Resolution of discordant HIV-1 protease resistance rankings using molecular dynamics simulations

Ensembles Are Required to Handle Aleatoric and Parametric Uncertainty in Molecular Dynamics Simulation

Classical molecular dynamics is a computer simulation technique that is in widespread use across many areas of science, from physics and chemistry to materials, biology, and medicine. The method continues to attract criticism due its oft-reported lack of reproducibility which is in part due to a failure to submit it to reliable uncertainty quantification (UQ). Here we show that the uncertainty arises from a combination of (i) the input parameters and (ii) the intrinsic stochasticity of the method controlled by the random seeds. To illustrate the situation, we make a systematic UQ analysis of a widely used molecular dynamics code (NAMD), applied to estimate binding free energy of a ligand-bound to a protein. In particular, we replace the usually fixed input parameters with random variables, systematically distributed about their mean values, and study the resulting distribution of the simulation output. We also perform a sensitivity analysis, which reveals that, out of a total of 175 parameters, just six dominate the variance in the code output. Furthermore, we show that binding energy calculations dampen the input uncertainty, in the sense that the variation around the mean output free energy is less than the variation around the mean of the assumed input distributions, if the output is ensemble-averaged over the random seeds. Without such ensemble averaging, the predicted free energy is five times more uncertain. The distribution of the predicted properties is thus strongly dependent upon the random seed. Owing to this substantial uncertainty, robust statistical measures of uncertainty in molecular dynamics simulation require the use of ensembles in all contexts.

Application of the ESMACS Binding Free Energy Protocol to a Multi-Binding Site Lactate Dehydogenase A Ligand Dataset

Over the past two decades, the use of fragment-based lead generation has become a common, mature approach to identify tractable starting points in chemical space for the drug discovery process. This approach naturally involves the study of the binding properties of highly heterogeneous ligands. Such datasets challenge computational techniques to provide comparable binding free energy estimates from different binding modes. The performance of a range of statistically robust ensemble-based binding free energy calculation protocols, called ESMACS (enhanced sampling of molecular dynamics with approximation of continuum solvent), is evaluated. Ligands designed to target two binding pockets in the lactate dehydogenase, a target protein, which vary in size, charge, and binding mode, are studied. When compared to experimental results, excellent statistical rankings are obtained across this highly diverse set of ligands. In addition, three approaches to account for entropic contributions are investigated: 1) normal mode analysis, 2) weighted solvent accessible surface area (WSAS), and 3) variational entropy. Normal mode analysis and WSAS correlate strongly with each other-although the latter is computationally far cheaper-but do not improve rankings. Variational entropy corrects exaggerated discrimination of ligands bound in different pockets but creates three outliers which reduce the quality of the overall ranking.

Application of ESMACS binding free energy protocols to diverse datasets: Bromodomain-containing protein 4

As the application of computational methods in drug discovery pipelines becomes more widespread it is increasingly important to understand how reproducible their results are and how sensitive they are to choices made in simulation setup and analysis. Here we use ensemble simulation protocols, termed ESMACS (enhanced sampling of molecular dynamics with approximation of continuum solvent), to investigate the sensitivity of the popular molecular mechanics Poisson-Boltzmann surface area (MMPBSA) methodology. Using the bromodomain-containing protein 4 (BRD4) system bound to a diverse set of ligands as our target, we show that robust rankings can be produced only through combining ensemble sampling with multiple trajectories and enhanced solvation via an explicit ligand hydration shell.

Effects of PRE and POST therapy drug-pressure selected mutations on HIV-1 protease conformational sampling

Conformational sampling of pre- and post-therapy subtype B HIV-1 protease sequences derived from a pediatric subject infected via maternal transmission with HIV-1 were characterized by double electron-electron resonance spectroscopy. The conformational ensemble of the PRE construct resembles native-like inhibitor bound states. In contrast, the POST construct, which contains accumulated drug-pressure selected mutations, has a predominantly semi-open conformational ensemble, with increased populations of open-like states. The single point mutant L63P, which is contained in PRE and POST, has decreased dynamics, particularly in the flap region, and also displays a closed-like conformation of inhibitor-bound states. These findings support our hypothesis that secondary mutations accumulate in HIV-1 protease to shift conformational sampling to stabilize open-like conformations, while maintaining the predominant semi-open conformation for activity.

Computing Clinically Relevant Binding Free Energies of HIV-1 Protease Inhibitors

The use of molecular simulation to estimate the strength of macromolecular binding free energies is becoming increasingly widespread, with goals ranging from lead optimization and enrichment in drug discovery to personalizing or stratifying treatment regimes. In order to realize the potential of such approaches to predict new results, not merely to explain previous experimental findings, it is necessary that the methods used are reliable and accurate, and that their limitations are thoroughly understood. However, the computational cost of atomistic simulation techniques such as molecular dynamics (MD) has meant that until recently little work has focused on validating and verifying the available free energy methodologies, with the consequence that many of the results published in the literature are not reproducible. Here, we present a detailed analysis of two of the most popular approximate methods for calculating binding free energies from molecular simulations, molecular mechanics Poisson-Boltzmann surface area (MMPBSA) and molecular mechanics generalized Born surface area (MMGBSA), applied to the nine FDA-approved HIV-1 protease inhibitors. Our results show that the values obtained from replica simulations of the same protease-drug complex, differing only in initially assigned atom velocities, can vary by as much as 10 kcal mol-1, which is greater than the difference between the best and worst binding inhibitors under investigation. Despite this, analysis of ensembles of simulations producing 50 trajectories of 4 ns duration leads to well converged free energy estimates. For seven inhibitors, we find that with correctly converged normal mode estimates of the configurational entropy, we can correctly distinguish inhibitors in agreement with experimental data for both the MMPBSA and MMGBSA methods and thus have the ability to rank the efficacy of binding of this selection of drugs to the protease (no account is made for free energy penalties associated with protein distortion leading to the over estimation of the binding strength of the two largest inhibitors ritonavir and atazanavir). We obtain improved rankings and estimates of the relative binding strengths of the drugs by using a novel combination of MMPBSA/MMGBSA with normal mode entropy estimates and the free energy of association calculated directly from simulation trajectories. Our work provides a thorough assessment of what is required to produce converged and hence reliable free energies for protein-ligand binding.

Patient-specific modelling in drug design, development and selection including its role in clinical decision-making

Genomics has made enormous progress in the twelve years since the publication of the first draft human genome sequence, but it has not yet been translated into the clinic. Despite spiralling development costs, the number of new drug registrations is not increasing. One reason for this lies in the genetic complexity of disease. Most diseases involve dysregulation in pathways that involve many genes, and many (including most cancers) are themselves genetically heterogeneous. Systems biology involves the multi-level simulation of physiology, cell biology and biochemistry using complex computational techniques. We show here using case studies in cancer and HIV how such computational models, and particularly models based on individual patient data, can be used for drug design and development, and in the selection of the appropriate treatment for a given patient in the face of resistance mutations. If these techniques are to be adopted in routine clinical practice, clinicians will need better training in modern approaches to the integrated analysis of large-scale heterogeneous data and multi-scale models, while developers will need to provide much more usable tools. Investment in computational infrastructure is needed so that results can be returned on clinically relevant timescales and data warehouses designed with data protection as well as accessibility in mind.

Elucidating a relationship between conformational sampling and drug resistance in HIV-1 protease

Enzyme targets in rapidly replicating systems, such as retroviruses, commonly respond to drug-selective pressure with mutations arising in the active site pocket that limit inhibitor effectiveness by introducing steric hindrance or by eliminating essential molecular interactions. However, these primary mutations are disposed to compromising pathogenic fitness. Emerging secondary mutations, which are often found outside of the binding cavity, may or can restore fitness while maintaining drug resistance. The accumulated drug pressure selected mutations could have an indirect effect in the development of resistance, such as altering protein flexibility or the dynamics of protein-ligand interactions. Here, we show that accumulation of mutations in a drug-resistant HIV-1 protease (HIV-1 PR) variant, D30N/M36I/A71V, changes the fractional occupancy of the equilibrium conformational sampling ensemble. Correlations are made among populations of the conformational states, namely, closed-like, semiopen, and open-like, with inhibition constants, as well as kinetic parameters. Mutations that stabilize a closed-like conformation correlate with enzymes of lowered activity and with higher affinity for inhibitors, which is corroborated by a further increase in the fractional occupancy of the closed state upon addition of inhibitor or substrate-mimic. Cross-resistance is found to correlate with combinations of mutations that increase the population of the open-like conformations at the expense of the closed-like state while retaining native-like occupancy of the semiopen population. These correlations suggest that at least three states are required in the conformational sampling model to establish the emergence of drug resistance in HIV-1 PR. More importantly, these results shed light on a possible mechanism whereby mutations combine to impart drug resistance while maintaining catalytic activity.

From base pair to bedside: molecular simulation and the translation of genomics to personalized medicine

Despite the promises made that genomic sequencing would transform therapy by introducing a new era of personalized medicine, relatively few tangible breakthroughs have been made. This has led to the recognition that complex interactions at multiple spatial, temporal, and organizational levels may often combine to produce disease. Understanding this complexity requires that existing and future models are used and interpreted within a framework that incorporates knowledge derived from investigations at multiple levels of biological function. It also requires a computational infrastructure capable of dealing with the vast quantities of data generated by genomic approaches. In this review, we discuss the use of molecular modeling to generate quantitative and qualitative insights at the smallest scales of the systems biology hierarchy, how it can play an important role in the development of a systems understanding of disease and in the application of such knowledge to help discover new therapies and target existing ones on a personal level.

Global Conformational Dynamics of HIV-1 Reverse Transcriptase Bound to Non-Nucleoside Inhibitors

HIV-1 Reverse Transcriptase (RT) is a multifunctional enzyme responsible for the transcription of the RNA genome of the HIV virus into DNA suitable for incorporation within the DNA of human host cells. Its crucial role in the viral life cycle has made it one of the major targets for antiretroviral drug therapy. The Non-Nucleoside RT Inhibitor (NNRTI) class of drugs binds allosterically to the enzyme, affecting many aspects of its activity. We use both coarse grained network models and atomistic molecular dynamics to explore the changes in protein dynamics induced by NNRTI binding. We identify changes in the flexibility and conformation of residue Glu396 in the RNaseH primer grip which could provide an explanation for the acceleration in RNaseH cleavage rate observed experimentally in NNRTI bound HIV-1 RT. We further suggest a plausible path for conformational and dynamic changes to be communicated from the vicinity of the NNRTI binding pocket to the RNaseH at the other end of the enzyme.

Molecular dynamics simulation in virus research

Virus replication in the host proceeds by chains of interactions between viral and host proteins. The interactions are deeply influenced by host immune molecules and anti-viral compounds, as well as by mutations in viral proteins. To understand how these interactions proceed mechanically and how they are influenced by mutations, one needs to know the structures and dynamics of the proteins. Molecular dynamics (MD) simulation is a powerful computational method for delineating motions of proteins at an atomic-scale via theoretical and empirical principles in physical chemistry. Recent advances in the hardware and software for biomolecular simulation have rapidly improved the precision and performance of this technique. Consequently, MD simulation is quickly extending the range of applications in biology, helping to reveal unique features of protein structures that would be hard to obtain by experimental methods alone. In this review, we summarize the recent advances in MD simulations in the study of virus–host interactions and evolution, and present future perspectives on this technique.