylamide gel electrophoresis. Oligopeptide substrates for HIV-1 PR were found to be suitable substrates for recombinant SIV-PR with the exception of a peptide containing the site identified for p66/p51 cleavage (Phe*Tyr) within HIV-1 reverse transcriptase (RT). Several synthetic peptide analogue inhibitors of HIV-1 PR were also potent inhibitors of SIV-PR, indicating that SIV infection in macaques and rhesus monkeys should be useful models for the preclinical evaluation of acquired immunodeficiency syndrome (AIDS) therapeutics targeted towards the virally encoded HIV-1 protease.

We have investigated the inhibitor specificity for the proteases of the human immunodeficiency viruses, types 1 and 2. Using a series of related inhibitors, the P1' side chain was confirmed to play a significant role in determining both the absolute and relative affinity for the enzymes. To further define the residues in the enzymes responsible for the difference in affinity, chimeric proteins were constructed in which domains of the respective proteases were exchanged at the genetic level. The results of these studies demonstrated that inhibitor affinity is conferred by a combination of the active site residues (32, 47, and 82) along with a loop comprised of residues 31 and 33-37, which lies outside of the active site cavity. These results are discussed in terms of existing structural data.

Analogues of peptides ranging in size from three to six amino acids and containing the hydroxyethylene dipeptide isosteres Phe psi Gly, Phe psi Ala, Phe psi NorVal, Phe psi Leu, and Phe psi Phe, where psi denotes replacement of CONH by (S)-CH(OH)CH2, were synthesized and studied as HIV-1 protease inhibitors. Inhibition constants (Ki) with purified HIV-1 protease depend strongly on the isostere in the order Phe psi Gly greater than Phe psi Ala greater than Phe psi NorVal greater than Phe psi Leu greater than Phe psi Phe and decrease with increasing length of the peptide analogue, converging to a value of 0.4 nM. Ki values are progressively less dependent on inhibitor length as the size of the P1' side chain within the isostere increases. The structures of HIV-1 protease complexed with the inhibitors Ala-Ala-X-Val-Val-OMe, where X is Phe psi Gly, Phe psi Ala, Phe psi NorVal, and Phe psi Phe, have been determined by X-ray crystallography (resolution 2.3-3.2 A). The crystals exhibit symmetry consistent with space group P6(1) with strong noncrystallographic 2-fold symmetry, and the inhibitors all exhibit 2-fold disorder. The inhibitors bind in similar conformations, forming conserved hydrogen bonds with the enzyme. The Phe psi Gly inhibitor adopts an altered conformation that places its P3' valine side chain partially in the hydrophobic S1' pocket, thus suggesting an explanation for the greater dependence of the Ki value on inhibitor length in the Phe psi Gly series. From the kinetic and crystallographic data, a minimal inhibitor model for tight-binding inhibition is derived in which the enzyme subsites S2-S2' are optimally occupied. The Ki values for several compounds are compared with their potencies as inhibitors of proteolytic processing in T-cell cultures chronically infected with HIV-1 (MIC values) and as inhibitors of acute infectivity (IC50 values). There is a rank-order correspondence, but a 20-1000-fold difference, between the values of Ki and those of MIC or IC50. IC50 values can approach those of Ki but are highly dependent on the conditions of the acute infectivity assay and are influenced by physiochemical properties of the inhibitors such as solubility.

The pH dependence of the peptidolytic reaction of recombinant human immunodeficiency virus type 1 protease has been examined over a pH range of 3-7 for four oligopeptide substrates and two competitive inhibitors. The pK values obtained from the pKis vs pH profiles for the unprotonated and protonated active-site aspartyl groups, Asp-25 and Asp-25', in the monoprotonated enzyme form were 3.1 and 5.2, respectively. Profiles of log V/K vs pH for all four substrates were "bell-shaped" in which the pK values for the unprotonated and protonated aspartyl residues were 3.4-3.7 and 5.5-6.5, respectively. Profiles of log V vs pH for these substrates were "wave-shaped" in which V was shifted to a constant lower value upon protonation of a residue of pK = 4.2-5.2. These results indicate that substrates bind only to a form of HIV-1 protease in which one of the two catalytic aspartyl residues is protonated. Solvent kinetic isotope effects were measured over a pH (D) range of 3-7 for two oligopeptide substrates, Ac-Arg-Ala-Ser-Gln-Asn-Tyr-Pro-Val-Val-NH2 and Ac-Ser-Gln-Asn-Tyr-Pro-Val-Val-NH2. The pH-independent value for DV/K was 1.0 for both substrates, and DV = 1.5-1.7 and 2.2-3.2 at low and high pH (D), respectively. The attentuation of both V and DV at low pH (D) is consistent with a change in rate-limiting step from a chemical one at high pH (D) to one in which a product release step or an enzyme isomerization step becomes partly rate-limiting at low pH (D). Proton inventory data is in accord with the concerted transfer of two protons in the transition state of a rate-limiting chemical step in which the enzyme-bound amide hydrate adduct collapses to form the carboxylic acid and amine products.

The peptidolytic reaction of HIV-1 protease has been investigated by using four oligopeptide substrates, Ac-Ser-Gln-Asn-Tyr-Pro-Val-Val-NH2, Ac-Arg-Ala-Ser-Gln-Asn-Tyr-Pro-Val-Val-NH2, Ac-Ser-Gln-Ser-Tyr-Pro-Val-Val-NH2, and Ac-Arg-Lys-Ile-Leu-Phe-Leu-Asp-Gly-NH2, that resemble two cleavage sites found within the naturally occurring polyprotein substrates Pr55gag and Pr160gag-pol. The values for the kinetic parameters V/KEt and V/Et were 0.16-7.5 mM-1 s-1 and 0.24-29 s-1, respectively, at pH 6.0, 0.2 M NaCl, and 37 degrees C. By use of a variety of inorganic salts, it was concluded that the peptidolytic reaction is nonspecifically activated by increasing ionic strength. V/K increased in an apparently parabolic fashion with increasing ionic strength, while V was either increased or decreased slightly. From product inhibition studies, the kinetic mechanism of the protease is either random or ordered uni-bi, depending on the substrate studied. The reverse reaction or a partial reverse reaction (as measured by isotope exchange of the carboxylic product into substrate) was negligible for most of the oligopeptide substrates, but the enzyme catalyzed the formation of Ac-Ser-Gln-Asn-Tyr-Phe-Leu-Asp-Gly-NH2 from the products Ac-Ser-Gln-Asn-Tyr and Phe-Leu-Asp-Gly-NH2. The protease-catalyzed exchange of an atom of 18O from H2 18O into the re-formed substrates occurred at a rate which was 0.01-0.12 times that of the forward peptidolytic reaction. The results of these studies are in accord with the formation of a kinetically competent enzyme-bound amide hydrate intermediate, the collapse of which is the rate-limiting chemical step in the reaction pathway.

Several divalent metal ions were used as kinetic probes of the beef heart mitochondrial adenosinetriphosphatase (F1) under a variety of conditions, and the relationship between the properties of the catalytic metal ion and the catalytic activity of the enzyme was examined. Vmax for ATP hydrolysis was largest when metal ions characterized by intermediate values of acidity of coordinated water molecules (pKa) and metal-nucleotide stability constants (Kstab) were present. As temperature increased, the peak of Vmax vs. pKa (or Kstab) shifted to lower initial values of pKa or Kstab. The solvent deuterium isotope effect on Vmax (DV) was normal and largest when the metal ion present during F1-catalyzed ATP hydrolysis was most acidic and the metal nucleotide stability constant was large. When an active site tyrosine on F1 was nitrated, Vmax was most affected when the metal ion present was least acidic and the metal nucleotide stability constant was small. The isotope effect on V/K (DV/K) was normal, small, and apparently independent of the metal ion present. ADP inhibition of F1-catalyzed ATP hydrolysis is competitive, and the Ki is independent of the metal ion present. The degree of Pi inhibition of F1 is dependent on the metal ion present. The inhibition by Pi is competitive at low temperature and becomes noncompetitive as temperature increases. These and previous results support a mechanism whereby a water molecule coordinated to the metal ion of an enzyme-bound gamma-monodentate metal-ATP complex is deprotonated to begin a series of events whereby a beta,gamma-bidentate metal-ATP complex is produced. Upon hydrolysis, the bond between the metal ion and the beta-phosphate of ADP in the Pi-metal-ADP complex is broken before products (ADP and metal-Pi) are released.