Skip to Main content Skip to Navigation

Fragment­ based modeling of protein­-GAG complexes

Abstract : Glycosaminoglycans (GAGs) are linear anionic periodic poly-saccharides that bind to their protein targets in the extracellular matrix, and so participate in many cell-signaling processes. They are very promising targets for the design of novel functional biomaterials with medical applications such as bone and skin regeneration [Scharnweber et al, JMSM 2015]. Useful biological and therapeutical insights can be obtained from the structure of complexes between GAGs and their target proteins. However the experimental resolution as well as the modeling of their structure is highly challenging. The reasons are both GAG intrinsic properties, such as the high flexibility and conformational diversity of glycans, and the lack for computational tools particularly designed for protein-GAG systems. This currently limits the success of GAG docking to very short fragments [Samsonov and Pisabarro, Glycobiology 2016]. We present here a new fragment-based method to dock GAGs on a coarsly known protein binding site. We combine flexible docking of trimeric GAG fragments with Autodock and combinatorial assembly of the compatible poses into GAGs chains, followed by fully flexible refinement. Tested on a benchmark of 13 complexes with various GAG types (heparin, chondroitin sulfate and hyaluronic acids), the method could model 5-mers to 7-mers with the accuracy of 5 Å RMSD to the experimental structure for all of them, and 3 Å RMSD for half of them. This is the first reported automatized fragment-based docking method to successfully dock such diverse GAGs. In principle, the independence of this approach on the ligand's length allows to dock very long GAG chains, which has been a bottleneck for previously applied docking approaches for these systems. The results of this work contribute to enrich the sparse pool of computational tools specifically developed for protein-GAG complexes.
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01927283
Contributor : Isaure Chauvot de Beauchene <>
Submitted on : Monday, November 19, 2018 - 5:15:07 PM
Last modification on : Tuesday, December 18, 2018 - 4:40:22 PM
Document(s) archivé(s) le : Wednesday, February 20, 2019 - 3:54:37 PM

File

Poster-GGMM2017-A0.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01927283, version 1

Collections

Citation

Isaure Chauvot de Beauchêne, Sergey A Samsonov, Martin Zacharias. Fragment­ based modeling of protein­-GAG complexes. GGMM 2017 - 20e congrès du Groupe de Graphisme et Modelisation Moleculaire, May 2017, Reims, France. pp.1. ⟨hal-01927283⟩

Share

Metrics

Record views

95

Files downloads

26