Can molecular dynamics simulations improve predictions of protein-ligand binding affinity with machine learning?

Binding affinity prediction largely determines the discovery efficiency of lead compounds in drug discovery. Recently, machine learning (ML)-based approaches have attracted much attention in hopes of enhancing the predictive performance of traditional physics-based approaches. In this study, we evaluated the impact of structural dynamic information on the binding affinity prediction by comparing the models trained on different dimensional descriptors, using three targets (i.e. JAK1, TAF1-BD2 and DDR1) and their corresponding ligands as the examples. Here, 2D descriptors are traditional ECFP4 fingerprints, 3D descriptors are the energy terms of the Smina and NNscore scoring functions and 4D descriptors contain the structural dynamic information derived from the trajectories based on molecular dynamics (MD) simulations. We systematically investigate the MD-refined binding affinity prediction performance of three classical ML algorithms (i.e. RF, SVR and XGB) as well as two common virtual screening methods, namely Glide docking and MM/PBSA. The outcomes of the ML models built using various dimensional descriptors and their combinations reveal that the MD refinement with the optimized protocol can improve the predictive performance on the TAF1-BD2 target with considerable structural flexibility, but not for the less flexible JAK1 and DDR1 targets, when taking docking poses as the initial structure instead of the crystal structures. The results highlight the importance of the initial structures to the final performance of the model through conformational analysis on the three targets with different flexibility.

Structural flexibility and heterogeneity of recombinant human glial fibrillary acidic protein (GFAP).

Glial fibrillary acidic protein (GFAP) is a promising biomarker for brain and spinal cord disorders. Recent studies have highlighted the differences in the reliability of GFAP measurements in different biological matrices. The reason for these discrepancies is poorly understood as our knowledge of the protein's 3-dimensional conformation, proteoforms, and aggregation remains limited. Here, we investigate the structural properties of GFAP under different conditions. For this, we characterized recombinant GFAP proteins from various suppliers and applied hydrogen-deuterium exchange mass spectrometry (HDX-MS) to provide a snapshot of the conformational dynamics of GFAP in artificial cerebrospinal fluid (aCSF) compared to the phosphate buffer. Our findings indicate that recombinant GFAP exists in various conformational species. Furthermore, we show that GFAP dimers remained intact under denaturing conditions. HDX-MS experiments show an overall decrease in H-bonding and an increase in solvent accessibility of GFAP in aCSF compared to the phosphate buffer, with clear indications of mixed EX2 and EX1 kinetics. To understand possible structural interface regions and the evolutionary conservation profiles, we combined HDX-MS results with the predicted GFAP-dimer structure by AlphaFold-Multimer. We found that deprotected regions with high structural flexibility in aCSF overlap with predicted conserved dimeric 1B and 2B domain interfaces. Structural property predictions combined with the HDX data show an overall deprotection and signatures of aggregation in aCSF. We anticipate that the outcomes of this research will contribute to a deeper understanding of the structural flexibility of GFAP and ultimately shed light on its behavior in different biological matrices.

Hydrazone-Linked Covalent Organic Frameworks.

Hydrazone-linked covalent organic frameworks (COFs) with structural flexibility, heteroatomic sites, post-modification ability and high hydrolytic stability have attracted great attention from scientific community. Hydrazone-linked COFs, as a subclass of Schiff-base COFs, was firstly reported in 2011 by Yaghi's group and later witnessed prosperous development in various aspects. Their adjustable structures, precise pore channels and plentiful heteroatomic sites of hydrazone-linked structures possess much potential in diverse applications, for example, adsorption/separation, chemical sensing, catalysis and energy storage, etc. Up to date, the systematic reviews about the reported hydrazone-linked COFs are still rare. Therefore, in this review, we will summarize their preparation methods, characteristics and related applications, and discuss the opportunity or challenge of hydrazone-linked COFs. We hope this review could provide new insights about hydrazone-linked COFs for exploring more appealing functions or applications.

Carboxyl group-modified myoglobin shows membrane-permeabilizing activity.

In this study, we investigated whether modification of the carboxyl group with semicarbazide-enabled myoglobin (Mb) exhibits membrane-perturbing activity in physiological solutions. Mass spectrometry analysis showed that semicarbazide molecules were coupled to 19 of the 22 carboxyl groups in semicarbazide-modified Mb (SEM-Mb). Measurements of the absorption and circular dichroism spectra indicated that SEM-Mb lost its heme group and reduced the content of the α-helix structure in Mb. The microenvironment surrounding Trp residues in Mb changes after blocking negatively charged residues, as shown by fluorescence quenching studies. The results of the trifluoroethanol-induced structural transition indicated that SEM-Mb had higher structural flexibility than that of Mb. SEM-Mb, but not Mb, induced the permeability of bilayer membranes. Both proteins showed similar lipid-binding affinities. The conformation of SEM-Mb and Mb changed upon binding to lipid vesicles or a membrane-mimicking environment composed of SDS micelles, suggesting that membrane interaction modes differ. Unlike lipid-bound Mb, Trp residues in lipid-bound SEM-Mb are located at the protein-lipid interface. Altogether, our data indicate that modifying negatively charged groups relieves the structural constraints in Mb, consequently switching Mb structure to an active conformation that exhibits membrane-permeabilizing activity.

Investigations on the influence of the structural flexibility of nanoliposomes on their properties.

In the present study, nanoliposomes with tuneable structure elasticity were prepared by reverse-phase evaporation. Both Fluorescence Polarization and Fluorescence Resonance Energy Transfer was employed to characterize the structural elasticity of resultant nanoliposomes. Temozolomide, a kind of hydrophilic drug as the first-line treatment choice for glioblastoma, was encapsulated into nanoliposomes. The results showed that the flexibility of nanoliposomes gradually increased with sodium cholate, while decreasing with cholesterol, Labrafac CC and Labrafac PG adding. Furthermore, the structural flexibility of nanoliposomes was positively correlated with the encapsulation efficiency and release rate and cellular uptake. Our research reveals the structural flexibility of nanoliposomes could affect in vitro characteristics and thereafter in vivo behaviors of nanoliposomes.

The child as a medium. Breakdown and possible resurgence of children's agency in the era of pandemic.

This paper deals with the unpredictable outbreak of the pandemic, explaining its impact on the education system, and with structural flexibility as a way to face unpredictability, based on the generalisability and coordination of manifestations of agency. The pandemic has enhanced a narrative of the child as a medium of learning, which undermines children's agency. The example of the research project CHILD-UP (Children Hybrid Integration: Learning Dialogue as a way of Upgrading policies of Participation) is used to show how children's agency and structural flexibility in classroom interactions can be supported and analysed.

Temperature-Programmed Separation of Hexane Isomers by a Porous Calcium Chloranilate Metal-Organic Framework.

The full separation of alkane isomers as a function of different degrees of branching remains a daunting challenge due to its stringent requirement with respect to pore dimensions of the adsorbents. In this work, we report a novel microporous coordination network built on calcium (II) and chloranilate. The compound has a flexible framework and exhibits temperature-dependent adsorption behavior toward hexane isomers. At 30 °C, it accommodates substantial amounts of linear and monobranched hexanes but fully excludes their dibranched isomer, and at elevated temperatures such as 150 °C, it acts as a splitter for linear and branched alkanes. Its capability of efficient discrimination of hexane isomers as a function of branching is verified by experimental breakthrough measurements. Ab initio calculations have uncovered the underlying selective size-exclusion separation mechanism.

Neutron scattering in photosynthesis research: recent advances and perspectives for testing crop plants.

The photosynthetic performance of crop plants under a variety of environmental factors and stress conditions, at the fundamental level, depends largely on the organization and structural flexibility of thylakoid membranes. These highly organized membranes accommodate virtually all protein complexes and additional compounds carrying out the light reactions of photosynthesis. Most regulatory mechanisms fine-tuning the photosynthetic functions affect the organization of thylakoid membranes at different levels of the structural complexity. In order to monitor these reorganizations, non-invasive techniques are of special value. On the mesoscopic scale, small-angle neutron scattering (SANS) has been shown to deliver statistically and spatially averaged information on the periodic organization of the thylakoid membranes in vivo and/or, in isolated thylakoids, under physiologically relevant conditions, without fixation or staining. More importantly, SANS investigations have revealed rapid reversible reorganizations on the timescale of several seconds and minutes. In this paper, we give a short introduction into the basics of SANS technique, advantages and limitations, and briefly overview recent advances and potential applications of this technique in the physiology and biotechnology of crop plants. We also discuss future perspectives of neutron crystallography and different neutron scattering techniques, which are anticipated to become more accessible and of more use in photosynthesis research at new facilities with higher fluxes and innovative instrumentation.

Structural Plasticity of LL-37 Indicates Elaborate Functional Adaptation Mechanisms to Bacterial Target Structures.

The human cathelicidin LL-37 is a multifunctional peptide of the human innate immune system. Among the various functions of LL-37, its antimicrobial activity is important in controlling the microorganisms of the human body. The target molecules of LL-37 in bacteria include membrane lipids, lipopolysaccharides (LPS), lipoteichoic acid (LTA), proteins, DNA and RNA. In this mini-review, we summarize the entity of LL-37 structural data determined over the last 15 years and specifically discuss features implicated in the interactions with lipid-like molecules. For this purpose, we discuss partial and full-length structures of LL-37 determined in the presence of membrane-mimicking detergents. This constantly growing structural database is now composed of monomers, dimers, tetramers, and fiber-like structures. The diversity of these structures underlines an unexpected plasticity and highlights the conformational and oligomeric adaptability of LL-37 necessary to target different molecular scaffolds. The recent co-crystal structures of LL-37 in complex with detergents are particularly useful to understand how these molecules mimic lipids and LPS to induce oligomerization and fibrillation. Defined detergent binding sites provide deep insights into a new class of peptide scaffolds, widening our view on the fascinating world of the LL-37 structural factotum. Together, the new structures in their evolutionary context allow for the assignment of functionally conserved residues in oligomerization and target interactions. Conserved phenylalanine and arginine residues primarily mediate those interactions with lipids and LPS. The interactions with macromolecules such as proteins or DNA remain largely unexplored and open a field for future studies aimed at structures of LL-37 complexes. These complexes will then allow for the structure-based rational design of LL-37-derived peptides with improved antibiotic properties.

Controlling the Substrate Specificity of an Enzyme through Structural Flexibility by Varying the Salt-Bridge Density.

Many enzymes, particularly in one single family, with highly conserved structures and folds exhibit rather distinct substrate specificities. The underlying mechanism remains elusive, the resolution of which is of great importance for biochemistry, biophysics, and bioengineering. Here, we performed a neutron scattering experiment and molecular dynamics (MD) simulations on two structurally similar CYP450 proteins; CYP101 primarily catalyzes one type of ligands, then CYP2C9 can catalyze a large range of substrates. We demonstrated that it is the high density of salt bridges in CYP101 that reduces its structural flexibility, which controls the ligand access channel and the fluctuation of the catalytic pocket, thus restricting its selection on substrates. Moreover, we performed MD simulations on 146 different kinds of CYP450 proteins, spanning distinct biological categories including Fungi, Archaea, Bacteria, Protista, Animalia, and Plantae, and found the above mechanism generally valid. We demonstrated that, by fine changes of chemistry (salt-bridge density), the CYP450 superfamily can vary the structural flexibility of its member proteins among different biological categories, and thus differentiate their substrate specificities to meet the specific biological needs. As this mechanism is well-controllable and easy to be implemented, we expect it to be generally applicable in future enzymatic engineering to develop proteins of desired substrate specificities.

The 3A6-TCR/superagonist/HLA-DR2a complex shows similar interface and reduced flexibility compared to the complex with self-peptide.

T-cell receptor (TCR) recognition of the myelin basic protein (MBP) peptide presented by major histocompatibility complex (MHC) protein HLA-DR2a, one of the MHC class II alleles associated with multiple sclerosis, is highly variable. Interactions in the trimolecular complex between the TCR of the MBP83-99-specific T cell clone 3A6 with the MBP-peptide/HLA-DR2a (abbreviated TCR/pMHC) lead to substantially different proliferative responses when comparing the wild-type decapeptide MBP90-99 and a superagonist peptide, which differs mainly in the residues that point toward the TCR. Here, we investigate the influence of the peptide sequence on the interface and intrinsic plasticity of the TCR/pMHC trimolecular and pMHC bimolecular complexes by molecular dynamics simulations. The intermolecular contacts at the TCR/pMHC interface are similar for the complexes with the superagonist and the MBP self-peptide. The orientation angle between TCR and pMHC fluctuates less in the complex with the superagonist peptide. Thus, the higher structural stability of the TCR/pMHC tripartite complex with the superagonist peptide, rather than a major difference in binding mode with respect to the self-peptide, seems to be responsible for the stronger proliferative response.

Structural basis for producing selective MAP2K7 inhibitors.

Mitogen-activated protein kinase kinase 7 (MAP2K7) in the c-Jun N-terminal kinase signal cascade is an attractive drug target for a variety of diseases. The selectivity of MAP2K7 inhibitors against off-target kinases is a major barrier in drug development. We report a crystal structure of MAP2K7 complexed with a potent covalent inhibitor bearing an acrylamide moiety as an electrophile, which discloses a structural basis for producing selective and potent MAP2K7 inhibitors.

Collective Breathing in an Eightfold Interpenetrated Metal-Organic Framework: From Mechanistic Understanding towards Threshold Sensing Architectures.

Functional materials that respond to chemical or physical stimuli through reversible structural transformations are highly desirable for the integration into devices. Now, a new stable and flexible eightfold interpenetrated three-dimensional (3D) metal-organic framework (MOF) is reported, [Zn(oba)(pip)]n (JUK-8) based on 4,4'-oxybis(benzenedicarboxylate) (oba) and 4-pyridyl functionalized benzene-1,3-dicarbohydrazide (pip) linkers, featuring distinct switchability in response to guest molecules (H2 O and CO2 ) or temperature. Single-crystal X-ray diffraction (SC-XRD), combined with density functional theory (DFT) and grand canonical Monte Carlo (GCMC) simulations, reveal a unique breathing mechanism involving collective motions of eight mixed-linker diamondoid subnetworks with only minor displacements between them. The pronounced stepwise volume change of JUK-8 during water adsorption is used to construct an electron conducting composite film for resistive humidity sensing.

Influences of conformations of peptides on stereoinversions and/or isomerizations of aspartic acid residues.

Recently, non-enzymatic stereoinversions of aspartic acid (Asp) residues in proteins and peptides have been reported. Here, we performed replica exchange molecular dynamics (REMD) simulations of model peptides (exon 6, 26A-1, and 26A-2) extracted from elastin to investigate their structural features, thereby revealing the factor that influences stereoinversions. For REMD trajectories, we calculated distances between carboxyl carbon in Asp and amide nitrogen in the (n + 1) residue (CN distances). Because bond formation between carbon and nitrogen is indispensable to the formation of a succinimide intermediate the distance between them seems to play an important role in stereoinversion. Moreover, we calculated polar surface areas (PSAs) for the trajectories, finding that CN distances and PSA were different for each peptide, with the longest CN distance and smallest PSA observed for exon 6 peptide, where stereoinversion of Asp is the slowest. Although the average CN distance was shorter for exon 26A-1 peptide than for exon 26A-2 peptide, the number of conformations with CN distances <3.0 Šwas greater for exon 26A-2 peptide than for exon 26A-1 peptide. Furthermore, PSA for amide nitrogen of the (n + 1) residue was larger for exon 26A-2 peptide than for exon 26A-1 peptide. These results indicated that the flexibility of Asp and (n + 1) residues and hydrophilicity of peptides, especially in the (n + 1) residue, play important roles in the stereoinversion of Asp. This article is part of a Special Issue entitled: D-Amino acids: biology in the mirror, edited by Dr. Loredano Pollegioni, Dr. Jean-Pierre Mothet and Dr. Molla Gianluca.

The molecular basis for lipase stereoselectivity.

Lipases are among the most applied biocatalysts in organic synthesis to catalyze the kinetic resolution of a wide range of racemic substrates to yield optically pure compounds. Due to the rapidly increased demands for optically pure compounds, deep understanding of the molecular basis for lipase stereoselectivity and how to obtain lipases with excellent asymmetric selectivity have become one of primary research goals in this field. This review is focused on the molecular factors that have impacts on the stereoselectivity of lipases including the steric complementarity between the lipase topological structure and its substrate, the regional structural flexibility, the hydrogen bonds between the residues around the catalytic site and the tetrahedral intermediates, and the electrostatic interactions between surface residues. Moreover, the synergistic effects of these structural factors on the catalytic properties including stereoselectivity, activity, and stability are also discussed.

Modeling of Protein Structural Flexibility and Large-Scale Dynamics: Coarse-Grained Simulations and Elastic Network Models.

Fluctuations of protein three-dimensional structures and large-scale conformational transitions are crucial for the biological function of proteins and their complexes. Experimental studies of such phenomena remain very challenging and therefore molecular modeling can be a good alternative or a valuable supporting tool for the investigation of large molecular systems and long-time events. In this minireview, we present two alternative approaches to the coarse-grained (CG) modeling of dynamic properties of protein systems. We discuss two CG representations of polypeptide chains used for Monte Carlo dynamics simulations of protein local dynamics and conformational transitions, and highly simplified structure-based elastic network models of protein flexibility. In contrast to classical all-atom molecular dynamics, the modeling strategies discussed here allow the quite accurate modeling of much larger systems and longer-time dynamic phenomena. We briefly describe the main features of these models and outline some of their applications, including modeling of near-native structure fluctuations, sampling of large regions of the protein conformational space, or possible support for the structure prediction of large proteins and their complexes.

Synthesis of Densely Packaged, Ultrasmall Pt02 Clusters within a Thioether-Functionalized MOF: Catalytic Activity in Industrial Reactions at Low Temperature.

The gram-scale synthesis, stabilization, and characterization of well-defined ultrasmall subnanometric catalytic clusters on solids is a challenge. The chemical synthesis and X-ray snapshots of Pt02 clusters, homogenously distributed and densely packaged within the channels of a metal-organic framework, is presented. This hybrid material catalyzes efficiently, and even more importantly from an economic and environmental viewpoint, at low temperature (25 to 140 °C), energetically costly industrial reactions in the gas phase such as HCN production, CO2 methanation, and alkene hydrogenations. These results open the way for the design of precisely defined catalytically active ultrasmall metal clusters in solids for technically easier, cheaper, and dramatically less-dangerous industrial reactions.

Selective Catalytic Performances of Noble Metal Nanoparticle@MOF Composites: The Concomitant Effect of Aperture Size and Structural Flexibility of MOF Matrices.

Noble metal nanoparticles embedded in metal-organic frameworks (MOFs) are composite catalysts with enhanced or novel properties compared to the pristine counterparts. In recent years, to determine the role of MOFs during catalytic process, most studies have focussed on the confinement effect of MOFs, but ignored the structural flexibility of MOFs. In this study, we use two composite catalysts, Pt@ZIF-8 [Zn(mIM)2 , mIM=2-methyl imidazole] with flexible structure and Pt@ZIF-71 [Zn(DClIM)2 , DClIM=4,5-dichloroimidazole] with rigid structure, and hydrogenation of cinnamaldehyde as model reaction, to show the confinement effect and the structure flexibility of MOF matrices on the catalytic performance of composite catalysts. Both catalysts showed high selectivity for cinnamic alcohol with the confinement effect of the aperture. But, compared to Pt@ZIF-71, Pt@ZIF-8 exhibited higher conversion but lower selectivity owing to the flexible structure. The above results remind us that we will have to consider both the aperture size of MOFs and structure flexibility to select the proper MOF matrices for the composite materials to achieve the optimized performance.

A structural perspective of RNA recognition by intrinsically disordered proteins.

Protein-RNA recognition is essential for gene expression and its regulation, which is indispensable for the survival of the living organism at one hand, on the other hand, misregulation of this recognition may lead to their extinction. Polymorphic conformation of both the interacting partners is a characteristic feature of such molecular recognition that promotes the assembly. Many RNA binding proteins (RBP) or regions in them are found to be intrinsically disordered, and this property helps them to play a central role in the regulatory processes. Sequence composition and the length of the flexible linkers between RNA binding domains in RBPs are crucial in making significant contacts with its partner RNA. Polymorphic conformations of RBPs can provide thermodynamic advantage to its binding partner while acting as a chaperone. Prolonged extensions of the disordered regions in RBPs also contribute to the stability of the large cellular machines including ribosome and viral assemblies. The involvement of these disordered regions in most of the significant cellular processes makes RBPs highly associated with various human diseases that arise due to their misregulation.

Hybrid simulation approach incorporating microscopic interaction along with rigid body degrees of freedom for stacking between base pairs.

Stacking interaction between the aromatic heterocyclic bases plays an important role in the double helical structures of nucleic acids. Considering the base as rigid body, there are total of 18 degrees of freedom of a dinucleotide step. Some of these parameters show sequence preferences, indicating that the detailed atomic interactions are important in the stacking. Large variants of non-canonical base pairs have been seen in the crystallographic structures of RNA. However, their stacking preferences are not thoroughly deciphered yet from experimental results. The current theoretical approaches use either the rigid body degrees of freedom where the atomic information are lost or computationally expensive all atom simulations. We have used a hybrid simulation approach incorporating Monte-Carlo Metropolis sampling in the hyperspace of 18 stacking parameters where the interaction energies using AMBER-parm99bsc0 and CHARMM-36 force-fields were calculated from atomic positions. We have also performed stacking energy calculations for structures from Monte-Carlo ensemble by Dispersion corrected density functional theory. The available experimental data with Watson-Crick base pairs are compared to establish the validity of the method. Stacking interaction involving A:U and G:C base pairs with non-canonical G:U base pairs also were calculated and showed that these structures were also sequence dependent. This approach could be useful to generate multiscale modeling of nucleic acids in terms of coarse-grained parameters where the atomic interactions are preserved. This method would also be useful to predict structure and dynamics of different base pair steps containing non Watson-Crick base pairs, as found often in the non-coding RNA structures. © 2015 Wiley Periodicals, Inc. Biopolymers 105: 212-226, 2016.