Comprehensive bioinformatic analysis of the specificity of human immunodeficiency virus type 1 protease

Rapidly developing viral resistance to licensed human immunodeficiency virus type 1 (HIV-1) protease inhibitors is an increasing problem in the treatment of HIV-infected individuals and AIDS patients. A rational design of more effective protease inhibitors and discovery of potential biological substrates for the HIV-1 protease require accurate models for protease cleavage specificity. In this study, several popular bioinformatic machine learning methods, including support vector machines and artificial neural networks, were used to analyze the specificity of the HIV-1 protease. A new, extensive data set (746 peptides that have been experimentally tested for cleavage by the HIV-1 protease) was compiled, and the data were used to construct different classifiers that predicted whether the protease would cleave a given peptide substrate or not. The best predictor was a nonlinear predictor using two physicochemical parameters (hydrophobicity, or alternatively polarity, and size) for the amino acids, indicating that these properties are the key features recognized by the HIV-1 protease. The present in silico study provides new and important insights into the workings of the HIV-1 protease at the molecular level, supporting the recent hypothesis that the protease primarily recognizes a conformation rather than a specific amino acid sequence. Furthermore, we demonstrate that the presence of 1 to 2 lysine residues near the cleavage site of octameric peptide substrates seems to prevent cleavage efficiently, suggesting that this positively charged amino acid plays an important role in hindering the activity of the HIV-1 protease.     

Fuzzy neural network for the control of high cell density cultivation of recombinant Escherichia coli

A five-layer fuzzy neural network (FNN) was developed for the control of fed-batch cultivation of recombinant Escherichia coli JM103 harboring plasmid pUR 2921. The FNN was believed to represent the membership functions of the fuzzy subsets and to implement fuzzy inference using previous experimental data. This FNN was then used for compensating the exponential feeding rate determined by the feedforward control element. The control system is therefore a feedforward-feedback type. The change in pH of the culture broth and the specific growth rate were used as the inputs to FNN to calculate the glucose feeding rate. A cell density of 84 g DWC/l in the fed-batch cultivation of the recombinant E. coli was obtained with this control strategy. Two different FNNs were then employed before and after induction to enhance plasmid-encoded β-galactosidase production. Before induction the specific growth rate was set as 0.31 h⁻¹, while it was changed to 0.1 h⁻¹ after induction. Compared to when only one FNN was used, the residual glucose concentration could be tightly controlled at an appropriate level by employing two FNNs, resulting in an increase in relative activity of β-galactosidase which was about four times greater. The present investigation demonstrates that a feedforward-feedback control strategy with FNN is a promising control strategy for the control of high cell density cultivation and high expression of a target gene in fed-batch cultivation of a recombinant strain.     

Substituent effects on P2-cyclopentyltetrahydrofuranyl urethanes: design, synthesis, and X-ray studies of potent HIV-1 protease inhibitors

The design, synthesis, and biological evaluation of novel C3-substituted cyclopentyltetrahydrofuranyl (Cp-THF)-derived HIV-1 protease inhibitors are described. Various C3-functional groups on the Cp-THF ligand were investigated in order to maximize the ligand-binding site interactions in the flap region of the protease. Inhibitors 3c and 3d have displayed the most potent enzyme inhibitory and antiviral activity. Both inhibitors have maintained impressive activity against a panel of multidrug resistant HIV-1 variants. A high-resolution X-ray crystal structure of 3c-bound HIV-1 protease revealed a number of important molecular insights into the ligand-binding site interactions.     

Artificial neural network method for predicting HIV protease cleavage sites in protein

Knowledge of the polyprotein cleavage sites by HIV protease will refine our understanding of its specificity, and the information thus acquired will be useful for designing specific and efficient HIV protease inhibitors. The search for inhibitors of HIV protease will be greatly expedited if one can find and accurate, robust, and rapid method for predicting the cleavage sites in proteins by HIV protease. In this paper, Kohonen’s self-organization model, which uses typical artificial neural networks, is applied to predict the cleavability of oligopeptides by proteases with multiple and extended specificity subsites. We selected HIV-1 protease as the subject of study. We chose 299 oligopeptides for the training set, and another 63 oligopeptides for the test set. Because of its high rate of correct prediction (58/63=92.06%) and stronger fault-tolerant ability, the neural network method should be a useful technique for finding effective inhibitors of HIV protease, which is one of the targets in designing potential drugs against AIDS. The principle of the artificial neural network method can also be applied to analyzing the specificity of any multisubsite enzyme.     

Artificial neural network method for predicting HIV protease cleavage sites in protein

Knowledge of the polyprotein cleavage sites by HIV protease will refine our understanding of its specificity, and the information thus acquired will be useful for designing specific and efficient HIV protease inhibitors. The search for inhibitors of HIV protease will be greatly expedited if one can find and accurate, robust, and rapid method for predicting the cleavage sites in proteins by HIV protease. In this paper, Kohonen’s self-organization model, which uses typical artificial neural networks, is applied to predict the cleavability of oligopeptides by proteases with multiple and extended specificity subsites. We selected HIV-1 protease as the subject of study. We chose 299 oligopeptides for the training set, and another 63 oligopeptides for the test set. Because of its high rate of correct prediction (58/63=92.06%) and stronger fault-tolerant ability, the neural network method should be a useful technique for finding effective inhibitors of HIV protease, which is one of the targets in designing potential drugs against AIDS. The principle of the artificial neural network method can also be applied to analyzing the specificity of any multisubsite enzyme.     

Design, synthesis, and biological evaluation of monopyrrolinone-based HIV-1 protease inhibitors possessing augmented P2' side chains

A series of monopyrrolinone-based HIV-1 protease inhibitors possessing rationally designed P2' side chains have been synthesized and evaluated for activity against wild-type HIV-1 protease. The most potent inhibitor displays subnanomolar potency in vitro for the wild-type HIV-1 protease. Additionally, the monopyrrolinone inhibitors retain potency in cellular assays against clinically significant mutant forms of the virus. X-ray structures of these inhibitors bound in the wild-type enzyme reveal important insights into the observed biological activity.     

Structure-activity relationship of HIV-1 protease inhibitors containing alpha-hydroxy-beta-amino acids. Detailed study of P1 site.

The structure-activity relationship of HIV-1 protease (HIV-1 PR) inhibitors containing alpha-hydroxy-beta-amino acids is discussed. We demonstrated that substituent groups on the P1 aromatic rings of the inhibitors exert significant influence on their biological activity. Inhibitors bearing an alkyl or a fluorine atom at the meta and para position on their P1 benzene ring were found to be good inhibitors. We also discovered that the substitution positions of the P2 benzamides were crucial for good antiviral potency. In this study, inhibitor 48 was the most potent [IC90 (CEM/HIV-1 IIIB) 27 nM] and showed good pharmacokinetics in rats.     

An immunoenzymatic solid-phase assay for quantitative determination of HIV-1 protease activity

A novel immunoenzymatic procedure for the quantitative determination of HIV protease activity is provided. An N-terminal biotinylated peptide (DU1) that comprises an HIV-1 protease (HIV-PR) cleavage sequence was bound to streptavidin-coated microtiter plates. The bound peptide can be quantified by an immunoenzymatic procedure (enzyme-linked immunosorbent assay, ELISA) that includes a monoclonal antibody (Mab 332) against the peptide (DU1) C-terminal. The incubation of the bound peptide with HIV-PR in solution resulted in a signal decrement, as the peptide was hydrolyzed and the released C-terminal segment washed away. An equation that relates the amount of added enzyme to the kinetics of the reaction was written in order to describe this heterogeneous enzyme-quasi-saturable system. This equation allows quantitative determination of protease activity, a feature widely underrated in previous similar assays. The assay also allows evaluation of the inhibitory activity of HIV-PR inhibitors. Due to the intrinsic advantages of the ELISA format, this method could be used in high-throughput screening of HIV protease inhibitors. The assay can be extended to other proteolytic enzymes.     

Design and synthesis of HIV-1 protease inhibitors. Novel tetrahydrofuran P2/P2'-groups interacting with Asp29/30 of the HIV-1 protease. Determination of binding from X-ray crystal structure of inhibitor protease complex

A series of HIV-1 protease inhibitors having new tetrahydrofuran P2/P2' groups have been synthesised and tested for protease inhibition and antiviral activity. Six novel 4-aminotetrahydrofuran derivatives were prepared starting from commercially available isopropylidene-alpha-D-xylofuranose yielding six symmetrical and six unsymmetrical inhibitors. Promising sub nanomolar HIV-1 protease inhibitory activities were obtained. The X-ray crystal structure of the most potent inhibitor (23, K(i) 0.25 nM) co-crystallised with HIV-1 protease is discussed and the binding compared with inhibitors 1a and 1b.     

A mechanistic study of 3-aminoindazole cyclic urea HIV-1 protease inhibitors using comparative QSAR

Comparative QSAR studies on P2/P2' and P1/P1' substituted symmetrical and nonsymmetrical 3-aminoindazole cyclic urea HIV-1 protease inhibitors were performed. The protease inhibitory activity of these compounds was found to decrease with larger and more hydrophobic molecules, whereas the antiviral potency and translation across the cell membrane increases with increase in hydrophobicity and size. These results provide mechanistic insight about the mode of interaction of these compounds with HIV-1 protease receptor and would help in further improving the biological activity.     

HIV-1 integrase pharmacophore model derived from diverse classes of inhibitors

A three-dimensional pharmacophore model has been generated for HIV-1 integrase (HIV-1 IN) from known inhibitors. A dataset consisting of 26 inhibitors was selected on the basis of the information content of the structures and activity data as required by the catalyst/HypoGen program. Our model was able to predict the activity of other known HIV-1 IN inhibitors not included in the model generation, and can be further used to identify structurally diverse compounds with desired biological activity by virtual screening.     

Discovery of potent HIV-1 protease inhibitors incorporating sulfoximine functionality

Based on the unique property of sulfoximine and the homodimeric C(2) structural symmetry of HIV-1 protease, a novel class of sulfoximine-based pseudosymmetric HIV-1 protease inhibitors was designed and synthesized. The sulfoximine moiety was demonstrated to be important for HIV-1 protease inhibitor potency. The most active stereoisomer (2S,2'S) displays a potency of 2.5 nM (IC(50)) against HIV-1 protease and an anti-HIV-1 activity of 408 nM (IC(50)). A possible mode of action is proposed.     

Crystal structures of a multidrug-resistant human immunodeficiency virus type 1 protease reveal an expanded active-site cavity

The goal of this study was to use X-ray crystallography to investigate the structural basis of resistance to human immunodeficiency virus type 1 (HIV-1) protease inhibitors. We overexpressed, purified, and crystallized a multidrug-resistant (MDR) HIV-1 protease enzyme derived from a patient failing on several protease inhibitor-containing regimens. This HIV-1 variant contained codon mutations at positions 10, 36, 46, 54, 63, 71, 82, 84, and 90 that confer drug resistance to protease inhibitors. The 1.8-angstrom (A) crystal structure of this MDR patient isolate reveals an expanded active-site cavity. The active-site expansion includes position 82 and 84 mutations due to the alterations in the amino acid side chains from longer to shorter (e.g., V82A and I84V). The MDR isolate 769 protease "flaps" stay open wider, and the difference in the flap tip distances in the MDR 769 variant is 12 A. The MDR 769 protease crystal complexes with lopinavir and DMP450 reveal completely different binding modes. The network of interactions between the ligands and the MDR 769 protease is completely different from that seen with the wild-type protease-ligand complexes. The water molecule-forming hydrogen bonds bridging between the two flaps and either the substrate or the peptide-based inhibitor are lacking in the MDR 769 clinical isolate. The S1, S1', S3, and S3' pockets show expansion and conformational change. Surface plasmon resonance measurements with the MDR 769 protease indicate higher k(off) rates, resulting in a change of binding affinity. Surface plasmon resonance measurements provide k(on) and k(off) data (K(d) = k(off)/k(on)) to measure binding of the multidrug-resistant protease to various ligands. This MDR 769 protease represents a new antiviral target, presenting the possibility of designing novel inhibitors with activity against the open and expanded protease forms.     

Design and synthesis of highly potent HIV-1 protease inhibitors with novel isosorbide-derived P2 ligands

The design, synthesis, and biological evaluation of a series of six HIV-1 protease inhibitors incorporating isosorbide moiety as novel P2 ligands are described. All the compounds are very potent HIV-1 protease inhibitors with IC₅₀ values in the nanomolar or picomolar ranges (0.05–0.43 nM). Molecular docking studies revealed the formation of an extensive hydrogen-bonding network between the inhibitor and the active site. Particularly, the isosorbide-derived P2 ligand is involved in strong hydrogen bonding interactions with the backbone atoms.     

Evaluation of triazolamers as active site inhibitors of HIV-1 protease

Proteases typically recognize their peptide substrates in extended conformations. General approaches for designing protease inhibitors often consist of peptidomimetics that feature this conformation. Herein we discuss a combination of computational and experimental studies to evaluate the potential of triazole-linked β-strand mimetics as inhibitors of HIV-1 protease activity.     

Characterization of a set of HIV-1 protease inhibitors using binding kinetics data from a biosensor-based screen

The interaction between 290 structurally diverse human immunodeficiency virus type 1 (HIV-1) protease inhibitors and the immobilized enzyme was analyzed with an optical biosensor. Although only a single concentration of inhibitor was used, information about the kinetics of the interaction could be obtained by extracting binding signals at discrete time points. The statistical correlation between the biosensor binding data, inhibition of enzyme activity (K(i)), and viral replication (EC(50)) revealed that the association and dissociation rates for the interaction could be resolved and that they were characteristic for the compounds. The most potent inhibitors, with respect to K(i) and EC(50) values, including the clinically used drugs, all exhibited fast association and slow dissociation rates. Selective or partially selective binders for HIV-1 protease could be distinguished from compounds that showed a general protein-binding tendency by using three reference target proteins. This biosensor-based direct binding assay revealed a capacity to efficiently provide high-resolution information on the interaction kinetics and specificity of the interaction of a set of compounds with several targets simultaneously.     

Nine Crystal Structures Determine the Substrate Envelope of the MDR HIV-1 Protease

Under drug selection pressure, emerging mutations render HIV-1 protease drug resistant, leading to the therapy failure in anti-HIV treatment. It is known that nine substrate cleavage site peptides bind to wild type (WT) HIV-1 protease in a conserved pattern. However, how the multidrug-resistant (MDR) HIV-1 protease binds to the substrate cleavage site peptides is yet to be determined. MDR769 HIV-1 protease (resistant mutations at residues 10, 36, 46, 54, 62, 63, 71, 82, 84, and 90) was selected for present study to understand the binding to its natural substrates. MDR769 HIV-1 protease was co-crystallized with nine substrate cleavage site hepta-peptides. Crystallographic studies show that MDR769 HIV-1 protease has an expanded substrate envelope with wide open flaps. Furthermore, ligand binding energy calculations indicate weaker binding in MDR769 HIV-1 protease-substrate complexes. These results help in designing the next generation of HIV-1 protease inhibitors by targeting the MDR HIV-1 protease.     

Predicting human immunodeficiency virus protease cleavage sites in nonlinear projection space

HIV-1 protease has a broad and complex substrate specificity. The discovery of an accurate, robust, and rapid method for predicting the cleavage sites in proteins by HIV protease would greatly expedite the search for inhibitors of HIV protease. During the last two decades, various methods have been developed to explore the specificity of HIV protease cleavage activity. However, because little advancement has been made in the understanding of HIV-1 protease cleavage site specificity, not much progress has been reported in either extracting effective methods or maintaining high prediction accuracy. In this article, a theoretical framework is developed, based on the kernel method for dimensionality reduction and prediction for HIV-1 protease cleavage site specificity. A nonlinear dimensionality reduction kernel method, based on manifold learning, is proposed to reduce the high dimensions of protease specificity. A support vector machine is applied to predict the protease cleavage. Superior performance in comparison to that previously published in literature is obtained using numerical simulations showing that the basic specificities of the HIV-1 protease are maintained in reduction feature space, and by combining the nonlinear dimensionality reduction algorithm with a support vector machine classifier.