Three-Dimensional Structures of Carbohydrates and Where to Find Them

Abstract

Analysis and systematization of accumulated data on carbohydrate structural diversity is a subject of great interest for structural glycobiology. Despite being a challenging task, development of computational methods for efficient treatment and management of spatial (3D) structural features of carbohydrates breaks new ground in modern glycoscience. This review is dedicated to approaches of chemo- and glyco-informatics towards 3D structural data generation, deposition and processing in regard to carbohydrates and their derivatives. Databases, molecular modeling and experimental data validation services, and structure visualization facilities developed for last five years are reviewed.

Keywords: PDB glycans; carbohydrate; database; glycoinformatics; model build; molecular modeling; spatial structure; structure validation; structure visualization; web-tool.

Figure 1

Typical components of a carbohydrate 3D structure exemplified on sucrose: (a) primary structure (in Symbol Nomenclature for Glycans (SNFG)); (b) superimposed conformational states and Cremer–Pople diagram; (c) conformational space of a two-torsion glycosidic linkage (Ramachandran plot); (d) transitions of glycosidic dihedrals.

Figure 2

Networking between glycoinformatics projects and related services that promotes achievement of data integration in glycomics. Reproduced with permission from [29], © 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.

Figure 3

NMR-validated conformational analysis of high-mannose oligosaccharide GM9 based on replica-exchange molecular dynamics (REMD) simulation results. (a) Superimposition of 260 GM9 conformers extracted from REMD trajectory (black—GlcNAc, green—Man, blue—Glc). (b) primary structure of the GM9 oligosaccharide (SNFG representation). (c) REMD density maps for φ-ψ torsions of GM9 branch (Glc₁Man₃). Red dots locate glycosidic torsion angles derived from crystallographic data of Glc₁Man₃ tetrasaccharide ligand complexed with the lectin domain of calreticulin (PDB ID: 3O0W). Panels (a) and (c) were reproduced with permission from [149], © 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.

Figure 4

Citations of dedicated force fields in carbohydrate studies for the recent five years, according to Scopus. Outer circle shows total citations (number of citing publications) of force fields in 2015–2020. Inner circle shows citations in articles filtered by a carbohydrate topic. See detailed data, references to original publications, absolute values, and carbohydrate filer details in Supplementary Table S2.

Figure 5

Digest of the most commonly used carbohydrate force fields with parameterization protocol comparison. Reproduced with permission from [138], © 2020 Elsevier Inc.

Figure 6

Interplay of the instrumental and computational methods in the 3D structure determination of carbohydrates, proteins, and protein–glycoconjugate complexes. Reproduced from [285] © 2020 The authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA.

Figure 7

X-ray diffraction data refinement of N-glycan moiety from PDB ID 2Z62. 2mF_o–DF_c electron density map contoured at 1σ is displayed in grey; positive and negative mF_o–DF_c difference electron density maps contoured at 3σ are displayed in green and red, respectively. (a) Original glycan structure model from the PDB entry. (b) PDB-REDO model with properly renamed fucose residue and improved fit to the electron density. (c) Manually rebuilt model based on PDB-REDO results. (d) CARP distribution plot for glycosidic φ-ψ torsions of FUC(1-6)NAG (from panel (a)) in PDB. Characteristic points: R, model refined with PDB-REDO; P, original PDB model; M, manually rebuilt model. Reproduced from [295], © 2020 The authors. Published by John Wiley & Sons, Inc.

Figure 8

M. catarrhalis lgt2Δ structure validation based on NOE data analysis. (a) Characteristic proton-proton contacts; (b) NOE-filtered (blue boxes) sampling of proton-proton distances from MD simulation (grey shades). Reproduced from [314], © 2020 The authors. Licensee MDPI, Basel, Switzerland.

Figure 9

Distribution of D- (shown in blue) and L-pyranoside (shown in yellow) ring conformations as function of resolution for all sugar moieties in N-glycosylated proteins in PDB (on April 2019) solved with (a) X-ray crystallography and (b) electron cryo-microscopy. Non-chair conformations are bordered by dotted line boxes for 0.0-6.0 Å (green) and 6.0-10.0 Å (red) resolution ranges; the percentage of structures is given in the boxes. Reproduced with permission from [301], © 2020 Elsevier Ltd.

Figure 10

Deposition statistics of carbohydrate-containing structures in Protein Data Bank based on carbohydrate remediated list data. Data for 2020 cover seven of twelve months. See detailed data in Supplementary Tables S3–S4.

Figure 11

Glycan structure colored according to SNFG, or superimposed with 3D SNFG, as implemented in SweetUnityMol (a), GLYCOSCIENCES.de (via JSmol) (b), and CSDB (via JSmol) (c,d), see text. Panel (a) was reproduced with permission from [372], © Springer Japan 2017.

Figure 12

Glycan structure colored according to SNFG, or superimposed with 3D SNFG, as implemented in 3D-SNFG (a), LiteMol (b), Mol* (c); monosaccharide presentation in Glycoblocks (d). Panel (a) was reproduced with permission from [366], © 2020, Oxford University Press. Panel (d) was reproduced from [369], © 2020 The authors. Published by John Wiley & Sons, Inc.