Discovery of multiple hidden allosteric sites by combining Markov state models and experiments

A Genetically Encoded β-Lactamase Reporter for Ultrasensitive (129) Xe NMR in Mammalian Cells

Molecular imaging holds considerable promise for elucidating biological processes in normal physiology as well as disease states, but requires noninvasive methods for identifying analytes at sub-micromolar concentrations. Particularly useful are genetically encoded, single-protein reporters that harness the power of molecular biology to visualize specific molecular processes, but such reporters have been conspicuously lacking for in vivo magnetic resonance imaging (MRI). Herein, we report TEM-1 β-lactamase (bla) as a single-protein reporter for hyperpolarized (HP) (129) Xe NMR, with significant saturation contrast at 0.1 μm. Xenon chemical exchange saturation transfer (CEST) interactions with the primary allosteric site in bla give rise to a unique saturation peak at 255 ppm, well removed (≈60 ppm downfield) from the (129) Xe-H2 O peak. Useful saturation contrast was also observed for bla expressed in bacterial cells and mammalian cells.

Revealing Atomic-Level Mechanisms of Protein Allostery with Molecular Dynamics Simulations

Molecular dynamics (MD) simulations have become a powerful and popular method for the study of protein allostery, the widespread phenomenon in which a stimulus at one site on a protein influences the properties of another site on the protein. By capturing the motions of a protein's constituent atoms, simulations can enable the discovery of allosteric binding sites and the determination of the mechanistic basis for allostery. These results can provide a foundation for applications including rational drug design and protein engineering. Here, we provide an introduction to the investigation of protein allostery using molecular dynamics simulation. We emphasize the importance of designing simulations that include appropriate perturbations to the molecular system, such as the addition or removal of ligands or the application of mechanical force. We also demonstrate how the bidirectional nature of allostery-the fact that the two sites involved influence one another in a symmetrical manner-can facilitate such investigations. Through a series of case studies, we illustrate how these concepts have been used to reveal the structural basis for allostery in several proteins and protein complexes of biological and pharmaceutical interest.

Emerging Computational Methods for the Rational Discovery of Allosteric Drugs

Allosteric drug development holds promise for delivering medicines that are more selective and less toxic than those that target orthosteric sites. To date, the discovery of allosteric binding sites and lead compounds has been mostly serendipitous, achieved through high-throughput screening. Over the past decade, structural data has become more readily available for larger protein systems and more membrane protein classes (e.g., GPCRs and ion channels), which are common allosteric drug targets. In parallel, improved simulation methods now provide better atomistic understanding of the protein dynamics and cooperative motions that are critical to allosteric mechanisms. As a result of these advances, the field of predictive allosteric drug development is now on the cusp of a new era of rational structure-based computational methods. Here, we review algorithms that predict allosteric sites based on sequence data and molecular dynamics simulations, describe tools that assess the druggability of these pockets, and discuss how Markov state models and topology analyses provide insight into the relationship between protein dynamics and allosteric drug binding. In each section, we first provide an overview of the various method classes before describing relevant algorithms and software packages.

CryptoSite: Expanding the Druggable Proteome by Characterization and Prediction of Cryptic Binding Sites

Many proteins have small-molecule binding pockets that are not easily detectable in the ligand-free structures. These cryptic sites require a conformational change to become apparent; a cryptic site can therefore be defined as a site that forms a pocket in a holo structure, but not in the apo structure. Because many proteins appear to lack druggable pockets, understanding and accurately identifying cryptic sites could expand the set of drug targets. Previously, cryptic sites were identified experimentally by fragment-based ligand discovery and computationally by long molecular dynamics simulations and fragment docking. Here, we begin by constructing a set of structurally defined apo-holo pairs with cryptic sites. Next, we comprehensively characterize the cryptic sites in terms of their sequence, structure, and dynamics attributes. We find that cryptic sites tend to be as conserved in evolution as traditional binding pockets but are less hydrophobic and more flexible. Relying on this characterization, we use machine learning to predict cryptic sites with relatively high accuracy (for our benchmark, the true positive and false positive rates are 73% and 29%, respectively). We then predict cryptic sites in the entire structurally characterized human proteome (11,201 structures, covering 23% of all residues in the proteome). CryptoSite increases the size of the potentially "druggable" human proteome from ~40% to ~78% of disease-associated proteins. Finally, to demonstrate the utility of our approach in practice, we experimentally validate a cryptic site in protein tyrosine phosphatase 1B using a covalent ligand and NMR spectroscopy. The CryptoSite Web server is available at http://salilab.org/cryptosite.

Conformational Sampling and Binding Site Assessment of Suppression of Tumorigenicity 2 Ectodomain

Suppression of Tumorigenicity 2 (ST2), a member of the interleukin-1 receptor (IL-1R) family, activates type 2 immune responses to pathogens and tissue damage via binding to IL-33. Dysregulated responses contribute to asthma, graft-versus-host and autoinflammatory diseases and disorders. To study ST2 structure for inhibitor development, we performed the principal component (PC) analysis on the crystal structures of IL1-1R1, IL1-1R2, ST2 and the refined ST2 ectodomain (ST2ECD) models, constructed from previously reported small-angle X-ray scattering data. The analysis facilitates mapping of the ST2ECD conformations to PC subspace for characterizing structural changes. Extensive coverage of ST2ECD conformations was then obtained using the accelerated molecular dynamics simulations started with the IL-33 bound ST2ECD structure as instructed by their projected locations on the PC subspace. Cluster analysis of all conformations further determined representative conformations of ST2ECD ensemble in solution. Alignment of the representative conformations with the ST2/IL-33 structure showed that the D3 domain of ST2ECD (containing D1-D3 domains) in most conformations exhibits no clashes with IL-33 in the crystal structure. Our experimental binding data informed that the D1-D2 domain of ST2ECD contributes predominantly to the interaction between ST2ECD and IL-33 underscoring the importance of the D1-D2 domain in binding. Computational binding site assessment revealed one third of the total detected binding sites in the representative conformations may be suitable for binding to potent small molecules. Locations of these sites include the D1-D2 domain ST2ECD and modulation sites conformed to ST2ECD conformations. Our study provides structural models and analyses of ST2ECD that could be useful for inhibitor discovery.

A chemical chaperone induces inhomogeneous conformational changes in flexible proteins

Organic osmolytes are major cellular compounds that favor protein's compaction and stabilization of the native state. Here, we have examined the chaperone effect of the naturally occurring trimethylamine N-oxide (TMAO) osmolyte on a flexible protein.

Exposing a β-Lactamase "Twist": the Mechanistic Basis for the High Level of Ceftazidime Resistance in the C69F Variant of the Burkholderia pseudomallei PenI β-Lactamase

Around the world, Burkholderia spp. are emerging as pathogens highly resistant to β-lactam antibiotics, especially ceftazidime. Clinical variants of Burkholderia pseudomallei possessing the class A β-lactamase PenI with substitutions at positions C69 and P167 are known to demonstrate ceftazidime resistance. However, the biochemical basis for ceftazidime resistance in class A β-lactamases in B. pseudomallei is largely undefined. Here, we performed site saturation mutagenesis of the C69 position and investigated the kinetic properties of the C69F variant of PenI from B. pseudomallei that results in a high level of ceftazidime resistance (2 to 64 mg/liter) when expressed in Escherichia coli. Surprisingly, quantitative immunoblotting showed that the steady-state protein levels of the C69F variant β-lactamase were ∼4-fold lower than those of wild-type PenI (0.76 fg of protein/cell versus 4.1 fg of protein/cell, respectively). However, growth in the presence of ceftazidime increases the relative amount of the C69F variant to greater than wild-type PenI levels. The C69F variant exhibits a branched kinetic mechanism for ceftazidime hydrolysis, suggesting there are two different conformations of the enzyme. When incubated with an anti-PenI antibody, one conformation of the C69F variant rapidly hydrolyzes ceftazidime and most likely contributes to the higher levels of ceftazidime resistance observed in cell-based assays. Molecular dynamics simulations suggest that the electrostatic characteristics of the oxyanion hole are altered in the C69F variant. When ceftazidime was positioned in the active site, the C69F variant is predicted to form a greater number of hydrogen-bonding interactions than PenI with ceftazidime. In conclusion, we propose "a new twist" for enhanced ceftazidime resistance mediated by the C69F variant of the PenI β-lactamase based on conformational changes in the C69F variant. Our findings explain the biochemical basis of ceftazidime resistance in B. pseudomallei, a pathogen of considerable importance, and suggest that the full repertoire of conformational states of a β-lactamase profoundly affects β-lactam resistance.

FAST Conformational Searches by Balancing Exploration/Exploitation Trade-Offs

Molecular dynamics simulations are a powerful means of understanding conformational changes. However, it is still difficult to simulate biologically relevant time scales without the use of specialized supercomputers. Here, we introduce a goal-oriented sampling method, called fluctuation amplification of specific traits (FAST), for extending the capabilities of commodity hardware. This algorithm rapidly searches conformational space for structures with desired properties by balancing trade-offs between focused searches around promising solutions (exploitation) and trying novel solutions (exploration). FAST was inspired by the hypothesis that many physical properties have an overall gradient in conformational space, akin to the energetic gradients that are known to guide proteins to their folded states. For example, we expect that transitioning from a conformation with a small solvent-accessible surface area to one with a large surface area will require passing through a series of conformations with steadily increasing surface areas. We demonstrate that such gradients are common through retrospective analysis of existing Markov state models (MSMs). Then we design the FAST algorithm to exploit these gradients to find structures with desired properties by (1) recognizing and amplifying structural fluctuations along gradients that optimize a selected physical property whenever possible, (2) overcoming barriers that interrupt these overall gradients, and (3) rerouting to discover alternative paths when faced with insurmountable barriers. To test FAST, we compare its performance to other methods for three common types of problems: (1) identifying unexpected binding pockets, (2) discovering the preferred paths between specific structures, and (3) folding proteins. Our conservative estimate is that FAST outperforms conventional simulations and an adaptive sampling algorithm by at least an order of magnitude. Furthermore, FAST yields both the proper thermodynamics and kinetics, allowing for a direct connection with kinetic experiments that is impossible with many other advanced sampling algorithms because they provide only thermodynamic information. Therefore, we expect FAST to be of great utility for a wide range of applications.

Molecular dynamics simulations: advances and applications

Molecular dynamics simulations have evolved into a mature technique that can be used effectively to understand macromolecular structure-to-function relationships. Present simulation times are close to biologically relevant ones. Information gathered about the dynamic properties of macromolecules is rich enough to shift the usual paradigm of structural bioinformatics from studying single structures to analyze conformational ensembles. Here, we describe the foundations of molecular dynamics and the improvements made in the direction of getting such ensemble. Specific application of the technique to three main issues (allosteric regulation, docking, and structure refinement) is discussed.

PyEMMA 2: A Software Package for Estimation, Validation, and Analysis of Markov Models

Markov (state) models (MSMs) and related models of molecular kinetics have recently received a surge of interest as they can systematically reconcile simulation data from either a few long or many short simulations and allow us to analyze the essential metastable structures, thermodynamics, and kinetics of the molecular system under investigation. However, the estimation, validation, and analysis of such models is far from trivial and involves sophisticated and often numerically sensitive methods. In this work we present the open-source Python package PyEMMA ( http://pyemma.org ) that provides accurate and efficient algorithms for kinetic model construction. PyEMMA can read all common molecular dynamics data formats, helps in the selection of input features, provides easy access to dimension reduction algorithms such as principal component analysis (PCA) and time-lagged independent component analysis (TICA) and clustering algorithms such as k-means, and contains estimators for MSMs, hidden Markov models, and several other models. Systematic model validation and error calculation methods are provided. PyEMMA offers a wealth of analysis functions such that the user can conveniently compute molecular observables of interest. We have derived a systematic and accurate way to coarse-grain MSMs to few states and to illustrate the structures of the metastable states of the system. Plotting functions to produce a manuscript-ready presentation of the results are available. In this work, we demonstrate the features of the software and show new methodological concepts and results produced by PyEMMA.

Small Molecule Targeting of Protein-Protein Interactions through Allosteric Modulation of Dynamics

The protein–protein interaction (PPI) target class is particularly challenging, but offers potential for “first in class” therapies. Most known PPI small molecules are orthosteric inhibitors but many PPI sites may be fundamentally intractable to this approach. One potential alternative is to consider more attractive, remote small molecule pockets; however, on the whole, allostery is poorly understood and difficult to discover and develop. Here we review the literature in order to understand the basis for allostery, especially as it can apply to PPIs. We suggest that the upfront generation of sophisticated and experimentally validated dynamic models of target proteins can aid in target choice and strategy for allosteric intervention to produce the required functional effect.

Accurately modeling nanosecond protein dynamics requires at least microseconds of simulation

Advances in hardware and algorithms have greatly extended the timescales accessible to molecular simulation. This article assesses whether such long timescale simulations improve our ability to calculate order parameters that describe conformational heterogeneity on ps-ns timescales or if force fields are now a limiting factor. Order parameters from experiment are compared with order parameters calculated in three different ways from simulations ranging from 10 ns to 100 μs in length. Importantly, bootstrapping is employed to assess the variability in results for each simulation length. The results of 10-100 ns timescale simulations are highly variable, possibly explaining the variation in levels of agreement between simulation and experiment in published works examining different proteins. Fortunately, microsecond timescale simulations improve both the accuracy and precision of calculated order parameters, at least so long as the full exponential fit or truncated average approximation is used instead of the common long-time limit approximation. The improved precision of these long timescale simulations allows a statistically sound comparison of a number of modern force fields, such as Amber03, Amber99sb-ILDN, and Charmm27. While there is some variation between these force fields, they generally give very similar results for sufficiently long simulations. The fact that so much simulation is required to precisely capture ps-ns timescale processes suggests that extremely extensive simulations are required for slower processes. Advanced sampling techniques could aid greatly, however, such methods will need to maintain accurate kinetics if they are to be of value for calculating dynamical properties like order parameters. © 2015 Wiley Periodicals, Inc.