Lei Yu

Cyclic peptides are an emerging therapeutic modality that can target protein-protein interaction sites with high affinity and selectivity. A common medicinal chemistry strategy for the optimization of peptide hits is conformational stabilization through macrocyclization. We present a method based on explicit solvent enhanced sampling molecular dynamics simulations for estimating the impact of varying linker lengths and chemistry on the conformational stability of a peptide. The method is demonstrated on three cyclic peptide series that bind to proteins PCSK9, trypsin, and MDM2 adopting loop, β-sheet, and helical secondary structures. In general, the simulations show greater solution stability of the receptor-bound conformation for the higher-affinity peptides, consistent with the idea that preorganizing a ligand for binding can enhance binding affinity. The impact of the force field and sampling is discussed for one series that does not follow this trend. We have successfully applied this method to internal discovery programs to design peptides with increased potency and chemical stability.

Intrinsically disordered proteins (IDPs) are highly dynamic systems that play an important role in cell signaling processes and their misfunction often causes human disease. Proper understanding of IDP function not only requires the realistic characterization of their three-dimensional conformational ensembles at atomic-level resolution but also of the time scales of interconversion between their conformational substates. Large sets of experimental data are often used in combination with molecular modeling to restrain or bias models to improve agreement with experiment. It is shown here for the N-terminal transactivation domain of p53 (p53TAD) and Pup, which are two IDPs that fold upon binding to their targets, how the latest advancements in molecular dynamics (MD) simulations methodology produces native conformational ensembles by combining replica exchange with series of microsecond MD simulations. They closely reproduce experimental data at the global conformational ensemble level, in terms of the distribution properties of the radius of gyration tensor, and at the local level, in terms of NMR properties including ¹⁵N spin relaxation, without the need for reweighting. Further inspection revealed that 10–20% of the individual MD trajectories display the formation of secondary structures not observed in the experimental NMR data. The IDP ensembles were analyzed by graph theory to identify dominant inter-residue contact clusters and characteristic amino-acid contact propensities. These findings indicate that modern MD force fields with residue-specific backbone potentials can produce highly realistic IDP ensembles sampling a hierarchy of nano- and picosecond time scales providing new insights into their biological function.

The power of structural information for informing biological mechanisms is clear for stable folded macromolecules, but similar structure-function insight is more difficult to obtain for highly dynamic systems such as intrinsically disordered proteins (IDPs) which must be described as structural ensembles. Here, we present IDPConformerGenerator, a flexible, modular open-source software platform for generating large and diverse ensembles of disordered protein states that builds conformers that obey geometric, steric, and other physical restraints on the input sequence. IDPConformerGenerator samples backbone phi (φ), psi (ψ), and omega (ω) torsion angles of relevant sequence fragments from loops and secondary structure elements extracted from folded protein structures in the RCSB Protein Data Bank and builds side chains from robust Monte Carlo algorithms using expanded rotamer libraries. IDPConformerGenerator has many user-defined options enabling variable fractional sampling of secondary structures, supports Bayesian models for assessing the agreement of IDP ensembles for consistency with experimental data, and introduces a machine learning approach to transform between internal and Cartesian coordinates with reduced error. IDPConformerGenerator will facilitate the characterization of disordered proteins to ultimately provide structural insights into these states that have key biological functions.

Intrinsically disordered proteins (IDPs) are highly dynamic systems that play an important role in cell signaling processes and their misfunction often causes human disease. Proper understanding of IDP function not only requires the realistic characterization of their three-dimensional conformational ensembles at atomic-level resolution but also of the time scales of interconversion between their conformational substates. Large sets of experimental data are often used in combination with molecular modeling to restrain or bias models to improve agreement with experiment. It is shown here for the N-terminal transactivation domain of p53 (p53TAD) and Pup how the latest advancements in molecular dynamics (MD) simulations methodology produces native conformational ensembles by combining replica exchange with series of microsecond MD simulations. They closely reproduce experimental data at the global conformational ensemble level, in terms of the distribution properties of the radius of gyration tensor, and at the local level, in terms of NMR properties including ¹⁵ N spin relaxation, without the need for reweighting. The IDP ensembles were analyzed by graph theory to identify dominant inter-residue contact clusters and characteristic amino-acid contact propensities. These findings indicate that modern MD force fields with residue-specific backbone potentials can produce highly realistic IDP ensembles sampling a hierarchy of nano- and picosecond time scales providing new insights into their biological function. AUTHOR SUMMARY: Accurate prediction of the conformational ensemble dynamics sans bias is shown for intrinsically disordered proteins including the transactivation domain of p53.

Amino-acid side-chain properties in proteins are key determinants of protein function. NMR spin relaxation of side chains is an important source of information about local protein dynamics and flexibility. However, traditional solution NMR relaxation methods are most sensitive to sub-nanosecond dynamics lacking information on slower ns−μs time-scale motions. Nanoparticle-assisted NMR spin relaxation (NASR) of methyl-side chains is introduced here as a window into these ns−μs dynamics. NASR utilizes the transient and nonspecific interactions between folded proteins and slowly tumbling spherical nanoparticles (NPs), whereby the increase of the relaxation rates reflects motions on time scales from ps all the way to the overall tumbling correlation time of the NPs ranging from hundreds of ns to μs. The observed motional amplitude of each methyl group can then be expressed by a model-free NASR S² order parameter. The method is demonstrated for ²H-relaxation of CH₂D methyl moieties and cross-correlated relaxation of CH₃ groups for proteins Im7 and ubiquitin in the presence of anionic silica-nanoparticles. Both types of relaxation experiments, dominated by either quadrupolar or dipolar interactions, yield highly consistent results. Im7 shows additional dynamics on the intermediate time scales taking place in a functionally important loop, whereas ubiquitin visits the majority of its conformational substates on the sub-ns time scale. These experimental observations are in good agreement with 4–10 μs all-atom molecular dynamics trajectories. NASR probes side-chain dynamics on a much wider range of motional time scales than previously possible, thereby providing new insights into the interplay between protein structure, dynamics, and molecular interactions that govern protein function.

The prevalence of intrinsically disordered proteins (IDPs) and protein regions in structural biology has prompted the recent development of molecular dynamics (MD) force fields for the more realistic representations of such systems. Using experimental nuclear magnetic resonance backbone scalar ³J-coupling constants of the intrinsically disordered proteins α-synuclein and amyloid-β in their native aqueous environment as a metric, we compare the performance of four recent MD force fields, namely, AMBER ff14SB, CHARMM C36m, AMBER ff99SB-disp, and AMBER ff99SBnmr2, by partitioning the polypeptides into an overlapping series of heptapeptides for which a cumulative total of 276 μs MD simulations were performed. The results show substantial differences between the different force fields at the individual residue level. Except for ff99SBnmr2, the force fields systematically underestimate the scalar ³J(H^N,Hα)-couplings due to an underrepresentation of β-conformations and an overrepresentation of either α- or PP_II conformations. The study demonstrates that the incorporation of coil library information in modern MD force fields, as shown here for ff99SBnmr2, provides substantially improved performance and more realistic sampling of the local backbone dihedral angles of IDPs as reflected by the good accuracy of the computed scalar ³J(H^N,Hα)-couplings with less than 0.5 Hz error. Such force fields will enable a better understanding of how structural dynamics and thermodynamics influence the IDP function. Although the methodology based on heptapeptides used here does not allow the assessment of potential intramolecular long-range interactions, its computational affordability permits well-converged simulations that can be easily parallelized. This should make the quantitative validation of intrinsic disorder observed in MD simulations of polypeptides with experimental scalar J-couplings widely applicable.

The quantitative and comprehensive description of the internal dynamics of proteins is critical for understanding their function. Nanoparticle‐assisted ¹⁵N NMR spin relaxation spectroscopy is a new method for the observation of picosecond to microsecond dynamics of proteins when transiently interacting with the surface of the nanoparticles (NPs). The method is applied here to the protein ubiquitin in the presence of anionic and cationic silica NPs (SNPs) of different sizes. The backbone dynamics profiles are reproducible and strikingly similar to each other, indicating that specific protein–SNP interactions are unimportant. The dynamics profiles closely match the sub‐nanosecond dynamics S² values observed by model‐free analysis of standard ¹⁵N relaxation of ubiquitin in free solution, indicating that the bulk of the ubiquitin backbone dynamics in solution is confined to sub‐nanosecond timescales and, hence, it is dynamically more restrained than previous NMR studies have suggested.

The quantitative and comprehensive description of the internal dynamics of proteins is critical for understanding their function. Nanoparticle‐assisted ¹⁵N NMR spin relaxation spectroscopy is a new method for the observation of picosecond to microsecond dynamics of proteins when transiently interacting with the surface of the nanoparticles (NPs). The method is applied here to the protein ubiquitin in the presence of anionic and cationic silica NPs (SNPs) of different sizes. The backbone dynamics profiles are reproducible and strikingly similar to each other, indicating that specific protein–SNP interactions are unimportant. The dynamics profiles closely match the sub‐nanosecond dynamics S² values observed by model‐free analysis of standard ¹⁵N relaxation of ubiquitin in free solution, indicating that the bulk of the ubiquitin backbone dynamics in solution is confined to sub‐nanosecond timescales and, hence, it is dynamically more restrained than previous NMR studies have suggested.

Molecular dynamics (MD) simulations provide a unique atomic-level description of the structure and dynamics of proteins, which is essential for the mechanistic understanding of protein interactions and function in living organisms. However, traditional MD force fields that are optimized for folded proteins often generate overly compact structures and incorrect characteristics of intrinsically disordered proteins (IDPs) and protein regions (IDRs), thereby limiting the quantitative insights that can be gained from MD simulations. We introduce the residue-specific protein force field, ff99SBnmr1cl, which is derived from ff99SBnmr1 by balancing the backbone dihedral angle potentials in a residue-specific manner to quantitatively reproduce dihedral angle distributions from an experimental coil library. The new force field substantially improves the backbone conformational ensembles of disordered proteins, protein regions, and peptides while keeping well-defined protein structures stable and accurate. This balanced new force field should enable a myriad of applications that require quantitative descriptions of IDPs, IDRs, loop dynamics, and folding/unfolding equilibria in the presence and absence of interaction partners.

Nanoparticle-assisted protein NMR opens window to observation of novel functional dynamics in the nano- to microsecond range.