Atomistic simulation of protein evolution reveals sequence covariation and time-dependent fluctuations of site-specific substitution rates.

Thermodynamic stability is a crucial fitness constraint in protein evolution and is a central factor in shaping the sequence landscapes of proteins. The correlation between stability and molecular fitness depends on the mechanism that relates the biophysical property with biological function. In the simplest case, stability and fitness are related by the amount of folded protein. However, when proteins are toxic in the unfolded state, the fitness function shifts, resulting in higher stability under mutation-selection balance. Likewise, a higher population size results in a similar change in protein stability, as it magnifies the effect of the selection pressure in evolutionary dynamics. This study investigates how such factors affect the evolution of protein stability, site-specific mutation rates, and residue-residue covariation. To simulate evolutionary trajectories with realistic modeling of protein energetics, we develop an all-atom simulator of protein evolution, RosettaEvolve. By evolving proteins under different fitness functions, we can study how the fitness function affects the distribution of proposed and accepted mutations, site-specific rates, and the prevalence of correlated amino acid substitutions. We demonstrate that fitness pressure affects the proposal distribution of mutational effects, that changes in stability can largely explain variations in site-specific substitution rates in evolutionary trajectories, and that increased fitness pressure results in a stronger covariation signal. Our results give mechanistic insight into the evolutionary consequences of variation in protein stability and provide a basis to rationalize the strong covariation signal observed in natural sequence alignments.

Amyloid ß 42 fibril structure based on small-angle scattering.

Amyloid fibrils are associated with a number of neurodegenerative diseases, including fibrils of amyloid ß42 peptide (Aß42) in Alzheimer's disease. These fibrils are a source of toxicity to neuronal cells through surface-catalyzed generation of toxic oligomers. Detailed knowledge of the fibril structure may thus facilitate therapeutic development. We use small-angle scattering to provide information on the fibril cross-section dimension and shape for Aß42 fibrils prepared in aqueous phosphate buffer at pH = 7.4 and pH 8.0 under quiescent conditions at 37 °C from pure recombinant Aß42 peptide. Fitting the data using a continuum model reveals an elliptical cross-section and a peptide mass-per-unit length compatible with two filaments of two monomers, four monomers per plane. To provide a more detailed atomistic model, the data were fitted using as a starting state a high-resolution structure of the two-monomer arrangement in filaments from solid-state NMR (Protein Data Bank ID 5kk3). First, a twofold symmetric model including residues 11 to 42 of two monomers in the filament was optimized in terms of twist angle and local packing using Rosetta. A two-filament model was then built and optimized through fitting to the scattering data allowing the two N-termini in each filament to take different conformations, with the same conformation in each of the two filaments. This provides an atomistic model of the fibril with twofold rotation symmetry around the fibril axis. Intriguingly, no polydispersity as regards the number of filaments was observed in our system over separate samples, suggesting that the two-filament arrangement represents a free energy minimum for the Aß42 fibril.

A double S shape provides the structural basis for the extraordinary binding specificity of Dscam isoforms.

Drosophila Dscam encodes a vast family of immunoglobulin (Ig)-containing proteins that exhibit isoform-specific homophilic binding. This diversity is essential for cell recognition events required for wiring the brain. Each isoform binds to itself but rarely to other isoforms. Specificity is determined by "matching" of three variable Ig domains within an approximately 220 kD ectodomain. Here, we present the structure of the homophilic binding region of Dscam, comprising the eight N-terminal Ig domains (Dscam(1-8)). Dscam(1-8) forms a symmetric homodimer of S-shaped molecules. This conformation, comprising two reverse turns, allows each pair of the three variable domains to "match" in an antiparallel fashion. Structural, genetic, and biochemical studies demonstrate that, in addition to variable domain "matching," intramolecular interactions between constant domains promote homophilic binding. These studies provide insight into how "matching" at all three pairs of variable domains in Dscam mediates isoform-specific recognition.

ZEAL: protein structure alignment based on shape similarity.

MOTIVATION: Most protein-structure superimposition tools consider only Cartesian coordinates. Yet, much of biology happens on the surface of proteins, which is why proteins with shared ancestry and similar function often have comparable surface shapes. Superposition of proteins based on surface shape can enable comparison of highly divergent proteins, identify convergent evolution and enable detailed comparison of surface features and binding sites. RESULTS: We present ZEAL, an interactive tool to superpose global and local protein structures based on their shape resemblance using 3D (Zernike-Canterakis) functions to represent the molecular surface. In a benchmark study of structures with the same fold, we show that ZEAL outperforms two other methods for shape-based superposition. In addition, alignments from ZEAL were of comparable quality to the coordinate-based superpositions provided by TM-align. For comparisons of proteins with limited sequence and backbone-fold similarity, where coordinate-based methods typically fail, ZEAL can often find alignments with substantial surface-shape correspondence. In combination with shape-based matching, ZEAL can be used as a general tool to study relationships between shape and protein function. We identify several categories of protein functions where global shape similarity is significantly more likely than expected by random chance, when comparing proteins with little similarity on the fold level. In particular, we find that global surface shape similarity is particular common among DNA binding proteins. AVAILABILITY AND IMPLEMENTATION: ZEAL can be used online at https://andrelab.org/zeal or as a standalone program with command line or graphical user interface. Source files and installers are available at https://github.com/Andre-lab/ZEAL. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Crystal Structure of Non-Structural Protein 10 from Severe Acute Respiratory Syndrome Coronavirus-2.

Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2), causing Coronavirus Disease 19 (COVID-19), emerged at the end of 2019 and quickly spread to cause a global pandemic with severe socio-economic consequences. The early sequencing of its RNA genome revealed its high similarity to SARS, likely to have originated from bats. The SARS-CoV-2 non-structural protein 10 (nsp10) displays high sequence similarity with its SARS homologue, which binds to and stimulates the 3'-to-5' exoribonuclease and the 2'-O-methlytransferase activities of nsps 14 and 16, respectively. Here, we report the biophysical characterization and 1.6 Å resolution structure of the unbound form of nsp10 from SARS-CoV-2 and compare it to the structures of its SARS homologue and the complex-bound form with nsp16 from SARS-CoV-2. The crystal structure and solution behaviour of nsp10 will not only form the basis for understanding the role of SARS-CoV-2 nsp10 as a central player of the viral RNA capping apparatus, but will also serve as a basis for the development of inhibitors of nsp10, interfering with crucial functions of the replication-transcription complex and virus replication.

Bayesian inference of protein conformational ensembles from limited structural data.

Many proteins consist of folded domains connected by regions with higher flexibility. The details of the resulting conformational ensemble play a central role in controlling interactions between domains and with binding partners. Small-Angle Scattering (SAS) is well-suited to study the conformational states adopted by proteins in solution. However, analysis is complicated by the limited information content in SAS data and care must be taken to avoid constructing overly complex ensemble models and fitting to noise in the experimental data. To address these challenges, we developed a method based on Bayesian statistics that infers conformational ensembles from a structural library generated by all-atom Monte Carlo simulations. The first stage of the method involves a fast model selection based on variational Bayesian inference that maximizes the model evidence of the selected ensemble. This is followed by a complete Bayesian inference of population weights in the selected ensemble. Experiments with simulated ensembles demonstrate that model evidence is capable of identifying the correct ensemble and that correct number of ensemble members can be recovered up to high level of noise. Using experimental data, we demonstrate how the method can be extended to include data from Nuclear Magnetic Resonance (NMR) and structural energies of conformers extracted from the all-atom energy functions. We show that the data from SAXS, NMR chemical shifts and energies calculated from conformers can work synergistically to improve the definition of the conformational ensemble.

Computational assessment of folding energy landscapes in heterodimeric coiled coils.

The coiled coil structural motif consists of alpha helices supercoiling around each other to form staggered knobs-into-holes packing. Such structures are deceptively simple, especially as they often can be described with parametric equations, but are known to exist in various conformations. Even the simplest systems, consisting of 2 monomers, can assemble into a wide range of states. They can form canonical as well as noncanonical coiled coils, be parallel or antiparallel, where helices associate with different degrees of shift, tilt, and rotation. Here, we investigate the energy landscape of heterodimeric coiled coils by carrying out de novo folding simulations starting from amino acid sequence. We folded a diverse set of 22 heterodimers and demonstrate that the approach is capable of identifying the atomic details in the experimental structure in the majority of cases. Our methodology also enables exploration of alternative states that can be accessible in solution beyond the experimentally determined structure. For many systems, we observe folding energy landscapes with multiple energy minima and several isoenergetic states. By comparing coiled coils from single domains and those extracted from larger proteins, we find that standalone coiled coils have deeper energy wells at the experimentally determined conformation. By folding the competing homodimeric states in addition to the heterodimers, we observe that the structural specificity towards the heteromeric state is often small. Taken together, our results demonstrate that de novo folding simulations can be a powerful tool to characterize structural specificity of coiled coils when coupled to assessment of energy landscapes.

Automated determination of fibrillar structures by simultaneous model building and fiber diffraction refinement.

For highly oriented fibrillar molecules, three-dimensional structures can often be determined from X-ray fiber diffraction data. However, because of limited information content, structure determination and validation can be challenging. We demonstrate that automated structure determination of protein fibers can be achieved by guiding the building of macromolecular models with fiber diffraction data. We illustrate the power of our approach by determining the structures of six bacteriophage viruses de novo using fiber diffraction data alone and together with solid-state NMR data. Furthermore, we demonstrate the feasibility of molecular replacement from monomeric and fibrillar templates by solving the structure of a plant virus using homology modeling and protein-protein docking. The generated models explain the experimental data to the same degree as deposited reference structures but with improved structural quality. We also developed a cross-validation method for model selection. The results highlight the power of fiber diffraction data as structural constraints.

A Protein-Based Encapsulation System with Calcium-Controlled Cargo Loading and Detachment.

Protein-based encapsulation systems have a wide spectrum of applications in targeted delivery of cargo molecules and for chemical transformations in confined spaces. By engineering affinity between cargo and container proteins it has been possible to enable the efficient and specific encapsulation of target molecules. Missing in current approaches is the ability to turn off the interaction after encapsulation to enable the cargo to freely diffuse in the lumen of the container. Separation between cargo and container is desirable in drug delivery applications and in the use of capsids as catalytic nanoparticles. We describe an encapsulation system based on the hepatitis B virus capsid in which an engineered high-affinity interaction between cargo and capsid proteins can be modulated by Ca2+ . Cargo proteins are loaded into capsids in the presence of Ca2+ , while ligand removal triggers unbinding inside the container. We observe that confinement leads to hindered rotation of cargo inside the capsid. Application of the designed container for catalysis was also demonstrated by encapsulation of an enzyme with ß-glucosidase activity.

Computational design of a leucine-rich repeat protein with a predefined geometry.

Structure-based protein design offers a possibility of optimizing the overall shape of engineered binding scaffolds to match their targets better. We developed a computational approach for the structure-based design of repeat proteins that allows for adjustment of geometrical features like length, curvature, and helical twist. By combining sequence optimization of existing repeats and de novo design of capping structures, we designed leucine-rich repeats (LRRs) from the ribonuclease inhibitor (RI) family that assemble into structures with a predefined geometry. The repeat proteins were built from self-compatible LRRs that are designed to interact to form highly curved and planar assemblies. We validated the geometrical design approach by engineering a ring structure constructed from 10 self-compatible repeats. Protein design can also be used to increase our structural understanding of repeat proteins. We use our design constructs to demonstrate that buried Cys play a central role for stability and folding cooperativity in RI-type LRR proteins. The computational procedure presented here may be used to develop repeat proteins with various geometrical shapes for applications where greater control of the interface geometry is desired.

Exploring alternate states and oligomerization preferences of coiled-coils by de novo structure modeling.

Homomeric coiled-coils can self-assemble into a wide range of structural states with different helix topologies and oligomeric states. In this study, we have combined de novo structure modeling with stability calculations to simultaneously predict structure and oligomeric states of homomeric coiled-coils. For dimers an asymmetric modeling protocol was developed. Modeling without symmetry constraints showed that backbone asymmetry is important for the formation of parallel dimeric coiled-coils. Collectively, our results demonstrate that high-resolution structure of coiled-coils, as well as parallel and antiparallel orientations of dimers and tetramers, can be accurately predicted from sequence. De novo modeling was also used to generate models of competing oligomeric states, which were used to compare stabilities and thus predict the native stoichiometry from sequence. In a benchmark set of 33 coiled-coil sequences, forming dimers to pentamers, up to 70% of the oligomeric states could be correctly predicted. The calculations demonstrated that the free energy of helix folding could be an important factor for determining stability and oligomeric state of homomeric coiled-coils. The computational methods developed here should be broadly applicable to studies of sequence-structure relationships in coiled-coils and the design of higher order assemblies with improved oligomerization specificity.

Automated de novo phasing and model building of coiled-coil proteins.

Models generated by de novo structure prediction can be very useful starting points for molecular replacement for systems where suitable structural homologues cannot be readily identified. Protein-protein complexes and de novo-designed proteins are examples of systems that can be challenging to phase. In this study, the potential of de novo models of protein complexes for use as starting points for molecular replacement is investigated. The approach is demonstrated using homomeric coiled-coil proteins, which are excellent model systems for oligomeric systems. Despite the stereotypical fold of coiled coils, initial phase estimation can be difficult and many structures have to be solved with experimental phasing. A method was developed for automatic structure determination of homomeric coiled coils from X-ray diffraction data. In a benchmark set of 24 coiled coils, ranging from dimers to pentamers with resolutions down to 2.5 Å, 22 systems were automatically solved, 11 of which had previously been solved by experimental phasing. The generated models contained 71-103% of the residues present in the deposited structures, had the correct sequence and had free R values that deviated on average by 0.01 from those of the respective reference structures. The electron-density maps were of sufficient quality that only minor manual editing was necessary to produce final structures. The method, named CCsolve, combines methods for de novo structure prediction, initial phase estimation and automated model building into one pipeline. CCsolve is robust against errors in the initial models and can readily be modified to make use of alternative crystallographic software. The results demonstrate the feasibility of de novo phasing of protein-protein complexes, an approach that could also be employed for other small systems beyond coiled coils.

Structure of the basal components of a bacterial transporter.

Proteins SpoIIQ and SpoIIIAH interact through two membranes to connect the forespore and the mother cell during endospore development in the bacterium Bacillus subtilis. SpoIIIAH consists of a transmembrane segment and an extracellular domain with similarity to YscJ proteins. YscJ proteins form large multimeric rings that are the structural scaffolds for the assembly of type III secretion systems in gram-negative bacteria. The predicted ring-forming motif of SpoIIIAH and other evidence led to the model that SpoIIQ and SpoIIIAH form the core components of a channel or transporter through which the mother cell nurtures forespore development. Therefore, to understand the roles of SpoIIIAH and SpoIIQ in channel formation, it is critical to determine whether SpoIIIAH adopts a ring-forming structural motif, and whether interaction of SpoIIIAH with SpoIIQ would preclude ring formation. We report a 2.8-Å resolution structure of a complex of SpoIIQ and SpoIIIAH. SpoIIIAH folds into the ring-building structural motif, and modeling shows that the structure of the SpoIIQ-SpoIIIAH complex is compatible with forming a symmetrical oligomer that is similar to those in type III systems. The inner diameters of the two most likely ring models are large enough to accommodate several copies of other integral membrane proteins. SpoIIQ contains a LytM domain, which is found in metalloendopeptidases, but lacks residues important for metalloprotease activity. Other LytM domains appear to be involved in protein-protein interactions. We found that the LytM domain of SpoIIQ contains an accessory region that interacts with SpoIIIAH.

Improved prediction of site-rates from structure with averaging across homologs.

Variation in mutation rates at sites in proteins can largely be understood by the constraint that proteins must fold into stable structures. Models that calculate site-specific rates based on protein structure and a thermodynamic stability model have shown a significant but modest ability to predict empirical site-specific rates calculated from sequence. Models that use detailed atomistic models of protein energetics do not outperform simpler approaches using packing density. We demonstrate that a fundamental reason for this is that empirical site-specific rates are the result of the average effect of many different microenvironments in a phylogeny. By analyzing the results of evolutionary dynamics simulations, we show how averaging site-specific rates across many extant protein structures can lead to correct recovery of site-rate prediction. This result is also demonstrated in natural protein sequences and experimental structures. Using predicted structures, we demonstrate that atomistic models can improve upon contact density metrics in predicting site-specific rates from a structure. The results give fundamental insights into the factors governing the distribution of site-specific rates in protein families.

Accurate prediction of protein assembly structure by combining AlphaFold and symmetrical docking.

AlphaFold can predict the structures of monomeric and multimeric proteins with high accuracy but has a limit on the number of chains and residues it can fold. Here we show that a combination of AlphaFold and all-atom symmetric docking simulations enables highly accurate prediction of the structure of complex symmetrical assemblies. We present a method to predict the structure of complexes with cubic - tetrahedral, octahedral and icosahedral - symmetry from sequence. Focusing on proteins where AlphaFold can make confident predictions on the subunit structure, 27 cubic systems were assembled with a median TM-score of 0.99 and a DockQ score of 0.72. 21 had TM-scores of above 0.9 and were categorized as acceptable- to high-quality according to DockQ. The resulting models are energetically optimized and can be used for detailed studies of intermolecular interactions in higher-order symmetrical assemblies. The results demonstrate how explicit treatment of structural symmetry can significantly expand the size and complexity of AlphaFold predictions.

Facile Method for High-throughput Identification of Stabilizing Mutations.

Characterizing the effects of mutations on stability is critical for understanding the function and evolution of proteins and improving their biophysical properties. High throughput folding and abundance assays have been successfully used to characterize missense mutations associated with reduced stability. However, screening for increased thermodynamic stability is more challenging since such mutations are rarer and their impact on assay readout is more subtle. Here, a multiplex assay for high throughput screening of protein folding was developed by combining deep mutational scanning, fluorescence-activated cell sorting, and deep sequencing. By analyzing a library of 2000 variants of Adenylate kinase we demonstrate that the readout of the method correlates with stability and that mutants with up to 13 °C increase in thermal melting temperature could be identified with low false positive rate. The discovery of many stabilizing mutations also enabled the analysis of general substitution patterns associated with increased stability in Adenylate kinase. This high throughput method to identify stabilizing mutations can be combined with functional screens to identify mutations that improve both stability and activity.

Simultaneous prediction of protein folding and docking at high resolution.

Interleaved dimers and higher order symmetric oligomers are ubiquitous in biology but present a challenge to de novo structure prediction methodology: The structure adopted by a monomer can be stabilized largely by interactions with other monomers and hence not the lowest energy state of a single chain. Building on the Rosetta framework, we present a general method to simultaneously model the folding and docking of multiple-chain interleaved homo-oligomers. For more than a third of the cases in a benchmark set of interleaved homo-oligomers, the method generates near-native models of large alpha-helical bundles, interlocking beta sandwiches, and interleaved alpha/beta motifs with an accuracy high enough for molecular replacement based phasing. With the incorporation of NMR chemical shift information, accurate models can be obtained consistently for symmetric complexes with as many as 192 total amino acids; a blind prediction was within 1 A rmsd of the traditionally determined NMR structure, and fit independently collected RDC data equally well. Together, these results show that the Rosetta "fold-and-dock" protocol can produce models of homo-oligomeric complexes with near-atomic-level accuracy and should be useful for crystallographic phasing and the rapid determination of the structures of multimers with limited NMR information.

DnaK response to expression of protein mutants is dependent on translation rate and stability.

Chaperones play a central part in the quality control system in cells by clearing misfolded and aggregated proteins. The chaperone DnaK acts as a sensor for molecular stress by recognising short hydrophobic stretches of misfolded proteins. As the level of unfolded protein is a function of protein stability, we hypothesised that the level of DnaK response upon overexpression of recombinant proteins would be correlated to stability. Using a set of mutants of the λ-repressor with varying thermal stabilities and a fluorescent reporter system, the effect of stability on DnaK response and protein abundance was investigated. Our results demonstrate that the initial DnaK response is largely dependent on protein synthesis rate but as the recombinantly expressed protein accumulates and homeostasis is approached the response correlates strongly with stability. Furthermore, we observe a large degree of cell-cell variation in protein abundance and DnaK response in more stable proteins.

A memetic algorithm enables efficient local and global all-atom protein-protein docking with backbone and side-chain flexibility.

Protein complex formation is encoded by specific interactions at the atomic scale, but the computational cost of modeling proteins at this level often requires use of simplified energy models and limited conformational flexibility. In particular, use of all-atom energy functions and backbone and side-chain flexibility results in rugged energy landscapes that are difficult to explore. In this study, we develop a protein-protein docking algorithm, EvoDOCK, that combines the strength of a differential evolution algorithm for efficient exploration of the global search space with the benefits of a local optimization method to refine detailed atomic interactions. EvoDOCK enabled accurate and fast local and global protein-protein docking using an all-atom energy function with side-chain flexibility. Comparison with a standard method built on Monte Carlo optimization demonstrated improved accuracy and increases in computational speed of up to 35 times. The evolutionary algorithm also enabled efficient atomistic docking with backbone flexibility.

Determination of the structures of symmetric protein oligomers from NMR chemical shifts and residual dipolar couplings.

Symmetric protein dimers, trimers, and higher-order cyclic oligomers play key roles in many biological processes. However, structural studies of oligomeric systems by solution NMR can be difficult due to slow tumbling of the system and the difficulty in identifying NOE interactions across protein interfaces. Here, we present an automated method (RosettaOligomers) for determining the solution structures of oligomeric systems using only chemical shifts, sparse NOEs, and domain orientation restraints from residual dipolar couplings (RDCs) without a need for a previously determined structure of the monomeric subunit. The method integrates previously developed Rosetta protocols for solving the structures of monomeric proteins using sparse NMR data and for predicting the structures of both nonintertwined and intertwined symmetric oligomers. We illustrated the performance of the method using a benchmark set of nine protein dimers, one trimer, and one tetramer with available experimental data and various interface topologies. The final converged structures are found to be in good agreement with both experimental data and previously published high-resolution structures. The new approach is more readily applicable to large oligomeric systems than conventional structure-determination protocols, which often require a large number of NOEs, and will likely become increasingly relevant as more high-molecular weight systems are studied by NMR.