Conformational free-energy landscapes of a Na+/Ca2+ exchanger explain its alternating-access mechanism and functional specificity.

Secondary-active transporters catalyze the movement of myriad substances across all cellular membranes, typically against opposing concentration gradients, and without consuming any ATP. To do so, these proteins employ an intriguing structural mechanism evolved to be activated only upon recognition or release of the transported species. We examine this self-regulated mechanism using a homolog of the cardiac Na+/Ca2+ exchanger as a model system. Using advanced computer simulations, we map out the complete functional cycle of this transporter, including unknown conformations that we validate against existing experimental data. Calculated free-energy landscapes reveal why this transporter functions as an antiporter rather than a symporter, why it specifically exchanges Na+ and Ca2+, and why the stoichiometry of this exchange is exactly 3:1. We also rationalize why the protein does not exchange H+ for either Ca2+ or Na+, despite being able to bind H+ and its high similarity with H+/Ca2+ exchangers. Interestingly, the nature of this transporter is not explained by its primary structural states, known as inward- and outward-open conformations; instead, the defining factor is the feasibility of conformational intermediates between those states, wherein access pathways leading to the substrate binding sites become simultaneously occluded from both sides of the membrane. This analysis offers a physically coherent, broadly transferable route to understand the emergence of function from structure among secondary-active membrane transporters.

Multilevel superposition for deciphering the conformational variability of protein ensembles.

The dynamics and variability of protein conformations are directly linked to their functions. Many comparative studies of X-ray protein structures have been conducted to elucidate the relevant conformational changes, dynamics and heterogeneity. The rapid increase in the number of experimentally determined structures has made comparison an effective tool for investigating protein structures. For example, it is now possible to compare structural ensembles formed by enzyme species, variants or the type of ligands bound to them. In this study, the author developed a multilevel model for estimating two covariance matrices that represent inter- and intra-ensemble variability in the Cartesian coordinate space. Principal component analysis using the two estimated covariance matrices identified the inter-/intra-enzyme variabilities, which seemed to be important for the enzyme functions, with the illustrative examples of cytochrome P450 family 2 enzymes and class A $\beta$-lactamases. In P450, in which each enzyme has its own active site of a distinct size, an active-site motion shared universally between the enzymes was captured as the first principal mode of the intra-enzyme covariance matrix. In this case, the method was useful for understanding the conformational variability after adjusting for the differences between enzyme sizes. The developed method is advantageous in small ensemble-size problems and hence promising for use in comparative studies on experimentally determined structures where ensemble sizes are smaller than those generated, for example, by molecular dynamics simulations.

Investigating direct current potentials that affect native protein conformation during trapped ion mobility spectrometry-mass spectrometry.

Trapped ion mobility spectrometry-time-of-flight mass spectrometry (TIMS-TOFMS) has emerged as a tool to study protein conformational states. In TIMS, gas-phase ions are guided across the IM stages by applying direct current (DC) potentials (D1-6), which, however, might induce changes in protein structures through collisional activation. To define conditions for native protein analysis, we evaluated the influence of these DC potentials using the metalloenzyme bovine carbonic anhydrase (BCA) as primary test compound. The variation of DC potentials did not change BCA-ion charge and heme content but affected (relative) charge-state intensities and adduct retention. Constructed extracted-ion mobilograms and corresponding collisional cross-section (CCS) profiles gave useful insights in (alterations of) protein conformational state. For BCA, the D3 and D6 potential (which are applied between the deflection transfer and funnel 1 [F1] and the accumulation exit and the start of the ramp, respectively) had most profound effects, showing multimodal CCS distributions at higher potentials indicating gradual unfolding. The other DC potentials only marginally altered the CCS profiles of BCA. To allow for more general conclusions, five additional proteins of diverse molecular weight and conformational stability were analyzed, and for the main protein charge states, CCS profiles were constructed. Principal component analysis (PCA) of the obtained data showed that D1 and D3 exhibit the highest degree of correlation with the ratio of folded and unfolded protein (F/U) as extracted from the mobilograms obtained per set D potential. The correlation of D6 with F/U and protein charge were similar, and D2, D4, and D5 showed an inverse correlation with F/U but were correlated with protein charge. Although DC boundary values for induced conformational changes appeared protein dependent, a set of DC values could be determined, which assured native analysis of most proteins.

Terms and conditions.

Brian Plosky provides some context for a debate over the use of "intrinsically disordered" to describe regions of proteins.

Intrinsic disorder: A term to define the specific physicochemical characteristic of protein conformational heterogeneity.

In his commentary in this issue of Molecular Cell,1 Struhl reasons that the term "intrinsically disordered regions" represents a vague and confusing concept for protein function. However, the term "intrinsically disordered" highlights the important physicochemical characteristic of conformational heterogeneity. Thus, "intrinsically disordered" is the counterpart to the term "folded, " with neither term having specific functional implications.

Intrinsically disordered regions (IDRs): A vague and confusing concept for protein function.

The term "intrinsically disordered region" (IDR) in proteins has been used in numerous publications. However, most proteins contain IDRs, the term refers to very different types of structures and functions, and many IDRs become structured upon interaction with other biomolecules. Thus, IDR is an unnecessary, vague, and ultimately confusing concept.

Investigating the ability of deep learning-based structure prediction to extrapolate and/or enrich the set of antibody CDR canonical forms.

Deep learning models have been shown to accurately predict protein structure from sequence, allowing researchers to explore protein space from the structural viewpoint. In this paper we explore whether "novel" features, such as distinct loop conformations can arise from these predictions despite not being present in the training data. Here we have used ABodyBuilder2, a deep learning antibody structure predictor, to predict the structures of ~1.5M paired antibody sequences. We examined the predicted structures of the canonical CDR loops and found that most of these predictions fall into the already described CDR canonical form structural space. We also found a small number of "new" canonical clusters composed of heterogeneous sequences united by a common sequence motif and loop conformation. Analysis of these novel clusters showed their origins to be either shapes seen in the training data at very low frequency or shapes seen at high frequency but at a shorter sequence length. To evaluate explicitly the ability of ABodyBuilder2 to extrapolate, we retrained several models whilst withholding all antibody structures of a specific CDR loop length or canonical form. These "starved" models showed evidence of generalisation across CDRs of different lengths, but they did not extrapolate to loop conformations which were highly distinct from those present in the training data. However, the models were able to accurately predict a canonical form even if only a very small number of examples of that shape were in the training data. Our results suggest that deep learning protein structure prediction methods are unable to make completely out-of-domain predictions for CDR loops. However, in our analysis we also found that even minimal amounts of data of a structural shape allow the method to recover its original predictive abilities. We have made the ~1.5 M predicted structures used in this study available to download at https://doi.org/10.5281/zenodo.10280181.

Activation and friction in enzymatic loop opening and closing dynamics.

Protein loop dynamics have recently been recognized as central to enzymatic activity, specificity and stability. However, the factors controlling loop opening and closing kinetics have remained elusive. Here, we combine molecular dynamics simulations with string-method determination of complex reaction coordinates to elucidate the molecular mechanism and rate-limiting step for WPD-loop dynamics in the PTP1B enzyme. While protein conformational dynamics is often represented as diffusive motion hindered by solvent viscosity and internal friction, we demonstrate that loop opening and closing is activated. It is governed by torsional rearrangement around a single loop peptide group and by significant friction caused by backbone adjustments, which can dynamically trap the loop. Considering both torsional barrier and time-dependent friction, our calculated rate constants exhibit very good agreement with experimental measurements, reproducing the change in loop opening kinetics between proteins. Furthermore, we demonstrate the applicability of our results to other enzymatic loops, including the M20 DHFR loop, thereby offering prospects for loop engineering potentially leading to enhanced designs.

AlphaFold-assisted structure determination of a bacterial protein of unknown function using X-ray and electron crystallography.

Macromolecular crystallography generally requires the recovery of missing phase information from diffraction data to reconstruct an electron-density map of the crystallized molecule. Most recent structures have been solved using molecular replacement as a phasing method, requiring an a priori structure that is closely related to the target protein to serve as a search model; when no such search model exists, molecular replacement is not possible. New advances in computational machine-learning methods, however, have resulted in major advances in protein structure predictions from sequence information. Methods that generate predicted structural models of sufficient accuracy provide a powerful approach to molecular replacement. Taking advantage of these advances, AlphaFold predictions were applied to enable structure determination of a bacterial protein of unknown function (UniProtKB Q63NT7, NCBI locus BPSS0212) based on diffraction data that had evaded phasing attempts using MIR and anomalous scattering methods. Using both X-ray and micro-electron (microED) diffraction data, it was possible to solve the structure of the main fragment of the protein using a predicted model of that domain as a starting point. The use of predicted structural models importantly expands the promise of electron diffraction, where structure determination relies critically on molecular replacement.

High-throughput prediction of protein conformational distributions with subsampled AlphaFold2.

This paper presents an innovative approach for predicting the relative populations of protein conformations using AlphaFold 2, an AI-powered method that has revolutionized biology by enabling the accurate prediction of protein structures. While AlphaFold 2 has shown exceptional accuracy and speed, it is designed to predict proteins' ground state conformations and is limited in its ability to predict conformational landscapes. Here, we demonstrate how AlphaFold 2 can directly predict the relative populations of different protein conformations by subsampling multiple sequence alignments. We tested our method against nuclear magnetic resonance experiments on two proteins with drastically different amounts of available sequence data, Abl1 kinase and the granulocyte-macrophage colony-stimulating factor, and predicted changes in their relative state populations with more than 80% accuracy. Our subsampling approach worked best when used to qualitatively predict the effects of mutations or evolution on the conformational landscape and well-populated states of proteins. It thus offers a fast and cost-effective way to predict the relative populations of protein conformations at even single-point mutation resolution, making it a useful tool for pharmacology, analysis of experimental results, and predicting evolution.

Protein Ensemble Generation Through Variational Autoencoder Latent Space Sampling.

Mapping the ensemble of protein conformations that contribute to function and can be targeted by small molecule drugs remains an outstanding challenge. Here, we explore the use of variational autoencoders for reducing the challenge of dimensionality in the protein structure ensemble generation problem. We convert high-dimensional protein structural data into a continuous, low-dimensional representation, carry out a search in this space guided by a structure quality metric, and then use RoseTTAFold guided by the sampled structural information to generate 3D structures. We use this approach to generate ensembles for the cancer relevant protein K-Ras, train the VAE on a subset of the available K-Ras crystal structures and MD simulation snapshots, and assess the extent of sampling close to crystal structures withheld from training. We find that our latent space sampling procedure rapidly generates ensembles with high structural quality and is able to sample within 1 Å of held-out crystal structures, with a consistency higher than that of MD simulation or AlphaFold2 prediction. The sampled structures sufficiently recapitulate the cryptic pockets in the held-out K-Ras structures to allow for small molecule docking.

How nuclear receptors transition between active and inactive forms: An energetic perspective.

Nuclear receptors regulate transcriptional programs in response to the binding of natural and synthetic ligands. These ligands modulate the receptor by inducing dynamic changes in the ligand binding domain that shift the C-terminal helix (H12) between active and inactive conformations. Despite decades of study, many questions persist regarding the nature of the inactive state and how ligands shift receptors between different states. Here, we use molecular dynamics (MD) simulations to investigate the timescale and energetic landscape of the conformational transition between inactive and active forms of progesterone receptor (PR) bound to a partial agonist. We observe that the microsecond timescale is insufficient to observe any transitions; only at millisecond timescales achieved via accelerated MD simulations do we find the inactive PR switches to the active state. Energetic analysis reveals that both active and inactive PR states represent energy minima separated by a barrier that can be traversed. In contrast, little or no transition is observed between active and inactive states when an agonist or antagonist is bound, confirming that ligand identity plays a key role in defining the energy landscape of nuclear receptor conformations.

FRETpredict: a Python package for FRET efficiency predictions using rotamer libraries.

Förster resonance energy transfer (FRET) is a widely-used and versatile technique for the structural characterization of biomolecules. Here, we introduce FRETpredict, an easy-to-use Python software to predict FRET efficiencies from ensembles of protein conformations. FRETpredict uses a rotamer library approach to describe the FRET probes covalently bound to the protein. The software efficiently and flexibly operates on large conformational ensembles such as those generated by molecular dynamics simulations to facilitate the validation or refinement of molecular models and the interpretation of experimental data. We provide access to rotamer libraries for many commonly used dyes and linkers and describe a general methodology to generate new rotamer libraries for FRET probes. We demonstrate the performance and accuracy of the software for different types of systems: a rigid peptide (polyproline 11), an intrinsically disordered protein (ACTR), and three folded proteins (HiSiaP, SBD2, and MalE). FRETpredict is open source (GPLv3) and is available at github.com/KULL-Centre/FRETpredict and as a Python PyPI package at pypi.org/project/FRETpredict .

[Modulation of Aggregation and Immunogenicity of a Protein: Based on the Study of Hen Lysozyme].

This study focuses on the modulation of protein aggregation and immunogenicity. As a starting point for investigating long-range interactions within a non-native protein, the effects of perturbing denatured protein states on their aggregation, including the formation of amyloid fibrils, were evaluated. The effects of adducts, sugar modifications, and stabilization on protein aggregation were then examined. We also investigated how protein immunogenicity was affected by enhancing protein conformational stability and other factors.

Mechanistic Insights Revealed by YbtPQ in the Occluded State.

Recently, several ATP-binding cassette (ABC) importers have been found to adopt the typical fold of type IV ABC exporters. Presumably, these importers would function under the transport scheme of "alternating access" like those exporters, cycling through inward-open, occluded, and outward-open conformations. Understanding how the exporter-like importers move substrates in the opposite direction requires structural studies on all the major conformations. To shed light on this, here we report the structure of yersiniabactin importer YbtPQ from uropathogenic Escherichia coli in the occluded conformation trapped by ADP-vanadate (ADP-Vi) at a 3.1 Å resolution determined by cryo-electron microscopy. The structure shows unusual local rearrangements in multiple helices and loops in its transmembrane domains (TMDs). In addition, the dimerization of the nucleotide-binding domains (NBDs) promoted by the vanadate trapping is highlighted by the "screwdriver" action at one of the two hinge points. These structural observations are rare and thus provide valuable information to understand the structural plasticity of the exporter-like ABC importers.

Classifying protein kinase conformations with machine learning.

Protein kinases are key actors of signaling networks and important drug targets. They cycle between active and inactive conformations, distinguished by a few elements within the catalytic domain. One is the activation loop, whose conserved DFG motif can occupy DFG-in, DFG-out, and some rarer conformations. Annotation and classification of the structural kinome are important, as different conformations can be targeted by different inhibitors and activators. Valuable resources exist; however, large-scale applications will benefit from increased automation and interpretability of structural annotation. Interpretable machine learning models are described for this purpose, based on ensembles of decision trees. To train them, a set of catalytic domain sequences and structures was collected, somewhat larger and more diverse than existing resources. The structures were clustered based on the DFG conformation and manually annotated. They were then used as training input. Two main models were constructed, which distinguished active/inactive and in/out/other DFG conformations. They considered initially 1692 structural variables, spanning the whole catalytic domain, then identified ("learned") a small subset that sufficed for accurate classification. The first model correctly labeled all but 3 of 3289 structures as active or inactive, while the second assigned the correct DFG label to all but 17 of 8826 structures. The most potent classifying variables were all related to well-known structural elements in or near the activation loop and their ranking gives insights into the conformational preferences. The models were used to automatically annotate 3850 kinase structures predicted recently with the Alphafold2 tool, showing that Alphafold2 reproduced the active/inactive but not the DFG-in proportions seen in the Protein Data Bank. We expect the models will be useful for understanding and engineering kinases.

Xplor-NIH: Better parameters and protocols for NMR protein structure determination.

The present work describes an update to the protein covalent geometry and atomic radii parameters in the Xplor-NIH biomolecular structure determination package. In combination with an improved treatment of selected non-bonded interactions between atoms three bonds apart, such as those involving methyl hydrogens, and a previously developed term that affects the system's gyration volume, the new parameters are tested using structure calculations on 30 proteins with restraints derived from nuclear magnetic resonance data. Using modern structure validation criteria, including several formally adopted by the Protein Data Bank, and a clear measure of structural accuracy, the results show superior performance relative to previous Xplor-NIH implementations. Additionally, the Xplor-NIH structures compare favorably against originally determined NMR models.

Structure and function of the Arabidopsis ABC transporter ABCB19 in brassinosteroid export.

Brassinosteroids are steroidal phytohormones that regulate plant development and physiology, including adaptation to environmental stresses. Brassinosteroids are synthesized in the cell interior but bind receptors at the cell surface, necessitating a yet to be identified export mechanism. Here, we show that a member of the ATP-binding cassette (ABC) transporter superfamily, ABCB19, functions as a brassinosteroid exporter. We present its structure in both the substrate-unbound and the brassinosteroid-bound states. Bioactive brassinosteroids are potent activators of ABCB19 ATP hydrolysis activity, and transport assays showed that ABCB19 transports brassinosteroids. In Arabidopsis thaliana, ABCB19 and its close homolog, ABCB1, positively regulate brassinosteroid responses. Our results uncover an elusive export mechanism for bioactive brassinosteroids that is tightly coordinated with brassinosteroid signaling.

Role of Protonation States in the Stability of Molecular Dynamics Simulations of High-Resolution Membrane Protein Structures.

Classical molecular dynamics (MD) simulations provide unmatched spatial and time resolution of protein structure and function. However, the accuracy of MD simulations often depends on the quality of force field parameters and the time scale of sampling. Another limitation of conventional MD simulations is that the protonation states of titratable amino acid residues remain fixed during simulations, even though protonation state changes coupled to conformational dynamics are central to protein function. Due to the uncertainty in selecting protonation states, classical MD simulations are sometimes performed with all amino acids modeled in their standard charged states at pH 7. Here, we performed and analyzed classical MD simulations on high-resolution cryo-EM structures of two large membrane proteins that transfer protons by catalyzing protonation/deprotonation reactions. In simulations performed with titratable amino acids modeled in their standard protonation (charged) states, the structure diverges far from its starting conformation. In comparison, MD simulations performed with predetermined protonation states of amino acid residues reproduce the structural conformation, protein hydration, and protein-water and protein-protein interactions of the structure much better. The results support the notion that it is crucial to perform basic protonation state calculations, especially on structures where protonation changes play an important functional role, prior to the launch of any conventional MD simulations. Furthermore, the combined approach of fast protonation state prediction and MD simulations can provide valuable information about the charge states of amino acids in the cryo-EM sample. Even though accurate prediction of protonation states in proteinaceous environments currently remains a challenge, we introduce an approach of combining pKa prediction with cryo-EM density map analysis that helps in improving not only the protonation state predictions but also the atomic modeling of density data.

Hairpin trimer transition state of amyloid fibril.

Protein fibril self-assembly is a universal transition implicated in neurodegenerative diseases. Although fibril structure/growth are well characterized, fibril nucleation is poorly understood. Here, we use a computational-experimental approach to resolve fibril nucleation. We show that monomer hairpin content quantified from molecular dynamics simulations is predictive of experimental fibril formation kinetics across a tau motif mutant library. Hairpin trimers are predicted to be fibril transition states; one hairpin spontaneously converts into the cross-beta conformation, templating subsequent fibril growth. We designed a disulfide-linked dimer mimicking the transition state that catalyzes fibril formation, measured by ThT fluorescence and TEM, of wild-type motif - which does not normally fibrillize. A dimer compatible with extended conformations but not the transition-state fails to nucleate fibril at any concentration. Tau repeat domain simulations show how long-range interactions sequester this motif in a mutation-dependent manner. This work implies that different fibril morphologies could arise from disease-dependent hairpin seeding from different loci.