Uncovering Protein Ensembles: Automated Multiconformer Model Building for X-ray Crystallography and Cryo-EM.

In their folded state, biomolecules exchange between multiple conformational states that are crucial for their function. Traditional structural biology methods, such as X-ray crystallography and cryogenic electron microscopy (cryo-EM), produce density maps that are ensemble averages, reflecting molecules in various conformations. Yet, most models derived from these maps explicitly represent only a single conformation, overlooking the complexity of biomolecular structures. To accurately reflect the diversity of biomolecular forms, there is a pressing need to shift towards modeling structural ensembles that mirror the experimental data. However, the challenge of distinguishing signal from noise complicates manual efforts to create these models. In response, we introduce the latest enhancements to qFit, an automated computational strategy designed to incorporate protein conformational heterogeneity into models built into density maps. These algorithmic improvements in qFit are substantiated by superior Rfree and geometry metrics across a wide range of proteins. Importantly, unlike more complex multicopy ensemble models, the multiconformer models produced by qFit can be manually modified in most major model building software (e.g. Coot) and fit can be further improved by refinement using standard pipelines (e.g. Phenix, Refmac, Buster). By reducing the barrier of creating multiconformer models, qFit can foster the development of new hypotheses about the relationship between macromolecular conformational dynamics and function.

Automated multiconformer model building for X-ray crystallography and cryo-EM.

In their folded state, biomolecules exchange between multiple conformational states that are crucial for their function. Traditional structural biology methods, such as X-ray crystallography and cryogenic electron microscopy (cryo-EM), produce density maps that are ensemble averages, reflecting molecules in various conformations. Yet, most models derived from these maps explicitly represent only a single conformation, overlooking the complexity of biomolecular structures. To accurately reflect the diversity of biomolecular forms, there is a pressing need to shift toward modeling structural ensembles that mirror the experimental data. However, the challenge of distinguishing signal from noise complicates manual efforts to create these models. In response, we introduce the latest enhancements to qFit, an automated computational strategy designed to incorporate protein conformational heterogeneity into models built into density maps. These algorithmic improvements in qFit are substantiated by superior Rfree and geometry metrics across a wide range of proteins. Importantly, unlike more complex multicopy ensemble models, the multiconformer models produced by qFit can be manually modified in most major model building software (e.g., Coot) and fit can be further improved by refinement using standard pipelines (e.g., Phenix, Refmac, Buster). By reducing the barrier of creating multiconformer models, qFit can foster the development of new hypotheses about the relationship between macromolecular conformational dynamics and function.

Mapping kinase domain resistance mechanisms for the MET receptor tyrosine kinase via deep mutational scanning.

Mutations in the kinase and juxtamembrane domains of the MET Receptor Tyrosine Kinase are responsible for oncogenesis in various cancers and can drive resistance to MET-directed treatments. Determining the most effective inhibitor for each mutational profile is a major challenge for MET-driven cancer treatment in precision medicine. Here, we used a deep mutational scan (DMS) of ~5,764 MET kinase domain variants to profile the growth of each mutation against a panel of 11 inhibitors that are reported to target the MET kinase domain. We identified common resistance sites across type I, type II, and type I ½ inhibitors, unveiled unique resistance and sensitizing mutations for each inhibitor, and validated non-cross-resistant sensitivities for type I and type II inhibitor pairs. We augment a protein language model with biophysical and chemical features to improve the predictive performance for inhibitor-treated datasets. Together, our study demonstrates a pooled experimental pipeline for identifying resistance mutations, provides a reference dictionary for mutations that are sensitized to specific therapies, and offers insights for future drug development.

Comparison of side-chain dispersion in protein structures determined by cryo-EM and X-ray crystallography.

An evaluation of systematic differences in local structure and conformation in the interior of protein tertiary structures determined by crystallography and by cryo-electron microscopy (cryo-EM) is reported. The expectation is that any consistent differences between the derived atomic models could provide insights into variations in side-chain packing that result from differences in specimens prepared for analysis between these two methods. By computing an atomic packing score, which provides a quantitative measure of clustering of side-chain atoms in the core of the tertiary structures, it is found that, in general, for structures determined by cryo-EM, side chains are more dispersed than in structures determined by X-ray crystallography over a similar resolution range. This trend is also observed in the packing comparison at subunit interfaces. Similar trends were observed in the packing comparison at the core of tertiary structures of the same proteins determined by both X-ray and cryo-EM methods. It is proposed here that the reduced dispersion of side chains in protein crystals could be due to some level of dehydration in 3D crystals prepared for X-ray crystallography and also because the higher rate of freezing of protein samples for cryo-EM may enable preservation of a more native conformation.

Hypervariability of accessible and inaccessible conformational space of proteins.

Proteins undergo motions in a range of amplitudes, from domain motions to backbone rotations, leading to changes in (φ,ψ) torsion angles and small-scale bond vibrations and angle bending. Here, we study the extent of variations in (φ,ψ) values in proteins and the effects of bond geometry variations due to vibrational motions in a protein on the accessible, (steric clash-free) (φ,ψ) space. We perform 1-fs timestep unconstrained molecular dynamics simulations on super-high-resolution protein structures. Extent of variations in bond geometry during the simulation is within acceptable ranges of bond lengths and angles. However, the steric clash-free (φ,ψ) space continuously changes as seen in bond geometry-specific (φ,ψ) steric maps at the residue level during simulations. (φ,ψ) regions that have steric clash at one timepoint can become steric clash-free at a different timepoint through minor adjustments to backbone bond lengths and angles. Also instances of (φ,ψ) transitions from the left to right half of the (φ,ψ) map in consecutive snapshots of the trajectory are seen. Although the two quadrants are separated by a steric clash-prone region, corresponding to a high-energy barrier, height of this barrier is lowered by adjusting the bond geometry such that a bridging region of steric clash-free, low-energy (φ,ψ) values is formed. We demonstrate the idea of dynamically varying nature of acceptable and accessible (φ,ψ) steric space in proteins, which has implications for protein folding; proteins could sample (φ,ψ) space which is originally considered to be inaccessible, during folding, through minor adjustments to their backbone bond geometry.

Conformational Strain Indicated by Ramachandran Angles for the Protein Backbone Is Only Weakly Related to the Flexibility.

Studies on energy associated with free dipeptides have shown that conformers with unfavorable (ϕ,ψ) torsion angles have higher energy compared to conformers with favorable (ϕ,ψ) angles. It is expected that higher energy confers higher dynamics and flexibility to that part of the protein. Here, we explore a potential relationship between conformational strain in a residue due to unfavorable (ϕ,ψ) angles and its flexibility and dynamics in the context of protein structures. We compared flexibility of strained and relaxed residues, which are recognized based on outlier/allowed and favorable (ϕ,ψ) angles respectively, using normal-mode analysis (NMA). We also performed in-depth analysis on flexibility and dynamics at catalytic residues in protein kinases, which exhibit different strain status in different kinase structures using NMA and molecular dynamics simulations. We underline that strain of a residue, as defined by backbone torsion angles, is almost unrelated to the flexibility and dynamics associated with it. Even the overall trend observed among all high-resolution structures in which relaxed residues tend to have slightly higher flexibility than strained residues is counterintuitive. Consequently, we propose that identifying strained residues based on (ϕ,ψ) values is not an effective way to recognize energetic strain in protein structures.

Stereochemical Assessment of (φ,ψ) Outliers in Protein Structures Using Bond Geometry-Specific Ramachandran Steric-Maps.

Ramachandran validation of protein structures is commonly performed using developments, such as MolProbity. We suggest tailoring such analyses by position-wise, geometry-specific steric-maps, which show (φ,ψ) regions with steric-clash at every residue position. These maps are different from the classical steric-map because they are highly sensitive to bond length and angle values that are used, in our steric-maps, as observed in the residue positions in super-high-resolution peptide and protein structures. (φ,ψ) outliers observed in such structures seldom have steric-clash. Therefore, we propose that a (φ,ψ) outlier is unacceptable if it is located within the steric-clash region of a bond geometry-specific steric-map for a residue position. These steric-maps also suggest position-specific accessible (φ,ψ) space. The PARAMA web resource performs in-depth position-wise analysis of protein structures using bond geometry-specific steric-maps.

Protein stabilization by tuning the steric restraint at the reverse turn.

Reverse turns are solvent-exposed motifs in proteins that are crucial in nucleating ß-sheets and drive the protein folding. The solvent-exposed nature makes reverse turns more amenable to chemical modifications than α-helices or ß-sheets towards modulating the stability of re-engineered proteins. Here, we utilize van der Waals repulsive forces in tuning the steric restraint at the reverse turn. The steric restraint induced upon N-methylation of the i+1-i+2 amide bond at the reverse turn results in well-folded and stable ß-sheets in aqueous solution at room temperature. The developed superactive turn inducing motif is tolerant to a wide variety of functional groups present on coded amino acids making the designed turn fully compatible with bioactive loops in proteins. We demonstrate that the steric restraint and the functional groups at the reverse turn act in synergy to modulate the folding of re-engineered ß-sheets. Introduction of the turn motifs onto a three-stranded ß-sheet protein, Pin 1 WW domain, resulted in various analogs showing a cooperative two-state transition with thermal stability (TM) ranging from 62 °C to 82 °C. Despite modulating the stability of Pin 1 variants by ∼2.8 kcal mol-1 (ΔΔGf), the native fold in all the protein variants was found to be unperturbed. This structural stability is brought about by conformational preorganization at the engineered reverse turn that results in strong intramolecular hydrogen bonds along the three dimensional structure of the protein. Thus, this simple loop engineering strategy via two amino acid substitution provides us a "toolkit" to modulate the stability of ß-sheet containing peptides and proteins in aqueous solution that will greatly expand the scope of de novo protein and foldamer design.