Conformational Landscape Of Proteins Research Articles

Mapping protein conformational landscapes from crystallographic drug fragment screens.

Proteins are dynamic macromolecules. Knowledge of a protein's thermally accessible conformations is critical to determining important transitions and designing therapeutics. Accessible conformations are highly constrained by a protein's structure such that concerted structural changes due to external perturbations likely track intrinsic conformational transitions. These transitions can be thought of as paths through a conformational landscape. Crystallographic drug fragment screens are high-throughput perturbation experiments, in which thousands of crystals of a drug target are soaked with small-molecule drug precursors (fragments) and examined for fragment binding, mapping potential drug binding sites on the target protein. Here, we describe an open-source Python package, COLAV (COnformational LAndscape Visualization), to infer conformational landscapes from such large-scale crystallographic perturbation studies. We apply COLAV to drug fragment screens of two medically important systems: protein tyrosine phosphatase 1B (PTP-1B), which regulates insulin signaling, and the SARS CoV-2 Main Protease (MPro). With enough fragment-bound structures, we find that such drug screens also enable detailed mapping of proteins' conformational landscapes.

FLEXR GUI: a graphical user interface for multi-conformer modeling of proteins.

Proteins are well known 'shapeshifters' which change conformation to function. In crystallography, multiple conformational states are often present within the crystal and the resulting electron-density map. Yet, explicitly incorporating alternative states into models to disentangle multi-conformer ensembles is challenging. We previously reported the tool FLEXR, which, within a few minutes, automatically separates conformational signal from noise and builds the corresponding, often missing, structural features into a multi-conformer model. To make the method widely accessible for routine multi-conformer building as part of the computational toolkit for macromolecular crystallography, we present a graphical user interface (GUI) for FLEXR, designed as a plugin for Coot 1. The GUI implementation seamlessly connects FLEXR models with the existing suite of validation and modeling tools available in Coot. We envision that FLEXR will aid crystallographers by increasing access to a multi-conformer modeling method that will ultimately lead to a better representation of protein conformational heterogeneity in the Protein Data Bank. In turn, deeper insights into the protein conformational landscape may inform biology or provide new opportunities for ligand design. The code is open source and freely available on GitHub at https://github.com/TheFischerLab/FLEXR-GUI.

Recent advances in protein conformation sampling by combining machine learning with molecular simulation

Abstract The rapid advancement and broad application of machine learning (ML) have driven a groundbreaking revolution in computational biology. One of the most cutting-edge and important applications of ML is its integration with molecular simulations to improving the sampling efficiency of the vast conformational space of large biomolecules. This review focuses on recent studies that utilize ML-based techniques in the exploration of protein conformational landscape. We first highlight the recent development of ML-aided enhanced sampling methods, including heuristic algorithms and neural networks that are designed to refine the selection of reaction coordinates for the construction of bias potential, or facilitate the exploration of the unsampled region of the energy landscape. Further, we review the development of autoencoder based methods which combine molecular simulations and deep learning to expand the search of protein conformations. Lastly, we discuss the cutting-edge methodologies for the one-shot generation of protein conformations with precise Boltzmann weights. Collectively, this review demonstrates the promising potential of machine learning in revolutionizing our insight into the complex conformational ensembles of proteins.

Searching for Structure: Characterizing the Protein Conformational Landscape with Clustering-Based Algorithms.

The identification and characterization of the main conformations from a protein population are a challenging and inherently high-dimensional problem. Here, we evaluate the performance of the Secondary sTructural Ensembles with machine LeArning (StELa) double-clustering method, which clusters protein structures based on the relationship between the φ and ψ dihedral angles in a protein backbone and the secondary structure of the protein, thus focusing on the local properties of protein structures. The classification of states as vectors composed of the clusters' indices arising naturally from the Ramachandran plot is followed by the hierarchical clustering of the vectors to allow for the identification of the main features of the corresponding free energy landscape (FEL). We compare the performance of StELa with the established root-mean-squared-deviation (RMSD)-based clustering algorithm, which focuses on global properties of protein structures and with Combinatorial Averaged Transient Structure (CATS), the combinatorial averaged transient structure clustering method based on distributions of the φ and ψ dihedral angle coordinates. Using ensembles of conformations from molecular dynamics simulations of intrinsically disordered proteins (IDPs) of various lengths (tau protein fragments) or short fragments from a globular protein, we show that StELa is the clustering method that identifies many of the minima and relevant energy states around the minima from the corresponding FELs. In contrast, the RMSD-based algorithm yields a large number of clusters that usually cover most of the FEL, thus being unable to distinguish between states, while CATS does not sample well the FELs for long IDPs and fragments from globular proteins.

Assessments of Variational Autoencoder in Protein Conformation Exploration.

Molecular dynamics (MD) simulations have been extensively used to study protein dynamics and subsequently functions. However, MD simulations are often insufficient to explore adequate conformational space for protein functions within reachable timescales. Accordingly, many enhanced sampling methods, including variational autoencoder (VAE) based methods, have been developed to address this issue. The purpose of this study is to evaluate the feasibility of using VAE to assist in the exploration of protein conformational landscapes. Using three modeling systems, we showed that VAE could capture high-level hidden information which distinguishes protein conformations. These models could also be used to generate new physically plausible protein conformations for direct sampling in favorable conformational spaces. We also found that VAE worked better in interpolation than extrapolation and increasing latent space dimension could lead to a trade-off between performances and complexities.

Temporal and spatial resolution of distal protein motions that activate hydrogen tunneling in soybean lipoxygenase

The enzyme soybean lipoxygenase (SLO) provides a prototype for deep tunneling mechanisms in hydrogen transfer catalysis. This work combines room temperature X-ray studies with extended hydrogen-deuterium exchange experiments to define a catalytically-linked, radiating cone of aliphatic side chains that connects an active site iron center of SLO to the protein-solvent interface. Employing eight variants of SLO that have been appended with a fluorescent probe at the identified surface loop, nanosecond fluorescence Stokes shifts have been measured. We report a remarkable identity of the energies of activation (Ea) for the Stokes shifts decay rates and the millisecond C-H bond cleavage step that is restricted to side chain mutants within an identified thermal network. These findings implicate a direct coupling of distal protein motions surrounding the exposed fluorescent probe to active site motions controlling catalysis. While the role of dynamics in enzyme function has been predominantly attributed to a distributed protein conformational landscape, the presented data implicate a thermally initiated, cooperative protein reorganization that occurs on a timescale faster than nanosecond and represents the enthalpic barrier to the reaction of SLO.

Applications of hydrogen deuterium exchange ensemble reweighting to model native protein ensembles, macromolecular interactions and protein drug interactions.

Hydrogen-deuterium exchange mass spectrometry (HDX-MS) is a powerful approach to probe the structural dynamics of a protein in solution. It simultaneously provides time-resolved information on the structural environment of the amide hydrogen of every residue of a protein except Prolines. As such, HDX-MS has been successfully applied to provide insight into the characterization of native state ensembles of proteins, macromolecular interactions, protein-small molecule interactions or protein folding and unfolding studies. However, it remains largely limited to peptide-level resolution information. On the other hand, molecular dynamics (MD) simulations provide a high-resolution view of a protein's structural dynamics by simulating the motion of a protein's atoms over time starting from a high-resolution structure. However, the time scales accessible to MD simulations are limited and can lead to inadequate exploration of a protein's conformational landscape. To bridge this gap, efforts have aimed at developing quantitative and integrative approaches that leverage HDX-MS data to assess and refine computationally generated structural ensembles. HDX ensemble reweighting (HDXer) is one such approach which uses a post hoc maximum entropy approach to adjusts the weights of individual frames of a structural ensemble to conform to HDX-MS data. We had previously applied HDXer to model the native ensemble of the cytoplasmic heme binding protein, PhuS. Here, we have validated the proposed model ensemble by using structure-based drug design approach to successfully target novel, cryptic drug binding sites that would have otherwise not been targetable using HDX-MS or MD simulations alone. Finally, we demonstrate the broader applicability of such HDX-MS guided integrative approach to model macromolecular interactions, missing electron density and protein-small molecule drug interactions

Optimization of hydrogen-deuterium exchange ensemble reweighting (HDXer) to aid the biophysical characterization of proteins.

Hydrogen-deuterium exchange mass spectrometry (HDX-MS) experiments provide unique insights into native protein structure and dynamics in solution but have limited spatial resolution. Molecular dynamics (MD) simulations, in contrast, allow for the exploration of protein conformational landscapes with an atomistic level of resolution. However, forcefield inaccuracies, crystallographic biases, and timescale limitations may compromise the validity of the final structural ensemble. Hydrogen Deuterium eXchange Ensemble Reweighting (HDXer) is a maximum entropy approach that was recently developed to (1) quantitatively predict hydrogen-deuterium exchange from a modeled conformational ensemble and (2) refine the candidate ensemble to best fit an experimental HDX-MS dataset. Here, we compare how various computational and experimental parameters impact the performance of HDXer, using extracellular regulated kinase 2 (ERK2) and hen egg white lysozyme (HEWL) as model proteins. Conventional molecular dynamics simulations of ERK2 and HEWL were performed in triplicate using the CHARMM36m, AMBER FF19SB, and Drude forcefields. After predicting the HDX rates of each candidate ensemble, HDXer was used to fit the simulated structures to the corresponding HDX-MS datasets. To assess the influence of simulation length, truncated and strided copies of the full-length simulations were generated. Other factors, including peptide charge state selection, ion mobility mass spectrometry, and independent HDX-MS data acquisitions, were also tested. We find that while some variables (e.g., high vs. low peptide charge states) do not have a significant effect, other parameters (e.g., early vs. late timepoint selection) directly correlate with the disagreement between calculated and experimental deuterium uptake. The results from this benchmarking inform the development of a recommended HDXer workflow and best practices for optimal data acquisition.

Dynamic conformational states of apo, ATP and cabozantinib bound TAM kinases to differentiate active-inactive kinetic models

The dynamically active and inactive conformations of kinases play a crucial role in the activation of intracellular downstream signaling pathways. The all-atom molecular dynamics (MD) simulations at microsecond (µs) timescale and longer provide robust insights into the structural details of conformational alterations in kinases that contribute to their cellular metabolic activities and signaling pathways. Tyro3, Axl and Mer (TAM) receptor tyrosine kinases (RTKs) are overexpressed in several types of human cancers. Cabozantinib, a small molecule inhibitor constrains the activity of TAM kinases at nanomolar concentrations. The apo, complexes of ATP (active state) and cabozantinib (active and inactive states) with TAM RTKs were studied by 1 µs MD simulations followed by trajectory analyses. The dynamic mechanistic pathways intrinsic to the kinase activity and protein conformational landscape in the cabozantinib bound TAM kinases are revealed due to the alterations in the P-loop, α-helix and activation loop that result in breaking the regulatory (R) and catalytic (C) spines, while the active states of ATP bound TAM kinases are retained. The co-existence of dynamical states when bound to cabozantinib was observed and the long-lived kinetic transition states of distinct active and inactive structural models were deciphered from MD simulation trajectories that have not been revealed so far. Communicated by Ramaswamy H. Sarma

Large-Scale Ligand Perturbations of the Protein Conformational Landscape Reveal State-Specific Interaction Hotspots.

Protein flexibility is important for ligand binding but often ignored in drug design. Considering proteins as ensembles rather than static snapshots creates opportunities to target dynamic proteins that lack FDA-approved drugs, such as the human chaperone, heat shock protein 90 (Hsp90). Hsp90α accommodates ligands with a dynamic lid domain, yet no comprehensive analysis relating lid conformations to ligand properties is available. To date, ∼300 ligand-bound Hsp90α crystal structures are deposited in the Protein Data Bank, which enables us to consider ligand binding as a perturbation of the protein conformational landscape. By estimating binding site volumes, we classified structures into distinct major and minor lid conformations. Supported by retrospective docking, each conformation creates unique hotspots that bind chemically distinguishable ligands. Clustering revealed insightful exceptions and the impact of crystal packing. Overall, Hsp90α's plasticity provides a cautionary tale of overinterpreting individual crystal structures and motivates an ensemble-based view of drug design.

Temperature-dependent hydrogen deuterium exchange shows impact of analog binding on adenosine deaminase flexibility but not embedded thermal networks

The analysis of hydrogen deuterium exchange by mass spectrometry as a function of temperature and mutation has emerged as a generic and efficient tool for the spatial resolution of protein networks that are proposed to function in the thermal activation of catalysis. In this work, we extend temperature-dependent hydrogen deuterium exchange from apo-enzyme structures to protein–ligand complexes. Using adenosine deaminase as a prototype, we compared the impacts of a substrate analog (1-deaza-adenosine) and a very tight-binding inhibitor/transition state analog (pentostatin) at single and multiple temperatures. At a single temperature, we observed different hydrogen deuterium exchange-mass spectrometry properties for the two ligands, as expected from their 106-fold differences in strength of binding. By contrast, analogous patterns for temperature-dependent hydrogen deuterium exchange mass spectrometry emerge in the presence of both 1-deaza-adenosine and pentostatin, indicating similar impacts of either ligand on the enthalpic barriers for local protein unfolding. We extended temperature-dependent hydrogen deuterium exchange to a function-altering mutant of adenosine deaminase in the presence of pentostatin and revealed a protein thermal network that is highly similar to that previously reported for the apo-enzyme (Gao et al., 2020, JACS 142, 19936-19949). Finally, we discuss the differential impacts of pentostatin binding on overall protein flexibility versus site-specific thermal transfer pathways in the context of models for substrate-induced changes to a distributed protein conformational landscape that act in synergy with embedded protein thermal networks to achieve efficient catalysis.

Characterizing Protein Conformational Spaces using Efficient Data Reduction and Algebraic Topology

Datasets representing the conformational landscapes of protein structures are high-dimensional and hence present computational challenges. Efficient and effective dimensionality reduction of these datasets is therefore paramount to our ability to analyze the conformational landscapes of proteins and extract important information regarding protein folding, conformational changes, and binding. Representing the structures with fewer attributes that capture the most variance in the data makes for a quicker and more precise analysis of these structures. In this study, we make use of dimensionality reduction methods for reducing the number of instances and for feature reduction. The reduced dataset that is obtained is then subjected to topological and quantitative analysis. In this step, we perform hierarchical clustering to obtain different sets of conformation clusters that may correspond to intermediate structures. The structures represented by these conformations are then analyzed by studying their high-dimensional topological properties to identify truly distinct conformations and holes in the conformational space that may represent high energy barriers. Our results show that the clusters closely follow known experimental results about intermediate structures as well as binding and folding events. Doi: 10.28991/HEF-SP2022-01-01 Full Text: PDF

Probabilistic inference of distance distributions in DEER spectroscopy

Double electron-electron (DEER) spectroscopy measures distance distributions between pairs of spin labels site-selectively attached to proteins, thereby revealing structural and energetic information about protein conformational landscapes. We present a Bayesian inference approach for analyzing DEER data, based on a model that assumes a nonparametric distance distribution P with a Tikhonov-like smoothness prior. It uses Markov chain Monte Carlo (MCMC) sampling that combines Gibbs and Hamilton Monte Carlo (HMC) sampling to generate samples of the posterior distribution of P and all other model parameters, given an experimental dataset.

Investigating the protein conformational landscape associated with cognitive decline in aging with mass spectrometry

Cognitive decline in aging is becoming a major issue for families and healthcare providers as the population ages. Protein misfolding and aggregation are frequently implicated as driving many neurodegenerative diseases; however, this view is informed by a relatively small number of proteins, such as Aβ, α-synuclein, and tau. Limited proteolysis mass spectrometry (LiP-MS) is an emerging technique which can provide structural information about protein conformational states proteome-wide. In this approach, pulse proteolysis with a non-specific protease enables accessible regions within proteins to be recorded as cleavage sites, which can be subsequently read out with mass spectrometry.

Protein conformation and biomolecular condensates.

Protein conformation and cell compartmentalization are fundamental concepts and subjects of vast scientific endeavors. In the last two decades, we have witnessed exciting advances that unveiled the conjunction of these concepts. An avalanche of studies highlighted the central role of biomolecular condensates in membraneless subcellular compartmentalization that permits the spatiotemporal organization and regulation of myriads of simultaneous biochemical reactions and macromolecular interactions. These studies have also shown that biomolecular condensation, driven by multivalent intermolecular interactions, is mediated by order-disorder transitions of protein conformation and by protein domain architecture. Conceptually, protein condensation is a distinct level in protein conformational landscape in which collective folding of large collections of molecules takes place. Biomolecular condensates arise by the physical process of phase separation and comprise a variety of bodies ranging from membraneless organelles to liquid condensates to solid-like conglomerates, spanning lengths from mesoscopic clusters (nanometers) to micrometer-sized objects. In this review, we summarize and discuss recent work on the assembly, composition, conformation, material properties, thermodynamics, regulation, and functions of these bodies. We also review the conceptual framework for future studies on the conformational dynamics of condensed proteins in the regulation of cellular processes.

The epitope arrangement on flavivirus particles contributes to Mab C10’s extraordinary neutralization breadth across Zika and dengue viruses

SummaryThe human monoclonal antibody C10 exhibits extraordinary cross-reactivity, potently neutralizing Zika virus (ZIKV) and the four serotypes of dengue virus (DENV1–DENV4). Here we describe a comparative structure-function analysis of C10 bound to the envelope (E) protein dimers of the five viruses it neutralizes. We demonstrate that the C10 Fab has high affinity for ZIKV and DENV1 but not for DENV2, DENV3, and DENV4. We further show that the C10 interaction with the latter viruses requires an E protein conformational landscape that limits binding to only one of the three independent epitopes per virion. This limited affinity is nevertheless counterbalanced by the particle’s icosahedral organization, which allows two different dimers to be reached by both Fab arms of a C10 immunoglobulin. The epitopes’ geometric distribution thus confers C10 its exceptional neutralization breadth. Our results highlight the importance not only of paratope/epitope complementarity but also the topological distribution for epitope-focused vaccine design.

Comment on the Optimal Parameters to Derive Intrinsically Disordered Protein Conformational Ensembles from Small-Angle X-ray Scattering Data Using the Ensemble Optimization Method.

The Ensemble Optimization Method (EOM) is a popular approach to describe small-angle X-ray scattering (SAXS) data from highly disordered proteins. The EOM algorithm selects subensembles of coexisting states from large pools of randomized conformations to fit the SAXS data. Based on the unphysical bimodal radius of gyration (Rg) distribution of conformations resulting from the EOM analysis, a recent article (Fagerberg et al. J. Chem. Theory Comput. 2019, 15 (12), 6968-6983) concluded that this approach inadequately described the SAXS data measured for human Histatin 5 (Hst5), a peptide with antifungal properties. Using extensive experimental and synthetic data, we explored the origin of this observation. We found that the one-bead-per-residue coarse-grained representation with averaged scattering form factors (provided in the EOM as an add-on to represent disordered missing loops or domains) may not be appropriate for EOM analyses of scattering data from short (below 50 residues) proteins/peptides. The method of choice for these proteins is to employ atomistic models (e.g., from molecular dynamics simulations) to sample the protein conformational landscape. As a convenient alternative, we have also improved the coarse-grained approach by introducing amino acid specific form factors in the calculations. We also found that, for small proteins, the search for relatively large subensembles of 20-50 conformers (as implemented in the original EOM version) more adequately describes the conformational space sampled in solution than the procedures optimizing the ensemble size. Our observations have been added as recommendations into the information for EOM users to promote the proper utilization of the program for ensemble-based modeling of SAXS data for all types of disordered systems.

Macromolecular room temperature crystallography.

X-ray crystallography enables detailed structural studies of proteins to understand and modulate their function. Conducting crystallographic experiments at cryogenic temperatures has practical benefits but potentially limits the identification of functionally important alternative protein conformations that can be revealed only at room temperature (RT). This review discusses practical aspects of preparing, acquiring, and analyzing X-ray crystallography data at RT to demystify preconceived impracticalities that freeze progress of routine RT data collection at synchrotron sources. Examples are presented as conceptual and experimental templates to enable the design of RT-inspired studies; they illustrate the diversity and utility of gaining novel insights into protein conformational landscapes. An integrative view of protein conformational dynamics enables opportunities to advance basic and biomedical research.

A Simple Cross-Linking/Mass Spectrometry Workflow for Studying System-wide Protein Interactions.

We present a cross-linking/mass spectrometry workflow for performing proteome-wide cross-linking analyses within 1 week. The workflow is based on the commercially available mass spectrometry-cleavable cross-linker disuccinimidyl dibutyric urea and can be employed by every lab having access to a mass spectrometer with tandem mass spectrometry capabilities. We provide an updated version 2.0 of the freeware software tool MeroX, available at www.StavroX.com , that allows us to conduct fully automated and reliable studies delivering insights into protein-protein interaction networks and protein conformations at the proteome level. We exemplify our optimized workflow for mapping protein-protein interaction networks in Drosophila melanogaster embryos on a system-wide level. From cross-linked Drosophila embryo extracts, we detected 29931 cross-link spectrum matches corresponding to 7436 unique cross-linked residues in biological triplicate experiments at a 1% false discovery rate. Among these, 1611 interprotein cross-linking sites were identified and yielded valuable information about protein-protein interactions. The 5825 remaining intraprotein cross-links yield information about the conformational landscape of proteins in their cellular environment.

Pressure-Temperature Analysis of the Stability of the CTL9 Domain Reveals Hidden Intermediates

The observation of two-state unfolding for many small single-domain proteins by denaturants has led to speculation that protein sequences may have evolved to limit the population of partially folded states that could be detrimental to fitness. How such strong cooperativity arises from a multitude of individual interactions is not well understood. Here, we investigate the stability and folding cooperativity of the C-terminal domain of the ribosomal protein L9 in the pressure-temperature plane using site-specific NMR. In contrast to apparent cooperative unfolding detected with denaturant-induced and thermal-induced unfolding experiments and stopped-flow refolding studies at ambient pressure, NMR-detected pressure unfolding revealed significant deviation from two-state behavior, with a core region that was selectively destabilized by increasing temperature. Comparison of pressure-dependent NMR signals from both the folded and unfolded states revealed the population of at least one invisible excited state at atmospheric pressure. The core destabilizing cavity-creating I98A mutation apparently increased the cooperativity of the loss of folded-state peak intensity while also increasing the population of this invisible excited state present at atmospheric pressure. These observations highlight how local stability is subtly modulated by sequence to tune protein conformational landscapes and illustrate the ability of pressure- and temperature-dependent studies to reveal otherwise hidden states.