Download Hi-Res ImageDownload to MS-PowerPointCite This:J. Chem. Inf. Model. 2024, 64, 19, 7816-7825

BioinformaticsOctober 3, 2024

Modeling Protein–Glycan Interactions with HADDOCK
Click to copy article linkArticle link copied!

Anna Ranaudo
Anna Ranaudo
Department of Earth and Environmental Sciences, University of Milano-Bicocca, Piazza Della Scienza 1, Milan 20126, Italy
Bijvoet Centre for Biomolecular Research, Faculty of Science - Chemistry, Utrecht University, Padualaan 8, Utrecht 3584CH, The Netherlands
More by Anna Ranaudo
https://orcid.org/0000-0002-8733-7938
Marco Giulini
Marco Giulini
Bijvoet Centre for Biomolecular Research, Faculty of Science - Chemistry, Utrecht University, Padualaan 8, Utrecht 3584CH, The Netherlands
More by Marco Giulini
Angela Pelissou Ayuso
Angela Pelissou Ayuso
Bijvoet Centre for Biomolecular Research, Faculty of Science - Chemistry, Utrecht University, Padualaan 8, Utrecht 3584CH, The Netherlands
More by Angela Pelissou Ayuso
Alexandre M. J. J. Bonvin*
Alexandre M. J. J. Bonvin
Bijvoet Centre for Biomolecular Research, Faculty of Science - Chemistry, Utrecht University, Padualaan 8, Utrecht 3584CH, The Netherlands
*Email: [email protected]
More by Alexandre M. J. J. Bonvin
https://orcid.org/0000-0001-7369-1322

Open PDF Supporting Information (2)

Journal of Chemical Information and Modeling

Cite this: J. Chem. Inf. Model. 2024, 64, 19, 7816–7825

Click to copy citationCitation copied!

https://doi.org/10.1021/acs.jcim.4c01372

Published October 3, 2024

CC-BY 4.0 .

Abstract

Click to copy section linkSection link copied!

The term glycan refers to a broad category of molecules composed of monosaccharide units linked to each other in a variety of ways, whose structural diversity is related to different functions in living organisms. Among others, glycans are recognized by proteins with the aim of carrying information and for signaling purposes. Determining the three-dimensional structures of protein–glycan complexes is essential both for the understanding of the mechanisms glycans are involved in and for applications such as drug design. In this context, molecular docking approaches are of undoubted importance as complementary approaches to experiments. In this study, we show how high ambiguity-driven DOCKing (HADDOCK) can be efficiently used for the prediction of protein–glycan complexes. Using a benchmark of 89 complexes, starting from their bound or unbound forms, and assuming some knowledge of the binding site on the protein, our protocol reaches a 70% and 40% top 5 success rate on bound and unbound data sets, respectively. We show that the main limiting factor is related to the complexity of the glycan to be modeled and the associated conformational flexibility.

This publication is licensed under

CC-BY 4.0 .

License Summary*
You are free to share(copy and redistribute) this article in any medium or format and to adapt(remix, transform, and build upon) the material for any purpose, even commercially within the parameters below:
- Creative Commons (CC): This is a Creative Commons license.
- Attribution (BY): Credit must be given to the creator.
View full license
*Disclaimer
This summary highlights only some of the key features and terms of the actual license. It is not a license and has no legal value. Carefully review the actual license before using these materials.
License Summary*
You are free to share(copy and redistribute) this article in any medium or format and to adapt(remix, transform, and build upon) the material for any purpose, even commercially within the parameters below:
- Creative Commons (CC): This is a Creative Commons license.
- Attribution (BY): Credit must be given to the creator.
View full license
*Disclaimer
This summary highlights only some of the key features and terms of the actual license. It is not a license and has no legal value. Carefully review the actual license before using these materials.

Subjects

Introduction

Click to copy section linkSection link copied!

Glycans are organic compounds with a polymeric structure consisting of monosaccharides, small building blocks covalently linked to each other in various ways through glycosidic bonds, in linear or branched arrangements. Depending on the number of monosaccharide units, glycans are named disaccharides (2 units), oligosaccharides (3–10 units), and polysaccharides (>10 units). (1) For simplicity, we use here the term “glycan” for referring to compounds with any number of monosaccharide units.

Glycans can form complex structures. This complexity already lies in monosaccharides themselves, which have a high degree of intrinsic chemical variability. The second source of complexity arises from the way monosaccharides are linked to each other as each glycosidic bond can form two possible stereoisomers at the anomeric carbon, i.e., the carbon whose asymmetric center is formed upon cyclization of the monosaccharide. Regioisomers may also exist because of the many hydroxyl groups, which allow two monosaccharides to be linked in several ways. (2) Monosaccharides, having the ability to create more than two glycosidic bonds, can give rise to branched chains. The frequent occurrence of branched patterns, as opposed to the linear patterns typically found in most peptides and oligonucleotides, contributes to the complexity of the glycans’ structural landscape. Besides the very large amount of glycans that can be built starting from a few monosaccharides, glycans show large conformational variability at room temperature due to the low torsional energy barriers around the glycosidic bonds. (3)

Glycans are ubiquitously found in living organisms, where they can be freestanding or linked to proteins and lipids, giving rise to glycoproteins and glycolipids, respectively. They are involved in a variety of biological processes that can be classified into three main categories. (4) First, they can play structural roles by either assisting in creating extracellular scaffolds, such as cell walls and matrices, or by being involved in protein folding and function. Second, they can serve as crucial sources of energy metabolism. Third, they can play the role of information carriers, being recognized by glycan-binding proteins (GBPs). Glycans can bind proteins through non-covalent, reversible interactions for signaling purposes, initiating a range of biological processes in both plants and animals. (5) The role of glycans has recently been highlighted in the context of the well-known SARS-CoV-2 spike protein. This protein, which enters host cells by connecting to the angiotensin-converting enzyme (ACE2), is surrounded by a layer of glycans, which hides it from the immune system. Casalino et al. (6) showed that specific glycans play a crucial role in the movement and structure of the part of the spike protein that binds to ACE2. Their removal results in diminished binding to ACE2.

It is evident that understanding the way glycans interact with proteins is of crucial importance. Experimental methods such as X-ray crystallography and cryo-electron microscopy face challenges when dealing with glycans because of their intrinsic flexibility and heterogeneity. Computational approaches, such as molecular docking, can provide an alternative, less expensive, and faster way of generating three-dimensional (3D) models of protein–glycan complexes.

Despite the growing attention that is being given to the carbohydrates field, (7,8) the modeling of protein–glycan complexes by docking has only received limited attention compared to that of other biomolecular complexes. A few protocols have been developed in the past years, such as GlycanDock based on Rosetta, (9) a heparin-specific protocol based on PIPER/ClusPro, (10) Vina-Carb, (11) GlycoTorch Vina, (12) RosettaCarbohydrate, (13) and ATTRACT. (14) Recently, a new version of Alphafold, (15) Alphafold3, (16) that can handle glycosylated proteins, has been published. It, however, does not allow one to model protein–glycan complexes. State-of-the-art protein–ligand docking software encounters challenges in addressing the conformational variability of glycans as they are usually developed for dealing with small, more rigid molecules. (9)

In this study, we use high ambiguity-driven DOCKing (HADDOCK) (17,18) to address the protein–glycan interaction prediction problem. HADDOCK is an information-driven docking approach that can harvest knowledge about binding sites to drive the docking process (see below). Note that HADDOCK (18) has already been applied to glycan modeling (19,20) but, to date, without an exhaustive benchmarking. Here, we first benchmark the baseline performance of HADDOCK in predicting the 3D structures of protein–glycan complexes on a bound data set composed of 89 high-resolution experimental complexes from the Protein Data Bank (PDB) (21) assuming an ideal scenario in which the binding interfaces on both protein and glycan are known in order to drive the docking process. A protocol is then proposed to deal with a realistic scenario in which the bound conformations of the partners are unknown. The GLYCAM-Web web server (22,23) is used for the generation of glycan unbound structures, while the unbound protein structures are retrieved from the PDB. Finally, to address the conformational flexibility challenge of glycans, we assess whether providing an ensemble of glycan conformations generated through a short conformational sampling carried out with HADDOCK prior to the docking process can improve the performance of the protocol, next to the standard semi-flexible refinement of the interface.

Methods

Click to copy section linkSection link copied!

Benchmark Data Set Preparation

HADDOCK’s (17,18) performance in reproducing the binding geometries of protein–glycan complexes was evaluated by exploiting an adapted version of the data set provided in GlycanDock. (9) It is composed of 109 experimentally determined high-resolution (<2.0 Å) protein–glycan complexes collected from the PDB. By discarding the entries containing glycans not yet supported by HADDOCK (refer to https://rascar.science.uu.nl/haddock2.4/library for a list of supported glycans), a data set of 89 complexes was obtained, which will be referred to as the bound data set from now on. The protein receptors in this data set include 8 antibodies, 21 carbohydrate-binding modules, 18 enzymes, 27 lectins or glycan-binding proteins (GBPs), and 15 viral glycan-binding proteins. The length of the glycans ranges from 2 to 7 monosaccharide units, 72 of which are linear glycans and the remaining 17 branched ones. This structural diversity is considered in the analysis of the docking performance.

With the aim of evaluating HADDOCK’s performance in a realistic unbound docking scenario, a subset of 55 complexes (out of 89) was defined for which unbound protein forms were available in the PDB. The glycans’ unbound conformations were generated with the GLYCAM-Web web server. (22,23) This data set of 55 unbound conformations of both proteins and glycans will be referred to as the unbound data set from now on. It contains 47 linear and 8 branched glycans; 25 glycans are composed of three or fewer monosaccharide units, while 30 have more than three units. For the evaluation of HADDOCK’s performance on the unbound data set, all the complexes including glycans made up of three or fewer monosaccharide units are treated together and referred to with the label SL-SB (short linear-short branched). For the larger bound data set, linear (SL) and branched (SB) short glycans are analyzed separately. The complexes including glycans composed of more than three units (L for long) are divided into linear (LL, 23 cases) and branched (LB, 7 cases). Glycans and protein structures were prepared for docking as described in Supporting Information Text S1. Details on the two data sets are reported in Supporting Information File 1, and the Symbol Nomenclature for Glycans (24,25) (SNFG) is given in Supporting Information Table S1. All HADDOCK-ready bound and unbound conformations, together with restraints, HADDOCK configuration files, and analysis scripts, are provided in the following GitHub repository: https://github.com/haddocking/protein-glycans.

HADDOCK General Protocol and Scoring Function

Docking calculations were performed with HADDOCK3 (https://github.com/haddocking/haddock3, DOI:10.5281/zenodo.10527751), the new, modular version of the well-established HADDOCK 2.X software. (26) The original HADDOCK protocol consists of three successive steps: (i) full randomization of the orientations and docking by rigid-body energy minimization; (ii) semi-flexible refinement by simulated annealing in torsion angle space during which the interfaces are considered flexible; and (iii) final refinement, either by energy minimization (current default) or by a short molecular dynamics simulation in explicit solvent. HADDOCK3 overcomes this rigid workflow structure as its constituent modules can be freely combined and interchanged by the user. A description of the HADDOCK3 modules used in this study is given in Supporting Information Text S2.

HADDOCK’s scoring function (HS) includes the intermolecular electrostatic (E_elec) and van der Waals (E_vdW) energies (calculated with the OPLS force field (27)), an empirical desolvation energy term (E_desolv), (28) the buried surface area (E_BSA), and the ambiguous interaction restraint energy (E_air) (see below). The combination and weights of those terms depend on the stage of the protocol. In the present study, we compare the default scoring function at the rigid-body stage (eq 1) with one in which the weight of the van der Waals energy term was increased from 0.01 to 1.0 (eq 2) as recommended for small-molecule docking with HADDOCK. (29) The scoring function in eq 2 will be referred to with the name vdW to be distinguished from the default one. For the flexible refinement stage, the default scoring function is used (eq 3).

H S (r i g i d ‐ b o d y d e f a u l t) = 0.01 E_{v d w} + 1.0 E_{e l} + 1.0 E_{d e s o l v} + 0.01 E_{a i r} - 0.01 E_{B S A}

(1)

H S (r i g i d ‐ b o d y v d W) = 1.0 E_{v d w} + 1.0 E_{e l} + 1.0 E_{d e s o l v} + 0.01 E_{a i r} - 0.01 E_{B S A}

(2)

H S (f l e x r e f d e f a u l t) = 1.0 E_{v d w} + 1.0 E_{e l} + 1.0 E_{d e s o l v} + 0.1 E_{a i r} - 0.01 E_{B S A}

(3)

Restraints to Drive the Docking

Experimental or predicted information about binding sites can be introduced as restraints for guiding the docking process and scoring the generated models (E_air in eqs 1–3). Interface information is typically introduced as ambiguous interaction restraints (AIRs), (18) which are defined from lists of residues divided into two groups: active and passive. Active residues are those of central importance for the interaction. They are restrained to be part of the interface throughout the docking and refinement process; if an active residue is not part of the interface in a model, it generates restraint energy (and corresponding forces). Passive residues are those that could contribute to the interaction, being among the possible interaction partners of the active residues defined on the other molecules; if a passive residue is not in the interface of a given model, there is no restraint energy generated.

The functional form of the restraint energy function is similar to the distance restraining function introduced by Nilges, (30) namely, a flat bottom function which is harmonic for short violations and then transits to a linear form, thus ensuring stable force computations.

In this work, two scenarios in terms of AIRs were considered: (i) true-interface scenario (ti-aa), where active residues, corresponding to the interface residues within 3.9 Å (31,32) from the partner, are defined for both the protein and the glycan; (ii) true-interface-protein–full glycan passive scenario (tip-ap), where active residues are still defined for the protein interface, but all residues of the glycan are listed as passive. By default, HADDOCK randomly discarded 50% of the defined AIRs for each docking model. This is done to account for possible wrong information (false positives) in the experimental data.

Docking Protocols for Bound and Unbound Data Sets

Two different protocols were used for the bound and unbound data sets. For the bound data set, a workflow consisting of the following six modules was defined for running HADDOCK3 (details in Supporting Information Table S2 and the corresponding config file available in the GitHub repository):

1	topoaa: creation of the topologies of the two partners during which any missing atoms are automatically added.
2	rigidbody: AIR-driven generation of rigid-body models.
3	caprieval: models’ quality analysis (see below).
4	rmsdmatrix: calculation of the RMSD matrix between all the models based on either all the interface residues (when ti-aa AIRs are used) or the protein interface residues and the whole glycan (when tip-ap AIRs are used).
5	clustrmsd: RMSD-based agglomerative hierarchical clustering of the models using the average linkage criterion and a distance cutoff of 2.5 Å. Only clusters containing four or more models are evaluated.
6	caprieval: cluster-based evaluation of the quality of the models.

As the starting structures are already in their bound conformations, no flexible refinement was performed.

For the unbound data set (details in Table S3 and the corresponding config file available in the GitHub repository), the workflow consisted of the following 12 modules:

1	topoaa: creation of the topologies of the two partners during which any missing atoms are automatically added.
2	rigidbody: AIR-driven generation of rigid-body docking models (1000 by default) with increased sampling (200 per conformation) when starting from an ensemble of conformations.
3	caprieval: models’ quality analysis (see below).
4	rmsdmatrix: calculation of the RMSD matrix between all the models based on either all the interface residues (when ti-aa AIRs are used) or the protein interface residues and the whole glycan (when tip-ap AIRs are used).
5	clustrmsd: RMSD-based agglomerative hierarchical clustering of the models using the average linkage criterion. Here, 50 (150 when the ensemble of glycan conformations is used) clusters are created.
6	seletopclusts: selection of the top 5 models of the existing clusters.
7	caprieval: cluster-based evaluation of the quality of the models.
8	flexref: semi-flexible refinement through a simulated annealing protocol in torsion angle space in which first side-chains and then side-chains and backbone of interface residues are treated as flexible.
9	caprieval: models’ quality analysis.
10	rmsdmatrix: calculation of the RMSD matrix between all the models, as in point 4.
11	clustrmsd: RMSD-based clustering of the models as in point 4 but here using a distance cutoff of 2.5 Å as in the bound scenario.
12	caprieval: cluster-based evaluation of the quality of the models.

The workflow for the unbound data set with the RMSD clustering of the rigid-body models followed by semi-flexible refinement is schematically represented in Figure 1.

Figure 1. Schematic representation of the docking protocol for the *unbound* data set. First, rigid-body docking is performed. The models are then clustered based on RMSD. The best-scoring models of each cluster are then subjected to a flexible refinement (interface), and the resulting models are again clustered and analyzed.

Model Quality Assessment

The quality of the models was evaluated using interface-ligand RMSD (IL-RMSD) with respect to the experimental structures. We did not use the fraction of native contacts (Fnat) as its values are typically quite high irrespective of the pairwise orientation between the molecules and thus unable to highlight small conformational differences, especially for small glycans. The IL-RMSD is calculated by first superimposing the model onto the reference structure by using the backbone atoms of the protein interface residues and then calculating the RMSD on the heavy atoms of the oligosaccharide. This is motivated by the fact that the protein interface is larger compared to the glycan interface and would dominate the RMSD calculation if the standard CAPRI (33) interface RMSD (I-RMSD) metric was used. The IL-RMSD gives a better measure of variations in the position of the ligand compared to the I-RMSD. The cutoffs used to define the quality of the models based on IL-RMSD are

high-quality models: IL-RMSD ≤ 1.0 Å
medium-quality models: IL-RMSD ≤ 2.0 Å
acceptable-quality models: IL-RMSD ≤ 3.0 Å
near acceptable-quality models: IL-RMSD ≤ 4.0 Å

Supporting Information Figure S1 shows examples of docking models falling into these four categories for short- and long-linear glycans and for branched glycans, highlighting how the near acceptable threshold is only appropriate when dealing with long glycans, being too permissive for short monosaccharide chains.

HADDOCK3′s performance is evaluated by calculating success rates (SRs), defined as the fraction of complexes having at least one high-, medium-, acceptable-, or near-acceptable-quality model among the top N models ranked according to the HADDOCK score. The cluster-based SR is also calculated by considering the top 4 scoring models within each cluster and assigning a given quality to a cluster if any of the top 4 members of the cluster reaches the corresponding quality cutoff. Note that the clustering step after rigid-body docking considers the 5 top models of each cluster.

Glycans’ Conformational Sampling

Conformational sampling of the glycans was carried out with the HADDOCK3 water refinement module (mdref (26)), starting from the models generated with the GLYCAM-Web web server. (22,23) Different scenarios were tested in terms of the number of molecular dynamics integration steps and the number of models generated (each simulation starting with a different random seed). The RMSD of the generated models with respect to the bound glycan conformation was calculated with the rmsdmatrix module. The generated models were clustered using RMSD-based hierarchical clustering, (34,35) requesting 20 clusters. The centers of each cluster were then used to define an ensemble of starting unbound conformations for docking. Further details of the conformational sampling can be found in Supporting Information Text S3.

Results and Discussion

Click to copy section linkSection link copied!

This section is structured as follows. First, the impact of the rigid-body scoring function on HADDOCK3’s performance is discussed based on docking calculations performed on the bound data set. The performance of the docking is then analyzed considering structural features of the glycans (length and branching) and the definition of the AIRs. We then focus on the more realistic scenario of unbound docking, assessing the best way of selecting the rigid-body models to be refined and the impact of the flexible refinement on the quality of the final models. Finally, the impact and limitations of using an ensemble of glycans as starting points for the docking are discussed.

Bound Docking Performance and the Impact of the Rigid-Body Scoring Function in the Ranking of Models

We first assessed the accuracy of the rigid-body scoring function in ranking the generated models by comparing the default scoring function (eq 1) to the one recommended for small ligands, in which the van der Waals energy weight is increased from 0.01 to 1.0 (vdW scoring function, eq 2). This was assessed by running docking calculations on the bound data set with true interface restraints (ti-aa AIRs). A comparison of success rates (SRs) obtained with the default (eq 1) and vdW (eq 2) scoring functions is shown in Figure 2.

Figure 2. Comparison of bound docking success rates obtained with the *default* (w_vdW = 0.01) and *vdW* (w_vdW = 1.0) scoring functions (eqs 1 and 2) as a function of the number of top-ranked models (T = 1, 5, 10, 50, 100, and 200) selected using true interface residues of both protein and glycan to define the ambiguous interaction restraints (**ti-aa** AIRs).

The vdW scoring function with increased van der Waals energy weight (w_vdW = 1.0) performs much better than the default function (w_vdW = 0.01) with remarkably higher success rates. This highlights the importance of the van der Waals energy term in scoring protein–glycan models. For example, considering T1 and T5 high-quality models, the improvement is around 30%. Even when considering the top 200 models (the default number of models passed to the flexible refinement stage in a standard HADDOCK2.X workflow), a slight improvement is observed in both number and quality of acceptable or better models. Results are consistent if we consider the cluster-based success rate (see Methods), with 66% of top-ranked clusters falling into the acceptable category in the vdW settings, compared to 33% of the default scenario. Glycans, despite all of their polar groups, have quite hydrophobic properties, especially in their ring structure. A similar behavior was observed in previous work when docking cyclic peptides (36) and small ligands. (29) Based on these results, all subsequent docking calculations were performed with the vdW scoring function.

Impact of Glycan Structural Features and AIR Definition on the Bound Docking Performance

The dependency of the structural features of the glycans on the SR was then assessed using the true-interface AIR restraints (ti-aa) and vdW scoring function scenario. SRs were calculated grouping the complexes based on the size of the glycans (glycans composed of three or fewer units, labeled with S, or more than three, with L) and connectivity (linear, L, or branched, B). The SR, shown in the first column of Supporting Information Figure S2, indicates that HADDOCK3 performs better for long-linear (LL) and long-branched (LB) glycans. The bound SR for T1 high-quality models is 45, 17, 74, and 73% for SL, SB, LL, and LB glycans, respectively (50, 33, 79, and 82%, respectively, for T1 acceptable or better-quality models). When considering a larger number of models, the SRs become rather similar for the various types of glycans (70–80% for T50 and 80–90% for T200), except for the short-branched (SB) ones (∼70%). The lower performance on the short-branched glycans and, to a lesser extent, on the SL ones could be due to the fact that smaller ligands could be accommodated into the protein-binding site with a greater variability of positions and orientations, while the range of possible docking orientations is more restricted for larger ligands.

The impact of defining the glycan as passive (tip-ap) was then investigated. A slight decrease in the SR can be observed (second column of Supporting Information Figure S2), with respect to the ti-aa scenario, for LB glycans and, to a lesser extent, for SB ones. This can be explained by noting that most of the linear-long (LL) glycans (27/34) have all of their residues involved in the binding to the protein, whereas the same is true for only half (5/11) of the long-branched (LB) ones. Branched glycans are thus more difficult to model when no information about their interface is available. Overall, HADDOCK3’s performance on protein–glycan complexes is not much affected by defining the glycan as passive (tip-ap AIRs). As experimental interface information for the glycans might not always be available in a realistic scenario, the remainder of the paper will discuss docking calculations only with tip-ap AIRs. Note that, for example, nuclear magnetic resonance (NMR) can provide specific information about which groups of a glycan are involved in the binding, as demonstrated for the modeling of the complex between sialic acid and the N-terminal domain of the SARS-Cov2 spike protein using NMR saturation transfer experiments. (37)

Unbound Docking

In a realistic scenario, the bound conformations of the docking partners are unknown, and only unbound structures or models will be available. As such, conformational rearrangements may be required during binding. We therefore assessed HADDOCK3’s performance on the unbound data set consisting of 55 complexes. The starting point for the docking was the unbound form of the protein taken from the PDB and models of the glycans obtained with the GLYCAM-Web web server (22,23) (see Methods section Benchmark Data Set Preparation). When dealing with unbound structures, the flexible refinement stage of the workflow might allow some conformational changes to occur. As only a fraction (typically around 20%) of the models is subjected to flexible refinement, ranking and selection of models after the rigid-body docking stage become crucial. We tested two scenarios: (1) the “classical” scenario in which the top 200 ranked models are passed to the flexible refinement stage; (2) a cluster-based selection, made possible by the modular and flexible structure of HADDOCK3. In the second scenario, the rigid-body models are first clustered based on their RMSD similarity; then, the top 5 models of all 50 clusters (150 when the ensemble is used, see below) are selected. In this way, models that might not rank high enough to be selected in an individual model ranking might still get selected and subjected to the flexible refinement, thus increasing the diversity of refined models (potentially at the cost of decreasing the overall number of acceptable or better models in the final set of refined models, which is not an issue provided that the scoring function can identify the near native models). Since we are using an agglomerative hierarchical clustering method allowing the desired number of clusters to be defined, clusters with less than 5 models might be obtained, and for this reason, a different number of models might be subjected to the flexible refinement for each complex. The overall docking performance and the performance by glycan size and branching are shown in Figure 3.

Figure 3. HADDOCK3’s performance on the *unbound* data set using the *vdW* scoring function and **tip-ap** AIRs (true interface on the protein defined as active and the glycan residues as passive). The success rates (SRs) (Y axis), defined as the percentage of complexes for which acceptable-, medium-, or high-quality models are generated, are calculated on the top 1 (T1) to top 200 (T200) ranked rigid-body models (column “rigid”), T1 to T50 rigid-body clusters, considering the top 5 models of each cluster (column “rigid + clustering”), the T1 to T200 ranked refined models (column “flexref”), and the T1 to T10 refined clusters, considering the top 4 models of each cluster (column “flexref + clustering”). SRs are shown separately for the whole data set (first row) and for the three categories of complexes grouped by glycan size and connectivity: **SL-SB** (second row), LL (third row), and LB (fourth row).

The comparison of the rigid-body SR obtained on the single models (column “rigid” in Figure 3) with the SR obtained on the clustered models (column “rigid + clustering” in Figure 3) reveals that, overall, the selection of clustered models is beneficial to the docking success rate. This is particularly helpful for the long-branched (LB) glycans (fourth row of Figure 3), with the clustering of the rigid-body models allowing us to retrieve more than 40% acceptable-quality models (70% near-acceptable-quality) compared to 15% acceptable-quality models (30% near-acceptable-quality) for the single model selection. Overall, for all types of glycans, selecting the rigid-body models after clustering is the best way to choose the structures for the refinement stage.

The introduction of flexibility in the interface region strongly improves the quality of the models. Comparing the T200 refined models (column “flexref” in Figure 3) with the selected T50 rigid-body clusters (column “rigid + clustering”), for short (SL-SB) glycans, almost 10% high-quality models are obtained, while there were none before, and almost 60% of the models fall within the medium-quality cutoff, to be compared with around 30% at the rigid-body stage. For the long-linear (LL) glycans, the acceptable-quality success rate increases from around 25% (rigid-body stage) to 61%. The improvement is also substantial for the LB glycans, for which the flexible refinement allows us to obtain medium-quality models, which were not present at the rigid-body stage, and an increase in acceptable-quality SR from 43 to 57%.

Finally, clustering of the refined models (fourth column in Figure 3) slightly improves the success rate compared to T10 single refined models, with around 50% of the glycans having an acceptable or better model in the top 10 clusters.

To demonstrate how flexible refinement affects both glycans’ conformations and models’ ranking, the best-scoring refined models of three representative complexes are shown in Figure 4, superimposed onto their corresponding rigid-body models and reference structures. These are representative of the SL-SB, LL, and LB groups. The refinement stage results in both better ranking and a better quality (lower IL-RMSD values) of the models. Overall, longer glycans are more challenging to refine than short ones. A high-quality model is produced for the 1C1L complex (SL, IL-RMSD = 0.52 Å after refinement), an acceptable-quality model for 5VX5 (LL, IL-RMSD = 2.22 Å), while for 1OH4 (LB, IL-RMSD = 4.42 Å), the quality is still not acceptable, although it improved after the refinement stage.

Figure 4. Superimposition of the best-scoring flexible refinement models (orange) and the rigid-body models (teal) to the reference structures (gray) for the complexes 1OH4 (LB), 5VX5 (LL), and 1C1L (SL) and the unbound docking scenario carried out with *vdW* scoring potential and **tip-ap** AIRs. Oxygen atoms of the glycans are shown in red in all the structures, nitrogen atoms in blue, and hydrogens are not shown. Ranking and IL-RMSD values with respect to the reference structures for the flexref and rigid-body models are shown as well.

We assess the overall impact of the refinement by measuring how much closer to the target structure is the best refined model with respect to the best model prior to the refinement stage; this analysis shows an average improvement of 0.43 Å on our data set with a maximum observed improvement of 3.73 Å. Of course, not all models improve after refinement, but those that do improve also tend to be ranked higher by the HADDOCK scoring function, as shown in Figure 3. Proteins containing long-linear glycans show the most substantial improvement (1.0 Å on average, with a maximum of 3.02 Å). Short- (SL-SB) and long-branched (LB) glycans show a negligible improvement (0.03 and 0.02 Å, respectively), although this is dependent on the specific glycan. For example, considering SL-SB glycans, we observe a maximum improvement of 3.73 Å (PDB 3AOF), while for some other models, the flexible refinement worsens their quality. For LB glycans, a maximum improvement of 2.13 Å is obtained (3AP9). Overall, the flexible refinement affects most LL glycans.

We investigated whether increasing the length of the flexible refinement protocol would further improve the glycan conformations. This did not improve the success rates significantly or in a uniform way as it seemed to depend on the number of models passed to the refinement and the group of glycans considered (data not shown). It further comes at increased computational costs. As this behavior could not be rationalized in a simple way, this approach was discarded.

Can an Ensemble of Presampled Glycan Conformations Improve the Docking Performance?

While the performance in the unbound docking scenario is already quite high, unsurprisingly, it does not reach the bound docking performance. One limiting factor here could be the unbound glycan conformations used for docking. Comparison of the conformations generated by the GLYCAM-Web web server with respect to the bound ones reveals a strong dependency of the RMSD to the bound form on both the glycan’s size and connectivity with mean RMSDs increasing from 0.89 Å for short-linear glycans to 1.74 Å for long-branched ones (see Supporting Information Figure S3).

As HADDOCK can take an ensemble of conformations as the starting point for the docking, we investigated if sampling conformations prior to docking could improve the overall docking performance. We used the water refinement module of HADDOCK, varying the number of models generated and the length of the molecular dynamics sampling. Six different protocols were investigated (Table S4). The RMSD distributions from the bound glycan conformation of the various protocols are shown as box plots in Supporting Information Figure S4 and compared to the distribution of the original GLYCAM server conformations. From this analysis, the protocol that generates conformations closest to the bound form consists of sampling 400 models, with a 16 times longer refinement protocol (sf400-x16) (which still remains a very short refinement protocol). With this sampling scenario, the average RMSD to the bound conformation decreases from 0.93 to 0.54 Å for SL-SB glycans, from 1.65 to 1.26 Å for LL glycans, and from 1.70 to 1.25 Å for LB glycans. An analysis of the glycosidic linkages indicates that the force field does describe favorable conformations (also during the flexible refinement stage of the docking), but the sampling protocol is too limited to generate conformations very different from those of the starting models generated by GLYCAM (see Supporting Information Figures S8 and S9). Some minor improvements toward the bound form are however observed in some cases.

For docking, we want to limit the number of models in an ensemble to allow for sufficient sampling of each starting model combination without having to increase the sampling too much. To this end, we clustered the sampled glycan conformations using the rmsdmatrix and clustrmsd modules in HADDOCK3, requesting either 10 or 20 clusters from the ensemble of conformations. RMSDs from the bound conformation were calculated for the centers of each cluster with the rmsdmatrix module. Requesting 20 clusters results in the best sampling of the overall glycan RMSD distribution of the ensemble of 400 models, retaining low RMSD conformations. The cluster centers, however, rarely correspond to the best RMSD sampled, which results in some rather limited loss in RMSD to the bound form (see Supporting Information Figure S5). Examples of such conformational ensembles for the same glycans reported in Figure 4 are shown in Supporting Information Figure S6.

The centers of those 20 clusters were provided as an ensemble for unbound docking, following the same rigid-body, cluster-based workflow described above. Only limited improvements are observed in single-structure ranking performance after flexible refinement compared to that of the single unbound conformer protocol (see Supporting Information Figure S7). For example, a higher number of medium-quality models for long-linear glycans is obtained, together with more high-quality models for SL-SB glycans. No improvement is observed for long-branched glycans. This can be attributed to the rather limited conformational sampling during the still short water refinement protocol (Supporting Information Figure S6). Better sampling strategies, possibly based on more extensive MD simulations, will be required. Another issue is related to the selection of the relevant representative conformations while limiting their number for docking purposes.

Conclusions

Click to copy section linkSection link copied!

In this study, we have assessed HADDOCK’s ability in modeling the 3D structures of protein–glycan complexes. First, the baseline performance was evaluated on the rather simple, unrealistic scenario starting from the partners in their bound conformations and giving full information about the interface (ti-aa AIRs). This allowed us to improve the rigid-body scoring function for protein–glycan complexes by increasing the weight of the intermolecular van der Waals energy term to 1.0, as done for the docking of small molecules to proteins. (29) An analysis of the bound docking performance per type of glycan revealed that longer glycans (in their bound conformation) are easier to model. This is probably a consequence of the lower number of possible orientations that longer glycans can assume when binding the protein. In the more realistic unbound docking scenario, in which the glycan conformations were modeled with the GLYCAM-Web web server and free form structures of the proteins are used, conformational changes are required to generate native-like poses. In this case, longer glycans are more difficult to model with only (near-)acceptable predictions generated in most cases.

Making use of the new modular version of HADDOCK3, we introduced a protocol in which rigid-body models are clustered prior to refinement, with representatives of all clusters being passed to the flexible refinement stage of HADDOCK. This strategy enables HADDOCK to retain 51% acceptable or better models after rigid-body selection compared to 33% using the standard single model selection. After flexible refinement, the overall performance reaches an overall success rate of 16% (respectively, 18%) and 38% (respectively, 44%) for T1 and T5 single-structure models (respectively, clusters).

Single-structure or cluster-based selection of models at the end of the workflow shows almost similar performance. This is in line with what was also observed in small-molecule docking with HADDOCK. (29,38) There is, however, still room to improve the scoring of models. One possible extension could be the incorporation of CH-π stacking interactions into the scoring function, as suggested in previous works. (39,40)

While the flexible refinement does improve the quality of the models, it is not sufficient in cases in which large conformational changes are required. We therefore investigated whether a limited sampling of glycan conformations prior to docking using the water refinement module of HADDOCK could generate conformations closer to the bound form. While we do observe slight improvements, their impact on the docking performance in an ensemble docking scenario is limited, thus highlighting the need for more extensive conformational sampling strategies, together with methods to identify the most relevant conformations. For example, the use of a database of glycan conformations such as GlycoShape (41) could help in improving HADDOCK’s performance in modeling protein–glycan complexes.

Data Availability

Click to copy section linkSection link copied!

The software used in this manuscript (HADDOCK3) is publicly available at https://github.com/haddocking/haddock3. All input data, analysis scripts, and results presented in the paper can be accessed at https://github.com/haddocking/protein-glycans. A tutorial describing the modeling of a protein–glycan complex using HADDOCK3 is hosted at https://bonvinlab.org/education/HADDOCK3/HADDOCK3-protein-glycan. The full runs, including docking models from all modules of a workflow, have been deposited in our lab collection (https://data.sbgrid.org/labs/32/1138) at the SBGRID data repository. (42)

Supporting Information

Click to copy section linkSection link copied!

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jcim.4c01372.

Data set used in this study (XLSX)
Details of the preparation of protein and glycan structures; description of HADDOCK3 modules used in this work; details of the glycan conformational sampling protocol; SNFG representation of the glycans; modules and parameters used for bound docking; modules and parameters used for unbound docking; glycan conformational sampling scenarios; example of HADDOCK models satisfying the quality thresholds; HADDOCK3’ performance on the bound data set; glycans’ RMSD to their bound conformations; impact of mdref on glycan conformations; impact of the clustering on glycans’ lowest RMSD; examples of glycan ensembles of conformations; HADDOCK3’ performance with the ensembles of glycans; torsion angle analysis of glycosidic linkages: a comparison to HADDOCK flexible refinement models; and torsion angle analysis of glycosidic linkages: a comparison to HADDOCK short molecular dynamics refinement models (PDF)

Terms & Conditions

Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.

Author Information

Click to copy section linkSection link copied!

Corresponding Author
- Alexandre M. J. J. Bonvin - Bijvoet Centre for Biomolecular Research, Faculty of Science - Chemistry, Utrecht University, Padualaan 8, Utrecht 3584CH, The Netherlands; https://orcid.org/0000-0001-7369-1322; Email: [email protected]
Authors
- Anna Ranaudo - Department of Earth and Environmental Sciences, University of Milano-Bicocca, Piazza Della Scienza 1, Milan 20126, Italy; Bijvoet Centre for Biomolecular Research, Faculty of Science - Chemistry, Utrecht University, Padualaan 8, Utrecht 3584CH, The Netherlands; https://orcid.org/0000-0002-8733-7938
- Marco Giulini - Bijvoet Centre for Biomolecular Research, Faculty of Science - Chemistry, Utrecht University, Padualaan 8, Utrecht 3584CH, The Netherlands
- Angela Pelissou Ayuso - Bijvoet Centre for Biomolecular Research, Faculty of Science - Chemistry, Utrecht University, Padualaan 8, Utrecht 3584CH, The Netherlands
Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.
Funding
This work was supported by grants from the European High Performance Computing Joint Undertaking project BioExcel (823,830) and The Netherlands e-Science Center (027.020.G13).
Notes
The authors declare no competing financial interest.

Abbreviations

Click to copy section linkSection link copied!

HADDOCK	high ambiguity-driven DOCKing
IL-RMSD	interface-ligand root mean squared deviation
SL	short-linear
SB	short-branched
LL	long-linear
LB	long-branched
AIRs	ambiguous interaction restraints
PDB	Protein Data Bank.

References

Click to copy section linkSection link copied!

This article references 42 other publications.

1
Gold, V. The IUPAC Compendium of Chemical Terminology; International Union of Pure and Applied Chemistry (IUPAC): Research Triangle Park, NC, 2019.
Google Scholar
There is no corresponding record for this reference.
2
Seeberger, P. H. Monosaccharide Diversity. In Essentials of Glycobiology; Varki, A., Cummings, R. D., Esko, J. D., Eds.; Cold Spring Harbor Laboratory Press: New York, 2022;,pp 21– 32. DOI: 10.1101/glycobiology.4e.2 .
Google Scholar
There is no corresponding record for this reference.
3
Lebrilla, C. B.; Liu, J.; Widmalm, G.; Prestegard, J. H. Oligosaccharides and Polysaccharides. In Essentials of Glycobiology; Varki, A., Cummings, R. D., Esko, J. D., Stanley, P., Hart, G. W., Aebi, M., Mohnen, D., Kinoshita, T., Packer, N. H., Prestegard, J. H., Schnaar, R. L., Seeberger, P. H., Eds.; Cold Spring Harbor Laboratory Press: New York, 2022, pp 33– 42. DOI: 10.1101/glycobiology.4e.3 .
Google Scholar
There is no corresponding record for this reference.
4
Varki, A.; Gagneux, P. Biological Functions of Glycans. In Essentials of Glycobiology; Varki, A., Cummings, R. D., Esko, J. D., Eds.; Cold Spring Harbor Laboratory Press: New York, 2022. DOI: 10.1101/glycobiology.4e.7 .
Google Scholar
There is no corresponding record for this reference.
5
Molina, A.; O’Neill, M. A.; Darvill, A. G.; Etzler, M. E.; Mohnen, D.; Hahn, M. G.; Esko, J. D. Free Glycans as Bioactive Molecules. In Essentials of Glycobiology; Varki, A., Cummings, R. D., Esko, J. D., Stanley, P., Hart, G. W., Aebi, M., Mohnen, D., Kinoshita, T., Packer, N. H., Prestegard, J. H., Schnaar, R. L., Seeberger, P. H., Eds.; Cold Spring Harbor Laboratory Press: New York, 2022, pp 539– 548. DOI: 10.1101/glycobiology.4e.40 .
Google Scholar
There is no corresponding record for this reference.
6
Casalino, L.; Gaieb, Z.; Goldsmith, J. A.; Hjorth, C. K.; Dommer, A. C.; Harbison, A. M.; Fogarty, C. A.; Barros, E. P.; Taylor, B. C.; McLellan, J. S.; Fadda, E.; Amaro, R. E. Beyond Shielding: The Roles of Glycans in the SARS-CoV-2 Spike Protein. ACS Cent. Sci. 2020, 6 (10), 1722– 1734, DOI: 10.1021/acscentsci.0c01056
Google Scholar
6
Beyond Shielding: The Roles of Glycans in the SARS-CoV-2 Spike Protein
Casalino, Lorenzo; Gaieb, Zied; Goldsmith, Jory A.; Hjorth, Christy K.; Dommer, Abigail C.; Harbison, Aoife M.; Fogarty, Carl A.; Barros, Emilia P.; Taylor, Bryn C.; McLellan, Jason S.; Fadda, Elisa; Amaro, Rommie E.
ACS Central Science (2020), 6 (10), 1722-1734CODEN: ACSCII; ISSN:2374-7951. (American Chemical Society)
The ongoing COVID-19 pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has resulted in more than 28,000,000 infections and 900,000 deaths worldwide to date. Antibody development efforts mainly revolve around the extensively glycosylated SARS-CoV-2 spike (S) protein, which mediates host cell entry by binding to the angiotensin-converting enzyme 2 (ACE2). Similar to many other viral fusion proteins, the SARS-CoV-2 spike utilizes a glycan shield to thwart the host immune response. Here, we built a full-length model of the glycosylated SARS-CoV-2 S protein, both in the open and closed states, augmenting the available structural and biol. data. Multiple microsecond-long, all-atom mol. dynamics simulations were used to provide an atomistic perspective on the roles of glycans and on the protein structure and dynamics. We reveal an essential structural role of N-glycans at sites N165 and N234 in modulating the conformational dynamics of the spike's receptor binding domain (RBD), which is responsible for ACE2 recognition. This finding is corroborated by biolayer interferometry expts., which show that deletion of these glycans through N165A and N234A mutations significantly reduces binding to ACE2 as a result of the RBD conformational shift toward the "down" state. Addnl., end-to-end accessibility analyses outline a complete overview of the vulnerabilities of the glycan shield of the SARS-CoV-2 S protein, which may be exploited in the therapeutic efforts targeting this mol. machine. Overall, this work presents hitherto unseen functional and structural insights into the SARS-CoV-2 S protein and its glycan coat, providing a strategy to control the conformational plasticity of the RBD that could be harnessed for vaccine development. The glycan shield is a sugary barrier that helps the viral SARS-CoV-2 spikes to evade the immune system. Beyond shielding, two of the spike's glycans are discovered to prime the virus for infection.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BB3cXhvVOlsb3N&md5=52d499afcd7e3caa7d9e6017ffa86e45
7
Woods, R. J. Predicting the Structures of Glycans, Glycoproteins, and Their Complexes. Chem. Rev. 2018, 118 (17), 8005– 8024, DOI: 10.1021/acs.chemrev.8b00032
Google Scholar
7
Predicting the Structures of Glycans, Glycoproteins, and Their Complexes
Woods, Robert J.
Chemical Reviews (Washington, DC, United States) (2018), 118 (17), 8005-8024CODEN: CHREAY; ISSN:0009-2665. (American Chemical Society)
A review. Complex carbohydrates are ubiquitous in nature, and together with proteins and nucleic acids they comprise the building blocks of life. But unlike proteins and nucleic acids, carbohydrates form nonlinear polymers, and they are not characterized by robust secondary or tertiary structures but rather by distributions of well-defined conformational states. Their mol. flexibility means that oligosaccharides are often refractory to crystn., and NMR spectroscopy augmented by mol. dynamics (MD) simulation is the leading method for their characterization in soln. The biol. importance of carbohydrate-protein interactions, in organismal development as well as in disease, places urgency on the creation of innovative exptl. and theor. methods that can predict the specificity of such interactions and quantify their strengths. Addnl., the emerging realization that protein glycosylation impacts protein function and immunogenicity places the ability to define the mechanisms by which glycosylation impacts these features at the forefront of carbohydrate modeling. This review will discuss the relevant theor. approaches to studying the three-dimensional structures of this fascinating class of mols. and interactions, with ref. to the relevant exptl. data and techniques that are key for validation of the theor. predictions.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC1cXhsVKhsbjF&md5=5b0fa8d3380516b34a09bf941e2853ce
8
Perez, S.; Makshakova, O. Multifaceted Computational Modeling in Glycoscience. Chem. Rev. 2022, 122 (20), 15914– 15970, DOI: 10.1021/acs.chemrev.2c00060
Google Scholar
8
Multifaceted Computational Modeling in Glycoscience
Perez, Serge; Makshakova, Olga
Chemical Reviews (Washington, DC, United States) (2022), 122 (20), 15914-15970CODEN: CHREAY; ISSN:0009-2665. (American Chemical Society)
A review. Glycoscience assembles all the scientific disciplines involved in studying various mols. and macromols. contg. carbohydrates and complex glycans. Such an ensemble involves one of the most extensive sets of mols. in quantity and occurrence since they occur in all microorganisms and higher organisms. Once the compns. and sequences of these mols. are established, the detn. of their three-dimensional structural and dynamical features is a step toward understanding the mol. basis underlying their properties and functions. The range of the relevant computational methods capable of addressing such issues is anchored by the specificity of stereoelectronic effects from quantum chem. to mesoscale modeling throughout mol. dynamics and mechanics and coarse-grained and docking calcns. The Review leads the reader through the detailed presentations of the applications of computational modeling. The illustrations cover carbohydrate-carbohydrate interactions, glycolipids, and N- and O-linked glycans, emphasizing their role in SARS-CoV-2. The presentation continues with the structure of polysaccharides in soln. and solid-state and lipopolysaccharides in membranes. The full range of protein-carbohydrate interactions is presented, as exemplified by carbohydrate-active enzymes, transporters, lectins, antibodies, and glycosaminoglycan binding proteins. A final section features a list of 150 tools and databases to help address the many issues of structural glycobioinformatics.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BB38Xhs1Ort7jK&md5=9ab27d5d4e828eb9b5d7760c6c55d60b
9
Nance, M. L.; Labonte, J. W.; Adolf-Bryfogle, J.; Gray, J. J. Development and Evaluation of GlycanDock: A Protein–Glycoligand Docking Refinement Algorithm in Rosetta. J. Phys. Chem. B 2021, 125 (25), 6807– 6820, DOI: 10.1021/acs.jpcb.1c00910
Google Scholar
There is no corresponding record for this reference.
10
Mottarella, S. E.; Beglov, D.; Beglova, N.; Nugent, M. A.; Kozakov, D.; Vajda, S. Docking Server for the Identification of Heparin Binding Sites on Proteins. J. Chem. Inf. Model. 2014, 54 (7), 2068– 2078, DOI: 10.1021/ci500115j
Google Scholar
10
Docking Server for the Identification of Heparin Binding Sites on Proteins
Mottarella, Scott E.; Beglov, Dmitri; Beglova, Natalia; Nugent, Matthew A.; Kozakov, Dima; Vajda, Sandor
Journal of Chemical Information and Modeling (2014), 54 (7), 2068-2078CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
Many proteins of widely differing functionality and structure are capable of binding heparin and heparan sulfate. Since crystg. protein-heparin complexes for structure detn. is generally difficult, computational docking can be a useful approach for understanding specific interactions. Previous studies used programs originally developed for docking small mols. to well-defined pockets, rather than for docking polysaccharides to highly charged shallow crevices that usually bind heparin. The authors have extended the program PIPER and the automated protein-protein docking server ClusPro to heparin docking. Using a mol. mechanics energy function for scoring and the fast Fourier transform correlation approach, the method generates and evaluates close to a billion poses of a heparin tetrasaccharide probe. The docked structures are clustered using pairwise root-mean-square deviations as the distance measure. Clustering of heparin mols. close to each other but having different orientations and selecting the clusters with the highest protein-ligand contacts reliably predicts the heparin binding site. In addn., the centers of the five most populated clusters include structures close to the native orientation of the heparin. These structures can provide starting points for further refinement by methods that account for flexibility such as mol. dynamics. The heparin docking method is available as an advanced option of the ClusPro server at http://cluspro.bu.edu/.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2cXhtVKnsLjK&md5=82dedc6c1e29961cbe8eb8a86957e582
11
Nivedha, A. K.; Thieker, D. F.; Makeneni, S.; Hu, H.; Woods, R. J. Vina-Carb: Improving Glycosidic Angles during Carbohydrate Docking. J. Chem. Theory Comput. 2016, 12 (2), 892– 901, DOI: 10.1021/acs.jctc.5b00834
Google Scholar
11
Vina-Carb: Improving Glycosidic Angles during Carbohydrate Docking
Nivedha, Anita K.; Thieker, David F.; Hu, Huimin; Woods, Robert J.
Journal of Chemical Theory and Computation (2016), 12 (2), 892-901CODEN: JCTCCE; ISSN:1549-9618. (American Chemical Society)
Mol. docking programs are primarily designed to align rigid, drug-like fragments into the binding sites of macromols. and frequently display poor performance when applied to flexible carbohydrate mols. A crit. source of flexibility within an oligosaccharide is the glycosidic linkages. Recently, Carbohydrate Intrinsic (CHI) energy functions are reported that attempt to quantify the glycosidic torsion angle preferences. The CHI-energy functions have been incorporated into the AutoDock Vina (ADV) scoring function, subsequently termed Vina-Carb (VC). Two user-adjustable parameters have been introduced, namely, a CHI- energy wt. term (chi_coeff) that affects the magnitude of the CHI-energy penalty and a CHI-cutoff term (chi_cutoff) that negates CHI-energy penalties below a specified value. A data set consisting of 101 protein-carbohydrate complexes and 29 apoprotein structures was used in the development and testing of VC, including antibodies, lectins, and carbohydrate binding modules. Accounting for the intramol. energies of the glycosidic linkages in the oligosaccharides during docking led VC to produce acceptable structures within the top five ranked poses in 74% of the systems tested, compared to a success rate of 55% for ADV. An enzyme system was employed to illustrate the potential application of VC to proteins that may distort glycosidic linkages of carbohydrate ligands upon binding. VC represents a significant step toward accurately predicting the structures of protein-carbohydrate complexes. Furthermore, the described approach is conceptually applicable to any class of ligands that populate well-defined conformational states.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XltlCrsQ%253D%253D&md5=cf018be5775bfaa2ad035ed1b82305b6
12
Boittier, E. D.; Burns, J. M.; Gandhi, N. S.; Ferro, V. GlycoTorch Vina: Docking Designed and Tested for Glycosaminoglycans. J. Chem. Inf. Model. 2020, 60 (12), 6328– 6343, DOI: 10.1021/acs.jcim.0c00373
Google Scholar
There is no corresponding record for this reference.
13
Labonte, J. W.; Adolf-Bryfogle, J.; Schief, W. R.; Gray, J. J. Residue-centric Modeling and Design of Saccharide and Glycoconjugate Structures. J. Comput. Chem. 2017, 38 (5), 276– 287, DOI: 10.1002/jcc.24679
Google Scholar
13
Residue-centric modeling and design of saccharide and glycoconjugate structures
Labonte, Jason W.; Adolf-Bryfogle, Jared; Schief, William R.; Gray, Jeffrey J.
Journal of Computational Chemistry (2017), 38 (5), 276-287CODEN: JCCHDD; ISSN:0192-8651. (John Wiley & Sons, Inc.)
The RosettaCarbohydrate framework is a new tool for modeling a wide variety of saccharide and glycoconjugate structures. This report describes the development of the framework and highlights its applications. The framework integrates with established protocols within the Rosetta modeling and design suite, and it handles the vast complexity and variety of carbohydrate mols., including branching and sugar modifications. To address challenges of sampling and scoring, RosettaCarbohydrate can sample glycosidic bonds, side-chain conformations, and ring forms, and it uses a glycan-specific term within its scoring function. Rosetta can work with std. PDB, GLYCAM, and GlycoWorkbench (.gws) file formats. Saccharide residue-specific chem. information is stored internally, permitting glycoengineering and design. Carbohydrate-specific applications described herein include virtual glycosylation, loop-modeling of carbohydrates, and docking of glyco-ligands to antibodies. Benchmarking data are presented and compared to other studies, demonstrating Rosetta's ability to predict glyco-ligand binding. The framework expands the tools available to glycoscientists and engineers.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XhvFKjsLnO&md5=822f98946c2d848b46f06b581beb11b4
14
Glashagen, G.; de Vries, S.; Uciechowska-Kaczmarzyk, U.; Samsonov, S. A.; Murail, S.; Tuffery, P.; Zacharias, M. Coarse-grained and Atomic Resolution Biomolecular Docking with the ATTRACT Approach. Proteins: Struct., Funct., Bioinf. 2020, 88 (8), 1018– 1028, DOI: 10.1002/prot.25860
Google Scholar
There is no corresponding record for this reference.
15
Jumper, J.; Evans, R.; Pritzel, A.; Green, T.; Figurnov, M.; Ronneberger, O.; Tunyasuvunakool, K.; Bates, R.; Žídek, A.; Potapenko, A.; Bridgland, A.; Meyer, C.; Kohl, S. A. A.; Ballard, A. J.; Cowie, A.; Romera-Paredes, B.; Nikolov, S.; Jain, R.; Adler, J.; Back, T.; Petersen, S.; Reiman, D.; Clancy, E.; Zielinski, M.; Steinegger, M.; Pacholska, M.; Berghammer, T.; Bodenstein, S.; Silver, D.; Vinyals, O.; Senior, A. W.; Kavukcuoglu, K.; Kohli, P.; Hassabis, D. Highly Accurate Protein Structure Prediction with AlphaFold. Nature 2021, 596 (7873), 583– 589, DOI: 10.1038/s41586-021-03819-2
Google Scholar
15
Highly accurate protein structure prediction with AlphaFold
Jumper, John; Evans, Richard; Pritzel, Alexander; Green, Tim; Figurnov, Michael; Ronneberger, Olaf; Tunyasuvunakool, Kathryn; Bates, Russ; Zidek, Augustin; Potapenko, Anna; Bridgland, Alex; Meyer, Clemens; Kohl, Simon A. A.; Ballard, Andrew J.; Cowie, Andrew; Romera-Paredes, Bernardino; Nikolov, Stanislav; Jain, Rishub; Adler, Jonas; Back, Trevor; Petersen, Stig; Reiman, David; Clancy, Ellen; Zielinski, Michal; Steinegger, Martin; Pacholska, Michalina; Berghammer, Tamas; Bodenstein, Sebastian; Silver, David; Vinyals, Oriol; Senior, Andrew W.; Kavukcuoglu, Koray; Kohli, Pushmeet; Hassabis, Demis
Nature (London, United Kingdom) (2021), 596 (7873), 583-589CODEN: NATUAS; ISSN:0028-0836. (Nature Portfolio)
Proteins are essential to life, and understanding their structure can facilitate a mechanistic understanding of their function. Through an enormous exptl. effort, the structures of around 100,000 unique proteins have been detd., but this represents a small fraction of the billions of known protein sequences. Structural coverage is bottlenecked by the months to years of painstaking effort required to det. a single protein structure. Accurate computational approaches are needed to address this gap and to enable large-scale structural bioinformatics. Predicting the three-dimensional structure that a protein will adopt based solely on its amino acid sequence-the structure prediction component of the 'protein folding problem'-has been an important open research problem for more than 50 years. Despite recent progress, existing methods fall far short of at. accuracy, esp. when no homologous structure is available. Here we provide the first computational method that can regularly predict protein structures with at. accuracy even in cases in which no similar structure is known. We validated an entirely redesigned version of our neural network-based model, AlphaFold, in the challenging 14th Crit. Assessment of protein Structure Prediction (CASP14), demonstrating accuracy competitive with exptl. structures in a majority of cases and greatly outperforming other methods. Underpinning the latest version of AlphaFold is a novel machine learning approach that incorporates phys. and biol. knowledge about protein structure, leveraging multi-sequence alignments, into the design of the deep learning algorithm.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BB3MXhvVaktrrL&md5=25964ab1157cd5b74a437333dd86650d
16
Abramson, J.; Adler, J.; Dunger, J.; Evans, R.; Green, T.; Pritzel, A.; Ronneberger, O.; Willmore, L.; Ballard, A. J.; Bambrick, J.; Bodenstein, S. W.; Evans, D. A.; Hung, C.-C.; O’Neill, M.; Reiman, D.; Tunyasuvunakool, K.; Wu, Z.; Žemgulytė, A.; Arvaniti, E.; Beattie, C.; Bertolli, O.; Bridgland, A.; Cherepanov, A.; Congreve, M.; Cowen-Rivers, A. I.; Cowie, A.; Figurnov, M.; Fuchs, F. B.; Gladman, H.; Jain, R.; Khan, Y. A.; Low, C. M. R.; Perlin, K.; Potapenko, A.; Savy, P.; Singh, S.; Stecula, A.; Thillaisundaram, A.; Tong, C.; Yakneen, S.; Zhong, E. D.; Zielinski, M.; Žídek, A.; Bapst, V.; Kohli, P.; Jaderberg, M.; Hassabis, D.; Jumper, J. M. Accurate Structure Prediction of Biomolecular Interactions with AlphaFold 3. Nature 2024, 630 (8016), 493– 500, DOI: 10.1038/s41586-024-07487-w
Google Scholar
There is no corresponding record for this reference.
17
HADDOCK3, Bonvin’s Lab , 2022. https://github.com/haddocking/haddock3.
Google Scholar
There is no corresponding record for this reference.
18
Dominguez, C.; Boelens, R.; Bonvin, A. M. J. J. HADDOCK: A Protein–Protein Docking Approach Based on Biochemical or Biophysical Information. J. Am. Chem. Soc. 2003, 125 (7), 1731– 1737, DOI: 10.1021/ja026939x
Google Scholar
18
HADDOCK: A protein-protein docking approach based on biochemical or biophysical information
Dominguez, Cyril; Boelens, Rolf; Bonvin, Alexandre M. J. J.
Journal of the American Chemical Society (2003), 125 (7), 1731-1737CODEN: JACSAT; ISSN:0002-7863. (American Chemical Society)
The structure detn. of protein-protein complexes is a rather tedious and lengthy process, by both NMR and x-ray crystallog. Several methods based on docking to study protein complexes have also been well developed over the past few years. Most of these approaches are not driven by exptl. data but are based on a combination of energetics and shape complementarity. Here, we present an approach called HADDOCK (High Ambiguity Driven protein-protein Docking) that makes use of biochem. and/or biophys. interaction data such as chem. shift perturbation data resulting from NMR titrn. expts. or mutagenesis data. This information is introduced as Ambiguous Interaction Restraints (AIRs) to drive the docking process. An AIR is defined as an ambiguous distance between all residues shown to be involved in the interaction. The accuracy of our approach is demonstrated with three mol. complexes. For two of these complexes, for which both the complex and the free protein structures have been solved, NMR titrn. data were available. Mutagenesis data were used in the last example. In all cases, the best structures generated by HADDOCK, i.e., the structures with the lowest intermol. energies, were the closest to the published structure of the resp. complexes (within 2.0 Å backbone RMSD).
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3sXkvFGquw%253D%253D&md5=9cf6fd9c690afb0181579485ae263197
19
Wu, A. M.; Singh, T.; Liu, J.-H.; Krzeminski, M.; Russwurm, R.; Siebert, H.-C.; Bonvin, A. M. J. J.; André, S.; Gabius, H.-J. Activity–Structure Correlations in Divergent Lectin Evolution: Fine Specificity of Chicken Galectin CG-14 and Computational Analysis of Flexible Ligand Docking for CG-14 and the Closely Related CG-16. Glycobiology 2007, 17 (2), 165– 184, DOI: 10.1093/glycob/cwl062
Google Scholar
There is no corresponding record for this reference.
20
Krzeminski, M.; Singh, T.; André, S.; Lensch, M.; Wu, A. M.; Bonvin, A. M. J. J.; Gabius, H.-J. Human Galectin-3 (Mac-2 Antigen): Defining Molecular Switches of Affinity to Natural Glycoproteins, Structural and Dynamic Aspects of Glycan Binding by Flexible Ligand Docking and Putative Regulatory Sequences in the Proximal Promoter Region. Biochim. Biophys. Acta, Gen. Subj. 2011, 1810 (2), 150– 161, DOI: 10.1016/j.bbagen.2010.11.001
Google Scholar
There is no corresponding record for this reference.
21
Berman, H. M. The Protein Data Bank. Nucleic Acids Res. 2000, 28 (1), 235– 242, DOI: 10.1093/nar/28.1.235
Google Scholar
21
The Protein Data Bank
Berman, Helen M.; Westbrook, John; Feng, Zukang; Gilliland, Gary; Bhat, T. N.; Weissig, Helge; Shindyalov, Ilya N.; Bourne, Philip E.
Nucleic Acids Research (2000), 28 (1), 235-242CODEN: NARHAD; ISSN:0305-1048. (Oxford University Press)
The Protein Data Bank (PDB; http://www.rcsb.org/pdb/)is the single worldwide archive of structural data of biol. macromols. This paper describes the goals of the PDB, the systems in place for data deposition and access, how to obtain further information, and near-term plans for the future development of the resource.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3cXhvVKjt7w%253D&md5=227fb393f754be2be375ab727bfd05dc
22
Woods Group. Complex Carbohydrate Research Center; GLYCAM Web: University of Georgia, Athens, GA, 2023.
Google Scholar
There is no corresponding record for this reference.
23
Kirschner, K. N.; Yongye, A. B.; Tschampel, S. M.; González-Outeiriño, J.; Daniels, C. R.; Foley, B. L.; Woods, R. J. GLYCAM06: A Generalizable Biomolecular Force Field. Carbohydrates. J. Comput. Chem. 2008, 29 (4), 622– 655, DOI: 10.1002/jcc.20820
Google Scholar
23
GLYCAM06: a generalizable biomolecular force field. Carbohydrates
Kirschner Karl N; Yongye Austin B; Tschampel Sarah M; Gonzalez-Outeirino Jorge; Daniels Charlisa R; Foley B Lachele; Woods Robert J
Journal of computational chemistry (2008), 29 (4), 622-55 ISSN:.
A new derivation of the GLYCAM06 force field, which removes its previous specificity for carbohydrates, and its dependency on the AMBER force field and parameters, is presented. All pertinent force field terms have been explicitly specified and so no default or generic parameters are employed. The new GLYCAM is no longer limited to any particular class of biomolecules, but is extendible to all molecular classes in the spirit of a small-molecule force field. The torsion terms in the present work were all derived from quantum mechanical data from a collection of minimal molecular fragments and related small molecules. For carbohydrates, there is now a single parameter set applicable to both alpha- and beta-anomers and to all monosaccharide ring sizes and conformations. We demonstrate that deriving dihedral parameters by fitting to QM data for internal rotational energy curves for representative small molecules generally leads to correct rotamer populations in molecular dynamics simulations, and that this approach removes the need for phase corrections in the dihedral terms. However, we note that there are cases where this approach is inadequate. Reported here are the basic components of the new force field as well as an illustration of its extension to carbohydrates. In addition to reproducing the gas-phase properties of an array of small test molecules, condensed-phase simulations employing GLYCAM06 are shown to reproduce rotamer populations for key small molecules and representative biopolymer building blocks in explicit water, as well as crystalline lattice properties, such as unit cell dimensions, and vibrational frequencies.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A280%3ADC%252BD1c%252FotlSmtQ%253D%253D&md5=fd32f08518d8e3568627c36eaa958efb
24
Varki, A.; Cummings, R. D.; Aebi, M.; Packer, N. H.; Seeberger, P. H.; Esko, J. D.; Stanley, P.; Hart, G.; Darvill, A.; Kinoshita, T.; Prestegard, J. J.; Schnaar, R. L.; Freeze, H. H.; Marth, J. D.; Bertozzi, C. R.; Etzler, M. E.; Frank, M.; Vliegenthart, J. F.; Lütteke, T.; Perez, S.; Bolton, E.; Rudd, P.; Paulson, J.; Kanehisa, M.; Toukach, P.; Aoki-Kinoshita, K. F.; Dell, A.; Narimatsu, H.; York, W.; Taniguchi, N.; Kornfeld, S. Symbol Nomenclature for Graphical Representations of Glycans. Glycobiology 2015, 25 (12), 1323– 1324, DOI: 10.1093/glycob/cwv091
Google Scholar
24
Symbol nomenclature for graphical representations of glycans
Varki, Ajit; Cummings, Richard D.; Aebi, Markus; Packer, Nicole H.; Seeberger, Peter H.; Esko, Jeffrey D.; Stanley, Pamela; Hart, Gerald; Darvill, Alan; Kinoshita, Taroh; Prestegard, James J.; Schnaar, Ronald L.; Freeze, Hudson H.; Marth, Jamey D.; Bertozzi, Carolyn R.; Etzler, Marilynn E.; Frank, Martin; Vliegenthart, Johannes F. G.; Lutteke, Thomas; Perez, Serge; Bolton, Evan; Rudd, Pauline; Paulson, James; Kanehisa, Minoru; Toukach, Philip; Aoki-Kinoshita, Kiyoko F.; Dell, Anne; Narimatsu, Hisashi; York, William; Taniguchi, Naoyuki; Kornfeld, Stuart
Glycobiology (2015), 25 (12), 1323-1324CODEN: GLYCE3; ISSN:0959-6658. (Oxford University Press)
Symbol nomenclature for graphical representations of glycans.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28Xhs1OqurzK&md5=6d906f3912a24468074ded519e6354c9
25
Neelamegham, S.; Aoki-Kinoshita, K.; Bolton, E.; Frank, M.; Lisacek, F.; Lütteke, T.; O’Boyle, N.; Packer, N. H.; Stanley, P.; Toukach, P.; Varki, A.; Woods, R. J. Updates to the Symbol Nomenclature for Glycans guidelines. Glycobiology 2019, 29 (9), 620– 624, DOI: 10.1093/glycob/cwz045
Google Scholar
25
Updates to the symbol nomenclature for glycans guidelines
Neelamegham, Sriram; Aoki-Kinoshita, Kiyoko; Bolton, Evan; Frank, Martin; Lisacek, Frederique; Lutteke, Thomas; O'Boyle, Noel; Packer, Nicolle H.; Stanley, Pamela; Toukach, Philip; Varki, Ajit; Woods, Robert J.; Darvill, Alan; Dell, Anne; Henrissat, Bernard; Bertozzi, Carolyn; Hart, Gerald; Narimatsu, Hisashi; Freeze, Hudson; Yamada, Issaku; Paulson, James; Prestegard, James; Marth, Jamey; Vliegenthart, Jfg; Etzler, Marilynn; Aebi, Markus; Kanehisa, Minoru; Taniguchi, Naoyuki; Edwards, Nathan; Rudd, Pauline; Seeberger, Peter; Mazumder, Raja; Ranzinger, Rene; Cummings, Richard; Schnaar, Ronald; Perez, Serge; Kornfeld, Stuart; Kinoshita, Taroh; York, William; Knirel, Yuriy
Glycobiology (2019), 29 (9), 620-624CODEN: GLYCE3; ISSN:1460-2423. (Oxford University Press)
A review. The Symbol Nomenclature for Glycans (SNFG) is a community-curated std. for the depiction of monosaccharides and complex glycans using various colored-coded, geometric shapes, along with defined text addns. It is hosted by the National Center for Biotechnol. Information (NCBI) at the NCBI-Glycans Page. Several changes have been made to the SNFG page in the past year to update the rules for depicting glycans using the SNFG, to include more examples of use, particularly for non-mammalian organisms, and to provide guidelines for the depiction of ambiguous glycan structures. This Glycoforum article summarizes these recent changes.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BB3cXps1Witrk%253D&md5=b7bb24d9cf3e0b8ff3edb88960186d77
26
de Vries, S. J.; van Dijk, A. D. J.; Krzeminski, M.; van Dijk, M.; Thureau, A.; Hsu, V.; Wassenaar, T.; Bonvin, A. M. J. J. HADDOCK versus HADDOCK: New Features and Performance of HADDOCK2.0 on the CAPRI Targets. Proteins: Struct., Funct., Bioinf. 2007, 69 (4), 726– 733, DOI: 10.1002/prot.21723
Google Scholar
There is no corresponding record for this reference.
27
Jorgensen, W. L.; Tirado-Rives, J. The OPLS [Optimized Potentials for Liquid Simulations] Potential Functions for Proteins, Energy Minimizations for Crystals of Cyclic Peptides and Crambin. J. Am. Chem. Soc. 1988, 110 (6), 1657– 1666, DOI: 10.1021/ja00214a001
Google Scholar
27
The OPLS [optimized potentials for liquid simulations] potential functions for proteins, energy minimizations for crystals of cyclic peptides and crambin
Jorgensen, William L.; Tirado-Rives, Julian
Journal of the American Chemical Society (1988), 110 (6), 1657-66CODEN: JACSAT; ISSN:0002-7863.
A complete set of intermol. potential functions was developed for use in computer simulations of proteins in their native environment. Parameters are reported for 25 peptide residues as well as the common neutral and charged terminal groups. The potential functions have the simple Coulomb plus Lennard-Jones form and are compatible with the widely used models for water, TIP4P, TIP3P, and SPC. The parameters were obtained and tested primarily in conjunction with Monte Carlo statistical mechanics simulations of 36 pure org. liqs. and numerous aq. solns. of org. ions representative of subunits in the side chains and backbones of proteins. Bond stretch, angle bend, and torsional terms were adopted from the AMBER united-atom force field. As reported here, further testing involved studies of conformational energy surfaces and optimizations of the crystal structures for 4 cyclic hexapeptides and a cyclic pentapeptide. The av. root mean square deviation from the x-ray structures of the crystals is only 0.17 Å for the at. positions and 3% for the unit cell vols. A more crit. test was then provided by performing energy minimizations for the complete crystal of the protein crambin, including 182 water mols. that were initially placed via a Monte Carlo simulation. The resultant root mean square deviation for the non-H atoms is still only 0.17 Å and the variation in the errors for charged, polar, and nonpolar residues is small. Significant improvement is apparent over the AMBER united-atom force field which was previously demonstrated to be superior to many alternatives.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaL1cXht1yjt7Y%253D&md5=b6c901d8c295b3b37329a7faef527e12
28
Fernández-Recio, J.; Totrov, M.; Abagyan, R. Identification of Protein–Protein Interaction Sites from Docking Energy Landscapes. J. Mol. Biol. 2004, 335 (3), 843– 865, DOI: 10.1016/j.jmb.2003.10.069
Google Scholar
28
Identification of Protein-Protein Interaction Sites from Docking Energy Landscapes
Fernandez-Recio, Juan; Totrov, Maxim; Abagyan, Ruben
Journal of Molecular Biology (2004), 335 (3), 843-865CODEN: JMOBAK; ISSN:0022-2836. (Elsevier)
Protein recognition is one of the most challenging and intriguing problems in structural biol. Despite all the available structural, sequence and biophys. information about protein-protein complexes, the physico-chem. patterns, if any, that make a protein surface likely to be involved in protein-protein interactions, remain elusive. Here, we apply protein docking simulations and anal. of the interaction energy landscapes to identify protein-protein interaction sites. The new protocol for global docking based on multi-start global energy optimization of an all-atom model of the ligand, with detailed receptor potentials and at. solvation parameters optimized in a training set of 24 complexes, explores the conformational space around the whole receptor without restrictions. The ensembles of the rigid-body docking solns. generated by the simulations were subsequently used to project the docking energy landscapes onto the protein surfaces. We found that highly populated low-energy regions consistently corresponded to actual binding sites. The procedure was validated on a test set of 21 known protein-protein complexes not used in the training set. As much as 81% of the predicted high-propensity patch residues were located correctly in the native interfaces. This approach can guide the design of mutations on the surfaces of proteins, provide geometrical details of a possible interaction, and help to annotate protein surfaces in structural proteomics.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3sXpvVWltLg%253D&md5=8a7687b9ecf2f39625e0511d3ec71cc5
29
Basciu, A.; Koukos, P. I.; Malloci, G.; Bonvin, A. M. J. J.; Vargiu, A. V. Coupling Enhanced Sampling of the Apo-Receptor with Template-Based Ligand Conformers Selection: Performance in Pose Prediction in the D3R Grand Challenge 4. J. Comput. Aided Mol. Des 2020, 34 (2), 149– 162, DOI: 10.1007/s10822-019-00244-6
Google Scholar
29
Coupling enhanced sampling of the apo-receptor with template-based ligand conformers selection: performance in pose prediction in the D3R Grand Challenge 4
Basciu, Andrea; Koukos, Panagiotis I.; Malloci, Giuliano; Bonvin, Alexandre M. J. J.; Vargiu, Attilio V.
Journal of Computer-Aided Molecular Design (2020), 34 (2), 149-162CODEN: JCADEQ; ISSN:0920-654X. (Springer)
We report the performance of our newly introduced Ensemble Docking with Enhanced sampling of pocket Shape (EDES) protocol coupled to a template-based algorithm to generate near-native ligand conformations in the 2019 iteration of the Grand Challenge (GC4) organized by the D3R consortium. Using either AutoDock4.2 or HADDOCK2.2 docking programs (each software in two variants of the protocol) our method generated native-like poses among the top 5 submitted for evaluation for most of the 20 targets with similar performances. The protein selected for GC4 was the human beta-site amyloid precursor protein cleaving enzyme 1 (BACE-1), a transmembrane aspartic-acid protease. We identified at least one pose whose heavy-atoms RMSD was less than 2.5 Å from the native conformation for 16 (80%) and 17 (85%) of the 20 targets using AutoDock and HADDOCK, resp. Dissecting the possible sources of errors revealed that: (i) our EDES protocol (with minor modifications) was able to sample sub-angstrom conformations for all 20 protein targets, reproducing the correct conformation of the binding site within ∼ 1 Å RMSD; (ii) as already shown by some of us in GC3, even in the presence of near-native protein structures, a proper selection of ligand conformers is crucial for the success of ensemble-docking calcns. Importantly, our approach performed best among the protocols exploiting only structural information of the apo protein to generate conformations of the receptor for ensemble-docking calcns.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC1MXitFGit73J&md5=af95398263b2f4c630dc50847867a518
30
Nilges, M. A calculation strategy for the structure determination of symmetric demers by ¹H NMR. Proteins: Struct., Funct., Bioinf. 1993, 17 (3), 297– 309, DOI: 10.1002/prot.340170307
Google Scholar
There is no corresponding record for this reference.
31
Wallace, A. C.; Laskowski, R. A.; Thornton, J. M. LIGPLOT: A Program to Generate Schematic Diagrams of Protein-Ligand Interactions. Protein Eng., Des. Sel. 1995, 8 (2), 127– 134, DOI: 10.1093/protein/8.2.127
Google Scholar
There is no corresponding record for this reference.
32
McDonald, I. K.; Thornton, J. M. Satisfying Hydrogen Bonding Potential in Proteins. J. Mol. Biol. 1994, 238 (5), 777– 793, DOI: 10.1006/jmbi.1994.1334
Google Scholar
32
Satisfying hydrogen bonding potential in proteins
McDonald, Ian K.; Thornton, Janet M.
Journal of Molecular Biology (1994), 238 (5), 777-93CODEN: JMOBAK; ISSN:0022-2836.
The authors have analyzed the frequency with which potential hydrogen bond donors and acceptors are satisfied in protein mols. There are a small percentage of nitrogen or oxygen atoms that do not form hydrogen bonds with either solvent or protein atoms, when std. criteria are used. For high resoln. structures, 9.5% and 5.1% of buried main-chain nitrogen and oxygen atoms, resp., fail to hydrogen bond under the authors' std. criteria, representing 5.8% and 2.1% of all main-chain nitrogen and oxygen atoms. The authors find that as the resoln. of the data improves, the percentages fall. If the hydrogen bond criteria are relaxed many of these unsatisfied atoms form weak hydrogen bonds. However, there remain some buried atoms (1.3% NH and 1.8% CO) that fail to hydrogen bond without any immediately obvious compensating interactions.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaK2cXltVagt7k%253D&md5=0f21bf6f6b1ae5f5f70235da81f426b5
33
Méndez, R.; Leplae, R.; De Maria, L.; Wodak, S. J. Assessment of Blind Predictions of Protein–Protein Interactions: Current Status of Docking Methods. Proteins: Struct., Funct., Bioinf. 2003, 52 (1), 51– 67, DOI: 10.1002/prot.10393
Google Scholar
There is no corresponding record for this reference.
34
Giulini, M.; Menichetti, R.; Shell, M. S.; Potestio, R. An Information-Theory-Based Approach for Optimal Model Reduction of Biomolecules. J. Chem. Theory Comput. 2020, 16 (11), 6795– 6813, DOI: 10.1021/acs.jctc.0c00676
Google Scholar
34
An Information-Theory-Based Approach for Optimal Model Reduction of Biomolecules
Giulini, Marco; Menichetti, Roberto; Shell, M. Scott; Potestio, Raffaello
Journal of Chemical Theory and Computation (2020), 16 (11), 6795-6813CODEN: JCTCCE; ISSN:1549-9618. (American Chemical Society)
In theor. modeling of a phys. system, a crucial step consists of the identification of those degrees of freedom that enable a synthetic yet informative representation of it. While in some cases this selection can be carried out on the basis of intuition and experience, straightforward discrimination of the important features from the negligible ones is difficult for many complex systems, most notably heteropolymers and large biomols. The authors here present a thermodn.-based theor. framework to gauge the effectiveness of a given simplified representation by measuring its information content. The authors employ this method to identify those reduced descriptions of proteins, in terms of a subset of their atoms, that retain the largest amt. of information from the original model; these highly informative representations share common features that are intrinsically related to the biol. properties of the proteins under examn., thereby establishing a bridge between protein structure, energetics, and function.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BB3cXit1alt73P&md5=03a82b725a5399cc9502639f62485100
35
Sokal, R. R.; Michener, C. D. A Statistical Method for Evaluating Systematic Relationships; University of Kansas Science Bulletin, 1958; Vol. 38, pp 1409– 1438.
Google Scholar
There is no corresponding record for this reference.
36
Charitou, V.; van Keulen, S. C.; Bonvin, A. M. J. J. Cyclization and Docking Protocol for Cyclic Peptide–Protein Modeling Using HADDOCK2.4. J. Chem. Theory Comput. 2022, 18 (6), 4027– 4040, DOI: 10.1021/acs.jctc.2c00075
Google Scholar
There is no corresponding record for this reference.
37
Buchanan, C. J.; Gaunt, B.; Harrison, P. J.; Yang, Y.; Liu, J.; Khan, A.; Giltrap, A. M.; Le Bas, A.; Ward, P. N.; Gupta, K.; Dumoux, M.; Tan, T. K.; Schimaski, L.; Daga, S.; Picchiotti, N.; Baldassarri, M.; Benetti, E.; Fallerini, C.; Fava, F.; Giliberti, A.; Koukos, P. I.; Davy, M. J.; Lakshminarayanan, A.; Xue, X.; Papadakis, G.; Deimel, L. P.; Casablancas-Antràs, V.; Claridge, T. D. W.; Bonvin, A. M. J. J.; Sattentau, Q. J.; Furini, S.; Gori, M.; Huo, J.; Owens, R. J.; Schaffitzel, C.; Berger, I.; Renieri, A.; Naismith, J. H.; Baldwin, A. J.; Davis, B. G.; Davis, B. G. Pathogen-Sugar Interactions Revealed by Universal Saturation Transfer Analysis. Science 2022, 377 (6604), eabm3125 DOI: 10.1126/science.abm3125
Google Scholar
There is no corresponding record for this reference.
38
Koukos, P. I.; Réau, M.; Bonvin, A. M. J. J. Shape-Restrained Modeling of Protein–Small-Molecule Complexes with High Ambiguity Driven DOCKing. J. Chem. Inf. Model. 2021, 61 (9), 4807– 4818, DOI: 10.1021/acs.jcim.1c00796
Google Scholar
There is no corresponding record for this reference.
39
Kerzmann, A.; Neumann, D.; Kohlbacher, O. SLICK – Scoring and Energy Functions for Protein–Carbohydrate Interactions. J. Chem. Inf. Model. 2006, 46 (4), 1635– 1642, DOI: 10.1021/ci050422y
Google Scholar
39
SLICK - Scoring and Energy Functions for Protein-Carbohydrate Interactions
Kerzmann, Andreas; Neumann, Dirk; Kohlbacher, Oliver
Journal of Chemical Information and Modeling (2006), 46 (4), 1635-1642CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
Protein-carbohydrate interactions are increasingly being recognized as essential for many important biomol. recognition processes. From these, numerous biomedical applications arise in areas as diverse as drug design, immunol., or drug transport. The authors introduce SLICK, a package contg. a scoring and an energy function, which were specifically designed to predict binding modes and free energies of sugars and sugarlike compds. to proteins. SLICK accounts for van der Waals interactions, solvation effects, electrostatics, hydrogen bonds, and CH···π interactions, the latter being a particular feature of most protein-carbohydrate interactions. Parameters for the empirical energy function were calibrated on a set of high-resoln. crystal structures of protein-sugar complexes with known exptl. binding free energies. The authors show that SLICK predicts the binding free energies of predicted complexes (through mol. docking) with high accuracy. SLICK is available as part of the authors' mol. modeling package BALL (www.ball-project.org).
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD28Xksl2ksbY%253D&md5=af9f4a1e22709fcf8e39e6595b11a501
40
Kerzmann, A.; Fuhrmann, J.; Kohlbacher, O.; Neumann, D. BALLDock/SLICK: A New Method for Protein-Carbohydrate Docking. J. Chem. Inf. Model. 2008, 48 (8), 1616– 1625, DOI: 10.1021/ci800103u
Google Scholar
40
BALLDock/SLICK: A New Method for Protein-Carbohydrate Docking
Kerzmann, Andreas; Fuhrmann, Jan; Kohlbacher, Oliver; Neumann, Dirk
Journal of Chemical Information and Modeling (2008), 48 (8), 1616-1625CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
Protein-ligand docking is an essential technique in computer-aided drug design. While generally available docking programs work well for most drug classes, carbohydrates and carbohydrate-like compds. are often problematic for docking. The authors present a new docking method specifically designed to handle docking of carbohydrate-like compds. BALLDock/SLICK combines an evolutionary docking algorithm for flexible ligands and flexible receptor side chains with carbohydrate-specific scoring and energy functions. The scoring function has been designed to identify accurate ligand poses, while the energy function yields accurate ests. of the binding free energies of these poses. On a test set of known protein-sugar complexes the authors demonstrate the ability of the approach to generate correct poses for almost all of the structures and achieve very low mean errors for the predicted binding free energies.
>> More from SciFinder ^®
https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1cXovVWitb8%253D&md5=e34cd7fe7693e397ad796306b8479314
41
Ives, C.; Singh, O.; D’Andrea, S.; Fogarty, C.; Harbison, A.; Satheesan, A.; Tropea, B.; Fadda, E. Restoring Protein Glycosylation with GlycoShape. bioRxiv 2023, DOI: 10.1101/2023.12.11.571101
Google Scholar
There is no corresponding record for this reference.
42
Meyer, P. A.; Socias, S.; Key, J.; Ransey, E.; Tjon, E. C.; Buschiazzo, A.; Lei, M.; Botka, C.; Withrow, J.; Neau, D.; Rajashankar, K.; Anderson, K. S.; Baxter, R. H.; Blacklow, S. C.; Boggon, T. J.; Bonvin, A. M. J. J.; Borek, D.; Brett, T. J.; Caflisch, A.; Chang, C.-I.; Chazin, W. J.; Corbett, K. D.; Cosgrove, M. S.; Crosson, S.; Dhe-Paganon, S.; Di Cera, E.; Drennan, C. L.; Eck, M. J.; Eichman, B. F.; Fan, Q. R.; Ferré-D’Amaré, A. R.; Christopher Fromme, J.; Garcia, K. C.; Gaudet, R.; Gong, P.; Harrison, S. C.; Heldwein, E. E.; Jia, Z.; Keenan, R. J.; Kruse, A. C.; Kvansakul, M.; McLellan, J. S.; Modis, Y.; Nam, Y.; Otwinowski, Z.; Pai, E. F.; Pereira, P. J. B.; Petosa, C.; Raman, C. S.; Rapoport, T. A.; Roll-Mecak, A.; Rosen, M. K.; Rudenko, G.; Schlessinger, J.; Schwartz, T. U.; Shamoo, Y.; Sondermann, H.; Tao, Y. J.; Tolia, N. H.; Tsodikov, O. V.; Westover, K. D.; Wu, H.; Foster, I.; Fraser, J. S.; Maia, F. R. N. C.; Gonen, T.; Kirchhausen, T.; Diederichs, K.; Crosas, M.; Sliz, P. Data Publication with the Structural Biology Data Grid Supports Live Analysis. Nat. Commun. 2016, 7 (1), 10882, DOI: 10.1038/ncomms10882
Google Scholar
There is no corresponding record for this reference.

Cited By

Click to copy section linkSection link copied!

This article has not yet been cited by other publications.

Download PDF

Get e-Alerts

Journal of Chemical Information and Modeling

Cite this: J. Chem. Inf. Model. 2024, 64, 19, 7816–7825

Click to copy citationCitation copied!

https://doi.org/10.1021/acs.jcim.4c01372

Published October 3, 2024

CC-BY 4.0 .

Article Views

Altmetric

Citations

Learn about these metrics

Article Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.

Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.

The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated.

Abstract
High Resolution Image
Download MS PowerPoint Slide
Figure 1
Figure 1. Schematic representation of the docking protocol for the unbound data set. First, rigid-body docking is performed. The models are then clustered based on RMSD. The best-scoring models of each cluster are then subjected to a flexible refinement (interface), and the resulting models are again clustered and analyzed.
High Resolution Image
Download MS PowerPoint Slide
Figure 2
Figure 2. Comparison of bound docking success rates obtained with the default (w_vdW = 0.01) and vdW (w_vdW = 1.0) scoring functions (eqs 1 and 2) as a function of the number of top-ranked models (T = 1, 5, 10, 50, 100, and 200) selected using true interface residues of both protein and glycan to define the ambiguous interaction restraints (ti-aa AIRs).
High Resolution Image
Download MS PowerPoint Slide
Figure 3
Figure 3. HADDOCK3’s performance on the unbound data set using the vdW scoring function and tip-ap AIRs (true interface on the protein defined as active and the glycan residues as passive). The success rates (SRs) (Y axis), defined as the percentage of complexes for which acceptable-, medium-, or high-quality models are generated, are calculated on the top 1 (T1) to top 200 (T200) ranked rigid-body models (column “rigid”), T1 to T50 rigid-body clusters, considering the top 5 models of each cluster (column “rigid + clustering”), the T1 to T200 ranked refined models (column “flexref”), and the T1 to T10 refined clusters, considering the top 4 models of each cluster (column “flexref + clustering”). SRs are shown separately for the whole data set (first row) and for the three categories of complexes grouped by glycan size and connectivity: SL-SB (second row), LL (third row), and LB (fourth row).
High Resolution Image
Download MS PowerPoint Slide
Figure 4
Figure 4. Superimposition of the best-scoring flexible refinement models (orange) and the rigid-body models (teal) to the reference structures (gray) for the complexes 1OH4 (LB), 5VX5 (LL), and 1C1L (SL) and the unbound docking scenario carried out with vdW scoring potential and tip-ap AIRs. Oxygen atoms of the glycans are shown in red in all the structures, nitrogen atoms in blue, and hydrogens are not shown. Ranking and IL-RMSD values with respect to the reference structures for the flexref and rigid-body models are shown as well.
High Resolution Image
Download MS PowerPoint Slide
References
This article references 42 other publications.
1. 1
  Gold, V. The IUPAC Compendium of Chemical Terminology; International Union of Pure and Applied Chemistry (IUPAC): Research Triangle Park, NC, 2019.
  There is no corresponding record for this reference.
2. 2
  Seeberger, P. H. Monosaccharide Diversity. In Essentials of Glycobiology; Varki, A., Cummings, R. D., Esko, J. D., Eds.; Cold Spring Harbor Laboratory Press: New York, 2022;,pp 21– 32. DOI: 10.1101/glycobiology.4e.2 .
  There is no corresponding record for this reference.
3. 3
  Lebrilla, C. B.; Liu, J.; Widmalm, G.; Prestegard, J. H. Oligosaccharides and Polysaccharides. In Essentials of Glycobiology; Varki, A., Cummings, R. D., Esko, J. D., Stanley, P., Hart, G. W., Aebi, M., Mohnen, D., Kinoshita, T., Packer, N. H., Prestegard, J. H., Schnaar, R. L., Seeberger, P. H., Eds.; Cold Spring Harbor Laboratory Press: New York, 2022, pp 33– 42. DOI: 10.1101/glycobiology.4e.3 .
  There is no corresponding record for this reference.
4. 4
  Varki, A.; Gagneux, P. Biological Functions of Glycans. In Essentials of Glycobiology; Varki, A., Cummings, R. D., Esko, J. D., Eds.; Cold Spring Harbor Laboratory Press: New York, 2022. DOI: 10.1101/glycobiology.4e.7 .
  There is no corresponding record for this reference.
5. 5
  Molina, A.; O’Neill, M. A.; Darvill, A. G.; Etzler, M. E.; Mohnen, D.; Hahn, M. G.; Esko, J. D. Free Glycans as Bioactive Molecules. In Essentials of Glycobiology; Varki, A., Cummings, R. D., Esko, J. D., Stanley, P., Hart, G. W., Aebi, M., Mohnen, D., Kinoshita, T., Packer, N. H., Prestegard, J. H., Schnaar, R. L., Seeberger, P. H., Eds.; Cold Spring Harbor Laboratory Press: New York, 2022, pp 539– 548. DOI: 10.1101/glycobiology.4e.40 .
  There is no corresponding record for this reference.
6. 6
  Casalino, L.; Gaieb, Z.; Goldsmith, J. A.; Hjorth, C. K.; Dommer, A. C.; Harbison, A. M.; Fogarty, C. A.; Barros, E. P.; Taylor, B. C.; McLellan, J. S.; Fadda, E.; Amaro, R. E. Beyond Shielding: The Roles of Glycans in the SARS-CoV-2 Spike Protein. ACS Cent. Sci. 2020, 6 (10), 1722– 1734, DOI: 10.1021/acscentsci.0c01056
  6
  Beyond Shielding: The Roles of Glycans in the SARS-CoV-2 Spike Protein
  Casalino, Lorenzo; Gaieb, Zied; Goldsmith, Jory A.; Hjorth, Christy K.; Dommer, Abigail C.; Harbison, Aoife M.; Fogarty, Carl A.; Barros, Emilia P.; Taylor, Bryn C.; McLellan, Jason S.; Fadda, Elisa; Amaro, Rommie E.
  ACS Central Science (2020), 6 (10), 1722-1734CODEN: ACSCII; ISSN:2374-7951. (American Chemical Society)
  The ongoing COVID-19 pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has resulted in more than 28,000,000 infections and 900,000 deaths worldwide to date. Antibody development efforts mainly revolve around the extensively glycosylated SARS-CoV-2 spike (S) protein, which mediates host cell entry by binding to the angiotensin-converting enzyme 2 (ACE2). Similar to many other viral fusion proteins, the SARS-CoV-2 spike utilizes a glycan shield to thwart the host immune response. Here, we built a full-length model of the glycosylated SARS-CoV-2 S protein, both in the open and closed states, augmenting the available structural and biol. data. Multiple microsecond-long, all-atom mol. dynamics simulations were used to provide an atomistic perspective on the roles of glycans and on the protein structure and dynamics. We reveal an essential structural role of N-glycans at sites N165 and N234 in modulating the conformational dynamics of the spike's receptor binding domain (RBD), which is responsible for ACE2 recognition. This finding is corroborated by biolayer interferometry expts., which show that deletion of these glycans through N165A and N234A mutations significantly reduces binding to ACE2 as a result of the RBD conformational shift toward the "down" state. Addnl., end-to-end accessibility analyses outline a complete overview of the vulnerabilities of the glycan shield of the SARS-CoV-2 S protein, which may be exploited in the therapeutic efforts targeting this mol. machine. Overall, this work presents hitherto unseen functional and structural insights into the SARS-CoV-2 S protein and its glycan coat, providing a strategy to control the conformational plasticity of the RBD that could be harnessed for vaccine development. The glycan shield is a sugary barrier that helps the viral SARS-CoV-2 spikes to evade the immune system. Beyond shielding, two of the spike's glycans are discovered to prime the virus for infection.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BB3cXhvVOlsb3N&md5=52d499afcd7e3caa7d9e6017ffa86e45
7. 7
  Woods, R. J. Predicting the Structures of Glycans, Glycoproteins, and Their Complexes. Chem. Rev. 2018, 118 (17), 8005– 8024, DOI: 10.1021/acs.chemrev.8b00032
  7
  Predicting the Structures of Glycans, Glycoproteins, and Their Complexes
  Woods, Robert J.
  Chemical Reviews (Washington, DC, United States) (2018), 118 (17), 8005-8024CODEN: CHREAY; ISSN:0009-2665. (American Chemical Society)
  A review. Complex carbohydrates are ubiquitous in nature, and together with proteins and nucleic acids they comprise the building blocks of life. But unlike proteins and nucleic acids, carbohydrates form nonlinear polymers, and they are not characterized by robust secondary or tertiary structures but rather by distributions of well-defined conformational states. Their mol. flexibility means that oligosaccharides are often refractory to crystn., and NMR spectroscopy augmented by mol. dynamics (MD) simulation is the leading method for their characterization in soln. The biol. importance of carbohydrate-protein interactions, in organismal development as well as in disease, places urgency on the creation of innovative exptl. and theor. methods that can predict the specificity of such interactions and quantify their strengths. Addnl., the emerging realization that protein glycosylation impacts protein function and immunogenicity places the ability to define the mechanisms by which glycosylation impacts these features at the forefront of carbohydrate modeling. This review will discuss the relevant theor. approaches to studying the three-dimensional structures of this fascinating class of mols. and interactions, with ref. to the relevant exptl. data and techniques that are key for validation of the theor. predictions.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC1cXhsVKhsbjF&md5=5b0fa8d3380516b34a09bf941e2853ce
8. 8
  Perez, S.; Makshakova, O. Multifaceted Computational Modeling in Glycoscience. Chem. Rev. 2022, 122 (20), 15914– 15970, DOI: 10.1021/acs.chemrev.2c00060
  8
  Multifaceted Computational Modeling in Glycoscience
  Perez, Serge; Makshakova, Olga
  Chemical Reviews (Washington, DC, United States) (2022), 122 (20), 15914-15970CODEN: CHREAY; ISSN:0009-2665. (American Chemical Society)
  A review. Glycoscience assembles all the scientific disciplines involved in studying various mols. and macromols. contg. carbohydrates and complex glycans. Such an ensemble involves one of the most extensive sets of mols. in quantity and occurrence since they occur in all microorganisms and higher organisms. Once the compns. and sequences of these mols. are established, the detn. of their three-dimensional structural and dynamical features is a step toward understanding the mol. basis underlying their properties and functions. The range of the relevant computational methods capable of addressing such issues is anchored by the specificity of stereoelectronic effects from quantum chem. to mesoscale modeling throughout mol. dynamics and mechanics and coarse-grained and docking calcns. The Review leads the reader through the detailed presentations of the applications of computational modeling. The illustrations cover carbohydrate-carbohydrate interactions, glycolipids, and N- and O-linked glycans, emphasizing their role in SARS-CoV-2. The presentation continues with the structure of polysaccharides in soln. and solid-state and lipopolysaccharides in membranes. The full range of protein-carbohydrate interactions is presented, as exemplified by carbohydrate-active enzymes, transporters, lectins, antibodies, and glycosaminoglycan binding proteins. A final section features a list of 150 tools and databases to help address the many issues of structural glycobioinformatics.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BB38Xhs1Ort7jK&md5=9ab27d5d4e828eb9b5d7760c6c55d60b
9. 9
  Nance, M. L.; Labonte, J. W.; Adolf-Bryfogle, J.; Gray, J. J. Development and Evaluation of GlycanDock: A Protein–Glycoligand Docking Refinement Algorithm in Rosetta. J. Phys. Chem. B 2021, 125 (25), 6807– 6820, DOI: 10.1021/acs.jpcb.1c00910
  There is no corresponding record for this reference.
10. 10
  Mottarella, S. E.; Beglov, D.; Beglova, N.; Nugent, M. A.; Kozakov, D.; Vajda, S. Docking Server for the Identification of Heparin Binding Sites on Proteins. J. Chem. Inf. Model. 2014, 54 (7), 2068– 2078, DOI: 10.1021/ci500115j
  10
  Docking Server for the Identification of Heparin Binding Sites on Proteins
  Mottarella, Scott E.; Beglov, Dmitri; Beglova, Natalia; Nugent, Matthew A.; Kozakov, Dima; Vajda, Sandor
  Journal of Chemical Information and Modeling (2014), 54 (7), 2068-2078CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
  Many proteins of widely differing functionality and structure are capable of binding heparin and heparan sulfate. Since crystg. protein-heparin complexes for structure detn. is generally difficult, computational docking can be a useful approach for understanding specific interactions. Previous studies used programs originally developed for docking small mols. to well-defined pockets, rather than for docking polysaccharides to highly charged shallow crevices that usually bind heparin. The authors have extended the program PIPER and the automated protein-protein docking server ClusPro to heparin docking. Using a mol. mechanics energy function for scoring and the fast Fourier transform correlation approach, the method generates and evaluates close to a billion poses of a heparin tetrasaccharide probe. The docked structures are clustered using pairwise root-mean-square deviations as the distance measure. Clustering of heparin mols. close to each other but having different orientations and selecting the clusters with the highest protein-ligand contacts reliably predicts the heparin binding site. In addn., the centers of the five most populated clusters include structures close to the native orientation of the heparin. These structures can provide starting points for further refinement by methods that account for flexibility such as mol. dynamics. The heparin docking method is available as an advanced option of the ClusPro server at http://cluspro.bu.edu/.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC2cXhtVKnsLjK&md5=82dedc6c1e29961cbe8eb8a86957e582
11. 11
  Nivedha, A. K.; Thieker, D. F.; Makeneni, S.; Hu, H.; Woods, R. J. Vina-Carb: Improving Glycosidic Angles during Carbohydrate Docking. J. Chem. Theory Comput. 2016, 12 (2), 892– 901, DOI: 10.1021/acs.jctc.5b00834
  11
  Vina-Carb: Improving Glycosidic Angles during Carbohydrate Docking
  Nivedha, Anita K.; Thieker, David F.; Hu, Huimin; Woods, Robert J.
  Journal of Chemical Theory and Computation (2016), 12 (2), 892-901CODEN: JCTCCE; ISSN:1549-9618. (American Chemical Society)
  Mol. docking programs are primarily designed to align rigid, drug-like fragments into the binding sites of macromols. and frequently display poor performance when applied to flexible carbohydrate mols. A crit. source of flexibility within an oligosaccharide is the glycosidic linkages. Recently, Carbohydrate Intrinsic (CHI) energy functions are reported that attempt to quantify the glycosidic torsion angle preferences. The CHI-energy functions have been incorporated into the AutoDock Vina (ADV) scoring function, subsequently termed Vina-Carb (VC). Two user-adjustable parameters have been introduced, namely, a CHI- energy wt. term (chi_coeff) that affects the magnitude of the CHI-energy penalty and a CHI-cutoff term (chi_cutoff) that negates CHI-energy penalties below a specified value. A data set consisting of 101 protein-carbohydrate complexes and 29 apoprotein structures was used in the development and testing of VC, including antibodies, lectins, and carbohydrate binding modules. Accounting for the intramol. energies of the glycosidic linkages in the oligosaccharides during docking led VC to produce acceptable structures within the top five ranked poses in 74% of the systems tested, compared to a success rate of 55% for ADV. An enzyme system was employed to illustrate the potential application of VC to proteins that may distort glycosidic linkages of carbohydrate ligands upon binding. VC represents a significant step toward accurately predicting the structures of protein-carbohydrate complexes. Furthermore, the described approach is conceptually applicable to any class of ligands that populate well-defined conformational states.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XltlCrsQ%253D%253D&md5=cf018be5775bfaa2ad035ed1b82305b6
12. 12
  Boittier, E. D.; Burns, J. M.; Gandhi, N. S.; Ferro, V. GlycoTorch Vina: Docking Designed and Tested for Glycosaminoglycans. J. Chem. Inf. Model. 2020, 60 (12), 6328– 6343, DOI: 10.1021/acs.jcim.0c00373
  There is no corresponding record for this reference.
13. 13
  Labonte, J. W.; Adolf-Bryfogle, J.; Schief, W. R.; Gray, J. J. Residue-centric Modeling and Design of Saccharide and Glycoconjugate Structures. J. Comput. Chem. 2017, 38 (5), 276– 287, DOI: 10.1002/jcc.24679
  13
  Residue-centric modeling and design of saccharide and glycoconjugate structures
  Labonte, Jason W.; Adolf-Bryfogle, Jared; Schief, William R.; Gray, Jeffrey J.
  Journal of Computational Chemistry (2017), 38 (5), 276-287CODEN: JCCHDD; ISSN:0192-8651. (John Wiley & Sons, Inc.)
  The RosettaCarbohydrate framework is a new tool for modeling a wide variety of saccharide and glycoconjugate structures. This report describes the development of the framework and highlights its applications. The framework integrates with established protocols within the Rosetta modeling and design suite, and it handles the vast complexity and variety of carbohydrate mols., including branching and sugar modifications. To address challenges of sampling and scoring, RosettaCarbohydrate can sample glycosidic bonds, side-chain conformations, and ring forms, and it uses a glycan-specific term within its scoring function. Rosetta can work with std. PDB, GLYCAM, and GlycoWorkbench (.gws) file formats. Saccharide residue-specific chem. information is stored internally, permitting glycoengineering and design. Carbohydrate-specific applications described herein include virtual glycosylation, loop-modeling of carbohydrates, and docking of glyco-ligands to antibodies. Benchmarking data are presented and compared to other studies, demonstrating Rosetta's ability to predict glyco-ligand binding. The framework expands the tools available to glycoscientists and engineers.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28XhvFKjsLnO&md5=822f98946c2d848b46f06b581beb11b4
14. 14
  Glashagen, G.; de Vries, S.; Uciechowska-Kaczmarzyk, U.; Samsonov, S. A.; Murail, S.; Tuffery, P.; Zacharias, M. Coarse-grained and Atomic Resolution Biomolecular Docking with the ATTRACT Approach. Proteins: Struct., Funct., Bioinf. 2020, 88 (8), 1018– 1028, DOI: 10.1002/prot.25860
  There is no corresponding record for this reference.
15. 15
  Jumper, J.; Evans, R.; Pritzel, A.; Green, T.; Figurnov, M.; Ronneberger, O.; Tunyasuvunakool, K.; Bates, R.; Žídek, A.; Potapenko, A.; Bridgland, A.; Meyer, C.; Kohl, S. A. A.; Ballard, A. J.; Cowie, A.; Romera-Paredes, B.; Nikolov, S.; Jain, R.; Adler, J.; Back, T.; Petersen, S.; Reiman, D.; Clancy, E.; Zielinski, M.; Steinegger, M.; Pacholska, M.; Berghammer, T.; Bodenstein, S.; Silver, D.; Vinyals, O.; Senior, A. W.; Kavukcuoglu, K.; Kohli, P.; Hassabis, D. Highly Accurate Protein Structure Prediction with AlphaFold. Nature 2021, 596 (7873), 583– 589, DOI: 10.1038/s41586-021-03819-2
  15
  Highly accurate protein structure prediction with AlphaFold
  Jumper, John; Evans, Richard; Pritzel, Alexander; Green, Tim; Figurnov, Michael; Ronneberger, Olaf; Tunyasuvunakool, Kathryn; Bates, Russ; Zidek, Augustin; Potapenko, Anna; Bridgland, Alex; Meyer, Clemens; Kohl, Simon A. A.; Ballard, Andrew J.; Cowie, Andrew; Romera-Paredes, Bernardino; Nikolov, Stanislav; Jain, Rishub; Adler, Jonas; Back, Trevor; Petersen, Stig; Reiman, David; Clancy, Ellen; Zielinski, Michal; Steinegger, Martin; Pacholska, Michalina; Berghammer, Tamas; Bodenstein, Sebastian; Silver, David; Vinyals, Oriol; Senior, Andrew W.; Kavukcuoglu, Koray; Kohli, Pushmeet; Hassabis, Demis
  Nature (London, United Kingdom) (2021), 596 (7873), 583-589CODEN: NATUAS; ISSN:0028-0836. (Nature Portfolio)
  Proteins are essential to life, and understanding their structure can facilitate a mechanistic understanding of their function. Through an enormous exptl. effort, the structures of around 100,000 unique proteins have been detd., but this represents a small fraction of the billions of known protein sequences. Structural coverage is bottlenecked by the months to years of painstaking effort required to det. a single protein structure. Accurate computational approaches are needed to address this gap and to enable large-scale structural bioinformatics. Predicting the three-dimensional structure that a protein will adopt based solely on its amino acid sequence-the structure prediction component of the 'protein folding problem'-has been an important open research problem for more than 50 years. Despite recent progress, existing methods fall far short of at. accuracy, esp. when no homologous structure is available. Here we provide the first computational method that can regularly predict protein structures with at. accuracy even in cases in which no similar structure is known. We validated an entirely redesigned version of our neural network-based model, AlphaFold, in the challenging 14th Crit. Assessment of protein Structure Prediction (CASP14), demonstrating accuracy competitive with exptl. structures in a majority of cases and greatly outperforming other methods. Underpinning the latest version of AlphaFold is a novel machine learning approach that incorporates phys. and biol. knowledge about protein structure, leveraging multi-sequence alignments, into the design of the deep learning algorithm.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BB3MXhvVaktrrL&md5=25964ab1157cd5b74a437333dd86650d
16. 16
  Abramson, J.; Adler, J.; Dunger, J.; Evans, R.; Green, T.; Pritzel, A.; Ronneberger, O.; Willmore, L.; Ballard, A. J.; Bambrick, J.; Bodenstein, S. W.; Evans, D. A.; Hung, C.-C.; O’Neill, M.; Reiman, D.; Tunyasuvunakool, K.; Wu, Z.; Žemgulytė, A.; Arvaniti, E.; Beattie, C.; Bertolli, O.; Bridgland, A.; Cherepanov, A.; Congreve, M.; Cowen-Rivers, A. I.; Cowie, A.; Figurnov, M.; Fuchs, F. B.; Gladman, H.; Jain, R.; Khan, Y. A.; Low, C. M. R.; Perlin, K.; Potapenko, A.; Savy, P.; Singh, S.; Stecula, A.; Thillaisundaram, A.; Tong, C.; Yakneen, S.; Zhong, E. D.; Zielinski, M.; Žídek, A.; Bapst, V.; Kohli, P.; Jaderberg, M.; Hassabis, D.; Jumper, J. M. Accurate Structure Prediction of Biomolecular Interactions with AlphaFold 3. Nature 2024, 630 (8016), 493– 500, DOI: 10.1038/s41586-024-07487-w
  There is no corresponding record for this reference.
17. 17
  HADDOCK3, Bonvin’s Lab , 2022. https://github.com/haddocking/haddock3.
  There is no corresponding record for this reference.
18. 18
  Dominguez, C.; Boelens, R.; Bonvin, A. M. J. J. HADDOCK: A Protein–Protein Docking Approach Based on Biochemical or Biophysical Information. J. Am. Chem. Soc. 2003, 125 (7), 1731– 1737, DOI: 10.1021/ja026939x
  18
  HADDOCK: A protein-protein docking approach based on biochemical or biophysical information
  Dominguez, Cyril; Boelens, Rolf; Bonvin, Alexandre M. J. J.
  Journal of the American Chemical Society (2003), 125 (7), 1731-1737CODEN: JACSAT; ISSN:0002-7863. (American Chemical Society)
  The structure detn. of protein-protein complexes is a rather tedious and lengthy process, by both NMR and x-ray crystallog. Several methods based on docking to study protein complexes have also been well developed over the past few years. Most of these approaches are not driven by exptl. data but are based on a combination of energetics and shape complementarity. Here, we present an approach called HADDOCK (High Ambiguity Driven protein-protein Docking) that makes use of biochem. and/or biophys. interaction data such as chem. shift perturbation data resulting from NMR titrn. expts. or mutagenesis data. This information is introduced as Ambiguous Interaction Restraints (AIRs) to drive the docking process. An AIR is defined as an ambiguous distance between all residues shown to be involved in the interaction. The accuracy of our approach is demonstrated with three mol. complexes. For two of these complexes, for which both the complex and the free protein structures have been solved, NMR titrn. data were available. Mutagenesis data were used in the last example. In all cases, the best structures generated by HADDOCK, i.e., the structures with the lowest intermol. energies, were the closest to the published structure of the resp. complexes (within 2.0 Å backbone RMSD).
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3sXkvFGquw%253D%253D&md5=9cf6fd9c690afb0181579485ae263197
19. 19
  Wu, A. M.; Singh, T.; Liu, J.-H.; Krzeminski, M.; Russwurm, R.; Siebert, H.-C.; Bonvin, A. M. J. J.; André, S.; Gabius, H.-J. Activity–Structure Correlations in Divergent Lectin Evolution: Fine Specificity of Chicken Galectin CG-14 and Computational Analysis of Flexible Ligand Docking for CG-14 and the Closely Related CG-16. Glycobiology 2007, 17 (2), 165– 184, DOI: 10.1093/glycob/cwl062
  There is no corresponding record for this reference.
20. 20
  Krzeminski, M.; Singh, T.; André, S.; Lensch, M.; Wu, A. M.; Bonvin, A. M. J. J.; Gabius, H.-J. Human Galectin-3 (Mac-2 Antigen): Defining Molecular Switches of Affinity to Natural Glycoproteins, Structural and Dynamic Aspects of Glycan Binding by Flexible Ligand Docking and Putative Regulatory Sequences in the Proximal Promoter Region. Biochim. Biophys. Acta, Gen. Subj. 2011, 1810 (2), 150– 161, DOI: 10.1016/j.bbagen.2010.11.001
  There is no corresponding record for this reference.
21. 21
  Berman, H. M. The Protein Data Bank. Nucleic Acids Res. 2000, 28 (1), 235– 242, DOI: 10.1093/nar/28.1.235
  21
  The Protein Data Bank
  Berman, Helen M.; Westbrook, John; Feng, Zukang; Gilliland, Gary; Bhat, T. N.; Weissig, Helge; Shindyalov, Ilya N.; Bourne, Philip E.
  Nucleic Acids Research (2000), 28 (1), 235-242CODEN: NARHAD; ISSN:0305-1048. (Oxford University Press)
  The Protein Data Bank (PDB; http://www.rcsb.org/pdb/)is the single worldwide archive of structural data of biol. macromols. This paper describes the goals of the PDB, the systems in place for data deposition and access, how to obtain further information, and near-term plans for the future development of the resource.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3cXhvVKjt7w%253D&md5=227fb393f754be2be375ab727bfd05dc
22. 22
  Woods Group. Complex Carbohydrate Research Center; GLYCAM Web: University of Georgia, Athens, GA, 2023.
  There is no corresponding record for this reference.
23. 23
  Kirschner, K. N.; Yongye, A. B.; Tschampel, S. M.; González-Outeiriño, J.; Daniels, C. R.; Foley, B. L.; Woods, R. J. GLYCAM06: A Generalizable Biomolecular Force Field. Carbohydrates. J. Comput. Chem. 2008, 29 (4), 622– 655, DOI: 10.1002/jcc.20820
  23
  GLYCAM06: a generalizable biomolecular force field. Carbohydrates
  Kirschner Karl N; Yongye Austin B; Tschampel Sarah M; Gonzalez-Outeirino Jorge; Daniels Charlisa R; Foley B Lachele; Woods Robert J
  Journal of computational chemistry (2008), 29 (4), 622-55 ISSN:.
  A new derivation of the GLYCAM06 force field, which removes its previous specificity for carbohydrates, and its dependency on the AMBER force field and parameters, is presented. All pertinent force field terms have been explicitly specified and so no default or generic parameters are employed. The new GLYCAM is no longer limited to any particular class of biomolecules, but is extendible to all molecular classes in the spirit of a small-molecule force field. The torsion terms in the present work were all derived from quantum mechanical data from a collection of minimal molecular fragments and related small molecules. For carbohydrates, there is now a single parameter set applicable to both alpha- and beta-anomers and to all monosaccharide ring sizes and conformations. We demonstrate that deriving dihedral parameters by fitting to QM data for internal rotational energy curves for representative small molecules generally leads to correct rotamer populations in molecular dynamics simulations, and that this approach removes the need for phase corrections in the dihedral terms. However, we note that there are cases where this approach is inadequate. Reported here are the basic components of the new force field as well as an illustration of its extension to carbohydrates. In addition to reproducing the gas-phase properties of an array of small test molecules, condensed-phase simulations employing GLYCAM06 are shown to reproduce rotamer populations for key small molecules and representative biopolymer building blocks in explicit water, as well as crystalline lattice properties, such as unit cell dimensions, and vibrational frequencies.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A280%3ADC%252BD1c%252FotlSmtQ%253D%253D&md5=fd32f08518d8e3568627c36eaa958efb
24. 24
  Varki, A.; Cummings, R. D.; Aebi, M.; Packer, N. H.; Seeberger, P. H.; Esko, J. D.; Stanley, P.; Hart, G.; Darvill, A.; Kinoshita, T.; Prestegard, J. J.; Schnaar, R. L.; Freeze, H. H.; Marth, J. D.; Bertozzi, C. R.; Etzler, M. E.; Frank, M.; Vliegenthart, J. F.; Lütteke, T.; Perez, S.; Bolton, E.; Rudd, P.; Paulson, J.; Kanehisa, M.; Toukach, P.; Aoki-Kinoshita, K. F.; Dell, A.; Narimatsu, H.; York, W.; Taniguchi, N.; Kornfeld, S. Symbol Nomenclature for Graphical Representations of Glycans. Glycobiology 2015, 25 (12), 1323– 1324, DOI: 10.1093/glycob/cwv091
  24
  Symbol nomenclature for graphical representations of glycans
  Varki, Ajit; Cummings, Richard D.; Aebi, Markus; Packer, Nicole H.; Seeberger, Peter H.; Esko, Jeffrey D.; Stanley, Pamela; Hart, Gerald; Darvill, Alan; Kinoshita, Taroh; Prestegard, James J.; Schnaar, Ronald L.; Freeze, Hudson H.; Marth, Jamey D.; Bertozzi, Carolyn R.; Etzler, Marilynn E.; Frank, Martin; Vliegenthart, Johannes F. G.; Lutteke, Thomas; Perez, Serge; Bolton, Evan; Rudd, Pauline; Paulson, James; Kanehisa, Minoru; Toukach, Philip; Aoki-Kinoshita, Kiyoko F.; Dell, Anne; Narimatsu, Hisashi; York, William; Taniguchi, Naoyuki; Kornfeld, Stuart
  Glycobiology (2015), 25 (12), 1323-1324CODEN: GLYCE3; ISSN:0959-6658. (Oxford University Press)
  Symbol nomenclature for graphical representations of glycans.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC28Xhs1OqurzK&md5=6d906f3912a24468074ded519e6354c9
25. 25
  Neelamegham, S.; Aoki-Kinoshita, K.; Bolton, E.; Frank, M.; Lisacek, F.; Lütteke, T.; O’Boyle, N.; Packer, N. H.; Stanley, P.; Toukach, P.; Varki, A.; Woods, R. J. Updates to the Symbol Nomenclature for Glycans guidelines. Glycobiology 2019, 29 (9), 620– 624, DOI: 10.1093/glycob/cwz045
  25
  Updates to the symbol nomenclature for glycans guidelines
  Neelamegham, Sriram; Aoki-Kinoshita, Kiyoko; Bolton, Evan; Frank, Martin; Lisacek, Frederique; Lutteke, Thomas; O'Boyle, Noel; Packer, Nicolle H.; Stanley, Pamela; Toukach, Philip; Varki, Ajit; Woods, Robert J.; Darvill, Alan; Dell, Anne; Henrissat, Bernard; Bertozzi, Carolyn; Hart, Gerald; Narimatsu, Hisashi; Freeze, Hudson; Yamada, Issaku; Paulson, James; Prestegard, James; Marth, Jamey; Vliegenthart, Jfg; Etzler, Marilynn; Aebi, Markus; Kanehisa, Minoru; Taniguchi, Naoyuki; Edwards, Nathan; Rudd, Pauline; Seeberger, Peter; Mazumder, Raja; Ranzinger, Rene; Cummings, Richard; Schnaar, Ronald; Perez, Serge; Kornfeld, Stuart; Kinoshita, Taroh; York, William; Knirel, Yuriy
  Glycobiology (2019), 29 (9), 620-624CODEN: GLYCE3; ISSN:1460-2423. (Oxford University Press)
  A review. The Symbol Nomenclature for Glycans (SNFG) is a community-curated std. for the depiction of monosaccharides and complex glycans using various colored-coded, geometric shapes, along with defined text addns. It is hosted by the National Center for Biotechnol. Information (NCBI) at the NCBI-Glycans Page. Several changes have been made to the SNFG page in the past year to update the rules for depicting glycans using the SNFG, to include more examples of use, particularly for non-mammalian organisms, and to provide guidelines for the depiction of ambiguous glycan structures. This Glycoforum article summarizes these recent changes.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BB3cXps1Witrk%253D&md5=b7bb24d9cf3e0b8ff3edb88960186d77
26. 26
  de Vries, S. J.; van Dijk, A. D. J.; Krzeminski, M.; van Dijk, M.; Thureau, A.; Hsu, V.; Wassenaar, T.; Bonvin, A. M. J. J. HADDOCK versus HADDOCK: New Features and Performance of HADDOCK2.0 on the CAPRI Targets. Proteins: Struct., Funct., Bioinf. 2007, 69 (4), 726– 733, DOI: 10.1002/prot.21723
  There is no corresponding record for this reference.
27. 27
  Jorgensen, W. L.; Tirado-Rives, J. The OPLS [Optimized Potentials for Liquid Simulations] Potential Functions for Proteins, Energy Minimizations for Crystals of Cyclic Peptides and Crambin. J. Am. Chem. Soc. 1988, 110 (6), 1657– 1666, DOI: 10.1021/ja00214a001
  27
  The OPLS [optimized potentials for liquid simulations] potential functions for proteins, energy minimizations for crystals of cyclic peptides and crambin
  Jorgensen, William L.; Tirado-Rives, Julian
  Journal of the American Chemical Society (1988), 110 (6), 1657-66CODEN: JACSAT; ISSN:0002-7863.
  A complete set of intermol. potential functions was developed for use in computer simulations of proteins in their native environment. Parameters are reported for 25 peptide residues as well as the common neutral and charged terminal groups. The potential functions have the simple Coulomb plus Lennard-Jones form and are compatible with the widely used models for water, TIP4P, TIP3P, and SPC. The parameters were obtained and tested primarily in conjunction with Monte Carlo statistical mechanics simulations of 36 pure org. liqs. and numerous aq. solns. of org. ions representative of subunits in the side chains and backbones of proteins. Bond stretch, angle bend, and torsional terms were adopted from the AMBER united-atom force field. As reported here, further testing involved studies of conformational energy surfaces and optimizations of the crystal structures for 4 cyclic hexapeptides and a cyclic pentapeptide. The av. root mean square deviation from the x-ray structures of the crystals is only 0.17 Å for the at. positions and 3% for the unit cell vols. A more crit. test was then provided by performing energy minimizations for the complete crystal of the protein crambin, including 182 water mols. that were initially placed via a Monte Carlo simulation. The resultant root mean square deviation for the non-H atoms is still only 0.17 Å and the variation in the errors for charged, polar, and nonpolar residues is small. Significant improvement is apparent over the AMBER united-atom force field which was previously demonstrated to be superior to many alternatives.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaL1cXht1yjt7Y%253D&md5=b6c901d8c295b3b37329a7faef527e12
28. 28
  Fernández-Recio, J.; Totrov, M.; Abagyan, R. Identification of Protein–Protein Interaction Sites from Docking Energy Landscapes. J. Mol. Biol. 2004, 335 (3), 843– 865, DOI: 10.1016/j.jmb.2003.10.069
  28
  Identification of Protein-Protein Interaction Sites from Docking Energy Landscapes
  Fernandez-Recio, Juan; Totrov, Maxim; Abagyan, Ruben
  Journal of Molecular Biology (2004), 335 (3), 843-865CODEN: JMOBAK; ISSN:0022-2836. (Elsevier)
  Protein recognition is one of the most challenging and intriguing problems in structural biol. Despite all the available structural, sequence and biophys. information about protein-protein complexes, the physico-chem. patterns, if any, that make a protein surface likely to be involved in protein-protein interactions, remain elusive. Here, we apply protein docking simulations and anal. of the interaction energy landscapes to identify protein-protein interaction sites. The new protocol for global docking based on multi-start global energy optimization of an all-atom model of the ligand, with detailed receptor potentials and at. solvation parameters optimized in a training set of 24 complexes, explores the conformational space around the whole receptor without restrictions. The ensembles of the rigid-body docking solns. generated by the simulations were subsequently used to project the docking energy landscapes onto the protein surfaces. We found that highly populated low-energy regions consistently corresponded to actual binding sites. The procedure was validated on a test set of 21 known protein-protein complexes not used in the training set. As much as 81% of the predicted high-propensity patch residues were located correctly in the native interfaces. This approach can guide the design of mutations on the surfaces of proteins, provide geometrical details of a possible interaction, and help to annotate protein surfaces in structural proteomics.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD3sXpvVWltLg%253D&md5=8a7687b9ecf2f39625e0511d3ec71cc5
29. 29
  Basciu, A.; Koukos, P. I.; Malloci, G.; Bonvin, A. M. J. J.; Vargiu, A. V. Coupling Enhanced Sampling of the Apo-Receptor with Template-Based Ligand Conformers Selection: Performance in Pose Prediction in the D3R Grand Challenge 4. J. Comput. Aided Mol. Des 2020, 34 (2), 149– 162, DOI: 10.1007/s10822-019-00244-6
  29
  Coupling enhanced sampling of the apo-receptor with template-based ligand conformers selection: performance in pose prediction in the D3R Grand Challenge 4
  Basciu, Andrea; Koukos, Panagiotis I.; Malloci, Giuliano; Bonvin, Alexandre M. J. J.; Vargiu, Attilio V.
  Journal of Computer-Aided Molecular Design (2020), 34 (2), 149-162CODEN: JCADEQ; ISSN:0920-654X. (Springer)
  We report the performance of our newly introduced Ensemble Docking with Enhanced sampling of pocket Shape (EDES) protocol coupled to a template-based algorithm to generate near-native ligand conformations in the 2019 iteration of the Grand Challenge (GC4) organized by the D3R consortium. Using either AutoDock4.2 or HADDOCK2.2 docking programs (each software in two variants of the protocol) our method generated native-like poses among the top 5 submitted for evaluation for most of the 20 targets with similar performances. The protein selected for GC4 was the human beta-site amyloid precursor protein cleaving enzyme 1 (BACE-1), a transmembrane aspartic-acid protease. We identified at least one pose whose heavy-atoms RMSD was less than 2.5 Å from the native conformation for 16 (80%) and 17 (85%) of the 20 targets using AutoDock and HADDOCK, resp. Dissecting the possible sources of errors revealed that: (i) our EDES protocol (with minor modifications) was able to sample sub-angstrom conformations for all 20 protein targets, reproducing the correct conformation of the binding site within ∼ 1 Å RMSD; (ii) as already shown by some of us in GC3, even in the presence of near-native protein structures, a proper selection of ligand conformers is crucial for the success of ensemble-docking calcns. Importantly, our approach performed best among the protocols exploiting only structural information of the apo protein to generate conformations of the receptor for ensemble-docking calcns.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BC1MXitFGit73J&md5=af95398263b2f4c630dc50847867a518
30. 30
  Nilges, M. A calculation strategy for the structure determination of symmetric demers by ¹H NMR. Proteins: Struct., Funct., Bioinf. 1993, 17 (3), 297– 309, DOI: 10.1002/prot.340170307
  There is no corresponding record for this reference.
31. 31
  Wallace, A. C.; Laskowski, R. A.; Thornton, J. M. LIGPLOT: A Program to Generate Schematic Diagrams of Protein-Ligand Interactions. Protein Eng., Des. Sel. 1995, 8 (2), 127– 134, DOI: 10.1093/protein/8.2.127
  There is no corresponding record for this reference.
32. 32
  McDonald, I. K.; Thornton, J. M. Satisfying Hydrogen Bonding Potential in Proteins. J. Mol. Biol. 1994, 238 (5), 777– 793, DOI: 10.1006/jmbi.1994.1334
  32
  Satisfying hydrogen bonding potential in proteins
  McDonald, Ian K.; Thornton, Janet M.
  Journal of Molecular Biology (1994), 238 (5), 777-93CODEN: JMOBAK; ISSN:0022-2836.
  The authors have analyzed the frequency with which potential hydrogen bond donors and acceptors are satisfied in protein mols. There are a small percentage of nitrogen or oxygen atoms that do not form hydrogen bonds with either solvent or protein atoms, when std. criteria are used. For high resoln. structures, 9.5% and 5.1% of buried main-chain nitrogen and oxygen atoms, resp., fail to hydrogen bond under the authors' std. criteria, representing 5.8% and 2.1% of all main-chain nitrogen and oxygen atoms. The authors find that as the resoln. of the data improves, the percentages fall. If the hydrogen bond criteria are relaxed many of these unsatisfied atoms form weak hydrogen bonds. However, there remain some buried atoms (1.3% NH and 1.8% CO) that fail to hydrogen bond without any immediately obvious compensating interactions.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADyaK2cXltVagt7k%253D&md5=0f21bf6f6b1ae5f5f70235da81f426b5
33. 33
  Méndez, R.; Leplae, R.; De Maria, L.; Wodak, S. J. Assessment of Blind Predictions of Protein–Protein Interactions: Current Status of Docking Methods. Proteins: Struct., Funct., Bioinf. 2003, 52 (1), 51– 67, DOI: 10.1002/prot.10393
  There is no corresponding record for this reference.
34. 34
  Giulini, M.; Menichetti, R.; Shell, M. S.; Potestio, R. An Information-Theory-Based Approach for Optimal Model Reduction of Biomolecules. J. Chem. Theory Comput. 2020, 16 (11), 6795– 6813, DOI: 10.1021/acs.jctc.0c00676
  34
  An Information-Theory-Based Approach for Optimal Model Reduction of Biomolecules
  Giulini, Marco; Menichetti, Roberto; Shell, M. Scott; Potestio, Raffaello
  Journal of Chemical Theory and Computation (2020), 16 (11), 6795-6813CODEN: JCTCCE; ISSN:1549-9618. (American Chemical Society)
  In theor. modeling of a phys. system, a crucial step consists of the identification of those degrees of freedom that enable a synthetic yet informative representation of it. While in some cases this selection can be carried out on the basis of intuition and experience, straightforward discrimination of the important features from the negligible ones is difficult for many complex systems, most notably heteropolymers and large biomols. The authors here present a thermodn.-based theor. framework to gauge the effectiveness of a given simplified representation by measuring its information content. The authors employ this method to identify those reduced descriptions of proteins, in terms of a subset of their atoms, that retain the largest amt. of information from the original model; these highly informative representations share common features that are intrinsically related to the biol. properties of the proteins under examn., thereby establishing a bridge between protein structure, energetics, and function.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BB3cXit1alt73P&md5=03a82b725a5399cc9502639f62485100
35. 35
  Sokal, R. R.; Michener, C. D. A Statistical Method for Evaluating Systematic Relationships; University of Kansas Science Bulletin, 1958; Vol. 38, pp 1409– 1438.
  There is no corresponding record for this reference.
36. 36
  Charitou, V.; van Keulen, S. C.; Bonvin, A. M. J. J. Cyclization and Docking Protocol for Cyclic Peptide–Protein Modeling Using HADDOCK2.4. J. Chem. Theory Comput. 2022, 18 (6), 4027– 4040, DOI: 10.1021/acs.jctc.2c00075
  There is no corresponding record for this reference.
37. 37
  Buchanan, C. J.; Gaunt, B.; Harrison, P. J.; Yang, Y.; Liu, J.; Khan, A.; Giltrap, A. M.; Le Bas, A.; Ward, P. N.; Gupta, K.; Dumoux, M.; Tan, T. K.; Schimaski, L.; Daga, S.; Picchiotti, N.; Baldassarri, M.; Benetti, E.; Fallerini, C.; Fava, F.; Giliberti, A.; Koukos, P. I.; Davy, M. J.; Lakshminarayanan, A.; Xue, X.; Papadakis, G.; Deimel, L. P.; Casablancas-Antràs, V.; Claridge, T. D. W.; Bonvin, A. M. J. J.; Sattentau, Q. J.; Furini, S.; Gori, M.; Huo, J.; Owens, R. J.; Schaffitzel, C.; Berger, I.; Renieri, A.; Naismith, J. H.; Baldwin, A. J.; Davis, B. G.; Davis, B. G. Pathogen-Sugar Interactions Revealed by Universal Saturation Transfer Analysis. Science 2022, 377 (6604), eabm3125 DOI: 10.1126/science.abm3125
  There is no corresponding record for this reference.
38. 38
  Koukos, P. I.; Réau, M.; Bonvin, A. M. J. J. Shape-Restrained Modeling of Protein–Small-Molecule Complexes with High Ambiguity Driven DOCKing. J. Chem. Inf. Model. 2021, 61 (9), 4807– 4818, DOI: 10.1021/acs.jcim.1c00796
  There is no corresponding record for this reference.
39. 39
  Kerzmann, A.; Neumann, D.; Kohlbacher, O. SLICK – Scoring and Energy Functions for Protein–Carbohydrate Interactions. J. Chem. Inf. Model. 2006, 46 (4), 1635– 1642, DOI: 10.1021/ci050422y
  39
  SLICK - Scoring and Energy Functions for Protein-Carbohydrate Interactions
  Kerzmann, Andreas; Neumann, Dirk; Kohlbacher, Oliver
  Journal of Chemical Information and Modeling (2006), 46 (4), 1635-1642CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
  Protein-carbohydrate interactions are increasingly being recognized as essential for many important biomol. recognition processes. From these, numerous biomedical applications arise in areas as diverse as drug design, immunol., or drug transport. The authors introduce SLICK, a package contg. a scoring and an energy function, which were specifically designed to predict binding modes and free energies of sugars and sugarlike compds. to proteins. SLICK accounts for van der Waals interactions, solvation effects, electrostatics, hydrogen bonds, and CH···π interactions, the latter being a particular feature of most protein-carbohydrate interactions. Parameters for the empirical energy function were calibrated on a set of high-resoln. crystal structures of protein-sugar complexes with known exptl. binding free energies. The authors show that SLICK predicts the binding free energies of predicted complexes (through mol. docking) with high accuracy. SLICK is available as part of the authors' mol. modeling package BALL (www.ball-project.org).
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD28Xksl2ksbY%253D&md5=af9f4a1e22709fcf8e39e6595b11a501
40. 40
  Kerzmann, A.; Fuhrmann, J.; Kohlbacher, O.; Neumann, D. BALLDock/SLICK: A New Method for Protein-Carbohydrate Docking. J. Chem. Inf. Model. 2008, 48 (8), 1616– 1625, DOI: 10.1021/ci800103u
  40
  BALLDock/SLICK: A New Method for Protein-Carbohydrate Docking
  Kerzmann, Andreas; Fuhrmann, Jan; Kohlbacher, Oliver; Neumann, Dirk
  Journal of Chemical Information and Modeling (2008), 48 (8), 1616-1625CODEN: JCISD8; ISSN:1549-9596. (American Chemical Society)
  Protein-ligand docking is an essential technique in computer-aided drug design. While generally available docking programs work well for most drug classes, carbohydrates and carbohydrate-like compds. are often problematic for docking. The authors present a new docking method specifically designed to handle docking of carbohydrate-like compds. BALLDock/SLICK combines an evolutionary docking algorithm for flexible ligands and flexible receptor side chains with carbohydrate-specific scoring and energy functions. The scoring function has been designed to identify accurate ligand poses, while the energy function yields accurate ests. of the binding free energies of these poses. On a test set of known protein-sugar complexes the authors demonstrate the ability of the approach to generate correct poses for almost all of the structures and achieve very low mean errors for the predicted binding free energies.
  >> More from SciFinder ^®
  https://chemport.cas.org/services/resolver?origin=ACS&resolution=options&coi=1%3ACAS%3A528%3ADC%252BD1cXovVWitb8%253D&md5=e34cd7fe7693e397ad796306b8479314
41. 41
  Ives, C.; Singh, O.; D’Andrea, S.; Fogarty, C.; Harbison, A.; Satheesan, A.; Tropea, B.; Fadda, E. Restoring Protein Glycosylation with GlycoShape. bioRxiv 2023, DOI: 10.1101/2023.12.11.571101
  There is no corresponding record for this reference.
42. 42
  Meyer, P. A.; Socias, S.; Key, J.; Ransey, E.; Tjon, E. C.; Buschiazzo, A.; Lei, M.; Botka, C.; Withrow, J.; Neau, D.; Rajashankar, K.; Anderson, K. S.; Baxter, R. H.; Blacklow, S. C.; Boggon, T. J.; Bonvin, A. M. J. J.; Borek, D.; Brett, T. J.; Caflisch, A.; Chang, C.-I.; Chazin, W. J.; Corbett, K. D.; Cosgrove, M. S.; Crosson, S.; Dhe-Paganon, S.; Di Cera, E.; Drennan, C. L.; Eck, M. J.; Eichman, B. F.; Fan, Q. R.; Ferré-D’Amaré, A. R.; Christopher Fromme, J.; Garcia, K. C.; Gaudet, R.; Gong, P.; Harrison, S. C.; Heldwein, E. E.; Jia, Z.; Keenan, R. J.; Kruse, A. C.; Kvansakul, M.; McLellan, J. S.; Modis, Y.; Nam, Y.; Otwinowski, Z.; Pai, E. F.; Pereira, P. J. B.; Petosa, C.; Raman, C. S.; Rapoport, T. A.; Roll-Mecak, A.; Rosen, M. K.; Rudenko, G.; Schlessinger, J.; Schwartz, T. U.; Shamoo, Y.; Sondermann, H.; Tao, Y. J.; Tolia, N. H.; Tsodikov, O. V.; Westover, K. D.; Wu, H.; Foster, I.; Fraser, J. S.; Maia, F. R. N. C.; Gonen, T.; Kirchhausen, T.; Diederichs, K.; Crosas, M.; Sliz, P. Data Publication with the Structural Biology Data Grid Supports Live Analysis. Nat. Commun. 2016, 7 (1), 10882, DOI: 10.1038/ncomms10882
  There is no corresponding record for this reference.
- PDB: 3AOF
Supporting Information
Supporting Information
The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jcim.4c01372.
- Data set used in this study (XLSX)
- Details of the preparation of protein and glycan structures; description of HADDOCK3 modules used in this work; details of the glycan conformational sampling protocol; SNFG representation of the glycans; modules and parameters used for bound docking; modules and parameters used for unbound docking; glycan conformational sampling scenarios; example of HADDOCK models satisfying the quality thresholds; HADDOCK3’ performance on the bound data set; glycans’ RMSD to their bound conformations; impact of mdref on glycan conformations; impact of the clustering on glycans’ lowest RMSD; examples of glycan ensembles of conformations; HADDOCK3’ performance with the ensembles of glycans; torsion angle analysis of glycosidic linkages: a comparison to HADDOCK flexible refinement models; and torsion angle analysis of glycosidic linkages: a comparison to HADDOCK short molecular dynamics refinement models (PDF)
- ci4c01372_si_001.xlsx (17.55 kb)
- ci4c01372_si_002.pdf (5.43 MB)
Terms & Conditions

Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.

Modeling Protein–Glycan Interactions with HADDOCKClick to copy article linkArticle link copied!

Journal of Chemical Information and Modeling

License Summary*

Abstract

License Summary*

License Summary*

License Summary*

Introduction

Methods

Benchmark Data Set Preparation

HADDOCK General Protocol and Scoring Function

Restraints to Drive the Docking

Docking Protocols for Bound and Unbound Data Sets

Figure 1

Model Quality Assessment

Glycans’ Conformational Sampling

Results and Discussion

Bound Docking Performance and the Impact of the Rigid-Body Scoring Function in the Ranking of Models

Figure 2

Impact of Glycan Structural Features and AIR Definition on the Bound Docking Performance

Unbound Docking

Figure 3

Figure 4

Can an Ensemble of Presampled Glycan Conformations Improve the Docking Performance?

Conclusions

Data Availability

Supporting Information

Terms & Conditions

Author Information

Abbreviations

References

Cited By

Journal of Chemical Information and Modeling

License Summary*

Article Views

Altmetric

Citations

Recommended Articles

Abstract

Figure 1

Figure 2

Figure 3

Figure 4

References

Supporting Information

Supporting Information

Terms & Conditions

Modeling Protein–Glycan Interactions with HADDOCK
Click to copy article linkArticle link copied!