Modeling large protein–glycosaminoglycan complexes using a fragment‐based approach

• Published in: Journal of computational chemistry Vol. 40; no. 14; pp. 1429 - 1439
• Main Authors: Samsonov, Sergey A., Zacharias, Martin, Chauvot de Beauchene, Isaure
• Format: Journal Article
• Language: English
• Published: Hoboken, USA John Wiley & Sons, Inc 30-05-2019; Wiley Subscription Services, Inc; Wiley
• Subjects: Binding sites; Biochemistry, Molecular Biology; Bioinformatics; Chemical Sciences; Cheminformatics; Computer Science; Docking; Fragmentation; fragment‐based docking; glycans modeling; Glycosaminoglycans; glycosaminoglycans docking; glycosaminoglycans‐protein complex; Life Sciences; Molecular dynamics; Proteins; Signaling; Structural Biology; glycans modeling; glycosaminoglycans-protein complex; glycosaminoglycans docking; fragment-based docking; Glycosaminoglycans-protein complex; Glycosaminoglycans docking
• DOI: 10.1002/jcc.25797
• EISSN: 1096-987X
• ISSN: 0192-8651

Glycosaminoglycans (GAGs), a major constituent of the extracellular matrix, participate in cell‐signaling by binding specific proteins. Structural data on protein–GAG interactions are crucial to understand and modulate these signaling processes, with potential applications in regenerative medicine. However, experimental and theoretical approaches used to study GAG–protein systems are challenged by GAGs high flexibility limiting the conformational sampling above a certain size, and by the scarcity of GAG‐specific docking tools compared to protein–protein or protein–drug docking approaches. We present for the first time an automated fragment‐based method for docking GAGs on a protein binding site. In this approach, trimeric GAG fragments are flexibly docked to the protein, assembled based on their spacial overlap, and refined by molecular dynamics. The method appeared more successful than the classical full‐ligand approach for most of 13 tested complexes with known structure. The approach is particularly promising for docking of long GAG chains, which represents a bottleneck for classical docking approaches applied to these systems. © 2019 Wiley Periodicals, Inc.
The high flexibility of GAGs makes their conformational sampling challenging above a GAG certain length, which is a main bottleneck in GAG–protein docking. A fragment‐based approach can tackle this problem in successive steps: local sampling by flexible fragment docking, assembly of the docked fragments based on sequence and spacial compatibility, and merging the assembled fragments. This method reveals a better initial placement of long GAGs on a protein binding site than classical docking methods.