Binding pathways of ligands to HIV-1 protease: coarse-grained and atomistic simulations

Multiscale Approach for Computing Gated Ligand Binding from Molecular Dynamics and Brownian Dynamics Simulations

We develop an approach to characterize the effects of gating by a multiconformation protein consisting of macrostate conformations that are either accessible or inaccessible to ligand binding. We first construct a Markov state model of the apo-protein from atomistic molecular dynamics simulations from which we identify macrostates and their conformations, compute their relative macrostate populations and interchange kinetics, and structurally characterize them in terms of ligand accessibility. We insert the calculated first-order rate constants for conformational transitions into a multistate gating theory from which we derive a gating factor γ that quantifies the degree of conformational gating. Applied to HIV-1 protease, our approach yields a kinetic network of three accessible (semi-open, open, and wide-open) and two inaccessible (closed and a newly identified, "parted") macrostate conformations. The parted conformation sterically partitions the active site, suggesting a possible role in product release. We find that the binding kinetics of drugs and drug-like inhibitors to HIV-1 protease falls in the slow gating regime. However, because γ = 0.75, conformational gating only modestly slows ligand binding. Brownian dynamics simulations of the diffusional association of eight inhibitors to the protease─having a wide range of experimental association constants (∼10⁴-10¹⁰ M^-1 s^-1)─yields gated rate constants in the range of ∼0.5-5.7 × 10⁸ M^-1 s^-1. This indicates that, whereas the association rate of some inhibitors could be described by the model, for many inhibitors either subsequent conformational transitions or alternate binding mechanisms may be rate-limiting. For systems known to be modulated by conformational gating, the approach could be scaled computationally efficiently to screen association kinetics for a large number of ligands.

Multiscale simulation approaches to modeling drug-protein binding

Simulations can provide detailed insight into the molecular processes involved in drug action, such as protein-ligand binding, and can therefore be a valuable tool for drug design and development. Processes with a large range of length and timescales may be involved, and understanding these different scales typically requires different types of simulation methodology. Ideally, simulations should be able to connect across scales, to analyze and predict how changes at one scale can influence another. Multiscale simulation methods, which combine different levels of treatment, are an emerging frontier with great potential in this area. Here we review multiscale frameworks of various types, and selected applications to biomolecular systems with a focus on drug-ligand binding.

Inhibition of the activity of HIV-1 protease through antibody binding and mutations probed by molecular dynamics simulations

HIV-1 protease is an essential enzyme in the life cycle of the HIV-1 virus. The conformational dynamics of the flap region of the protease is critical for the ligand binding mechanism, as well as for the catalytic activity. The monoclonal antibody F11.2.32 raised against HIV-1 protease inhibits its activity on binding. We have studied the conformational dynamics of protease in its free, inhibitor ritonavir and antibody bound forms using molecular dynamics simulations. We find that upon Ab binding to the epitope region (residues 36-46) of protease, the overall flexibility of the protease is decreased including the flap region and the active site, which is similar to the decrease in flexibility observed by inhibitor binding to the protease. This suggests an allosteric mechanism to inhibit protease activity. Further, the protease mutants G40E and G40R are known to have decreased activity and were also subjected to MD simulations. We find that the loss of flexibility in the mutants is similar to that observed in the protease bound to the Ab/inhibitor. These insights highlight the role played by dynamics in the function of the protease and how control of flexibility through Ab binding and site specific mutations can inhibit protease activity.

Enzymatic activity of human immunodeficiency virus type 1 protease in crowded solutions

Cells are crowded with various macromolecules and metabolites, which affect biochemical reactions in many ways, from the diffusion of substrates to catalytic activities of enzymes. We herein investigated the proteolytic activity of the human immunodeficiency virus type 1 protease (HIV-1 PR) under non-crowded and crowded conditions. The latter environment was mimicked with various (poly)ethylene glycol molecules as crowding agents. We found that these crowding agents affect the kinetic parameters of the HIV-1 PR catalyzed reaction by increasing the Michaelis-Menten constant and decreasing the maximum velocity. The influence of crowding was concentration dependent. We explain this effect by the dynamics of the HIV-1 PR flexible flaps that cover the peptide substrate binding site and are crucial for enzyme activity, and by a possibly slower substrate-enzyme association time in the crowded conditions.

Toward Multiscale Modeling of Proteins and Bioagglomerates: An Orientation-Sensitive Diffusion Model for the Integration of Molecular Dynamics and the Discrete Element Method

Most processes involved in biological and biotechnological systems spread over many scales in space and time. For example, the interaction of multiple enzymes in heterogeneous enzymatic agglomerates or clusters, necessary for efficient enzymatic conversion, is of high interest for research and enzyme engineering. In order to understand and predict their overall behavior and performance, it is important to describe these scales as completely as possible, known as multiscale modeling. While many different approaches have been presented in recent years, knowledge about protein formation and bioagglomeration at the micro scale is still very limited. In an attempt to address such systems, we propose a bottom- up multiscale modeling methodology, bridging the gaps between molecular dynamics (MD) with an explicit solvent and the larger scale discrete element method (DEM) using an implicit solvent and abstracting macromolecules (e.g., proteins) as objects with anisotropic properties. We term this approach the molecular discrete element method (MDEM). For this, we present an orientation-sensitive diffusion model for DEM, which describes the dynamics of anisotropic translational and rotational diffusion, while implicitly considering solvent molecules and enforcing a canonical ensemble. A general-purpose model and parametrization approach is presented, which can be used to simulate any process involving diffusion of discrete particles. Effects of temperature and viscosity changes can be considered, and guidance is provided concerning time step selection. This model is generally applicable and serves as a precondition to enforce the proper dynamics (i.e., diffusion characteristics and canonical ensemble, similar to a thermostat in MD) for the proposed multiscale modeling methodology with anisotropic properties. Thereby, it presents a first step toward modeling at the micro scale and is integral to enforcing dynamics of such systems and therefore extensively validated. As a next step, interaction models are to be defined and added to the presented model. In comparison to atomistic and coarse-grained (CG) MD, a speedup of 5-7 orders of magnitude can be achieved. The approach is demonstrated on multiple components of the pyruvate dehydrogenase enzyme complex, a multienzymatic machinery that involves very different types of enzymes and is of high value to further elucidate the mechanisms of bioagglomeration and metabolic channeling.

Computational modeling approaches to quantitative structure-binding kinetics relationships in drug discovery

Simple comparative correlation analyses and quantitative structure-kinetics relationship (QSKR) models highlight the interplay of kinetic rates and binding affinity as an essential feature in drug design and discovery. The choice of the molecular series, and their structural variations, used in QSKR modeling is fundamental to understanding the mechanistic implications of ligand and/or drug-target binding and/or unbinding processes. Here, we discuss the implications of linear correlations between kinetic rates and binding affinity constants and the relevance of the computational approaches to QSKR modeling.

Ligand Binding Pathways and Conformational Transitions of the HIV Protease

It is important to determine the binding pathways and mechanisms of ligand molecules to target proteins to effectively design therapeutic drugs. Molecular dynamics (MD) is a promising computational tool that allows us to simulate protein-drug binding at an atomistic level. However, the gap between the time scales of current simulations and those of many drug binding processes has limited the usage of conventional MD, which has been reflected in studies of the HIV protease. Here, we have applied a robust enhanced simulation method, Gaussian accelerated molecular dynamics (GaMD), to sample binding pathways of the XK263 ligand and associated protein conformational changes in the HIV protease. During two of 10 independent GaMD simulations performed over 500-2500 ns, the ligand was observed to successfully bind to the protein active site. Although GaMD-derived free energy profiles were not fully converged because of insufficient sampling of the complex system, the simulations still allowed us to identify relatively low-energy intermediate conformational states during binding of the ligand to the HIV protease. Relative to the X-ray crystal structure, the XK263 ligand reached a minimum root-mean-square deviation (RMSD) of 2.26 Å during 2.5 μs of GaMD simulation. In comparison, the ligand RMSD reached a minimum of only ∼5.73 Å during an earlier 14 μs conventional MD simulation. This work highlights the enhanced sampling power of the GaMD approach and demonstrates its wide applicability to studies of drug-receptor interactions for the HIV protease and by extension many other target proteins.

Mechanism of the Association Pathways for a Pair of Fast and Slow Binding Ligands of HIV-1 Protease

Equilibrium constants, together with kinetic rate constants of binding, are key factors in the efficacy and safety of drug compounds, informing drug design. However, the association pathways of protein-ligand binding, which contribute to their kinetic behaviors, are little understood. In this work, we used unbiased all-atom molecular dynamics (MD) simulations with an explicit solvent model to study the association processes of protein-ligand binding. Using the HIV protease (HIVp)-xk263 and HIVp-ritonavir protein-ligand systems as cases, we observed that ligand association is a multistep process involving diffusion, localization, and conformational rearrangements of the protein, ligand, and water molecules. Moreover, these two ligands preferred different routes of binding, which reflect two well-known binding mechanisms: induced-fit and conformation selection models. Our study shows that xk263 has a stronger capacity for desolvating surrounding water molecules, thereby inducing a semiopen conformation of the HIVp flaps (induced-fit model). In contrast, the slow dehydration characteristic of ritonavir allows for gradual association with the binding pocket of HIVp when the protein's flap conformation is fully open (conformation selection model). By studying the mechanism of ligand association and understanding the role of solvent molecules during the binding event, we can obtain a different perspective on the mechanism of macromolecule recognition, providing insights into drug discovery.

Enzyme Tunnels and Gates As Relevant Targets in Drug Design

Many enzymes contain tunnels and gates that are essential to their function. Gates reversibly switch between open and closed conformations and thereby control the traffic of small molecules-substrates, products, ions, and solvent molecules-into and out of the enzyme's structure via molecular tunnels. Many transient tunnels and gates undoubtedly remain to be identified, and their functional roles and utility as potential drug targets have received comparatively little attention. Here, we describe a set of general concepts relating to the structural properties, function, and classification of these interesting structural features. In addition, we highlight the potential of enzyme tunnels and gates as targets for the binding of small molecules. The different types of binding that are possible and the potential pharmacological benefits of such targeting are discussed. Twelve examples of ligands bound to the tunnels and/or gates of clinically relevant enzymes are used to illustrate the different binding modes and to explain some new strategies for drug design. Such strategies could potentially help to overcome some of the problems facing medicinal chemists and lead to the discovery of more effective drugs.

Developing HIV-1 Protease Inhibitors through Stereospecific Reactions in Protein Crystals

Protease inhibitors are key components in the chemotherapy of HIV infection. However, the appearance of viral mutants routinely compromises their clinical efficacy, creating a constant need for new and more potent inhibitors. Recently, a new class of epoxide-based inhibitors of HIV-1 protease was investigated and the configuration of the epoxide carbons was demonstrated to play a crucial role in determining the binding affinity. Here we report the comparison between three crystal structures at near-atomic resolution of HIV-1 protease in complex with the epoxide-based inhibitor, revealing an in-situ epoxide ring opening triggered by a pH change in the mother solution of the crystal. Increased pH in the crystal allows a stereospecific nucleophile attack of an ammonia molecule onto an epoxide carbon, with formation of a new inhibitor containing amino-alcohol functions. The described experiments open a pathway for the development of new stereospecific protease inhibitors from a reactive lead compound.

A Quantitative Model for cAMP Binding to the Binding Domain of MloK1

Ligand-protein binding processes are essential in biological systems. A well-studied system is the binding of cyclic adenosine monophosphate to the cyclic nucleotide binding domain of the bacterial potassium channel MloK1. Strikingly, the measured on-rate for cyclic adenosine monophosphate binding is two orders of magnitude slower than a simple Smoluchowski diffusion model would suggest. To resolve this discrepancy and to characterize the ligand-binding path in structural and energetic terms, we calculated 1100 ligand-binding molecular dynamics trajectories and tested two scenarios: In the first scenario, the ligand transiently binds to the protein surface and then diffuses along the surface into the binding site. In the second scenario, only ligands that reach the protein surface in the vicinity of the binding site proceed into the binding site. Here, a binding funnel, which increasingly confines the translational as well as the rotational degrees of freedom, determines the binding pathways and limits the on-rate. From the simulations, we identified five surface binding states and calculated the rates between these surface binding states, the binding site, and the bulk. We find that the transient binding of the ligands to the surface binding states does not affect the on-rate, such that this effect alone cannot explain the observed low on-rate. Rather, by quantifying the translational and rotational degrees of freedom and by calculating the binding committor, our simulations confirmed the existence of a binding funnel as the main bottleneck. Direct binding via the binding funnel dominates the binding kinetics, and only ∼10% of all ligands proceed via the surface into the binding site. The simulations further predict an on-rate between 15 and 40μs-1(mol/l)-1, which agrees with the measured on-rate.

Distributions of experimental protein structures on coarse-grained free energy landscapes

Predicting conformational changes of proteins is needed in order to fully comprehend functional mechanisms. With the large number of available structures in sets of related proteins, it is now possible to directly visualize the clusters of conformations and their conformational transitions through the use of principal component analysis. The most striking observation about the distributions of the structures along the principal components is their highly non-uniform distributions. In this work, we use principal component analysis of experimental structures of 50 diverse proteins to extract the most important directions of their motions, sample structures along these directions, and estimate their free energy landscapes by combining knowledge-based potentials and entropy computed from elastic network models. When these resulting motions are visualized upon their coarse-grained free energy landscapes, the basis for conformational pathways becomes readily apparent. Using three well-studied proteins, T4 lysozyme, serum albumin, and sarco-endoplasmic reticular Ca(2+) adenosine triphosphatase (SERCA), as examples, we show that such free energy landscapes of conformational changes provide meaningful insights into the functional dynamics and suggest transition pathways between different conformational states. As a further example, we also show that Monte Carlo simulations on the coarse-grained landscape of HIV-1 protease can directly yield pathways for force-driven conformational changes.

Investigation of Structural Dynamics of Enzymes and Protonation States of Substrates Using Computational Tools

This review discusses the use of molecular modeling tools, together with existing experimental findings, to provide a complete atomic-level description of enzyme dynamics and function. We focus on functionally relevant conformational dynamics of enzymes and the protonation states of substrates. The conformational fluctuations of enzymes usually play a crucial role in substrate recognition and catalysis. Protein dynamics can be altered by a tiny change in a molecular system such as different protonation states of various intermediates or by a significant perturbation such as a ligand association. Here we review recent advances in applying atomistic molecular dynamics (MD) simulations to investigate allosteric and network regulation of tryptophan synthase (TRPS) and protonation states of its intermediates and catalysis. In addition, we review studies using quantum mechanics/molecular mechanics (QM/MM) methods to investigate the protonation states of catalytic residues of β-Ketoacyl ACP synthase I (KasA). We also discuss modeling of large-scale protein motions for HIV-1 protease with coarse-grained Brownian dynamics (BD) simulations.

Flap Dynamics in Aspartic Proteases: A Computational Perspective

Recent advances in biochemistry and drug design have placed proteases as one of the critical target groups for developing novel small-molecule inhibitors. Among all proteases, aspartic proteases have gained significant attention due to their role in HIV/AIDS, malaria, Alzheimer's disease, etc. The binding cleft is covered by one or two β-hairpins (flaps) which need to be opened before a ligand can bind. After binding, the flaps close to retain the ligand in the active site. Development of computational tools has improved our understanding of flap dynamics and its role in ligand recognition. In the past decade, several computational approaches, for example molecular dynamics (MD) simulations, coarse-grained simulations, replica-exchange molecular dynamics (REMD) and metadynamics, have been used to understand flap dynamics and conformational motions associated with flap movements. This review is intended to summarize the computational progress towards understanding the flap dynamics of proteases and to be a reference for future studies in this field.

Switches of hydrogen bonds during ligand-protein association processes determine binding kinetics

Revealing the processes of ligand-protein associations deepens our understanding of molecular recognition and binding kinetics. Hydrogen bonds (H-bonds) play a crucial role in optimizing ligand-protein interactions and ligand specificity. In addition to the formation of stable H-bonds in the final bound state, the formation of transient H-bonds during binding processes contributes binding kinetics that define a ligand as a fast or slow binder, which also affects drug action. However, the effect of forming the transient H-bonds on the kinetic properties is little understood. Guided by results from coarse-grained Brownian dynamics simulations, we used classical molecular dynamics simulations in an implicit solvent model and accelerated molecular dynamics simulations in explicit waters to show that the position and distribution of the H-bond donor or acceptor of a drug result in switching intermolecular and intramolecular H-bond pairs during ligand recognition processes. We studied two major types of HIV-1 protease ligands: a fast binder, xk263, and a slow binder, ritonavir. The slow association rate in ritonavir can be attributed to increased flexibility of ritonavir, which yields multistep transitions and stepwise entering patterns and the formation and breaking of complex H-bond pairs during the binding process. This model suggests the importance of conversions of spatiotemporal H-bonds during the association of ligands and proteins, which helps in designing inhibitors with preferred binding kinetics.

Discover binding pathways using the sliding binding-box docking approach: application to binding pathways of oseltamivir to avian influenza H5N1 neuraminidase

Drug binding and unbinding are transient processes which are hardly observed by experiment and difficult to analyze by computational techniques. In this paper, we employed a cost-effective method called "pathway docking" in which molecular docking was used to screen ligand-receptor binding free energy surface to reveal possible paths of ligand approaching protein binding pocket. A case study was applied on oseltamivir, the key drug against influenza a virus. The equilibrium pathways identified by this method are found to be similar to those identified in prior studies using highly expensive computational approaches.

Molecular basis for the dissociation dynamics of protein A-immunoglobulin G1 complex

Staphylococcus aureus protein A (SpA) is the most popular affinity ligand for immunoglobulin G1 (IgG1). However, the molecular basis for the dissociation dynamics of SpA-IgG1 complex is unclear. Herein, coarse-grained (CG) molecular dynamics (MD) simulations with the Martini force field were used to study the dissociation dynamics of the complex. The CG-MD simulations were first verified by the agreement in the structural and interactional properties of SpA and human IgG1 (hIgG1) in the association process between the CG-MD and all-atom MD at different NaCl concentrations. Then, the CG-MD simulation studies focused on the molecular insight into the dissociation dynamics of SpA-hIgG1 complex at pH 3.0. It is found that there are four steps in the dissociation process of the complex. First, there is a slight conformational adjustment of helix II in SpA. This is followed by the phenomena that the electrostatic interactions provided by the three hot spots (Glu143, Arg146 and Lys154) of helix II of SpA break up, leading to the dissociation of helix II from the binding site of hIgG1. Subsequently, breakup of the hydrophobic interactions between helix I (Phe132, Tyr133 and His137) in SpA and hIgG1 occurs, resulting in the disengagement of helix I from its binding site of hIgG1. Finally, the non-specific interactions between SpA and hIgG1 decrease slowly till disappearance, leading to the complete dissociation of the SpA-hIgG1 complex. This work has revealed that CG-MD coupled with the Martini force field is an effective method for studying the dissociation dynamics of protein-protein complex.

Combining coarse-grained protein models with replica-exchange all-atom molecular dynamics

We describe a combination of all-atom simulations with CABS, a well-established coarse-grained protein modeling tool, into a single multiscale protocol. The simulation method has been tested on the C-terminal beta hairpin of protein G, a model system of protein folding. After reconstructing atomistic details, conformations derived from the CABS simulation were subjected to replica-exchange molecular dynamics simulations with OPLS-AA and AMBER99sb force fields in explicit solvent. Such a combination accelerates system convergence several times in comparison with all-atom simulations starting from the extended chain conformation, demonstrated by the analysis of melting curves, the number of native-like conformations as a function of time and secondary structure propagation. The results strongly suggest that the proposed multiscale method could be an efficient and accurate tool for high-resolution studies of protein folding dynamics in larger systems.

Ligand Binding Pathway Elucidation for Cryptophane Host-Guest Complexes

Modeling binding pathways can provide insight into molecular recognition, including kinetic mechanisms, barriers to binding, and gating effects. This work represents a novel computational approach, Hopping Minima, for the determination of conformational transitions of single molecules as well as binding pathways for molecular complexes. The method begins by thoroughly sampling a set of conformational minima for a molecular system. The natural motions of the system are modeled using the normal modes of the sampled minima. The natural motions are utilized to connect conformational minima and are finally combined to form association/binding pathways in the case of molecular complexes. We provide an implementation and example application of the method using alanine dipeptide and a set of chemical host-guest systems: two cryptophane hosts with two guest cations, trimethylammonium and tetramethylammonium. Our results demonstrate that conformational transitions can be modeled and extended to find binding pathways as well as energetic information relevant to the minimum conformations involved. This approach has advantages over simulation-based methods for studying systems with slow binding processes and can help design molecules with preferred binding kinetics.

A generic force field for protein coarse-grained molecular dynamics simulation

Coarse-grained (CG) force fields have become promising tools for studies of protein behavior, but the balance of speed and accuracy is still a challenge in the research of protein coarse graining methodology. In this work, 20 CG beads have been designed based on the structures of amino acid residues, with which an amino acid can be represented by one or two beads, and a CG solvent model with five water molecules was adopted to ensure the consistence with the protein CG beads. The internal interactions in protein were classified according to the types of the interacting CG beads, and adequate potential functions were chosen and systematically parameterized to fit the energy distributions. The proposed CG force field has been tested on eight proteins, and each protein was simulated for 1000 ns. Even without any extra structure knowledge of the simulated proteins, the Cα root mean square deviations (RMSDs) with respect to their experimental structures are close to those of relatively short time all atom molecular dynamics simulations. However, our coarse grained force field will require further refinement to improve agreement with and persistence of native-like structures. In addition, the root mean square fluctuations (RMSFs) relative to the average structures derived from the simulations show that the conformational fluctuations of the proteins can be sampled.

Mechanistic insights into phosphopeptide--BRCT domain association: preorganization, flexibility, and phosphate recognition

Promiscuous proteins are commonly observed in biological systems, for example, in modular domains that recognize phosphopeptides during signal transduction. This promiscuous recognition is of fundamental interest in chemistry and biology but is challenging when designing phosphopeptides in silico for cell biology studies. To investigate promiscuous recognition and binding processes of phosphopeptides and the modular domain, we selected a domain essential in breast cancer-the breast-cancer-associated protein 1 (BRCA1) C-terminal (BRCT) repeats as our model system. We performed molecular dynamics simulations and detailed analyses of the dihedral space to study protein fluctuation and conformational changes with phosphopeptide binding. We also studied the association processes of phosphorylated and unphosphorylated peptides using Brownian dynamics with a coarse-grained model. We found that the BRCT domain is preorganized for phosphopeptide binding but has a moderate arrangement of side chains to form complexes with various types of phosphopeptides. Phosphopeptide binding restricts the system motion in general, while the nonpolar phosphopeptide becomes more flexible in the bound state. Our analysis found that the BRCT domain utilizes different mechanisms, usually termed lock and key, induced-fit, and population-shift/conformational-selection models, to recognize peptides with different features. Brownian dynamics simulations revealed that the charged phosphate group may not always accelerate peptide association processes, but it helps the phosphopeptide orient into binding pockets accurately and stabilizes the complex. This work provides insights into molecular recognition in the promiscuous protein system.

Identifying the molecular mechanics and binding dynamics characteristics of potent inhibitors to HIV-1 protease

Human immunodeficiency virus type 1 protease (HIV-1 PR) is one of the primary inhibition targets for chemotherapy of AIDS because of its critical role in the replication cycle of the HIV. In this work, a combinatory coarse-grained and atomistic simulation method was developed for dissecting molecular mechanisms and binding process of inhibitors to the active site of HIV-1 PR, in which 35 typical inhibitors were trialed. We found that the molecular size and stiffness of the inhibitors and the binding energy between the inhibitors and PR play important roles in regulating the binding process. Comparatively, the smaller and more flexible inhibitors have larger binding energy and higher binding rates; they even bind into PR without opening the flaps. In contrast, the larger and stiffer inhibitors have lower binding energy and lower binding rate, and their binding is subjected to the opening and gating of the PR flaps. Furthermore, the components of binding free energy were quantified and analyzed by their dependence on the molecular size, structures, and hydrogen bond networks of inhibitors. Our results also deduce significant dynamics descriptors for determining the quantitative structure and property relationship in potent drug ligands for HIV-1 PR inhibition.

Computational studies of darunavir into HIV-1 protease and DMPC bilayer: necessary conditions for effective binding and the role of the flaps

Human immunodeficiency virus type 1 protease (HIV-1 PR) is one of the main targets toward AIDS therapy. We have selected the potent drug darunavir and a weak inhibitor (fullerene analog) as HIV-1 PR substrates to compare protease's conformational features upon binding. Molecular dynamics (MD), molecular mechanics Poisson-Boltzmann surface area (MM-PBSA), and quantum-mechanical (QM) calculations indicated the importance of the stability of HIV-1 PR flaps toward effective binding: a weak inhibitor may induce flexibility to the flaps, which convert between closed and semiopen states. A water molecule in the darunavir-HIV-1 PR complex bridged the two flap tips of the protease through hydrogen bonding (HB) interactions in a stable structure, a feature that was not observed for the fullerene-HIV-1 PR complex. Additionally, despite that van der Waals interactions and nonpolar contribution to solvation favored permanent fullerene entrapment into the cavity, these interactions alone were not sufficient for effective binding; enhanced electrostatic interactions as observed in the darunavir-complex were the crucial component of the binding energy. An alternative pathway to the usual way of a ligand to access the cavity was also observed for both compounds. Each ligand entered the binding cavity through an opening between the one flap of the protease and a neighboring loop. This suggested that access to the cavity is not necessarily regulated by flap opening. Darunavir exerts its biological action inside the cell, after crossing the membrane barrier. Thus, we also initiated a study on the interactions between darunavir and the DMPC bilayer to reveal that the drug was accommodated inside the bilayer in conformations that resembled its structure into HIV-1 PR, being stabilized via HBs with the lipids and water molecules.

Molecular dynamics applied in drug discovery: the case of HIV-1 protease

Molecular dynamics (MD) is a way to computationally simulate the movement of particles and it is widely used to provide a dynamic perspective on biomolecules. Nowadays, the ever-growing computer power and the improvement in methodology further strengthen the role of MD in drug discovery. In this chapter, an overview of MD's application in drug discovery will be given first, using HIV-1 protease as an example. Then, the underlying theories of MD will be briefly outlined. The second half of this chapter will provide a practical protocol on how to simulate a soluble protein in solvent. All-atom simulation with either implicit solvent or explicit solvent will be covered. The former samples global conformational change more efficiently, and post-processing including angle/distance measurement, structural deviation measurement, Ramachandran plot, and secondary structure analysis will be introduced. The latter is more realistic/expensive and is generally used to finely examine local conformational rearrangement and water-mediated interactions. Post-processing including water density analysis will be described.

Gating and Intermolecular Interactions in Ligand-Protein Association: Coarse-Grained Modeling of HIV-1 Protease

Most biological processes are initiated or mediated by the association of ligands and proteins. This work studies multistep, ligand-protein association processes by Brownian dynamics simulations with coarse-grained models for HIV-1 protease (HIVp) and its neutral ligands. We report the average association times when the ligand concentration is 100 μM. The influence of crowding on the simulated binding time was also studied. HIVp has flexible loops that serve as a gate during the ligand binding processes. It is believed that the flaps are partially closed most of the time in its free state. To accelerate our simulations, we fixed a part of the HIVp and reparameterized our coarse-grained model, using atomistic molecular dynamics simulations, to reproduce the "gating" motions of HIVp. HIVp-ligand interactions changed the gating behavior of HIVp and helped ligands diffuse on HIVp surface to accelerate binding. The structural adjustment of the ligand toward its final stable state was the limiting step in the binding processes, which is highly system dependent. The intermolecular attraction between the ligands and crowder proteins contributes the most to the crowding effects. The results highlight broader implications in recognition pathways under more complex environment that considers molecular dynamics and conformational changes. This work brings insights into ligand-protein associations and is helpful in the design of targeted ligands.

Strength of hydrogen bond network takes crucial roles in the dissociation process of inhibitors from the HIV-1 protease binding pocket

To understand the underlying mechanisms of significant differences in dissociation rate constant among different inhibitors for HIV-1 protease, we performed steered molecular dynamics (SMD) simulations to analyze the entire dissociation processes of inhibitors from the binding pocket of protease at atomistic details. We found that the strength of hydrogen bond network between inhibitor and the protease takes crucial roles in the dissociation process. We showed that the hydrogen bond network in the cyclic urea inhibitors AHA001/XK263 is less stable than that of the approved inhibitor ABT538 because of their large differences in the structures of the networks. In the cyclic urea inhibitor bound complex, the hydrogen bonds often distribute at the flap tips and the active site. In contrast, there are additional accessorial hydrogen bonds formed at the lateral sides of the flaps and the active site in the ABT538 bound complex, which take crucial roles in stabilizing the hydrogen bond network. In addition, the water molecule W301 also plays important roles in stabilizing the hydrogen bond network through its flexible movement by acting as a collision buffer and helping the rebinding of hydrogen bonds at the flap tips. Because of its high stability, the hydrogen bond network of ABT538 complex can work together with the hydrophobic clusters to resist the dissociation, resulting in much lower dissociation rate constant than those of cyclic urea inhibitor complexes. This study may provide useful guidelines for design of novel potent inhibitors with optimized interactions.

Dynamic flaps in HIV-1 protease adopt unique ordering at different stages in the catalytic cycle

The flexibility of HIV-1 protease flaps is known to be essential for the enzymatic activity. Here we attempt to capture a multitude of conformations of the free and substrate-bound HIV-1 protease that differ drastically in their flap arrangements. The substrate binding process suggests the opening of active site gate in conjunction with a reversal of flap tip ordering, from the native semiopen state. The reversed-flap, open-gated enzyme readily transforms to a closed conformation after proper placement of the substrate into the binding cleft. After substrate processing, the closed state protease which possessed opposite flap ordering relative to the semiopen state, encounters another flap reversal via a second open conformation that facilitates the evolution of native semiopen state of correct flap ordering. The complicated transitional pathway, comprising of many high and low energy states, is explored by combining standard and activated molecular dynamics (MD) simulation techniques. The study not only complements the existing findings from X-ray, NMR, EPR, and MD studies but also provides a wealth of detailed information that could help the structure-based drug design process.

Gated Diffusion-controlled Reactions

The binding and active sites of proteins are often dynamically occluded by motion of the nearby polypeptide. A variety of theoretical and computational methods have been developed to predict rates of ligand binding and reactivity in such cases. Two general approaches exist, "protein centric" approaches that explicitly treat only the protein target, and more detailed dynamical simulation approaches in which target and ligand are both treated explicitly. This mini-review describes recent work in this area and some of the biological implications.

Minimalist models for proteins: a comparative analysis

The last decade has witnessed a renewed interest in the coarse-grained (CG) models for biopolymers, also stimulated by the needs of modern molecular biology, dealing with nano- to micro-sized bio-molecular systems and larger than microsecond timescale. This combination of size and timescale is, in fact, hard to access by atomic-based simulations. Coarse graining the system is a route to be followed to overcome these limits, but the ways of practically implementing it are many and different, making the landscape of CG models very vast and complex. In this paper, the CG models are reviewed and their features, applications and performances compared. This analysis, restricted to proteins, focuses on the minimalist models, namely those reducing at minimum the number of degrees of freedom without losing the possibility of explicitly describing the secondary structures. This class includes models using a single or a few interacting centers (beads) for each amino acid. From this analysis several issues emerge. The difficulty in building these models resides in the need for combining transferability/predictive power with the capability of accurately reproducing the structures. It is shown that these aspects could be optimized by accurately choosing the force field (FF) terms and functional forms, and combining different parameterization procedures. In addition, in spite of the variety of the minimalist models, regularities can be found in the parameters values and in FF terms. These are outlined and schematically presented with the aid of a generic phase diagram of the polypeptide in the parameter space and, hopefully, could serve as guidelines for the development of minimalist models incorporating the maximum possible level of predictive power and structural accuracy.

On the orientation of the catalytic dyad in aspartic proteases

The recent re-refinement of the X-ray structure of apo plasmepsin II from Plasmodium falciparum suggests that the two carboxylate groups in the catalytic dyad are noncoplanar, (Robbins et al., Acta Crystallogr D Biol Crystallogr 2009;65: 294-296) in remarkable contrast with the vast majority of structures of aspartic proteases. Here, evidence for the noncoplanarity of the catalytic aspartates is provided by analysis of multiple explicit water molecular dynamics (MD) simulations of plasmepsin II, human beta-secretase, and HIV-protease. In the MD runs of plasmepsin II, the angle between the planes of the two carboxylates of the catalytic dyad is almost always in the range 60 degrees -120 degrees , in agreement with the perpendicular orientation in the re-refined X-ray structure. The noncoplanar arrangement is prevalent also in the beta-secretase simulations, as well as in the runs with the inhibitor-bound proteases. Quantum-mechanics calculations provide further evidence that before catalysis the noncoplanar arrangement is favored energetically in eukaryotic aspartic proteases. Remarkably, the coplanar orientation of the catalytic dyad is observed in MD simulations of HIV-protease at 100 K but not at 300 K, which indicates that the noncoplanar arrangement is favored by conformational entropy. This finding suggests that the coplanar orientation in the crystal structures of apo aspartic proteases is promoted by the very low temperature used for data collection (usually around 100 K).

Structure-based design of carbon nanotubes as HIV-1 protease inhibitors: atomistic and coarse-grained simulations

Nanoparticles such as fullerenes and carbon nanotubes have been extensively studied for biomedical applications. In this paper, we report the design of carbon nanotubes as HIV-1 protease inhibitors. Docking and molecular dynamics calculations are performed using an atomistic model to explore the optimal interaction structure and free energy between the nanotube and HIV-1 protease. A coarse-grained model is then developed based on the atomistic model, allowing us to investigate the dynamic behaviors of the protease in the bound and unbound states. The dynamic process reveals that the carbon nanotube is able to bind to the active site of the protease and prevent the active flaps from opening up, thus blocking the function of the protease. This process is strongly influenced by the size of the nanotube. The binding of carbon nanotubes to an alternative binding site other than the active site is also explored. Therefore, carbon nanotube-based inhibitors have great potential for application as HIV-1 protease inhibitors.

Multiscale modeling of proteins

The activity within a living cell is based on a complex network of interactions among biomolecules, exchanging information and energy through biochemical processes. These events occur on different scales, from the nano- to the macroscale, spanning about 10 orders of magnitude in the space domain and 15 orders of magnitude in the time domain. Consequently, many different modeling techniques, each proper for a particular time or space scale, are commonly used. In addition, a single process often spans more than a single time or space scale. Thus, the necessity arises for combining the modeling techniques in multiscale approaches. In this Account, I first review the different modeling methods for bio-systems, from quantum mechanics to the coarse-grained and continuum-like descriptions, passing through the atomistic force field simulations. Special attention is devoted to their combination in different possible multiscale approaches and to the questions and problems related to their coherent matching in the space and time domains. These aspects are often considered secondary, but in fact, they have primary relevance when the aim is the coherent and complete description of bioprocesses. Subsequently, applications are illustrated by means of two paradigmatic examples: (i) the green fluorescent protein (GFP) family and (ii) the proteins involved in the human immunodeficiency virus (HIV) replication cycle. The GFPs are currently one of the most frequently used markers for monitoring protein trafficking within living cells; nanobiotechnology and cell biology strongly rely on their use in fluorescence microscopy techniques. A detailed knowledge of the actions of the virus-specific enzymes of HIV (specifically HIV protease and integrase) is necessary to study novel therapeutic strategies against this disease. Thus, the insight accumulated over years of intense study is an excellent framework for this Account. The foremost relevance of these two biomolecular systems was recently confirmed by the assignment of two of the Nobel prizes in 2008: in chemistry for the discovery of GFP and in medicine for the discovery of HIV. Accordingly, these proteins were studied with essentially all of the available modeling techniques, making them ideal examples for studying the details of multiscale approaches in protein modeling.

Large conformational changes in proteins: signaling and other functions

Guanine and adenine nucleotide triphosphatases, such as Ras proteins and protein kinases, undergo large conformational changes upon ligand binding in the course of their functions. New computer simulation methods have combined with experimental studies to deepen our understanding of these phenomena. In particular, a 'conformational selection' picture is emerging, where alterations in the relative populations of pre-existing conformations can best describe the conformational switching activity of these important proteins.

Substrate binding mechanism of HIV-1 protease from explicit-solvent atomistic simulations

The binding mechanism of a peptide substrate (Thr-Ile-Met-Met-Gln-Arg, cleavage site p2-NC of the viral polyprotein) to wild-type HIV-1 protease has been investigated by 1.6 micros biased all-atom molecular dynamics simulations in explicit water. The configuration space has been explored biasing seven reaction coordinates by the bias-exchange metadynamics technique. The structure of the Michaelis complex is obtained starting from the substrate outside the enzyme within a backbone rmsd of 0.9 A. The calculated free energy of binding is -6 kcal/mol, and the kinetic constants for association and dissociation are 1.3 x 10(6) M(-1) s(-1) and 57 s(-1), respectively, consistent with experiments. In the main binding pathway, the flaps of the protease do not open sizably. The substrate slides inside the enzyme cavity from the tight lateral channel. This may contrast with the natural polyprotein substrate which is expected to bind by opening the flaps. Thus, mutations might influence differently the binding kinetics of peptidomimetic ligands and of the natural substrate.

Dynamics of "flap" structures in three HIV-1 protease/inhibitor complexes probed by total chemical synthesis and pulse-EPR spectroscopy

The unliganded form of nitroxide spin-labeled HIV-1 protease and three different complexes with inhibitors were studied by pulse-EPR spectroscopy to determine "interflap" distance distributions in solution. In the unliganded enzyme, we observed a rather broad distribution with three maxima corresponding to three flap conformers; the principal form is a "semiopen/semiopen" conformer. In the complexes with inhibitors, the dominant conformer is an asymmetric "closed/semiopen" form. Moreover, the distance distribution profile is significantly varied among the different inhibitors, which mimic different species on the reaction coordinate for enzyme catalyzed proteolysis.

Interpretation of protein/ligand crystal structure using QM/MM calculations: case of HIV-1 protease/metallacarborane complex

Deltahedral metallacarborane compounds have recently been discovered as potent, specific, stable, and nontoxic inhibitors of HIV-1 protease (PR), the major target for AIDS therapy. The 2.15 A-resolution X-ray structure has exhibited a nonsymmetrical binding of the parental compound [Co(3+)-(C2B9H11)2](-) (GB-18) into PR dimer and a symmetrical arrangement in the crystal of two PR dimer complexes into a tetramer. In order to explore structural and energetic details of the inhibitor binding, quantum mechanics coupled with molecular mechanics approach was utilized. Realizing the close positioning of anionic inhibitors in the active site cavity, the possibility of an exchange of structural water molecules Wat50 and Wat128 by Na+ counterions was studied. The energy profiles for the rotation of the GB-18 molecules along their longitudinal axes in complex with PR were calculated. The results show that two Na+ counterions are present in the active site cavity and provide energetically favorable and unfavorable positions for carbon atoms within the carborane cages. Eighty-one rotamer combinations of four molecules of GB-18 bound to PR out of 4 x 10(5) are predicted to be highly populated. These results lay ground for further calculations of interaction energies between GB-18 and amino acids of PR active site and will make it possible to interpret computationally the binding of similar metallacarborane molecules to PR as well as to resistant PR variants. Moreover, this computational tool will allow the design of new, more potent metallacarborane-based HIV-1 protease inhibitors.

Clarifying allosteric control of flap conformations in the 1TW7 crystal structure of HIV-1 protease

The 1TW7 crystal structure of HIV-1 protease shows the flaps placed wider and more open than what is seen in other examples of the semi-open, apo form. It has been proposed that this might be experimental evidence of allosteric control, because crystal packing creates contacts to the "elbow region" of the protease, which may cause deformation of the flaps. Recent dynamics simulations have shown that the conformation seen in 1TW7 relaxes into the typical semi-open conformation in the absence of the crystal contacts, definitively showing that the crystal contacts cause the deformation (Layten et al., J Am Chem Soc 2006;128:13360-13361). However, this does not prove or disprove allosteric modulation at the elbow. In this study, we have conducted additional simulations, supplemented with experimental testing, to further probe the possibility of 1TW7 providing an example of allosteric control of the flap region. We show that the contacts are unstable and do not restrict the conformational sampling of the flaps. The deformation seen in the 1TW7 crystal structure is simply opportunistic crystal packing and not allosteric control.

A poke in the eye: inhibiting HIV-1 protease through its flap-recognition pocket

A novel mechanism of inhibiting HIV-1 protease (HIVp) is presented. Using computational solvent mapping to identify complementary interactions and the Multiple Protein Structure method to incorporate protein flexibility, we generated a receptor-based pharmacophore model of the flexible flap region of the semiopen, apo state of HIVp. Complementary interactions were consistently observed at the base of the flap, only within a cleft with a specific structural role. In the closed, bound state of HIVp, each flap tip docks against the opposite monomer, occupying this cleft. This flap-recognition site is filled by the protein and cannot be identified using traditional approaches based on bound, closed structures. Virtual screening and dynamics simulations show how small molecules can be identified to complement this cleft. Subsequent experimental testing confirms inhibitory activity of this new class of inhibitor. This may be the first new inhibitor class for HIVp since dimerization inhibitors were introduced 17 years ago.